UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Emerging powers and systemic change : China's impact on global commodity markets Massot, Pascale 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2015_november_massot_pascale.pdf [ 16.34MB ]
Metadata
JSON: 24-1.0165771.json
JSON-LD: 24-1.0165771-ld.json
RDF/XML (Pretty): 24-1.0165771-rdf.xml
RDF/JSON: 24-1.0165771-rdf.json
Turtle: 24-1.0165771-turtle.txt
N-Triples: 24-1.0165771-rdf-ntriples.txt
Original Record: 24-1.0165771-source.json
Full Text
24-1.0165771-fulltext.txt
Citation
24-1.0165771.ris

Full Text

   EMERGING POWERS AND SYSTEMIC CHANGE: CHINA’S IMPACT ON GLOBAL COMMODITY MARKETS    by    Pascale Massot   B.A., University of Montreal, 2006 M.A., The University of British Columbia, 2009     A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF   DOCTOR OF PHILOSOPHY   in   The Faculty of Graduate and Postdoctoral Studies  (Political Science)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   August 2015    © Pascale Massot, 2015   ii ABSTRACT How are global economic institutions transformed at times of power transition? Why have some international markets for important raw materials undergone fundamental change in the way they operate as a result of China’s emergence, while other such markets have been more resilient to change? The goal of this dissertation is to explain diverging global outcomes from the dramatic and contemporary expansion of China’s economy. By doing so, I shed new light on the political economy of global markets, why they operate the way they do now, and how they have evolved over time. I trace key variances in China’s effect on global markets to the interaction of Chinese domestic industrial structures and the pre-existing structures of global commodity markets. The structure of key industries within China varies: some are concentrated, some fragmented, some very sensitive to price signals, and others less so. Likewise, the structures of various global commodity markets varied significantly before China’s emergence as a dominant global consumer in the twenty-first century.  I argue that transformations in market power relations between consumers and suppliers increase the likelihood of institutional change in global markets. Price trends influence market stakeholders’ preferences for global pricing regimes, but they cannot fully explain the direction of change. Market power – including the capacity to coordinate others and the capacity to extract rent – also motivates behaviour.  Combining comparative case analysis of the iron ore, potash and uranium markets with careful process-tracing, I unveil the full picture, from domestic variables to international-level outcomes. I show the tremendous concentration of market power in global markets prior to China’s emergence; that China’s market power, despite its economic size, is in many ways weak; that some of the largest systemic changes have been the result of this Chinese position of weakness; and that China’s emergence has led to marketization, despite it being a state-led hybrid economy. This is a study of institutional resonance and complementarity between global markets and their systemically relevant consumers. More broadly, this dissertation seeks to contribute to ongoing debates about the systemic resilience of global market structures, and the domestic determinants of global economic power.      iii PREFACE  This dissertation is an original intellectual product of the author, Pascale Massot. Field research for this dissertation was conducted under the title “China and Global Commodity Markets” (H11-03440), which was approved by the UBC Behavioural Research Ethics Board on January 23, 2012 (Principal Investigator, Yves Tiberghien).   My field research in China was made possible through official affiliations with the Chinese Academy of Social Sciences in Beijing, as well as with the Center of International Political Economy at Peking University.    Some of the work in this dissertation has been published elsewhere.   Components of my theory chapter were previously published here:  Massot, Pascale. 2013. The Financialization of global commodity markets. Global Summitry Project, Munk School of Global Affairs, University of Toronto. Accessed May 14, 2015, http://globalsummitryproject.com.s197331.gridserver.com/blog/global-summitry-report/the-financialization-of-global-commodity-markets/ and here: Massot, Pascale. 2012. China's liberalizing impact on global commodity markets. Asia Pacific Memo. Accessed May 14, 2015, http://www.asiapacificmemo.ca/chinas-liberalizing-impact-on-global-commodity-markets  A portion of Chapter 4 was originally published here:  Massot, Pascale. 2014. Rough sailing in the iron ore shipping business. East Asia Forum, October 3. Accessed May 14, 2015, http://www.eastasiaforum.org/2014/10/03/rough-sailing-in-the-iron-ore-shipping-business/   Versions of Chapter 3 and 4 were presented here:   Massot, Pascale. 2014. “Emerging states and systemic change in global markets: China’s impact on the global iron ore market and related shipping industry,” International Studies Association Annual Conference, March 28, Toronto, Canada and here:  Massot, Pascale. 2015. “Power transitions and systemic change in global markets: Japanese and Chinese emergence as number one iron ore consumers,” International Studies Association Annual Conference, February 20, New Orleans, U.S.  A small portion of the empirical content of Chapter 5 was previously published here: Massot, Pascale. 2011. “Chinese state investments in Canada: Lessons from the potash saga,” Canada-Asia Agenda, Asia Pacific Foundation of Canada. Accessed May 14, 2015, http://www.asiapacific.ca/canada-asia-agenda/chinese-state-investments-canada-lessons-potash-saga  Earlier versions of Chapter 3 and 5 were presented here:   iv Massot, Pascale. 2014. “Emerging states and systemic change in global markets: A two-level political economy of China's impact on the global potash market,” International Studies Association Annual Conference, March 26, Toronto, Canada. and here: Massot, Pascale. 2013. “China’s liberalizing impact on global commodity markets: The cases of iron ore and potash,” Association of Asian Studies Annual Conference, March 22, San Diego, U.S.  Some of the empirical work and Chapter 6 was initially published in Chinese in an op-ed: Massot, Pascale. 2012. “How can China deal with dependence on imported uranium?” (), China Business News Daily (), April 23, A6, http://www.yicai.com/news/2012/04/1654156.html  and here: Massot, Pascale. 2012. China and the global uranium market: Prospects for development and systemic risks (	). China Business News Research Report (), Beijing, Shanghai Business News Media Ltd., April 16, Energy and Resource series, Issue 9.  An earlier version of Chapter 6 was also previously published here:  Massot, Pascale and Zhanming Chen 2013. China and the global uranium market: Prospects for peaceful coexistence. Scientific World Journal. 2013: 1-11. As lead author, I was responsible for the drafting, argument, theory, structure, case study, and data gathering.  A later version of Chapter 6 was also presented here:  Massot, Pascale. 2014. “Two-level political economy of Chinese global uranium procurement patterns,” CESS Workshop: China’s Environment: Policy, Science, and Society in a Changing World, Centre for Chinese Research, Institute of Asian Research, UBC, April 28, Vancouver, Canada.             v TABLE OF CONTENTS ABSTRACT'........................................................................................................................................................'ii!PREFACE'.........................................................................................................................................................'iii!TABLE'OF'CONTENTS'...................................................................................................................................'v!LIST'OF'TABLES'.........................................................................................................................................'viii!LIST'OF'FIGURES'...........................................................................................................................................'ix!ACKNOWLEDGEMENTS'..............................................................................................................................'xi!DEDICATION'...............................................................................................................................................'xiv!CHAPTER'1'–'EMERGING'POWERS'AND'SYSTEMIC'CHANGE'..........................................................'1!1.1!EXISTING!APPROACHES!TO!THE!STUDY!OF!POWER!TRANSITION!AND!GLOBAL!MARKET!CHANGE!...........!4!1.2!RESEARCH!QUESTIONS!.............................................................................................................................................!7!1.3!ARGUMENT!.................................................................................................................................................................!8!1.4!WHY!EXTRACTIVE!COMMODITY!MARKETS?!......................................................................................................!13!1.5!WHY!CHINA?!............................................................................................................................................................!15!1.6!METHODOLOGY!........................................................................................................................................................!17!1.6.1$Case$selection$....................................................................................................................................................$19!1.6.2$Data$.......................................................................................................................................................................$20!1.7!DISSERTATION!OUTLINE!........................................................................................................................................!21!CHAPTER'2'–'GLOBAL'MARKETS'..........................................................................................................'23!2.1!WHAT!MARKETS!ARE!.............................................................................................................................................!27!2.2!GLOBAL!MARKETS!AS!INSTITUTIONAL!SYSTEMS!...............................................................................................!29!2.3!TYPOLOGY!OF!GLOBAL!MARKETS!.........................................................................................................................!32!2.4!THE!IMPACT!OF!A!DOMINANT!EMERGING!CONSUMER!.....................................................................................!35!2.5!DEPENDENT!VARIABLE:!GLOBAL!PRICING!REGIMES!........................................................................................!36!2.6!MARKETIZATION!......................................................................................................................................................!39!2.7!WHEN!DO!MARKET!ACTORS!PREFER!MARKETIZATION?!................................................................................!46!2.7.1$Open$economy$politics$...................................................................................................................................$47!2.7.2$German$historical$school$..............................................................................................................................$48!2.7.3$Relationship$with$price$trends$...................................................................................................................$49!CHAPTER'3'–'TWO?LEVEL'POLITICAL'ECONOMY'OF'GLOBAL'MARKET'CHANGE'.................'54!3.1!ALTERNATIVE!EXPLANATIONS!..............................................................................................................................!54!3.1.1$Liberal$approaches$.........................................................................................................................................$55!3.1.2$Hegemonic$Stability$Theory$........................................................................................................................$57!3.2!FOUNDATIONAL!SCHOLARSHIP!.............................................................................................................................!60!3.2.1$Domestic$vs$twoIlevel$explanations$.........................................................................................................$60!3.2.2$International$political$economy$of$resources$.....................................................................................$63!3.2.3$Comparative$political$economy$–$convergence/divergence$debate$.........................................$63!3.2.4$IPE$literature$on$the$domestic$determinants$of$global$economic$trends$...............................$64!3.3!THE!KEY!PLAYERS!...................................................................................................................................................!65!3.3.1$Beyond$the$unitary$actor$.............................................................................................................................$65!3.3.2$The$Chinese$state$.............................................................................................................................................$66!3.4!METHODOLOGY!........................................................................................................................................................!68!3.4.1$Case$selection$....................................................................................................................................................$68!3.4.2$Explanatory$variables$at$both$levels$are$necessary$to$explain$outcomes$..............................$74!3.5!ARGUMENT!...............................................................................................................................................................!76!  vi 3.5.1$Argument:$Likelihood$of$change$...............................................................................................................$78!3.5.2$Argument:$Direction$of$change$.................................................................................................................$80!3.6!INDEPENDENT!VARIABLES:!GLOBAL!AND!DOMESTIC!CONCENTRATION!LEVELS!........................................!83!3.7!CAUSAL!MECHANISM:!COORDINATION!CAPACITY!.............................................................................................!89!3.8!PROCESS!TRACING:!THE!POLITICS!OF!FRAGMENTATION!AND!COORDINATION!..........................................!93!3.9!SUMMARY!OF!CASES!................................................................................................................................................!94!3.9.1$Iron$ore$case$summary$..................................................................................................................................$94!3.9.2$Potash$case$summary$....................................................................................................................................$95!CHAPTER'4'–'CHINA’S'IMPACT'ON'THE'GLOBAL'IRON'ORE'MARKET'......................................'97!4.1!ARGUMENT!–!THE!IRON!ORE!CASE!......................................................................................................................!99!4.2!DOMESTICILEVEL!MARKET!STRUCTURES!.......................................................................................................!101!4.2.1$Iron$ore:$Relevance$to$China$...................................................................................................................$103!4.3!CASE!1:!THE!FALL!OF!THE!GLOBAL!IRON!ORE!BENCHMARK!PRICING!REGIME!........................................!103!4.4!INDEPENDENT!VARIABLE:!THE!GLOBAL!IRON!ORE!MARKET!......................................................................!104!4.5!GLOBAL!IRON!ORE!MARKET!PLAYERS:!COORDINATION!CAPACITY!............................................................!106!4.6!CHINA’S!DOMESTIC!IRON!ORE!MARKET!..........................................................................................................!109!4.7!INDEPENDENT!VARIABLE:!DOMESTIC!LEVELS!OF!CONCENTRATION!..........................................................!110!4.8!DOMESTIC!MARKET:!COORDINATION!LEVELS!................................................................................................!112!4.8.1$High$fragmentation$at$the$interface$between$Chinese$and$global$iron$ore$markets$.....$112!4.8.2$Process$tracing:$Iron$ore$importing$licenses$–$a$short$history$.................................................$112!4.8.3$The$fall$of$the$iron$ore$benchmarking$regime$.................................................................................$116!4.8.4$The$three$international$producers’$response$...................................................................................$121!4.9!CASE!2:!CHINA’S!IMPACT!ON!THE!GLOBAL!IRON!ORE!SHIPPING!PRICING!REGIME!.................................!127!4.9.1$Dependent$variable:$The$iron$ore$shipping$industry$pricing$regime$....................................$127!4.9.2$Independent$variables:$Concentrated$international$and$domestic$markets$.....................$129!4.9.3$The$Valemax$shipping$impasse$..............................................................................................................$130!4.10!CASE!3:!JAPAN'S!IMPACT!ON!THE!GLOBAL!IRON!ORE!MARKET!IN!THE!SECOND!HALF!OF!THE!TWENTIETH!CENTURY!................................................................................................................................................!136!4.11!INDEPENDENT!VARIABLES:!CONCENTRATION!AT!THE!GLOBAL!LEVEL!...................................................!138!4.12!INDEPENDENT!VARIABLES:!CONCENTRATION!AT!THE!DOMESTIC!LEVEL!..............................................!139!4.13!CAUSAL!STORY:!JAPAN’S!IMPACT!ON!THE!INTERNATIONAL!IRON!ORE!PRICING!REGIME!...................!141!4.14!CAUSAL!STORY:!JAPAN’S!IMPACT!ON!THE!INTERNATIONAL!SHIPPING!REGIME!....................................!143!4.15!CONCLUSION!.......................................................................................................................................................!145!CHAPTER'5'–'CHINA’S'IMPACT'ON'THE'GLOBAL'POTASH'MARKET'.......................................'147!5.1!ARGUMENT!IN!THE!POTASH!CASE!.....................................................................................................................!148!5.2!POTASSIUM:!A!VITAL!PLANT!NUTRIENT!.........................................................................................................!150!5.3!CHINA’S!POTASH!IMPORTS:!A!SYSTEMIC!WEIGHT!ON!THE!GLOBAL!POTASH!MARKET!..........................!151!5.4!DEPENDENT!VARIABLE:!THE!INTERNATIONAL!POTASH!BENCHMARK!PRICING!SYSTEM!......................!154!5.5!INDEPENDENT!VARIABLE:!GLOBAL!POTASH!MARKET!CONCENTRATION!LEVELS!...................................!156!5.6!GLOBAL!POTASH!MARKET!PLAYERS:!COORDINATION!CAPACITY!...............................................................!158!5.7!CHINA’S!DOMESTIC!POTASH!MARKET!..............................................................................................................!162!5.8!INDEPENDENT!VARIABLE:!DOMESTIC!CONCENTRATION!LEVELS!...............................................................!165!5.9!CHINESE!DOMESTIC!COORDINATION!LEVELS!IN!THE!POTASH!MARKET!...................................................!166!5.9.1$Concentration$at$the$interface$of$Chinese$and$global$potash$markets$.................................$166!5.9.2$Process$tracing:$Potash$importing$licenses$–$a$short$history$...................................................$167!5.9.3$Causal$mechanism:$Levels$of$coordination$.......................................................................................$176!5.10!THREE!EVENTS!IN!THE!GLOBAL!POTASH!MARKET!.....................................................................................!179!5.10.1$Increase$in$the$frequency$of$benchmark$pricing$..........................................................................$179!5.10.2$China’s$refusal$to$sign$a$procurement$contract$with$the$global$potash$producers$.....$180!  vii 5.10.3$TakeIover$failures$and$breakdown$of$RussiaIBelarus$cartel$(BPC)$....................................$183!5.11!LOOKING!AHEAD!–!LESSONS!FROM!THE!IRON!ORE!CASE!...........................................................................!189!5.12!CONCLUSION!.......................................................................................................................................................!191!CHAPTER'6'–'CHINA’S'IMPACT'ON'THE'GLOBAL'URANIUM'MARKET'....................................'192!6.1!ARGUMENT!IN!THE!URANIUM!CASE!..................................................................................................................!192!6.2!CHINESE!DEMAND!FOR!URANIUM!.....................................................................................................................!195!6.3!INDEPENDENT!VARIABLE:!GLOBAL!URANIUM!MARKET!CONCENTRATION!LEVELS!................................!200!6.3.1$Lack$of$interest$and$investment$in$the$industry$in$recent$decades$........................................$203!6.4!INDEPENDENT!VARIABLE:!DOMESTIC!CONCENTRATION!LEVELS!...............................................................!205!6.4.1$Overlap$between$SOE$and$government$priorities$in$the$sector$...............................................$206!6.5!CHINESE!GLOBAL!URANIUM!PROCUREMENT!PATTERNS!..............................................................................!207!6.5.1$Active$participation$in$international$initiatives$.............................................................................$210!6.6!CONCLUSION!..........................................................................................................................................................!213!CHAPTER'7'?'POLICY'RESPONSES'TO'GLOBAL'MARKET'CONCENTRATION'.........................'216!7.1!CHINA’S!ATTEMPT!TO!REDUCE!ITS!DEPENDENCE!RELATIVE!TO!THE!BIG!THREE!IRON!ORE!SUPPLIERS!.........................................................................................................................................................................................!216!7.2!ATTEMPTS!TO!INCREASE!COMPETITION!IN!THE!GLOBAL!POTASH!MARKET!.............................................!219!CONCLUSION'..............................................................................................................................................'222!REFERENCES'..............................................................................................................................................'230!APPENDIX'1:'LIST'OF'INTERVIEWEES'...............................................................................................'247!APPENDIX'2:'CODE'CO?OCCURRENCE'IN'INTERVIEW'EXCERPTS'.............................................'249!APPENDIX'3:'TOP'10'IMPORTERS'OF'POTASH,'AND'THEIR'AFFILIATION'...........................'250!'      viii LIST OF TABLES  TABLE!1!I!TWOILEVEL!MARKET!POWER!ASYMMETRIES!IN!DOMESTIC!AND!INTERNATIONAL!MARKETS!......................................................................................................................................................................!10!TABLE!2!I!CHINA’S!PROJECTED!DEPENDENCE!IN!MINERALS!UP!TO!2050!...................................................!16!TABLE!3!I!KEY!CHARACTERISTICS!OF!GLOBAL!MARKETS!.............................................................................!33!TABLE!4!–!DV:!GLOBAL!PRICING!REGIME!CHANGE!AS!A!RESULT!OF!CHINA'S!AND!JAPAN'S!EMERGENCE!......................................................................................................................................................................!37!TABLE!5:!CODE!COIOCCURRENCE!....................................................................................................................!38!TABLE!6!I!PRICING!REGIMES!(DV!INDICATORS)!............................................................................................!40!TABLE!7:!PRICING!REGIMES!(DV!INDICATORS),!OPERATIONALIZATION!....................................................!41!TABLE!8!I!DEDUCTIVE!PRICING!REGIME!PREFERENCES!(PRODUCERS!AND!CONSUMERS),!BY!PRICE!TREND!..........................................................................................................................................................!50!TABLE!9!I!LEVEL!OF!ANALYSIS!IN!THE!IPE!OF!RESOURCES!..........................................................................!63!TABLE!10!I!CASES!...............................................................................................................................................!69!TABLE!11!I!SUMMARY!OF!SIMILARITIES!ACROSS!THE!GLOBAL!POTASH!AND!IRON!ORE!MARKETS!PRIOR!TO!CHINA'S!EMERGENCE!...........................................................................................................................!70!TABLE!12!I!ASYMMETRIES!AND!CASES!............................................................................................................!72!TABLE!13!I!TWOILEVEL!ASYMMETRIES!IN!DOMESTIC!AND!INTERNATIONAL!MARKETS,!AND!ENSUING!OUTCOMES!..................................................................................................................................................!77!TABLE!14!I!IMPLICATIONS!OF!THEORETICAL!MODEL!...................................................................................!83!TABLE!15!I!GLOBAL!COMMODITY!MARKET!CONCENTRATION!BEFORE!AND!AFTER!CHINA'S!EMERGENCE!AS!DOMINANT!CONSUMER!........................................................................................................................!85!TABLE!16!I!INDEPENDENT!VARIABLE,!INTERNATIONAL!MARKETS!/!CONCENTRATION!LEVELS!...........!87!TABLE!17!I!INDEPENDENT!VARIABLE:!DOMESTIC!MARKETS!/!CONCENTRATION!LEVELS!.....................!88!TABLE!18!I!INTERNATIONAL!MARKETS!I!COORDINATION!LEVELS!.............................................................!89!TABLE!19!I!DOMESTIC!MARKETS!–!COORDINATION!LEVELS!.......................................................................!90!TABLE!20!I!DOMESTIC!MARKETS!I!CONCENTRATION!AND!COORDINATION!LEVELS!................................!91!TABLE!21!I!CODE!COIOCCURRENCE!IN!INTERVIEWS!.....................................................................................!93!TABLE!22!I!TWOILEVEL!MARKET!POWER!ASYMMETRIES!IN!THE!IRON!ORE!MARKET!...........................!99!TABLE!23!–!CHINA’S!IRON!ORE!IMPORTS!BY!COUNTRY,!WITH!PRICE!PAID!(2008)!.............................!108!TABLE!24!I!COMPARATIVE!FRAGMENTATION!LEVELS!IN!THE!CHINESE!IRON!ORE!AND!POTASH!INDUSTRIES!...............................................................................................................................................!111!TABLE!25!I!CHINESE!IRON!ORE!IMPORTS!BY!COUNTRY!AND!PRICE!PAID!(2008)!................................!118!TABLE!26!I!TWOILEVEL!MARKET!POWER!ASYMMETRIES!IN!THE!POTASH!CASE!...................................!149!TABLE!27!I!CHINESE!POTASH!IMPORTS!AND!PRICE!PAID!(2001)!...........................................................!155!TABLE!28!I!COMPARATIVE!FRAGMENTATION!OF!THE!IRON!ORE!AND!POTASH!INDUSTRIES!IN!CHINA!(2007)!......................................................................................................................................................!166!TABLE!29!I!TWOILEVEL!MARKET!POWER!ASYMMETRIES!IN!THE!URANIUM!CASE!................................!194!TABLE!30!I!REACTORS!OPERABLE,!UNDER!CONSTRUCTION,!PLANNED!AND!PROPOSED,!AND!URANIUM!REQUIREMENTS!........................................................................................................................................!195!TABLE!31!I!URANIUM!MINES!IN!OPERATION!IN!MAINLAND!CHINA!(2010)!..........................................!198!TABLE!32!I!GLOBAL!NATURAL!URANIUM!PRODUCTION!(2009!I!2011)!...............................................!199!     ix LIST OF FIGURES   FIGURE!1!I!SYSTEMIC!CHANGE!AT!THE!GLOBAL!LEVEL!AS!A!RESULT!OF!CHINA'S!EMERGENCE!AS!DOMINANT!PLAYER!......................................................................................................................................!8!FIGURE!2!I!RELATIVE!IMPACT!OF!CHINA!AND!INDIA!ON!GLOBAL!MINING!MARKETS!...............................!17!FIGURE!3!I!CHINA'S!POTASH!AND!IRON!ORE!IMPORT!DEPENDENCY!RATIOS!(2000I2011)!................!20!FIGURE!4!I!TYPOLOGY!OF!GLOBAL!MARKETS!..................................................................................................!34!FIGURE!5!I!CHINA!AS!SHARE!OF!GLOBAL!IRON!ORE!IMPORTS!......................................................................!36!FIGURE!6!I!EXTENT!AND!DIRECTION!OF!PRICING!REGIME!CHANGE!AS!A!RESULT!OF!CHINA'S!AND!JAPAN'S!EMERGENCE!.................................................................................................................................!38!FIGURE!7!I!COMMODITY!PRICES!(1980I2012)!............................................................................................!49!FIGURE!8!I!CHINESE!IMPORT!DEPENDENCE!IN!THE!MARKETS!OF!IRON!ORE,!POTASH!AND!URANIUM!..!71!FIGURE!9!I!NATIONALITY!OF!INTERVIEWEES!..................................................................................................!73!FIGURE!10!I!WORK!UNITS,!BY!INTERVIEWEE!.................................................................................................!73!FIGURE!11!–!NATURE!OF!OWNERSHIP,!BY!WORK!UNIT!................................................................................!74!FIGURE!12!I!LANGUAGE!USED,!BY!INTERVIEW!...............................................................................................!74!FIGURE!13!I!CHINESE!DOMESTIC!COMMODITY!MARKETS,!COMPARATIVE!CONCENTRATION!LEVELS!...!88!FIGURE!14!I!CHINA'S!IRON!ORE!IMPORT!DEPENDENCE!..............................................................................!103!FIGURE!15!I!IRON!ORE!EXPORTERS!WORLD!MARKET!SHARE!...................................................................!104!FIGURE!16!I!CHINESE!IRON!ORE!IMPORTS!BY!COUNTRY!(2001I2010)!.................................................!105!FIGURE!17!I!TOP!FIVE!SEABORNE!IRON!ORE!PRODUCERS!(1997!I!2007)!............................................!105!FIGURE!18!I!MAIN!STEEL!CONSUMING!SECTORS!IN!CHINA!(2011)!.........................................................!109!FIGURE!19!I!CHINESE!IRON!ORE!ANNUAL!PRODUCTION!AND!IMPORTS!(2000I2010)!.......................!110!FIGURE!20!I!DOMESTIC!FRAGMENTATION!LEVELS!I!TOP!FOUR!CHINESE!CRUDE!STEEL!PRODUCERS!(2000!I!2010)!........................................................................................................................................!111!FIGURE!21!I!SHARE!OF!TOTAL!IRON!ORE!IMPORTS!BY!TOP!2!CHINESE!FIRMS!(2001!I!2011)!.........!115!FIGURE!22!I!IRON!ORE!PRICES!BENCHMARK!AND!SPOT!PRICING!REGIMES!(1975!I!2015)!...............!124!FIGURE!23!I!DRY!BULK!VESSEL!CATEGORIES,!IRON!ORE!TRADE!WITH!CHINA!.......................................!129!FIGURE!24!I!SHARE!OF!GLOBAL!POTASH!CONSUMPTION!BY!COUNTRY!(2011)!.....................................!152!FIGURE!25!I!CHINESE!SHARE!OF!GLOBAL!IMPORTS!AS!%!OF!WORLD!EXPORTS!(1992I2013)!.........!152!FIGURE!26!I!CHINESE!POTASH!IMPORTS!AS!SHARE!OF!GLOBAL!EXPORTS!1980!I!2013!.....................!153!FIGURE!27!I!CHINA'S!POTASH!IMPORT!DEPENDENCE!(2000!I!2011)!...................................................!154!FIGURE!28!I!POTASH!PRICES!(1960!I!2014)!.............................................................................................!156!FIGURE!29!I!WORLD!SHARES!OF!POTASH!EXPORTS!(2012)!.....................................................................!157!FIGURE!30!I!CHINESE!POTASH!PRODUCTION,!CONSUMPTION!AND!IMPORTS!..........................................!163!FIGURE!31!I!CHINESE!POTASH!SUPPLY!SOURCES!.........................................................................................!163!FIGURE!32!I!POTASH!PRICES!IN!CHINA!(2011!I!2012)!............................................................................!165!FIGURE!33!–!SHARE!OF!POTASH!IMPORTS!BY!TOP!TWO!CHINESE!COMPANIES!(2001!I!2011)!.........!174!FIGURE!34!I!SHARE!OF!POTASH!IMPORTS!BY!THE!TOP!10!CHINESE!COMPANIES!(2001!–!2011)!....!175!FIGURE!35!I!URANIUM!PRODUCTION!IN!CHINA!(1998!I!2011)!..............................................................!197!FIGURE!36!I!CHINESE!URANIUM!IMPORTS!(2009!I!2013)!.......................................................................!200!FIGURE!37!I!URANIUM!PRODUCTION!BY!COUNTRY!(2011)!......................................................................!201!FIGURE!38!I!URANIUM!PRICES!(1969!I!2012)!..........................................................................................!204!FIGURE!39!I!CHINESE!URANIUM!IMPORTS!BY!COMPANY!(2011)!.............................................................!205!  x FIGURE!40!I!CHINESE!URANIUM!PROCUREMENT!BY!COUNTRY!(2006)!..................................................!208!FIGURE!41!I!CHINESE!URANIUM!IMPORTS!BY!COUNTRY!(2013)!.............................................................!208!FIGURE!42!I!CHINESE!IRON!ORE!IMPORTS!BY!COUNTRY!(2001)!.............................................................!218!FIGURE!43!I!CHINESE!IRON!ORE!IMPORTS!BY!COUNTRY!(2012)!.............................................................!218!FIGURE!44!I!CHINESE!POTASH!IMPORTS!BY!COUNTRY!(2001)!................................................................!220!FIGURE!45!I!CHINESE!POTASH!IMPORTS!BY!COUNTRY!(2011)!................................................................!221!     xi ACKNOWLEDGEMENTS  Only as I come to the end, do I realize this is another beginning.   My supervisor, Yves Tiberghien, knew before I did that I would pursue a PhD. He and Brian Job, alongside many other mentors and colleagues, are one of the main reasons I elected to pursue doctoral studies in the Department of Political Science at UBC. Yves is a great scholar, an exceptional human being, a passionate communicator and an extraordinary mentor. His passion, rigour, intelligence, ambition, joie de vivre and humanity have been an inspiration for me since our first meeting at the beginning of my MA degree in 2007. His unwavering belief in my potential has been the best incentive for me to work hard. He – and his family – have also made the experience truly enjoyable. Any measure of success I enjoy is in no small part the result of his kind support. Brian Job is humanity at its very best, a perfect mixture of wisdom, kindness, humour, sharpness of reason, prescience, and straightforwardness. He has never once given me advice that was not worth following, and is the kind of person I would like to have with me every day, for his feedback, company and editing skills. Both Yves and Brian have affected my life beyond my academic pursuits and I am a better person for having had the privilege of being mentored by them. Heartfelt thanks also go to my third committee member, Lou Pauly, for going beyond the call of duty in agreeing to be on my committee. I deeply appreciate his time, support and the wisdom of his comments. It has been a real privilege to receive his feedback, rich from his many decades as a central figure in the field. During my dissertation defense, I was blessed with exceptional feedback from University Examiners Paul Evans and Amy Hanser, as well as from External Examiner David Zweig. Their comments will be invaluable as I move ahead with turning this dissertation into a book.  I have also benefited from numerous unofficial advisors, who have provided support, feedback and encouragement above and beyond what was required of them. Chief among them are Paul Evans, Alan Jacobs, Tim Cheek, and Pitman Potter, but also Ben Nyblade, Richard Johnston, Peter Dauvergne, Katia Coleman, Alison Bailey, Jack Austin, Julian Dierkes, Abidin Kusno, and Cesi Cruz. Earlier in my PhD program, I benefitted from the mentorship and support of Max Cameron, Richard Price, Antje Ellerman, Kathy Harrison, Philippe Le Billon, Fred Cutler and Joseph Caron, as well as Yuen Pau Woo from the Asia Pacific Foundation of Canada. A special word of recognition is also due to Angel O’Mahony for her support and mentorship at a critical point in my trajectory. She played an important role in my decision to pursue doctoral studies. Richard Taylor provided the last push I needed to embark on the doctoral adventure, when, near the summit of Grouse Mountain one evening, he told me to close my eyes and follow my heart. From then on, my decision was inevitable.    One of the best aspects of my doctoral journey at UBC were my colleagues, many of whom have now become steadfast friends, all of whom have enhanced my intellectual journey: Sule Yaylaci, Elena Caprioni, Linting Zhang, Yoel Kornreich, Konrad Kalicki, Jan Luedert, Matt Gravelle, Susan Yikun Dong, Brent Sutton, Agustin Goenaga, Erin Williams, Stewart Prest, Kate Neville, Aim Sinpeng, Carla Winston, Tommi Rebien, Charlie Rogers, Leanne Smythe, Yana Gorokhovskaia, Jane Lister, Evelyn Chan, Yingqiu Kuang, Anastasia Shesterinina, Guo Li, Conrad King, Jen Allan, Grégoire Legault, Matt Galway, Beth Schwartz, Deborah Farias,   xii Chunman Zhang, Katie Meridith and Sandra Salt. Sincere thanks are due to the Gingras-Andrade family for such a warm welcome to Vancouver.    Thank you to my co-organizers Sophia Murphy, Sara Elder and Carry Wu, for a great collaborative experience putting together the interdisciplinary workshop on the role of trust in global commodity markets at UBC in 2014.   My field research in China would not have been possible without the stellar support and generosity of my host supervisor at the Chinese Academy of Social Sciences Institute for World Economics and Politics in Beijing, the Head of the World Energy Research Division, Xu Xiaojie, as well as the research team – Wan Jun, Liu Xiulian and Zhang Chunyu – for welcoming me as one of their own. My thanks also go to my host supervisor at Peking University, Wang Yong, Director of the Center of International Political Economy, for his excellent IPE seminar, his feedback and his collegiality. The support of Xu Hongcai, Deputy Director of the Information Department at the China Center for International Economic Exchanges, was also invaluable and I am indebted to him for his kind welcome and multiple introductions. Sincere thanks also go to the Research Division of China Business News for publishing my work and for being so welcoming, in particular Vice-President Xu Yisheng, Li Dongchao for his superb research and writing skills, as well as Zhang Meng and colleagues.   I thank the Department of Foreign Affairs, Trade and Development Canada for inviting me to speak at an Energy Security roundtable in Nanjing and for their welcome; Liu Shuchen at the Chinese Ministry of Land and Resources Information Center for inviting me to speak at the China Mining Congress in Tianjin; Qi Zhaoying at the Potash Branch of China Inorganic Salt Industry Association for inviting me to speak at their annual conference in Shanghai; Wang Yong at the Peking University Center of International Political Economy for inviting me to give a seminar; and Zhen Xinye from the Renmin University Department of Economics for inviting me to give a talk, all in 2012. I also thank Guan Qingyou, Research Division Chief at the China Offshore Oil Corporation; Shen Lei, Deputy Director of the Natural Resources and Environment Security Research Centre at the Chinese Academy of Sciences; Diao Xungang from Beihang University; and diplomats from the Brazilian Embassy in Beijing for their feedback and generosity, and too many others to list here. I also thank Gerard Hervouet and Louis Bélanger from Laval University, as well as Gordon Houlden from the University of Alberta. My field research in China and Canada was a transformative experience. I conducted over a hundred interviews with Chinese and international government officials, journalists, industry insiders and academics. I am eternally grateful for my interviewees’ time, openness, frankness and friendliness. I hope this work does them justice. A special mention goes to Nicholas Martin, for his collegiality and friendship during our many months of field research in China.   My participation in the Global Political Economy workshop at the Balsillie School of International Affairs was a highlight of my PhD years. There, I benefitted from the support and feedback of all participants, including Eric Helleiner, Herman Schwartz, Derek Hall and Kathryn Hochstetler, as well as Veronica Rubio Vega and Julian Germann. I also benefitted from the comments of – and forged continuing collegial bonds with – Sarah Martin and Kathleen Sexsmith. At international conferences, I benefitted from the comments and support of Eric Helleiner, Tony Porter, Thomas Rawski, Loren Brandt and Regina Abrami.   xiii  I would also like to extend my thanks for the wonderful help and support from the staff in the Institute of Asian Research – Karen Jew, Kerry Ross and Marietta Lao – and in the Department of Political Science – Josephine Calazan and Dory Urbano.       Three research assistants helped me with data gathering. Their research support was excellent and I thank them for their commitment to good work: Carry Wu, Haochen Li and Elise Sammons. My deepest gratitude also goes to Erin Williams for stellar editing work – all remaining errors are mine.   