UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Modelling inter-provincial migration between British Columbia and other Canadian provinces during 1966-1970 Dabestani-Sharifabad, Alayar 1975

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

Item Metadata

Download

Media
831-UBC_1975_A4_6 D32_4.pdf [ 5.73MB ]
Metadata
JSON: 831-1.0099981.json
JSON-LD: 831-1.0099981-ld.json
RDF/XML (Pretty): 831-1.0099981-rdf.xml
RDF/JSON: 831-1.0099981-rdf.json
Turtle: 831-1.0099981-turtle.txt
N-Triples: 831-1.0099981-rdf-ntriples.txt
Original Record: 831-1.0099981-source.json
Full Text
831-1.0099981-fulltext.txt
Citation
831-1.0099981.ris

Full Text

MODELLING INTER-PROVINCIAL MIGRATION BETWEEN BRITISH COLUMBIA AND OTHER CANADIAN PROVINCES DURING 1966-1970  by ALAYAR DABESTANI-SHARIFABAD B.A., University of Tehran, Iran, 1969 M.A., University of Tehran, Iran, 1972  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN BUSINESS ADMINISTRATION in the Faculty of Commerce and Business Administration  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA JANUARY, 1975  I n p r e s e n t i n g this thesis  i n p a r t i a l fulfillment, of the requirements  fbr an advanced degree a t t h e U n i v e r s i t y o f B r i t i s h Columbia,  I agree  t h a t t h e L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s t u d y . I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e copying o f t h i s  thesis  for s c h o l a r l y p u r p o s e s may be g r a n t e d by t h e Head o f my Department o r h i s r e p r e s e n t a t i v e . I t i s understood  that copying o r p u b l i c a t i o n o f  t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l n o t be a l l o w e d w i t h o u t my w r i t t e n permission.  Faculty of Commerce and Business Administrati  The University of British Columbia Vancouver, B.C.  January, 1975  ii ' ABSTRACT  . In studying population, the migration component i s the most d i f f i c u l t component to analyze e f f e c t i v e l y .  The main purpose of t h i s t h e s i s i s to  analyze and quantify the migration flow between B r i t i s h Columbia and other Canadian provinces during the period 1966-70 and develop a model which i s capable of forecasting t h i s i n t e r - p r o v i n c i a l flow with some known p r o b a b i l i t y of accuracy. In t h i s regard the d i f f e r e n t data sources which are a v a i l a b l e are evaluated. Income Tax Return data are found to be more r e l i a b l e and are a v a i l a b l e i n more d e t a i l ; therefore, t h i s source i s used.  To f i n d the best possible method of  measuring migration flow d i f f e r e n t methods of migration were considered and t h e o r e t i c a l models were found to be more sophisticated and accurate than others. Amongst these various t h e o r e t i c a l models, the push-pull theories are the basis of the most t h e o r e t i c a l hypotheses of this model.  Linear step-wise and ordinary  multiple regressions analysis, using computer "UBC  TRIP" package' are the methods  of i n v e s t i g a t i o n used to test the hypotheses.  The two complete and simpler models  of i n - and out-migration and the complete model of net-migration f o r B.C. during 1966-70 have been tested.  The combination of cross-section and  time  s e r i e s analysis with 30 observations for 5 years and 6 regions i n Canada (B.C. excluded) at 1 percent s i g n i f i c a n t l e v e l with the t o t a l of 8 independent variables are the major components underlying the operational r e s u l t s .  Distance,  unemployment rate, climate, friends and r e l a t i v e s i n other provinces were found to be the best explanatory v a r i a b l e s .  Also, climate was found to be an  v a r i a b l e to a t t r a c t the migrants to B.C.,  important  while high unemployment rate has been  the major factor of out-migration from B.C.  F i n a l l y , some area f o r further  research toward improving the quality of data which can have s i g n i f i c a n t outcome on modelling i n t e r n a l migrations have been sketched.  iii TABLE OF CONTENTS  Title  Page  ABSTRACT  'i  Table of Contents  i i  L i s t of Tables  /  7  i i i  CHAPTER I  Introduction  1  CHAPTER I I  Basic Concepts and D e f i n i t i o n s  4  CHAPTER I I I Quality Evaluation and Explanation of the Data Sources:  8  on Internal Migration 1.  Census Data  2.  Registration System  11  3.  B.C. T e l Data  11  4.  Family-Allowance Transfers Data  .5. 6.  8  '  12  Income Tax Return Data  15  The Relationship between Income Tax Data and  20  Family-Allowance Data  CHAPTER IV  Conclusion  21  Methods of Measuring Internal Migration  22  1.  Comparative Forecasting  22  2.  H i s t o r i c a l Trends as an Instrument to Forecast  23  Future Migration  3.  (a)  Graphic and Mathematical Extrapolation  23  (b)  Ratio Method  24  Using Direct Census Data  25  iv Title  Page ••  CHAPTER IV  /  Methods of. Measuring Internal Migration (Cont'd.) 4.  5.  Residual Methods or Using Indirect Census Data  27  (a)  V i t a l S t a t i s t i c s Method  28  (b)  S u r v i v a l Ratio Method  29  T h e o r e t i c a l Models  /  Conclusion  CHAPTER V  CHAPTER VI  '  / /  29 31  A Review of Current Inter-regional Migration Models  32  1.  Size-Distance Theories Model  33  2.  Market Mechanism or Economic Opportunity Model  37  3.  Push-Pull Theories Model  3  4.  Input-Output Model of Internal Migration  42  5.  P r o b a b i l i s t i c Theories Model  4  6.  Cost-Benefit Model  •  9  3  44  Conclusion  47  The Model  48  Introduction  48  A Theoretical Approach to Migration  48  T h e o r e t i c a l Hypotheses  50  An Operational Test of the Model:  The Regression  56  Results (a)  (b)  The Complete Model  56  1.  In-Migration Flow  58  2.  Out-Migration Flow  63  3.  Net-Migration Flow  66  The Simpler Model  70  V Title  Page  CHAPTER VI  t}  I  The Model (Cont'd.) T h e o r e t i c a l Problems and Deficiencies of the Model  73  Conclusion  74  Area f o r Further Research and Recommendations on Data Improvements 75  BIBLIOGRAPHY  77  APPENDIX A  83  APPENDIX B  90  vi LIST OF TABLES Table  Page 1.  2.  3. 4. 5.  6.  I n t e r - p r o v i n c i a l Migrants Aged 5 Years and Over by Province of Residence i n 1966 and 1971 .  10  I n t e r - p r o v i n c i a l Movement of Family as indicated by Family-Allowance Transfers Data 1946-73  14  Out-Migration from B.C. to Other Provinces and Foreign Countries i n 1966-71 Using Income Tax Data  17  In-Migration to B.C. from Other Canadian Provinces and Foreign Countries i n 1966-71 Using Income Tax Data  18  Net-Migration i n B.C. Tax Data  19  during 1966-71 Using Income  Hypothetical Relationships Between Dependent and Independent Variables  .  55  7.  Mean and Standard Deviation of the Variables  8.  C o e f f i c i e n t Correlation Matrix of In-Migration to B.C. .  59  M u l t i p l e Regression Results of the Complete Model of In-Migration to B.C  60  9.  /  10.. P a r t i a l Correlations between each Independent and Dependent Variables 11.  12. 13.  14.  15.  .57  62  Correlation C o e f f i c i e n t Matrix of Out-Migration from B.C  64  The C o e f f i c i e n t s and Multiple Regression Equations of Out-Migration from B.C ,  65  P a r t i a l Correlations Matrix of Out-Migration from B.C  67  C o e f f i c i e n t Correlation and P a r t i a l Correlations of Net-Migration i n B.C  68  M u l t i p l e Regression C o e f f i c i e n t s f o r Net-Migration i n B.C. during 1966-70  69  CHAPTER I.  Introduction  I n t e r - p r o v i n c i a l and i n t e r - s t a t e migration are a subject of increasing study i n Canada and the United States.  Numerous attempts have been made to  model s p a t i a l m o b i l i t y and thereby s c i e n t i f i c a l l y account f o r observed r e g u l a r i t i e s i n pattern, flows and composition.  1  On the importance of  studying migration and i t s e f f e c t s on the planning and decision-making  process,  2 L.O.  Stone says:  Migration i s an important component of population change, p a r t i c u l a r l y when viewed from the standpoint of a l o c a l community. I t i s at once an i n d i c a t o r and a generator of s o c i a l and economic changes, a l t e r i n g the s i z e and the demographic and socio-economic compositions of population. Through such a l t e r a t i o n i t influences the growth p o t e n t i a l of a community and the extent to which the community experiences c e r t a i n s o c i a l and economic problems. Migration i n a s o c i e t y may resources.  I t may  or may  .  not r e s u l t i n e f f i c i e n t a l l o c a t i o n of  contribute to the rapid development or underdevelopment of 3  a region.  Simon Kuznets  summarizes the r e l a t i o n s between migration and economic  development: Internal migration and the r e d i s t r i b u t i o n of population by residence among various parts of the country are a major way i n which people respond to changing economic opportunities emerging i n the course of economic growth.  1.  Andrei Rogers. An Analysis of Inter-regional Migration i n C a l i f o r n i a , (Berkeley: U n i v e r s i t y of C a l i f o r n i a , 1965), page 1.  2.  L.O. Stone. page 4.  3..  Simon Kuznets "Introduction", i n Hope T. Eldridge and Dorothy Swaine Thomas, Population R e d i s t r i b u t i o n and Economic Growth: United States, 1870-1950, V o l . I I I . P h i l a d e l p h i a , 1964, page 23.  Migration i n Canada:  Regional Aspects, (Ottawa, DBS,  1969),  2 People do not migrate s o l e l y f o r economic reasons.  S o c i a l and other non-economic  factors influence an i n d i v i d u a l ' s decision to migrate  (or not to migrate).  In  the present study economic and s o c i a l ; demographic and environmental factors 4 rather than only economic factors have been considered.  For the purposes of  migration projections, attempts have been made to b u i l d a model which t r i e s to , .. r e l a t e migration in a systematic way  to a set of measurable explanatory v a r i a b l e s  f o r which reasonable projections already exist or can be prepared  readily.  The main purpose of t h i s study i s to describe and analyse some of the major factors of migration, namely, inward and outward migration i n B r i t i s h Columbia. An attempt has been made to b u i l d a mathematical model which can predict future trends of migration with exogenous variables that are easy to p r e d i c t . •  Most migration models, f o r the lack of data, supposedly,  consider only net  migration flow (the d i f f e r e n c e between in-migration and out-migration This report views i n - and out- migration separately.  flow.)  Consequently, i n this  report, two models of in-migration and out-migration between B r i t i s h Columbia and s i x other Canadian Regions have been'produced. 5  Furthermore, i n t e r n a l migration^ may of view:  be approached from two d i f f e r e n t points  migraton streams and migration d i f f e r e n t i a l s .  These are not mutually  4.  As w i l l be shown i n following sections, most of the migration studies have considered only labour force population as a c t i v e and representative of the whole migrant population.  5.  These s i x Canadian Regions are: 1-Alberta, 2-Saskatchewan, 3-Manitoba, 4-0ntarip, 5-Quebec, 6-Atlantic (Substitute f o r the 4 Provinces of Newfoundland, Nova Scotia, New Brunswick and Prince Edward Island.)  6.  Internal Migration Uses versus Of External Migration: Considering Immigration and Emigration of people which i s the movement out of the boundaries of a country.  3 e x c l u s i v e c o n c e p t u a l i z a t i o n s ; each c o n c e n t r a t e s on a p a r t i c u l a r a s p e c t of tion.  A m i g r a t i o n stream  o f o r i g i n f o r any  i s t h a t volume o f m i g r a n t s  d e p a r t i n g from any one  area of d e s t i n a t i o n during a migration i n t e r v a l .  d i f f e r e n t i a l i s the d i f f e r e n c e i n c h a r a c t e r i s t i c s between m i g r a n t s migrants  and  the d i f f e r e n c e between v a r i o u s types o f m i g r a n t s .  