UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Political participation on the prairies : historical institutionalism and the economic determinants of… James, Hammond 2018

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

Item Metadata


24-ubc_2018_february_hammond_james.pdf [ 601.64kB ]
JSON: 24-1.0363070.json
JSON-LD: 24-1.0363070-ld.json
RDF/XML (Pretty): 24-1.0363070-rdf.xml
RDF/JSON: 24-1.0363070-rdf.json
Turtle: 24-1.0363070-turtle.txt
N-Triples: 24-1.0363070-rdf-ntriples.txt
Original Record: 24-1.0363070-source.json
Full Text

Full Text

    Political Participation on the Prairies Historical Institutionalism and the Economic Determinants of Voter Turnout in Alberta and Saskatchewan     by   JAMES HAMMOND  B.A., (Honours, First Class), The University of Durham, 2016     A thesis submitted in partial fulfilment of the requirements for the degree of:  MASTER OF ARTS  in  THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES  (Political Science)      THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)    January 2018    © James Hammond, 2018   ii Abstract  Turnout at Canadian elections has been declining considerably since the 1980s yet much of the literature on turnout is dominated by behaviouralism which offers value in predicting the voting behaviour of individuals but struggles to account for long-term trends at the aggregate level. This is particularly pressing in Canada where turnout differs substantially between provinces, nowhere more so than between Alberta and Saskatchewan whose political differences, despite their similarities elsewhere, illustrates the regionalized nature of politics in Canada. This paper will seek to apply the two most prevalent theoretical understandings of regionalism, formal institutional approaches and political culture approaches, to demonstrate that neither offers entirely adequate explanations when tested against such challenging cases. Instead it will be argued that to resolve the undetermined causal logic of these approaches to understand turnout in Alberta and Saskatchewan, the historical agricultural development of the provinces must be examined to understand the formation of cleavages along class and ownership of land that continue to determine political behaviour. Such a re-conceptualization will be presented through the lens of historical institutionalism, which argues that historical economic processes create informal institutions, as norms and rules of behaviour, that are relatively static and durable in the face of subsequent change but continue to frame perceptions and guide decision making, thereby determining the political behaviour of individuals. By considering the historical sources of political differences in Canada and how these attitudes emerged through relationships of power and economic exploitation with other parts of the country, it is hoped that more productive steps can be taken to begin rebuilding trust in state institutions and creating the conditions for the evolution of new norms which may foster greater participation.   iii Lay Summary Turnout at elections in Canada is falling substantially yet existing understandings of why people decide to vote struggle to offer answers beyond the individual. This is particularly important in Canada where different regions of the country display markedly different levels in turnout, particularly between the otherwise similar provinces of Alberta and Saskatchewan. While theories that focus on the effect of institutions or the political culture of each province provide some understanding of the reasons for these differences, the key determining factor is often not clear. This paper will instead argue that in the case of Alberta and Saskatchewan these other differences emerge from economic and agricultural patterns of development, which historically created different classes, based around land ownership and agricultural activities, which developed differing political attitudes, particularly regarding voting at elections.   iv Preface This thesis is original, unpublished, independent work by the author, James Hammond.   v  Table of Contents Abstract.................................................................................................................................... ii Lay Summary ......................................................................................................................... iii Preface..................................................................................................................................... iv Table of Contents .....................................................................................................................v List of Tables .......................................................................................................................... vi List of Figures........................................................................................................................ vii Dedication ............................................................................................................................. viii Introduction..............................................................................................................................1 The Puzzle on the Prairies.......................................................................................................4 Literature Review ....................................................................................................................9 Behaviouralism.....................................................................................................................9 Institutionalism ..................................................................................................................11 Political Culture .................................................................................................................13 Political Economy...............................................................................................................18 Assessing Institutions.............................................................................................................22 Assessing Political Culture ....................................................................................................30 Assessing Political Economy .................................................................................................38 Conclusion ..............................................................................................................................54 Bibliography ...........................................................................................................................58    vi List of Tables Table 1: Selected birthplace of population by proportion, Alberta, Census 1906-1921............... 35	Table 2: Selected birthplace of population by proportion, Saskatchewan, Census 1906-1921.... 35	  vii List of Figures Figure 1: Western Canada Turnout 1975-2016 .............................................................................. 7	Figure 2: Alberta 1975-2015 - Electoral Competitiveness and Voting ........................................ 25	Figure 3: Saskatchewan 1975-2016 - Electoral Competitivenss and Voting ............................... 26	Figure 4: Total Acres of Wheat (1921-2006) ............................................................................... 44	Figure 5: Proportion of all Farm Area, in Acres, Growing Wheat (1921-2006) .......................... 45	Figure 6: Total Number of Cattle and Calves (1921-2006) .......................................................... 46	Figure 7: Proportion of Total Farms Reporting Cattle and Calves (1941-2006) .......................... 46	Figure 8: Average Number of Cattle and Calves per Farm (1921-2006) ..................................... 47	   viii Dedication    In memory of Cyril Doward, who grew up amongst the wheat fields and under the eternal sky.      Wheat kings and pretty things Let’s just see what the morning brings        1 Introduction Turnout at Canadian elections is falling considerably. While the most recent federal election in 2015 saw an increased turnout of 68.3%, the general trend in recent elections has been downward (Elections Canada). Between 1993 and 2011, turnout at federal general elections averaged just 63.3%, compared to an average of 74.8% throughout the post-war period until 1993. The 58.8% of registered voters who turned out to vote in 2008 was the lowest in Canadian history. Comparatively, this decline is not as serious as that experienced by many other established democracies, yet it suggests a worrying trend that mirrors the phenomena of growing political apathy in other democracies. Such a decline is troubling; Arend Lijphart highlighted the significant challenge to the legitimacy of the democratic process that such trends present, emphasising in particular the corrosive influence of growing voter apathy on systematic biases and unequal class influence on the political process (Lijphart, 1997). Yet attempts to explain the causes for this marked decline in turnout remain focused of behaviouralism. While productive at the individual level, the attempt to locate causal explanations with individuals risks imposing a positivist philosophy of science without the ability to explain durable long-term trends. This work will seek to challenge these explanations in the Canadian context where such individualised accounts for voting behaviour break down at the sub-national level, as Canada’s rich regional political cultures and differing institutional systems display marked differences in levels of turnout, demonstrating a need to consider the aggregate level. Indeed given that such differences hold even when typically cited individual determinants are controlled for (see Simeon & Elkins, 1974), it would seem that such variance requires an institutional focus to   2 explain the determinants of differing levels of participation. Moreover, in the tradition of Bent Flyvbjerg’s Making Social Science Matter, rejecting a purely positivist social science and focusing on the clearly identifiable determinants of voting behaviour potentially makes for research which is more relevant for policy makers seeking to reverse declining turnout and therefore engaged in the production of practical knowledge (Flyvbjerg, 2001, p.4). If the causes of such phenomenon can be accounted for, then this research becomes more than simply a process of explaining an interesting political phenomenon but rather engaged in renewing the Canadian political system and maintaining the health of its democratic process.  Given this intention to account for the identifiable causes of a decline in turnout at the aggregate level, this paper will undertake a comparative approach to identify the causes of markedly different levels of turnout in provincial elections in two Canadian provinces, Alberta and Saskatchewan. The sharp differences in turnout of these two provinces alongside what appears to be similar social and economic features, with an early economy focused on agriculture and unicameral, single member plurality representative institutions, presents an empirical puzzle in understanding turnout and political apathy in Canada. After reviewing the major literature on the behavioural approach to demonstrate its shortcomings in explaining such stark regional differences in Canada, this paper will strive to apply the theoretical assumptions of prevalent aggregate level approaches in Canadian political science literature on voting behaviour. Institutionalism, which seeks to account for the role of formal institutions in shaping voting behaviour will be analyzed using aggregate voting data from Alberta and Saskatchewan, demonstrating that formal institutional accounts which focus on the competitiveness of elections or the party system do not adequately explain identifiable patterns in this data. The respective political cultures of each province, in terms of attitudes and   3 orientations to participation will then be studied, demonstrating that to understand regional variations in turnout, cultural differences, particularly informed by Putnam’s social capital thesis, must be accounted for. Yet it will be argued that while successfully raising the role played by cultural attitudes, these accounts remain fixed on patterns of immigration and demographic change, neither of which is entirely adequate for cases that demonstrate long-term divergence in turnout. To adequately account for the continued persistence of divergent attitudes in the political cultures of the two provinces, this paper will seek to utilize the framework of ‘new institutionalism’ to analyse the historical economic structures which formed such attitudes, determining relatively static institutional norms of behaviour that have persisted even under conditions of subsequent sociological and economic change. In particular, by examining the different agricultural trends during the early years of the two provinces, it will be argued that early rural settlement and differing focuses on wheat cultivation and cattle ranching significantly shaped the emergence of class based institutions which continue to influence electoral behaviour. In this light, understanding and resoling the challenges of declining turnout requires more than studying the institutional processes in which elections take place or generalizing observable political attitudes as fixed cultural obstacles and instead requires acknowledging the pervasive influence of early economic forces of exploitation in Western Canada, particularly in shaping class and the institutionalized and deeply embedded norms of behaviour which are reflected in observed cultural and institutional differences.   