This project benefited from the financial support of the Social Sciences and Humanities Research Council of Canada; the Chiang Ching-kuo Foundation; the Institute of Asian Research, the Liu Institute of Global Issues and the Department of Political Science and the Faculty of Graduate and Postgraduate Studies, all at UBC; and the Fukien Chinese Association.   During the last year of my program, as the Cadieux-Léger Fellow at the Department of Foreign Affairs, Trade and Development Canada, I benefitted from the collegiality and support of the whole team at the Foreign Policy Research Division, the Foresight and Global Trends Analysis group, and beyond. The warmth of my colleagues there carried me through the last miles of this dissertation-writing marathon. Thank you to Valérie La Traverse, Jordan Zed, Martin Roy, Andréanne Goyette, Nicole Favreau, Marketa Geislerova, Frédéric Miville-Deschênes, Mary Halloran, Greg Donaghy, Tim Hodges, Sarah MacLeod, Jonathan Yendall, Milan Konopek and also Cindy Termorshuizen, David Hartman and Susan Gregson. Thank you also to Julia Bentley for her warm welcome. I will not live the joy of sharing this dissertation with Sumeeta Chandavarkar, who touched my soul. I imagine her reading it over my shoulder.    Lastly, I would like to thank my parents, Marie-France and Alain, for being two of the most intense, passionate and unique individuals I have ever known, for instilling in me the value of education and hard work and for always insisting on excellence, fairness and genuineness. Merci maman, for believing in me and supporting me no matter what. My maternal grandparents, Hélène and Paul, as well as my uncle Maurice, played a critical role in my desire to excel at university. Much of what I do, I do for them. The boundless love from Monic, Jo-Anne, Johanne and the rest of the Arsenault family has provided me with fuel for life.   My first journey to Asia started in 2001 on a bulk cargo ship across the Pacific and on rough seas – I can thank my father’s St-Malo origins for this appel du large. I came back with two of life’s most precious gifts: an intellectual passion for China, and, most precious of all, love and companionship. I seriously doubt whether I would have been able to accomplish any of this without David Needham, who makes me whole. Certainly, none of it would have been so colourful, nor would I have been so profoundly happy while doing it.         xiv  DEDICATION     To all in my family – and there are many –  who made the pursuit of higher education irresistible,  through encouragement and example,  while allowing me the freedom,  and the responsibility,  to own my decisions.     1 CHAPTER 1 – EMERGING POWERS AND SYSTEMIC CHANGE   How are global economic institutions transformed by dominant emerging economies at times of power transition? Why have some international markets for important raw materials undergone fundamental change in the way they operate as a result of China’s emergence, while other such markets have been more resilient to change? The goal of this dissertation is to explain diverging global outcomes from the dramatic contemporary expansion of China’s economy. By doing so, my objective is to shed new light on the political economy of global markets; why they operate the way they do now, and how they have evolved and changed over time.   The structure of key industries within China varies: some are concentrated, some fragmented, some very sensitive to price signals, and others less so. Likewise, the structures of global markets in various commodities varied significantly before the emergence of China as dominant global consumer in the twenty-first century. My central thesis supports the view that China has had a powerful effect on global markets, but suggests that the effect has varied from market to market. I trace key variances to the interaction of Chinese domestic industrial structures and the pre-existing structures of global commodity markets.   We stand at a time of transition, when emerging economies are playing an increasingly important role in the global economy. One of the areas of the global economy over which emerging powers have the most palpable impact is global commodity markets. A case in point is China, which has gone from being an almost complete outsider to the most important player in many such global markets within a very short period of time. The (re)emergence of China offers a great opportunity to investigate global market change, since rarely does a country have such a decisive impact on global markets in such a short period of time. Yet the political economy of China’s behaviour and its impact on global markets is poorly understood. There is still much space within the literature for a more nuanced assessment of the diverse impacts China is having on global markets.   Since the institutional arrangements underpinning global markets were brokered during the second half of the twentieth century, before China’s rise, and since China has quickly become the key stakeholder and actor in many of these markets, we are essentially observing a quasi-experiment. China’s emergence and impact has occurred quickly and has been dramatic in scope across a variety of distinct, yet comparable, market structures. This is   2 an ideal empirical situation for a comparative study, and a great opportunity to investigate exactly how dominant emerging states impact the fabric of the international economy and how their impacts vary across markets.  The dynamics of China’s impacts on global commodity markets force us to profoundly re-evaluate several widespread assumptions. This includes assumptions about global market institutions prior to China’s arrival (contrarily to widespread assumptions, they were not competitive in a classical sense); Chinese domestic markets (contrarily to widespread assumptions, there is great variation in concentration levels across commodity markets); China’s global market power (contrarily to widespread assumptions, despite its economic size, its position is often weak, both domestically and in terms of global economic relations of power); and the direction of institutional change that is expected as a result of China’s emergence (contrarily to widespread assumptions, China’s state-led hybrid economy does not necessarily have a mercantilistic effect on global markets). The impacts observed are substantial and do destabilize established global power relationships, but they consist at times of a “marketization impulse.”   This dissertation investigates the intensity and direction of change in comparable international markets as a result of China’s emergence as a dominant consumer over the past decade. I argue that the impact of China’s emergence as a systemically relevant player in a global market is explained by the interplay of system-level and domestic-level variables. More specifically, I argue that 1) asymmetries between domestic (Chinese) and international market structures have a determining impact on global market institutional stability, and that 2) positions of market power at the domestic level influence preferences for pricing regimes at the global level. By tracing China’s global impacts back to its domestic roots, I unveil the full picture, from Chinese domestic market variables to the international-level outcomes in global markets.   In this dissertation, I develop conceptual tools to theorize about the divergent impacts a systemically relevant country can have on market institutions at the global level. First, I develop a typology of global market institutions, and identify their level of concentration and coordination as determining features. I also develop a multi-pronged definition of marketization as a measure of institutional change in markets at the global level. Lastly, I   3 provide a comparative assessment of procurement dynamics at the interface between Chinese and global commodity markets.  This project is unique because it combines classical and comparative political economy, economic sociology and Chinese politics approaches to answer classical International Political Economy (IPE) questions, with an empirical focus on commodity markets. China’s impact on the political economy of global commodity market institutions (their pricing regimes, power relations between key actors, oligopolistic practices, transport structures) has been identified by some as an important area of research (Zweig and Bi, 2005; Garnaut and Song, 2006; Song and Li, 2009; Rosen and Houser, 2007; Helleiner and Clapp, 2012; Economy and Levi, 2014).  This research contributes to four key debates in the political science literature. First, I propose a reconceptualization of global market institutions (Fligstein, 2001, 2008; Ebner and Beck, 2008; Krasner, 1974; Rodrik, 1982). Second, I contribute to the theorization of systemic change by working within the IPE inside-out literature and incorporating interactive components (Katzenstein, 1985; Krasner, 1976; Putnam, 1988; Gilpin, 1977), as well as institutional comparative political economy insights (Hall and Soskice, 2001; Esping-Andersen, 1990; Thelen and Hall, 2009; Aoki, 2001; Streeck and Yamamura, 2001). Third, I contribute to the existing literature on “fragmented” foreign economic engagement, specifically in the context of Chinese politics (Lieberthal, 2004; Lieberthal and Lampton, 1992; Oksenberg and Lieberthal, 1988; Mertha, 2009; Shambaugh, 2011; Tiberghien, 2012; Kennedy and Cheng, 2012). To do this, I draw from a wide range of scholarship on Chinese foreign relations (e.g., status quo vs. revisionist power, rule taker vs. rule maker vs. rule shaper), but develop a nuanced and counter-intuitive model of Chinese behaviour.   Much has been written on China’s struggle to secure the supply of commodities that are key to its continued economic development. But this dissertation takes the discussion one step further, and seeks to build a systematic political economic theory of the variation in Chinese behaviour and impacts on commodity markets, and to draw generalizable conclusions.   The vast majority of the empirical material gathered for this dissertation concerns the interactions between China and global commodity markets (the exception is a comparison with Japan’s emergence as a dominant commodity consumer in Chapter 4). However, the   4 theoretical model is constructed with the aim of capturing the interactional dynamics between any dominant emerging consumer and its global suppliers. The scope of cases it applies to is very specific: large consuming emerging economies at times of power transition within a given commodity market. As such, this research reflects awareness of the main protagonist’s stage of economic development when its impacts on global markets are being evaluated. In fact, many of the dynamics observed are inextricable from the underlying conditions of economic development (Gerschenkron, 1962; List, 1909).   Significant theoretical value can be derived from considering China as a basis for comparison. It is up to future researchers to test the full extent of the model’s theoretical reach, an obvious candidate being the United States when it emerged as key consumer of global commodities.1   The study of institutional resonance and complementarity between global markets and their systemically relevant consumers points to broader patterns we can expect to see as China tries to carve out a place for itself that is commensurate with its purchasing power in the global economy. This dissertation seeks to contribute to ongoing debates about the systemic resilience of global market structures, as well as on the domestic determinants of global economic power.   1.1 Existing Approaches to the Study of Power Transition and Global Market Change  Dominant theories of China’s impact on the global economy usually fall into one of two camps. On the one hand, conflict scenarios foresee China’s rise, its thirst for resources and its state-centred economic development as leading to destabilizing, protectionist or mercantilist behaviour on the global stage, and to economic conflict (Vivoda, 2009; Economy, 2010; Calder, 1996; Kane and Serewicz, 2001; Andrews-speed et al., 2004; Bremmer, 2010). For instance, Vivoda argues that “China is… undermining the Western-dominated neoliberal capitalist system” (Vivoda, 2009). As enunciated – and then refuted – by Kenney and Cheng: “A more powerful China, the thinking goes, is likely to promote a statist and more hierarchical brand of international governance that is inconsistent with the                                                  1 See Foot and Walter for an interesting comparison between the US and Chinese behaviour on the global stage (Foot and Walter, 2010).   5 open, multilateral governance architecture that has emerged and developed since the end of World War II under American and European guidance” (Kennedy and Cheng, 2012, 9).   On the other hand, status quo scenarios predict that China will not disrupt existing international institutions (Ikenberry, 2008; Johnston, 2013; Buzan and Cox, 2013; Steinfeld, 2010; Bergsten et al., 2008). For instance, Ikenberry argues that “China [will] integrate into the liberal international order” (Ikenberry, 2008).   I argue that the two sides of the debate outlined here have three key shortcomings. First, both view global markets as uniform. In other words, both sides of the debate tend to overlook the political economy of global markets. In contrast, I make the case that global markets vary significantly in their institutional structure. A critical precondition to correctly diagnosing China’s impact on global market structures is an accurate description of the variation across market structures internationally. There is a variety of pricing regimes in commodity markets today. Some commodities, such as copper, are more financialized and are traded on spot and futures markets. Other commodities, such as uranium or potash, still use long-term negotiated benchmark pricing systems in which prices are negotiated between producers and consumers, are valid over longer periods of time, and in which only a small percentage of total trades occur on the spot market.  Second, both sides of this debate characterize global markets as liberal. This is explained in part by the powerful liberal tradition, which highlights the continuing presence of post-hegemonic yet dominant US economic power, combined with path-dependent global liberal economic institutions that remain too entrenched for any challenger to disrupt (Ruggie, 1982; Ikenberry, 2008). I make the case that characterizing current global markets as liberal is an oversimplification. Indeed, extractive markets in particular are very oligopolistic and leave more room for strategic coordination than is usually acknowledged (Nolan, 2012; Fligstein, 1996; Foster and McChesney, 2012). If all markets are not liberal, it follows that China’s emergence can actually lead to the liberalization of market institutions, a possibility that is overlooked by the debate in its most common incarnations. This dissertation in fact turns Ikenberry’s liberal argument on its head. Ikenberry predicts a co-optation of emerging economies into existing liberal global economic institutions (Ikenberry, 2008); I argue that an emerging economy has led to the liberalization of global economic institutions.    6  Third, both arguments expect China’s impact on the functioning of the global economy to be uniform across markets: either statist or nil. In contrast, I will show that China’s impacts vary significantly across markets. On the liberal side of the debate, arguments put the agency firmly in the hands of the Chinese state and business leaders as making a conscious rational policy choice to integrate into existing global institutions (Steinfeld, 2010; Bergsten et al., 2008; Johnston, 2003; Johnston, 2013). Others in the liberal tradition predict that China will integrate into global liberal market institutions because it has a weak capacity for shaping global institutions or effecting global change. For instance, Nolan argues that China will not be able to have much impact on the global economy, as global firms and institutions are too powerful. “We are already inside them, they are not inside us” (Nolan, 2012). Both of these variants make the case that China’s effect (or lack thereof) will be similar across markets. The modern liberal argument, derived from the theory of economic interdependence, predicts that China’s dependence on the global economy will lead to its integration into existing market structures and stability in global economic regimes (Keohane and Nye, 1977).  On the other side of the debate, Hegemonic Stability Theory (HST) scholars argue “that large dominant states possess strong preferences for free and open international exchange, and, in turn, coerce, induce or persuade other states into opening their markets to foreign trade and investment” (Krasner, 1976; Gilpin, 1987, in; Lake, 2009, 224, emphasis added). Although HST allows for a diagnosis of current global market institutions as illiberal in theory, which is worthy of note, recent realist endeavours do not tend to explore this possibility as a basis for argumentation. HST also posits that dominant emerging economies will tend to prefer open and free exchanges. Therefore, the argument does not allow for wide variation in the emerging power’s international preferences and behaviour, depending on the global context. My cases show that the diverging preferences and market power of Chinese domestic interest groups need to be taken in to account. Various market actors do not prefer the same global outcomes, a reality made clear by the Open Economy Politics framework (Bates, 1997; Rogowski, 1989), and this highlights the limitations of unitary state approaches in answering the questions motivating this study.    In sum, I argue that the likelihood of change in global market institutions, and the direction of change as a result of China’s rise, cannot be predicted based only on assumptions   7 that a) global market institutions are both uniform and b) liberal, and that c) Chinese market players have the same preferences across the board. By taking into account the variation in global and Chinese market institutions and their relative asymmetry, we can arrive at a better understanding and more satisfying explanations of outcomes at the global level. The question should not be whether China will cause institutional change at the global level, but when, how and why it does have an impact on global market institutions.  1.2 Research Questions This dissertation contributes to the current scholarship on the impact China is having on the global economy. As such, it inserts itself within classical political economic debates about power transition and the preferences and impacts of a dominant emerging economy on global markets. I use as my starting point the raison d’être of classical International Political Economy, which is to investigate “the reciprocal and dynamic interaction in international relations of the pursuit of wealth and the pursuit of power” (Gilpin, 1975: 43) in (Cohen, 2007).   This dissertation seeks to answer two research questions. The first has to do with the likelihood of change in global market institutions as a result of China’s emergence. Under what conditions will a large emerging economy cause disruption in global market institutions? Why have some global markets undergone systemic change following China’s emergence as dominant player, whereas other markets, under similar circumstances, have been more resilient to change?   The second question has to do with the direction of change in global market institutions as a result of China’s emergence as the dominant consumer. Indeed, some markets have undergone change in the direction of marketization, whereas others have undergone change in the opposite direction. Under what conditions will dominant market players push for change in one direction rather than another? What explains preferences for market coordination, as opposed to strategic coordination, in global market institutions? The figure below illustrates the context in which the relationships of interest in this dissertation take place, as well as the direction of change I am trying to explain.   8 Figure 1 - Systemic Change at the Global Level as a Result of China's Emergence as Dominant Player    As a result of the empirical research conducted for this dissertation, additional empirical puzzles have arisen. For instance, how is it that China, a state-led hybrid market economy, has led to the marketization of some global commodity markets? And why is it that we observe the most dramatic change in global market institutions in the cases in which key Chinese stakeholders were the weakest? This dissertation addresses these questions in turn. 1.3 Argument In the case of a power transition from one dominant consumer to another in a given market, international market structures are vulnerable to systemic change. I argue that market power asymmetries between Chinese and global market players have a determining impact on global market institutional stability. This is because changes in specific patterns of two-level encounters between the emerging country’s stakeholders and the global markets’ stakeholders recast relations of power, and incentives. Only by understanding relative positions of power within markets and internationally can we explain variation in behaviour and outcomes. International markets Level of concentration (T1) Domestic (Chinese) markets Level of concentration (T1) International impacts (T2)  Domestic impacts (T2) Chinese and international stakeholders’ procurement behaviour China’s emergence as dominant consumer   9 To explain the likelihood of change at the global level, I argue that power transitions that entail a profound change in consumer-producer power relations are more likely to be disruptive than power transitions that entail a continuation of two-level power relations. In addition, a change towards asymmetric positions of power tends to be more unstable than a change towards symmetric positions of power. An example of change that resulted from China’s emergence as the dominant consumer is the marketization (liberalization and financialization)2 of global commodity market pricing regimes, despite China’s own statist proclivities. Yet, we also see change in the other direction. Why? Price trends influence market stakeholders’ preferences for global pricing regimes, but they cannot account for all the variation observed. Market power – including the capacity to coordinate other market stakeholders’ behaviour, and the capacity to extract rent – also motivates behaviour.  To explain the direction of change at the global level, I argue that market players’ preferences for pricing regimes are influenced by the relative position of market power they occupy in their own market as well as relative to international market players. Asymmetric positions of power provide strategic advantages, since the dominant side’s preferences for market institutions at the global level are more easily expressed.  A domestic position of power occupied by key resource importers with rent-extracting privileges provides these importers with disincentives to push for the marketization of pricing regimes, regardless of price levels. Domestic market stakeholders’ preference for the marketization of pricing regimes is more likely to occur when they are occupying a weak position of market power relative to suppliers, as well as relative to their own peers. On the other hand, international market stakeholders’ preference for the marketization of pricing regimes is most likely to occur when they are occupying a dominant position of market power relative to the consumers, as well as relative to their own peers. At the domestic level, the opposite is true.   Going beyond simple inside-out and outside-in approaches, and leveraging the comparative method and in-depth case studies to shed light on key international political                                                  2 See p. 39 for a definition.   10 economy questions, I develop a two-level model to explain diverging outcomes at the global level.   For the purpose of this dissertation, I differentiate between global and domestic market structures according to their respective levels of market concentration. Structural power asymmetries set the conditions for systemic change. Fundamental market conditions at the domestic or international level – the fragmented or concentrated structures of markets – are the result of path-dependent historical set-ups. But these structural characteristics merely create the conditions for strategic behaviour. The key causal mechanism at play is the relative capacity for market coordination of commodity market stakeholders at both levels. It is really the strategic behaviour by key global and Chinese stakeholders (private firms, State-Owned Enterprises (SOEs), relevant state organs, industry associations, etc.) that determines whether attempts to coordinate market stakeholders’ behaviour succeed or not.  Table 1 - Two-Level Market Power Asymmetries in Domestic and International Markets International market (Producers) ! Domestic market (Consumers) " Fragmentation Concentration Fragmentation 1 Symmetry  Stable  2 Asymmetry in favour of producers Unstable Case: China/iron ore pricing Concentration 3 Asymmetry in favour of consumers Unstable Case: China/uranium market, Japan/iron ore pricing and shipping  4 Symmetry Stable Cases: China/iron ore shipping, China/potash pricing  Following Bates, “political institutions create, or deny, strategic opportunities. Within the game they define, they help to determine the power of particular interests. They do so by rendering them, or preventing them from becoming, politically pivotal” (Bates, 1997, 164).    11  Table 1 illustrates the schematic universe of possible two-level combinations of market power relations. When a power transition occurs from one dominant consumer to another, two-level power relations are either maintained from one period to the next, or they are transformed.   Transitions towards different two-level relations of market power (different quadrants) are more likely to bring instability to global pricing regimes. Transitions to asymmetric positions of power (Quadrants 2 and 3) are also more likely to bring instability than they are transitions towards symmetric positions of power. For instance, cases in which a fragmented Chinese domestic market emerges to face a concentrated and coordinated international market (Quadrant 2) create instability and open the way for systemic change at the global level. This also creates an opportunity for the expression of international suppliers’ preferences. In the opposite scenario, in cases in which concentrated Chinese domestic market actors emerge to face weakly concentrated international market actors (Quadrant 3), change remains highly likely. This time around, however, it creates space for Chinese stakeholders’ agency and expression of their preferences.   In cases of a change towards symmetry between market power structures in international and domestic realms, change is of course still possible (Quadrant 4). However, this change is less likely to be sudden, and more likely to be incremental, as the result of an alignment of preferences, or of strategic bargaining. Yet, as this combination involves equally powerful positions of power in both market realms, it is also the least predictable, as there is the possibility of deadlock or conflict between dominant players at the international and domestic levels.   The dynamics through which this gets expressed on the ground can be unexpected. Let us look at empirical narratives. China’s emergence has led to unequal systemic change in pricing regimes in the iron ore and potash markets.  In the iron ore market, China played a key part in the collapse of the benchmark pricing regime in 2010 and the subsequent move towards spot-market pricing. In contrast to the concentrated and coordinated nature of the Japanese iron ore importers prior to China’s emergence, the Chinese iron ore industry remains deeply fragmented. Thus, the transition from Japanese to Chinese dominance in the global iron ore import market saw a significant   12 change in two-level relations of market power between suppliers and consumers (Quadrant 3 to Quadrant 2).  At key transition moments between the Japanese and Chinese tenure as the world’s dominant consumer, Chinese negotiators were unable to coordinate iron ore procurement behaviour. This failure of collective action, coupled with the sheer volumes of Chinese imports, allowed small Chinese iron ore importers without a license to have a profound and destabilizing impact on global market institutions. After only a few repeated interactions, the global mining firms started breaking rank one by one and selling to individual Chinese importers on what became the quarterly, then spot, market. In fact, just six years after China replaced Japan as the world’s number one iron ore importer, the decades-long coordinated benchmarking pricing regime that had defined the global iron ore market had fallen apart. The fragmented behaviour of Chinese importers disrupted global benchmarking negotiations and provided the three main suppliers with a window of opportunity to usher in a more volatile pricing regime, against the will of larger Chinese iron ore importers and state negotiators.   In the potash market, China’s emergence has led to an increase in the frequency of benchmarking negotiations, the emergence of new producers, and the fall of a suppliers’ cartel; however, the benchmark pricing system itself has survived. In that case, we see successful coordination of the Chinese domestic interface with global market stakeholders and the continued tight management of procurement behaviour by Chinese firms. This has afforded the Chinese importers more room for manoeuvre in international benchmark negotiations, whereas the international behaviour of global suppliers has been unequal, despite the position of strength they started with in the early 2000s. In that case, we see more effective translation of Chinese procurement objectives into market outcomes.   China’s emergence has also had an unequal effect on the direction of change in pricing regimes in the iron ore and potash markets. As indicated above, the relative position of market power of given firms in their own markets explains preferences for global pricing regimes and how they differ from expected “price-determined preferences.” In the Chinese iron ore case, the presence of powerful rent-extracting privileges in China’s domestic market explains the reluctance of large steel firms to push for the marketization of the global pricing   13 regime in the 2000s. In 2009, large Chinese iron ore importers continued to express their preference for the iron ore benchmarking regime despite falling iron ore prices.   In the potash case, because of the tightly controlled licensing system, the top Chinese importers maintain a privileged position of market power by holding exclusive rights to the import and reselling of internationally acquired potash to domestic distributors. That position of market power explains at least in part why the license-holding potash importers have supported the global benchmarking regime despite price fluctuations (i.e., drop in prices) in their favour.  In sum, within- and across-market power asymmetries between domestic and global market structures profoundly affect market players’ incentive structures and preferences for pricing regimes, as well as create the conditions for systemic change. A deep understanding of the political economy of Chinese domestic commodity markets is key to understanding global market outcomes. Chinese domestic market dynamics have now become a determining feature of the global economy. 1.4 Why Extractive Commodity Markets?  Global extractive commodity markets are one of the areas of the global economy in which the impact of rising powers has been the most palpable.3 Global commodity markets are among the first in the global production chain to react to global changes in demand for finished products, and are critical for economic development. They are therefore a good candidate for examining an emerging economy’s impact on global markets. “Movements in commodity prices are a key determinant of the performance of the world economy. They affect the level and stability of export incomes earned by developing countries, the cost of inputs to production in industrial countries, the allocation (sectoral and spatial) of world capital flows, and in particular rates of national economic growth” (Cashin et al., 2000). Extractive commodities must be extracted from nature, rather than produced (Le Billon, 2012), and their geographical concentrations cannot be ignored without large amounts of                                                  3 Extractive commodity markets are defined here as markets for primary commodities or materials in their raw or unprocessed state. Their status as commodities is partly defined by the fact that they are, in their refined form, indistinguishable from one another (i.e., copper is copper, and it is measured by concentration). This gets more complicated, however, in many extractive commodity markets, as different moisture content, grades, or the overall quality of the resource play a role in differentiating one commodity’s source from another. However, it remains that the boundaries between commodity markets are easily identifiable and distinguishable from one another. Furthermore, it is difficult, if not impossible, to find a substitute for most commodities.   14 unproductive energy being spent. Global extractive commodity markets are also unique in that they have largely remained outside of global regulatory structures.   All major emerging economies have seen their demand for commodities increase dramatically in the past decade. “This is a topic of great importance, given that about 25 percent of world merchandise trade consists of primary commodities, and both long-term trends and short-term fluctuations in primary commodity prices are key determinants of developments in the world economy” (Cashin et al., 2000).  Commodity markets thus offer a unique angle for analysis, as they are essential to economic activity and national security.4 Their arrangements are also unequal and varied. Certain global commodity markets have evolved coordinated and strategic market institutions – cartels, in some cases, to regulate production and sales – whereas others have evolved market institutions based on global commodity exchanges.   Commodity markets are a promising area of research for another reason: whereas the current international political economy literature theorizes about international markets at length, it tends to focus on the global trade, monetary and financial architectures. A comparable body of literature cannot be found for physical commodities in the primary sector of the economy. Extractive commodity markets have undergone systemic institutional change in the past decade, yet the IPE/CPE literatures have not yet caught up with this intense empirical activity. The only exception to this is the IPE literature on the oil market, which has continued to produce scholarship, especially on the role of the Organization of Petroleum Exporting Countries (OPEC) and broader questions about energy security. In the field of the IPE of resources, more generally, there is a literature on the resource curse, and on the political economy of renewables, but much less on extractive commodity markets.   This dearth of analysis in the IPE literature on extractive markets dates from the 1980s, when commodity prices stabilized for two decades. Yet the field of IPE itself was born in part as a result of the inescapable relevance of politics in the study of global commodity markets following the oil crisis in the early 1970s. Indeed, the core question for much of the scholarship of that era was whether competition for resources would lead to war.                                                  4 The section above refers to the author’s previously published work: 2013. The Question of Commodity Price Volatility at the G20 St. Petersburg Summit, The Global Summitry Project. Munk School of Global Affairs, University of Toronto, http://www.globalsummitryproject.com/content/question-commodity-price-volatility- g20-st-petersburg- summit   15 The IPE of resources highlighted the need to reintroduce politics into the study of international economic patterns, and this unleashed a dynamic period of scholarship (Krasner, 1974; Rodrik, 1982; Olson, 1963; Bergsten, 1974; Huntington and Manshel, 1974; Moran, 1973; Zacher, 1987). This early IPE literature was also preoccupied with the potential for developing countries rich in natural resources to create powerful producers’ cartels. It was also aware of the diversity of existing natural resources market arrangements and included a debate about the normative value of cartelized markets (for the sake of stability).5 The end of the Cold War and the rise of the global liberal order established that the first preoccupation was overstated, but led to the neglect of the two other topics (the study of global market diversity and the normative value of market arrangements other than the classical liberal arrangement), with few exceptions (Bates, 1997).    China’s emergence has revived the need for an IPE of resources. The context is different – cartels are less numerous – but the role of power is still relevant, even with the disappearance of most producers’ cartels. And China’s interaction with global commodity markets highlights the remaining variation across markets. The IPE of resources has been identified in recent review articles as a research area in crucial need of revitalization (Helleiner and Clapp, 2012; Hancock and Vivoda, 2014). For recent endeavours, see (Petkova, 2006; Hughes, 2014; Richardson, 2009; Peine, 2013; Dauvergne and Neville, 2009; Moran, 2010; Economy and Levi, 2014). This dissertation thus revives the international political economy of resources scholarship that today has returned to relevance. 1.5 Why China?  China is a large and systemically significant economy. In terms of its economic weight in the global economy, China’s relevance is inescapable. But before the creation of the Asian Infrastructure Investment Bank in 2014, it was not easy to identify instances in which China was having a determining impact on global economic institutions. Yet, China has had an impact on the global market institutions of various commodities. Understanding how and why is key to understanding the dynamics behind one of the major drivers of change in the global economy for the foreseeable future.                                                   5 I thank Louis W. Pauly for his useful comments on this section.   16  We have witnessed a profound transformation in the balance of power of the global political economy in the past decade: the emergence of China from a complete outsider to the single most important economic actor in most global commodity markets within the span of the past 15 years. “Up until the late 1980s, China’s energy woes were essentially a domestic issue, in light of the country’s self-sufficiency with coal, oil and hydropower. But since the beginning of the reform and opening up era in 1978, China’s demand for natural resources and for imported energy has risen at an increasing pace, culminating in the late 1990s and early 2000s and turning China’s energy consumption patterns into a global issue” (Meidan, M., Ed. (2007). Table 2 - China’s projected dependence in minerals up to 2050  Dependence Number Main commodities Serious shortage 5 Chromite, Cobalt, Platinum, Potash, Diamond Unable to fulfill domestic needs 21 Iron ore, Manganese, Copper, Lead, Zinc, Thorium, Tin, Gold, Silver, Strontium, Fluorite, Boron, Barite, Oil, Uranium, Nickel, Antimony, Refractory Material, Sulphur, Cement Limestone, Kaolin Almost able to fulfill domestic needs 10 Coal, Titanium, Tungsten, Molybdenum, Phosphate, Siliceous Rock, Stone, Gypsum, Diatomite Able to fulfill domestic needs 9 Natural gas, Rare Earths, Magnesite, Sodium Salt, Mirabilite, Bentonite, Graphite, Talc, Refractory Silica Source: China up to 2050, Chinese Academy of Sciences, 2009. Author’s translation.  By 2005, China ranked first in the world in its consumption of all main metals (it accounted for between 15% and 33% of global consumption of aluminum, zinc, copper, iron ore, tin, nickel and lead), and it has since reached even higher levels. China is the world’s largest exporter of rare earths and aluminum and the largest importer of iron ore, potash and copper. In 2011, it became the largest energy consumer in the world, at 20.3% of global energy consumption. China’s average consumption growth per annum between 1990 and 2005 ranged between 16% and 24% for all main metals (Streifel, 2006). “Demand in the rich world may be dropping, but (…) steel consumption will not peak in China until 2026” (The lore of ore, 2012).   17  The Chinese Academy of Sciences has identified that the country will fall “seriously short,” or be “unable to fulfill domestic needs,” in 26 commodities, at least until the year 2050 (see Table 2). Potash and iron ore fall into the first and second of these two categories, respectively.  