and  non-  Studying  migration  and  therefore  c o n s i d e r e d i n t h i s s t u d y . T h i s s t u d y c o n c e n t r a t e s on the m i g r a t i o n streams  between B.C.  and  the o t h e r Canadian r e g i o n s and  f o c u s e s on the volume and  t i o n s i n p l a c e - t o - p l a c e movements r a t h e r than t h e demographic a s p e c t s and a c t e r i s t i c s of  To t e s t  the degree o f a c c u r a c y and  e s t i m a t i o n of the model s t e p w i s e and  a n a l y s e s w i t h 30 o b s e r v a t i o n s have b een used.  o f c r o s s - s e c t i o n and  To f i n d  the r e l a t i o n s h i p between the Income Tax F i l e r s Data and T r a n s f e r s Data, which has been d i s c u s s e d and found  direcchar-  migrants.  m u l t i p l e r e g r e s s i o n a n a l y s i s w i t h the combination  and  area  A migration  d i f f e r e n t i a l i s the s u b j e c t m a t t e r of demographic a n a l y s i s o f m i g r a n t not  migra-  the b e s t d a t a  ordinary time-series  sources  the F a m i l y Allowances .  shown i n the text,' have been t e s t e d  to be v e r y h i g h , but because the a v a i l a b i l i t y and b e i n g i n d e t a i l ,  former s o u r c e s of d a t a have been used.  the  4 CHAPTER I I .  B a s i c Concepts and D e f i n i t i o n s  ,  1  Before beginning the analysis of migration and attempting  .  to b u i l d a mathematical  model, i t i s d e s i r a b l e to e s t a b l i s h some basic d e f i n i t i o n s and to explain a few p r i n c i p l e s of migration research. A d d i t i o n a l concepts and d e f i n i t i o n s of migration w i l l be introduced when a p a r t i c u l a r problem i s discussed. Migration:  The e s s e n t i a l character of migration i s change i n the place  of usual residence i n a s p e c i f i e d period of time or as a change of residence from one c i v i l d i v i s i o n to another.  I t i s possible to get d i f f e r e n t  types  of migration, such as i n t e r - r e g i o n a l , i n t e r - p r o v i n c i a l , i n t r a - p r o v i n c i a l , inter-urban, rural-urban and the l i k e . —  Migrant:  A migrant i s a person who  ;  has changed h i s usual place of residence  from one migration defining area or boundary  to another at l e a s t once  during the migration i n t e r v a l .  1.  To obtain the concepts and d e f i n i t i o n s , the following sources have been used: - United Nations, Department of Economics and S o c i a l A f f a i r s , Methods of Measuring Internal Migration, Manual IV, 1961 (New York, U.N. P u b l i c a t i o n s ) , pages 1-4. - Hamilton, C.H., " P r a c t i c a l and Mathematical Considerations i n the Formulation and S e l e c t i o n of Migration Rates", Demography, No. 2 (1965), pages 429-443. - D.J. Bogue, HJS. Shryock and S.A. Hoermann, Subregional Migration i n the U.S., 1935-40, (Ohio: Miami U n i v e r s i t y , 1957), pages 1-14. - M.V. George, Internal Migration i n Canada:. Demographic Analysis (Ottawa, DBS P u b l i c a t i o n s , .1970), pages 5-8. - L.O. Stone, Migration i n Canada: Publications, 1969), pages 6-8.  2.  Regional Aspects,  (Ottawa, DBS  This boundary can be either municipal boundary or p r o v i n c i a l , county, r u r a l , urban and the l i k e .  Migration Interval: definite; e.g.,  T h i s i s t h e d u r a t i o n o f t h e study.  e.g., one y e a r ,  f i v e years,  e t c . o r i t can be i n d e f i n i t e ;  the l i f e - t i m e o f the population a l i v e a t a given  Area of O r i g i n (departure):  I t can be  For migration,  the area  date.  (or place)  from  which a move i s made i s t h e a r e a o f o r i g i n . Area of D e s t i n a t i o n ( a r r i v a l ) :  The a r e a i n which a move t e r m i n a t e s  p l a c e which a m i g r a n t s e l e c t e d f o r t h e b e g i n n i n g  o f a new l i f e  or the  i s the area  of d e s t i n a t i o n . Immigrant:  An immigrant i s an i n t e r n a t i o n a l migrant e n t e r i n g t h e a r e a o f  d e s t i n a t i o n from a p l a c e o u t s i d e t h e c o u n t r y . Emigrant:  Emigrant i s an i n t e r n a t i o n a l m i g r a n t , d e p a r t i n g  t o another  c o u n t r y by c r o s s i n g an i n t e r n a t i o n a l boundary. External or International Migration: and. e m i g r a n t s w i t h i n two d i f f e r e n t boundaries.  Refers  t o t h e movements o f immigrants  c o u n t r i e s by c r o s s i n g i n t e r n a t i o n a l  The d i f f e r e n c e between immigrants and emigrants i s r e f e r r e d  t o as n e t e x t e r n a l m i g r a t i o n . Internal Migration:  T h i s term r e f e r s t o t h e movements o f m i g r a n t s w i t h i n  a l i m i t e d boundary o f a c o u n t r y the same  o r g e o g r a p h i c u n i t t o another a r e a  within  country.  M i g r a t i o n Stream:  A m i g r a t i o n stream i s t h e t o t a l number o f moves made  during a given migration  i n t e r v a l t h a t have a common a r e a o f o r i g i n and  a common a r e a o f d e s t i n a t i o n . Migration D i f f e r e n t i a l or S e l e c t i v i t y :  T h i s term r e f e r s t o t h e c h a r a c t e r -  i s t i c s o f m i g r a n t s such as age, s e x , economic s t a t u s , e t h n i c o r i g i n , o f e d u c a t i o n a l achievement and t h e l i k e .  level  Migration d i f f e r e n t i a l s are usually  the common s u b j e c t o f demographic a n a l y s i s o f m i g r a t i o n and h i g h l y dependant upon t h e a v a i l a b i l i t y o f p a s t d a t a f o r comparison.  6 Migration Rates:  Some analysts use migration r a t e and some migration r a t i o .  In t h i s study migration rates w i l l be used and applied when the amount of migration i s divided by the population exposed to the l i k e l i h o o d  (risk)  of migrating, or, i n other words, those p o t e n t i a l population which have the same p r o b a b i l i t y of movements.  Choosing the needed rates depend upon  the kinds of migration data used, on the objective of analysis, and on such p r a c t i c a l matters as convenience or ease of c a l c u l a t i o n , customary p r a c t i c e , and popular understanding of what a migration rate means. (Hamilton, 1965, page 429). In-Migration:  Every movement i s an out-migration with respect to the area  of o r i g i n and an in-migration with respect to the area .of destination.  An  in-migrant i s thus a person who enters an area of destination by crossing i t s boundary from some point outside the areas but within the same country. In-Migration Rate =. the t o t a l number of in-migrants i n the s p e c i f i e d migration i n t e r v a l by the t o t a l population of the sending area. TIN-MI G In-Migration Rate =  ;  A-=»BC ?  Out-Migration:  Symbolically:  A  An out-migrant i s a person who departs from an area of  o r i g i n by crossing i t s boundary to a point outside i t , but within the same country. Out-Migration Rate = the r a t i o between the t o t a l number of out-migrants to an outside area to the t o t a l population of the o r i g i n . OUT-MIG Out-Migration Rate =•  BC—»A P  BC  Symbolically:  Net-Migration:  Net-migration refers to the balance of movements in the  opposing directions or the algebraic difference between i n - and outmigration i n the specified areas and time intervals.  Net migration i s a  purely mathematical concept and there i s no net migrant i n the real world. A plus sign denotes a net migration gain and a minus sign denotes a net migration loss, referred to as net in-migration and net out-migration. Net-Migration Rate = the ratio between the. total algebraic net-migration to the multiplication of the population of the area of origin and the area of destination. Symbolically: NET-MI G - , Net-Migration Rate = P  x A  BC P BC  :  Which each rate can be multiplied by 100 or as common practice by 1000 • or else depends upon the need of the researcher. Gross-Migration:  Gross migration refers to all'movements within the  specific definitions of migration that i s being applied or in other words, gross migration i s the algebraic total of i n - and out-migration in specified areas and time intervals.  Gross migration i s usually used to  Illustrate the significance of total moves and mobility of populations within two regions.  8 CHAPTER I I I . Q u a l i t y Evaluation and Explanation of the Data Sources on I n t e r n a l Migration  There are a number of sources of data that can be used to measure i n t e r n a l migration.  Each has i t s own advantages and disadvantages.  In this section  attempt w i l l be made to explain and evaluate each p o s s i b l e source of data i n Canada which can be e s p e c i a l l y u s e f u l f o r a study of B.C. m i g r a t i o n .  1.  Census Data:  1  Census data have been and s t i l l are the major sources of  2 information on i n t e r n a l migration i n most countries of the world.  The  •census data i n Canada conducted completely each ten years and p a r t i a l l y being modified each f i v e years since 1921 by the Federal Agency of Dominion Bureau of S t a t i s t i c s (established i n 1918) which i n 1971 has changed i t s name to S t a t i s t i c s Canada. questions about i n t e r n a l migration are:  In census enumeration, the usual place of b i r t h , place of l a s t  residence; duration of residence i n the place of enumeration and place of residence on a s p e c i f i c date before the census.  The main d e f i c i e n c i e s of  the census data are: - Errors on the respondents' ledge; misunderstanding  side such as:  f a u l t y memory; f a u l t y know-  of the question; deliberate f a l s i f i c a t i o n f o r  s o c i a l , p o l i t i c a l or prestige reasons and boundary changes of the geographical u n i t s without awareness of the respondents.  1.  At the end of t h i s research, a few recommendations w i l l be made with respect to possible improvement i n the q u a l i t y of data sources.  2.  United Nations, Department of Economics and S o c i a l A f f a i r s , Population Studies, Methods of Measuring Internal Migration, 1970 (New York, 1970), page 3 .  - Errors on the enumerators' side such as:  not giving the respondent enough  time to think (does not exist i n the case of mailing census); careless .entry on the schedule and misunderstanding of the question and answer. - One of the main problems i s that the timing of migration i s unknown and f i v e years census i s very long. - The census data do not r e f l e c t : A - the m o b i l i t y of persons who died or l e f t Canada between two censuses. B - M u l t i p l e moves during the f i v e or ten years period are not included and hence i t underestimates i n t e r n a l migration.  Change of d e f i n i t i o n s and the geographical units i n census, makes i t d i f f i c u l t e s p e c i a l l y f o r i n t r a - p r o v i n c i a l migration f o r comparison.  In the context of  Current Census data, t h i s deficiency may be i l l u s t r a t e d by the following example In B.C. ten census d i v i s i o n s were established i i i 1929 f o r the 1931 census.  They  were p a r t i a l l y modified f o r the 1956 census and have now been discontinued and 3 replaced by the 29 newly created regional d i s t r i c t s i n 1971 census. shows the i n t e r - p r o v i n c i a l migrants using census data.)  (Table 1  With the above d e f i -  ciencies i n census data, the r e g i s t r a t i o n system i s the better s o l u t i o n f o r measuring the i n t e r n a l migration.  3.  S t a t i s t i c s Canada, Catalog No. 92-712, V o l . 1 - Part 1, Feb.1973.  TABLE 1  o  Interprovincial Migrants Aged 5 Years and Over by Province of Residence in 1966 and 1971 Province of Residence i n 1966 Newfoundland Prince Edward Is. Nova Scotia New Brunswick Quebec Ontario Manitoba Saskatchewan Alberta British Columbia Yukon Northwest Terr. From Outside Canada Province of Res. in 1966 not stated Total In-Migrants Source:  Nfld.  P.E.I.  _  340  N.S.  225 2,380 1,140 2,145 6,290 210 180 425 420 15 140  1,975 1,310 755 3,060 400 95 190 330 15 -  3,440 2,180 8,315 6,090 18,805 1,825 840 2,000 2,490 80 105  4,035  1,500  13,150  6,165 23,770  1,285 11,255  10,165 69,485  -  N.B.  Province of Residence i n 1971 Que. Ont. Man. Sask.  1,740 2,230 1,335 635 7,635 4,345 7,905 9,315 12,450 48,370 1,405 4,630 475 1,515 1,130 3,305 1,195. 4,740 75 ? 90 130 220  Alta.  17,885 3,565 25,730 18,555 99,435 23,875 11,810 17,655 21,210 745 830  675 265 1,660 1,360 4,330 18,245 16,365 7,190 6,310 125 380  285 800 125 505 620 3,300 490 2,i55 1,575 7,755 6,845 ' 23,550 9,425 17,410 41,910 — 10,575 6,085 27,765 80 645 250 1,770  137,620  438,010  31,510  11,730  7,225 73,080 53,125, 288,685  90,240 770,035  9,015  12,130 9,805 100,545v 57,890  Statistics Canada, Migration Data, 1971 Census, May 31, 1974.  1. Excludes Canadian Stationed abroad i n the Armed Forces or Diplomatic Service.  59,880  '•• B.C.. 1,445 600 6,075 3,115 16,740 47,395 26,915 29,920 58,915 _.  1,960 1,105  Yukon  Total 0 N.W.T. Migrant  50 30 5 105 160 65 140 20 400 170 495 1,115 550 250 805 465 1,215 2,225 2,560 1,055 130 210  114,695  895  22,995 43,770 210,440- 352,660  1,045 7,445  925  11 2.  Registration System:  R e g i s t r a t i o n systems population as used i n d i f f e r e n t  European countries and Japan are p o t e n t i a l l y an excellent source of data for the study of past trends on migration.  The usefulness of the r e g i s t e r s  f o r analyses of i n t e r n a l migration depends upon the way i n which they are designed, i t s completeness and i t s periodic assembling of s t a t i s t i c a l data. Various techniques f o r population r e g i s t r a t i o n include:  legal identification  of i n d i v i d u a l s , e l e c t o r a l r o l l s , s e l e c t i o n f o r m i l i t a r y service,  social  4 security records and tax l i s t s .  As Walter Isard  (1960, page 59) has  ; mentioned r e g i s t e r s have never been used i n the United States and Canada. In the absence of continuous population r e g i s t r a t i o n , some l i m i t e d data of a s i m i l a r character may be a v a i l a b l e from such sources as s p e c i a l l o c a l censuses, school enrollment data, s o c i a l security and unemployment insurance data, c i t y d i r e c t o r i e s , wartime r a t i o n i n g r e g i s t r a t i o n s , draft r e g i s t r a t i o n and the l i k e .  Such a system can provide accurate, r e l i a b l e and s u f f i c i e n t data sources on i n t e r n a l migration.  In Canada attempts are being made to use t h i s  system such as using tax f i l e r s data which i s the main data sources of t h i s research.  3.  B.C. T e l Data:  Doing i n t r a - r e g i o n a l or i n t r a - p r o v i n c i a l studying of  i n t e r n a l migration, one of the r e l i a b l e sources of data can be the B r i t i s h Columbia Telephone Company which has a c t i v i t y almost a l l across the province of B.C. The information that could be c o l l e c t e d i s : - Month and year of telephone no. assignment (or the date of in-migrating to the area)  4.  Walter Isard, Methods of Regional A n a l y s i s : An Introduction to Regional Science (Massachusetts: The M.I.T, Press, 1960), page 59.  12 - Telephone  p r e f i x and a r e a code o f p r e v i o u s t e l e p h o n e  ( o r where d i d they  come from?) -  Occupation  - Marital status - Dwelling  tenure  U n f o r t u n a t e l y , as f a r as i n t e r n a l m i g r a t i o n i s concerned, information related  a l l o f t h e above  t o t h e head of t h e f a m i l y w i t h o u t knowing t h e number o f  dependents.  A l s o t h i s i n f o r m a t i o n i s a v a i l a b l e o n l y f o r persons who have  own phones.  T h i s d a t a c o v e r s m u l t i p l e moves and can be more a c c u r a t e than o t h e r  data sources. such a s :  their  B.C. T e l d a t a would be more u s e f u l i f q u e s t i o n s were i n c l u d e d ,  t h e number o f dependents; how l o n g d i d they s t a y i n t h e p r e v i o u s  a d d r e s s ; and b i r t h p l a c e .  I t would then be one o f t h e b e s t s o u r c e s o f d a t a on  internal migration.  4.  Family-Allowance  Another data.  T r a n s f e r s Data:  d a t a s o u r c e f o r i n t e r n a l m i g r a t i o n i n Canada i s F a m i l y - A l l o w a n c e These s t a t i s t i c s show t h e i n t e r n a l m i g r a t i o n o f f a m i l i e s w i t h  e l i g i b l e f o r a l l o w a n c e s by o r i g i n and d e s t i n a t i o n . by t h e F a m i l y A l l o w a n c e  Transfers  children  They a r e p u b l i s h e d monthly  D i v i s i o n o f t h e Department o f N a t i o n a l H e a l t h and W e l f a r e .  The d a t a a r e a d m i n i s t r a t i v e s t a t i s t i c s c o l l e c t e d by t h e f a m i l y a l l o w a n c e when a change i n a d d r e s s o c c u r s .  branch  I n t h i s r e p o r t t h e a n n u a l sums o f t h e s e monthly  t r a n s f e r s a s . i n d i c a t o r s o f m o b i l i t y f l o w s w i l l be used which means t h a t m u l t i p l e moves o f t h e f a m i l i e s a r e i n c l u d e d . - i t covers only f a m i l i e s with  The main d e f i c i e n c i e s o f t h e d a t a a r e : children;  - no d e t a i l i s a v a i l a b l e f o r each c i t y , s e p a r a t e l y ; - i t doesn't  i n c l u d e t h e presumably more m o b i l e group o f s i n g l e p e o p l e and  young f a m i l i e s w i t h o u t  children;  13 - i t gives Information only on migration streams and not migration differentials  Family-Allowance  Transfers data has a very high correlation coefficient with  the Income Tax F i l e r s data.  This correlation i s discussed further in Section 6.  Table 2 shows the interprovincial movement of families i n to and from British :  Columbia between A p r i l 1946 and March 1974 based on Family-Allowance data.  Transfers  14  Table 2 Interprovincial movement of family as indicated by Family-Allowance Transfers Data 1946-73 Period 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973  In-Migration to B.C.  Out-Migration From B.C.  6,950 8,262 6,119 5,101 4,551 5,996 6,248 6,044 6,103 7,082 9,505 5,887 5,731 5,852 5,722 5,831 6,653 5,378 5,559 10,906 11,810 10,304 9,563 11,183 10,720 11,106 11,014 10,877  3,344 3,144 3,772 4,333 4,459 4,377 4,742 5,250 4,871 4,156 4,543 6,137 5,910 5,280 5,481 5,546 5,227 7,544 8,577 5,779 6,575 6,958 6,541 6,983 7,199 7,312 7,080 9,626  Net-Migration In B.C.  + + + + + + + + + + +  -+  + + +  + + + + + + + + +  3,606 5,118 2,347 768 92 1,619 1,506 794 1,232 2,926 4,962 250 179 572 241 285 1,426 2,166 3,018 5,127 5,235 3,346 3,022 4,200 3,521 . 3,794 3,934 1,251  1.  Source: Department of National Health and Welfare. Statistics.  Family Allowances  2.  Each period begins from April of that year to the March of the immediate next year.  15 5.  Income Tax Return Data:  The Regional and Urban Research and Development  Division of Statistics Canada has developed useful information about the movement of tax f i l e r s by age and sex.  The basic data on population movement are year-to-year address changes of those persons f i l i n g personal income tax returns.  Each computer print-out table gives  the gross flows between two areas, the net flow, and the error range of the estimates.  Each table also contains, for comparison, the total out-flow from  each area.  Each f i l e has the following information: - Locality code for the current year.  t  - Locality code for the previous year. - Age of f i l e r . - Sex of f i l e r . - Marital status of f i l e r . - A mobility code indicating international movement.  In another sample f i l e , they tried to estimate the number of dependents and their sex and ages.  The advantages of the Data are:  - It gives information about In-, Out- and Net-Migration separately. - It i s in very flexible format and is in enough detail classification to be useful for a l l kinds of migration studies. - It gives the degree of the conservative error range of each printed figure.  The deficiencies of the Data are as follows: - The number of dependents and the age groups of each moving f i l e r i s an estimation of the available information in the f i l e s .  16  - It i s only i n annual bases and doesn't cover the multiple movement of tax f i l e r within a year. - It underestimates  the mobility of population.  - The delay factor i n the availability of the Data poses problems: at present, only 5 observations are available for regression analysis.  In this study Income Tax Return Data have been used, though the relationship between the Income Filers data and the Family Transfers data i s very high and they can be substituted. This relationship w i l l be shown i n the next section. Tables 3-5 show In-Migration, Out-Migration and Net-Migration i n B.C. during 1966-71 using Income Tax Return Data.  More detailed information about the migration flows which are based on the Income Tax Return Data has been shown i n Appendix A.  TABLE 3 Out-Migration from B.C. to other Provinces and Foreign Countries i n 1966 - 71 Using Income Tax Data From B.C. •  To:  •  Year  Alberta  Saskatchewan  Manitoba  Ontario  •  Quebec  Atlantic  Northern  Foreign  Total  1966-67  11,020  2,986  2,425  7,996  2,211  1,057  1,155  4,819  33,669  1967-68  11,550  3,107  3,018  8,960  2,012  1,123  1,322  7,298  38,390  1968-69  11,890  2,889 "  3,133  9,394  2,010  1,146  1,639  8,546  40,647  1969-70  11,806  2,214  2,460  9,489  2,150  1,420  1,907  9,377  40,823  1970-71  12,508  2,937  2,998  10,363  2,445  1,799  1,869  9,304  44,223  1966-71  58,774  14,129  14,033  46,202  10,828  6,545  7,892  39,345  197,748  Source:  S t a t i s t i c s Canada, Regional and Urban Research and Development, Dec. 1973.  TABLE A  In-Migration to B.C. from other Canadian Provinces and Foreign Countries in 1966 - 71 Using Income Tax Data From:  To B.C.: Year  Alberta  Saskatchewan  Manitoba  Ontario  Quebec  Atlantic  Total  1966-67  19,695  7,641  9,577  14,466  3,787  2,070  82,840  1967-68  16,721  7,019  7,509  13,723  4,039  1,917  80,350  1968-69  18,614  8,167  7,239  15,861  4,943  2,381  81,744  1969-70  23,071  12,706  8,651  18,189  6,486  3,209  97,753  1970-71  19,804  10,006  7,244  16,731  5,738  2,411  85,917  1966-71  97,907  45,538  40,220  78,968  24,944  11,987  428,603  Source:  |  Statistics Canada, Regional and Urban Research and Development, Dec. 1973.  TABLE 5 Net-Migration in B.C. during 1966 - 71 Using Income Tax Data Northern  Foreign  Total  +1,013  +280  +19,350  + 49,171  + 2,027  +  794  -155  +20,957  + 41,960  + 6,467  + 2,933  +1,235  -157  +14,511  + 41,097  + 6,191  + 8,700  + 4,336  +1,789  -395  +14,552  + 56,930  + 7,069  + 4,246  + 6,368  + 3,293  +  -289  +13,099  + 41,694  +31,409  +26,187  +32,766  +14,166  +5,442-  -716  +82,468  +230,855  Manitoba  Ontario  Quebec Atlantic  + 4,655  + 7.152  + 6,470  + 1,576  + 5,171  + 3,912  + 4,491  + 4,763  1968-•69  + 6,724  + 5,278  + 4,106  1969-•70  +11,265  +10,492  1970-•71  + 7,296  1966-•71  +39,133  Year  Alberta  Saskatchewan  1966-•67  + 8,675  1967-•68  Source:  612  Statistics Canada, Regional and Urban Research and Development, December 1973.  20 6.  The Relationship Between Income Tax Data and Family-Allowance  Data  Since most of the internal migration researchers in Canada have used the Family-Allowance Data (in the absence of the Income Tax Data) and using the Income Tax Data in this research, I was eager to see how these two sources of data are related and whether or not or in what degree of accuracy one can substitute one of them in the absence of another.  To test the  relationship, I have run a simple linear least-square regression analysis.  Y represents the income tax f i l e r s in region A moving to region B (dependent variable) and X i s the families of region A which moves to region B variable).  (independent  The regression of Y on X with the 18 observations for the years  1968-69 to 1970-71 are as follows: Independent  Correlation Matrix  CoStandard efficient Error of  Migration Flow  Dependent Variable  In-Migration  Income Tax  F. Allowance  .97  765.1  5.78  .37  , .05  .94  Out-Migration  Income Tax  F. Allowance  .98  -.38  4.60  .22  .05  .97  Gross-Migration Income Tax  F. Allowance  .96  -66 A2  5.63  .30  .05  .91  Variable  Intercept  Estimate  With these high correlations (.96-.98) and R the two data sources i s significant.  