4 The Puzzle on the Prairies Perhaps the most puzzling aspect of this striking decline in turnout at federal elections is the sharp variation is turnout levels between provinces. Donley Studlar highlights the significant absence of literature on turnout at the sub-national level in Canada, declaring that research needs to move beyond the national level (Studlar, 2001, p.318). Research that does exist of turnout at the provincial level, such as Siaroff and Wesley’s informative study of all 134 provincial elections between 1965 and 2014, display marked variations between provinces; from an average of 83.3% between 1965 and 2014 in Prince Edward Island to a just 56.1% in Alberta over the same period (see Siaroff & Wesley, 2015, p.151). For the authors, this variation illustrates the need to overcome the individual-level determinants that dominate the literature and instead emphasise the importance of context in Canadian elections (Siaroff & Wesley, 2015, p.148), suggesting a role for environmental circumstances in structuring voting decisions; that institutions as embedded norms influence voting behaviour beyond the rational judgements of individuals stressed by behavioural literature. In some respects these findings are not particularly novel, with a stress on institutions and regional variations in values returning to a rich body of literature on regional political cultures in Canada which suggested that regionalism overcame individual attitudes. In an influential piece, Simeon and Elkins described regional political culture as the “substantial regional divergences in some of the basic attitudes towards politics… which cannot be account for by an plausible control variables… differences in history, patterns of settlement, ethnic composition and even institutions” (Simeon & Elkins, 1974, p.398). Yet Siaroff and Wesley’s comparison of turnout across provinces does not conform to the typical regions presented in political culture literature; they find that the province with the lowest average turnout in provincial elections between 1965   5 and 2014 is Alberta, at 56.1% and while the highest over this period is Prince Edward Island, the second highest is Saskatchewan at 76.1% (Siaroff and Wesley, 2015, p.2015). Given the small size of Prince Edward Island, with a population just over 100,000, it could be argued therefore that Saskatchewan displays the highest rate of turnout amongst the provinces with mass political systems. It is deeply striking that Alberta and Saskatchewan, two neighbouring provinces, which both entered confederation in 1905 and whose border is simply a line that cuts across the prairies, should present with such widely divergent levels of turnout, a difference which averages 20% between 1965 and 2014. This puzzle therefore presents not only an intriguing political phenomenon in Canadian politics worthy of study, but amounts to a difference between the political apathy which so concerns democratic theorists and a healthy level of participation which still reaches the historical post-war levels. Moreover, given that this difference occurs not between distinct provinces from different regions of Canada, but from two provinces of ‘Western Canada’ typically held to display a similar political culture suggests that identifying the causes of this stark variation also offers the potential to overcome the focus on ‘regionalism’ in literature on political culture in Canada. The key question this research will seek to address therefore is whether the suitable explanatory variable which can explain the puzzling differences in turnout between Alberta and Saskatchewan can be explained as structured by formal institutional circumstances in which the elections take place, the wider cultural orientations of the citizens involved which serve as political cleavages or whether there are more deep, historically rooted structural forces which are the key dependent variable for this divergence. It will be demonstrated that while the factors presented in traditional institutional approaches, on party system and electoral competitiveness   6 seem to hold explanatory weight, they do not do enough to explain Saskatchewan’s trend for higher turnout. While an approach which identifies and measures the cultural differences between these two provinces, particularly regarding trust, begins to suggest the deeper causes of this trend, when the sources for such cultural divergence presented by existing literature is tested it increasingly appears that culture may simply be a dependent variable expressing more deeply rooted causes. It is for this reason that this paper will argue that a comprehensive explanation for variations in turnout between the two cases needs to turn to economic cleavages, particularly class. Informed by the work of historical institutionalists, a case will be made that these differences can be understood through how early patterns of settlement and economic development, particularly the agricultural differences between wheat cultivation and ranching developed informal institutions, as norms of behaviour in relation to trust and confidence in government, which have translated these early historical processes into behavioural orientations towards voting. It is productive therefore to begin this account by charting the differing trends in turnout, as measured by the proportion of registered electors who voted in an election, amongst the Western provinces, as Figure 1 does using data from 1975 to 2016, to illustrate the puzzle at the heart of this research. It should be noted that this comparison uses turnout at provincial elections to avoid the influence of specific electoral circumstances at the federal level and instead demonstrate the contained nature of the political circumstances in each province. Alberta consistently displays the lowest levels of turnout in recent provincial elections amongst Western provinces, dropping to its lowest historical level of only 40.49% in 2008.      7  Saskatchewan’s recent electoral history presents a more complex picture. Throughout most of this period, Saskatchewan displays the highest levels of turnout of any Western province, only falling below both British Columbia and Manitoba during the elections of 1995 and 1999, when turnout reached 64.4% and 65.5% respectively. Yet since a renewed peak of 76% in 2007, turnout in Saskatchewan has dropped sharply, eroding the significant share it has consistently held over Alberta. Of course in attempting to identify long-term trends in participation, such a recent change necessitates caution to avoid the risk of accounting for the influence of particular electoral contexts in turnout, with this downward trend which begins in 2011 coinciding with the first electoral victory for the Saskatchewan Party, yet it does begin to appear that perhaps something is changing in Saskatchewan and that levels of political participation are beginning to match trends elsewhere in Canada. It is for this reason therefore that the research question guiding this analysis will not be concerned specifically with the most recent rates of turnout, but attempt to identify the institutional or cultural causes of a long lasting historical era in which Figure 1   8 Saskatchewan and Alberta displayed comparatively divergent trends in turnout at provincial elections.     9 Literature Review Behaviouralism Existing theories of political participation that seek to identify the determinants of voting behaviour, including the causes for a decline in turnout, have largely assumed a behavioural approach to explaining levels of turnout at elections, seeking to identify such determinants at the individual rather than the aggregate level. In asserting that statistically observable relationships between measurable variables amounted to a scientifically robust means of identifying determinants of political behaviour, the approach quickly generated a whole set of individual characteristics, such as age, education and income, which could be used to predict how an individual voter behaved at elections. In particular, two theoretical frameworks for explaining voting behaviour came to dominate research; the sociological framework and the rational-choice framework. The sociological framework attempts to account for a series of observable relationships between certain social characteristics and voting patterns such as the statistically significant relationship between indicators of income, education and class and the probability of voting (see Lazarsfeld, Berelson & Gaudet, 1944). The most influential of such an approach, the Michigan model developed by Campbell et al, combined such sociological variables with a psychological approach, arguing that such behaviour was determined not just by membership of a certain sociological category, but that this identification is continuously reinforced through social networks forming a long-term psychological orientation (Campbell, 1960, p.18). As non-voting becomes normalised within certain sociological groups therefore, this behaviour becomes entrenched.   10  The alternative framework is based on rational choice models of voting behaviour, built on a tradition founded on the Downsian model which posits that voters judge the relative utility of voting, as a cost-benefit analysis of the potential personal benefits a government may offer (Downs, 1957). This model has been developed into an ‘instrumental’ theory of turnout which suggests that as the act of voting is a means to obtain desired aims, it is judged as being beneficial only as far as voting is assumed to impact electoral outcomes (see Aldrich, 1993).  While this highly instrumental approach to conceiving of the determinants of voting behaviour presents a convincing case for accounting for the range of factors that determine an individuals decision to vote and presents a functional means by which turnout could be increased, such as reducing the time and cognitive cost of the voting process, it struggles to grapple with the considerable evidence which suggests that the weak utility of voting makes doing so a marginal act (see Aldrich, 1993, p.274). Riker and Ordeshook have provided perhaps the most satisfactory counter to the claim that the weak possibility of influencing an electoral outcome makes voting irrational, but their attempt to account for the role ‘citizen duty’ plays in decisions to vote provides factors that are difficult to quantify (Riker and Ordeshook, 1968, p.36).  In sum therefore, both of these approaches have utility in explaining an individual’s decision to vote or not, but these frameworks struggle to account for the aggregate trend of declining turnout which concerns this research. The convincing evidence presented by Simeon and Elkins, amongst others, of distinct, regionalised, political cultures which effect voter’s perceptions on political participation certainly challenges that voting patterns can be explained at the individual level. While in some respects, utilizing rational choice models to explain how institutional circumstances, such as the competitiveness of elections, operate through calculations at the individual level hold considerable value in explaining the effects of institutions on turnout   11 levels, this makes such rational calculations an intermediary variable rather than the causal determinant of declining turnout. Moreover the assumptions of Aldrich and others, that the expected benefits in influencing electoral outcomes are so minimal as to make voting a marginal decision, would be expected to result in random and non-linear trends; yet as has been illustrated turnout has demonstrated a clearly identifiable and consistent downward trend for almost 25 years at federal elections. Unless the cost of voting has simultaneously risen with each election, this would suggest that other factors are influencing turnout beyond individual determinants. Institutionalism Perhaps the most promising source of alternative explanations for the puzzling variation in provincial turnout between Alberta and Saskatchewan, that moves such causal mechanisms from the individual level to the aggregate level, is through a focus on institutions and how the formal electoral process shapes voting behaviour. This shift towards a focus on institutions within the wider literature on electoral turnout was led by the pioneering work of Powell (1986) and Jackman (1987), who were expressly concerned with the variation in levels of turnout between different democratic states and were searching for a more explanatory approach. What both authors sought to highlight was that while individual political attitudes do shape voting patterns, their effect is limited because the voting behaviour of individuals is highly constrained by the institutional context, as Powell asserts; “participation is also facilitated or hindered by the institutional context within which individuals act… [that] present the individual with conditions that shape his or her choices, and are relatively difficult for the individual to change” (Powell, 1986, p.17). Building on Powell’s findings, Jackman sets out a specific set of institutional circumstances that he believes determine turnout. The most useful variables for this research found in his regression analysis of turnout levels in ‘industrial democracies’ are the presence of   12 competitive electoral districts, the proportionality of the electoral system and multi-party politics (Jackman, 1987, p.407-408). Of these variables, Jackman finds that competitive districts have the most significant effect on levels of turnout.  Jackman’s assertion of the important effect electoral competitiveness has on turnout conforms with a trend throughout the literature where positive correlations between competitiveness and voting are found when other variables are controlled for (see Hofstetter, 1973; Jewell & Olson, 1982). The assumptions throughout much of this research remains informed by rational choice, the mechanism by which competiveness influences the decision to vote as a rational calculation of the likelihood of voting influencing the outcome of an election. However, other scholars have sought to explain the mechanism by which competition influences turnout differently, with Hofstetter instead approaching the relationship from a political culture approach, arguing that competitive political environments induces norms of civic participation which therefore lead to high rates of turnout (Hofstetter, 1973, p.366).  Work in the Canadian context on the effect of formal institutions on turnout has placed an emphasis on competitiveness. Analysing turnout and the competitiveness of elections at the federal level between 1988 and 2004, Johnston, Matthews and Bittner examine variables of competitiveness and turnout at the level of each riding to demonstrate the role played by a voter’s perception of how competitive an election will be (Johnston, Matthews & Bittner, 2007). Their work does much to demonstrate that at the federal level, competitiveness does have an impact, pointing alternating patterns of competitiveness to account for variations in turnout over the period. However, when this data is run through a regression analysis, the authors find that competitiveness has less of an impact than attitudes of alienation or indifferences amongst voters, indicators that are taken from the Canadian Electoral Survey (Johnston, Matthews &   13 Bittner, 2007, p.741). Indeed throughout the study, the authors acknowledge the difficulty in identifying the mechanism by which strategic circumstances are translated into behaviour, highlighting the role that other attitudes may have in framing considerations on the utility of voting and the competitiveness of an election (Johnston, Matthews & Bittner, 2007, p.741).  