What about other emerging economies? The fact is that China has emerged as a dominant actor in the world’s commodity markets in a way that has almost eclipsed the significance of other emerging economies until now. For instance, between 1999 and 2005, China accounted for nearly two-thirds of the growth in global demand of refined metals consumption, compared with less than 10% for India (Figure 2) (Streifel 2006).  Figure 2 - Relative Impact of China and India on Global Mining Markets  Source: Development Prospects Group, World Bank, reproduced from Streifel, 2006, p. 19   This dissertation does not contend that China is the only variable of interest in studying commodity-market change over the past decade, but that its impact on commodity markets is large enough to be systemic. This means that changes in global commodity markets during this period are likely to have something to do with China’s rapid emergence as a dominant consumer (a process which can be traced).  1.6 Methodology “For social scientists who enjoy comparisons, happiness is finding a force or event which affects a number of societies at the same time. Like test-tube solutions that respond differently to the same agent, these societies reveal their 0!2000!4000!6000!8000!10000!12000!14000!16000!1990I1999! 1999I2005!Thousand!tons!Growth'of'ReRined'Metals'Consumption'1990?2005'Other!India!China!  18 characters in divergent responses to the same stimulus.” (Gourevitch, 1977, p.281)    This dissertation adopts a two-level (domestic-international) institutional political economy approach to unveil the conditions under which change in global market institutions is most likely, that is, in the context of a power transition.    My approach thus broadly builds on Peter Katzenstein’s early efforts to combine domestic and international levels of analysis (Katzenstein, 1978, 1985). As Cohen describes, “Waltz’s second image had to be synthesized with his third image for the picture to be complete” (Cohen, 2008, 126). Waltz (1959) outlined three categories of explanations for the causes of war. The first image referred to individual leaders, the second image to the role of domestic politics and the third image to systemic explanations. My project seeks to leverage both the second and third images by incorporating the role of domestic and system-level structures on global market outcomes. Cohen continues in saying that the frontier of research in IPE remains at the level of interaction between international and domestic variables.   The Open-Economy Politics (OEP) framework is also relevant to this approach (Bates, 1997; Lake, 2009), albeit with two qualifications. First, my approach is closer to classical political economy understandings of the role of power in markets (Gilpin, 1977; Hirschman, 1945; Viner, 1948), than are OEP’s assumptions of “market as arena” (see discussion in Chapter 2). Second, Cohen argues that over the years, OEP scholarship has tended to “drift toward mainly midlevel theory” and “focus on more and more narrowly defined relationships and variables” (Cohen, 2008, 127). Lake has argued that OEP should consider “relaxing the small country (…) assumption of traditional economic theory” (Lake, 2009). Similarly, Keohane has argued that OEP “fails to pay sufficient attention, or, worse, is blinded to big, macro-level changes in the international economy” (Keohane, 2009). This dissertation, taking note of these criticisms, aims to go beyond micro-level theorizing and small-state assumptions, and rejoin earlier classical political economy’s macro-level questions, attention to systemic change, and focus on major, systemically relevant, economies.   This research employs a mixed-methods approach (process-tracing with an emphasis on field research and interviews, document analysis and comparative/descriptive statistical analysis) to carry out a comparative study of Chinese and international commodity market   19 institutions and Chinese procurement policies in the markets of iron ore, potash and uranium. My case selection follows a small-n, comparative (method of difference) approach, and focuses on two key markets – iron ore and potash – which share important institutional characteristics, yet towards which China has developed distinct procurement policies. A shadow case – uranium – was selected to ensure a wider variation on the independent and dependent variables, as it does not share global market characteristics with the two others. A plausibility probe is conducted with the emergence of Japan as the number one iron ore consumer in the late 1960s.   I combine a two-level political economy approach with the power of small-n, comparative case studies, and use process tracing to unveil the causal pathways that lead to systemic market changes at the global level. In brief, I leverage two-level institutional political economic theorizing, the comparative method, and in-depth case studies and process tracing to resolve puzzles situated at the systemic level.  1.6.1 Case selection  In order to ensure that in the cases presented here, China is indeed the primary driver of institutional change over the past 15 years, I selected my cases among a universe of commodity markets where China has quickly become the dominant consumer and importer (and thus is having a critical and undeniable systemic impact on the relevant global markets). The two markets that form the main comparison for the purpose of this dissertation are the potash and iron ore markets.   The potash and iron ore markets exhibit strong comparative potential. They were selected according to Mill’s method of difference, as the two markets shared striking similarities, yet were transformed differently by China’s rise as the dominant importer of both resources. First, both markets are bulk commodities, and their extraction is a capital-intensive process. Second, the majority of their global supply is concentrated geographically in two or three countries: Brazil and Australia in the case of iron ore, and Russia, Belarus and Canada in the case of potash (one BRICS economy and one Asia-Pacific economy in each case). Third, both commodities are vital, one for industry and the other for agriculture, and both have no substitutes. What is more, China has a similar level of import dependence on the two commodities, hovering above 50% for both iron ore and potash. Finally, before   20 China’s emergence as dominant consumer, both markets were controlled by strong oligopolies, and were home to annual price benchmarking systems.      Figure 3 - China's Potash and Iron Ore Import Dependency Ratios (2000-2011)  Source: Ministry of Land and Resources, 2012  Despite the striking similarities between global iron ore and potash markets, China’s impact on these markets has varied significantly.   An additional benefit of focusing on the global potash market is that it is a key global fertilizer market that is absolutely essential for food security, yet it remains understudied. This dissertation may in fact be the first in-depth English-language political economic analysis of China’s potash procurement dynamics and associated global market institutions. 1.6.2 Data Much of the primary data for this dissertation was gathered during three stints of 2-4 months in China (for a total of over eight months), part of which was spent as a visiting scholar at the Chinese Academy of Social Sciences (World Institute of Economics and Politics) in Beijing, and part as a Senior Visiting Student ( ) at Peking University’s Center for International Political Economy.  My field research in China included interviews in Beijing, Shanghai, Nanning (Guangxi), Tianjin, and Nanjing. I also conducted interviews in Canada, at industry conferences, by phone and by email. The data gathered includes 127 interviews with Chinese and foreign officials (diplomats, bureaucrats, managers), industry practitioners, private-sector 0!10!20!30!40!50!60!70!80!90!100!%'China's'import'dependency'ratio'Potash'Iron'ore'  21 employees, managers and executives, industry analysts, specialized journalists and other experts. Interviews were always conducted in the interviewee’s preferred language, whether Mandarin Chinese, English or French. Data gathered also includes statistical data from various sources, including the Chinese Customs Department, Chinese and international industry associations, industry conference presentations by government officials, state-owned enterprise representatives, private industry insiders, Chinese statistical yearbooks, consultant firms and media sources. In addition, it includes a review of relevant official Chinese policy documents, notices, regulations, press releases and speeches, as well as a review of the Chinese-language academic literature on the respective Chinese resource procurement policies in each market. Finally, the data also includes Chinese- and English-language business literatures on the subject (corporate reports, strategic policy documents, press releases). This dissertation includes a review not only of novel raw interview data, but also of official Chinese policy documents as well as Chinese media, industry and academic literatures that for the most part have been scarcely referenced, if at all, in previous English-language publications.   1.7 Dissertation Outline  In Chapter 2, I look to push the frontier of conceptualization and analysis of global markets. To evaluate change at the global level, I develop a series of characteristics to define global market structures and a typology of global markets. The chapter also provides a definition of my dependent variable – pricing regimes – for which I develop a series of indicators. Finally, the chapter concludes with a discussion of the concept of marketization. Chapter 3 provides a review of my theoretical model and argument, as well as a discussion of possible alternative answers from the political economy literature on systemic change in the global economy. It includes a discussion of how the structural market characteristics at both levels have shaped coordination capacity and market behaviour of key stakeholders in both the Chinese domestic and global markets. It also provides a definition and measure of the market characteristics I select as explanatory variables. Then, I present variation over time across seven global commodity markets along some of the same indicators. This allows me to comment more broadly on the variation across different global commodity market structures. I also provide an analytical overview of the variation across my two main cases – the iron ore and potash markets – and a more in-depth justification of my case selection. Chapter 4   22 provides a discussion of the iron ore market, with an analysis of three within-market cases: the global iron ore pricing regime, the global iron ore shipping regime, and comparison with Japan’s impact on the global iron ore pricing regime in the 1960s. Chapter 5 discusses China’s impact on the global potash pricing regime, with a review of specific within-case events, including the fall of the Belarus-Russia cartel in 2013. Chapter 6 discusses the applicability of my theoretical framework to a third global market: the global uranium market. Chapter 7 provides a discussion of broader trends in China’s procurement dynamics overseas, beyond its impact on actual pricing regimes. Chapter 8 provides a discussion of conclusions and implications of this research for the study of China’s impact on the global economy.       23 CHAPTER 2 – GLOBAL MARKETS This chapter offers an in-depth discussion of the concepts of market and marketization. I first discuss the current state of the political science literature on markets. Second, I present a few foundational assumptions behind my definition of global markets, and identify a list of core characteristics of global markets. Third, I introduce a novel typology of global market structures, one that allows us to think about systemic changes in global markets in different ways. Fourth, I specify my dependent variable, marketization, define it in relation to other related terms, such as globalization, liberalization and financialization, and then specify observable implications derived from the global market typology I developed. Finally, the questions of whether marketization trends are natural, what influences preferences for pricing regimes, and whether we should expect such impact from China’s emergence in the global economy are discussed.    Markets are as central to political economy as states are to politics. They are ubiquitous in daily life, in policy-making, in domestic politics and in international relations. Yet, whereas defining the concept of “the state” has occupied much space in the literature (Levi, 2002; Evans et al., 1985; Scott, 1999; Spruyt, 1996), there has not been a similar collective effort devoted to defining what markets are.   The effort that comes closest can be found in the comparative political economy literature, in the Varieties of Capitalism (VOC) research program, which argues that there is systemic variation across national capitalist economies. This literature is not devoted to the comparative study of markets outside of their national polities, however.   The strength of the VOC literature lies in the fact that scholars from a wide variety of epistemological and methodological perspectives have tackled the same fundamental question: why are there persisting divergences in national systems of political economy? Why do these divergences persist despite the fact that states are being subjected to the forces of economic globalization? Marxist, rational institutionalist, historical or party politics approaches provide a variety of arguments, all of which build on an understanding that national systems of political economy have distinct complementary features (Hall and Soskice, 2001; Hall and Gingerich, 2004; Katzenstein, 1985; Zysman, 1983; Albert, 1991; Esping-Andersen, 1990; Garrett, 1998). “Complementarities is what makes taxonomies of capitalisms possible. Different configurations are associated with distinct comparative   24 institutional advantages for particular kinds of innovation, production strategies, or distributional outcomes” (Jackson and Deeg, 2011, p. 683).   The VOC literature provides us with ideal-type categories of domestic systems of political economy, whether categorized as Liberal and Coordinated Market Economies (Hall and Soskice, 2001), or Liberal, Corporatist and Social-Democratic Welfare Capitalisms (Esping-Anderson, 1990), among other typologies. We can use these categories to compare domestic types of political-economic systems, and theorize about their trajectories. However, the VOC literature’s typologies and tools are rooted in domestic politics and are not suited to investigating similarities and differences in transnational market structures.   Similarly, there is much valuable work on multinational corporations in IPE, which acknowledges variation; however, it also rests on the idea of structurally differentiated home markets,6 the key insight being that “different home environments produce different sorts of multinational corporations” (Lall, Sanjaya, 1983, p. 267-268, in Doremus et al., 1999, p.13).   In the international and comparative political economy literatures, there is a tendency to consider national markets as subjected to national political dynamics and institutions – and thus prone to divergence – and international markets as being subjected to international institutions, dynamics and regulatory structures – and thus less likely to diverge from one another. Indeed, much of the IPE literature since the 1980s has moved away from attempts to define distinct types of market structures, and overwhelmingly adopted the definition of “markets as arena.” An rare example of a formal definition of the “market as arena” is given by Gilpin when, quoting Cournot, he defines a market as: “the whole of any region in which buyers and sellers are in such free intercourse with one another that the prices of the same goods tend to equality easily and quickly” (Gilpin, 1987, p. 18).   More often than not, the definition of markets as “arenas” where economic exchange occurs is implicit in scholarly work. For example, the top five articles that result from a simple search (by relevancy) for “market” in the journal International Organization are as follows (as of 2014): “Food markets and their regulation,” Gary Steevers, 1978; “World oil marketing in transition,” Brian Levy, 1982; “Central American Common Market,” 1962, “The European Common Market Proposal,” Raymond Bertrand, 1956; and “World politics                                                  6 I thank Louis W. Pauly for bringing this to my attention.   25 and international energy markets,” Ernest Wilson, 1987. None of these articles make a point of defining the concept of market itself, which simply highlights that there is no tradition in the political science or IPE literatures of discussing and debating the nature and characteristics of global markets. By and large, the consideration of market structures themselves has gradually receded into the background, giving way to the study of the agency or ideas of individuals, companies or states that operate “within” them.  As an area of inquiry so devoted to the study of cooperation and conflict, why is it that we have not applied more of the tools of political science to the study of international market structures? As Gerring argues, “concepts are the building blocks of all inferences, and the formation of many concepts is clearly, and legitimately, theory-driven” (Gerring, 1999, p.364). Thus, it may be that the concept of market is underspecified in the political science literature because scholarship tends to ask questions that are set “in” or work “through” markets, but does not as a rule inquire about market structures as dependent or independent variables in themselves.   This trend is partly the result of international political developments in the last quarter of the twentieth century, including the rise of neoclassical economics, neoliberalism (in economic policy making as well as in international relations), and a gradual movement towards delegating the study of the functioning of markets per se to economists, or to political scientists who adopt econometric tools of analysis and a “market as arena” definition. Underlying these trends is the widespread assumption that markets are not a political economic structure in and of themselves, but rather simply a locus where “market forces” operate, or where the automatic coordination of individuals' utility-maximization impulses occur.  One body of literature in political science that devoted quite a bit of energy to global markets as differentiated institutions is the early IPE literature on resource markets (see discussion in Chapter 1). But as explained above, this body of literature has faded away as many producers’ cartels disappeared over the course of the second half of the twentieth century, as if the relevance of this inquiry was predicated on the survival of one particular form of market structure. Since then, there has been a gradual departure from inquiries into variation of global market structures. Susan Strange discussed cartels and the relationship between power and markets in the international political economy, and this dissertation is   26 very much in tune with her analysis. Even then, her 1996 analysis argued that “power over outcomes is exercised impersonally by markets and often unintentionally by those who buy and sell and deal in markets” (Strange, 1996, p. 149). As Van Schothorst argues, “after challenging the established definitions of power and state and broadening the scope of examination for these concepts, she fails to take the extra step to define the market” (Van Schothorst, 2005).   On the other hand, many in the economics literature have pointed to dynamics that complexify, bound or obfuscate neoclassical assumptions, among whom are scholars of the New Institutional Economics literature (North, 1990; Balleisen and Moss, 2010; Coase, 1988; Rodrik et al., 2004) and the philosophy of economics literatures (Schabas, 2005, Macloed, 2006). In some ways, and ironically, part of the economics literature is more sophisticated in its acknowledgement of the need for a more complicated view of markets, whereas “noneconomists have been fixated on a stylized reading of the neoclassical view of perfect competition” (Fligstein, 2001, p.8). The field of economics has also developed a literature in organizational economics and behavioural economics that seeks to identify varying market structures. However, the great majority of the economics literature functions under the assumption that outcomes are efficient, as opposed to historically derived and bound by path dependence, and underestimates the pursuit of power to the benefit of the pursuit of profits.   The emerging literature on global commodity/value chains is another interesting literature that is devoting some thought to market structures. It identifies variation among, and provides us with valuable insights on, the structure of global commodity chains (Gereffi et al., 2005). For instance, Gereffi and Sturgeon identify the governance type for five global values chains, from “market” to “hierarchical,” along a continuum of “degree of explicit coordination and power asymmetry.” This is an approach that is germane to the VOC literature and to the approach adopted here, with the exception that the continuum ends with the most integrated, hierarchical option being an individual, fully integrated, firm. In other words, whereas the VOC literature theorizes about national systems of political economy and not beyond, the Global Values Chains literature theorizes about singular value chains, which at one end of the spectrum, can be subsumed within one firm, as opposed to theorizing about global markets, which are often made up of many such integrated, hierarchical firms.   27  In the classical political economy literature, both Polanyi (1944) and Gilpin (1977) have provided rich theories of market systems, but they have proceeded to define distinct market systems in a temporally sequential way. For example, Gilpin provides a review of exchange systems that have existed throughout history: systems based on reciprocity, redistribution and eventually mobilization before the emergence of the current market economy (Fligstein, 2001, p. xiii), while Polanyi also discusses systems that predated attempts to usher in a market society. Few authors in political science have attempted to theorize about the existence of diverging yet concurrent global market systems. In IPE, Geoffrey Garrett’s analysis of globalization (2000) provides inspiration as he identifies different concurrent dynamics at play in the areas of trade, finance and multinational production at the global level, and different paths to their respective globalization.  Inasmuch as scholarship in political science has been critical of the “market as arena” approach, it has mostly tended to focus on dynamics that occur outside of the markets themselves. That is, scholars who have found the “market as arena” conceptualization insufficient have usually focused instead on something other than markets (e.g., importance of elections, culture or state power).   The need to define markets thus remains pressing. Paraphrasing Neil Fligstein, who asked a similar question from a sociology of economics perspective: What is a market from a political science point of view? A political science perspective can contribute to the study of markets by bringing back the importance of power and politics in the study of global markets.    2.1 What Markets Are  Without a fuller, more sophisticated definition of global markets, how are we to investigate their evolution over time? How are we to identify the market variables that have causal impact, or the market variables that can be fruitfully compared across markets? How can we create useful typologies of global markets? How can we even ask questions about market structures themselves?  This chapter is but a first step towards the creation of a research program devoted to the study of global markets. It by no means claims to have filled the gap. Citing Fligstein, who has launched such a program in the sociology of economics literature, the goal here is “to create an alternative sociological political economic view of markets in capitalist society   28 the global economy. But to do so requires that one look at these social relations power relations in markets from a sociological political economy point of view” (Fligstein, 2001, p. xiii) (italics and strikethroughs added by the author).  To fully account for China’s impact on global markets, we need to be able to describe these markets in a more multidimensional way. For this, the tools of comparative political economy can be fruitfully applied to the study of transnational markets, from defining them to comparing them and to theorizing about their embedded power relations. The VOC literature has argued that qualitatively different political economic systems coexist under the broader umbrella of a capitalist world economy and that they are not necessarily converging toward a single ideal type (Zysman, 1983, Hall and Soskice, 2001, Katzenstein, 1985, Esping-Andersen, 1990; Gerlach, 1992). My comparative analysis of global markets is similarly motivated by the observation that different global market systems, which exhibit different characteristics and dynamics, coexist in the global economy.   As such, my approach to characterizing global commodity markets is germane to the literature on the persistence of divergence between national systems of political economy despite powerful globalization forces, or the ‘divergence’ school, in IPE (Keohane et al., 1996; Tiberghien, 2007; Core et al., 1996). “Industrial countries are not converging toward a single form (Fligstein and Freeland 1995). Instead a plurality of social relations has been observed that structure markets within and across societies. These observations have challenged the neoclassical economists’ view that markets select efficient forms which, over time, converge to a single form.” (Fligstein, 1996, 657) In the words of Gourevitch: “Our account of incentives makes us skeptical about arguments that predict convergence towards a single model. Much talk about the world economy assumes a single, optimal pattern, a single equilibrium, a unique and perfect way of combining all the ingredients of the economy, so that market competition will force all countries to converge. We are doubtful. The economy is too complex, there are too many ways of putting the pieces together. (…) Since countries vary in their internal political dynamics – institutions, preferences, parties, and interest groups – their policy outcomes vary. And thus their corporate governance systems will differ. Change may occur, but not necessarily toward convergence on a single model of governance.” (Gourevitch et al. 2005, p.12)  Market structures, as well as the behaviour of stakeholders in the same global commodity market, vary. The global iron ore and uranium markets are structured quite   29 differently, but even within the global iron ore market, key actors also respond differently to the same event, such as the fall of the benchmarking regime in 2010.  The aim of this chapter is to contribute to a concept-refining, typology-building exercise for global markets.  2.2 Global Markets as Institutional Systems Key assumptions about markets 1) Markets are created (Gilpin, 1970, Polanyi, 1944, Fligstein, 2002, 2008). The idea of a self-regulating market is a myth. This is a position that contrasts with the perception of markets as natural, and of market stakeholders’ behaviour as naturally converging towards competitive markets. “Normatively, I value the critical spirit of British IPE, and of Susan Strange and Robert Cox, because, like them, I am unwilling to accept the contemporary political-economic system as either natural or good” (Keohane, 2011, p. 43).  I draw my analysis of markets as “created” as opposed to “natural” from Fligstein (2002, 2008), Polanyi (1944) and Gilpin (1970). The idea of self-regulating markets is not useful because it does not describe reality (it is useful in research, however, inasmuch as it consists of a belief system that influences individuals’ behaviours and/or policy). This is a position that contrasts with the neoliberal position (IR), which argues that the perfect market is an arena, or a space, not an institutional system.   2) There is a fragmentation of the global market economy into different market institutions or systems, which have different dynamics. I build on Garrett’s analysis of globalization (2000), in which he identifies different dynamics at play in the areas of trade, finance and multinational production. I also build on the VOC literature, which shows that qualitatively different political economic systems coexist under the umbrella of a global world economy (Zysman, 1983, Hall and Soskice, 2001, Katzenstein, 1985, Esping-Andersen, 1990, etc.).  3) Power relations7 and exchange relations are both partie prenante of all markets because individuals, firms, industry associations and states – all market stakeholders –                                                  7 I adopt Susan Strange’s definition of power: “Power is simply the ability of a person or group of persons so to affect outcomes that their preferences take precedence over the preferences of others” (Strange, 1996, p. 17).    30 negotiate these two fundamental forces concomitantly (Gilpin, 1977; Kirschner, 2003; Vilner; Carr, 1939; List, 1841).  Definition  The position adopted here is that global markets can be fruitfully conceived of as institutional systems (Fligstein 2002, Ebner and Beck 2008), which are the result of their material and geographical characteristics, their different institutional histories, the nature of the main actors, embedded power relationships, various local political systems, coordination mechanisms and shared values.   I define global markets as 1) distinct, coexisting, transnational institutional systems of exchange built on asymmetric relations of power, 2) which are subject to path dependence, vary along variables such as concentration, coordination, governance structures, role of the state and systemic stability (Fligstein, 2001; Polanyi, 1944; Rodrik, 1982; Peine, 2013; Krasner, 1974; Baumgartner et al., 1991; Fligstein, 2008) and 3) of which modes of price formation are a critical, but not exclusive, determining feature. This broader definition of global markets allows us to evaluate and characterize global market structures, as well as systemic change, more fully than by relying on a “market as arena” definition.   Therefore, I am not simply saying that markets are subjected to varying degrees of political distortion, which implies that if markets were free from politics, they would tend towards an ideal of competitiveness. This would be a “negative view of institutions” (Duina, 2011), which is nested in neoclassical economics assumptions about institutions and politics as constraints on the efficient functioning of markets.   Instead, this reconceptualization is nested in the view that “markets by themselves are neither morally nor politically neutral; they embody the values of society and the interests of powerful actors” (Gilpin, 2000, p.50). A “more complete picture helps us understand why a particular market structure came into existence [as well as] evaluate how the firms that dominate a particular industry came to occupy that position” (Fligstein, in Ebner and Beck, p.133).   As such, my research seeks to contribute to debates about the existence of multiple stable equilibria, the comparative study of distributive consequences of different market structures, as well as to discussions of the moral status of different market systems. By showing that global markets are fundamentally different complex institutional systems   31 consisting of deep and durable power asymmetries, this dissertation contributes to research which seeks to explain persisting variation among global markets, a research program that is not feasible if one assumes global markets to be symmetric in structure. It also seeks to contribute to debates about economic behaviour. This is in line with the view that “market rationality is neither universal nor apparent as the neoclassical paradigm assumes. On the contrary, the ideal types of market rationalities are multiple and contested” (Kwon, Hyeong-Ki, 2004, p.98). In thinking about global market types, it is important to differentiate between concentration levels and governance structures. The fragmentation of a market into many actors (low concentration levels) is not necessarily equivalent to the ideal notion of competitive markets. A fragmented market can be a market with high transaction costs, inefficient transportation structures, poor regulatory context, presence of information asymmetries, and other inefficiencies, just as in the case of a concentrated market. In other words, the concentration levels and the well-functioning of a market do not necessarily go hand in hand. Indeed, two markets can be equally concentrated without nearly the same amount or type of coordination among producers, and vice-versa. For instance, the top four firms in the global potash market controlled 68% of global exports in 2012 (Chen, 2014), whereas the top four firms in the global uranium market controlled 81.8% of exports in 2009 (Global Key Uranium, 2011). However, until 2013, the potash market was composed of two highly institutionalized market structures, cartel-like organizations, whereas the uranium market is much more fragmented and lacks formal coordination mechanisms among producers.   Finally, this research contributes to debates about the underlying tendencies of markets towards competition or concentration. Streeck (2009) argues that there is a natural tendency towards competition in markets. In the same line of argument, others, such as Krasner (1974), argue that there is a natural tendency towards the break-up of cartels. At the other end of the debate, others argue that the natural tendency in capitalist markets is towards concentration and coordination (Foster and McChesney, 2012; Nolan, 2012; Strange, 1996). Fligstein also argues that when legal frameworks allow it, cartel arrangements become the major strategy for coping with competition (1990, p.23), such as in Europe in the nineteenth century. He also argues that “the social structures of markets and the internal organizations of   32 firms are best viewed as attempts to mitigate the effect of competition with other firms” (Fligstein, 1996, p.657).   In this dissertation, I make the case that despite the difficulties in fully overcoming collective action problems, the momentum towards coordination remains quite strong in global commodity markets (Gilpin, 1977; Ostrom, 1997; Nolan, 2012). It has taken many shapes and forms throughout the twentieth century, even as the golden era of formal cartel arrangements faded away. In order to fully engage in these debates, we need to be able to compare and measure different global political economic market structures – and their evolution across time – along variables such as their levels of concentration or coordination.  2.3 Typology of Global Markets  This exercise builds on Rodrik (1982) and Krasner (1974), who developed multifaceted indicators to evaluate global extractive commodity markets. Rodrik defined global commodity markets along four variables: degree of concentration, vertical integration, state ownership and trade and price-formation mode. Krasner identifies the conditions that make durable cartel behaviour more likely: price inelastic demand, high barriers to entry, high market concentration, shared experience among producers, lack of consumer resistance, ability to work with an extended time horizon and shared values. His conceptualization of the variables likely to make cartel behaviour more likely is still useful today, although he did not proceed to develop measures for each indicator.   A first attempt at developing a useful list of key global market characteristics can be found below (see Table 3). The first category of descriptors is grouped under the label of the “shape” of the global markets. I define levels of concentration (Krasner, 1974; Rodrik, 1982) and geographical dispersion as key variables in that category.   In the second category, I am interested in defining the nature of major market actors (state-owned or private companies, industry associations, states, etc.) (Rodrik, 1982).   The third category is the coordination capacity of market actors, state or private. It includes the capacity to overcome collective action problems, or to coordinate behaviour, but also the alignment of expectations about how a market should work among market stakeholders. This notion is building on Fligstein’s “conceptions of control” or the understandings that structure perceptions of how a market works (2001). It is also germane to Ostrom’s concept of symmetry of interests, resources and time horizon (1997), and to   33 Krasner’s notion of shared experiences/values among producers (1974). This last characteristic is fundamental inasmuch as it determines whether market actors, concentrated or not, function with the same assumptions. This variable is especially crucial in the context of a power transition that brings actors functioning under different national systems of political economy together in a global market. It is closely linked with the capacity for coordination of market actors.  Table 3 - Key Characteristics of Global Markets Global market characteristics 1. Shape of producer’s market  1.1 Geographical distribution 1.2 Concentration of production  1.3 Tightness of supply 2. Nature of actors 2.1 Role of the state 2.2 Role of large firms 2.3 Role of small firms 2.4 International market organizations 3. Coordination capacity  3.1 Alignment of expectations (about how a market should work) 3.2 Capacity to overcome collective action problems 4. Governance structures 4.1 Pricing regimes  4.2 Transportation systems  4.3 Relevant regulatory context   The fourth category of market characteristics is a market’s governance structures. These structures include, but are not limited to, pricing regimes (price-setting mechanisms, or rules of exchange) (Fligstein, 2001); transportation systems; and relevant regulatory contexts   34 (rules that define relations of competition and cooperation among firms, including property rights)8 (Fligstein, 2001; Hall et al., 2001; Gourevitch et al., 2005).    Based on this initial list of key market characteristics, multiple typologies are possible. Typologies are tools that allow us to question existing, often implicit conceptual categories, and offer opportunities for novel theoretical research endeavours. They are not meant to be exclusive or definitive, but to contribute to the advancement of knowledge by provoking our thinking and providing a clear statement that we can question further, and refute, if necessary.   I create an initial typology of global markets by combining the first and last of the market characteristics: the level of concentration/fragmentation, and the capacity for coordination (see Figure 4).  Figure 4 - Typology of Global Markets                                                         8 I thank Wang Yong for raising this point with me.  Regional!blocks!Cartel/Oligopoly!Segmented!market!Competitive!market!Coordination Lack of coordination Concentration Fragmentation   35  This typology is a more fine-grained tool that allows us to think about markets in new ways and with counter-intuitive implications. For instance, it allows us to refine our perception of the fragmented market as one with two possible outcomes: a segmented, dysfunctional type (the Chinese domestic iron ore market is a good example) and a competitive type. Indeed, a competitive market is one that requires the alignment of expectations in how the market functions (agreement on property rights, the rule of law, access to resources and arbitration, fairness of the game, pricing regimes, etc.), or in other words, a certain degree of coordination of behaviour. It also allows us to refine our understanding of coordination. There are two types of coordinated markets: oligopolistic and competitive markets. Similarly, at the concentrated end of the spectrum, we find two types of markets: oligopolistic/cartelized markets (iron ore and potash) and segmented, regional blocks (uranium). The typology obfuscates certain assumptions about competitive markets, and introduces the possibility that a market can be concentrated and not coordinated, or fragmented but not competitive.   2.4 The Impact of a Dominant Emerging Consumer  For the purpose of this dissertation, the universe of cases I am referring to is “within-market power transition from one dominant consumer to another.” In other words, I consider “emerging dominant consumer” – “commodity market” dyads as one case, and argue that we can use the concepts of “major power transition” and “systemic change” within a given market system (as opposed to within the global economy at large). Japan taking over from Germany and the US and having a systemic impact on the global iron ore market in the 1960s is a relevant example (see Chapter 4).  