Significant Level  2 R  (.91-.97) the relationship between  They are therefore substitutes for one  another in relation to high correlation and low standard error of estimates in five percent significant level.  5.  Vanderkamp has found the similar relationship between the two sources of data in 1964 with the correlation coefficient of 0.98 and R =.961 which are quite close to the conclusion of this report. See John Vanderkamp, "Inter-regional Mobility in Canada: A study of the Time Pattern of Migration", Canadian Journal of Economics, Vol. 1, No. 3 (Aug. 1968), page 599. 2  21 Conclusion;  The studies of d i f f e r e n t data sources on migration i n Canada reveal the f a c t that, there are not any completely and absolutely r e l i a b l e and accurate  sources  of information yet, but among these a v a i l a b l e sources, the annual Income Tax Data are more f l e x i b l e and In more d e t a i l than others.  Population Registers i s the best t h e o r e t i c a l source of information on migration. Census data on i n t e r p r o v i n c i a l migration i s a r e l i a b l e source with c e r t a i n l i m i t a t i o n s , but, because of changes i n d e f i n i t i o n s , they are not u s e f u l f o r time s e r i e s a n a l y s i s of c e r t a i n periods i n i n t r a - p r o v i n c i a l studies of migration. Family Allowances Data are a v a i l a b l e only f o r net-migration and deal with the people who  have c h i l d r e n and are not representative of the highly mobile people  with no c h i l d r e n at a l l . B.C. persons who  T e l Data r e l a t e d only to the family heads and  the  own phones but show inter-movements of people i n c e r t a i n migration  i n t e r v a l s very w e l l .  The r e l a t i o n s h i p between the Income Tax Data and the Family-Allowance  Transfers  Data i s very high and i n the absence of one of them another can be used. s u b s t i t u t i o n has c e r t a i n b u i l t - i n l i m i t a t i o n s which should be considered.  This  22 CHAPTER IV.  Methods of Measuring Internal Migration  In population studies, migration seems to be the least explained of a l l and i s certainly the most unpredictable factor.  In this regard, N.A. Humphreys i n  the 1880's commented that*: Migration was rather distinguished for i t s lawlessness than for having any definite law. There are different types of measuring internal migration from very simple to the most d i f f i c u l t and sophisticated one.  Selecting any of them depends on the  time data sources, accessibility to the computer and knowledge of the researchers. Some of the methods are useful only for measuring the past trends and some for both past and the future. The advantages and disadvantages of the most common methods are discussed i n the following paragraphs.  1.  Comparative Forecasting: In this method, the future growth of the study area i s assumed to follow the pattern of another older area whose earlier growth has exhibited characteristics similar to those anticipated for the study area. By selecting a pattern area whose growth i s substantially completed, the 2 entire course of growth of the study area i s defined.  This method i s the least satisfactory method of measuring internal migration because: 1.  No exact comparable pattern areas are available;  1.  E.S. Lee, "A Theory of Migration", Demography, 3 (1966), page 47.  2.  Walter Isard, Methods of Regional Analysis: An Introduction to Regional Science, chapter 2, "Population Projection", and chapter 3, "Migration Estimate", pages 5-79, (MIT, 1960).  23 2.  The future trend of the growth of an area may not be similar to the past trend of a similar area;  3. Migration decisions are influenced by certain factors such as social, p o l i t i c a l , economic and natural phenomenon which can not be equal i n two areas;  Considering the above limitations of this method, there are not enough reasons to agree with the current popular opinion that the future trend of migration in B.C. w i l l follow the same trend as existed in California, or similar area.  2.  Historical Trends as an Instrument to Forecast Future Migration: There are different ways which migration can be predicted when using the past known trends.  These are:  (a) Graphic and Mathematical Extrapolations: Graphical extrapolation i s perhaps the easiest way of forecasting migration which most of the time has unsatisfactory results.  The  3 procedure has been described by Walter Isard as: determine migration totals or rates for some desired sequence of periods i n the past; f i f a trend line or curve to the data by use of freehand, graphic or mathematical methods; and extrapolate to obtain total migration for desired periods i n the future or to obtain future rates of movement. This process i s only valid i f we assume that the relationships that have existed i n the past w i l l continue to exist i n the future and with the same intensity.  Mathematical extrapolation assumes that past population growth has  followed some law to which future growth w i l l continue to conform and also mathematical refinements and s t a t i s t i c a l tests are possible with the expected future variations i n the factors of migration growth.  3.  Ibid, page 64.  In these methods, the  24 longer the past trend periods, the better the results, only i f anything extraordinary hasn't happened which, obviously i n this case, w i l l cause changes to the future trend.  These extrapolative methods are mostly used to check the forecasts made by other methods. (b) Ratio Method The ratio method i s similar to the comparative method but with improvements.' In this method, a present and future relationship among the factors causing growth is postulated while i n the comparative 3-1  technique the regions needed to be similar to each other.  The  general different assumptions of the ratio methods which may b.e applicable i n internal migrations can be summarized as: a.  It assumes that migration for a given area i n the future i s a simple function (e.g. a percentage) of total population growth of the area.  b.  It assumes that migration i n an area relates to total population growth of some other (larger) area, such as the nation, for which there exists a population forecast.  c.  It assumes that migration i n a given area can be predicted i f there exists forecasts of migration in a different area.  d.  And finally, migration can be determined by possible determinants of migration, i.e., economic opportunity, social and p o l i t i c a l phenomena may be natural factors.  Accuracy of this method depends upon the correctness of the particular relationship selected to establish the ratio.  3-1  I b i d , page 65.  The major weakness of this method as i s  25 true for the previous mentioned methods i s i t s dependence on the past trend of migration.  This method i s often used for open regions  such as cities whose growth may be  linked to national growth.  3.  Using Direct Census Data: This method i s useful only when attempts have been made to determine the past trends of migration.  In the census enumeration, questions such as:  place of birth, place of residence x years ago, duration of last residence w i l l be asked which are the basis of estimating intercensal migration.  The simple procedure i s to subtract the place of birth information for the earlier census from that of the later census.  To show i t symbolically^ assume: 1^, 0^, and  are the i n - , out- and net lifetime migrants, respectively at the  f i r s t census i n province A; 1^* (Xjt  ML, are the corresponding migrants in the second census in the same province  A; then the net number of intercensal migrants in A would be: M = ( I - 0 ) - (I 2  2  x  - 0) = M 1  2  - M  x  4.  An "open" area i s an area which migration movements i s not controlled or directly counted, such as a city, province or regional d i s t r i c t while a "close" area i s an area which control or enumeration are affected such as a nation.  5.  M.V. George, "Estimation of Inter-provincial Migration for Canada from Place of Birth by Residence Data, 1951-61", Demography, (Feb. 1971), Pages 123-139.  26  This procedure of estimating net intercensal migration i s subject to errors in the basic data, and errors caused by mortality, return multiple movements and international migration.  migration,  Also, this formula gives  an estimate of the number of surviving net migrants and not total intercensal net migrants.  The effect of mortality can be considered by studying i t s effect on the estimates of intercensal net in-migration (C^-O^).  (^-I^) and net out-migration  To do this let's further assume:  S i s the overall intercensal survival ratio; *  P i s the proportion of surviving in-migrants among 1^, who were enumerated at the second census;  SM ±  Surviving in-migrants to A during the intercensal period who were enumerated i n the second census;  SM = Surviving out-migrants from A during the intercensal period who 0  were enumerated i n the second census;  then, the number of life-time in-migrants to A at the second census, I^* i s equal to:  i  2  -.^-(Ij-S) I  ±  + SM - (l-DSI^ ±  27 and the number of out-migrants from A enumerated in the second census w i l l be : 6  °2  =  °1 "  (  1  "  S  )  0  l  +  SM  0  " < - ) i 1  p  S O  This method underestimates the volume of internal migration, because i t does not include multiple movements of migrants during the intercensal period.  In this method, the shorter the intercensal period, the better  the measurement of internal migration.  4.  Residual Methods or Using Indirect Census Data: This method estimates migration using intercensal demographic data.,  Any  change in the population between two dates for any given geographic area i s the result of natural increase (births minus deaths) and net migration. So, i f we know the changes in population at two points of time (by the census) and' knowing the number of births and deaths in the same periods (by either v i t a l statistics or survival ratio), the number of migrants are simply the difference of actual and expected^humber of population in the later census. Two residual methods are discussed here:  6.  Complete discussion of this method can be found in the following sources: - M.V. George, "Internal Migration in Canada: Demographic Analysis", DBS, 1970, pages 17-30 and pages 233-243. - Walter Isard, "Methods of Regional Analysis; An Introduction to Regional Science", MIT Press, 1960, pages 56-58. n  - M.V. George, "Estimation of Interprovincial Migration for Canada from Place of Birth by Residence Data, 1951-61", Demography, (Feb. 1971), pages 123-139.  28 (a) Vital Statistics Methods V i t a l Statistics method assumes that, in the absence of population migrations, the growth of the population of a given area over a given period of time w i l l be equal to the difference between births and deaths i n the area.  Any difference between this theoretical change and the  actual change, i s defined as net-migration during the period.  This method also assumes that reliable v i t a l statistics of births and deaths of the residents of the given area are available.  The estimate  of net migration, then would be: N e t  M  i - <  P  n,i " t,i> " P  t+  where for any given area, Net P  < W  = net migration for an age grouping i ,  , i s the population at the earlier census for the age grouping i , t, i  P... . i s the population at the later census i n age group i , B. and D. t+n,i i i are the number of -births and deaths within the age group i , respectively, that occurred to residents of the area during the intercensal period.  The accuracy of the estimates for any given level of age group classification depends on the accuracy of statistics on population, births, deaths, as well as the age grouping of the data.  