The literature that seeks to apply such an institutional analysis, with a focus on electoral competitiveness, is significantly smaller at the provincial level and it is here therefore that this study seeks to help address such a gap. One of the few works to analyse turnout at the provincial level is Siaroff and Wesley’s 2015 aforementioned study of turnout between 1965 and 2014 (Siaroff & Wesley, 2015). The authors assert that ‘context matters’ in determining political participation, including the context in which an election takes place, particularly the competitiveness of the election and the degree of multi-partisim of the election (Siaroff & Wesley, 2015, p. 159). In conclusion therefore, two particular explanations which seem relevant for the question posed by this study repeatedly emerge from the institutional literature, the effect the competitiveness of elections has on turnout and the influence of the structured party system in mobilizing and engaging with voters. Given that there are only a few examples of this prevalent approach being applied to differences between Canadian provinces, these two competing explanations seem to offer the most value in a direct comparison between the two cases of Alberta and Saskatchewan. Political Culture An alternative explanation for differences in political behaviour between regions is the political culture approach, which authors writing in the Canadian context have drawn on to posit that regionalized political attitudes can explain electoral behaviour (Simeon & Elkins, 1974;   14 Elkins & Simeon, 1980). The more traditional accounts presented by Simeon and Elkins serves a valuable role in both illustrating the existence of distinct, regionalized political cultures which act independent of their institutional setting and which offer measurable indicators of such political cultures. Drawing data from post-election surveys undertaken in the 1960s, Simeon and Elkins develop two indicators of political attitudes, efficacy and trust and apply these indicators comparatively across provinces. The authors posit that these two dimensions, efficacy as an individual’s sense that they can be personally effective, and trust, that a citizen feels governments are interested in their welfare, have a causal relationship with political involvement (Simeon & Elkins, 1974, p.404-405). As Simeon and Elkins’ writing is influenced by Almond and Verba’s concept of civic culture, they become focused on developing typologies of political orientations to categorize provinces’ political culture, as ‘citizen societies’ or ‘disaffected societies’, trying to find their own civic culture in Canadian provinces (Simeon & Elkins, 1974, p.415). This effort to enforce such a dichotomous classification on the range of political cultures across Canada serves to obscure rather than highlight cultural differences between provinces, with the authors admitting that both Saskatchewan and Alberta do not fit either group (Simeon & Elkins, 1974, p.434).  Yet within their work is also the helpful beginning of quantifying and measuring the cultural differences between provinces in the political attitudes of citizens. Their research brings together a broad range of survey questions to compile two basic variables of efficacy and trust and the quantification of these indicators illustrate the stark attitudinal differences between provinces; the authors find that in measurements of trust, “Alberta seems somewhat out of place”, scoring considerably lower levels than other prairie provinces (Simeon & Elkins, 1974,   15 p.434). Moreover, the authors confirm that these differences in political attitudes remain significant even when other variables, including sociological factors and party identification, are controlled for (Simeon & Elkins, 1974, p.417-421). This account of the existence of a distinct set of political attitudes in Alberta is supported by Elkins’ later work in Small Worlds, indexing the strength of attachment to a ‘Canadian national identity’, Elkins finds that Alberta displays the lowest behind only French speaking Quebec and Newfoundland (Elkins & Simeon, 1980, p.10) and that Alberta returns the highest proportion who report to identify more with their province than with Canada (Elkins & Simeon, 1980, p.17). This early literature on political culture in Canada serves a number of useful purposes for this work therefore. Firstly, it presents clear evidence that regional political cultures exist within Canada and that independent of other factors, differences in political orientations shape the strong regionalism of Canadian politics. Secondly, this literature demonstrates that these attitudes can be measured both to compare between provinces and to inform the reasoning for political behaviour, including participation, with regard to a specific set of attitudes concerning the individual’s attitude towards political institutions; Simeon and Elkins variables of efficacy and trust. Finally, this early literature suggests that there is something particular about the attitudes of citizens in Alberta, in both the remarkable absence of trust and weak attachment to national values that results in the province failing to conform to expected patterns. This peculiarity of political attitudes in Alberta and the low levels of trust, especially in comparison with neighbouring Saskatchewan, is particularly pertinent given that conceptualizations of trust have become a key element within the notion of social capital, developed by Robert Putnam, of how social connections and supporting norms serve to foster a participatory civic community that can overcome problems of collective action and opportunism   16 to develop strong and efficient institutions in which citizens have confidence (Putnam, 1993, p.167). In Making Democracy Work, Putnam conceives of three elements of social capital; trust, norms of reciprocity and networks of cooperation that “improve the efficiency of society by facilitating coordinated actions” (Putnam, 1993, p.167) and thereby foster a strong civic community with high levels of political participation. Interestingly however, Putnam singles out interpersonal trust as the ‘essential’ form of social capital, key to successful cooperation and serving to develop a ‘virtuous circle’ which further reinforces each element (Putnam, 1993, p.170). The empirically observed weakness in levels of trust in Alberta therefore would suggest a strong explanation for the distinct political culture in attitudes regarding voting. It is precisely this concern which motivated research undertaken by Pickup et al (2004), who used survey data from the Canadian Election Study and World Values Survey to directly apply this concept of civic culture to Alberta. Interestingly, the authors find that while Alberta appears to have a strong civic community, displaying high levels of interpersonal trust and civic participation, this does not translate into positive attitudes towards government or political participation as the model presented by Putnam suggests (Pickup et al, 2004, p.617). The authors argue that Putnam’s model of a self-supporting ‘virtuous cycle’ of civic engagement and participation breaks down in Alberta due to the province’s history as a “populist, agrarian community” which encouraged civic engagement with a suspicion for government which the authors feel was reinforced by Alberta’s ‘relative neglect’ (Pickup et al, 2004, p.618). Indeed Putnam himself argues that historical events and long term institutional performance can determine whether the conditions for cooperation can be reached and trust in government developed: “historical turning points thus can have extremely long-lived consequences… individuals responding rationally to the social context bequeathed to them by   17 history reinforce the social pathologies” (Putnam, 1993, p.179). Pointing to the work of ‘new institutionalists’, Putnam suggests that there is path dependence in the development of social capital as a result of relatively static institutions (Putnam, 1993, p.179). In this light, Pickup et al’s assertion that the absence of confidence in government in Alberta is a result of the historical failure of state institutions to respond to the demands of citizens, developing instead norms of independence and self-reliance, would seem to have merit. However, this explanation presented by both Putnam and Pickup et al illustrates a considerable challenge at the conceptual level for political culture explanations to electoral turnout; whether political culture is an independent or dependent variable, whether differences in culture between provinces affect the behaviour of voters, or if these cultural differences simply reflect other factors. The uncertainty of the causal relationship between culture and political behaviour is highlighted even in the early work by Simeon and Elkins, who acknowledge that their observed orientations may be dependent variables, providing a range of factors that shaped differences in culture; from differing historical experiences, differences in levels of development, immediate experience with government authorities or the structure of political institutions (Simeon & Elkins, 1974, p.433-435). Notwithstanding the conceptual concerns with political culture explanations that struggle to clarify the causal process between cultural variables and political behaviour, that such literature has successfully identified measurable differences in attitudes between provinces highlights that such an approach may hold considerable explanatory power. It is clear that the presence of trust in institutions is a key explanatory variable between the two cases studied here and if further research can confirm the suggested sources of these cultural differences posited by theorists, a conceivable explanation for differences in turnout could be provided. The challenge   18 for this research remains the methodological differences in operationalizing culture as comparative indicators beyond specific attitudinal data and resolving the conceptual problems posed by an unclear causal logic. This research will aim to identify if clear cultural differences, with identifiable origins are driving the divergence between Alberta and Saskatchewan or whether these are the expression of more deep-rooted forces. Political Economy The final theoretical approach that may offer an explanation for the puzzle of turnout between the two provinces and regionalized political differences in Canada is a focus on political economy. Janine Brodie argues for a revisionist tradition on the political economy of regionalism in Canada which rejects the centralist bias of the most prominent approaches and that instead understanding regions in Canada requires conceiving of the spatial relationship between areas, that regions are not constituted by their geographic boundaries but are the product of historical experience, human organization and social interaction (Brodie, 1989, p.141). In a sense Brodie is rejecting the formal institutional approach of writers such as Donald Smiley and Raymond Breton (see Smiley, 1976; Breton, 1981) who conceived of regions in Canada deriving largely from the formalized process of federalism in creating administrative boundaries. Instead, Brodie is drawing on the work of what could be termed the ‘hinterland’ theorists, a tradition which explores the process of uneven development in Canada and the creation of an unequal relationship of subordination and domination by the federal government over Western Canada through the mercantilist policies of early economic development (Brodie, 1989, p.147). Beginning to conceive of regions as defined by relationships with other regions, particularly economic relationships of domination rather than simply institutional separation offers considerable value for this work. This understanding developed initially under the work or   19 Harold Innis’ staples theory, whose pioneering study of the fur trade in Canada led him to conclude that “the economic history of Canada has been dominated by the discrepancy between the centre and the margin of western civilization” (Innis, 1962, p.385). Identifying not just furs but also cod, lumber, gold and later wheat, Innis argued that the economic development of Canada was marked by the search for valuable commodities in the periphery to supply demand in the metropolitan centres of capital and that much of Canada’s economic development was directed towards the trade and exploitation of these staples and transportation away from their source (Innis, 1962, p.384). For Innis the patterns of settlement and indeed the expansion of the country into new regions, along an east-west axis, can only be understood by reference to the constant source for further staples to meet the demands for export-led growth. Innis’ work has proven deeply influential and inspired a number of other authors to continue this account of how this structurally unequal relationship has shaped the economic and political growth of regions in Canada. W. L. Morton, writing on the rise of the Progressive Party in Canada directly points to how the practical disadvantages of grain growers both as a producer selling a commodity and a consumer seeking to purchase machinery produced a political movement which rejected Canadian economic policy as a “metropolitan economy designed, by the control of tariffs, railways and credit, to draw wealth from the hinterland ad the countryside into the commercial and industrial centres of central Canada” and the national parties as simply instruments of the financial interests of capital in central Canada (Morton, 1950, p.288) Nowhere is the effect this relationship of subordination had on the development of the wheat economy on the prairies better encapsulated then in V. C. Fowke’s work, The National Policy and the Wheat Economy, highlighting how persisted pursuit of industrialization through the tariff program of the ‘National Policy’, supported by both the Conservative and Liberal parties for much of   20 Canada’s early history placed wheat growers at a competitive disadvantage, demonstrating a “persistent disregard of the competitive inferiority of agriculture within the price system” (Fowke, 1957, p.290). Fowke goes to great lengths to highlight that this inequality created by the National Policy was not simply the indirect result of placing wheat growers at a competitive disadvantage on the market but that the very process of settlement of the west, in the construction of the railways and distribution of land, was pursued towards this end, indicative of the exploitative relationship long suffered by farmers in Western Canada. While reconceptualising regionalism in Canada as more than simply the institutional processes of federalism but instead representing an economic relationship of uneven development makes for a convincing account of understanding the strength of regionalism, this literature in of itself does not provide the theoretical basis for understanding how these historical processes continue to shape political behaviour today, in terms of electoral turnout and participation in politics. For this reason it is valuable to turn to the approach presented by ‘new institutionalism’ and in particular the school of historical institutionalism which seeks to re-conceive institutions as historically determined structures of behaviour, emerging through historical economic forces as relatively static norms and rules of behaviour. March and Olsen define such institutions as “relatively enduring collection of rules and organized practices, embedded in structures of meaning and resources that are relatively invariant in the face of turnover of individuals and changing external circumstances” (March & Olsen, 2011). Through this conception, institutions convey power structures, acting as a behavioural determinant by framing how individuals make decisions. Hall and Taylor write of an individual “as an entity deeply embedded in a world of institutions, composed of symbols, scripts and routines, which   21 provide filters for interpretation… out of which the course of action is constructed (Hall & Taylor, 1996, p.939). There is potential for this work therefore in repurposing the historical