I argue that evaluating the impact of a single country’s emergence on global markets is justified when two conditions are met: 1) its impact on commodity markets is large enough to be systemic9; and 2) its levels of import dependence are large enough for the country not to be able to be self-sufficient for extended periods of time.10 For instance, by 2012, China accounted for 62% of global iron ore imports (see Figure 5).                                                  9 For the purpose of this dissertation, I define an impact as systemic when a country becomes the largest consumer in the world, or when it starts importing more than 20% of global exports, whichever occurs first.  10 For the purpose of this dissertation, I define a significant level of import dependence as above 20% of consumption.    36 Figure 5 - China as Share of Global Iron Ore Imports  Source: UNCTAD, 2013  2.5 Dependent Variable: Global Pricing Regimes  The core of this analysis centres around the impact of China’s rise on one of the market characteristics (governance structures) derived from the typology of global markets developed above: global commodity market pricing regimes.    Pricing regimes matter for reasons of efficiency, as well as for questions of distribution and fairness at the global level. Chinese actors’ preferences for market or strategic pricing regimes will be a determining feature of long-term trends in how these markets evolve. I operationalize the dependent variable along a “market – strategic” continuum. This conceptualization is drawn from the VOC literature. Hall and Soskice, and Hall and Gingerich, define national systems of political economy along a “market – strategic” continuum11 (2001, 2004). A movement towards more market is labeled “marketization,” and in the opposite direction, “demarketization.”12 I define “marketization” as composed of two categories: “liberalization”13 and “financialization” (see discussion below, as well as                                                  11 Hall and Gingerich define the two modes of coordination as follows: “In one, firms coordinate with other actors primarily through competitive markets, characterized by arms-length relations and formal contracting. Here, equilibrium outcomes are dictated primarily by relative prices, market signals, and familiar marginalist considerations. In the other modality, firms coordinate with other actors through processes of strategic interaction of the kind typically modeled by game theory. Here, equilibrium outcomes depend on the institutional support available for the formation of credible commitments, including support for effective information-sharing, monitoring, sanctioning, and deliberation.” (Hall and Gingerich, 2004, p. 7-8)  12 I thank the participants of the Global Political Economy workshop at the Balsillie School of International Affairs in 2013 for their discussion of my Dependent Variable. In particular, I thank Herman Schwartz, Eric Helleiner and Kathryn Hochstetler who helped me broaden my definition from only “liberalization” to “marketization.” I also thank Lou Pauly for his comments on the matter.  13 See Hughes (2014, p.37), for an interesting typology of liberalization in a national context in the oil market.  China!65%!Japan!12%!EU!10%!South!Korea!6%!Others!7%!Iron'ore'importers'(world'market'shares),'2012'  37 Table 6 and 7). Values are compiled prior to and following China’s emergence as the dominant consumer. A qualitative description of outcomes is represented in Table 4, and illustrated in Figure 6 below.14 Table 4 – DV: Global Pricing Regime Change as a Result of China's and Japan's Emergence Market change 1. China – iron ore pricing 2000s  2. China – iron ore bulk shipping  2000s 3. China – potash pricing  2000s 4. Japan – iron ore (pricing and shipping) 1970s  5. China – uranium (pricing and market IOs) 2000s Pricing regimes (Benchmark vs. spot, short-term vs. long-term (L-T) contracts, volatility) Fall of benchmark, opening of new spot trading platforms, large increase in volatility and frequency of trades, fewer L-T contracts  Suppliers’ build ships to compensate for price differentials, government decree blocking docking of ships, impasse resolved through L-T international sharing agreements  Doubling in frequency of negotiations, fall of one cartel, caving in of global producers towards lower prices, benchmarking system survives Increased role of state, coordination of purchasing by Japanese firms, benchmarking pricing regime, regime stability, L-T contracts Increased role of state, joint ventures with state partners, long-term contracts, low volatility, increased participation in global market IOs Marketization (Market 0 - Strategic 10) Significant marketization 8/10 to 3/10 Moderate demarketization 5/10 to 7/10 Moderate marketization 9/10 to 7/10 Significant demarketization 6/10 to 9/10 Moderate demarketization N/A                                                   14 Colours are added to highlight the relationship between elements of Table 4 and Figure 6.   38 Figure 6 - Extent and Direction of Pricing Regime Change as a Result of China's and Japan's Emergence   Using mixed-methods data analysis software to code my interview transcripts, I conducted a “co-occurrence” search for particular concepts in my case studies (see Appendix 2 for a more complete table). Pulling the results from searches for the co-occurrence of concepts relating to global pricing regimes for each commodity market has yielded the results presented in Table 5 below.15   Table 5: Code Co-Occurrence !! Iron%Ore% Shipping% Potash% Uranium%Marketization! 27% 1% 8% 6%State!intervention! 4% 3% 13% 19%Financialization! 17% %% 4% 4%Increase!in!frequency!of!trades! 7% %% 2% %%Spot!market! 30% %% 2% 5%Demarketization! %% %% 3% 5% The use of co-occurrence searches in my interview transcripts is a first attempt to use a large volume of interview data more systematically. The preliminary results serve to confirm more impressionistic recollections of aggregate interviewees’ responses. For                                                  15 The total number of interviews which directly mentioned one of the four markets is 114, 38 of which mentioned the iron ore market, eight the shipping market, 36 the potash market and 32 the uranium market. I colour-coded the results for easier analysis (pale blue: 0-4 mentions; beige: 5-9 mentions; pale orange: 10-14 mentions; orange: 15-19 mentions; dark orange: 20 mentions and above).         39 instance, I can say with much certainty that interviewees speaking about the iron ore market mentioned the marketization of global pricing regimes more often than did interviewees speaking about the potash or uranium markets, although it is difficult to specify how much more without the use of more precise measuring techniques.   A “concept frequency” analysis and the identification of co-occurrences allowed me to say that the concept was mentioned more than four times more often in the context of the iron ore market than in the uranium market. Similarly, I recall that interviewees very seldom mentioned state intervention (the opposite of liberalization) in the context of the iron ore market, and mentioned it more often in the context of the potash and uranium markets. But how seldom? And was the concept mentioned equally in the context of the potash and uranium markets? A manual analysis of the interview transcripts, given the volume of these transcripts, provided only limited insights. The results of the preliminary code co-occurrences mentioned above is not meant as a test, but rather as a validation of broader claims I make in the dissertation about the aggregate views of interviewees for each particular market. Another analysis of code co-occurrence can be found in Table 21.     What emerges from this analysis is that interviewees who mentioned concepts related to the marketization and the financialization of global pricing regimes were much more likely to do so when speaking about the iron ore market than when they were speaking about the potash or uranium markets. The opposite is true of state interference (opposite of liberalization), which was mentioned more often in conversations about the uranium and potash markets. This merely serves to validate my assertion that interviewees identified marketization (liberalization and financialization) forces far more readily in the case of the iron market, followed by the potash market, and then the uranium market.     2.6 Marketization  The changes I observe in global commodity markets happen within the larger forces of economic globalization, defined as “the integration of international markets in goods, services and capital” (Streifel, 2006). I define marketization as being composed of two related but distinct notions: liberalization and financialization (see Table 6 and 7 below). As such, marketization is broader than the retreat of the state (which is the liberalization part) and can be conceived as a movement away from strategic interaction between market actors.    40 Table 6 - Pricing Regimes (DV Indicators) Marketization (Liberalization + Financialization) Liberalization Market Strategic Small role of the state Large role of the state Low level of state ownership High level of state ownership Financialization Market Strategic High frequency of pricing Low frequency of pricing New spot trading platforms No new spot trading platform Multiple prices Benchmark Short-term Long-term Included in major commodity indices16 Not included in major commodity indices   A movement towards the marketization of a pricing regime is a movement towards more liberalization (less state) and/or more financialization (more interaction with financial markets). In practice, the marketization of pricing regimes entails a movement towards spot pricing, short-term contracts, high frequency of trades, volatility and less state involvement.   A movement towards the demarketization (strategic end of the spectrum) of a pricing regime is a movement towards more state and/or more hierarchy and power-based dynamics. Here, demarketization of pricing regimes entails a movement towards benchmarking, long-term contracts, low frequency of trades, low volatility and more state involvement.                                                     16 These include the Bloomberg Commodity Index (BCOM) (formerly the Dow Jones-UBS Commodity Index), the Goldman Sachs Commodity Index, the Thomson Reuters/Jefferies Index and the Deutsche Bank Liquid Commodity Index. Among other commodities, the Bloomberg Commodity Index includes natural gas, Brent crude oil, corn, soybeans, wheat, sugar, cotton, copper, aluminum, zinc, nickel, gold, and silver.   41 Table 7: Pricing Regimes (DV Indicators), operationalization Marketization (Liberalization + Financialization) Iron ore Shipping Japan Potash Prior17  Post Prior  Post Prior  Post Prior  Post Liberalization (0-4, where 0 is liberalized) 1 1 1 3 1 2 2 2 Market (0) Strategic (4) Operationalization         Small role of the state Large role of the state Presence of unilateral actions, regulations, state influence, Yes (2)/No (0) 0 1 0 2 1 2 1 1 Low level of state ownership High level of state ownership % State ownership, Low (0)/High (2) 1 0 1 1 0 0 1 1 Financialization (0-10, where 0 is financialized) 10 3 6 7 8 10 10 8 Market (0) Strategic (10) Operationalization         High frequency of pricing Low frequency of pricing High volatility (0)/ Low volatility (2) 2 0 0 1 1 2 2 1 New spot trading platforms No new spot trading platform Yes (0)/ No (2) 2 0 2 2 2 2 2 2 Multiple prices Benchmark Multiple prices (0)/ Benchmark (2) 2 0 1 1 1 2 2 1 Short-term Long-term % Total trades conducted on spot market,  High (0)/Low (2) 2 1 1 1 2 2 2 2 Included in major commodity indices Not included in major commodity indices Yes (0)/ No (2) 2 2 2 2 2 2 2 2 Marketization (Liberalization + Financialization), compound variable: (0-10) 8 3 5 7 6 9 9 7   I define liberalization as “an outcome in which the market progressively replaces government as the mechanism through which resources are allocated” (Garrett, 2000, 941). I measure liberalization as the level of state ownership, and the presence or absence of state                                                  17 Prior and post China’s (or Japan’s) emergence as number one consumer.   42 influence on pricing regimes, whether through government influence, regulation or unilateral actions. I define financialization as the “strengthening interaction of commodities markets with the financial system over the last decade (…), i.e. returns from commodities are increasingly pooled with returns from pure financial assets (‘pooling effect’)” (Hughes, 2014). I measure financialization of pricing regimes as the frequency of trades; the presence of spot trading platforms; the presence of a benchmarking system that sets the price for long periods of time; the percentage of trades conducted on the spot market; and whether a commodity is included in major commodity indices.  Should we be surprised by the fact that the emergence of a state-led authoritarian economy such as China leads, in some cases, to the marketization of global market pricing regimes? Some may suggest that China should be expected to cause marketization, although little, if any, evidence of this can be found in writing. An exception is the 2014 book, By All Means Necessary: How China’s Resource Quest is Changing the World, by Elizabeth Economy and Michael Levi (pp. 36-40), which tells the story of the marketization of the global iron ore market (which I describe in more detail in Chapter 4), although the authors agree that this turn of events was rather unexpected.   China’s impacts on global markets are still poorly understood empirically because, among other reasons, the impacts are so recent, and because China itself is changing very quickly. But why, at least deductively, would some scholars expect marketization impacts?  One could point out that the last two hegemonic powers, the US and the UK, have had a liberalizing impact on a variety of global markets following their emergence as dominant powers (Valiante, 2013). Would marketization be a natural preference of major emerging dominant powers? Friederich List’s 1841 “National Systems of Political Economy” (Fallows, 1994; Chang, 2002; Rodrik, 2011) suggests a simple explanation for this behaviour: after reaching a certain level of development and after achieving a position of dominance in certain global markets, the dominant power’s interest becomes forcing subordinate economic powers to open their economies to global market forces. But the US and the UK had a global marketization impact at a markedly different stage in their domestic levels of development – they were home to quite different economic structures at the time, compared to China’s, and they faced different international environments.    43  Some may also argue that the marketization trends I observe are just part of a broader and natural process of global convergence. But my tracing of the process through which it actually happens unveils a more complicated story. Although many commodity markets are experiencing some levels of liberalization, the trends I observe are unequal and their patterns are substantively different. In fact, marketization trends are far from overwhelming (oligopolies are hard to displace). There are always winners and losers and tugs of war between various interest groups who fight for a particular outcome. There are also unintended consequences. It may also be that at certain times in history, outcomes, at least for a period of time, tend to resonate across markets and towards a certain distant horizon (globalization, financialization). But if we do not carefully trace the process of marketization (and its reversals), not only will the causal mechanisms behind institutional change remain obscured, but within-case reversals (such as the marketization - demarketization - marketization pathway in the iron ore market, following Japan’s and then China’s emergence), would also remain hidden from view.   “The ‘financialization’ wave that has swept global markets over the past 30 years has only recently included extractive commodities, and many of these commodities remain outside exchanges or are excluded from major commodity indices. Iron ore, tin, uranium, potash or rare earths, for instance, are not included in major commodity indices such as the Dow Jones-UBC Commodity Index, the Goldman Sachs Commodity Index, the Thomson Reuters/Jefferies Index or the Deutsche Bank Liquid Commodity Index. They are just beyond the financialization frontier” (Massot, 2013).   In other words, I argue here that just as markets are not “natural,” neither is marketization a “natural” process. If there is currently a general marketization tendency across markets, it is not apolitical. Market changes are a political economic process; they consist of an unsettling of existing power relations in global markets. What some see as a natural trend is in fact the displacement or reorganizing of existing power hierarchies, which are replaced by different power distributions, not a tendency towards more apolitical markets. Marketization occurs to the benefit of some actors and at the expense of others. Change in a particular direction is the result of a victory by certain interest groups, and as such, needs to be explained, much like the fall of the Bretton Woods regime in the 1970s needed to be explained (Helleiner, 1994; Abdelal, 2009; Best, 2004).   Global market movements, both over time and across markets, are not unilateral. There has been a push and pull of marketization/liberalization forces throughout the   44 twentieth and early twenty-first centuries, not a constant movement towards one type of market arrangement. Polanyi’s pendulum movement metaphor is apt here. Garrett’s study of different aspects of globalization (2000) also provides a good example of how different international economic regimes (trade or finance, for instance) have not globalized to the same extent or following the same causal logic.   In addition, China is a different domestic economic power than were the US or the UK at the time of their emergence. For this reason, many China scholars expect that China will have a statist impact on global markets, rather than the opposite impact.   In fact, many industry insiders persist in saying that certain commodity markets are not fungible enough to become fully commodified and financialized (some include iron ore in this category, and most include potash). In other words, we cannot assume financialization. To the extent that we witnessed a movement towards increasing marketization, liberalization and financialization from the 1980s until the Global Financial Crisis (GFC) of 2008, this coincides with a period of stability and lack of financialization in both of my cases. Only after the GFC, and after some retreat in liberalization and financialization globally, have the two markets experienced some movement towards increased marketization. How do we explain this timing?  China does not in fact have a uniform marketization impact on global markets (thus the argument that any emerging power would have a marketizing impact on global market structures is insufficient; see also the Japan case). Therefore, the burden of explanation remains relevant and necessary, from the moment we recognize that financialization is not unavoidable, or even equal, across cases.    Furthermore, countries’ foreign economic policy making is not necessarily an expression of their domestic economic policy making (for example, the US was home to liberalized domestic markets for a long time before it started promoting liberalization overseas) (Helleiner, 1994; Abdelal et al., 2010).  So the difficulty is predicting when and why China or any systemically relevant power would have a marketization impact on global market structures. For instance, Japan’s emergence had an opposite impact on the global iron ore market. Indeed, I argue that the state of the market prior to a given systemically relevant country’s emergence, as well as that country’s domestic market structures, are both crucial to explaining marketization outcomes.    45  Why else is China’s marketization impulse surprising? First, for marketization to even occur, one needs to admit that it is possible. This means admitting that global markets are not competitive to start with, which we have now seen is often the case in global extractive commodity markets and beyond (Nolan, 2012).  Second, if China is having a marketization impact on global markets, it is not simply because “Chinese actors are powerful, and they get what they want,” which would be one of the claims deriving from Hegemonic Stability Theory (that is, dominant states prefer global liberal economic regimes, and will coerce other powers in fulfilling their aims). In fact, my cases show that at critical moments, dominant Chinese stakeholders do not get their preferred outcomes internationally. In this particular light, whether China is actually in a powerful position at all is not something we should take for granted. In this case, what impact should we expect of a domestically weak China on powerful global oligopolies? The best bet would be “no change at the systemic level,” as there is no capacity for impact. Many liberal scholars are actually expecting this result (see Ruggie's power and purpose argument, 1982).  China’s marketizing impact is also surprising because the most powerful Chinese actors do not always want marketization. In the iron ore market, for instance, the fall of the benchmarking regime was not in the interest of the most powerful Chinese stakeholders, the large steel SOEs and holders of import licenses in 2010. It was also not a policy or goal of the relevant organs of power in this market in China, such as the NDRC, the State Council, or the lead negotiator, the China Iron and Steel Association (CISA). Changes in relations of power, including financialization, are always in the interest of some actors at the expense of others. In this case, however, marketization was going against the interests of the most powerful Chinese stakeholders, and was thus in many ways unintended.   Lastly, for those who argue that China’s domestic markets are already very “market-oriented” and that this is why we observe symmetrical marketization impacts globally, suffice it to say here that the fragmented Chinese domestic markets I study are not working efficiently under clear rule-of-law conditions (in other words, they do not meet the definition of marketization I develop here). The Chinese domestic markets that I am studying are not only fragmented and “illiberal,” but they are also messy, and in many ways out of the reach of Chinese top-level regulators. The iron ore market, for instance, is working under conditions that are not efficient and competitive, which makes the symmetry argument   46 difficult to sustain. The Chinese government has reacted to this problem by trying to encourage consolidation of the iron ore industry and the licensing process, only to then reverse course and liberalize the licensing process altogether in 2013. In other words, and as the rare earths industry illustrates, a higher number of producers is not always an indication of higher levels of efficiency, competitiveness or competition. Market characteristics are multidimensional.   On the other side of the debate, but following the same logic, are observers who expect China to have a statist effect on the global economy. This is because many expect certain illiberal market characteristics, or state-market relations present in China, to be “exported” by Chinese stakeholders through their behaviour at the global level. But this intuition is also overly simplistic. Statism does not get expressed to the same degree in all my cases, and while the impulse is present, it is certainly not translating into clear symmetric impacts on global market structures. In fact, the opposite outcome is arguably more evident empirically. Therefore, I propose that a given country’s foreign economic policy and global impact are not a reflection of the nature of its domestic economic system (a liberal market economy can certainly conduct illiberal foreign economic policies or have a demarketization impact on a certain global market, and vice-versa). Whether Chinese actors have a de/marketization impact globally, or even whether they decide to advocate for a larger/smaller role of the state in markets, need not be a reflection of the situation within their borders. Understanding the interaction between market structures at both domestic and international levels, and the resulting power asymmetries, is crucial for explaining outcomes. 2.7 When Do Market Actors Prefer Marketization? When do large consumer or producer firms prefer marketization? This question has not been answered conclusively in the economics literature (Levenstein and Suslow, 2006). There are many reasons for firms to coordinate behaviour, as well as to cheat against one another, and we have seen the establishment of coordinated pricing regimes in a variety of situations with contrasting fundamentals.  There are two major types of predictions with regards to marketization in the IPE literature, which will be discussed further in Chapter 3: the ideal-typical Liberal position, which posits that states gain by operating in open, free-market environments; and the Hegemonic Stability Theory (HST) position, which posits that dominants states will also   47 prefer open markets, and will especially seek to coerce international market players into more open and free-market practices. HST’s argument only holds if global markets are not already liberal, and if the domestic conditions of the dominant player in that particular market would benefit from such open and liberal markets. It adopts the perspective of a dominant state, even though global markets are now populated with larger and smaller firms that may have divergent preferences for more or less competitive markets.  2.7.1 Open economy politics The Open Economy Politics (OEP) literature has identified conditions under which firms within a national economy will prefer trade liberalization. This approach is based on an assessment of market actors’ interests, which in turn is based on their relative positioning in their national vs. the global economy (Lake, 2009). Within this literature, some scholars have focused on the impacts of globalization (or exogenous easing of international exchange (Frieden and Rogowski, 1996)) on domestic political economic outcomes. Here, authors have focused on factors of production, industry cleavages, or a combination of both, as main explanatory variables (Frieden and Rogowski, 1996; Rogowski, 1989; Hiscox, 2001). These outside-in approaches to the problem of exposure to international trade point to possible answers to the question of interest here: When and why do firms prefer open markets? However, an important distinction needs to be made as per contrasts in the objects of study. OEP research, as well as its scholarly ancestors (for instance, Katzenstein, 1978), is ultimately interested in explaining national economic policy. This dissertation is interested in explaining changes in market structures at the global level. As such, my dependent variable is located outside of the national policy-making arena of the state.   Despite the fact that OEP has been mostly interested in national-level economic policy making, and despite the fact that this dissertation poses an inside-out question, as opposed to an outside-in question, the literature offers potential responses. Taking inspiration from Frieden and Rogowski’s work, one could surmise that Chinese firms relying on abundant factors of production would tend to prefer open markets. Whether we consider the Chinese economy to be abundant in labour or capital, it becomes apparent that factor endowments do not differ enough between the markets under study in this dissertation to explain divergent international preferences and behaviours. Indeed, land is scarce for all three case studies (that   48 is, China is heavily dependent on imports for all three), whereas relative abundance of labour and/or capital, a priori, does not vary enough to explain divergences.  Pushing the reasoning a little further is interesting. One could argue that the potash market’s main constituents, for instance, are farmers, whereas one of the iron ore market’s key constituents is real estate owners in large cities, or owners of capital. Could this give us traction? It may, but following the logic of cleavages in factors of production, one would need to conclude that since China is land-scarce, it would favour liberalization in the iron ore market. In fact, as we will see, the story is quite a bit more complicated than this. In fact, key players in the iron ore market did not want the marketization of global pricing structures, and second, among other things, central Chinese state organs have been working for the past decade to liberalize the interface between the domestic and the global potash industry. As we can see, cleavages along factors of production offer blunt tools in the cases under study here. The study of within-market variation in the iron ore case also indicates that industry cleavages are insufficient to explain outcomes. There are both within-industry variation in preferences and variation across markets endowed with similar factors of production that are not best captured by the tools described above.   2.7.2 German historical school            Coming from a different angle, the German Historical School argues that dominant economies only start preferring open markets when it benefits their domestic industry. Emerging economies choose to protect their own domestic industries until their firms are competitive with firms in developed economies (List, 1909). The German Historical School introduced an aspect of power, interests and sequencing in firms’ likelihood to prefer free and open markets, which are important, and much needed components.  But the same limitation remains with this literature. Following this school of thought, an emerging economy should prefer protectionist strategies, which is to say protectionist state policies. Later on in their development process, when their national firms are dominant, emerging economies’ preferences will then evolve until they align with those of developed economies, in favour of the liberalization of the national interface (import restrictions, quotas, etc.) with the global economy. Again, the reality of Chinese preferences, behaviour and impact is more complex, not least because some impacts are unintended.   49 However, the built-in argument of the German Historical School, which other scholars have followed up on (Chang, 2003), is useful. It is cognizant of a country’s stages of economic development, and says that only when national economies have developed strong domestic firms will they support free and open markets abroad. This is an argument that resonates with Hegemonic Stability Theory and will be addressed again at various points throughout this dissertation. 2.7.3 Relationship with price trends   Describing and explaining the impact of China’s emergence on global market institutions, and their variation, is difficult without a thicker definition of markets. Price series analyses do not actually provide rich enough empirical evidence to allow us to understand the multidimensional changes that have occurred in market structures in the past 15 years.18  Figure 7 - Commodity Prices (1980-2012)  Source: IMF Data, 2012 Often, however, existing tools that seek to evaluate global commodity markets limit themselves to price trends. One of the few recent articles that offers a sophisticated political                                                  18 I thank Eric Helleiner for raising the question of market stakeholders’ preferences in relation to price trends. 0!50!100!150!200!250!300!350!400!450!500!550!600!650!700!USD/varying'quantities'Commodity'Prices'1980'?'2012'(real'$2005)'Copper,!Price!Index!grade!A!cathode,!LME!spot!price,!CIF!European!ports!Aluminum,!Price!Index!99.5%!minimum!purity,!LME!spot!price,!CIF!UK!ports!Iron!Ore,!Price!Index!China!import!Iron!Ore!Fines!62%!FE!spot!(CFR!Tianjin!port)!Tin,!Price!Index!standard!grade,!LME!spot!price!Nickel,!Price!Index!melting!grade,!LME!spot!price,!CIF!European!ports!Zinc,!Price!Index!high!grade!98%!pure!Lead,!99.97%!pure,!Price!index!LME!spot!price,!CIF!European!Ports!Uranium,!Price!index!u3o8!restricted!price,!Nuexco!exchange!spot!  50 economic analysis of global commodity regimes (concentrating on the oil market) was written by Colgan, Keohane and Van de Graaf, and identifies “dissatisfaction with existing energy regimes” as a variable in explaining timing of innovation. But the measure of dissatisfaction is solely based on price levels. Apart from the fact that producers are not always dissatisfied when prices drop (the current high production/low price scenario in the iron ore market is producer-led), price trends can also be similar across markets, for instance in the run-up to the Global Financial Crisis in 2008 (see Figure 7 above). Yet, these trends obscure very different dynamics across distinct commodity markets. Price trends do not provide a satisfactory explanation as to why the iron ore market pricing regime fell apart in 2010, or why the 2012 potash benchmark regime did not. Relying on deductive economic reasoning would suggest that large producers have an interest in marketization of pricing regimes when prices are rising (they are price makers, have the upper hand, and would rather play one consumer off against the other), and an interest in demarketization when prices are dropping (they are in a weaker position, are price takers, and would rather function under a more predictable and stable pricing regime). On the other hand, large consumers should have an interest in demarketization when prices are rising (they are price takers, and would rather rely on more stable prices), and in marketization when prices are dropping (they have the upper hand and would rather play one supplier off against the other, renege on contracts, and so forth) (see Table 8 below).  Table 8 - Deductive Pricing Regime Preferences (Producers and Consumers), by Price Trend  Price trend up Price trend down Producers Marketization Demarketization Consumer Demarketization Marketization  Prices absolutely affect market stakeholders’ incentives, and play an important role in determining market actors’ preferences for global market pricing regimes. We will see that in some cases, price trends do in fact provide a powerful indication of market actors’   51 preferences, but this logic alone does not explain outcomes in many other cases being examined here. It is worth noting that in a world of volatile commodity prices, it is difficult to establish when actors consider mid-term, cyclical trends, or short-term price cycles, or whether their calculations are even accurate. It is difficult to define what “price-based” preferences are a priori. For instance, Chinese small and medium enterprises (SMEs) chose to break rank and procure iron ore on the emerging spot market in 2010, despite the fact that this move would lead to higher and more volatile prices in the short term. To complicate things, just five years later, declining price trends have priced many of them out of the market altogether because they have higher operating costs.  But even presuming that actors look at short-term price trends (say, a few months’ horizon), the logic described above does not hold in many of my sub-cases. For one thing, large Chinese and Japanese iron ore importers have routinely behaved differently and expressed different preferences concurrently, and this is despite the same global price trends.  In another example, the global potash suppliers did not push for the marketization of the global potash market during the commodities boom leading to the Global Financial Crisis (or at the very least, were not successful in doing so), an outcome diverging from the iron ore case (which, arguably, was subject to similar expected price trends). Similarly, the Russia-Belarus potash exporting cartel fell apart during a downward price trend. Furthermore, Belaruskali and Uralkali continue to undercut each other, not the other way around, as the price trend logic would suggest. In addition, some Chinese policies that target global market structures have persisted throughout upward and downward price trends. At other times, the behaviour of the key Chinese importers has also run contrary to expected behaviour given the price trends. For instance, the large Chinese iron ore consumers had clear preferences against the marketization of the global iron ore pricing regime in 2009, despite strong downward trends in prices.  Also, it is worth remembering that a sharp decrease in global commodity price does not necessarily affect the large consuming firms in intuitive ways. In the iron ore case, the recent drop in prices is forcing the closure of many small Chinese iron ore producers, in effect benefitting the larger, more cost-effective mining firms. Finally, this kind of reasoning does not leave much room for the role of actors outside of the main consuming/producing   52 firms. For instance, in the Chinese iron ore benchmark case, the leading importers did not want the end of the pricing regime (marketization), but in the end, the behaviour of small firms led to that outcome.  In fact, this logic would suggest that producers would have the upper hand in times of rising prices, and consumers would have the upper hand in times of declining prices. In both of these cases, the preferred outcome would be marketization. What, then, explains instances of demarketization?  In other words, price trends are an important component of the equation, but taken alone they cannot fully explain behaviour in global markets, as they suffer from at least two core shortcomings:  First, they do not help us identify which market stakeholders’ preferences will prevail internationally, as we have seen that consumer and supplier firms should have opposite preferences in reaction to a given price trend. This is also true domestically, as large and small firms operating in the same domestic industry exhibit different preferences towards global pricing regimes. My theoretical model seeks to address this shortcoming. Second, they also cannot explain why consumers (producers) sometimes have preferences that do not align with expected preferences given the price trend. My argument also seeks to address this shortcoming (see Chapter 3).   Price trend-based rationales tend to predict the same behaviour given certain price trends or certain market positions. For instance, it might predict that the emergence of a large consumer in an oligopolistic market leads to a particular outcome. However, my empirical cases not only reveal that different emerging consumers have different preferences towards global pricing regimes, but that the same consumer, for example, China, can have different preferences towards global pricing regimes as well, depending on the market. In addition, we are witnessing an era of deep structural changes, and this is why a multifaceted political economic analysis brings to light more complex dynamics at play. “Prices alone may not help us to identify the onset of new periods in world commodities regimes history, and inferring causal relationships on the basis of price alone, and especially in times of structural break, is not sufficient. Qualitatively, new periods may be characterized by higher prices, but they may also be characterized by higher volatility, transfer of production locations, prices being supported by different dynamics or mechanisms, or even by characteristics that are not immediately expressed by prices. Therefore   53 projections based on past price trends alone may not be helpful.”(Cashin, Liang, and McDermott, 2000)  In conclusion, preferences for marketization or demarketization of global pricing regimes are determined to some extent, but not exclusively, by price trends. This dissertation seeks to further specify the conditions under which the dominant players on either side favour the marketization of existing pricing structures, and how to explain which side gains the upper hand.     54 CHAPTER 3 – TWO-LEVEL POLITICAL ECONOMY OF GLOBAL MARKET CHANGE   The goal of this chapter is to develop my two-level argument after situating it within the broader literature. I start by outlining two major alternative approaches, including their limits, to the questions I ask in this dissertation: the Liberal approach, and Hegemonic Stability Theory approach. I then provide a review of the literature on existing two-level frameworks in the IPE of resources literature. Finally, I present my two-level model on the impacts of China’s emergence as a dominant consumer of commodities. I make the case that market power asymmetries between suppliers and consumers influence both the likelihood and direction of change at the global level.  