The main deficiency of this method i s that i t gives only the estimates of net migration.  With this method i t i s not possible to determine the origin  of the in-migrants or the destination of the out-migrants.  Not knowing  the area of origin and destination, studying the inter-regional migration i s almost impossible.  29 With the method, i t is possible to find a relatively small net migration into an area while the amount of in-migrati.on and out-migration is very large or vice versa.  This method tends to underestimate the number of  migration, because i t doesn't include multiple movements of the migrants during the intercensal period.  (b)  Survival Ratio Method  When the v i t a l statistics on the births and deaths are inadequate, this method may be used. bilities.  This method involves the use of survivorship proba-  The basic information required are the number of people at two  successive cenuses and a set of survival ratio, obtained either from l i f e tables or census data which can be applied to the population at the f i r s t census in order to derive an estimate of the number of persons expected to survive to the second census.  The difference between the enumerated popu-  lation at the second census and the expected population i s the estimate of net migration.  This method has disadvantages similar to those mentioned in the Vital Statistics methods and i s less accurate.  Theoretical Models: In the recent years, most of the researchers in migration areas attempted to identify the factors which affects the migration in a given area.  In  such models, the relationship between these factors as independent variables with the migration rate as dependent variable and their solution either by a) multiple regression analysis, b) simultaneous equations, c) simulation analysis or d) systems approach are used to explain the past trend and forecast the future trend of migration.  30 In multiple regression analysis, which has been used i n this report, linearity has been assumed with the form of equation of: M  ij  =  b  0  +  b  l 1 X  Where x^, x^*  +  >  b  2  X  2 ™ +  +  b  n  X  n  +  u  0  represent the values taken by the several independent  variables such as, climate, income differentials, unemployment rates, and etc., and b^, b^t  » b^ are constant coefficient which can be either  positive or negative and M „ represents migration from i to j and U Q i s error term. Any coefficient, say b » indicates the change in M^. to be 2  associated with a unit change i n the corresponding variable x when 2  allowance has been made for the other independent variables.  Most of the  theoretical models emphasize the costs and returns of the migration i n the terms of cost-benefit analysis for a migrant.  Economic factors,  especially comparison of job opportunity i n the origin and destination are the most important factors which can influence the decision to migrate, but i n this study other factors such as social, natural and environmental variables as well as economic factors have been considered. disadvantages of this technique are:  The main  *"•  - It assumes that causal relationships existing i n the past or present w i l l continue to operate i n the future with the same relative intensity. - High correlation coefficients are sometimes misleading, since a high degree of correlation implies no necessary causal relationship between the dependent and independent variables. - Certain conditions are necessary i f the derived s t a t i s t i c a l i n ferences are to be valid.  These are such that multivariable  normality must exist for establishing the significance of coefficients.  Linear assumption i s not always true and multi-  31 collinearity between the independent variables are the most constraints of this method.  Conclusion:  There are different methods of measuring internal migration from the simplest one such as graphical extrapolation to the most d i f f i c u l t theoretical models. Any method of measuring internal migration i s not a complete method and each one has i t s own advantages and disadvantages.  The basis of selecting them  depends on the availability of data, time, cost and the knowledge of'researchers.  Comparative analysis and graphical and mathematical extrapolation methods, are the simplest and least satisfactory methods i f our objective i s to forecast migration flow with a certain degree of accuracy, they wholly depend upon the past trends and can not be the basis for forecasting.  Direct or indirect  measures of the census data with application of either v i t a l statistics or survival ratio are reliable measurements only for net migration which cannot be indicators of in-migration or out-migration.  Theoretical models are the  most sophisticated and elaborate methods and have significantly contributed in the past ten years to the explanation of the importance and magnitude of major general factors which can affect the decision of a migrant.  In this  report multiple linear regression analysis has been used to build a suitable theoretical model of internal migration in B.C., following chapter.  which i s discussed in the  32 CHAPTER V.  A Review of Current Inter-regional Migration Models  In the previous section we reviewed the different methods of measuring internal migration.  In this section most of the current migration models w i l l be reviewed.  The number of theoretic models of internal migration have greatly increased in recent years.  Researchers believe that migration i s clearly patterned  and with accurate data and advanced mathematical models may be forecast with a resonable degree of accuracy.  There are six distinct conceptual approaches to modelling migration: - Size-Distance theories models, - Market Mechanism or Economic opportunity models, - Push-Pull theories models, - Input-Output models, - Probabilistic theories models, and - Cost-Benefit models.  Each one of the above migration model categories has adopted specific mathematical expression:  for example Size-Distance theory models have been  formulated as gravity models, push-pull theories models as regression models, and probabilistic theories models as stochastic processes and simulation models. Each model views either economic factors, and or behavioral factors as important phenomena which influence migrants.  This section w i l l briefly review some of the most important models of internal migration.  Attempts w i l l be made to support each model with an example.  33 1.  Size-Distance Theories Models*: Size-Distance theories models (also called gravity models) are based on the Newtonian Gravity concept.  Gravity models of migration are quite  simplistic and do not have much explanatory power.  Size-Distance theories oriented models merely specify that migrants are attracted to an area because of the size or i t s attractiveness and better availability of job opportunity and that they are inhibited by their distance from i t .  The assumptions of a general gravity model can be  shown symbolically as: M.. -  K  where:  f(s)  = the number of migrants from i to j ; K  = constant derived from actual data;  f(s) = a function of size or the attractiveness of the two areas; and f(d) = a function of distance between i to j .  Here distance may be a proxy for transportation costs, earnings foregone while unemployed during a move, non-pecuniary or psychic costs of moving, differentials in psychic income associated with the sending and receiving areas and uncertainty 2 about prospects i n the new area due to lack of information.  Two well-known gravity types models worth mentioning here are Zipf's and Stouffer's models.  1.  I have borrowed the t i t l e s from: Andrei Rogers, An Analysis of Inter-regional Migration in California, (Berkeley, 1965), pages 2-19.  2.  P.R. Shaw, "Migration Theory and Fact: A Review and Bibliography of Current Literature", mimeograph, (1971).  34 Zipf hypothesizes that the total flow of migrants between two areas of i and j have direct relationship between population of i (P^) and population of j (P^) "and inverse relationship with the shortest transportation distance between them, (Dij).  These relationships may be shown symbolically as: P P = Klili  M  Dij  3  3  Zipf concluded that  "The theoretical reasons for expecting that the inter-  community movement of goods (by value) and of persons between any two communities, P^ and T^, are separated by an easiest transportation distance, D, w i l l directly proportionate to the product, P^ x V^*  an(  * inversely proportionate to the  distance, D." Stouffer in his famous work in 1940 assumed that "There i s no necessary relation4  ship between mobility and distance."  Instead Stouffer's model introduces the  concept of intervening opportunities.  It proposes that the number of persons  going a given distance i s directly-proportional to the number of opportunities at that distance and inversely proportional to the number of intervening opportunities . Here, opportunities are defined in terms of total in-migrants.  In a later  3.  George K. Zipf, "The l 2 Hypothesis: On the Intercity Movements of D Persons", American Sociological Review, (December 1946), page 686.  4.  Samuel A. Stouffer, "Intervening Opportunities: A Theory Relating Mobility and Distance", American Sociological Review, (Dec. 1940), page 846.  r  r  35 statement  5  Stouffer changed some of his original definitions.  Intervening  opportunities now are defined as the total number of in-migrants to points lying inside a c i r c l e , not centered at i , but one having as i t s diameter, D i j , the straight line connecting i and j .  His formula then becomes: M  where:  i_» j  =  c  M  l i i  Mi—*j = number of migrants flowing from i to j ; Mi.  = total out-migration from i ;  M.j  =* total in-migration to j ;  Kj  - t o t a l number of in-migrants from a l l places; and  MQ  = total number of out-migrants from a l l places.  Gravity models have been tested several times* and Stouffer's model provides better results.  /  The major deficiencies of the "Size-Distance" Model are: - They are not able to express causal structure of the movements. - The usual operational d i f f i c u l t i e s associated with gravity models are present.  *.  The bibliography of testing the Gravity Model i s enormous. The literature i s surveyed In Gerald A.P. Carrothers, "A Historical Review of the Gravity and Potential Concepts of Human Interaction", 22 Journal of the American Institute of Planners, 94-102, (May 1956); and i n Walter Isard and David E. Brahmall, "Gravity, Potential, and Spatial Interaction Models," Ch. 11 of Isard et a l . , Methods of Regional Analysis: an Introduction to Regional Science (The Technology Press and John Wiley and Sons: New York and London, 1960).  5.  , "Intervening Opportunities and Competing Migrants", Journal of Regional Science, 2 (Spring 1960), Pages 1-26.  36 - These models are essentially an extrapolation of observed data. - D remains constant as distance, while i t i s indicator of transportation and psychic costs, and cannot be constant.  Gravity models with the above deficiencies cannot be used for forecasting purposes.  Therefore, i n order to make i t more sensitive with the factors  which influence a migrant, i n the late 1950's a Dutch Demographer^, Somermeijer developed a new model on the basis of Zipf's Interactance Hypothesis with the help of indices of attractiveness (F) of each destination.  Somermeijer's  indices of attractiveness included such features"as per capita income, percent unemployed, degree of urbanization, recreational resources, and quality of dwellings^.  His formula with dividing directional flows are as follows: (In-Migrants to j from i)  M. 1  (Out-Migrants from j to i)  M  .  J*K + C(Fj -  *J  j—»l  Fi)l i P  P  j  >»(Dij)  a  %K - C(Fj - Fi) I i j P  P  Wj)*  Gross migration i s the sum of the above two formula, or the equivalent Zipf formula:  (Dij)  3  6.  An English-language discussion of Somermeijer's work w i l l be found i n H. ter Heide, "Migration Models and their Significance for Population Forecast", 41 Milbank Memorial Fund Quarterly, (Jan. 1963), pages 56-76.  