literature of the metropolitan-hinterland thesis, which stresses the fundamentally exploitative economic forces Alberta and Saskatchewan was exposed to, within the framework of new institutionalism to understand how these economic forces have shaped relatively durable norms of behaviour and framed actors understanding of the outside world and their inferior place within it. Through such an approach, explaining voter behaviour in Alberta and Saskatchewan requires understanding the role of institutions in Canadian provincial politics beyond the traditional focus on formal institutions to account for how economic and sociological patterns express power relations. This paper will therefore seek to repurposes the tradition of metropolitan-hinterland theorists to explicitly link this early economic development with persistent trends in political behaviour, particularly turnout at elections, which continues to this day.   22 Assessing Institutions In assessing the extent to which institutions, particularly electoral competitiveness and party systems, effect turnout in the two cases studied here, the first challenge concerns a methodological debate on how competitiveness in elections is conceptualized and therefore what indicators can be operationalized to measure competitiveness. The most commonly used indicator in the literature on electoral competitiveness is the margin of victory for the winning party, expressed as the difference in the overall percentage share of the vote between the two largest parties (Siaroff & Wesley, 2015, p.149; Hofstetter, 1973, p.355). Such a measurement is both comparative between different cases and has the advantage of being relatively easy to collect. However, conceptualising competitiveness as simply the margin of victory in the share of the vote is not without criticism; Gary Cox argues that in the context of turnout, a margin of victory measurement does not accurately reflect the probability that a single voter will affect the outcome (Cox, 1988, p.769). Conceptualising the process from the rational actor perspective, Cox reasons that such a variable acts through the calculations of voters on their capacity to affect an outcome. He argues that the margin of victory should be measured as the raw number of votes to reflect the probability of a single vote influencing the outcome (Cox, 1988, p.770). Other studies have taken this concern with how the rational calculation of individual voters before an election is operationalized further; Altman and Pérez-Liñán, studying the quality of democracies in Latin America, move away from votes entirely to measure the share of seats instead, developing a model of the weighted difference between the share of seats for the government and opposition parties (Altman & Pérez-Liñán, 2002, p.89).   23 Given that this study seeks to identify the effects of competitive elections on turnout in Alberta and Saskatchewan as part of a broader attempt to study the effect of institutional variables, in would perhaps remain useful to rely upon the margin of victory in vote share measurement. Not only is this data considerably easier to collect, it is also more suitable for comparison over a longer time period, allowing for long-term patterns of civic engagement and party mobilization stressed in other accounts to be visible (see Verba, Nie & Kim, 1978). Moreover, the considerable population differences between Alberta and Saskatchewan and within each electoral district make measuring the absolute vote not well suited to comparison. The tendency for the first-past-the-post system to distort election results makes measuring seats less indicative of the degree to which a vote could swing an election. In this light, given that measuring the margin of victory allows for more suitable comparison between provinces and across time periods while still reasonably illustrating the information held by voters calculating the utility to vote, this study will use this measurement as the measure of the competitiveness of elections.  Analysing province wide data on the difference in the share of the vote between parties in Alberta and Saskatchewan across the period covered by this study, 1975 to 2016, seems to confirm the hypothesis suggested in the institutional literature on electoral competitiveness acting as a determinant of voter turnout. Generally elections in Saskatchewan are more competitive than Alberta. In Saskatchewan that difference in the share of the vote between the first and second party in each election averages just 14.35% whereas over the same period in Alberta, this averages 23.95%. Yet to confirm a causal relationship between competitiveness and turnout requires identifying a clear pattern between these variables at specific elections, thereby supporting the notion that closer elections directly lead to increased turnout. To do so, the data   24 presented earlier on electoral turnout over this period is plotted historically to compare the relationship between the difference in the share of the vote, competitiveness, at each election, alongside the proportion of the registered electorate who did not vote. The assumption here would be that as the share of the vote separating the two largest parties rises, so too would the proportion of the electorate deciding not to vote. By tracing this trend across elections, we are also able to account for an issue raised by Johnston, Matthews and Bittner that the competitiveness of a specific election may feedback in subsequent turnout rates at later elections (Johnston, Matthews & Bittner, 2007).  Figures 2 and 3 plot these two variables for Alberta and Saskatchewan respectively. The Albertan case presents an interesting picture. Firstly it is clear that despite the high average difference in vote shares between the two largest parties, the competitiveness of elections in Alberta has varied considerably, with the difference between parties dropping substantially in the late 80s and early 90s to reach Alberta’s closest election in 1993, in which the Progressive Conservatives under Ralph Klein received only 4.76% more of the vote than the Liberal Party, yet the dominance of the PC’s subsequently rose again, albeit never reaching the level of the 1970s, before dropping in another close election in 2012, leading to the subsequent NDP victory in 2015 by a margin of 12.83%. In some respects, the trend in turnout mirrors this pattern as expected, with the numbers of non-voters dropping suddenly with the close election in 1993, before rising again as elections become less competitive and then falling again as more people participate in the close elections of the 2010s. Yet this is not always the case, the election with the lowest share of non-voters occurs at the beginning of the period covered, during the most uncompetitive elections; indeed in 1982 less people did not vote (34%), than the huge difference in votes between the PCs and NDP, at 43.53%. Unfortunately, the absence of reliable turnout   25 data for the period before 1975 means it is not possible to judge how far back this trend reaches. Yet taken over a long-term perspective, we can see that the proportion of electors not voting is increasing even as the elections seem to trend towards becoming more competitive. A similar exercise for Saskatchewan also challenges the hypothesis expected by institutional accounts of electoral competitiveness. For Saskatchewan the long term historical trend more closely matches the expected relationship; the number of electors not voting is clearly increasing as turnout declines, and there does appear to be an increasing trend of uncompetitive elections occurring. Yet it is difficult to be satisfied that electoral competitiveness is the causal explanation for this declining turnout; the incredibly close election of 1999, in which the Saskatchewan Party won only 0.88% more of than vote than the NDP, (losing the election in terms of seats won), exhibited a relatively high level of non-voting, at 34.5%, substantially more than the similarly close election of 1986, only three elections previously, in which non-voting only reached 17.9% of the electorate. Of course the risk here is reading too much into the varying levels of turnout without accounting for the differing electoral context of the specific Figure 2   26 elections, but it is suggestive that other variables may be influencing turnout if the two most competitive elections in Saskatchewan’s recent electoral history produce such starkly different levels of turnout.  The other prominent institutional factor determining voting behaviour pointed to in the literature is the party system. The challenge in accounting for the influence of the party system is measuring the number of parties in the system and the areas of cleavage between them. Yet here also the specific literature on Canada seems to diverge from the expectations of the wider literature on turnout, with Jackman’s suggestion that multi-party politics will lead to increases in turnout challenged by Siaroff and Wesley’s findings that the impact of the party system is most significant in a clear two party system. Perhaps the most significant challenge in categorizing the party system and determining the number of parties in the system of Alberta and Saskatchewan is the nature of party formation in Canada. In his classic theoretical conceptualization of party systems, Giovanni Sartori asserts that measuring party systems needs to account both for the numerical size of parties in vote share and the degree of fractionalism in the ideological distinctiveness of parties, what he terms the ‘position value’ of parties (Sartori, 2005, p.282). Yet Figure 3   27 as Johnston and Sharman find in their comprehensive coverage of party systems in Canada, the tradition of brokerage politics in Canadian parties, which seeks to promote unity across the divergent groups of the confederation by accepting high levels of autonomy at the local level and balancing ideological interests presents a distinct model of party systems (Johnston & Sharman, 2015, p.5). This phenomenon of political parties in Canada makes it more difficult to determine the ideological distance between parties and thereby measure the effective number of parties.  Perhaps the most useful categorization of party systems at the provincial level is undertaken through a more qualitative rather than quantitative approach with the classification of provincial party systems presented by Carty and Stewart. The authors categorize Alberta as the only province with a ‘one-party dominant system’, highlighting that between 1960 and 1995 Alberta meets the accepted definition of one-party dominance with a single party wining over 70% of seats (Carty & Stewart, 1996, p.73). Other features of the party system in Alberta they point to include extended periods of office-holding for single parties, few changes in government and one-sided legislatures, with these key features of the political system extending back to at least the election of the Social Credit party in 1935 (Carty & Stewart, 1996, p.78).  In comparison, Carty and Stewart argue that Saskatchewan belongs to the category of ‘polarized party systems’ where the NDP structures the polarization of the province’s party system and the other parties compete to become the opposition to the NDP (Carty & Stewart, 1996, p.83). Carty and Stewart point to the almost complete disappearance of the Liberal Party by 1978 to be replaced by the Progressive Conservatives, with the more recent rise of the Saskatchewan Party at the expense of the PCs seeming to confirm this prediction. However, while this categorization presents a convincing account of the long-term trends in the structure of   28 party competition, it does little to measure the number of effective parties within the system; there are elections where three parties receive sizeable shares of the vote.  The evidence presented here indicates that there does exist differing party systems within the two provinces and comparing the share of the vote in electoral data can identify this. Alberta experienced three elections over the period 1975-2015 in which only one party managed to receive greater than 20% of the vote, this occurring in 1975, 1979 and 1982. The dominance of a single party in these elections is illustrative of the particularity of the Albertan party system where other parties have frequently struggled to force a change of government, with the long dominance of the Progressive Conservatives between 1971 and 2015 the most recent manifestation of this trend. Yet as the data on competitiveness has identified, it is not certain the extent to which this has effected turnout. Is it really the nature of party competition and a highly polarized environment that encourages parties to mobilize supporters and thereby raise turnout or are these conditions themselves a result of a more participatory and engaged electorate? In this light, while it does seem that somewhat distinct party systems exist within the two provinces, at least in terms of the number of parties with a reasonable chance of winning elections, such a simple institutional explanation does not seem fully adequate. Given that these political systems were created simultaneously in 1905 and were both dominated in their early years by provincial offshoots of federal parties, it is not clear whether the differing party systems in each province is the key independent variable to explain such differences or merely an intermediary variable expressing more deep structural circumstances which led to such varied political developments. Indeed it is striking that while Carty and Stewart present probably the most convincing account of two divergent party systems in Alberta and Saskatchewan, their explanations for their origins return to the early patterns of political development, suggesting that   29 the true causes of the differing levels of turnout were determined before the eventual emergence of distinct formal institutional systems.    