First, I argue that the likelihood of market change is greatest in cases when power transitions transform relations of market power. Market change is also more likely in situations of great asymmetry between domestic and international market players. This is because asymmetries lead to higher levels of instability in market exchanges between large domestic market stakeholders and global market stakeholders. Second, I argue that the direction of market change is also determined by relative positions of market power. Asymmetric positions of power provide strategic advantages, since the dominant side’s preferences for market institutions at the global level are more easily expressed. But market players’ preferences for pricing regimes are influenced by the relative position of market power they occupy in their own market as well as relative to international market players.  3.1 Alternative Explanations The literature that looks at the impact of rising powers on the global economy has a long history. With respect to the current context, debates have focused on the potential impacts of China’s rise on the global economy. This debate has mostly been separated into two camps. In one camp, status quo scenarios predict that China will not disrupt existing international institutions (Ikenberry, 2008; Johnston, 2013; Buzan and Cox, 2013; Steinfeld, 2010; Bergsten et al., 2008). In the other, conflict scenarios foresee China’s rise, its thirst for resources and its state-centered economic development as leading to destabilizing, protectionist or mercantilist behaviour on the global stage, and thus to economic conflict (Vivoda, 2009; Economy, 2010; Calder, 1996; Kane and Serewicz, 2001; Andrews-speed et al., 2004; Bremmer, 2010).    55 I argue that the two sides in this debate have three key shortcomings. First, both view global markets as uniform. In contrast, I make the case that global markets vary significantly in their institutional structures. Second, both sides characterize global markets as liberal, whereas I will show that this is not always the best way to characterize these markets. Third, both arguments view China’s impact on the functioning of the global economy as uniform across markets, that is, its impact is either statist in nature, or it is neutral. In contrast, I will show that Chinese impacts vary significantly across markets. 3.1.1 Liberal approaches  Within Liberal approaches, there is an embedded claim that China will integrate into existing international economic institutions without effecting change. For example, the “China as a status quo power” side of the debate argues that China’s economic interdependence with the global economy will lead to a smooth integration into international market institutions and into the global economic order more generally (Steinfeld, 2010; Johnston, 2003; Ikenberry, 2008; Bergsten et al., 2008; Keohane and Nye, 1977). My argument, while remaining in close dialogue with dominant Liberal theories, departs from some of its conclusions (International Relations), which predict China’s integration into global liberal economic regimes (Ikenberry, 2008; Johnston, 2013; Buzan and Cox, 2013; Steinfeld, 2010; Bergsten et al., 2008).  Liberal theories assume that the existing economic order is liberal and market-based. I argue that this assumption ignores the political economy of global markets. In fact, many global commodity markets are oligopolistic or cartelized, and leave more room for strategic coordination than is usually acknowledged (Nolan, 2012; Fligstein, 1996; Foster and McChesney, 2012). As Strange highlighted, citing Adam Smith, we cannot ignore the “endemic tendency of business people to combine together against the consumer” (Strange, 1996, p.149). If all markets are not liberal, it follows that China’s emergence can actually lead to liberalization, or marketization of market regimes, a possibility that is overlooked by an approach that assumes that markets are liberal. A liberal approach is not well attuned to the possibility that China’s impact would weaken the positions of powerful resource companies, positions that have been established and strengthened over the latter half of the twentieth century, or in other words, that China’s impact would lead to marketization.    56  Some Liberals predict that China will integrate into global liberal market institutions based on the fact that that it has much to gain from doing so (i.e., it is not weak, and it wants to integrate). This version of the argument puts the agency more firmly in the hands of Chinese state and business leaders as making a conscious rational policy choice to integrate into existing global institutions because it is in their interest to do so (Steinfeld, 2010; Bergsten et al., 2008). But this is not necessarily how thing are seen from China’s perspective; China faces very oligopolistic global markets, which casts doubt on the assumption that China will integrate into global liberal market institutions because it feels it has much to gain from doing so, or that it will respond positively to related calls for it to liberalize its own domestic markets. Unsurprisingly, Chinese politics scholars have also debated the issue (Johnston, 2003; Johnston, 2013). This Liberal argument may hold true on some level, but not in the cases under investigation here, because it under-specifies global market dynamics. Asking China to integrate into oligopolistic global markets results in asymmetric relations of power when Chinese consumers are fragmented (which is the case in the iron ore market), and the Chinese players are very aware of this disadvantageous position. Other Liberal scholars predict that China will integrate into global liberal market institutions because it has a weak capacity for shaping global institutions or effecting global change (that is, existing economic institutions are too entrenched to be impacted in any meaningful way by emerging Chinese actors (Ruggie, 1982; Ikenberry, 2008)). This Liberal or Neoliberal approach derives its findings from the argument that the joint presence of post-hegemonic yet dominant US economic power, combined with the existing/path-dependent liberal world order, remains too powerful for any challenger to disrupt (Ruggie, 1982). The multipolar word theory also seeks to paint a picture of China supporting a transition toward a non-hegemonic world (Buzan, 2010). But this perspective usually leads its authors to agree in essence with the Liberal conclusion that a peaceful, mostly non-disruptive integration into existing institutions is most likely (Buzan and Cox, 2013).  Others argue that China will not in fact be able to impact global economic institutions because it lacks the capacity to do so, as “we are already inside them, they are not inside us” (Nolan, 2012). Aside from the fact that China’s weakness in shaping global institutions is unequal across markets, the major issue I have with this line of reasoning is that it equates weak capacity with an absence of impact. Weak capacity is definitely present in China, such   57 as when China was unable to carry out a coherent procurement policy because of its fragmented iron ore market. But where I diverge from these particular Liberal predictions is on the point that it was the very weakness of the Chinese domestic iron ore market structures that led to the fall of the benchmarking regime, which is a significant impact.  To summarize, my biggest point of disagreement with Liberal theories is not that some key Chinese actors do not want to integrate into global markets (and will do so when it is to their advantage), or even that in many ways China has little capacity to effect change. Rather, it is because 1) China is in fact having systemic impacts on global market structures; 2) those impacts are not uniform, but varied; and 3) this is sometimes happening despite the will of powerful Chinese interests, and in other cases, despite its weak capacity for steering the direction of change. In fact, the most dramatic change in global market structures – a change in the direction of marketization – resulted from a domestic position of weakness, rather than a position of strength. Interestingly, this turns Ikenberry’s 2008 liberal argument on its head. Ikenberry predicts a co-optation of emerging economies into existing global economic institutions (Ikenberry, 2008), since economic interdependence leads to integration within existing market structures (Keohane and Nye, 1977). I see a transformation of existing not-so-liberal global economic institutions as a result of China’s emergence. In order to fully understand how this is possible, I argue that we need to take into account market dynamics at both the domestic and international levels. Arguments about the likelihood or direction of systemic change cannot be made assuming uniform global liberal market institutions, even if based on detailed knowledge of Chinese domestic markets. 3.1.2 Hegemonic Stability Theory  Another main body of literature referenced in this dissertation is Hegemonic Stability Theory (HST) (Kindleberger, 1973; Krasner, 1976; Gilpin, 1987). HST, like Liberal theory, is composed of varied, and not always compatible, lines of reasoning. I would like to reflect on two particular arguments within HST.   First, HST predicts instability in the absence of an economic hegemon. Here, one contemporary application is that China’s rise, its thirst for resources and its state-centred economic development will lead to destabilizing, protectionist or mercantilist behaviour on the global stage, and thus to economic conflict in a context of declining US power (Vivoda,   58 2009; Economy, 2010; Calder, 1996; Kane and Serewicz, 2001; Andrews-speed et al., 2004; Bremmer, 2010).   This aspect of HST resonates with Realist scenarios, which are more pessimistic with regards to the likely global outcomes of China’s emergence. One claim embedded within this line of reasoning is that China has the capacity not only to fulfill procurement strategies and objectives, but also to influence global market structures to its advantage. The second embedded claim is that Chinese stakeholders actually want to effect change on global market structures in the direction of mercantilism. These arguments derive their conclusions from a positive assessment of China’s strengths, and assume that China has coherent ambitions that are rooted in a strong state. As defined (and then refuted) by Kennedy and Cheng, “a more powerful China, the thinking goes, is likely to promote a statist and more hierarchical brand of international governance that is inconsistent with the open, multilateral governance architecture that has emerged and developed since the end of World War II under American and European guidance” (Kennedy and Cheng, 2012, 9). This reading of HST is not the most accurate or useful when analyzing China’s current emergence as a key actor in global commodity markets. First, it assumes, as do Liberal arguments, that the existing economic order is liberal and market-based (the opposite of, say, state-led or oligopolistic economic orders). Second, it does not allow for the possibility that China’s impact on global resource markets lies somewhere in between two extremes – that is, either peace or conflict – and it displays an unequal pattern of integration in the global economy in different issue areas (or even within the same issue area). Here, I am channelling Garrett (2000) and his work showing unequal causal pathways and outcomes in distinct sub-components of globalization.  In addition, most revisionist power arguments do not allow for the possibility that China’s bold behaviour internationally could be the direct result of domestic fragmentation or weakness, rather than stemming from a position of strength. Indeed, “it is China’s internal fragility, not its growing strength that presents the greatest danger” (Shirk, 2007). However, whereas Shirk argues that Chinese leaders’ weakness can lead them to act brazenly in order to appear strong in front of a domestic audience, I argue that the situation can be even more complicated than that; in fact, bold actions by Chinese stakeholders sometimes result from miscalculations, or from different kinds of domestic weaknesses, such as collective action   59 problems (rather than arising from crises of legitimacy, for instance). Even in cases in which one might expect an authoritarian state to “take over” procurement policies – such as the iron ore case in which state organs (CISA, SOEs) tried to “take over” lead negotiations and keep the benchmark price – this tells us little about the likely outcome of these behaviours or about the likely direction of global market change. For this, we need to take into consideration domestic levels of fragmentation and international market characteristics.  This is where a second line of reasoning within HST is fruitful. Hegemonic Stability Theory scholars have also “posited that large dominant states possess strong preferences for free and open international exchange, and, in turn, coerce, induce or persuade other states into opening their markets to foreign trade and investment” (Krasner, 1976; Gilpin, 1987, in; Lake, 2009, 224, emphasis added). This side of the story, in line with Friedrich List’s 1841 National System of Political Economy (List, 1909), certainly seems to explain the UK’s and the US’s emergence to economic hegemony, and has to do with an emerging hegemon finding it in its interest to force open global markets over which it is dominant. Following this line of reasoning would lead one to argue that as China grows, its actors will want to coerce other global market actors towards free and open international exchange.  This reading of HST actually allows us to consider the possibility that global markets can be marketized, which is a very valuable insight. However, it also supports the argument that dominant emerging economies will have a default preference for open and free exchanges. This does not fit the China case either. First, despite China’s large – and systemically relevant – size in most global commodity markets, it has often remained in a position of weakness. Much of the Chinese domestic commodity industry is uncompetitive, very fragmented, often inefficient and wasteful, and unable to compete with the large commodity enterprises internationally. It would be a mistake to call the Chinese iron ore industry dominant globally simply because of its size.  There are two additional issues with this reading of HST. One is that China’s behaviour in different commodity markets is not uniform, even across similar cases, as we will see below. This shows either that China is not equally successful in all cases, or that it does not seek the same outcome in all cases. The important questions then become a) what explains its preferences for certain global economic regimes, and b) why and when it is successful. A second issue is that this aspect of HST theory does not delve into the   60 divergences between domestic interest groups. To say that the Chinese SOEs’ and CISA’s policy goals in the iron ore market were to “coerce, induce or persuade other states into opening their markets to foreign trade and investment” (Lake, 2009, 224) in 2010 would be false. After all, they were benefitting from their privileged access to the suppliers’ resources despite their oligopolistic situation, prior to the fall of the benchmarking regime. In other words, and despite HST’s useful insights, it is still a unitary state theory that does not allow for a divergence of interests within the Chinese state. It also does not allow for the weakest actors’ behaviour to have unwittingly caused marketization. Again, this only fully makes sense when we take into account the variation in international market structures. 3.2 Foundational Scholarship 3.2.1 Domestic vs two-level explanations  A review of major commodity markets in which China became systemically relevant suggests that dynamics are not easily explained by either exclusively inside-out or systemic explanations. There is good research on both domestic explanations and systemic explanations, but less on their interrelation (Buthe, 2014). This project seeks to go beyond the domestic/international dichotomy, and to theorize about the mutual influence of domestic-level and system-level market institutions.  The characterization of IR studies as an “inside-out” (second image) and “outside-in” (second image reversed) dichotomy really refers to the selection of different explanatory variables in response to very different kinds of research questions. “Outside-in” research delves into the role of global forces in explaining domestic outcomes, whereas “inside-out” research more often than not focuses on the role of domestic variables in explaining the foreign (economic) policies of states on the global stage (as opposed to global regimes in and of themselves). Katzenstein’s and Krasner’s work are relevant here (see Putnam, 1988, for a review). However, I am explaining more than “the determinants of the foreign economic policies of states” (Katzenstein et al., 1998), as my dependent variable, global market institutional change, is located outside of China’s national boundaries.  I propose that for those of us interested in explaining global outcomes, rather than working with Gourevitch’s dichotomy, we synthesize Waltz’s second and third images (Lake, 1997, 757, 762, in Cohen, 2008, 126). Indeed, my project seeks to leverage both the   61 second and third images at the same time, and study the interrelation between domestic and international market structures, and their impact on global outcomes.  To date, efforts that claim to go beyond the inside-out and outside-in divide in studying China’s impact on the global economy – such as work that appears in edited volumes like David Zweig & Chen Zhimin’s 2007 “China's Reforms and International Political Economy,” and Gregory Chin’s chapter in that volume, “Between Outside-in and Inside-out: The Internationalization of the Chinese State” – do not actually provide theoretical architectures, but rather empirical case studies, or narratives, but not measurable relationships and generalizable patterns.   Gourevitch’s seminal 1977 work on international trade and domestic coalitions (Gourevitch, 1977) and his 2005 book with James Shinn (Gourevitch and Shinn, 2005) also inform this research. However, whereas Gourevitch and Shinn seek to explain the variation of coordination or institutional variation, I do not so much seek to explain original institutional set-ups, but rather institutional change.  Because this type of work is trying to go beyond simple inside-out/outside-in approaches, it is broadly inspired by the legacy of Putnam’s two-level games (Putnam, 1988; Mo, 1994). In my cases however, unlike in Putnam’s cases, we are working beyond unitary state explanations because we are not dealing with two states, but rather one complex, fragmented authoritarian state and multiple global market stakeholders operating with varying degrees of coordination. The questions I ask, and the context my cases are situated in, thus warrant a different kind of theoretical framework.  In more ways than one, my project is inspired by the same motivations as Bates’ “Open-economy Politics: The Political Economy of the World Coffee Trade.” As he explains:  “Writing in 1976, Peter Katzenstein called for an end to the division between the study of international politics and domestic politics. A decade later, Stephan Haggard and Beth Simmons renewed the call. ‘We suggest,’ they wrote, ‘a research program that views international [politics] not only as the outcome of relations among states, but of the interaction between domestic and international games and coalitions that span national boundaries.’ On the one hand, the interval between these pieces underscores a lack of progress in the program set out by Katzenstein; on the other, it highlights its continued significance. In recent years, Frieden, Rogowski, Putnam, Simmons and others have contributed to this research agenda, which might be called the search for a framework for research into the politics of open economies. I, too, seek to   62 contribute to this framework and do so by focusing on the domestic politics of the international market for coffee.” (Bates, 1997, 3)  Bates argues that the policies of nations towards the international market are not defined by nations’ locations in the international environment (capabilities), but rather are defined domestically in processes structured by institutions. In (rightly) positioning himself against exclusively systemic/structural theories (such as the Hegemonic Stability Theory, more on this below), however, Bates may be straying too far into the “other camp,” that is, the exclusively “inside-out” camp. There may be a methodological reason for this. Bates studied the ebbs and flows of one international commodity market institution (the International Coffee Organization) and this allowed him to study its relationship with the varying domestic politics of producer countries. This set-up did not allow him to evaluate, however, whether there is something about this particular global market institution that has causal power as well. In this dissertation, by selecting two global markets that share some characteristics but not others, the causal power of international markets becomes apparent. In his 2009 critique of Open Economy Politics, David Lake paraphrases Keohane (Keohane, 2009) when he argues about OEP that “in its focus on micro-level and individual incentives, it fails to pay sufficient attention or, worse, is blinded to big, macro-level changes in the international economy and lacks a synthetic interpretation of change”(Lake, 2009, 231). He continues by saying that  “The issue of structural power and, often, big, macro-level change revolve around the rise and decline of major industrial and trading states. The supposed decline of the United States in the 1970s and 1980s, the rise of Japan in the same period, or the rise of China in the 21st century all unsettle international economic relations and should, by implication, be subjects of study in IPE. OEP has, to date, been largely silent on these changes, but this need not be the case.” (Lake, 2009, 233)  The aim of this dissertation is to contribute to this program of research in several ways. First, in Lake’s words, it contributes by “relaxing the small country assumption” of OEP. Second, it identifies both structural and interactive variables at play in China’s relationship with the global economy. Third, paraphrasing Lake, it seeks to answer questions about bigger, more macro-level changes in the international economy by studying the impact of one of the most disruptive events in the international political economy in the last decades: the rise of China.    63 3.2.2 International political economy of resources  In sum, there are roughly three possible levels of analysis that can help answer the question of how a dominant emerging economy impacts global markets. My argument is inspired by three specific contributions to the IPE of resources literature, which has provided theories on the resilience of global cooperative market structures. In this literature (see Table 9), some scholars adopt a systemic-level explanation for the changes in global extractive markets (Krasner, 1974), whereas others seek to explain foreign economic behaviours of states through domestic variables (Rodrik, 1982). Truly interactive inside-out endeavours are generally relatively rare (Putnam, 1988). In the literature on the IPE of resources, Bates argues that the formation of a global coffee market producers’ association is due to patterns in domestic democratic political institutions, specifically party competition. Domestic party competition variables are not adapted to the Chinese case, yet I still observe variation in patterns of engagement with global markets. I build on Putnam and Bates’ two-level approaches, and find explanatory leverage in the political economy of markets themselves at the domestic and international levels.  Table 9 - Level of Analysis in the IPE of Resources Reference Level Relevant thesis statement Cases Krasner, 1974 International Tacit assistance from multinational corporations, surfeit revenues, and highly salient shared values Oil, coffee, cocoa, copper Rodrik, 1982 Domestic Levels of coordination between government firms and banks in Japan explain Japan’s success Copper, iron ore, bauxite Bates, 1997 Two-level Strategic opportunities created by political institutions explain the formation of producers’ association Coffee  3.2.3 Comparative political economy – convergence/divergence debate  My approach to characterizing global commodity markets is germane to the literature on the persisting divergence between national systems of political economy, despite powerful globalization forces (Keohane et al., 1996; Dore et al., 1996).   64 “Our account of incentives makes us skeptical about arguments that predict convergence towards a single model. Much talk about the world economy assumes a single, optimal pattern, a single equilibrium, a unique and perfect way of combining all the ingredients of the economy, so that market competition will force all countries to converge. We are doubtful. The economy is too complex, there are too many ways of putting the pieces together.  (…) Since countries vary in their internal political dynamics - institutions, preferences, parties, and interest groups – their policy outcomes vary. And thus their corporate governance systems will differ. Change may occur, but not necessarily toward convergence on a single model of governance.” (Gourevitch et al. 2005, p.12)  Indeed, my inquiry into the variation of outcomes I observe across markets is in line with scholarship that identifies unequal patterns of integration in the global economy across issue areas (Garrett, 2000; Tiberghien, 2007). The behaviour of stakeholders from different countries in the same global commodity market varies. For instance, Japanese and Chinese actors responded differently to the fall of the benchmarking regime in the iron ore market in 2010. “Industrial countries are not converging toward a single form (Fligstein and Freeland 1995). Instead a plurality of social relations has been observed that structure markets within and across societies. These observations have challenged the neoclassical economists’ view that markets select efficient forms which, over time, converge to a single form” (Fligstein, 1996, 657). 3.2.4 IPE literature on the domestic determinants of global economic trends  The IPE literature on the domestic determinants of international economic patterns has mostly concentrated, with good reason, on the US economy’s linkages with the global economy. Jonathan Seabrooke, for instance, argues that how states treat low-income citizens and allow them to access credit and build wealth has an impact on a states’ size and the stability of its domestic pool of capital, as well as on its ability to attract capital. In turn, this affects a state’s capacity “to have a regulatory and normative influence on the character of the international financial order” (Seabrooke, 2007, p.1). Working to “reveal the domestic bases of different kinds of international financial orders,” Seabrooke is concentrating on the “everyday sources of financial power.” Such an analysis would yield interesting results with regards to the Chinese financial system and its capacity to influence the global financial system. However, this chapter concentrates on the extractive industries, and therefore is interested in the behaviour of larger relevant market stakeholders in this industry. My   65 approach is very relevant to Seabrooke’s approach, however, as I recognize that the domestic nature of market systems in a rising power such as China has a determining impact on the future of global markets.   Andrew Walter (Walter, 1993) has also sought to make the link between particular international patterns of systemic change and domestic political economic variables. Walter points to the particularly fragmented nature of the American financial system in the early twentieth century (large number of small banks, decentralized Federal Reserve System, decentralization of power in the private banking system), which he contrasts to the oligopolistic structure of the European system, as a factor that led to the Great Depression in 1929 (and in turn, to the emergence of the interventionist state, or, as Ruggie has outlined, the rise of “embedded liberalism” at a global level after WWII) (Walter, 1993, p.144). This article is in line with Walter’s analysis since I seek to illuminate relevant domestic dynamics at play in explaining outcomes at the global level.  Few authors have tried to identify specific domestic patterns of Chinese political economy as potential explanatory variables for change at the international level. Part of the reason is that few have identified specific institutional change at the global level that has resulted from China’s rise. McNally (McNally, 2012) provides one of the most interesting and nuanced explorations (and goes beyond dichotomous debates) of the ways China’s domestic hybrid capitalist system (and its reliance on informal networks and a strong role for the state) can impact the international economy. However, his efforts stop short of identifying specific causal mechanisms that would link specific domestic characteristics to specific systemic outcomes at the global level. McNally reserves judgment as to what kinds of impacts China will have on the global economy, beyond pointing to the fact that the domestic characteristics he identifies as unique to the Chinese domestic economy will most certainly be reflected eventually in global market institutions, and will challenge the Anglo-American system of capitalism.  3.3 The Key Players 3.3.1 Beyond the unitary actor My approach stands in contrast with endeavours that conceptualize the state as a unitary actor. Ikenberry argues that “the state – or the state elite – is the crucial actor within the adjustment process. The state may therefore be conceived of as a strategist in the context   66 of domestic and international structures and constraints” (Ikenberry, 1986, 54). “Studies purporting to develop an understanding of the global economy have generally analyzed just one, or perhaps two, types of agents, such as firms or industrial sectors. Other agents (such as states, labour organizations, and global regulatory bodies) and non-human intermediaries (for example port facilities, telecommunication infrastructure, policy documents and manuals) have been neglected or even dismissed as irrelevant and anachronistic” (Dicken et al., 2001, 91). Indeed, “(...) development is no longer paradigmatically determined by the presence/absence of national/international regulation, but by the collective strategic behavior of large (‘related’) actors along the value chains of today’s global industries” (Petkova, 2006, 315). The unitary state approach is useful given certain research questions, but it is at odds with empirical evidence of Chinese behaviour, at least in the cases being studied here. My approach is very much in line with Katzenstein’s analysis. His work draws “attention to the domestic political and institutional influences on a state’s policy behavior in the world economy. (...) Katzenstein’s aim was to open up the unitary state – to complement the systemic (‘outside-in’) level of analysis of realism with the domestic (‘inside-out’) level of analysis more characteristic of comparative politics” (Cohen, 2007, 204). 3.3.2 The Chinese state My research is in line with the fragmented authoritarianism literature in Chinese politics (Lieberthal and Lampton, 1992; Lieberthal and Oksengerg, 1986; Oksenberg and Lieberthal, 1988; Shambaugh, 2011; Paltiel, 2010; Foot and Walter, 2010; Mertha, 2009). There is a fragmentation of economic policy making within the Chinese state, the contours of which vary depending on the policy area (Shambaugh, 2011; Foot and Walter, 2010). It follows (and this is verified in my cases) that there is no coherent procurement behaviour across all global commodity markets; in other words, China is not behaving “as one.” Institutional structures vary across markets domestically, and the specific relations of power among key domestic market stakeholders give distinctive hues to overall Chinese procurement behaviour in specific markets (Gourevitch, 1977; Gourevitch, 1986; Gourevitch and Shinn, 2005; Simmons, 1994).  China’s behaviour in given markets is not exclusively dependent on the state’s position in the international system, or the distribution of capabilities (Waltz, 1979), nor is it homogenous across markets. Tiberghien (2012), Kennedy and Cheng (2012) and Wang and   67 Zheng (2008) are interesting recent attempts at taking a more multidimensional approach to China’s interactions with the global economy. Beyond international and domestic market structures, this research will consider the behaviour of small firms, large firms, and state organs (whether industry associations or relevant ministries) as relevant actors. As we will see below, state organs play a decisive role in China’s interaction with global commodity markets. As agents that are often intent on coordinating the behaviour of other Chinese actors, their capacity for successfully enhancing or hampering coordination has direct impacts on the overall Chinese patterns of procurement internationally. Chinese stakeholders’ behaviour varies in multiple ways, and this includes key actors’ domestic balance of power – whether between industry associations and small private enterprises, or between SOEs and regulatory agencies – varying across industries. As a consequence, patterns of interaction between domestic and international market stakeholders also vary.  With regards to resource procurement, Chinese planners have been working within the “Two Markets, Two Resources” framework (	) (Xiao and Gao, 2009; Gu and Wang, 2005). Coined in the 1990s, conceptually prior to, and broader than the Going-Out strategy, the framework emphasizes the need to leverage both domestic supply sources (this includes increased investment in prospection and mining) and international supply sources (through the whole array of options, including foreign acquisition, investment, and long-term and short-term purchasing contracts). At no point in this dissertation do I conceive of a state as a unitary actor. However, two caveats need to be made regarding this. First, I do not exclude the possibility that despite a non-unitary state approach, one may come to the conclusion that certain patterns of behaviour exhibit clustering characteristics that allow us to identify certain patterns of behaviour as typical of “Chinese” or “Japanese” firms in a particular market at a particular point in time. Second, for the sake of readability, I will sometimes use the expression “Chinese actors” or “Chinese behaviour,” although by that I mean the sum of all actors’ behaviour, or rather, the discernible patterns that emerge from their aggregate behaviour.    68 3.4 Methodology  I combine a two-level political economy approach with the power of small-n, comparative case studies, and use process-tracing to unveil the causal pathways that lead to systemic market changes at the global level. I trace the origins of change that I observe at the global level back to the domestic political economic roots of Chinese market structures. As Aoki argues,  “institutional arrangements can be diverse across economies even if they are exposed to the same technological knowledge and are linked through the same markets. Thus we need to rely on comparative and historical information to understand why particular institutional arrangements have evolved in one economy but not in others. By this we imply that an institutional analysis must be also comparative and historical, and thus we have hope to provide the groundwork form comparative institutional analysis (CIA).” (Aoki, 2001, p. 3) My approach goes beyond endeavours methodologically designed to compare national (unitary state) impacts on the global economy (Krasner, 1976; Gowa and Mansfield, 1993; Ikenberry, 1986; Gilpin, 1987). Indeed, my inquiry into the variation of outcomes I observe across markets is in line with scholarship that identifies unequal patterns of integration in the global economy across issue areas (Garrett, 2000; Tiberghien, 2007, 2012; Wang and Zheng, 2008; Foot and Walter, 2010), as well as with scholarship that assumes a fragmentation of policy making within China (Lieberthal and Lampton, 1992; Lieberthal and Oksenberg, 1986; Oksenberg and Lieberthal, 1988; Shambaugh, 2011; Paltiel, 2010; Foot and Walter, 2010).  3.4.1 Case selection  The universe of cases I am referring to is: a systemically relevant emerging economy’s impact on a specific global commodity market structure. The rise of the US after World War II and its impact on the global oil market would be an example.19 In other words, I consider  “systemically relevant emerging economy” - “commodity market pricing regime” dyads, and argue that one can use the concepts of “major power transition” and “systemic change” within a given market system (as opposed to only within the global                                                  19 Too few authors have made the comparison between the US’s and China’s emergences as dominant global powers. For an exception, see Foot and Walder, 2010.   69 economy at large). Japan taking over from Germany and having a systemic impact on the global iron ore market starting from the late 1960s is another example.  Table 10 - Cases Cases (across markets) Case (across time) Iron Ore 1. China’s impact on the global iron ore market (pricing) – 2000s 4. Japan’s impact on the global iron ore market (pricing and shipping regimes) – 1960s   2. China’s impact on the global iron ore market (shipping) – 2000s Potash 3. China’s impact on the global potash market (pricing regime) – 2000s  Uranium  5. China’s impact on the global uranium market (market shape) – 2000s  My comparative case selection strategy is based on Mill’s method of difference, as I compare divergent institutional outcomes within global markets otherwise structured similarly (see Table 10). My two principal cases are the markets of iron ore and potash. In order to gain comparative leverage, I study China’s impact on two “within-iron-ore-market” case studies: the global iron ore pricing system, and the global iron ore shipping pricing system. This within-market case selection allows me to control for multiple variables inside the iron ore case, and further leverage the process-tracing of the two stories. Then, still working within the same market, I conduct a plausibility probe into the generalizability of my findings over time and space, but as it applies to another systemically significant rising economy decades earlier: Japan’s emergence and impact on the global iron ore market as it became the world’s largest consumer of the commodity in the 1960s. To conclude, I evaluate the generalizability of my findings to other commodity markets by also evaluating the applicability of my theoretical model to the global uranium market.     70 Table 11 - Summary of Similarities across the Global Potash and Iron Ore Markets Prior to China's Emergence ! Iron%Ore% Potash%Market%shape% %Nature!of!the!commodity! Extractive, essential to economic development, low substitution potential, high barriers to entry Extractive, essential to economic development, low substitution potential, high barriers to entry Geographical!concentration! High. 65% of global exports from Brazil and Australia (2000) (one developed Asia-Pacific country and one BRIC country) High. 93% of global production in Russia, Belarus and Canada for potash (2000), among which are one developed Asia-Pacific country and one BRIC country Market%Governance%Pricing!regime! Long-term contracts and annual benchmarking pricing system Long-term contracts and annual benchmarking pricing system Chinese%dependence%on%trade%Import!dependence! High level of import dependence (58% in 2011) Essential to China’s economic development High level of import dependence (50% in 2011) Essential to China’s food security  The global iron ore and potash markets share many characteristics. First, both markets are bulk commodities, extractive markets with capital-intensive requirements, and have long-lead time horizons and high barriers to entry (Komesaroff, 2013). Iron ore and potash are both essential to economic and agricultural activity, respectively, as they have very low potential, if any at all, for substitution. In addition, both the global iron ore and potash markets have a very high concentration of reserves in just two or three countries. This was true at the time of China’s emergence as the dominant consumer. In 2000, 65% of iron ore exports came from Brazil and Australia20, and 93% of potash exports came from Canada, Belarus and Russia.21 Both markets’ suppliers include one dominant Asia-Pacific developed economy (Australia for iron ore, and Canada for potash) and one dominant BRICs economy (Brazil for iron ore, and Russia for potash). Before China’s emergence, at the beginning of                                                  20 UN COMTRADE. 21 USGS Minerals Yearbook, 2003.   71 the 2000s, both markets were controlled by a few large companies (Vale, BHP Billiton and Rio Tinto for iron ore; Potash Corp., Mosaic, Ukalkali and Belarukali for potash). Furthermore, both markets’ pricing regimes were comparable before China’s rise, as both had a long history of closed-door annual benchmarking pricing negotiations, long-term contracts, low prices and low volatility (see Table 11).  