7.  Ira S. Lowry, Migration and Metropolitan Growth: Two Analytical Models 1966, pages 9-11.  37 2.  Market Mechanism or Economic Opportunity Model: Comparative economic opportunity between the two areas i s the driving motivational force which i s manifested in inter-regional migration pattern. Migration of the labor force i s important here, and according to this thesis, employment, opportunity of better occupation and salary are the major consideration in any decision to move.  g In 1966 Ira S. Lowry proposed the following conceptual form of the model:  M  •  i-.j  U.  W.  L, L  Uj  W  Dij  K  ±  4  Where the symbols are defined as follows: M.  .  L. L.  = number of migrants from place i to place j ; - number of persons i n the non-agricultural labor force at i and j , respectively;  U. U.  = unemployment as a percentage of the c i v i l i a n non-agricultural labor force at i and j , respectively;  W. W.  = hourly manufacturing wage, in dollars, at i and j , respectively;  Dij  = a i r l i n e distance from i to j , in miles.  This formula can be transformed in log form which i s easy to use in a multiple linear regression: Log M _^ ±  8.  - Log K + Log U  ±  - Log  - Log W  Ira S. Lowry, op_, c i t . , page 12.  ±  + Log W^ + Log L  ±  + Log L^ - Log Dij  38 Andrei Rogers i n 1967 with a few modifications tested the Lowry's model for California and suggested the following modified formula as Lowry-Rogers model:  "My - K  l i F  IL  WS,  .  L F  Dij  1  Where: My  = number of migrants from i to j ;  U. U. = c i v i l i a n unemployment rate at i and j , respectively LF. LF. = labour force eligibles at i and j , respectively; WS. WS. = per capita wages and salaries at i and j , respectively, and; Dij  = shortest highway mileage between the major county seats at i and j .  Later, Rogers eliminated the unemployment variable (because of the unexplainable behaviour of i t ) and f i n a l l y suggested the following model:  WS.  LF  • Dij  1  WS,  The log-transformed  LF. J  form of Rogers' model suitable in multiple regression  anlaysis i s :  ta U  to8  M  =  b  0  +  b  i^ n  W S  1  +  b  2  L o g  n  W  S  j  +  b  3  L o g  n  L  F  i  +  b  u  L o g  n  L  F  j  +  b  5  L o g  n  D i j  The market mechanism model has an accessible causal interpretation and the model assumes that the interchange between each pair of places i s independent  9.  Andrei Rogers, "A regression Analysis of inter-regional Migration in California", The Review of Economics and Statistics, (May 1967), pages 262-264.  39 of that between each other pair.  This makes i s possible to apply cross-  sectional analysis on migration among places.  The major deficiencies of this model are: - The relationships between migration and the independent variables cannot be interpreted s t r i c t l y i n terms of undirectional causation, i n other words, i t cannot show the feedback relationships between dependent and independent variables. - I t does not identify population subgroups whose migration behaviour may not be related to the values of the variables mentioned above*". - In this model non-economic factors such as behavioral or natural factors have not been considered.  Although market mechanism, or economic opportunity model i s superior to the Size-Distance model but because of i t s deficiencies and i t s inability to explain non-economic variables i t cannot be the basis of our forecasting purposes; thus, next we w i l l try to explain the push-pull theory model which has overcorned some of the deficiencies in previous models.  3.  Push-Pull Theories Model: Push-Pull theories of internal migration views causes of migration i n both the origin and the area of destination for a migrant.  Andrei Rogers  summarized the theories as**:  10.  P.R. Shaw and Chris Guild, Elements of the Population Submodel HPS, (Nov. 1972), mimeograph, page 5.  11.  Andrei Rogers, An Analysis of Inter-regional Migration i n California, (Berkeley, 1965), page 7.  40 Most commonly this point of view emphasizes the thesis that internal migration can best be explained as a response to changing economic opportunities brought about by the differential impact of economic growth....Higher wages, improved income opportunities, and better working conditions are typical economic variables that are cited along with those of a principally non-economic nature such as educational opportunities, and housing conditions. The "push" factors pressure a migrant to move from the origin to find a better opportunity and the "pull" factors in the destination attract the migrant with factors such as better job opportunities, better salary, better education, entertainment, better climate and other personal satisfactions.  There are no "laws" of migration in this model, rather, each model-builder, depending on his study area, chooses certain factors such as socio-economic, psychological, p o l i t i c a l and natural environment and tries to relate them by the regression analysis.  In the analyses, an attempt has been made to determine the influence of each factor acting simultaneously but independently of the others.  The general form of the push-pull theories relations can be shown as:  Y = b + b, x, + b- x- + o 1 1 2 2  + b x +u n n  where: b  = intercept or constant term;  Q  b. b_ — - , b = partial regression coefficient; x  l x j  2  , X  q  = independent variables which reflect the characteristics of  migrants and both the place of origin and destinations;  41 Y *» dependent variable which can be any migration rate (in-, out- or netmigration) or an absolute migration number;. u = residual term.  12 Donald Bogue i n a very detailed analysis and tabulation of internal migration in the U.S., has pointed out that in studying migration, migration stream analysis should relate rates of i n - , out-, and net-migration to differences in the socio-economic destination area.  environmental conditions existing at both origin and  He distinguished between Metropolitan and  areas and considered 12 independent variables.  Non-Metropolitan  This model has many variables  which makes i t d i f f i c u l t to test, but with the level of i t s operating d i f f i c u l t i e s 13 we agree with Lowry's conclusion as : Bogue's findings are impressive. On the whole, they point to the driving influence of employment differentials, educational attainment, and the mobility of white-collar employment.  •-  Tarver's model*^ i s more sophisticated than Bogue's model.  He included  demographic variables as well as social and economic variables. He uses netmigration rate only and included interaction components in the model.  The general form of Tarver's model i s :  Y = b  Q  + Z  ±  + Z  2  + Z  3  + Z Z + Z^ X  2  + Z ,Z + Z ,Z ,Z + U 2  3  1  2  3  12.  Donald J. Bogue, Henry S. Shryock and Siegfried A. Hoermann, Subregional Migration in the United States, 1935-40, Vol. 1, 1957, page 65.  13.  Andrei Rogers, OP. CIT, page 9.  14.  James D. Tarver, "Predicting Migrations", Social Forces, 39 (March 1961), pages 207-213.  42 Where: Y = 1940 - 1950 state net-migration rate; t> • intercept; o  Z^ » set of economic variables; "* set of social variables; V  Z^ • set of demographic variables; U = error term.  In Tarver's model independent variables account for 72 percent of the variation ]  in Y while the introducing of an interdependence contribution, he improved 26 percent of the predictive accuracy of a regression migration model.  •Other push-pull models are similar though containing less independent variables and as a result most of them have less explanatory power.  In this report, the chosen model i s a series of push-pull models which w i l l be . f u l l y discussed in the next section.  4.  Input-Output Model of Internal Migration:  Esse Lovgren*^ in 1957 has followed an especially interesting and methodologically unique attempt to analyze the migration stream in Sweden using Leontief inputoutput model.  He hypothesized that there i s close association between a l l the in-migrants from different origins to a certain destination and a l l the out-migrants from this destination to a l l other places.  15. '  Esse Lovgren, Analysis", i n Bruno Odeving University of  "Mutual Relations Between Migration Fields: A Circulation Migration in Sweden: A Symposium, Eds. David Haunerberg, (Lunds, Sweden: Department of Geography, The Royal Lund, 1957), Pages 159-169.  43 In place of the usual focus on interindustry response to a given change in f i n a l demand i n a typical input-output model Lovgren sets up an analysis of Sweden's interarea migratory response to a given change in the migration to Stockholm*^.  The results are very impressive, however, Lovgren's method would  be d i f f i c u l t to apply i n Canada. The main reasons are:  1.  Sweden has a registration population system which can be accessed to provide accurate sources of data.  2.  The nature of the two countries are different (distance between the c i t i e s , information and others).  5.  Probabilistic Theories Model:  The probabilistic theories model generally adopts one of two directions: analytic or synthetic. The analytic approach views movement as a simple phenomenon wherein population units undergo changes of state according to a set of "transition" probabilities. implied.  In these models no causal structure i s  The synthetic approach considers movement as a complex chain of  events about which a great deal of empirical information i s available.  The  problem here becomes one of synthesizing this piecemeal information, concerning various classes of units, into a stochastic process which has a reasonably close correlation to happenings in the external world*^.  Probabilistic theories models are especially suitable for simulation on an electronic computer.  However, they are very expensive in terms of computer  time and need detailed data.  16.  Andrei Rogers, OP. CIT, page 14.  17.  Andrei Rogers, OP. CIT, pages 14-15.  44 Dan Price i n 1958 has suggested the following formula which gives the probability 18 of an individual moving from one state to another  P O ^ A ^ ) = f(A,B,C,D) + f(X,Y,Z) + f(A,B,C,D) (X,Y,Z)  where: I  ±  = i  t  individual;  h  Aj = j*"* area of origin; 1  A^ «= k*"* area of destination. 1  f(A,B,C,D) = function defining the individual's proclivity to migrate; f(X,Y,Z) = function selecting the appropriate point of destination i n relation to the individual point of origin; f(A,B,C,D) (X,Y,Z) = interactions between the individual's characteristics and other areas.  According to Price's conclusion, movement of a person from one area to another relates to the characteristics of the individual and the characteristics of the area of origin and destination.  6.  Cost-Benefit Model:  An early attempt to migrate may be considered within a general model of investment i n human capital.  The costs of the investment, such as moving  costs, must be compared with the potential returns, such as the income differentials accruing to the transferred workers.  18.  Dan 0. Price, "A Mathematical Model of Migration Suitable for Simulation on an Electronic Computer", i n Proceedings of the International Population Conference, (Vienna: International Union for the Scientific Study of Population, 1959).  45 19 Sjaastad's model • based on the cost-benefit model.  He assumes that a person  moves only If the value of a l l future monetary benefits from moving i s greater than the monetary costs involved.  20  The cost-benefit model can be expressed by the following equations  *1=1  (I + r )  J  X  where: Y ^ j = earnings i n the j  t  b  year at the destination;  Y J = earnings in the j  t  b  year at the origin;  Q  T = cost of moving;N =• total number of years in which future returns are expected; r = rate of interest used to discount future earnings. The determination of the interest rate and the total number o f years (N) to be used i n the above formula presents a problem which is typical of most costbenefit analyses. 21  Paul Shaw  gives a better illustration of a cost-benefit model including  elements of decision-making process and amenity considerations.  