30 Assessing Political Culture As identified in the literature review, while the political culture approach is able to provide perhaps the most convincing explanations for the differences in turnout between the two cases studied and since the early development of the approach authors have become quite successful in deploying attitudinal data derived from opinion surveys to offer quantifiable indicators of these differences, understanding these cultural forces remains difficult due to the unclear causal processes developed throughout the literature. Such conceptual problems for understanding the causal process of political culture on electoral behaviour also present a methodological challenge for this research. While the evidence presented by the authors covered prove that there exists distinct political attitudes and cultural orientations which shape how voters in both Alberta and Saskatchewan behave, there remains a significant challenge in operationalizing these differences beyond observable attitudes in opinion surveys to understand why such cultural differences exist and what factors make up these ‘regional political cultures’ without generating broad, generalized theories that can not be adequately proved. The challenge of the political culture approach therefore becomes one of indentifying where cultural differences emerge from and proving that these differences are not the result of other forces. Given the differences in turnout between Alberta and Saskatchewan displays a long-term historical trend, stretching back at least to the 1970s, immediate cultural explanations such as the generational account presented by Blais et al of differing attitudes towards voting as a civic duty (Blais et al, 2004, p.221) are unsuitable for such an explanation. It would seem therefore that the successful fostering of strong social capital in Saskatchewan compared to the mistrust of government prevalent in Alberta that is clearly   31 identifiable in the analysis of political culture writers is to some degree historically determined. Indeed Simeon and Elkins turn to the role played by historical forces to account for the differences in political attitudes: “the ethos or community norms deriving from particular historical forces and events” (Simeon & Elkins, 1974, p.397). Their suggestion that historical political events shape political culture represented a major rethinking of the causal logic outlined in early political culture literature, such as Almond and Verba who saw the orientations they were measuring as emerging from a wider societal culture. For Simeon and Elkins, the causal pattern is not so neat and political events can directly shape and alter the political culture of a society (Simeon & Elkins, 1974, p.400). Of course the challenge posed by understanding cultural differences between provinces to be historically determined is that it greatly increases the number of potential variables to study to adequately explain the divergence in cultural attitudes. Simeon and Elkins write of the structure of party competition and conflict as a factor in the development of regional attitudes across Canada. Such a conception of how culture becomes regionalized is undoubtedly informed by the landmark work of Alan Cairns, who argued that the distortion in parliamentary representation created by the single member plurality system overtime creates and reinforces regionalized cleavages of political support along provincial lines (Cairns, 1968, p.62). David Smith in Prairie Liberalism suggests a similar explanation specifically for the case of Alberta and Saskatchewan’s political dissimilarity, arguing that the source of the cultural differences between the two provinces lay in the continued electoral success of the Liberal Party in Saskatchewan (Smith, 1975, p.22). However, reconceptualising cultural differences between the two provinces in this study as reflecting the long-term structural forces of formal institutions such as the electoral system   32 and success of parties does little to resolve the challenge of clarifying the causal process between culture and electoral turnout. Indeed, Smith has come to acknowledge the weaknesses in such an explicit institutional explanation for cultural differences, not least because of the subsequent decline in support of the Liberal Party in Saskatchewan. Instead, he has reformulated his explanation for the divergent political development of Saskatchewan and Alberta as expressing the wider structural forces of federalism and inter-governmental relations between provinces and the federal government (Smith, 2010, p.7) which occurred to “two provinces [that] have very different histories – politically, economically and demographically” (Smith, 2012, p.12) and generated differing responses. In this light, while formal institutions may play a role in driving cultural divergence, the sources of such difference become historical and even geographic, placing each province on distinct paths that lead to differing cultural responses to the external forces of federalism. While such an explanation offers considerable value to this study in accounting for the long-term trend in voting behaviour identified, such an approach is not without challenges. Drawing on historical accounts and generalizing to develop a holistic narrative of how early patterns of settlement and development formed cultural attitudes risks over-generalizing the complex process of cultural evolution. Moreover, by stressing a historical determinism in the development of cultural differences it fails to acknowledge the effect subsequent events and the institutionalized relations between actors can have on political attitudes Nowhere are these challenges more pressing than the responses generated to the Hart-Horrowitz fragmentation theory (see Forbes, 1987), perhaps one of the most prevalent of such historical-cultural explanations, which sought to argue that the relative strength of socialism in Canada in comparison to the United States, and the differing political cultures between English   33 and French Canada, could be accounted for by the period in which these areas were settled, during which a different ‘fragment’ in the stages of European political development were spun off to the colonial political community, developing since then along a highly deterministic course (Horowitz, 1966, p.144). Horowitz highlights what he refers to as a ‘touch’ of Toryism in English Canada, brought by American loyalists and therefore distinguishing English Canada from the outright liberalism of the US, creating the conditions within conservatism for the development of socialism (Horowitz, 1966, p.157). Horowitz’s thesis, while undoubtedly identifying an important explanation for the development of regional cultures, serves to highlight how deterministic such an account can become in its expectations for cultural development. Yet despite the problems with fragmentation theory and in particular the risk of generalizing early immigration as the sole explanatory variable for a complex web of subsequent political and cultural development, this account does present a useful explanation for the case studied here, stressing the need to examine historical patterns of settlement. For Nelson Wiseman, fragmentation theory and an emphasis on historical patterns of immigration are the basis for explaining the political differences between Alberta and Saskatchewan. Responding to the classic texts on the political economy of Alberta and Saskatchewan by C. B. Macpherson and Seymour Lipset, Wiseman draws on the influential Hartz-Horowitz thesis but seeks to account for subsequent cultural influences brought by later immigrants, arguing that Saskatchewan was influenced by the immigration of “turn-of-the-twentieth-century British urban immigrants who brought socialist, labourist and Fabian notions with them to rural Saskatchewan” (Wiseman, 2007, p.132) while Alberta was influenced at the same time by the immigration of rural American farmers from south of the border (Wiseman, 2007, p.133). Rather than the class emphases by Macpherson and Lipset, the cultural divergence of the two provinces is ideological,   34 driven by the settlement of people with differing attitudes. Together these immigrant influences are enough for Wiseman to account for an individualist attitude and an orientation towards market liberal policies in Alberta and a co-operative attitude and preference for state-centred policies in Saskatchewan, potentially ideologically linked to the strong levels of trust, reciprocity and civic engagement in Saskatchewan compared to distrust of government in Alberta. However, adequately testing the hypothesis suggested by Wiseman poses considerable methodologically challenges given the almost total absence of early immigration data broken down by province and indeed Wiseman offers no robust method for operationalizing his hypothesis. In light of these difficulties, perhaps the most suitable indicator for measuring the levels of immigration in each province is the reported place of birth in the early censuses conducted in the new Prairie Provinces, which go some way towards illustrating the differing peoples settling the two provinces. Of course the reported place of birth fails to distinguish whether immigrants have arrived directly from their homeland, carrying their fragmented ideology in their luggage, or have previously settled elsewhere, but at a more general level this data presents a strong generalized picture of the ethnic make up of the two provinces in their early history and is also directly comparative. As such, the research undertaken here collects the reported place of birth directly from archived copies with Statistics Canada for the censuses of 1906, 1911, 1916 and 1921, with the 1906 and 1916 censuses being specially conducted solely on the new joined prairie provinces and therefore not offering comparative data for Canada as a whole (Statistics Canada). Table 1 and 2 below present the reported place of birth by proportion in both Alberta and Saskatchewan, with the responses filtered out to highlight only the most popular responses.   35 Table 1 Selected birthplace of population by proportion, Alberta, Census 1906-1921 Place of Birth 1906 1911 1916 1921 Austria-Hungary 6.0% 5.6% 5.3% - British Isles (incl Ireland) 12.8% 17.6% 17.5% 16.6% Canada 46.8% 43.3% 48.6% 53.5% Germany 1.7% 1.6% 1.1% 0.8% Russia 3.1% 2.7% 3.0% 2.0% United States 23.3% 21.7% 18.5% 17.0%  Table 2 Selected birthplace of population by proportion, Saskatchewan, Census 1906-1921 Place of Birth 1906 1911 1916 1921 Austria-Hungary 8.5% 7.2% 6.6% - British Isles (incl Ireland) 13.8% 15.6% 14.3% 13.1% Canada 50.0% 50.5% 54.5% 60.4% Germany 2.3% 1.7% 1.1% 0.8% Russia 6.4% 4.7% 4.8% 3.7% United States 13.8% 14.1% 13.6% 11.6% The findings from this data reveal a number of striking observations. Firstly, examining the data on Alberta it is clear that in some respects Wiseman’s expectations are matched, there is a significant proportion of residents in Alberta reporting the United States as their place of birth, ranging from a high of 23.3% in 1906 before dropping each census to 17% by 1921. Compared to Saskatchewan, this is considerably higher, with the highest proportion in Saskatchewan only reaching 13.8% in 1906. However, while understanding the absence of trust and confidence in government in Alberta as representing a fragmentation of the liberal, individualistic ideology of early rural immigrants from America may therefore hold weight, the second half of Wiseman’s thesis is not as well supported by the data collected here. Rather than Saskatchewan representing a pattern of settlement made up of twentieth century British immigrants carrying socialist and   36 cooperative ideas, throughout the early period of immigration covered here Alberta frequently reports a higher proportion of residents born in the British Isles compared to Saskatchewan. It is only in the first year included, 1906, that a higher proportion can be found in Saskatchewan and even then the difference is only 1% between the two cases. Indeed, rather than suggesting that in its early years Saskatchewan became a haven for British immigrants, the province appears remarkably diverse. In 1906, 8.5% of the population was born in Austria-Hungary and 6.4% in Russia, alongside a range of smaller but significant amounts not reported here. Certainly it is also clear that Saskatchewan was settled in its early years by a greater proportion of residents reporting as born in Canada, but it also appears that Saskatchewan was a more ethnically heterogeneous province than Alberta, where American and British immigrants largely dominated. Not only does the more complex web of immigrant sources presented by this data illustrate the risks in a political culture approach, particularly one such as the fragmentation theory, of generalizing the ethnic composition of historical events to suit the wider theory, but the notion of Saskatchewan as a more heterogeneous province directly challenges the assertion made by Smith that its relatively similar ethnic make up encouraged responses of solidarity and cooperation amongst citizens to the early unequal relationship with the federal government. As such, the findings from this data poses two critical challenges to resolving the question posed by this research; what caused the divergence in turnout and potentially a culture of trust, between the two provinces. While it is clear that American immigrants more heavily settled Alberta in its early years, potentially explaining the importation of ideological values of liberalism and individualism, the conceptual issues of a political culture approach remain. Both provinces joined confederation at exactly the same time and undertook settlement through the   37 same process. What attracted American immigrants to Alberta and created the conditions for their individualistic and non-participatory attitudes to thrive but failed to do so in Saskatchewan? And secondly, in Saskatchewan, if it was not the influx of British immigrants carrying a socialist ideological fragment which drove the province’s subsequent history as a highly mobilized, participatory and relatively socialist political community nor the province’s homogenous cultural make up which facilitated the ease by which these institutional norms were developed, why has Saskatchewan developed into such a relatively civic minded political culture? It would seem that the explanations provided by the political culture approach that can be measured and tested within the limits of this research have failed to solve the dilemmas outlined here. The political culture theories may point to observable phenomena in cultural differences but has struggled to explain these phenomena solely within the confines of culture. To truly understand the forces shaping these observable differences, it is perhaps necessary to turn to the historical economic conditions of settlement and development in the two provinces.   38 Assessing Political Economy It is striking that the thesis presented by Wiseman and challenged by research here is presented by the author as a direct counter to two of the most prominent works on the influence of economic forces on political behaviour in Alberta and Saskatchewan: Macpherson’s Democracy in Alberta and Lipset’s Agrarian Socialism which belong to the hinterland thesis and understands the sources of regionalism as deriving from an exploitative relationship between a metropolitan centre in industrialized central Canada and a dependent hinterland.  