Figure 8 - Chinese Import Dependence in the Markets of Iron Ore, Potash and Uranium  Source: Ministry of Land and Resources, 2012; World Nuclear Association; China Customs Department. Author’s calculations.22  Finally, China’s positioning towards both international markets is comparable inasmuch as its level of import dependence is very high in all cases (see Figure 8). When we look at China’s import dependency ratios in the three global commodity markets studied here, we see that the three ratios were above 50% during the years for which data is available (apart from a short dip under 50% for potash in 2009).  Despite the striking similarities between the global iron ore and potash markets, China’s impacts on these markets have diverged significantly. As we will see below, part of the reason is that the Chinese domestic political economic conditions in both markets exhibit fundamental divergences, especially at the interface with the global economy.  Beyond the key iron ore and potash cases, I investigate China’s impact on two additional cases: the iron ore bulk shipping case, and the uranium case. Therefore, this research design makes full use of the comparative approach by increasing variation across                                                  22 Potash and iron ore, as calculated by the Ministry of Land and Resources; uranium, calculated as share of consumption minus production for uranium, from World Nuclear Association; China Customs Department data 0!20!40!60!80!100!Dependence'on'Foreign'Trade'Iron!ore!Potash!Uranium!  72 independent and dependent variables (see Table 12). Indeed, I investigate the impact of the emergence of Japan, another dominant economy, on the global iron ore market decades prior to China’s emergence. The results constitute an early and encouraging test of the generalizability of my argument beyond the iron ore and potash cases, as well as beyond China.   Table 12 - Asymmetries and Cases International market (Producers) ! Domestic market (Consumers) " Fragmentation Concentration Fragmentation 1 Symmetry   2 Asymmetry in favour of producers  Case: China/iron ore pricing Concentration 3 Asymmetry in favour of consumers  Case: China/uranium market, Japan/iron ore pricing and shipping  4 Symmetry   Cases: China/iron ore shipping, China/potash pricing   This dissertation makes use of mixed methodologies, including an emphasis on interviews (119 interviews, 98 interviewees, 74 work units, see Figures 9-12),23 narrative and coding-based analysis, document analysis and comparative/descriptive statistical analysis. In addition to establishing correlational inferences across cases, thick case studies and process-tracing allowed me to establish the mechanisms and the sequencing at play. As I go through each case, I pay particular attention to the relationship between an emerging economy’s preferences and global actors’ preferences, capacity to coordinate behaviour, as well as the amplitude and direction of change at the global level.                                                  23 Although I have conducted 127 interviews in the course of this research, only 119 of them yielded material worth considering as “relevant content.” I have not coded the remaining eight interviews, hence the working total of 119. My interviewees were selected using a mixture of diversity and snowball sampling approaches (I conducted a variety of interviews with individuals from different work units, while leveraging growing networks to deepen my knowledge of a particular sub-group’s perspectives). My interviews were semi-structured. See Appendix 1 for more details.   73 Figure 9 - Nationality of interviewees  Figure 10 - Work Units, by Interviewee  43!52!0! 10! 20! 30! 40! 50! 60!Chinese!Other!Nationality'(N=98,'number'of'interviewees)'13!6!14!10!7!1!7!17!24!7!13!0! 5! 10! 15! 20! 25! 30!Academia!Central!government!agency!Consulting!Firm!Diplomacy!Industry!association!International!Organization!Media!Ministry!Private!iirm!SOE!Think!Tank/Research!Institute!Interviewee'work'unit'(N=119'interviews)'  74 Figure 11 – Nature of Ownership, by Work Unit   Figure 12 - Language Used, by Interview  3.4.2 Explanatory variables at both levels are necessary to explain outcomes It is the contention of this dissertation that to fully understand and explain systemic change at the global level, power relations in markets at both levels need to be understood. This dissertation’s case selection allows us to investigate procurement behaviours and pricing regime changes in two sub-sets of cases.  0! 5! 10! 15! 20! 25! 30! 35! 40! 45! 50! 55! 60!Private!sector!Public!sector!Nature'of'ownership'(N=74'work'units)'4!9!51!55!0! 10! 20! 30! 40! 50! 60!French!Chinese!and!English!Chinese!English!Language'used'in'interview'(N=119'interviews)'  75  First, there are cases in which the international markets exhibited very high levels of concentration, but where the domestic levels of concentration varied. Here, we compare the pricing regime changes in the iron ore and potash markets. When comparing cases in which the concentration levels were high at the global level prior to China’s emergence, China’s domestic-level market concentration becomes crucial. We thus delve into the Chinese domestic situation in more detail, while holding the global market side of the equation constant. I then investigate, in a comparative fashion, exactly how domestic market characteristics have shaped Chinese procurement behaviour. The two key cases that form the basis of this dissertation are ones in which the Chinese market stakeholders faced highly concentrated international suppliers, but also in which domestic markets exhibited low (iron ore pricing regime) or high levels of concentration (potash pricing regime, iron ore shipping regime). This case selection strategy allows us to show that international-level variables are insufficient to explain outcomes. In such cases, distinct dynamics across the Chinese domestic commodity markets of iron ore and potash result in distinct Chinese procurement behaviours and distinct two-level power asymmetries, ultimately resulting in distinct impacts on global pricing regimes. In sum, I argue here that different domestic dynamics in the Chinese iron ore and potash markets resulted in different aggregate Chinese procurement behaviours, and, given the international context, ultimately resulted in different impacts on global pricing regimes. The fragmentation of Chinese domestic market structures has led to regime change and marketization in the iron ore case, whereas a higher level of coordination in the Chinese domestic potash market has led to higher resilience in the international potash market pricing institutions. Second, there are cases in which the domestic levels of concentration are high, but the international market concentration levels vary. Here, we concentrate on two types of comparisons. The first is a comparison between two within-market cases: China’s impact on the iron ore bulk shipping pricing regime; and Japan’s impact on the iron ore pricing and bulk shipping pricing regime decades prior to China’s emergence. This within-market, across time and country comparison allows us to compare cases in which the domestic level exhibited high levels of concentration and coordination, but concentration and coordination levels internationally varied. The questions of interest here then become-: How exactly are   76 Chinese domestic commodity markets structured differently? The fact that the patterns of change differ between the two cases is an indication that domestic-level variables are insufficient to explain outcomes at the global level. The other type of comparison is between China’s impact on the global potash market and its impact on the global uranium market. Here again, similar levels of domestic concentration but different dynamics of change at the international level point to the importance of international-level variables in explaining outcomes. It becomes clear that it is necessary to pay particular attention to Chinese domestic market structures and coordination levels in order to understand the global market outcomes that result from China’s emergence. Chinese domestic market dynamics have become a determining feature of the global economy. In other words, this dissertation’s case selection allows us to investigate, at once and in turn, the importance of variation in market characteristics at both levels.  The goal here is to combine the efforts of Putnam’s two-level game with a political economy approach (essentially reversing Rogowski’s 1987 (class cleavages) and Frieden’s 1991 (industry cleavages) outside-in approaches to the problem of exposure to international trade). I am proposing an inside-out, interactive model, which theorizes about two-level strategic action within each market.   It is worth restating the two research questions at the origin of this paper. The first question is about the likelihood of change: Why have some international markets for important raw materials undergone fundamental change in the way they operate as a result of China’s emergence, while other such markets have been more resilient to change? The second question is about direction of change: Why have some global markets seen change in the direction of marketization as a result of the emergence of a large consumer, whereas others have seen change in the opposite direction? Under what conditions will dominant market players push for change in one direction rather than another at the international level? 3.5 Argument  Domestic-level and international market variables combine to create the conditions for institutional change at the global level. I trace key variances in China’s effect on global markets to the interaction of China’s domestic industrial structures and the pre-existing structures of global commodity markets. The structure of key industries within China varies: some are concentrated, some are fragmented, some are very sensitive to price signals, and   77 others less so. Likewise, the significant structural variance found among global markets for various commodities pre-dates China’s emergence as a dominant global consumer in the twenty-first century.  I argue that asymmetries between global and domestic market structures profoundly affect incentives and domestic and global market players’ capacity to act, and create the conditions for systemic change (see Table 13 below).    More specifically, in response to the first question about the likelihood of change, I argue that power transitions that entail a profound change in consumer-producer power relations are more likely to be disruptive and lead to institutional change at the global level than are power transitions that entail a continuation of two-level power relations. In addition, transitions towards asymmetric positions of power (Quadrants 2 and 3) tend to be more unstable than transitions towards symmetric positions of power. Transformations in market power relations between consumers and suppliers increase the likelihood of institutional change in global markets.  Table 13 - Two-Level Asymmetries in Domestic and International Markets, and Ensuing Outcomes International market (Producers) ! Domestic market (Consumers) " Fragmentation Concentration Fragmentation Symmetry Low likelihood of change Outcome: competition/consolidation incremental change   Asymmetry High likelihood of change Dominant preferences: Producers Outcome: Dominant actions on behalf of producers   Iron ore pricing in 2000s (China) Concentration Asymmetry High likelihood of change Dominant preferences: Consumers Outcome: Dominant actions on behalf of consumers  Iron ore pricing in 1960s (Japan) Uranium 2000s (China) Symmetry Medium likelihood of change Outcome: Clash, bargaining incremental change   Iron ore shipping in 2000s (China) Potash 2000s (China)   In response to the second question about the direction of change, I argue that market power asymmetries give dominant players the upper hand in influencing global market institutions, such as pricing regimes. Market players’ preferences for pricing regimes are   78 influenced by the relative position of market power they occupy in their own market as well as their position relative to international market players. Price trends influence market stakeholders’ preferences for global pricing regimes, but they cannot fully explain the direction of change.   The causal mechanism through which change is effected is the differential coordination capacity of actors at each level, or the differential ability of actors to solve their collective action problems. It includes the coordination capacity between public and private actors. It is through coordination, or lack thereof, that market actors can exercise market power and get what they want. Maintaining pricing regimes, and especially changing pricing regimes, requires coordination (see Hall and Gingerich, 2004; Hall and Soskice, 2001; Elinor Ostrom, 1997).   3.5.1 Argument: Likelihood of change  The argument about the likelihood of market institutional change – that is, change resulting from the power transition from one dominant consumer to another – rests on the notion that asymmetries engender instability. There are at least two situations in which power transitions increase the likelihood of change in global market institutions:  1) In power transitions that entail a profound change in consumer-producer power relations are more unstable, i.e. when the emerging power causes a movement from one two-level power position to another (from one quadrant to another). These power transitions are more likely to be disruptive than power transitions that entail a continuation of two-level power relations.  2) In power transitions that entail a change towards asymmetric positions of power tend to be more unstable than those that lead to symmetric positions of power, i.e. transitions toward Quadrants 2 and 3. There is an exception to this: power transitions that entail a continuation of previous two-level relations of market power (remain within the same quadrant), whether they are situated in an asymmetric quadrant or not, tend to be more stable.  In other words, a power transition that is accompanied by a movement from one quadrant to another (say a movement from the lower left to the upper right quadrant) is more unstable than those who do not involve a movement from one quadrant to another. The greater the change in relative two-level positions of power, the greater the resulting   79 instability of global market institutions. As a rule, asymmetric quadrants also tend to be more unstable. The exception is that the continuation of two-level market power dynamics (power transition that remains within one quadrant) can lead to the stability of global market institutions, at least in the short term, regardless of the quadrant.   The China - iron ore benchmark pricing case is one in which a fragmented and weakly coordinated new consumer emerged to face a global market with very high levels of concentration and coordination (Quadrant 2). Prior to China’s emergence, the dominant consumer, Japan, had shown high levels of concentration and coordination. In fact, the iron ore benchmark pricing regime was established at a time when the global iron ore market players were much more fragmented than they are today. This power transition from Japan to China as the world’s number one consumer constitutes a movement from Quadrant 3 to Quadrant 2, from one asymmetric position of power to another. Whereas Japanese consumers’ preferences of held more sway during Japan’s tenure, the preferences of the global iron ore suppliers became dominant as China’s fragmented domestic consumers emerged as the largest in the world. We can thus understand why that transition had the potential to be disruptive. It created instability and opened the way for systemic change at the global level. The collective action failure and fragmented procurement behaviour on behalf of the Chinese consumers gave the global iron ore suppliers the opportunity to take decisive action and usher in a more volatile pricing regime.   In cases of a transition towards a symmetric position between domestic and international market power structures (Quadrant 4), change is still possible. However, it is less likely to be sudden (more likely to be incremental), because the consumers and producers are both concentrated. The outcome is more likely to be the result of an alignment of preferences, conflict or strategic bargaining. Yet, as this combination involves equally powerful positions of power in both market realms, it is also the least predictable because there is the possibility of deadlock or conflict between dominant players. Indeed, in the case of China’s impact on the global iron ore shipping regime, bold moves from the global iron ore shipping firms were met by a concentrated and coordinated Chinese domestic industry. This allowed the Chinese actors to better coordinate behaviour across multiple platforms and to counter the global firms’ strategy. In that case, the deadlock was resolved through bargaining. The coordination of Chinese market actors’ behaviour allowed them more power   80 over the outcome of the clash with international stakeholders. The result in that case was incremental change in favour of the Chinese consumers.   In the case of a transition towards Quadrant 3, where the international market is fragmented and the emerging economy’s domestic market actors are highly concentrated and coordinated, change is again likely, but this time around, there is the greater allowance for the consumers’ agency and expression of their preferences which is greatest. Lastly, changes towards Quadrant 1 are relatively rare, and I could not immediately find a suitable case, given the high levels of concentration in many global and Chinese commodity markets. The copper market is a potential candidate for future enquiry.  In summary, power transitions that cause a movement between one quadrant and another are more likely to be unstable, but this is especially true of transitions towards asymmetric quadrants. Indeed, transitions towards both extremes of the graph (in blue), where the asymmetry is greatest between domestic and international market structures, tend to bring with them a higher likelihood of global market institutional change. This may be because profound asymmetries tend to be more unstable, especially between systemically relevant domestic and global market stakeholders.  3.5.2 Argument: Direction of change The second part of my argument seeks to explain variation in the direction of change in market institutions as a result of China’s emergence. At the core of the argument lies the notion that asymmetric positions of power provide strategic advantages, and that the dominant side’s preferences for market institutions at the global level are more easily expressed. This insight may seem self-evident, but it only fully makes sense when grounded in a relational, multilevel, definition of “dominance” or “market power.”  Market power is defined here as relational, at both the domestic and international levels. I adopt Susan Strange’s definition of power: “Power is simply the ability of a person or group of persons so to affect outcomes that their preferences take precedence over the preferences of others” (Strange, 1996, p. 17). Pricing regimes have distributional consequences: there are losers and winners during pricing regime changes. This leads to firm preferences for asymmetric, influential positions of power. I argue that positions of asymmetry provide the dominant side with more leeway to effect change in its preferred direction. To achieve this position of power, market actors try to consolidate and coordinate   81 behaviour, and a struggle for dominance within specific global markets ensues. Here I build on Nolan (2012), Fligstein (1996) and Gilpin (1977). In this situation, there are major concentrations of economic power, and many global commodity markets are oligopolistic or cartelized and they leave more room for strategic coordination than is usually acknowledged in the literature. Dominance can only be established through an understanding of market players’ domestic and international market power, which includes their share of consumption/production and their capacity to coordinate behaviour. Dominance is precious because it enables the expression of one’s favoured market institutions at the global level. But what explains preferences for one type of pricing regime over another (see discussion in Chapter 2)? Market players’ preferences for pricing regimes are influenced by their relative positions of market power. The relative positions of market power that actors occupy in their own market, as well as relative to international market players, have an impact on their particular preferences for pricing regimes at the global level. The importance of these dynamics explains the divergence from strictly “price-determined” preferences for global pricing regimes (see Chapter 2). Trying to model behaviour based strictly on the logic of price levels does not fully capture the complexity of preferences on the ground. The “price determined” logic would predict that large consumers (producers) have an interest in marketization (demarketization) of pricing regimes when prices are dropping, and an interest in demarketization (marketization) when prices are rising. In some cases, price trends do provide the strongest explanation for pricing regime preferences. For instance, global suppliers’ preferences for marketization of the iron ore market are in line with expectations of preferences, given the expectations of rising iron ore prices. But in the majority of the cases studied here, they do not. I will mention only a few of the limitations of this rationale. The first is that consumers in the same market and facing the same price trends do not always have the same preferences. The second is that within a domestic market, different types of consumers do not always have the same preferences. The final limitation is that consumers or producers sometimes simply do not follow this logic.   Empirical analysis shows that price is not the only variable influencing preferences and behaviour for particular types of global pricing regimes. I argue that a more complete answer lies in the joint consideration of price levels and relative positions of market power. This provides us with critical clues as to the preferences and behaviour of key actors.   82 Preferences for pricing regimes are complex, but the one key variable that interferes with straight “price-determined preferences” for global pricing regimes is the position of market power of given firms in their own market. A domestic position of power occupied by key resource importers with rent-extracting privileges provides importers with huge disincentives to push for the marketization of pricing regimes, regardless of price levels. Domestic market stakeholders’ preference for the marketization of pricing regimes, on the other hand, is more likely to occur when they are occupying a weak position of market power relative to suppliers, as well as relative to their own peers. At the international level, the opposite is true. International market stakeholders’ preference for the marketization of pricing regimes is most likely to occur when they are occupying a dominant position of market power relative to the consumers, as well as relative to their own peers. For instance, in the case of Japan’s impact on global iron ore pricing regimes in the second half of the twentieth century (Quadrant 3), we saw a successful coordination of the domestic players and interface with global market stakeholders. We also saw a more fragmented global market. This situation of dominance afforded the large Japanese firms (consumers) more leeway to express their preferences for global pricing regimes. The Japanese tenure as the dominant consumer saw the reinforcement and strengthening of a strategic/negotiated pricing regime. This systemic movement towards the demarketization of the iron ore pricing regime was in Japan’s favour. Japan’s emergence as a systemically relevant resource consumer in the 1960s illustrates the decisive impact a dominant, coordinated consumer can have on global markets.   In the case of China’s impact on global iron ore pricing regimes in the 2000s (Quadrant 2), the global producers had the upper hand, and in the end, we did see a marketization of pricing regimes. However, the preferences of domestic Chinese consumers were unexpected. The hope for the continuation of powerful rent-extracting privileges in the Chinese domestic iron ore market prior to the fall of the benchmarking regime explains large Chinese iron ore importers’ preferences for the iron ore benchmarking regime despite falling prices in 2009.  In the potash case (Quadrant 4, a situation of symmetry), dynamics are subtler. Because of the licensing system, the top Chinese potash importers extract rents by reselling internationally-acquired potash to domestic distributors. In that case, the capacity for   83 domestic purchasers and farmers to pay for potash (including consideration of state subsidies here) becomes as important as the relative international price levels in determining importers’ preferences. Indeed, falling potash prices should have been the window of opportunity for Chinese potash importers to push for a marketization of the global potash pricing regime, but we have not seen this in any conclusive sense. In turn, the fall of the Russia-Belarus potash cartel occurred in the middle of declining price trends, and this obviously put the suppliers in a disadvantageous position towards the Chinese importers. Prices have indeed continued to fall ever since. The general implications of this theoretical model can be found in Table 14 below.       Table 14 - Implications of Theoretical Model Implications of my theoretical model 1. Systemic change in market institutions is most likely to happen as a result of transformations in two-level power relations between consumers and producers in times of power transition. 2. Changes towards Quadrants 2 and 3 (asymmetric market power relations) are most disruptive. 3. Leading consumer (producer) firms behave differently depending on the global (domestic) market structures.  4. The dominant side’s preferences for market institutions at the global level are more easily expressed in situations of asymmetry (Quadrants 2 or 3). 5. The potential for rent extraction blunts the responsiveness of market actors to price signals. 6. International (domestic) market stakeholders’ preference for the marketization of pricing regimes is most likely to occur when they are occupying a dominant (weak) position of market power relative to the consumers (producers), as well as relative to their own peers, and vice versa. 3.6 Independent Variables: Global and Domestic Concentration Levels    My independent variables are the market structures as determinants of coordination capacity at the domestic and international levels. The measure I select to compare market structures at the domestic and international levels is the level of concentration (or fragmentation) (most importantly, at the export/import interface) (Rodrik, 1982; Krasner,   84 1974)24 (see Table 16 and 17). Concentration levels matter since they play a large role in determining market players’ coordination capacity, and thus market power and capacity to effect change in the preferred direction.                                                   24 Rodrik defined global commodity markets along four variables, including “degree of concentration” and “trade and price formation mode.” His argument is directly relevant to my own: he argues that the levels of coordination between the government, firms and banks in Japan explain Japan’s capacity to procure iron ore at more favourable terms than the US on the global market.     85 36  Table 15 - Global Commodity Market Concentration Before and After China's Emergence as Dominant Consumer Commodity Oil25  Copper Coal Uranium Iron ore  Soy Potash Chinese imports as share of global exports 2000 3.6%26  2013 10%27  2000 14.4%28 2012 35.5%* 2000 0.3%* 2012 24%* 2001 1.2%*  2012 24%* 2003  26%29 2010 65%30 2000 19.2%31 2013 66%32 1998  24%33 2012 37%34 Shape of market Geographical dispersion (% of exports in 4 top producing countries) 37.6%35 38.5%36 64.2%* 59%* 58.6%* 73%* 75%* 83%* 78.2%37 73%38 97.6%39 92.1%40 95.5%* 87%* Firm concentration (% of exports by 4 largest firms) 27%41 39.4%42 40%43 36%44 55%45 45%46 62%47 59%*48 70%49 83%50 76%51 80%52 83%53 79%54                                                  25 China has not reached dominance in the global oil market as per this dissertation’s measure. Yet, some would argue that 10% of global exports is a sign of systemic significance in the oil market.  26 US Department of Energy 27 Barclays Capital, (Areddy, 2009) 28 UN COMTRADE (*) 29 (Ericsson, 2004) 30 2012, UN COMTRADE 31 2000-2001, (Oilseeds: World Markets and Trade, 2003)  32 Soy imports as share of world trade (Oliveira and Schneider, 2014) 33 China Chemical Yearbooks, CISIA statistics (in 1997, China consumed 18% of global potash exports, in 1993, 10%) 34 China Chemical Yearbooks, CISIA statistics. Here, UNCTAD gives a much lower share of global imports by China in 2012, at 17%, one percent lower than in 2000, at 18%.   35 US Department of Energy 36 US Department of Energy, 2010 37 USGS (Iron ore, 2003) 38 http://www.issb.co.uk/global.html#IronOre  39 US Department of Agriculture 40 US Department of Agriculture 41 1998. Top four companies http://www.gravmag.com/companies.shtml#producers, share of global production http://www.worldenergyoutlook.org/media/weowebsite/2008-1994/weo1998.pdf. 42 The Street (2012), Top oil producing companies, http://www.thestreet.com/story/12065842/3/top-10-oil-producing-companies-of-2012.html 43 Top five companies, share of total production. http://investors.boliden.com/files/press/boliden/Boliden_Annualreport2002_en.pdf 44 Top five firms, copper output, 2012. http://www.mining.com/top-10-copper-companies-in-2012-85961/ 45 Top five suppliers account for 55% of the seaborne trade in metallurgical coal (BHP Billiton Mitsubishi Alliance, Fording, Rio Tinto, Anglo American and MIM) (Kirkby, 2001)  46 Top four firms (BHP Billiton, Xstrata, Anglo American, and Adarao), share of global trade in 2010 (Schernikau, 2010) 47 Of global production (2006) (List, 1909) The top four companies produced 63% of exports in 2009 (UNCOMTRADE). 48 The top four exporting companies are: KazAtomProm, Cameco, Areva, ARMZ – Uranium One. World Nuclear Association, 2013 49 Almost 70% for the world’s top three companies, Vale, Rio and BHPB, (Ericsson, 2004). 50 2007, Seaborne iron ore production, Baffinland Iron Mines Corporation 51 It very difficult to find export share by company in the soy market. Here this figure represents the share of the top five soybean crushers in 1997 in the US market: ADM, Cargill, Bunge, Ag Processing and Central Soya. At the time, the US controlled around 70% of world exports of soy (Freeman, 1995). 52 This number represents the share of the global grains trade controlled by the top four grains companies: ADM, Bunge, Cargill and Louis Dreyfus (ABCD: 4 international grain traders, 2009). 53 2006, UBS Investment Research (Canpotex counted as one firm) 54 2009, UBS Investment Research (Canpotex counted as one firm)   86 Discussion   Because global markets are not customarily seen as distinct political economic structures, especially within the same industry, it is difficult to find comparative data across markets, other than price (and even prices are not always compiled in ways that make it easy to compare across markets). In Table 15 (above) I provide an overview of concentration levels for seven global commodity markets before and after China’s emergence as the dominant consumer.55 One of the systematic trends across the seven commodities is the rapidity with which China’s imports as a share of global exports grew. China’s uranium imports as share of global exports went from 1.2% in 2001 to 24% in 2012. In the iron ore market, China’ share of global exports went from 26% to 65%, and in the soy market, from a little less than 20% to 66% during that same time period. China’s rise in consumption has been extremely rapid. This confirms the contention that the impact that China had on global commodity markets in the first decade of the 2000s was large enough to be systemic and to have impacts at the global level. It was also sudden enough to act as a shock concentrated in time and space, or analogous to a quasi-experiment.  Looking at the comparative table, another point that is immediately evident is the high level of geographic and industry concentration of global commodity markets, before and after China’s emergence as a systemically dominant consumer in the early twenty-first century. Some would point out that that is the case with the global trade of many other industries as well (Nolan, 2012). Four countries, or even just four companies, frequently control more than 50% of global trade in commodities. Table 14 above is ordered from the least concentrated to the most concentrated market. Apart from the oil market, in all other markets, the four top countries control more than 55% of exports. Looking at firm concentration, numbers are still high, but with a little more of a gradual progression between the least to the most concentrated. Indeed, the four top companies in the copper, coal and uranium markets controlled less than 60% of export shares in 2012, whereas they controlled more than 80% in the markets of iron ore, soy and potash.                                                    55 Amassing comparable statistics in different commodity markets has proven an unexpectedly arduous task. Markets were selected for their economic importance, and because of China’s relevance to them. Colour coding as follows: (pale blue: 0-50%; beige: 50-70%; orange: 70-80%; red: 80% and above).         87  It is worth noting that most markets have gone through a period of consolidation, mergers and acquisitions, and have seen their levels of concentration rise over the latter part of the twentieth century. As a case in point, the four largest iron ore exporters in 1965 controlled 50% of global exports (Brantley, 1965), compared to over 78% at the beginning of the twenty-first century. The consolidation trends are most visible when considering the rise in concentration by company. Raw Materials Group, a leading mining consultancy, estimates that the share of global iron ore production for the top three iron ore firms globally rose from 15% in 1985 to 35% in 2009, while the top three firms controlled 60% of global iron ore seaborne exports (Storm, 2011).           Another trend worth noting is that following China’s emergence as a systemically relevant consumer, concentration levels (by company) in the markets of copper, coal, uranium and potash actually dropped a few percentage points. In the other markets, trends have been towards further consolidation. Table 16 - Independent Variable, International Markets / Concentration Levels International Market Levels of concentration  IV: Concentration levels % of world exports by top 3 or 4 top companies56   0-10 Iron ore 70% (200357) 8 Potash 83% (199858) 9 Uranium 62% (200659) 5     In accordance with these results, I consider the potash and iron ore markets to be highly concentrated. Given the fact that as a general rule, global commodity markets are highly concentrated, industry concentration in the uranium market is considered more moderate, at 62%. Since my model is based on a binary evaluation of international and                                                  56 Inspired by Krasner, 1974 (he uses the market share of the four largest producers – by country, and concentrated on developing economies exclusively) and Rodrik, 1982 (he uses output shares of the four and eight largest producers in total non-socialist world capacity (output). 57 Almost 70% for the world’s top three companies, Vale, Rio and BHPB, (Ericsson, 2004). 58 2009, UBS Investment Research (Canpotex counted as one firm) 59 Of global production (2006) (List, 1909) The top four companies produced 63% of exports in 2009 (UNCOMTRADE).   88 domestic market concentration, for the purpose of this dissertation, I will consider the global potash and iron ore markets as concentrated, and the uranium market as fragmented (Tables 12 and 16). At the domestic level, the China Mining Yearbook provides us with data on the relative proportion of large and medium enterprises relative to the total number of enterprises operating in a given commodity market in China. It immediately appears that the major extractive industries in China – iron ore, coal and copper – are quite fragmented. A few industries have higher levels of concentration, potash among them (see Figure 13). Figure 13 - Chinese Domestic Commodity markets, Comparative Concentration Levels  Source: China Mining Yearbook 2008. Author’s calculations. Table 17 - Independent Variable: Domestic Markets / Concentration Levels Domestic market Level of market concentration 0 – 10 China Iron ore  Very fragmented 3 China bulk shipping Concentrated 7 China Potash  Few companies, especially at the interface 8 China Uranium  Very concentrated 9  In the binary framework used for the purpose of this dissertation, the markets for uranium, potash and bulk shipping are considered concentrated, and the market for iron ore is 0%#10%#20%#30%#40%#50%#60%#70%#80%#90%#100%#6%# 5%# 8%#67%#5%# 6%#52%#24%#73%#14%#68%#Proportion#of#medium#and#large#mining#enterprises#in#China,#on#total#number#of#enterprises#(2007)!  89 considered fragmented (for a discussion and more empirical detail, see the empirical Chapters 4-6) (see Table 17). 3.7 Causal Mechanism: Coordination Capacity  The Varieties of Capitalism’s “coordination index,” defined by Hall and Gingerich (Coordination Index, 2001; 2004) and used in Hall and Soskice (2001) and Gourevitch and Shin (2005), is a key building block here. Indeed, market coordination, or the level of strategic coordination used to solve collective-action problems/social dilemmas (see Elinor Ostrom, 1997), is a critical characteristic of commodity market structures (see Chapter 2) (see Tables 18 and 19). Elinor Ostrom also argues that patterns of behaviour in global markets have much to do with collective action problems.   At the global level, concentration is used as a preliminary, proxy measure of coordination capacity, but in the qualitative case studies, we pay particular attention to coordination as a quality that often, but not always, overlaps with concentration (Levenstein and Suslow, 2006). As a case in point, the market for uranium exhibits relatively high levels of market concentration (the four largest firms controlled a 62% share of global production in 2006, see Chapter 2). However, qualitative analysis reveals that the uranium market players, in contrast to the potash players, for instance, have not developed institutionalized cooperative relations. What is more, the emergence of large, yet new, players in the market in the 2000s is obscured by “top four firms’ share of global production” data. Table 18 - International Markets - Coordination Levels  International Market Capacity for coordination 0-10 Iron ore Strong oligopoly Annual benchmarking negotiations 8 Potash Two formal cartel-like organizations, annual benchmarking negotiations 9 Uranium Few producers, but low coordination, no global price negotiations 4    Transformations in global markets do not necessarily alter global market shares by the top four firms, but they do alter the likelihood of market players successfully coordinating behaviour. Hence, in the uranium market, we see the characteristics of a relatively concentrated, yet segmented, global market (see typology of global markets, Chapter 2). Historical legacies, such as the failed producers’ cartels (as in the case of the uranium market   90 in the early 1970s), also influence market players’ behaviour and openness towards coordination. As one uranium market industry insider explained: “They haven’t seen any coordination among uranium players, I can tell you that! That is absolutely the case. (…) Nobody will touch it with a ten foot pole. You mention the cartel in the 1970s, those were different days” (Interview 129, 2013).     Table 19 - Domestic Markets – Coordination Levels Domestic market Levels of coordination of procurement behaviour (Interface)  0 – 10 China Iron ore (pricing) Fragmented behaviour, little coordination 3 China iron ore (bulk shipping) Successful coordination between various platforms 8 China Potash  Coordination between actors with access to key interface access points 8 China Uranium  Very few actors, state involvement, some competition 7  At the domestic level, coordination has to do with the cohesion of procurement behaviour among diverse stakeholders. The fundamental conditions at the domestic level – the fragmented or coordinated structure of domestic markets – are in part the result of path-dependent industrial set-ups. But it is the resulting behaviours at both levels that fully allow us to explain outcomes. At the domestic level, the strategic behaviour by key domestic stakeholders (in the Chinese case, SOEs, SMEs and state organs, including industry associations and ministries) determines whether attempts to coordinate Chinese stakeholders’ procurement succeed or not (surmounting problems of collective action).  