In the basis  of his proposals, prospective migrants are not only concerned with, but are 19.  Larry A. Sjaastad, "The Costs and Returns of Human Migration", Journal of P o l i t i c a l Economy, Supplement on "Investment in Human Beings", Vol. 70, No. 5, Part 2 (1962), Pages 80-93.  20.  Alden Speare, "A Cost-Benefit Model of Rural to Urban Migration in Taiwan", Population Studies, 1971, 25(1), Pages 117-130.  21.  Paul Shaw and Chris Guild, OP. CIT, Pages 5-6.  46 able to calculate, expected pecuniary and non-pecuniary benefits of relocation. He uses the following decision-making  formula to predict migration:  (some threshold value);  where: M - migration of an individual (or homogeneous group) from i to j ; R\j = returns to the individual at place j ; = returns to the individual at i and costs foregone in relocating; I = earnings at place i and j specified to A = age, E = education, S = s k i l l level, Z = other socio-economic conditioners of earnings potential e.g. race, sex; U = probability of unemployment at place i or j ; A = non-pecuniary returns at place i , j ; ^  = costs incurred in moving to place i to j including physical  costs, earnings, foregone while travelling and costs of installation at j .  The cost-benefit model, attempts to explain internal migration in terms of i t s various causal factors. However, determining the time process for costs and returns as they occur and estimating actual accounting discount rates makes a definite conclusion very d i f f i c u l t to obtain.  47 Conclusion; Examining, explaining and predicting internal migration has been the purpose of 'many studies i n different countries during the past thirty years.  For  forecasting purposes, natural increase and ratio methods are not adequate, therefore theoretical models have been the subject of the most recent studies. Each model from the simplest one such as size-distance model to the most d i f f i c u l t probabilistic model has i t s own advantages and disadvantages.  Use-  fulness of each model depends on the knowledge of the model-builder, purposes of the study, availability of data and other factors. These factors w i l l a l l be considered in the following chapter describing the development of the B.C. model.  48 CHAPTER VI.  THE MODEL  Introduction: Most internal migration studies base their analytical framework on the classical model of factor mobility and resource allocation.  Migration from i to j w i l l  occur i f Income (wage, salary and other disposable income) increases.  Due to  this increase i n migration flow, allocation of resources are perfect and i n the long-run a country w i l l have economic adjustment or the efficinet allocation of resources.  The underlying classical assumptions are:  (1) the goal of  maximizing income for each individual, (2) perfect knowledge of migrant about different areas, (3) migrants are many in number and homogeneous in s k i l l s and tastes, and (4) there are no barriers to mobility.  This paper represents a model that considers economic aspects of migration, but also eliminates some of the unnecessary classical assumptions and substitutes more real l i f e variables.  The following section consists of a theoretical  approach to migration and some ex ante theoretical hypotheses.  These  hypotheses are tested with the aid-of multiple regression analysis. Finally, problems and deficiencies of this approach are examined and suggestions are made for further research.  A theoretical approach to migration: There i s no unique theoretical approach to migration. i s the migrant himself.  The important variable  Migrants are selective people who differ from each  other in terms of personal factors, their attitudes toward the future and meaning of l i f e and other values. his own unique solution.  Each migrant responds differently and finds  When we examine a l l the movements as aggregation,  49 there are four distinct factors which can influence any one decision. They are: 1)  Factors associated with the area of origin,  2)  Factors associated with the area of destination,  3)  Intervening obstacles, and  4)  Personal factors^.  The combination of positive and negative factors at both the origin and destination may be differently defined for every migrant or prospective 2 migrant.  Similarly, intervening obstacles may be perceived differently.  Intervening obstacles are such things as distance and inconvenience of traveling.  Personal sensitivities, Intelligence, and awareness of conditions 3 elsewhere also enter into the evaluation. It i s d i f f i c u l t to develop a general theory of migration.  The usual procedure  i s to examine the determinants of migration flow by making a theoretical hypothesis and then, testing the theories to find the causes of migration. However, causes and effects of migration may change over time and in the longrun may be different than in the short-run. 4  The difference i s often referred  t  to as return migration.  *"*  In bringing a l l factors together, there are two approaches to explain migration theory: 1)  Economic Behaviour Approach  2)  Non-Economic or Behavioristic Approach  1.  Everett S. Lee, "A Theory of Migration", Demography, 3 (1966), page 47.  2.  Ibid, pages 50-51.  3.  Ibid, page 48.  4.  For further information on return migration refer to: John Vanderkamp, "Migration Flows; Their Determinants and the Effects of Return Migration", Journal of P o l i t i c a l Economy, pages 1012-1031.  50 The economic approach assumes that individuals are rational in their decisions to move or to stay. future.  Their decisions are based on costs and benefits in the  The economic approach describes migration as a decision involving  investment in human capital."* The behavioristic approach assumes that a migrant doesn't necessarily want to increase his money income, but is attempting to satisfy his needs.^  These needs can be either economic or non-economic factors  such as climate, living with relatives or other natural and cultural phenomena.  Considering a l l of the possible combinations in explaining migration flows, attempt i s made here to hypothesize various factors of migration which seem most relevant for B.C.  Theoretical Hypotheses: A decision to migrate i s influenced by factors such as the area of origin, destination, personal factors and intervening obstacles.  Obviously, an  individual w i l l migrate only i f the present value of expected future stream of inccme i s greater than the cost of relocating and other foregone income at origin.  Income here means a l l the possible benefits, not only monetary and  tangible ones. Thus, an individual w i l l tend to move i f the net benefit (benefits minus costs) i s positive.  In this section attempts w i l l be made to  isolate push and pull factors at the origin and destination and give a series of hypotheses which can be tested.  Regional income differentials usually attract people (especially labour force) to move. The significance of this variable has been tested several times by 5.  See Sjaastad  6.  For more information on this approach see: Julian Wolpert, "Behavioural Aspects of the Decision to Migrate", Papers of the Regional Science Association, Vol. 14 (1965), pages 159-169.  (1962) for this approach.  51 different researchers^ and in general can be summarized as:  the present value  of expected future income at the destination is an attracting factor to move and the present value of expected future income in the origin i s an economic benefit foregone by moving.  Therefore, we expect that the migration rate w i l l  be positively associated with the present value of the stream of income differences between the area of origin and destination. Income in this model i s real disposable personal per capita income. calculation are reviewed in Appendix B.  Definition and detailed  Sources of hypotheses are the  result of a l l previous migration studies and the current economic, social, p o l i t i c a l and natural conditions of the Province of British Columbia.  Some of  these hypotheses at the f i n a l testing are not significant (i.e. Dwelling starts) which w i l l be shown later. Hypothesis I:  The main theoretical hypotheses are the following;  Internal migration i s positively related to the real disposable personal per capita income between two regions i and j , the greater the income differentials between i and j , the higher the migration that w i l l occur.  Hypothesis II:  The unemployment rate has a positive relationship with out-migration.  Hypothesis III:  The greater the unemployment rate in B.C. the smaller the in-migration to B.C.,  they are negatively related.  A high unemployment rate can be a significant "push" factor of migration but 8 Canadian experiences  show that when the unemployment rate i s low, migration  7.  See for example L.A. Sjaastad, "The Relationship between Income and Migration...", (1960), pages 37-64, and T.J. Courchene, "Interprovincial Migration and Economic Adjustment", (1970), pages 550-576.  8.  John Vanderkamp, "Internal Mobility in Canada: A Study of the Time Pattern of Migration", Canadian Journal of Economics, 1 (1968), pages 595-609, T.J. Courchene, op. c i t . , page 554.  52 flow does not function entirely to increase in-migration.  The reason i s that  the people i n the labour force are interested in a relatively stable situation in making any decision to migrate, not just the unemployment rate which can have different trends at different times.  Distance has been used in most of the migration studies and i t represents not only the costs of movements but as a proxy for the effect of the distance on information about economic opportunities i n other regions.  Also, i t indicates  the earnings foregone during transition and the psychic costs and distribution of moving and uncertainty about income and job opportunities at the destinations. Therefore, Hypothesis  IV:  Migration rate between i and j is negatively related with the relative distance between i and j in relation to a l l other regions.  In this view, relative distance has more effect than the simple distance between i and j . Hypothesis  V:  The lower (better) the climate index , the greater i s the rate of migration.  10  In studying internal migration  in Canada, specially for B.C.,  this variable seems to have  the great influence.  9.  The term lower climate index shouldn't be confused with the lower temperature. For more explanation refer to the Appendix B.  10.  For more information about the Index please refer to Appendix B. In this Index the general climate index for B.C. has assumed 100 and the other regions w i l l increase when the climate situation is not as good as B.C.  53 Hypothesis  VI:  The greater the number of friends and relatives i n B.C., the better i s the availability of information about B.C. and the higher w i l l be migration to B.C. Therefore, they are positively related.  As a proxy for friends and relatives,  the number of persons born in i and living i n j at the beginning of the period have been used.  For more detail  please see Appendix B. Hypothesis  VII:  The labour force participation rate i s positively related to migration.  The labour force participation rate i s the  ratio between the labour force and the population of 14 years of age and over. Hypothesis  VIII:  The greater the number of immigrants from other countries 0  to B.C., the higher i s migration rate i n B.C.  Immigration  rate i n B.C. i n relation to the natural increase of population i s relatively high and this high rate has influence on the mobility of the existing population. Hypothesis  IX:  A relative change in dwelling starts i n B.C. and other regions has positive relationship with migration rate. Dwelling starts rate was included, because other migration studies have considered i t to be important, but as we w i l l see later i t i s not significant in this model.  