Machperson seeks to explain what he describes as the non-party, or quasi-party tradition of Alberta, which as identified has seen voters in Alberta repeatedly unite under a single party with largely uncompetitive elections. Pointing to both the United Farmers of Alberta (UFA) and Social Credit, Macpherson conceives of the long dominance of these parties as indicative of a demand in Alberta for mass movements without the partisanship of the traditional federal parties that, with members acting as delegates, express the will of Albertans (Machperson, 1953, p.4-5). Macpherson argues that the fundamental causes of this non-party tradition are the ‘quasi-colonial status’ of Alberta and its economic development and the relative homogenous class composition of small-property land owning petit-bourgeois (Macpherson, 1953, p.6). This argument is typified in much of the literature on the political economy of Western Canada, that Alberta was dependent on a colonial relationship of subordination by the federal government as an area for profitable investment of eastern capital and left victimised by the tariffs of the National Policy, making the federal government to Albertans according to Macpherson “not only a federal but an imperial government” making it essential that the provincial government “should be in effect an offensive and defensive weapon against the   39 imperial power” (Macpherson, 1953, p.21). The evidence that Macpherson and others points to, federal control over natural resources, the direction of settlement and railway construction by the federal government and the imposition of the mercantilist National Policy which greatly increased the costs of importing agricultural machinery and weakened the selling power of wheat farmers are all indicative of this relationship. What is perhaps more interesting, particularly given that the unequal relationship with the metropolitan centre was similarly imposed on Saskatchewan, is Macpherson’s second argument regarding the homogenous class composition of the province. Analyzing the size of farms in the early years and number of labourers employed in the agricultural sector, Macpherson points to the relatively small average size of farms in the early years of the province (Macpherson, 1953, p.11) and that hired labour made up only a small fraction of the total labour undertaken by farm owners, with a majority of farms not employing any hired labour in the early years covered (Macpherson, 1953, p.14). Macpherson surmises that the majority of farmers, the dominant area of employment, were engaged in independent commodity production, being small propertied landowners and self-employed in the production of agricultural goods. The result was a relatively homogenous class composition of petit-bourgeois farmers, producing agricultural goods for export, without the significant presence of a large base of proletariat labourers seeking wage employment or bourgeois capitalists with large land ownership (Macpherson, 1953, pp.12-13). It is this unique emergence of a single class in the dominant sector, of land-owning petit-bourgeois, which for Macpherson helped shaped the populist, mass movement culture of politics in Alberta, as the “absence of any serious opposition of class interests” meant the usual function of parties to express and mitigate conflicts of interests was not required (Macpherson, 1953, p.21). Moreover Macpherson highlights the popularity of the highly liberal free market ideology   40 of the UFA and Social Credit which could be understood to have contributed towards the highly individualistic and non-participatory political culture that is reflected in political culture literature (Macpherson, 1953, p.232-234). Indeed interpreting this rejection of the orthodox party system in Alberta as not just a rejection of parties but a rejection of the democratic politics they represent potentially points to how the early emergence of agrarian populism manifested itself in low engagement in elections. Macpherson writes of how this petit-bourgeois liberalism resulted in a deterioration of Albertan democracy as independent commodity producers persisted in demands beyond the limits the provincial government could satisfy, ultimately leading to the sudden fall of these movements (Macpherson, 1963, p.237). Re-examining Macpherson’s work through the lens of historical institutionalism and understanding how these early economic forces of a highly competitive individualism developed institutions as relatively static norms of behaviour, it is not difficult to identify how this disillusionment and even direct opposition to party politics also generated weakened norms of trust and confidence in government and instead developed prevailing norm of political apathy and low engagement in elections which continues to influence political behaviour. The other prominent study of political culture on the prairies that Wiseman sought to disprove was Seymour Lipset’s account of the deeply cooperative and participatory nature of politics in Saskatchewan in Agrarian Socialism. Wiseman’s challenge is that Lipset examines similar economic conditions that so concerned Macpherson and finds them to be the causal factors behind Saskatchewan’s fundamentally different political development as a socialist and communitarian political body.   41 Like Macpherson’s interest in the emergence of distinct populist parties in Alberta, Lipset is concerned primarily with the rise to power in 1944 of the Cooperative Commonwealth Federation (CCF) in Saskatchewan, the “first electorally successful North American socialist movement” (Lipset, 1950, p.3). However, while Lipset also finds class as the explanatory variable for the emergence of this distinct political culture in the early agricultural development of the province, his work emphasises the importance of the dominance of wheat in agricultural production. Pointing to the extreme fluctuations in the price of wheat on global markets in the period between 1900 and 1947 as Saskatchewan was first settled and developed, with some years seeing drops in price by more than 50% on the previous year, alongside the wide variations in yield as a result of difficult growing conditions, Lipset paints a picture of the existence of early wheat farmers as marked by uncertainty and instability (Lipset, 1950, p.28). Lipset conceives of wheat farmers as living “the life of gamblers” whose investment and labour on a small plot of land “may arbitrarily leave them either poverty-stricken or comparatively well to do at the end of the harvest” (Lipset, 1950, p.27). It is this exposure to the price system of wheat farmers, particularly in the context of the federally imposed National Policy which Fowke identified as incredibly damaging to wheat growers by placing them in a position of competitive inequality (see Fowke, 1957, p.290), which Lipset indentifies as the source of a cooperative culture in Saskatchewan. Arguing that in seeing themselves as a marginal frontier group, not receiving their just share of national wealth, an agrarian class consciousness developed which perceived their conflict as being with external capital forces and restricted the emergence of the usual cleavages between classes. Instead, this position on the periphery and the shared experience of hardship growing a single crop in harsh   42 climatic conditions gave wheat farmers in Saskatchewan “the structural basis for collective action” (Lipset, 1950, p.35). This unifying purpose of a struggle for better wheat prices and lower freight rates for exportation resulted in a sense of solidarity and community not found in areas of the country with greater economic diversification. Lipset points to remarkably high memberships in civil organizations and in particular cooperative organizations that sought to support fellow wheat farmers, writing elegantly of his time spent studying communities in Saskatchewan where after the harvest has taken place “thousands of meetings are held throughout the province… in which farmers discuss economic and political problems” (Lipset, 1950, p.xv). It is here that we can most immediately identify the sources for the emergence of norms of trust and civic participation, the key elements of Saskatchewan’s strong social capital crucial to supporting political participation (Lipset, 1950, pp.26-29). Indeed Lipset makes it clear that these cooperative organizations, beginning as self-help measures in the economic sphere, soon translated into wider political movements calling for radical action to collectively manage the sale of grain and control prices. While Lipset traces the emergence of political organizations such as the Saskatchewan Cooperative Elevator Company and the Grain Growers’ Grain Company to understand the rise of a self-conscious socialist class in Saskatchewan, his findings also neatly identify the process in which voters in Saskatchewan came to believe that participation in political action was the most effective means of resolving their economic demands (Lipset, 1950, p.61). Yet while to an extent both of the arguments presented by Macpherson and Lipset appear convincing explanations for the growth of such different attitudes towards politics in the two provinces, the criticism levied by Wiseman and others remains. How did the same variables, the   43 petit-bourgeois values of independent commodity producers and their exploitation by the federal government, produce such different outcomes? As already identified, the solution offered by Wiseman that the ideology of differing immigrants shaped the cultural evolution of the provinces is insufficient to explain this divergence. However while a solely cultural explanation may not be enough, the observation that higher numbers of American born immigrants settled in Alberta does suggest a variable operating which encouraged American settlement in Alberta and the work of Macpherson and Lipset suggests a potential economic explanation for this difference. For the two authors are not entirely analyzing the same variables; while Macpherson is concerned with the general economic conditions of early agriculturalists in Alberta, Lipset is specifically concerned with the experience of wheat growers in Saskatchewan. As Smith and other authors have tended to suggest, Alberta experienced greater economic diversification and in agriculture this was particularly marked by the prominence of ranching, only suited to the geographic and climatic conditions of the foothills of southern Alberta. Utilizing data from the Census of Agriculture provided by Statistics Canada and conducted every 10 years from 1921 onwards, it is possible to begin developing a picture of the composition of the agricultural sector in each province in terms of the agricultural goods produced. Figure 4 provides the total area in acres of wheat farmed in each province between 1921 and 2006 and Figure 5 provides this in proportion to the total farmed area of each to better reflect the overall share of wheat farming in the two provinces. The results are clear; throughout the period wheat was grown in considerably larger numbers in Saskatchewan and this constituted a greater share of the total farmed land in the province than it ever did in Alberta. This difference amounts to almost 7 thousand additional acres of wheat being grown in Saskatchewan as early as 1921, well over double the amount in Alberta, and this difference continues to rise. While   44 Alberta never saw more than 20% of the total farmed land growing wheat, Saskatchewan averages 26% across the period covered. When only the proportion of farmland growing crops, rather than other uses, is considered the difference becomes even greater; Saskatchewan averages 59.8% of all crops as wheat between 1921 and 2006, Alberta only 37.6%. If we are to understand the relative strength of social capital in Saskatchewan as emerging from the cooperative response to the difficulties of wheat production, these clear differences in the predominance of wheat indicate why such effects occurred in Saskatchewan and not Alberta.        Figure 4   45  If the Census of Agriculture is then used to examine the strength of ranching in each province, the notion that differences in agricultural practices may explain the province’s divergence becomes increasingly clear. Figure 6 presents the total number of cattle and calves reported on farms in each province between 1921 and 2006 and again to better reflect the relative share in each province, Figure 7 presents the proportion of all farms reporting ownership of at least one head of cattle, which due to limitations in the data stretches from 1941 to 2006. This data suggests that while between 1921 and 1941 the total number of cattle on farms was relatively equal between Alberta and Saskatchewan, after 1951 the number in Alberta begins to rise at an increasingly higher rate than Saskatchewan, ultimately reaching a point by the new century where there were more than double the numbers of cattle in Alberta. The significance of ranching in Alberta is further reinforced when the proportion of all farms reporting cattle is compared, while a considerably high rate for both provinces, Alberta consistently reports a Figure 5   46 greater proportion, averaging a difference of 12.2%, indicative of the prominence of the cattle industry in the province.   Figure 6 Figure 7   47 The challenge posed by the data provided in the Census of Agriculture is that it is not directly comparative, wheat production is reported in acres planted while cattle is indicated by the total number, without giving an indication of how much land is given over to pasture. To resolve this, a final two data sets presented here attempt to illustrate the different nature of farms in Alberta and Saskatchewan. Figure 8 presents the average number of cattle per farm reporting cattle ownership, between 1921 and 2006. The data shown here clearly suggests that, other than the initial years covered, farms in Alberta tend to have larger herds than in Saskatchewan and are therefore likely to derive more of their income from ranching. To support this comparison is the proportion of total farmland, in acres, given over to growing all crops. While farmland not being used for the cultivation of crops could be serving other purposes, or lying fallow, this measurement does go someway in indicating how much land is used for crops in relation to potential use for pasture. Here the evidence of the differing nature of farming is even more indicative, Saskatchewan consistently reports a considerably greater share of farm area engaged in crop cultivation, with the shares in 1921 amounting to 29.1% in Alberta and 40.5% in Saskatchewan. While this difference does decline, Saskatchewan continues to see on average a greater share of land growing crops by 6.6%. Figure 8   48 Taken together, this data seems to suggest a plausible variable, rooted in the economic history of the two provinces, which could explain their divergence in political attitudes; that is the size of the wheat and ranching industries relative to the overall agricultural economy. The structural forces which emerged due to the un-diversified nature of the agricultural economy in the early years of settlement, embedding institutionalized norms of trust and civic engagement in Saskatchewan were not as predominant in Alberta where wheat production constituted a smaller share of farming activities. While hardships produced by these structural forces in Saskatchewan by exposing farmers to price fluctuations demanded a reaction of organized political action to break their dependency on wheat, farmers in Alberta had the geographic and climatic conditions in much of the southern province to engage in other activities, particularly ranching. With the explanatory variable now potentially identified, a variable which may help understand much of the observed differences between the provinces, from the political attitudes of residents through to the institutionalization of this culture in distinct party systems and electoral trends which engaged much of the institutional literature, the remaining task is offering a potential theoretical understanding for why ranchers, despite the continued economic   49 dominance of the prairies to the interests of external capital, did not develop similar attitudes to wheat farmers. In some respects it is here that Wiseman’s assertion, which has been confirmed by this research, that American immigrants who brought individualistic and liberal attitudes settled Alberta in greater numbers begins to hold greater weight. Writing on the early development and settlement of the Canadian prairies, Jeremy Adelman highlights that the methods in which land was distributed by the federal government, both in the free homesteading process under the Dominion Lands Act but particularly through the government granting substantial shares of land in each township plot to the Canadian Pacific railway company and Hudson Bay Company, who quickly sold their land undervalued, proved highly attractive to American farms who could quickly raise capital through existing land ownership and engaged in considerable land speculation (Adelman, 1999, p.37). Adelman identifies that much of the southern parts of the provinces were initially focused on ranching when Palliser’s triangle was at first dismissed as unsuitable for growing crops and it is easy therefore to see the geographic attraction of southern Alberta, prime land for keeping cattle, to American settlers experienced in ranching and seeking to exploit the relative availability of cheap land at a time that prices were rising in the US. Moreover, Adelman highlights that the extent of land speculation in early settlement forced many farmers into considerable debt in the expectation of further rises in land value and that when this did not occur left may farmers further exposed to the demands of external capital (Adelman, 1999, p.40). For Adelman, this conditions created by the settlement of land directly challenges the image presented by Macpherson and Lipset of a relatively homogenous petit-bourgeois; land ownership was never this equal nor homogenous (Adelman, 1999, p.49).   