Market fragmentation/concentration provides us with a strong initial proxy, but it is indicators that provide us with the complete picture of domestic coordination in each market. This includes the number of import licenses; the number of companies actually importing the resource, despite the absence of a license; the share of imports by the top domestic companies; the state’s capacity for control and appointment of official negotiators; the role of industry associations and leadership small groups; and the level of preferences alignment (see Table 20).    Levels of concentration and coordination do not always match perfectly. This is why qualitative case studies are important. Particular care is devoted to assessing the capacity and quality of coordination at the interface with the global markets (this is inspired by Putnam’s   91 concept of “area of entanglement”) (Putnam, 1988).60 For instance, the Chinese domestic potash market is relatively concentrated, but not as concentrated as the uranium market. However, at the interface with the international market, it is strongly coordinated – indeed a “small group” is tasked with coordinating behaviour. Therefore, in this case, Chinese potash importers have arguably exhibited slightly higher coordination capacity than have uranium importers, despite the lower levels of domestic industry concentration. Table 20 - Domestic Markets - Concentration and Coordination Levels  Iron ore (2012) Potash (2012) Uranium (2012)  Number of domestic extractive enterprises  - Many thousands - Few dozen - Very few Number of import licenses - 105 until 2013 (but many more companies use them) -10, but really 3 companies lead negotiations - 2 (recently 3) Interface - Collective action problem due to incapacity for industry association to control private Chinese companies - Importing relationships are strictly controlled - Presence of coordinating “small groups”  - Only 2 companies allowed to import until very recently (now 3) - Some competition internationally Industry Association role -  Important role in the market, lead international negotiations, but at the same time, ineffective and suffering from lack of coordination capacity - No direct role in annual negotiations internationally, but important coordinating role prior to annual negotiations  - Not important  The fundamental structural conditions at the domestic level – the fragmented or concentrated structure of Chinese domestic markets – are the result of path-dependent historical set-ups. But these structural characteristics merely set the conditions for action. I argue that it is the ensuing strategic behaviour of key actors at both levels that allows us to fully explain outcomes. It is the strategic behaviour of key Chinese stakeholders (SOEs, SMEs and state coordinating agents, industry associations, the NDRC, MOFCOM, etc.) and their capacity for coordination of behaviour that determine whether Chinese stakeholders are able (or not able) to obtain their preferred direction of change at the global level. In cases of collective action failure, such as in the iron ore case, we see the highjacking of larger Chinese                                                  60 However, my focus is different from Putnam’s, as I include a wider range of relevant actors at the junction of the domestic and the international, and not exclusively state presence at the international level.   92 actors’ interests by a set of smaller domestic actors (non-licensed SMEs), opening the door for decisive behaviour by large international producers. In the case of potash, we see a more successful coordination of the interface with global market stakeholders.  An analysis of code co-occurrence in interview transcripts confirms these trends (see Table 21).61 Indeed, interviewees mentioned the fragmentation of the iron ore market and its interface much more often than they did for the other cases. Conversely, 16 interviewees mentioned the concentration of the domestic market in the case of uranium. Strikingly, a perception of global market weakness was much more likely to be mentioned in the case of iron ore, despite the fact that oligopolistic structures were more often mentioned in the case of potash. This means that despite higher levels of concentration and coordination in the global potash market, more interviewees felt frustrated in the iron ore case, which hints at the asymmetric position of weakness experienced in that market. This confirms the fact that the analysis of one level of market dynamics (in this case, the international level), is insufficient to explain procurement dynamics and preferences. The joint study of domestic and international market dynamics is key to explaining preferences, behaviour and outcomes.  These broad dynamics were also confirmed by a senior Chinese business journalist, who explained that “the domestic markets for potash and uranium are very concentrated. We were successful in those markets. The market for iron ore is another story” (Interview 8, 2012).                                                    61 See Appendix 2 for more details. From a total of 38 interviews for iron ore, 36 for potash and 32 for uranium, in other words, the opportunities to express the listed concepts were relatively similar across markets. The vastly different results show that interviewees who spoke about the uranium market, for instance, spoke about concepts other than the ones listed here. It is important to note that a few excerpts had to do with Japan, as well as, or in contrast to, China, which influences the results (16 excerpts mentioned Japan, against 78 mentioning China). In certain cases, the diagnosis was the same and it does not skew the results (for instance, an interviewee may have been saying that the Chinese and Japanese consumers both faced a concentrated global iron ore market in the 2000s). At times, however, an interviewee may have been contrasting both cases, for instance, saying that whereas the Chinese iron ore market is very fragmented, the Japanese iron ore market is very concentrated. This would yield two entries, one for “iron ore” and “domestic market fragmented” and one for “iron ore” and “domestic market concentrated,” which I suspect in the case in the two instances of the co-occurrence of “iron ore” and “domestic market concentrated” below. Overall, a qualitative analysis of the content of the interviews confirms the broad tendencies observed in this table. I colour-coded the results for easier analysis (pale blue: 0-4 mentions; beige: 5-9 mentions; pale orange: 10-14 mentions; orange: 15-19 mentions; dark orange: 20 mentions and above).         93 Table 21 - Code Co-Occurrence in Interviews !! Iron%Ore% Shipping% Potash% Uranium%Global!Market:!Fragmented! 4% %% 1% 9%Global!Market:!Oligopolistic!structure! 11% 1% 22% 4%Global!Market:!Coordinated! 10% %% 10% %%Perception!towards!global!markets:!Strength! 4% %% 1% 6%Perception!towards!global!markets:!Weakness! 20% 1% 4% 3%Diverging!interests! 9% 4% 1% 1%Difficulty!to!control! 10% %% 3% %%Lack!of!information! 15% 2% 11% 1%Collective!action!problem! 4% %% %% %%Extracting!rent! 5% %% 3% %%Domestic!Market:!Fragmented! 18% 1% 7% 2%Domestic!Market:!Concentrated! 2% 2% 4% 16%Interface:!Fragmented! 13% %% 5% 2%Interface:!Concentrated! %% 2% 5% 9%Interface:!Coordinated! 1% 3% 10% 6%Again, I argue that studying structural relations between domestic and international stakeholders, as well as the domestic power asymmetries between different domestic stakeholders, is necessary to help explain the global outcomes I observe as a result of China’s emergence. The likelihood of change in market institutions at the global level is the result of the coming together of market characteristics at both the domestic and international levels.  The fact remains that ironically, under certain circumstances, Chinese stakeholders’ procurement behaviour has positioned the country to be a catalyst for an increased marketization of global commodity market pricing regimes. This is a counter-intuitive finding for a state-led hybrid economy. China’s sheer size and its arrival in global market structures that offer little room for manoeuvre put it in a position in which disorganized procurement behaviours have had the unintended effect of giving global producers a strong hand in pursuing their pricing regime preferences.   3.8 Process Tracing: The Politics of Fragmentation and Coordination  Establishing the structural context in which Chinese market actors evolve allows us to get a general idea of the context in which procurement is affected in China. However, it is   94 important to delve into the details of Chinese market actors’ procurement behaviour to understand the following: How have levels of fragmentation at the interface of Chinese and international markets come to differ so much across markets? What was the role of key Chinese state organs in either facilitating or curbing these levels of fragmentation? What role did the relative fragmentation of the interface between Chinese market actors and the global marketplace play in influencing Chinese behaviour, and, in turn, systemic impacts at the global level?   I do this through a detailed study of the history of the import licensing process in China in the iron ore and potash markets. What emerges from this analysis is that the Chinese government tried to curtail import licenses in the iron ore industry from 2004 until 2013, but these attempts were ineffective and ultimately unsuccessful, and hundreds of Chinese iron ore importers continued to import the resource from overseas suppliers without official licensing documents. In the potash case, in contrast, the Chinese government has in fact been trying to increase the number of Chinese companies that can import potash. But these efforts have also remained ineffective; despite increasing numbers of potash import licenses in China, to date only two or three companies continue to import the bulk of the resource.  3.9 Summary of Cases 3.9.1 Iron ore case summary   In the iron ore case, the level of domestic fragmentation in the Chinese iron ore industry (fragmentation) and the lack of state organs' capacity to enforce national strategies (lack of coordination) had an impact on China’s capacity to implement consistent and coherent procurement behaviour and opened the door for decisive actions on behalf of global market stakeholders. The domestic market for iron ore is highly fragmented, despite multifaceted attempts to remedy this by several organs of the central Chinese government over the last decades (NDRC, State Council, MOFCOM, CISA, etc.).  What was even more critical was that its interface with the global market is also highly fragmented, and this resulted in a clear asymmetrical relationship between domestic and global market stakeholders, a relationship that favoured the global stakeholders. This has led to bold, accusatory rhetoric and fragmented behaviour, which made it difficult for China to implement its international iron ore procurement policy. At key moments in the transition period from Japanese to Chinese dominance in the global iron ore market in the mid-2000s,   95 the Chinese lead negotiators were unable to coordinate procurement behaviour by Chinese firms on the international market (collective action failure in the Chinese domestic iron ore industry). CISA underestimated three things: the global market players’ eagerness to sell; the international suppliers’ commitment to the benchmark system; and its own capacity to control access to the import interface. The ensuing fragmented procurement behaviour (or lack of coordination) by hundreds of Chinese iron ore importers, coupled with the sheer volumes of Chinese imports, had a profound destabilizing impact on global market institutions.   By 2010, only six years after China became the world’s number one importer of iron ore, the behaviour of Chinese stakeholders destabilized a decades-long benchmarking regime, and the system fell apart. The market then moved towards a quarterly, and eventually, a spot pricing system. In sum, a collective action failure in the Chinese domestic industry yielded behaviours that profoundly affected incentive structures of global market players, and after only a few repeated interactions, had the deepest kinds of repercussions for what had been until then a stable market regime.   3.9.2 Potash case summary  In the case of potash, the Chinese domestic market and its interface with the global market are significantly more concentrated and coordinated than in the case of iron ore. I argue that this is because the potash market was more successfully consolidated during the reform era, and thus presented a significantly more concentrated and coordinated interface with the global market when Chinese stakeholders became a key player in the global market. This allowed Chinese stakeholders more control over the aggregate behaviour of its market stakeholders. This two-level symmetry (concentrated domestic and international market players) translated into more stable bargaining patterns, as well as moderate and incremental market change, much of which favoured the Chinese market players. In that case, we see effective translation of policy priorities into market outcomes, at least up to a point.  In the global potash market, Chinese negotiators also adopted a very bold – and this time successful – negotiation strategy. In August 2012, the Chinese negotiator (Sinochem) refused to sign the benchmarking contract with the Canadian negotiators (Canpotex), and asked for a markdown. The Chinese potash consumers held firm during the negotiation period, and did not contravene directives not to buy. In December 2012, global potash   96 suppliers agreed to a 15% markdown in price, a significant concession. In contrast to the iron ore case, the annual negotiations and the pricing regime, as well as the prevalence of long-term pricing contracts, survived China’s bold behaviour and its emergence as the number one consumer. The frequency of the annual negotiations has also increased to twice a year, a change that has been supported by Chinese and global market players. Finally, whereas the North American global marketing cartel survives until today, the Russian-Belarusian cartel (BPC) fell apart in the fall of 2013. This dissertation explains the divergent outcomes in the iron ore and potash markets, two global markets exhibiting striking levels of similarities prior to China’s emergence, and it does so through the analysis of market power asymmetries. In the potash market, Chinese consumers are both more concentrated and better coordinated, and this has allowed Chinese stakeholders a more effective translation of policy priorities into market outcomes.       97 CHAPTER 4 – CHINA’S IMPACT ON THE GLOBAL IRON ORE MARKET It is worth restating the two research questions posed at the beginning of this dissertation. The first is about likelihood of change: Why have some global markets experienced systemic change following the emergence of a large consumer, whereas other markets have been more stable? The second question is about direction of change: Why have some global markets undergone marketization as a result of the emergence of a large consumer, whereas others, under similar circumstances, have changed in the opposite direction?  I argue that structural power asymmetries between domestic (Chinese or Japanese) and global market players determine global market institutional stability, and that they create the conditions for institutional market change at the global level. More specifically, I argue that extreme asymmetry between global and domestic positions tend to be unstable, and that rapid changes in market power relations during periods of power transition are particularly disruptive. Furthermore, positions of market power at the domestic level influence preferences for pricing regimes at the global level. Asymmetric positions of power provide strategic advantages, since the dominant side’s preferences for market institutions at the global level are more easily expressed. This has led to unexpected outcomes, including the marketization of certain pricing regimes, as a consequence of China’s emergence. The key causal mechanism at play is the relative capacity of commodity market stakeholders for market coordination, which I measure as their respective levels of market concentration. When looking at domestic market structures, my approach is similar to that developed in Chapter 2 to define global markets, although with a few adjustments. The goal here is not to provide a comprehensive analysis of China’s domestic political economy, but rather to point to a number of characteristics that emerge to play a crucial role in shaping Chinese actors’ interaction with their global market counterparts.    In this chapter, I combine within-market comparative case analysis with careful process-tracing to critically assess the role of domestic and international variables in leading to a change in the global iron ore market pricing regime. I focus on China and Japan’s iron ore consumption because it is one of the most critical components of infrastructure development. Moreover, the global iron ore market has experienced dramatic levels of   98 change following the emergence of Japan, and then China, as the world’s number one iron ore importer in the 1960s and 2000s, respectively, but in different directions. I investigate two related cases involving China’s emergence: China’s impact on the global iron ore market, and the iron ore bulk shipping pricing regimes. The Japan case will serve as an across-time comparison, but its within-market nature (both cases occur within the iron ore market) allows us to keep many variables constant.   The dynamics through which pricing regime changes occur can be unexpected. For instance, at key moments in the transition from Japanese to Chinese dominance of the global iron ore pricing regime, China’s lead negotiators were unable to coordinate procurement behaviour on the international market. This lack of coordination capacity created opportunities for large global iron ore suppliers to push for their market preferences. Merely six years after China’s emergence as the world’s number one iron ore importer, the decades-old benchmarking pricing regime had fallen apart and spot pricing had emerged. In contrast, relative symmetry in market coordination capacities at the domestic and international levels in the case of China’s impact on the global iron ore bulk shipping pricing regime translated into deadlock, a bargaining dynamic, and moderate and incremental market change away from marketization (that is, an ownership sharing agreement). Japan’s emergence as the number one consumer of iron ore in the 1960s was a case in which highly coordinated iron ore consumers faced a more fragmented global market. This provided Japan with the room to manoeuvre to coordinate behaviour and achieve preferred results at the global level: a demarketization of global pricing regimes.  In the first section, I introduce my argument as it plays out specifically in the iron ore market. I then characterize the change in the iron ore benchmark pricing regime that I try to explain. Third, I present key characteristics of China’s domestic iron ore industry. Then, I illustrate my causal narratives by tackling two within-market case studies involving China: the fall of the iron ore benchmarking regime, and the Valemax shipping saga. Finally, I investigate the causal story in the emergence of Japan, another dominant player in the iron ore industry, 50 years prior to China’s emergence as the number one consumer.   99 4.1 Argument – The Iron Ore Case  I argue that structural asymmetries between domestic and global markets profoundly affect incentives and global market players’ capacity to act and create the conditions for systemic change (see Table 22 below).   In response to the question about the likelihood of change at the global level, I argue that power transitions that entail a profound change in consumer-producer power relations are more likely to be disruptive and lead to institutional change at the global level, than are power transitions that entail a continuation of two-level power relations. In addition, a change towards asymmetric positions of power (Quadrants 2 and 3) tends to be more unstable than a change towards symmetric positions of power.  Table 22 - Two-Level Market Power Asymmetries in the Iron Ore Market International market ! (Producers) Domestic market " (Consumers) Fragmentation Coordination Fragmentation 1 Symmetry    2 Asymmetry in favour of producers  Case: China – iron ore benchmark Outcome: marketization  Coordination 3 Asymmetry in favour of consumers  Case: Japan – iron ore benchmark and shipping Outcome: demarketization  4 Symmetry   Case: China – shipping Clash, bargaining Outcome: indeterminate    The first Chinese case examined in this chapter, the China – iron ore/benchmark case, illustrates the conditions necessary for an extensive change to occur at the global level. Not only does the global iron ore market exhibit high levels of concentration, but the Chinese domestic iron ore industry exhibits very low levels of concentration, and in turn, low levels of coordination. This has placed the market players in positions of market power asymmetry with one another, with the global iron ore market players having the upper hand. Furthermore, the pricing regime that Chinese stakeholders disrupted in the 2000s had been   100 established decades earlier during Japan’s tenure as the dominant consumer. During that time, the market power asymmetry positions had been reversed. Therefore, China’s rise not only occurred in an unstable quadrant, but the Chinese stakeholders exhibited a profile that was markedly different from that of their Japanese counterparts, and this was also destabilizing. These patterns led to rapid and substantial change in the global iron ore market as a result of China’s emergence.   In this chapter’s China cases, the international markets exhibited very high levels of coordination prior to China’s rise. Because we are looking at two-level variation between cases, variation in China’s domestic market structures becomes determinant. I investigate how domestic market characteristics constrained China’s options, shaped its procurement behaviour, and, as a result of interaction with global market stakeholders, durably transformed global iron ore market pricing institutions.   In response to the question about the direction of change, I argue that market players’ preferences for pricing regimes are influenced by the relative position of market power they occupy in their own market as well as relative to international market players. Asymmetric positions of power provide strategic advantages, since the dominant side’s preferences for market institutions at the global level are more easily expressed. The specific pattern of China’s emergence and behaviour, coupled with that of the international producers in the iron ore market, led to the fall of the benchmark pricing regime, or to a marketization of global market institutions. In the case in which consumers had the upper hand, such as at the time of Japan’s emergence as the number one consumer, we saw a demarketization of global market institutions.  Case summary  China has already promoted systemic change in global iron ore market institutions. The first case in this chapter focuses on the fall of the decades-old international benchmarking regime in the global market, which was enabled by the behaviour of Chinese stakeholders. Indeed, in 2009, the China Iron and Steel Association (CISA), an association which exists to represent interests of both its member companies and to coordinate with central government organs, adopted a tough negotiating strategy with the three biggest global iron ore producers (Rio Tinto, BHPB and Vale). It also asked all domestic steel companies to refrain from buying on the international market while the negotiations were ongoing (CISA   101 was asking for significant price markdowns). But CISA's strategy backfired. A multitude of small Chinese steel firms ignored the industry association's request and purchased iron ore on the international market through individual "spot" contracts. The three big suppliers, led by BHP, were keen on ushering in an era of shorter-term contracts and spot pricing. This combination of behaviours eventually led to the fall of the decades-old international benchmarking regime.   One counter-intuitive finding in this instance is that China’s impact on the global iron ore benchmarking pricing regime was the result of a domestic position of weakness rather than a position of strength.   The second case examined here is that of China’s impact on global shipping pricing regimes. China’s emergence led to a shift of shipping pricing conventions, and to a strategy by Vale that entailed building very large ore carriers (VLOCs) to ship the ore to China. The response of Chinese stakeholders was bold and firm: a government edict blocked the docking of Vale’s super cargoes in Chinese ports between 2011 and 2013. Evidence shows that this is best understood not as a top-down central government policy, a CISA policy, or even a result of large SOEs’ coordination; rather, the strongest opposition, lobbying, and coordination capacity came from Chinese ship owners-operators, chiefly COSCO, which resented the fact that it was losing shipping business. In other words, the blocking of the VLOCs was the result of the consolidated power of the relevant Chinese stakeholder, COSCO, and the coordination of the interface with the global industry by a player that was actually acting against broader national priorities. 4.2 Domestic-Level Market Structures At the domestic level, fragmentation is key, as it impedes coordination of behaviour. But there are two key distinctions to make in the domestic context. First, the appropriate focal point when evaluating domestic-level fragmentation is the interface with the global industry. Since we are looking at commodity imports, the points of contact between domestic and international stakeholders are more pertinent than fragmentation writ large in China’s domestic markets. Levels of fragmentation are ultimately meaningful as they enable or impede Chinese stakeholders’ capacity for coordination. In evaluating this coordination capacity more fully, I take into account the role of key state institutions (relevant to specific commodity markets), key large importers and small and medium enterprises, when relevant.     102 I have identified the critical characteristic of commodity market structures as their levels of market concentration. At the domestic level, I conceptualize this at the interface with global players. This is best defined as the number of stakeholders with procurement access to global resource markets. The number of import licenses is a good proxy for this, even if in the iron ore-benchmark case there were many more companies importing the ore than there were official licenses.  The levels of market fragmentation are relevant because they allow us to estimate the control and coherence of behaviour at the interface with the global industry. The causal mechanism through which behaviour and change happens at the global level is the differential coordination capacity of actors at each level, or the differential ability of actors to solve their collective action problems. It is through coordination (or strategic coordination, used to solve collective action problems/social dilemmas), or lack thereof, that market actors get what they want. Maintaining pricing regimes, and especially changing pricing regimes, requires coordination (see Hall and Gingerich, 2004; Hall and Soskice, 2001; Elinor Ostrom, 1997). In my qualitative case studies, as I carefully evaluate Chinese actors’ capacity for coordination, I take into account, where relevant, the role of state organs and industry associations (which are different in each market), large SOEs, small private firms and the overlap of preferences between actors (there can be few actors with little overlap of preferences, or many actors with preference overlap that could make it easier to collaborate). In the China – iron ore / benchmark case, as China’s need to import iron ore soared, there was an effort at coordinating procurement behaviour at the interface of the domestic and international iron ore markets. This was evident, first, through the selection of a SOE representative (Baosteel) to lead the international negotiations, then through CISA’s behaviour during the 2009 and 2010 rounds of negotiations. However, as we will see, coordination attempts were largely ineffective. My case study shows, however, that the level of domestic fragmentation in the Chinese iron ore industry and the resulting lack of coordinated procurement strategy impaired China’s position and led to international suppliers gaining the upper hand. In this case, I show that Chinese stakeholders’ procurement dynamics ironically positioned the country as a catalyst for the marketization (liberalization and financialization) of the iron ore market pricing regime. This is a counter-intuitive twist of fate for a state-led hybrid economy.   103 4.2.1 Iron ore: Relevance to China   China’s level of iron ore import dependence is high and increasing (see Figure 14 below).   Figure 14 - China's Iron Ore Import Dependence  Source: Chinese Ministry of Land and Resources data (2012). Author’s calculations. 4.3 Case 1: The fall of the Global Iron Ore Benchmark Pricing Regime  The benchmark pricing regime prior to China’s emergence   In 2003, China’s steel imports surpassed those of Japan for the first time, and the country became the number one iron ore importer (Wan, 2010). The global iron ore market at the time of China’s emergence as the dominant consumer operated under a negotiated benchmark pricing regime. For 50 years until 2009-2010, the iron ore market had been operating under a mixture of long-term (10 to 20 years) and shorter-term (one-year) contracts. In both cases, prices were determined as a result of annual benchmarking negotiations between the three main iron ore producers, BHP Billiton and Rio Tinto and Vale, and the main international importers, which by then had long been the Japanese importers (Zhang, 2009a). In 2003, however, the first benchmark price agreement was signed between Vale and Arcelor, with Nippon Steel following closely behind (the next day) at more than 15% over the previous year (Ericsson, 2004). A small emerging spot market for iron ore came into existence as China started to import iron ore from India (neither of which was participating in the benchmark negotiations at the time). 58%# 59%# 57%#0#20#40#60#80#100#2000# 2001# 2002# 2003# 2004# 2005# 2006# 2007# 2008# 2009# 2010# 2011#China's!iron!ore!import!dependence!  104 4.4 Independent Variable: The Global Iron Ore Market  China faces particularly high levels of concentration in the global iron ore market. The “iron-ore industry has (…) been consolidating more or less continuously since the 1970s” (Ericsson, 2004). The market is heavily concentrated among two producing countries (Australia and Brazil), and three producing companies (Rio Tinto, BHP Billiton and Vale). Together, Australia and Brazil control more than 70% of the world’s iron ore exports.62 (Figure 15).  “To measure corporate control at the production stage underestimates the concentration of the iron ore sector because large amounts of production do not enter the market, but are produced in captive mines or mines which have a protected or restricted market. An alternative way to measure the control is to monitor the share of global seaborne trade of the leading companies. Measured this way, the shares of the major companies are considerably higher. Vale alone controls 26% of the total world market for seaborne iron ore. With the market shares of Rio Tinto and BHP Billiton dropping in 2010, the overall share controlled by the Big 3 fell from 60% in 2009 to 58% in 2010.” (Ericsson et al., 2011) What is more, the concentration in the global iron ore business had been increasing over the previous 20 years, until 2007 (Figure 17).  Figure 15 - Iron Ore Exporters World Market Share  Source: UNCTAD, 2013  Chinese imports mirror global average export shares (see Figure 16 below).                                                   62 International Steel Statistics Bureau. Australia#45%#Brazil#29%#South#Africa#5%#India#3%#Canada#3%#Sweden#2%#Others#13%#Iron!ore!exporters!(world!market!shares),!2012!  105 Figure 16 - Chinese Iron Ore Imports by Country (2001-2010)  Source: Chinese Steel Statistical Yearbook. Author’s calculations. Figure 17 - Top Five Seaborne Iron Ore Producers (1997 - 2007)  Source: Baffinland; Tex Report; AME; Roskill; UNCTAD; BMO Capital Markets.  0#5000#10000#15000#20000#25000#30000#10!000!tonnes!Chinese!Iron!Ore!Imports!by!Country!Australia# Brazil## India# South#Africa# Ukraine# Russia# Others##Vale#19%# BHP#14%#Rio#Tinto#12%#Robe#River#7%#MBR#5%#Other#43%#Top!5!seaborne!iron!ore!producers!in!1997:!56%!Vale#41%#Rio#Tinto#24%#BHPB#14%#Kumba#4%#LKAB#3%#Other#14%#Top!5!seaborne!iron!ore!producers!in!2007:!86%!  106 4.5 Global Iron Ore Market Players: Coordination Capacity  Two comments can be made in analyzing levels of coordination. Market concentration is high in the global iron ore market, but coordination between the three major firms, although taken as fact by many interviewees, is slightly harder to establish.  First, says Ericsson,  “while the international steel industry is not cohesive and producers do not act in unison, the three large iron ore producers don't have to collude in order to exercise considerable control over the market and ensure that they are pursuing mutually consistent strategies. Their objective is obvious—to maximize profits— and their method of achieving the objective equally so: keeping prices high enough to pay for new investment but low enough so that new entrants do not become realistic alternative sources of product.” (Ericsson et al., 2011)     Supranational regulatory gaps have also given established multinational corporations more leeway to influence the outcome of the game; for a recent example, see the current scandal over Goldman Sachs’ aluminum market shuffle, hoarding and double-play (Kaminska, 2013). One Chinese academic commented: “The problem with global markets is that there is no regulatory oversight on market actors’ behaviour (...), at the WTO for instance, there is nothing regulating the oligopolistic behaviour of the three big iron ore companies. This is a problem, we have not found the regulatory solutions to this issue” (Interview 7, Chinese academic, 2012).  Second, to the extent that perceptions and trust matter in influencing market stakeholders’ behaviour, it is worth noting the fact that the Chinese iron ore and steel market stakeholders, at the central level, have exhibited extremely high levels of mistrust towards global iron ore producers. China, as a newcomer, is particularly sensitive to perceptions of unfairness in global markets. The fact that global commodity markets operate under high levels of concentration, and that China is, as it were, entering the game after half-time, is adding to that perception (Nolan, 2012).   Although international suppliers themselves point to the presence of strong rivalry between the three big mining firms (interviews with industry insiders, 2012), suffice it to say at this point that Chinese stakeholders perceive the firms’ behaviour as collusive. In March 2013, China’s National and Development Reform Commission (NDRC) released a report that accused the big iron ore mining companies of “artificially inflating the price of iron ore,   107 by delaying and controlling shipments, delaying sales, and causing a temporary illusion of shortage in the market” (Zhong, 2013). BHP was also accused in January 2013 of buying 100,000 tonnes of iron ore to support prices.63 CISA has also accused the global miners of collusion (Paul and Serapio Jr, 2013).   China is skeptical of multinational corporations' commitment not to coordinate production to maximize profits. A senior Chinese journalist commenting on “the behaviour of international cartels or informal cartels in commodity markets, such as oligopolies,” said that: “Speaking of the dynamics among the major producers, even if iron ore is not controlled by a government-to-government cartel, we think it is just a different type, with private companies involved” (Interview 8, 2012). Some of my interviewees go even further. A Chinese official and industry insider exclaimed that: “Price making is in Western hands!!! By monopolies!! But the demand is in emerging countries!!! And [we have] no price making power! Very frustrating!” (Interview 65, 2012, emphasis noted by author). Even Platts, the price index company, does not escape this perception: “From a China point of view, the Platts index is really opaque; they can’t explain how they come up with the price, and its links with the three big iron ore companies and with companies such as Goldman Sachs and Merill Lynch are evident!” (Interview 59, Chinese government insider, 2012, emphasis noted by author). Another Chinese industry insider concurred: “People here do not think that Platts is transparent, or objective” (Interview 111, 2012).  As a case in point, in March 2013, China’s National and Development Reform Commission (NDRC) released a report that outlined four factors that caused the US$90-150 price rally in the second half of 2012 and said that the most recent surge in iron ore prices was partly the result of “unreasonable pricing methods” by the biggest iron ore producers. The NDRC accused the big iron ore mining companies of ‘artificially inflating the price of iron ore, by delaying and controlling shipments, delaying sales, and causing a temporary illusion of shortage in the market.” These behaviours by the three major mining companies were developed in order to foster an illusion of shortages and to send fake market signals. BHP was accused in January 2013 of buying 100,000 tonnes of iron ore to support prices. The economic agency went as far as to say that as a consequence, the Platts iron ore index                                                  63 BHP and Platts immediately issued a rebuttal (Riseborough, 2013).   108 could not be fully trusted because it uses opaque and unrepresentative pricing systems, which pushes prices higher, and because these inflated prices were the basis for its calculation (Zhong, 2013). BHP and Platts immediately issued a rebuttal (Riseborough, 2013). “Chinese steel mills (...) continue to claim that the health of the sector is being harmed by the "monopoly practices" of the big three global iron ore suppliers” (2011a). “The China Iron & Steel Association, an industry body representing large Chinese steelmakers, [has also recently] lobbied the government to investigate possible price manipulation by miners and traders, a senior official said” (Paul and Serapio Jr, 2013). See Table 23 below for a snapshot of price levels and quantities imported in 2008.   The relationship between transparency and commodity price volatility is complicated. The above-mentioned accusations of cheating by the NDRC destabilized the relationship between China and global market stakeholders at a time when markets were undergoing a massive transition. “Deng Qilin, the president of Wuhan Iron & Steel Group [and] the chairman of the China Iron & Steel Association (CISA), said: ‘Although the prices of steel products have picked up this year, they still cannot digest the pressure from rising iron ore prices. (…) Chinese steel mills are suffering from slender profits while iron ore miners only have to dig about a meter underground. What costs them $10, they sell for $100. Is it fair?’” (Zhang, 2010). Table 23 – China’s Iron Ore Imports by Country, with Price Paid (2008) Country Quantity (ton)  Share of total (%)  Total dollar amount (thousand of USD)  Share of total amount (%)  Price per ton  Australia 183400058 41.4 22447295 37.1 122.4 Brazil 100619615 22.7 14940815 24.7 148.5 India 90963204 20.5 13391744 22.1 147.2 South Africa 14523975 3.3 1908485 3.2 131.4 Indonesia 6756925 1.5 609035 1 90.1 Russian Federation 5790016 1.3 967505 1.6 167.1 Peru 5341970 1.2 668864 1.1 125.2 Iran 5144032 1.2 852912 1.4 165.8 Ukraine 4596880 1 813948 1.3 177.1 Canada 3710411 0.8 762438 1.3 205.5 Chile 3579471 0.8 550408 0.9 153.8 Afghanistan 3189947 0.7 443361 0.7 139.0 Venezuela 3185870 0.7 589064 1 184.9   109 Country Quantity (ton)  Share of total (%)  Total dollar amount (thousand of USD)  Share of total amount (%)  Price per ton  Mauritania 2492590 0.6 328296 0.5 131.7 North Korea 1881901 0.4 172494 0.3 91.7 Other 8328217 1.9 1106744 1.8 132.9 Total  443505080 100 60553407 100 Average = 144.6 Source: China Mining Yearbook 2010  In sum, coordination levels in the iron ore market are not formalized like they are in the case of the oil market or the potash market, both of which have created formal producers’ organizations. Implicit coordination is, however, mentioned repeatedly in interviews, and is certainly perceived to be an issue from the point of view of the Chinese market stakeholders.  