50 Indeed it is worth highlighting the practical differences in wheat cultivation and cattle ranching that drove this economic divergence. While Lipset’s historical study of the volatile price of wheat in the early twentieth century and the difficult growing conditions of the Canadian prairies successfully demonstrates that growing wheat involved considerable risk and uncertainty for farmers, which coupled with the considerable debt of many early settlers that Adelman points to and the small size of land owned made further investments in wheat difficult, this was not the case with ranching. A review of historical commodity prices illustrates that while wheat was incredibly volatile over the first half of the twentieth century, the price of beef remained largely stable and at a higher value than wheat (Jacks, 2013). This relative security in producing beef, coupled with the considerably lower capital costs of ranching which was not dependent on the importation of farming machinery rendered unduly costly by the effects of the National Policy made investing in beef and expanding ranching operations comparatively more secure. While a wheat farmer would, as Lipset asserts, be taking a considerable gamble in seeking to expand and buy more land, a rancher would likely be more certain that buying further plots of land and expanding the size of their herd would see a secure return in capital. Given the relative availability of land for those who had the capital and the almost unlimited potential size of herds ranchers could amass without increasing too greatly productivity costs, it would be expected that these favourable conditions would see the exponential growth in cattle herds and the considerable accumulation of land and capital by ranchers and this seems particularly borne out by the data already studied, with the size of cattle herds on farms in Alberta growing considerably. Adelman identifies that homesteaders, limited to their initial 160 acres and the difficult work of ‘proving up’ their plot to earn ownership largely engaged in wheat farming. With such   51 an equal allocation of land and the early hardships of homesteading, it is perhaps possible to identify Lipset’s wheat community, developing norms of equality and cooperation through their shard experiences. On the other hand, Adelman highlights that those engaged in land speculation and buying up tracts of land sold below value by the rail companies were often ranchers, where the reduced risk compared to wheat growing made the expansion of ranchers by accumulating more land a tempting prospect (Adelman, 1999, p.49). It is perhaps here then that we can identify how the conditions of land distribution helped develop a petit-bourgeois land owning class, directly engaged in a competitive relationship for land ownership with their neighbours. Given that Macpherson’s thesis is rooted in the nature of a property owning class of petit bourgeois in Alberta, the differing role played by land ownership in the ranching and wheat industries is therefore worth considering. The result of the Dominion Lands Act in Alberta and Saskatchewan was the systemised distribution and settlement of land that was simultaneously one of the most equal allocations of private property in human history, with each homesteader entitled to an identical 160 acres quarter-section unit and yet through the allocation by the federal government of large sections of each township to the Canadian Pacific Railway, Hudson’s Bay Company and school boards, created the conditions for the large accumulation of land amongst those few settlers with the capital to finance such acquisitions and therefore the generation of two distinct classes, both petit-bourgeois land owners but differing based on their capital ownership and relative security.  It is this accumulation of land and capital, made possible by the climatic conditions of southern Alberta and the land distribution policies of the federal government, which potentially made possible a more secure ranching community directly engaged in a competitive relationship   52 with other ranchers for the best grazing land and with a highly liberal and independent petit-bourgeois mindset not found amongst wheat farmers. A second potential explanation makes a more direct link between the early economic history of ranching and the later emergence of the now dominant oil industry. Drawing on the rentier theory, developed to explain low participation in oil rich nations, Nazih Richani makes the claim that in Colombia ranching amounts to a similar form of rentier capitalism (Richani, 2012, p.52). The author makes the argument that the relatively stable income that can be derived from the exportation of beef makes investing in land a speculative exercise based on the assumption of long term rents and that owning land becomes a means of accumulating capital (Richani, 2012, p.64). Given the context already discussed in Alberta, it could be seen how such a theory of rentierism in ranching could be translated to the prairies, whether the relative stable price of beef compared to wheat and the low productivity costs in ranching allowed a rentier economy to develop. The key notion behind rentiersm, as famously presented by Hazem Beblawi and others suggests that such an economic model and the accumulation of wealth under it creates a mentality of low engagement in politics because the distribution of this wealth stifle the sources of conflicts normally expressed through political participation (Beblawi, 1987). It is possible therefore that the ease by which land could be accumulated and rent derived from it through ranching could generate institutional norms of non-participation which would seamlessly translate with the growth of the oil industry in Alberta and the continued source of land rents from this new industry, a model of land ownership which remained rooted in the accumulation of capital for the individual. Indeed writing on the decline of the ranching frontier after the 1920s, David Breen highlights that as cattle ranchers often held the mineral rights to   53 their land, unlike homesteaders, it was former ranchers who became most involved in the development of the petroleum industry (Breen, 1983, p.238).   54 Conclusion In conclusion, it has been demonstrated that approaching the puzzling empirical variations in provincial electoral turnout between Alberta and Saskatchewan firstly needs to move beyond behaviouralism, which seeks to focus explanations of voting behaviour at the individual level, due to the difficultly in explaining the broad variations in attitudes and behaviour which the distinct regional political cultures in Canada display. However, upon testing the expected outcomes of many of the most prominent explanations found within formal institutional accounts, in particular regarding the competitiveness of elections and the party system, voting data has failed to clearly conform to the expectations of such accounts. While elections in Saskatchewan were generally more competitive than Alberta during this period Saskatchewan displays a more competitive two-party system, these trends do not appear to have had a direct effect on turnout over this period, suggesting explanations at the aggregate level are insufficient when they remain grounded in formal institutionalism as it remains difficult to determine the causal mechanisms of whether such institutional variations are the cause of differing voting behaviour or the resulting institutionalization of existing patters caused by deeper structural determinants.  Indeed, despite the methodological challenges of identifying robust empirical indicators of the kinds of cultural differences stressed in much of the literature on regional political cultures in Canada, the literature covered in this paper clearly points to a range of divergent cultural attitudes which exist between Canadian provinces and that these attitudes directly influence political behaviour. In particular, this literature makes it strikingly clear that distinctions in such attitudes exist between Alberta and Saskatchewan. The principle challenge of this study therefore   55 has been an attempt to distinguish the causal mechanisms of such divergence in political attitudes. Doing so has placed this research between the competing theoretical schools of old and new institutionalism; seeking to ask whether these differing political cultures exist as a result of formal institutional systems and processes which determine the behaviour of individual actors or whether these identifiable institutions represent the ordering and codification of more deeply set social norms and rules of behaviour whose origins can be traced to historical economic and social patterns.  As such, this paper has argued that explanations need to move away from a focus on the particular context of individual elections and instead look at the history of the provinces, particular the economic and social history, to understand how the so-called ‘twin provinces’, calved out at the same moment in Canadian history of the same open prairie, came to diverge so dramatically. Crucial here to explaining the causes of voter participation has been Putnam’s social capital thesis, identifying the mechanisms by which values of trust and reciprocity, supported through civic engagement, lead to higher rates of voting participation. Yet while the political culture literature has been found to be valuable in identifying the difference in attitudes between the two provinces, the conceptual weaknesses in understanding where these cultural variations emerge from has posed a challenge to such an explanation. After studying historical data sourced from the first censuses conducted in each province to examine levels of immigration, it is not immediately clear if the cultural explanations presented by the literature are adequate in of themselves to explain the emergence of attitudinal differences.  Instead, turning to the historical institutional approach which seeks to highlight how such norms of behaviour become institutionalized through historical structural economic forces, it has been argued that the origins of such a divergence can only be explaining by examining the early   56 economic patterns of development. In this light, the work of Macpherson and Lipset, informed by the tradition of understanding Western Canada’s economic relationship as one of dependency on and exploitation by capital in the east, proved particularly valuable. By examining the economic data of the Census of Agriculture to develop a detailed picture of the agricultural sector in each province from 1921 to 2006, the considerable difference between the two provinces in wheat cultivation and cattle ranching was illustrated. This evidence suggested that while the dominance of wheat in the early agricultural development of Saskatchewan helped to institutionalize norms of cooperation and political participation identified by Lipset, forming high levels of social capital, in norms of trust and reciprocity as a determining factor in electoral turnout, this process never occurred in Alberta where wheat never dominated agricultural production to the same degree. Instead, the province’s early history of ranching continued to see a substantial share of the agricultural sector engaged in raising cattle. Without the same levels of risk and uncertainty as wheat it was suggested that this diversification helped develop the norms of independence and possessive individualism which resulted in weak social capital, particularly in trust and confidence in government and can therefore explain low levels of turnout. While a number of hypothesis were presented to explain how ranching managed to escape the hardship and exploitation of the wheat industry and instead form a distinct political culture built on a class of land-owning petit bourgeois, engaged in a competitive rather than cooperative relationship with their fellow farmers, it remains for further research to test these explanations and further explore the relationship between ranching and political apathy. What remains clear from this research however, as highlighted by the historical institutional approach, is the deep structural effect early economic forces can have on the long term political development of a province, forming relatively static institutional norms which continue to effect   57 political behaviour long after these economic forces have changed. The unique nature of the two cases explored here, provinces that were created simultaneously, demonstrate the extent to which early economic forces can determine the future economic and political direction of development and shape norms which guide behaviour and formal institutional processes, such as elections, long after these early economic structures have lost their significance. Ultimately, by offering an explanation of this phenomenon which understands institutions to include the norms and rules of behaviour that exist as historically established and enduring value systems this research has sought to demonstrate that accounting for the sources of regionalized political behaviour in Canada requires understanding the historical processes of class and the province’s place in the generation of wealth and capital. It is only once norms of turnout and participation are placed in the context of these economic relationships that research can begin truly offering the possibility of reversing concerning declines in turnout.    