4.6 China’s Domestic Iron Ore Market Chinese production and consumption levels The total value of iron ore imports in China in 2011 was USD$112.43 billion. Most of the steel is used in the construction sector (see Figure 18). While Chinese iron ore production has been growing, it has not been growing fast enough to close the gap between consumption and production, which has been filled with ever increasing imports (see Figure 19). The growth in iron ore imports has been slowing down, but absolute import amounts keep growing, albeit at a slower pace. Figure 18 - Main Steel Consuming Sectors in China (2011)  Source: China International Iron and Steel Raw Materials Conference 2012 Construction#63%#Industry#34%#Energy#3%#Main!steel!consuming!sectors,!2011!  110 Figure 19 - Chinese Iron Ore Annual Production and Imports (2000-2010)  Source: Chinese Steel Statistical Yearbook. Author’s calculations. 4.7 Independent Variable: Domestic Levels of Concentration  The domestic iron ore industry in China is so fragmented that it is hard to get a clear picture of the companies that are operating at any one time. The China Metallurgical Press cites a total number of 7,373 steel-producing enterprises in 2007, up from 6,999 enterprises in 2006, and 2,997 enterprises in 2000 (Yang, 2010). In 2008, 92.2% of those enterprises were small and medium enterprises (see Table 24). The iron ore industry is equally fragmented, with 40 large iron ore enterprises (5 MT and above, 35.3% of national output), 187 medium size enterprises (1-5 MT, 28.8% of national output) and 1,171 small private mines (up to 1MT, 36% of national output) (Mayfield, 2013). The China Steel Statistical Yearbook counted a total of 12,143 steel and iron ore producing enterprises in the country in 2010 (China Steel Yearbook (ȗȚí"ôȕ), 2011).  The iron ore industry itself is also extremely fragmented in China, with Yang Jiasheng, the chairman of the Metallurgical Mines Association of China (MMAC), confirming that large-scale mines make up only around 3% of the country’s 4,037 mines (Stanway, 2015). Interviewees often used the term “messy, chaotic” (0), to describe the Chinese domestic iron ore industry (Interview 111, Chinese government official, 2012).  0#20000#40000#60000#80000#100000#120000#2000# 2001# 2002# 2003# 2004# 2005# 2006# 2007# 2008# 2009# 2010#10!000!tonnes!Chinese!iron!ore!production!and!imports!2000!J!2010!Domestic#Iron#Ore#Production# Iron#Ore#Import#  111 Table 24 - Comparative Fragmentation Levels in the Chinese Iron Ore and Potash Industries  Chinese potash industry Chinese iron ore industry Chinese mineral industry average Proportion of large-scale and medium-scale enterprises 73.3% 6.4% 7.8% Proportion of small-scale and small enterprises  2.67% 93.6% 92.2% Source: Chinese Mining Statistical Yearbook 2008. Author’s calculations.  Significantly, in China, the three largest companies only occupy 14% of total domestic production, compared to 70% by Japan’s three biggest steel makers, and 60% by the top three steel makers in the US (Bergsten et al., 2008). In 2010, although “key enterprises” (or large and medium size enterprises combined (ȒźÌ'ÁJ")) produced 87% of crude steel in the country, they produced 65% of total finished steel products, and 5% of ferroalloys (China Steel Yearbook (ȗȚí"ôȕ), 2011).  Figure 20 - Domestic Fragmentation Levels - Top Four Chinese Crude Steel Producers (2000 - 2010)  Source: Chinese Steel Statistical Yearbook. Author’s calculations.  The top four Chinese crude steel producers produced 25.87% of total Chinese steel in 2010 (see Figure 20). Consolidation efforts in the iron ore industry have been laborious and largely unsuccessful (Zhou, 2008). One Chinese industry insider commented: “One of the 32%#17.92%#25.87%#0%#20%#40%#60%#80%#100%#2000# 2005# 2010#Top!4!Chinese!Crude!Steel!Producers!(share!of!total)!  112 great failings of the last five years is that China could not reach its objectives of consolidation in the iron ore market. The objective was to reach 50% of the market for the 10 biggest steel companies by 2010, and 70% in 2020. In 2010, they were only at 42%. One of the biggest critics of this [situation] was the Youth Daily; it was very vocal” (Interview 98, 2012).  4.8 Domestic Market: Coordination Levels  4.8.1 High fragmentation at the interface between Chinese and global iron ore markets   Consolidation efforts in the Chinese iron ore industry have been laborious and unsuccessful (Zhou, 2008). The key variable of relevance here, however, is the number of Chinese companies that import iron ore and are situated at the interface of the Chinese and global iron ore market. The Chinese state has tried, though largely unsuccessfully, to reduce this number over the years. One resource-industry SOE official remarked: “The steel industry is not concentrated in China and the market structure influences negotiating power” (Interview 3, with Chinese SOE official, 2011).   4.8.2 Process tracing: Iron ore importing licenses – a short history  In 2004, the People's Republic of China enacted the Foreign Trade Law (People's Republic of China Foreign Trade Law, 2004). Article 15 stipulates that the Foreign Trade Department of the State Council can implement automatic import licenses in order to be able to monitor the country’s import situation. The law also stipulates that the Customs Department should not allow another entity to import a product without an import license (should the product be listed as an automatic import license good). The same department manages special import licenses (or quota licenses).  The Chinese Ministry of Commerce (MOFCOM) and the General Administration of Customs issued notice 26 on December 9, 2004, introducing the automatic import license management system, to be implemented on January 1, 2005 (The Ministry of Commerce and the General Administration of Customs issue notice 2004/26, "The Automatic Commodity Import License Management System" (²}ȎūbċƾD 2004ôƪ 26Ÿ "ǰżlj~ȆšǢœƭƃ{ť"), 2005). At the time, there were issues regarding the iron ore import process, including the fact that large steel enterprises often hoarded the ore and resold it at a higher value to small and   113 medium enterprises. The “automatic import licensing system” for iron ore was announced by the MOFCOM and the General Administration of Customs’ Announcement Number 9, on March 1, 2005 (ȚƗƙlj~ȆšǢœǥƇȴĻǕ{ť) (Provisional Procedures, 2005). Pertaining to the new licensing system, companies that conform to the various criteria outlined by MOFCOM could obtain an import license.  In 2005, the fifth criteria stipulated that to obtain a license, companies needed to conform to the qualification standards established by CCCMC and CISA (“Ʃ¢7Ɨ…í²LȗȚí"ŒLwǜƍ@4ȚƗƖƴǒǴDZŏk”), issued on February 28, 2005 (standards and reporting procedures for qualifying iron ore importers “ȚƗƖȆšJ"ǴDZŏk¯ƇĘƢø) (Provisional Procedures, 2005). Such standards included the need for companies to:  - Meet the steel industry access standards (Ʃ¢ȗȚǕ"k^ŏk), as well as a notice sent by the General Office of the State Council to the NDRC and other departments on December 23, 2002 on: “Ideas on how to put a stop to blind investments in the steel, aluminum and cement industries”64; - For steel companies, last year’s crude steel production should be at least one million tons; for distribution companies, they should have registered capital of at least 10 million RMB, imported at least 300,000 tons of iron ore the previous year, or more than 100,000 tons in the first two months of 2005; - Products to meet national quality standards, as well as ISO standards; - Imported iron ore should not go to enterprises that are not supported by state development policy or have backward facilities; - Steel enterprises meeting the above criteria should apply to the China Steel Industry Association for a license, and others (ȗȚƄ:EÊ) should apply to the CCCMC.  In 2004, prior to the establishment of the license system, there were 523 companies importing iron ore in China (Zhuang, 2006). In 2005, MOFCOM’s notice established the number of iron ore importing licenses at 118, more than a four-fold decrease.                                                   64 This notice includes a directive to strictly enforce market access (consolidate the industry) to avoid excessive competition and wasting of resources. Minimum conditions for investments include the size of the facilities, furnaces, energy efficiency standards, environmental standards, etc.   114 In 2007, CISA and CCCMC issued an announcement following a joint working meeting. The number of firms allowed a license to import iron ore was to be reduced again from 118 to 112 (the number then remained stable until 2010) (Liang, 2007). “CISA started reducing iron ore import licenses in 2005. The number of steel mills and trading companies possessing licenses in China has been reduced from 500 to 112 [in 2009], of which trading companies' licenses are down from 250 to 40” (Zhang, 2009b). Speaking about the automatic licensing process in general in 2013, MOFCOM spokesperson Shen Danyang also specified that the licensing system was separate from other issues. MOFCOM remained silent on other considerations by relevant industry associations and organizations regarding the rules, conventions and standards with regards to iron ore import governance, which get resolved via means of consensus, and even on official industry governance measures (Chinese Ministry of Commerce, 2013). This confirms that the substance of qualifying standards for iron ore imports is not established by MOFCOM, but by CISA and CCCMC.  In April 2010, CISA and CCCMC established a “Joint office of iron ore imports” (“ȆšȚƗƖǀ¢{`Ü”) specifically to “control the flow of iron ore imports and curb speculative behavior” (Li, 2010). Indeed, the illicit use of official licenses by more than one company was widespread at the time (interviews, 2012). The result was that a number of companies well in excess of the number of official licenses, were purchasing iron ore on the global market, and this was creating a problem that was labeled “ŹƗǕ+”, or the “stir-fry ore” behaviour (speculating behaviour in the mining industry). On February 18, 2008, in order to remediate the soaring iron ore prices and consolidate and reorganize (“ĭȳ”) domestic iron ore importers, CISA issued a notice that would see stringent controls on iron ore importers (the “Convention on the order and discipline of the steel industry’s iron ore import market” ȗȚǕ"ȆšȚƗƖDzĸƟølją`Ʊ) (Li, 2011a).  In April 2010, CISA and CCCMC also issued three key notices: “2010 Iron Ore Import Enterprises qualifying standards and application process” (	ôȚƗƖȆšJ"ǴDZŏk•ƇĘƢø; “Regarding Iron Ore Imports Representatives Regulations” (b6ȆšȚƗƖCƃwÛijƳs and “Iron Ore Importing Contracts and Reporting Flow   115 Registration Guidelines” (ȆšȚƗƖ¢¤VčĘƌǠ•cŨ§ƌǠÉŒǚÚ (Li, 2010).  Among the newly tightened “Iron Ore Import Qualifying Guidelines,” the minimum quantity of imports of iron ore by the enterprise the previous year increases from 700,000 tons to 1,000,000 tons, the amount of registered capital increases to 50 million Yuan, and the minimum credit rating is established at 2A. The number of firms allowed a license to import iron ore thus dropped further from 112 to 105, of which there were 65 steel producers and 40 traders at the time (Zhao, 2013). The overall pattern of these guidelines from 2005 to 2010 was to gradually decrease the number of enterprises with access to an import license (Where are the Chinese iron ore negotiations heading? 2010). The goal of these efforts was framed as reducing China’s dependence on the foreign iron ore export “monopoly” (“Àİ ”) (Zhao, 2013) by strengthening the hand of Chinese importers.  Figure 21 - Share of Total Iron Ore Imports by Top 2 Chinese Firms (2001 - 2011) Source: China’s Customs Department data (2001, 2006, 2011). Author’s calculations. Despite these efforts, the share of imports by specific Chinese domestic steel companies continued to decrease over the same period. Data from the China Customs Department allows us to observe the increasing level of fragmentation at the interface with the global iron ore market, even with the companies officially licensed to import the ore. In 2001, the two largest importers of iron ore in China (Baosteel and WISCO) imported 24% of 2001# 2006# 2011#Number#1#importer#(Baosteel)# 19# 11# 6#Number#2#importer#(WISCO,#Sinosteel,#Hebei#Iron#&#Steel)# 5# 5# 4#0#10#20#30#40#50#60#70#80#90#100#%"Share!of!total!iron!ore!imports!by!top!2!Chinese!Oirms!  116 total imports. By 2011, the proportion of imports by the two largest importers (Baosteel and Hebei Iron & Steel) had dropped to only 10% (see Figure 21).  As we can see, at the time of China’s emergence as the number one iron ore importer, the country was firmly situated in Quadrant 2: fragmented domestic consumers facing highly concentrated global producers. 4.8.3 The fall of the iron ore benchmarking regime  The spot market did not really exist before 2000, and got its start mainly for iron ore from India to China, as established producers (Brazil, Canada, Australia) kept the annual pricing system (Interview 126, Jan van Veelen, iron ore industry insider and former Sales Director at several key iron ore mining companies, 2013).   In 2003, China’s iron ore imports surpassed those of Japan. Until then, Japan had been the lead benchmark price negotiator. The “Japanese steel mills [were] leading the discussions, (…) in close cooperation with the Germans and other major buyers. I often read that these discussions were secretive, [but] only for the outsiders; the sellers and buyers [producers and consumers] knew exactly what was going on!” (Interview 126, Jan van Veelen, iron ore industry insider, 2013).   The Chinese were uneasy sitting on the sidelines and accepting the benchmark price every year, as they now were the world’s largest importer of iron ore.   “The involvement of Chinese buyers in the negotiation process carried the effect of weakening the bargaining power of Japanese firms on the buyers’ side, delivering the Big 3 miners a whip hand to push for rapid price increases in the context of booming Chinese demand. As a result, prices for internationally traded iron ore and coking coal started to boom, with iron ore alone increasing five-fold from USD 36 to 144 per tonne in four years between 2005 and 2008.” (Wilson, 2010)     The leading Chinese importer, Baosteel, initially took the lead in representing the Chinese purchasers in the annual benchmarking negotiations. After failing to reach an agreement for three years in a row, in December 2006, for the first time, Baosteel (which was acting as the representative of 260 Chinese buyers at the time) agreed to a benchmark with the Brazilian company, now known as Vale, ahead of the Japanese buyers (Craze and Blount, 2006).    117  Baosteel negotiated an increase of 19% from the previous year, well below expectations at the time (Wilson, 2012). Baosteel’s lead continued in 2007 (10% increase) (Craze and Blount, 2006). In 2008, Baosteel led the negotiation again, and was forced to agree to a 65% rise over the previous year’s benchmark price (Baosteel, 2008). This time around, the price rise was significant. Baosteel’s share of the domestic Chinese market was small – it accounted for 6% of Chinese imports at the time – and this weakened the position of the steel mill (The lore of ore, 2012). In 2009, the Chinese government then “attempted to use its massive demand for iron ore as a lever against iron-ore producers by uniting all of China's steel companies under a single negotiator: state-run CISA. CISA wrestled control from Baosteel to lead negotiations in 2009. Collective action problems were to plague the industry over the next two years.   In 2009, in the midst of the Global Financial Crisis, CISA took a tough negotiating strategy with the three biggest global iron ore producers (Rio Tinto, BHPB and Vale), and boldly asked for a 40%-50% markdown (China: Lessons, 2009). Despite the fact that Japanese and Korean mills signed contracts in early 2010, with an appreciation of between 90 and 100% over the year before, CISA gave strong assurances that it would continue negotiations, refuse the “sky high” prices and look to secure long-term contracts (Where are the Chinese iron ore negotiations heading?, 2010).  The three major iron ore suppliers respond to China’s strategy by proposing that pricing follow an emerging “index pricing formula” (Wilson, 2012). CISA refused, and asked all domestic steel companies to refrain from buying on the international market while the negotiations were ongoing. But CISA's strategy backfired and negotiations stalled. CISA “asked its steel companies to refrain from importing iron ore from three major mining companies in an effort to pressure them during price negotiations” (China: An Iron Ore Ultimatum?, 2010). Despite the central government urging its national steel makers not to negotiate directly with the iron ore producers, “the major Chinese steel companies had broken ranks with the CISA and signed contracts of their own with foreign suppliers. (…) This forced the remaining steel companies to fend for themselves on the spot market. (…) Spot prices rocketed up to well over $100 per metric ton in the months following the failed negotiations” (China: Lessons, 2009).   118 Table 25 - Chinese Iron Ore Imports by Country and Price Paid (2008) Country Quantity (tonnes) Share of total (%) Price per ton Australia 183,400,058 41.4 122.4 Brazil 100,619,615 22.7 148.5 Other 159,485,407 35.9 Average = 137.4 Total  443,505,080 100 Average = 144.6 Source: China Mining Yearbook 2010  A multitude of small Chinese steel firms thus ignored CISA’s request and purchased iron ore on the international market in individual "spot" contracts (2009 was the only year contract prices were above spot prices, with the spot below $60, see Table 25 and Figure 22). One Chinese iron ore industry insider from the private sector confirmed this to me when he said that: “There are many actors on the Chinese side, and they do not act together. For instance, the big steel mills act through CISA, whereas the private sector is more nimble” (Interview 69, international iron ore industry high-level executive, 2012).   The lack of capacity on behalf of the lead negotiator for coordinating domestic iron ore importers was compounded by infighting between two key stakeholders. On April 5, 2010, CISA and CCCMC held a closed-door meeting in Henan Building to discuss the “establishment of the iron ore market’s orderly alliance” (ǚǐȚƗƖð¼Ɵøǀ¢Lǟ). The result was the adoption of the three industry self-regulation documents (mentioned above), which were established to put an end to the reselling of iron ore by license-holding companies to smaller domestic companies. The Stern Hu affair, in which a dual citizen from Australia and China and a Rio Tinto employee convicted for bribery in China in 2010, was directly linked to the domestic difficulties surrounding the iron ore import license system. At the time, Chinese companies, including Sinochem and China National Building Materials Group, along with 20 others, were exposed in the Rio Tinto bribery scandal. As discussed in the Chinese media, mostly private Chinese companies were suspected of conspiring with Rio Tinto employees and licensed Chinese companies to import iron ore (in exchange for a bribe) to then resell on the Chinese market for a large profit (Xie, 2011).   In the middle of an international benchmarking negotiations deadlock, the failed investment in Rio Tinto, and the Stern Hu scandal, CISA and CCCMC attempted to wrestle control over the domestic iron ore importing situation and to make peace (media reports   119 hinted at previous infighting and efforts at “making peace” (Øe) (Where are the Chinese iron ore negotiations heading?, 2010). This is referred to in another news article even more directly as “Making internal peace is necessary before resisting foreign aggression!” (ĦÊć\Øe!) (Li, 2010).  Newspaper reports mention a “joint petition” to the upper levels of government by the leaders of Chinese steel SOEs, asking that the iron ore (negotiation) difficulties be raised to a “national level issue” (4ÛŘyĿŬčơĬÞ'¸ȗJƿċǀ¥“/”ǙŝâȚƗƖȞȶ‰v¸Þçȭ(Where are the Chinese iron ore negotiations heading?, 2010). CISA and CCCMC eventually established a joint office (ȆšȚƗƖǀ¢{`Ü) to control the imports of iron ore resources, collect all import contract information and monitor iron ore flows in an attempt to curb speculative iron ore behaviour (ĒpŹƗǕ+) (Li, 2010).  CISA Secretary General Shang Shanhua was reported saying that: “Rectifying/consolidating the iron ore import market is necessary for two reasons: first, to prevent lesser quality ore from entering the country; and second, to prevent iron ore reselling/speculative practices”65 (Li, 2010).  Despite the legal efforts to curb the number of domestic companies importing ore, many companies, including those without licenses, continued to bypass the domestic Chinese regulatory system. Articles refer to the fact that since 2007, despite the fact that there have been 112 official import licenses, at least 200 companies were in fact importing iron ore (Where are the Chinese iron ore negotiations heading?, 2010). A journalist interviewed a Beijing-based iron ore company that at the time had been selling between 500,000 and 2,000,000 tons of iron ore a year on the Chinese markets for three years, not with an official license, but with a “purchased license.” The process was very simple: a company without a license would arrange the contact, price and delivery with a foreign seller, then ask a domestic license holder to apply their own official seal on the contract in exchange for a fee. As per the journalist’s interview, fees ranged from 0.1 to 0.7% on a 50,000-ton cargo, which would be worth upwards of 20,000,000 RMB. In some cases the fee could reach 140,000 RMB (Zhao, 2013).                                                   65 “ĭȳȆšȚƗƖð¼ȹ\åĹǙňƶȅ8¢őƍƗ:°Ũ§¸e¤Ķ.âȌw¹DzĸȴÄƍŹƗǕ+”Author’s translation.   120  Another senior journalist confirmed that if only a few steel companies had the required permit to import the ore, other politically connected companies were still managing to import the ore and resell it domestically. As such, he explained, “many companies in China have an interest in iron ore prices going up. This is a problem” (Interview, 8, 2012). Indeed, the large SOEs benefitted from the benchmarking system (Li, 2010). “Some steel mills make more profit by selling ore to small ones than selling steel” (Zhang, 2009b). A Chinese academic summarized the situation as follows: “The fragmentation of the iron ore market explains a lot. It is so difficult for China to have an iron ore strategy, since all the iron ore companies don’t listen! The competition among Chinese companies leads to higher prices” (Interview 60, 2012).   During the 2009-2010 benchmark negotiations, CISA did not have enough sway over Chinese importers. It proposed taking away the licenses from the renegade Chinese steel mills and traders, a move which MOFCOM opposed for fear of endangering China’s resource security (Zhang, 2009b). It also tried to block the import of ore by individual Chinese purchasers who were paying well above the previous year’s price (Bunker and Ciccantell, 2007), with no success. In addition, CISA had overestimated the three big suppliers’ need to sell (Gu, 2009), compared to the Chinese mills’ need to buy iron ore.   CISA faced “harsh criticism (…) for mishandling the 2008-2009 negotiations as an out-of-touch government entity with no experience in the world of trade” (China: Lessons, 2009). Following the failure of its negotiations led by CISA in 2009,  “Executive Deputy Chairman Luo Bingsheng, General Secretary Shan Shanghua, and Chen Xianwen, the director of CISA’s market investigation department who was lead negotiator during the failed 2009 price negotiations, all submitted their resignations at the organization’s annual meeting Feb. 20. The three are blamed for CISA’s failed strategy of leveraging its role as megaconsumer in the global market to extract deep discounts in which CISA refused to make concessions. (...) From China’s point of view, the reason CISA failed in previous negotiations was its inability to prevent small steel miners from breaking ranks and striking their own deals with iron ore producers. The government is attempting to hasten the long-running consolidation of the steel sector to enable CISA to present a united front in future.” (Chinese Dependence on Foreign Iron Ore: A Special Report, 2011).  A Chinese industry insider explained: “During the 2008 negotiations, CISA had a cultural problem. The perception was that the foreign companies were there to trick us (Ƚďȝ). There was a big feeling of a struggle against a foreign discrimination. They hadn’t   121 thought through the consequences” (Interview 46, 2012). Another Chinese government official confirmed: “CISA supported the benchmarking system; they were conservative, and it was Australia who wasn’t willing” (Interview 59, 2012). One high-level international industry insider commented that “the broad direction towards spot pricing globally had a lot to do with the arrival of China.” He continued:  “There is a saying in Chinese: “You lift the rock and it falls on your foot.” This was the situation with iron ore. (…) The fall of the benchmarking system is particularly bad for the purchasers. The Chinese had no intention of making the benchmarking system fall! What happened is that the spot market started to be higher than the benchmark, after the arrival of China on the scene, whereas the benchmark had been very stable for decades. Every year, the benchmark would go up, but the spot price would be even higher. The only year when the spot was a little below the benchmark, the Chinese said: “We are going for the spot price!” It was a mistake!! BHP and others were only too happy! It was more in their favour. So the benchmark system fell down, and this was a big blow to the Chinese steel companies. So short-sighted.” (Interview 86, 2012)   The fact that the Chinese negotiators did not intend to cause the fall of the benchmark pricing system was confirmed yet again by a senior Chinese government official (Interview 95, 2012). But the strongest confirmation came from a Chinese government official with intimate knowledge of the events. I asked whether in 2009/2010, CISA wanted to end the benchmarking system when it led the negotiations. The interviewee responded, very firmly: “No. CISA did not want to end the benchmarking system at that point. They just thought that the price was too high. It wasn’t in their intention to see the end of the benchmark, nor was it Baogang’s” (Interview 111, 2012).    All in all, speaking with dozens of interviewees, I never once met someone who said that the fall of the benchmarking system was a well thought-out plan by the lead Chinese negotiators, or even powerful Chinese interests (government organs or large firms). In the end, as an indication of the perception of a failed mission, “the top CISA guys left after the fall of the benchmarking system” (Interview 112, 2012).  4.8.4 The three international producers’ response  Interviews with iron ore suppliers confirm that the three firms were initially divided, as the Global Financial Crisis unfolded, on whether to maintain the annual benchmarking system that had been in place for decades. A Chinese industry insider explained: “For a   122 moment, the spot price was above the benchmark and that was when the three big companies made the jump” (Interview 98, 2012). Other interviewees pointed to a longer-term strategy.  Of the three, BHP Billiton was most clearly in favour of a transition towards a quarterly, and eventually a spot market, system, whereas Rio Tinto’s management took a more traditional approach, in part because its main Asian clients (Japanese and South Korean steel mills) valued the stability brought by the benchmarking pricing system. When I asked a high-level executive in the iron ore industry whether BHP was the most vocal about its interest in having the benchmarking system fall apart, he replied: “Yes, they were very vocal about it” (Interview 69, 2012). At the time, the CEO of BHP (Marius Kloppers), is said to have been “pretty keen on seeing the fall of the benchmark” (Interview with international long-time iron ore industry insider, 2013), because he estimated the spot price would rise above contract prices. He was right. “Vale, at the beginning, didn’t want to know anything about the spot prices and indexes. But remember, they changed management completely66. (…) The president and commercial directors and other top managements left and a new team was brought in, who was a more open to the spot cargoes and the volatile prices, etc. (…) so they changed the bidding culture” (Interview, international mining industry insider, 2013). Rio Tinto followed BHP’s and Vale’s lead, the last to make the move in 2010, and the quarterly contract was introduced. One iron ore industry insider close to Rio Tinto said: “During the fall of the benchmark, was Rio more reticent? Yes, BHP was the early mover. But now we wonder if we shouldn’t have made the move earlier! Why? We figured out that it was indeed a profitable move. BHP saw it before us, and we now have come to the realization that it was indeed beneficial and we could have moved earlier” (Interview 134, Iron ore industry insider, 2014).   A small emerging spot market for iron ore had come into existence a few years earlier as China started to import iron ore from India, both of whom were outside of the benchmark pricing regime negotiations at the time. As Chinese firms reneged on their lead negotiator’s demands, and the three large iron ore suppliers were only too happy to follow suit, spot pricing gained pre-eminence, and thus the decades-old benchmarking era had come to an end.                                                   66 Vale was privatized in 1997 (Ericsson, 2004).   123  While it is clear that the three big suppliers, led at first by BHP and Vale, actively pushed for spot pricing to gain pre-eminence, this would not have been possible without China. The Chinese firms created the opportunity for the global mining firms to financialize the market by rapidly emerging as number one consumers, disrupting the global market balance of power and changing the terms of trade. But their role went beyond this. The Chinese stakeholders’ uncoordinated procurement behaviour was the result of a failure of collective action, or in other words, a failure of the lead negotiators to enforce a procurement strategy. One high-level executive from one of the three big iron ore suppliers confirmed this in an interview with me: “It is the growth of the Chinese industry that caused the spot market to emerge” (Interview 69, 2012). But even beyond that, one Chinese industry insider added: “The first step was the market evolving from a benchmark to a quarterly contract system. But the Chinese walked from contracts and the pricing system evolved increasingly close to a spot system” (Interview 112, 2012). In other words, Chinese stakeholders’ behaviour may have been unintended, at least in the aggregate, but it was not passive.   Needless to say, the iron ore suppliers saw their profits skyrocket in subsequent years. By the time the negotiations fell apart and all Chinese consumers were forced to fend for themselves on the quarterly or spot markets, prices were already above US$100/ton. The shift in the iron ore pricing regime “gives major producers more leverage. (…) The control exercised by the Big 3 will, to some extent, counteract the tendency to greater price instability that will result from the new pricing methods” (Ericsson et al., 2011).   124 Figure 22 - Iron Ore Prices Benchmark and Spot Pricing Regimes (1975 - 2015)  Source: Michael Komesaroff, China Economic Quarterly, June 2010; Urandaline, 2011; Keith Tan, Platts, 2012; China Mining Conference, 2012; Reuters, 2013; Wall Street Journal, 2013, World Bank Data, 2015. Author’s calculations.  On July 1, 2013, MOFCOM stated that in order to facilitate trade, automatic import licenses for iron ore could be applied for online (Chinese Ministry of Commerce, 2013). A month earlier, in June 2013, MOFCOM released a statement assuring observers that this new automatic online importing system for iron ore would not constitute a new kind of restriction on iron ore trade. In fact, a MOFCOM employee was quoted as saying that: “After July 1, provided [one] is allowed to operate as an import and export business, one can apply for an iron ore automatic import license, there are no other restrictions or audit standards, the qualification assessment is canceled” (Zhao, 2013). One analysis explains that “providing the company has the right to conduct import-export business, it can automatically apply for an iron ore import license” (Zhao, 2013). 67 Further monitoring is necessary to ascertain the impact of the relaxation of Chinese iron ore importing rules.  Some industry analysts argue that the licensing system had slowly brought chaos to the industry since 2005. Chinese industry specialists such as Liu Wenlu, Deputy General Manager of Steel Home (a Chinese steel industry consultancy), said that recent trends would                                                  67 “ÉȆƴǒŅȴȚƗ	~ȆǢǥ” (Author’s translation) 0#20#40#60#80#100#120#140#160#180#200#1975#1980#1985#1990#1995#2000#2001#2002#2003#2004#2005#2006#2007#2008#2009#2010#2011#2012#2013#2014#2015#Nominal!USD!Iron!Ore!Prices!1975J2015!Spot#Benchmark#  125 go a long way toward breaking down the Chinese iron ore import monopoly, and would be in the interest of Chinese steel mills which can now procure the resource from more channels (Zhao, 2013). He continued by saying that the restricted import licensing system did not actually have a decreasing effect on prices or even an impact on supply diversification. A CISA senior official was quoted as saying that the removal of the import license qualifying restrictions was a positive step for the Chinese steel industry (Zhao, 2013). He then went on to say that the licensing system not only failed to fulfill its original goals of restricting iron ore importing agents (“ĢǕȚƗƖCƃwƍƑŏ), but also resulted in the “reselling issue” (“WŽǴDZ) (Zhao, 2013).  On May 8, 2012, CBMX, China’s own iron ore spot trading platform, was launched in Beijing. CISA backed the platform publicly. This public backing came years after CISA blocked an earlier attempt to create an indigenous iron ore trading platform in 2009, one year prior to the fall of the benchmarking regime. This shows that at the time, CISA was still invested in the benchmark system. GlobalOre, a rival platform backed by BHP, was launched in Singapore on May 30, 2012, only three weeks after the launch of CBMX. CBMX officials voiced confidence that the Chinese iron ore market dynamic is mature enough not to be overly affected by the cancellation of the iron ore license process. In fact, the new regulation merely ends up endorsing what had become a fait accompli. Discussion  The fall of the decades-old international benchmarking regime – which ushered in the era of quarterly, and then spot, pricing, as well as the emergence of spot trading platforms – was a consequence of the asymmetric encounter between fragmented Chinese iron ore consumers and concentrated global iron ore producers.  The uncoordinated, fragmented procurement behaviour of Chinese consumers between 2008-2010 disrupted established patterns of exchange between traditional benchmark negotiators. Whereas Japan’s consumers were well organized and coordinated, Chinese consumers presented fragmented, uncoordinated behaviour to the big three miners. This led to a shift in incentive structures and the emergence of opportunities for global iron ore producers to usher in their preferred pricing regime. The asymmetric position of the three big producers’ market power explains their room for manoeuvre in supporting the regime change. This led to the overall marketization of the iron ore pricing regime.    126  The Chinese side of the story is one of failure of collective action. Critically, the lead Chinese negotiators, Baosteel and CISA, were not able to coordinate the behaviour of the other Chinese iron ore consumers. Stakeholders in the domestic steel industry in China held different positions regarding the benefit of a benchmark system because of the licensing system put in place, which gave privileged access to the global iron ore market to some domestic Chinese consumers over others. This system existed for domestic reasons: “When MOFCOM issues more licenses for import, they have the domestic situation in mind, not the potential impacts on global market institutions... some issues they consider…competition among domestic providers … but they are unlikely to have thought through the likely international impacts” (Interview 128, Chinese government official at a central government agency, 2012).  It is important to underline that CISA and the large state-owned steel companies in China – by far the most powerful players overall – did not want the benchmark pricing system to fall. Indeed, they benefitted from the existing import system and were able to extract rents from other Chinese firms that had no access to the import interface by reselling the ore domestically at a profit.  On the other hand, the smaller mills had an interest in fending for themselves on the global market. In addition, there were power struggles between Baosteel and CISA between 2008-2010, all of which led to a fragmented interface with the global iron ore industry and the unraveling of a decades-long pricing regime.   Overall, the result is ironic as seen from China. It is by far the largest purchaser of iron ore globally, but unlike in the case of Japan (Japanese steel mills had superior pricing power over the Australian and Brazilian producers for decades, see below), it has very little control over the global iron ore pricing regime (Wu and Wu, 2009). This is the case even if the “Japanese quarterly price adjustment scheme reflects Chinese import prices” (Interview 136, international iron ore industry insider, 2014). One Chinese academic commented: “China is the largest iron ore importer, and it thought that this would provide it with a strong hand to influence the market, but it found itself in a position of weakness, and it doesn’t know what to do about it” (Interview 7, China, 2012). This has led to scores of bold comments by CISA, the NDRC and other Chinese agencies about the unfairness of the global iron ore market.   127  It is important to note that China’s and the three big producers’ behaviour has affected other iron ore consumers beyond themselves. “The traditional consumers of iron ore (European and Japanese steel mills) detest the development, as it appears not to be in their interest and has severed the relationship between producer-consumer” (Interview 126, Jan van Veelen, iron ore industry insider, 2013). In other words, China’s emergence as the number one iron ore consumer led to a systemic change in the iron ore pricing regime.   4.9 Case 2: China’s Impact on the Global Iron Ore Shipping Pricing Regime 4.9.1 Dependent variable: The iron ore shipping industry pricing regime  Another part of the international iron ore market that experienced changes as a result of China’s emergence was the shipping market. China’s emergence as the largest iron ore consumer transformed incentive structures in the global shipping market through dramatic increases in freight rates. Indeed, by the late 2000s, “consistent with the story of soaring iron ore demand, the dry bulk shipping market [had] registered the highest real freight rate for nearly 50 years” (Lu et al., 2009, p.359).  The commodities boom of the first decade of the 2000s had a dramatic impact on shipping costs. “The volume of seaborne trade has increased by a steeper trend in all major commodity categories since 2002. Among them, five major dry bulk sectors displayed the most conspicuous growth, averaging 9.5% annually. Iron ore, as the largest single item of dry cargo in seaborne trade, increased its share in world seaborne trade from 11% in 2002 to 15% in 2007 (UNCTAD 2008), a record never reached ever since iron ore became a major ocean shipping item” (Lu et al., 2009, p.357).    Prior to China’s emergence as the world’s top iron ore importer, Japan had established a very favourable and stable system by which iron ore was priced FOB68, and the Japanese steel mills chartered the bulk cargo ships themselves. This changed with the emergence of China, the meteoric rise in iron ore demand, and a parallel increase in shipping rates. The dramatic increases in freight rates shifted the global iron ore producers’ comparative                                                  68 “Free on Board (FOB) suggests that the seller pays for the transportation of the goods only to the port of shipment, which includes the cost of loading the goods on the cargo ship, inland haulage costs, customs clearance, origin documentation charges, demurrage, if any, and port handling charges. Beyond this point all costs are borne by the buyer of the goods. Cost and Freig