58 Bibliography Aldrich, John H., (1993), ‘Rational Choice and Turnout’, American Journal of Political Science, Vol. 37, No. 1, pp. 246-278 Almond, Gabriel and Verba, Sidney, (1963), The Civic Culture: Political Attitudes and Democracy in Five Nations, (Princeton, NJ: Princeton University Press) Altman, D. and Pérez-Liñán, (2002), ‘Assessing the Quality of Democracy: Freedom, Competitiveness and Participation in Eighteen Latin American Countries’, Democratization, Vol. 9, No. 2, pp. 85-100 Beblawi, Hazem, (1987), ‘The Reniter State in the Arab World’, Arab Studies Quarterly, Vol. 9, No. 4, pp. 383-398 Bellamy, David J., Pammett, Jon H. and Rowat, Donald C. (eds.), (1976), The Provincial Political Systems: Comparative Essays, (Toronto: Methuen) Blais, André and Dobrzynska, Aginesszka, (1998), ‘Turnout in Electoral Democracies’, European Journal of Political Research, Vol. 33, pp. 239-261 Blais, André, (2000), To Vote or Not to Vote: The Merits and Limits of Rational Choice Theory, (Pittsburgh, PA: University of Pittsburgh Press) Blais, André, Gidengil, Elisabeth, Nevitte, Neil and Nadeau, Richard, (2004), ‘Where Does Turnout Decline Come From?’, European Journal of Political Research, Vol. 43, pp. 221-236 Blais, André, (2006), ‘What Affects Voter Turnout?’, Annual Review of Political Science, Vol. 9, pp. 111-125  Blais, André and Lago, Ignacio, (2009), ‘A General Measure of District Competitiveness’, Electoral Studies, Vo. 28, pp. 94-100 Breen, David H., (1983), The Canadian Prairie West and the Ranching Frontier: 1874-1924, (Toronto: University of Toronto Press) Breton, Raymond, (1981), ‘Regionalism in Canada’, in David M. Cameron (ed.) Regionalism and Supranationalism: Challenges and Alternatives to the Nation State in Canada and Europe, (Montreal, QC: Institute for Research on Public Policy), pp. 57-80 Brodie, Janine, (1989), ‘The Political Economy of Regionalism’, in Clement Wallace and Glen Williams, (eds.), The New Canadian Political Economy, (Kingston, ON: McGill-Queen’s University Press), pp.138-159   59 Cairns, Alan C., (1968), ‘The Electoral System and the Party System in Canada, 1921-1965’, Canadian Journal of Political Science, Vol. 1, No. 1, pp.55-80 Cairns, Alan C., (1977), ‘The Governments and Societies of Canadian Federalism’, Canadian Journal of Political Science, Vol. 10, No. 4, pp. 695-725 Caldeira, Gregory A, Patterson, Samuel C. and Markko, Gregory A., ‘The Mobilization of Voters in Congressional Elections’, The Journal of Politics, Vol. 47, No. 2, pp.490-509 Campbell, Angus, Converse, Philip E., Miller, Warren E. and Stokes, Donald, (1960), The American Voter, (New York: John Wiley & Sons) Carty, R. Kenneth and Stewart, Nelson, (1996), ‘Parties and Party Systems’, in Christopher Dunn (ed.) Provinces: Canadian Provincial Politics, (Peterborough, ON: Broadview Press), pp. 63-94 Carty, R. Kenneth, (2004), ‘Parties as Franchise Systems: The Stratarchical Organizational Imperative’, Party Politics, Vol. 10, No. 1, pp. 5-24 Ceaser, James W. and Saldin, Robert P., (2005), ‘A New Measure of Party Strength’, Political Research Quarterly, Vol. 58, No. 2, pp. 245-256 Chaudhry, Kiren Aziz, (1994), ‘Economic Liberalization and the Lineages of the Rentier State’, Comparative Politics, Vol. 27, No. 1, pp. 1-25 Courtney, John C. and Smith, David E., (1972), ‘Saskatchewan: Parties in a Politically Competitive Province’, in Martin Robin (ed.) Canadian Provincial Politics: The Party Systems of the Ten Provinces, (Scarborough, ON: Prentince-Hall), pp. 290-318 Cox, Gary W., (1988), ‘Closeness and Turnout: A Methodological Note’, The Journal of Politics, Vol. 50, No. 3, pp. 768-775 Downs, Anthony, (1957), ‘An Economic Theory of Political Action’, Journal of Political Economy, Vol. 65, No. 2, pp. 135-150 Dunn, Christopher, (1996), Provinces: Canadian Provincial Politics, (Peterborough, ON: Broadview Press) Elections Alberta, Statement of Votes, [online], available at: http://www.elections.ab.ca/news-reports/reports/statistics/, accessed 23/04/2017 Elections BC, Statement of Votes, [online], available at: http://elections.bc.ca/resources/statistics/, accessed 23/04/2017 Elections Canada, Voter Turnout at Federal Elections and Referendums, [online], available at: http://www.elections.ca/content.aspx?dir=turn&document=index&lang=e&section=ele, accessed 23/04/2017   60 Elections Canada, Turnout and Reasons for Not Voting During the 42nd General Election: Results from the Labour Force Survey, [online], available at: http://elections.ca/content.aspx?section=res&dir=rec/eval/pes2015/lfs&document=index&lang=e, accessed 25/05/2017 Elections Manitoba, Statement of Votes, [online], available at: http://www.electionsmanitoba.ca/en/Results, accessed 23/04/2017 Elections Saskatchewan, Statement of Votes, [online], available at: http://www.elections.sk.ca/election-results/, accessed 23/04/2017 Elkins, David J. and Simeon, Richard (eds.), (1980), Small Worlds: Provinces and Parties in Canadian Political Life, (Agincourt, ON: Methuen Publications) Flanagan, Tom, (2001), ‘From Riel to Reform (And a Little Beyond): Politics in Western Canada’, American Review of Canadian Studies, Vol. 31, No. 4, pp. 623-638 Forbes, H. D., (1987), ‘Hartz-Horowitz at Twenty: Nationalism, Toryism and Socialism in Canada and the United States’, Canadian Journal of Political Science, Vol. 20, No. 2, pp. 287-315 Fowke, Vernon C., (1957), The National Policy and the Wheat Economy, (Toronto: University of Toronto Press) Fowke, Vernon C. and Fowke, Donald (1968), ‘Political Economy and the Canadian Wheat Grower’, in Norman Ward and Duff Spafford, (eds.), Politics in Saskatchewan, (Don Mills, ON: Longman’s Canada) Gerber, Alan S., Green, Donald P. and Larimer, Christopher W., (2008), ‘Social Pressure and Voter Turnout: Evidence from a Large-Scale Field Experiment’, American Political Science Review, Vol. 102, No. 1, pp. 33-48 Geys, Benny, (2006), ‘Explaining voter turnout: A review of aggregate-level research’, Electoral Studies, Vol. 25, No. 4, pp. 637-840 Gibbins, Roger, (1980), Prairie Politics and Society: Regionalism in Decline, (Toronto: Butterworths) Hall, Peter and Taylor, Rosemary, (1996), ‘Political Science and the Three New Institutionalisms’, Political Studies, pp. 936-957 Hepburn, Eve, (2010), ‘Small Worlds in Canada and Europe: A Comparison of Regional Party Systems in Quebec, Bavaria and Scotland’, Regional and Federal Studies, Vol. 20, No. 4, pp. 527-544   61 Hepburn, Eve, (2011), ‘Citizens of the Region’: Party Conceptions of Regional Citizenship and Immigrant Integration’, European Journal of Political Research, Vol. 50, No. 4, pp. 504-529 Hofstetter, C. Richard, (1973), ‘Inter-Party Competition and Electoral Turnout: The Case of Indiana’, American Journal of Political Science, Vol. 17, No. 2, pp. 351-366 Horowitz, G., (1966), ‘Conservatism, Liberalism and Socialism in Canada: An Interpretation’. The Canadian Journal of Economics and Political Science, Vol. 32, No. 2, pp. 143-171 Innis, Harold A., (1962), The Fur Trade in Canada: An Introduction to Canadian Economic History, Rev. Ed., by S. D. Clark and W. T. Easterbrook, (Toronto: University of Toronto Press) Irving, John A., (1959), The Social Credit Movement in Alberta, (Toronto: University of Toronto Press) Jackman, Robert W., (1987), ‘Political Institutions and Voter Turnout in Industrial Democracies’, American Political Science Review, Vol. 81, No. 2, pp. 405-424 Jacks, David S, (2013), ‘From Boom to Bust: A typology of Real Commodity Prices in the Long Run’, NBER Working Paper No. 18874, data available online at: http://www.sfu.ca/~djacks/data/boombust/index.html, accessed 05/12/2017 Jeffery, Charlie and Hough, Dan, (2003), ‘Regional Elections in Multi-Level Systems’, European Urban and Regional Studies, Vol. 10, No. 3, pp. 199-212 Jenson, Jane, (1976), ‘Party Systems’, in David J. Bellamy, Jon H. Pammett and Donald C. Rowat (eds) The Provincial Political Systems: Comparative Essays, (Toronto: Methuen) Jewell, Malcolm E. and Olson, David M., (1982), American State Political Parties and Elections, (Homewood, Ill: Dorsey Press) Johnston, Richard, (1980), ‘Federal and Provincial Voting: Contemporary Patterns and Historical Evolution’, in David J. Elkins and Richard Simeon (eds.), Small Worlds: Provinces and Parties in Canadian Political Life, (Agincourt, ON: Methuen Publications) Johnston, Richard, Blais, André, Brady, Henry E. and Crête, Jean, (1992), Letting the People Decide: Dynamics of a Canadian Election, (Montreal, QC: McGill-Queen’s University Press) Johnston, Richard, Matthews, J. Scott and Bittner, Amanda, (2007), ‘Turnout and the Party System in Canada, 1988-2004’, Electoral Studies, Vol. 26, pp. 735-745 Johnston, Richard and Sharman, Campbell (eds.), (2015), Parties and Party Systems: Structure and Content. (Vancouver, BC: UBC Press) Laxer, Gordon, (ed.), (1991), Perspectives on Canadian Economic Development: Class, Staples, Gender and Elites, (Toronto: Oxford University Press)   62 Lazarsfeld, Paul Felix, Berelson, Bernard and Gaudet, Hazel, (1944), The People’s Choice: How the Voter Makes Up His Mind in a Presidential Campaign, (New York: Columbia University Press) Lijphart, Arend, (1997), ‘Unequal Participation: Democracy’s Unresolved Dilemma’, American Political Science Review, Vol. 97, No. 1, 1997, pp. 1-14 Lipset, Seymour Martin, (1950), Agrarian Socialism: The Cooperative Commonwealth Federation in Saskatchewan: A Study in Political Sociology, (Berkeley, CA: University of California Press) Long, J. A. and Quo, F. Q. (1972) ‘Alberta: One Party Dominance’ in Martin Robin (ed.) Canadian Provincial Politics: The Party Systems of the Ten Provinces, (Scarborough, ON: Prentince-Hall), pp. 1-26 Macpherson, C. B., (1953), Democracy in Alberta: Social Credit and the Party System, (Toronto, ON: University of Toronto Press) Mallory, J. R., (1954), Social Credit and the Federal Power in Canada, (Toronto: University of Toronto Press) March, James and Olsen, Johan, (2008), Elaborating the 'New Institutionalism'. In R. Rhodes, S. Binder, & B. Rockman, The Oxford Handbook of Political Science, (pp. 3-20), (Oxford: Oxford University Press) Morton, W. L., (1950), The Progressive Party in Canada, (Toronto: University of Toronto Press) Murtazashvili, Ilia, (2013), The Political Economy of the American Frontier, (Cambridge: Cambridge University Press) Patterson, Samuel C. and Caldeira, Gregory A., (1983), ‘Getting Out the Vote: Participation in Gubernatorial Elections’, The American Political Science Review, Vol. 77, No. 3, pp. 675-689 Pickup, Mark, Sayers, Anthony, Knopff, Rainer and Archer, Keith, (2004), ‘Social Capital and Civic Community in Alberta’. Canadian Journal of Political Science, Vol. 37, No. 3, pp. 617-645 Powell, G. Bingham. (1986), ‘American Voter Turnout in Comparative Perspective’. American Political Science Review, Vol. 80, No. 1, pp. 17-43 Pratt, Larry (1976), The Tar Sands: Syncrude and the Politics of Oil, (Edmonton: Hurtig) Putnam, Robert, (1993), Making Democracy Work: Civic Traditions in Modern Italy, (Princeton: Princeton University Press)   63 Putnam, Robert, (1995), ‘Bowling Along: America’s Declining Social Capital’, Journal of Democracy, Vol. 6, No. 1 Richani, Nazih, (2012), ‘The Agrarian Rentier Political Economy: Land Concentration and Food Insecurity in Colombia’, Latin American Research Review, Vol. 47, No. 2, pp. 51-78 Richards, John, and Pratt, Larry (1979), Prairie Capitalism: Power and Influence in the New West, (Toronto: McClelland and Stewart) Riker, William H., and Ordeshook, Peter C., (1968), ‘A Theory of the Calculus of Voting’, The American Political Science Review, Vol. 62, No. 1, pp. 25-42 Robin, Martin, (ed.), (1972), Canadian Provincial Politics: The Party Systems of the Ten Provinces, (Scarborough, ON: Prentice-Hall) Sartori, Giovanni, (2005), Parties and Party Systems: A Framework for Analysis, (Colchester: ECPR Press) Smiley, Donald V., (1975), ‘Canada and the Quest for a National Policy’, Canadian Journal of Political Science, Vol. 8, No. 1, pp.40-62 Smiley, Donald V., (1976), Canada in Question: Federalism in the Seventies, (Toronto: McGraw-Hill Ryerson) Smith, David E., (1975), Prairie Liberalism: The Liberal Party in Saskatchewan 1905-71, (Toronto: University of Toronto Press) Smith, David E., (1976), ‘The Prairie Provinces’, in David J. Bellamy, Jon H. Pammett and Donald C. Rowat (eds) The Provincial Political Systems: Comparative Essays, (Toronto: Methuen) Smith, David E., (2010), ‘Parallel Provinces: Saskatchewan and Alberta’, Journal of Canadian Studies, Vol. 44, No. 3 Statistics Canada, ‘Census of Agriculture’, [online], available at: http://www.statcan.gc.ca/pub/95-632-x/95-632-x2007000-eng.htm, accessed 23/10/2017 Statistics Canada, (1907), Census of Population and Agriculture of the Northwest Provinces: Manitoba, Saskatchewan, Alberta: 1906, [archived online], available at: https://archive.org/details/1906981906F21907engfra, accessed 27/11/2017 Statistics Canada, (1912), Fifth Census of Canada: 1911, Vol. 2: ‘Religions, Origins, Birthplace, Citizenship, Literacy and Infirmities By Provinces, Districts and Sub-Districts’, [archived online], available at: https://archive.org/details/fifthcensusofcan02cana, accessed 27/11/2017   64 Statistics Canada, (1918), Census of the Prairie Provinces: Population and Agriculture: Manitoba, Saskatchewan, Alberta: 1916, [archived online], available at: https://archive.org/details/1916981916F1918eng, accessed 27/11/2017 Statistics Canada, (1925), Sixth Census of Canada, 1921, Vol. 2: ‘Age, Conjugal Condition, Birthplace, Immigration, Citizenship, Language, Educational Status, School Attendance, Blindness and Deaf Mutism’, [archived online], available at: https://archive.org/details/1921981921FV21925engfra, accessed 27/11/2017 Studlar, Donley T., (2001), ‘Canadian Exceptionalism: Explaining Differences over Time in Provincial and Federal Voter Turnout’, Canadian Journal of Political Science, Vol. 34, No. 2, pp. 299-319 Siaroff, Alan and Wesley, Jared J., (2015), ‘Comparative Voter Turnout in the Canadian Provinces since 1965: The Importance of Context’, Canadian Political Science Review, Vol. 9, No. 1, pp. 147-163 Simeon, Richard and Elkins, David J., (1974), ‘Regional Political Cultures in Canada’, Canadian Journal of Political Science, Vol. 7, No. 3, pp. 397-437 Verba, Sidney, Nie, Norman H. and Kim, Jae-on, (1978), Participation and Political Equality: A Seven-Nation Comparison, (Chicago: University of Chicago Press) Wallace, Clement and Williams, Glen, (eds.), (1989), The New Canadian Political Economy, (Kingston, ON: McGill-Queen’s University Press) Ward, Norman and Spafford, Duff, (eds.), (1968), Politics in Saskatchewan, (Don Mills, ON: Longman’s Canada) Wiseman, Nelson, (1996), ‘Provincial Political Cultures’, in Christopher Dunn (ed.) Provinces: Canadian Provincial Politics, (Peterborough, ON: Broadview Press), pp. 21-62 Wiseman, Nelson, (2007), In Search of Canadian Political Culture, (Vancouver, BC: UBC Press)  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items