As well, since  housing starts respond to demand factors, the implicit causality of using starts as an independent variable to predict migration i s suspect. Dwelling starts can by hypothesized as: The higher the dwelling starts i n a region, the better are job opportunities for different s k i l l s and therefore, the higher i s migration rate. Given this reasoning, employment would be a more suitable variable.  54 Hypothesis X:  The independent variables of migration have a different relationship with in-migration and out-migration.  Therefore,  net-migration (the difference between i n - and out-migration) can not be a good substitute.  Most migration studies have considered net-migration rate. reason i s the lack of d a t a .  11  The most important  Considering "push" and "pull" factors i n the area  of origin and the destination, each characterized by i t s own merits, i t i s necessary to distinguish between i n - and out-migration separately.  11.  In Canada, most of the interregional migration studies have used the Family Allowance Data which before 1968, were given by net-movements of families of children the region's.  55 The above hypotheses can be summarized as follows:  Table 6 Hypothesized Relationships Between Dependent and Independent Variables  Hypothesis In-Migration to B.C.  Hypothesis Out-Migration from B.C.  Hypothesis Net-Migration in B.C.  Variables  Operational Definition  DWELST  Dwelling Starts  pos. related  pos. related  pos. related  CLIMTE  Climate Index  neg. related  neg. related  neg. related  UNEMPR  Unemployment Rate  neg. related  pos. related  neg. related  DISTCE  Relative Distance  neg. related  neg. related  neg. related  INCOME  Real Disposable Personal Per Capita Income  pos. related  pos. related  pos. related  Labour Force Participation Rate.  pos. related  pos. related  pos. related  IMMIGR  Immigration Rate  pos. related  pos. related  pos. related  FRIEND  Friend & Relatives  pos. related  pos. related  pos. related  LFPART  pos. = positively  neg. = negatively  By testing those theoretical hypotheses with suitable s t a t i s t i c a l method, a model for internal migration i n B.C. has been accomplished.  56 An Operational test of the model:  the regression results  To make an operational test of the model, a regression analysis package been used.  12  has  To testing the complete model, step-wise regression was selected.  In step-wise regression, independent variables are ranked according to their a b i l i t y to reduce the variation in dependent variable that remain unexplained at each step. used.  In the simpler model, ordinary multiple regression analysis was  Normality of data and a simple linear relationship are assumed.  The  complete model using a l l 8 variables w i l l be discussed f i r s t and then the simpler model using only selected variables.  (a)  The Complete Model: In the complete model of internal migration in B.C., and 3 dependent variables have been used. detail in Appendix B.  8 independent variables  Data sources have been shown in  Distribution of each variable around i t s mean, and  the standard deviations of each variable are shown by the following table:  12.  James H. Bjerring and Paul Seagraves, "UBC TRIP: Triangular Regression Package", Computer Centre, U.B.C., Vancouver, June 1974.  57  Table 7:  Mean and Standard Deviation of the Variables  Name  Mean  Standard Deviation  In-Migration  5.74  4.74  Out-Migration  2.64  2.53  Net-Migration  0.0016  0.0014  Dwelling Starts  1.25  1.66  Climate  1.87  0.36  Unemployment Rate  0.82  0.39  Distance  1.38  0.70  Income Differentials  0.84  0.11  Labour Force Participation Rate  0.98  0.07  Immigration Rate  4.83  6.63  Friend & Relatives  0.17  0.10  The diversion of dwelling starts and immigration rates between the six regions outweigh a l l the rest. This model based on the combination of time series (5 years from 1966-70) and cross-sectional analysis.  The number of observations are 30 which i s the  result of multiplication of the 5 years i n 6 regions.  58 To find out the relationships between each variables their directions and magnitude of this relationship, the following Correlation Matrix table  the  has  13 been established.  1.  In-Migration Flow The correlation coefficient matrix of in-migration to B.C.  from the six  other Canadian regions between 1966-70 are summarized in Table 8. One  of the advantages of multiple regression analysis i s i t s power of  prediction. been used.  13.  In this sense, step-wise multiple regression analysis have The result can be summarized as Table 9.  A coefficient of correlation i s an index of the direction and magnitude of a relation. The formula for calculation i s :  where: x = fx - x and  Table 8; Correlation Coefficient Matrix of In-Migration to B.C.'  VARIABLE  IN-MIG  IN-MIG  1.00  DWELST  0.23  DWELST  CLIMTE  UNEMPR  DISTCE  INCOME  LFPART  IMMIGR  1.00  CLIMTE  -0.33  -0.11  1.00  UNEMPR  -0.67  -0.08  0.32  1.00  DISTCE  -0.96  - 0 : 2 5  0.27  0.59  1.00  INCOME  0.25  0.03  -0.61  -0.68  -0.10  1.00  LFPART  0.43  0.09  -0.70  -0.60  -0.30  0.86  1.00  IMMIGR  0.25  0.31  0.10  -0.07  -0.16  -0.02  -0.10  • 1.00  FRIEND  0.57  0.15  -0.67  -0.47  6.81  0.86  -0.03  -0.85  Since the relationship between in-, out-, and net-migration and independent variables are different the results of in-migration to B.C. w i l l be discussed f i r s t .  14.  FRIEND  For the definition of variables, please see the Appendix B. The above table not only illustrates the direction and magnitude of each independent variable with the in-migration but the direction and magnitude of any other independent variables with another independent variable as well. The direction of relation i n a l l variables are the same were hypothesized i n the Table 6. In conclusion, the operational test of the model verifies the theoretical hypothesis which was made before.  1.00  (15)  Table 9: Multiple Regression Results of the Complete Model of In-Migration to B.C.  Coefficient for Equation  Intercept  No.  DWELST  CLIMTE  Standard  X^  X^  X^  X^  UNEMPR  DISTCE  INCOME  LFPART  IMMIGR  FRIEND  R  Estimate  -0.451 (0.124)  -5.475 (73.7)  -4.294 (0.45)  11.689 (2.33)  Q.098 (8.15)  11.680 (1.97)  .961  1.099  -5.624 -2.889 (144.23) (0.35)  10.866 (2.32)  0.097 (8.36)  12.157 (2.28)  0.96  1.076  -5.802 (270.73)  8.694 (2.08)  0.092 (8.25)  9.047 (2.29)  0.96  1.061  8.562 (2.02)  0.083 (7.32)  9.197 (2.37)  0.959  1.06  2  Error of  1  -0.208  -0.128 (0.911)  1.975 (1.896)  2  -1.178  -0.133 (1.03)  2.177 (2.85)  3  0.021  -0.126 (0.96)  1.803 (2.65)  4  -0.198  1.874 (2.88)  -5.745 (273.3)  5  6.478  2.184 (3.92)  -5.621 (268.23)  0.078 (6.266)  15.303 (13.04)  0.955  1.082  6  11.958  -5.79 (273.1)  0.084 (6 .'61)  8.176 (12.02)  0.948  1.14  7  12.738  -5.97 (250.78)  0.932  1.26  •  15 - A l l of the variables are significant at .01 level. - Degree of Freedom is(30-J^=29. - Bracketed figures beneath the regression coefficients are F-Ratio.  7.43 (8.34)  61 The f u l l regression equation of in-migration to B.C. during 1966-70 from other 6 regions with using Equation No. 1 can be written as:  Y « -0.208 - O.mXj^ + 1.975X - 0.451X- - 5.475X. - 4.29X- + 11.689X + 3 4 5 6 0.099X + 11.68X 0.961 2  ?  £  g  R* = 0.95  1 6  In each step, the number of independent variables as has been shown i n the table 9 was reduced.  The 4 Independent variables:  Climate, Distance, Immigration to  B.C. and Relatives and Friends are the most significant variables, explaining 2 R = .955 or 96 percent of the variance of in-migration to B.C.  It remains to explain the partial correlations between the variables.  Partial  correlations show the relationship between an independent and dependent variable holding a l l other independent variables constant.  Partial correlations are  shown i n the following table 10:  16.  —2 2 R i s adjusted R for degree of freedom. K  2  -1 -  Its formula i s :  a-R ) 2  where n i s the number of observations, and k i s the number of independent variables.  CM  Table 10: Partial Correlations Between Each Independent and Dependent Variables  VARIABLE Y  - IN-MIG  Xj - DWELST  IN-MIG  DWELST  CLIMTE  UNEMPR  DISTCE  1.00  - CLIMTE  0.371  -0.03  Xg - UNEMPR  -0.239  * 0.125  -0.612  1.00  X, - DISTCE 4  -0.95  -0.207  -0.273  0.418  1.00  X  2  LFPART  IMMIGR  1.00 -0.038  X  INCOME  -  1.00  - INCOME  0.274  -0.042  0.522  -0.704  0.54  1.00  Xg - LFPART  0.246  -0.02  0.186  -0.178  0.229  0.51  X  ?  - IMMIGR  0.45  0.287  0.098  -0.038  -0.202  0.145  -0.01  1.00  X  g  - FRIEND  0.49  0.035  -0.852  -0.522  -0.47  0.869  0.85  -0.124  5  1.00  1.00  63  As has been shown in Table 10, the least explained variable i s dwelling starts and the best explained variables are distance, friend, immigration rate and climate.  2.  Out-Migration Flow:  To explain the complete operational model of out-migration from B.C. to other 6 Canadian regions during 1966-70 the same definitions and formula are used. The correlation coefficient matrix of out-migration which illustrates the direction and magnitudes of relations i s contained in Table 11.  The comparison of this table with Table 6 (theoretical hypothesis relationships) shows that a l l the hypothesized relations are correct except for the unemployment rate.  Here, again distance (-0.87), friend (0.69) and climate (-0.64) are the  most important variables i n explaining out-migration from B.C.  The relationship and coefficient quantities of variables and the multiple regression equations are shown in Table 12.  The complete regression equation considering a l l of the variables for out-migration from B.C. as shown i n Equation 1 of Table 12 i s :  Y - -1.969 + 0.35^ 6.857X.  - 0.22X + 1.912X - 3.106X + 0.226X + 0.637X - 0.014X + 2  3  R  2  4  = 0.976, << = .01  9  5  g  ?  VARIABLE  IN-MIG  OUT-MIG  1.00  DWELST  0.27  1.00  CLIMTE  -0.64  -0.1113  1.00  UNEMPR  -0.51  -0.081  0.32  1.00  DISTCE  -0.87  -0.25  0.27  0.588  INCOME  0.35  0.034  -0.605  -0.679  -0.1  1.00  LFPART  0.59  0.093  -0.7  -0.598  -0.298  0.858  IMMIGR  0.017  0.309  0.104  -0.067  -0.164  -0.019  -0.099  1.00  FRIEND  0.693  0.13  -0.919  -0.493  -0.332  0.773  0.818  -0.149  DWELST  CLIMTE  UNEMPR  DISTCE  INCOME  LFPART  IMMIGR  FRIEND  •  1.00  )  1.00  1.00  (17) Table 12; The Coefficients and Multiple Regression Equations of Out-Migration from B.C.  VD  Coefficient for: Equation  Intercept  No.  Standard  X^  X^  X^  X^  X^  Xg"  2  R  Error of  DWELST  CLIMTE  UNEMPR  DISTCE  INCOME  LFPART  IMMIGR  FRIEND  Estimate  0.226 (0.007)  6.537 (5.61)  -0.014 (1.01)  6.857 0.982 (8.27)  0.4  1  -1.968  0.035 (0.512)  -0.219 (0.126  1.912 (15.581)  -3.106 (167.01)  2  -1.978  0.035 (0.529)  -0.2 (0.126)  1.88 (38.8)  -3.09 (508.82)  6.68 (10.16)  -0.013 (1.22)  6.98 .982 (13.87)  0.39  3  -2.426  0.03 (0.5)  1.92 (48.54)  -3.1 (539.82)  6.64 (10.45)  -0.013 (1.17)  7.55 .981 (62.87)  0.38  4.  -2.42  1.94 (51.73)  -3.12 (581.95)  6.65 (10.73)  -0.01 (0.84)  7.62 .981 (66.01)  0.38  5  -2.51  1.96 (52.78)  -3.09 (595.12)  6.64 (10.76)  7.74 .980 (70.33)  0.378  17. -  I Significant level of a l l the variables i s .01. Degree of Freedom i s 29. Bracketed figures beneath the regression coefficients are F-Ratio.  66 R  2  i n these 5 equations in Table 12 shows that by eliminating 4 independent  variables, there i s not too much change in the explanatory power of the equation. As a result, we can write the out-migration equation with only 4 independent variables as follows:  Y - -2.5145 + 1.956X- - 3.095X. + 6.642X. + 7.745X3 4 6 8 R  2  = .975  oC  = .01  As was explained in the previous section, partial correlation illustrates the explanatory power of each independent variable. Distance alone, can explain 98 per cent of change in out-migration.  The  importance of distance in most of the migration studies relates to the cost of moving, uncertainty at the destination, information effect, psychic costs and so on.  Friend (86%), unemployment rate (82%), and labour force participation it  rate (55%) are also important i n explaining out-migration flow.  3.  Net-Migration Flow:  *•*  Most of the migration studies in Canada, have chosen net-migration as their dependent variable.  Gross migration data i s seldom available.  In this report  emphasis i s on gross-migration (in-, and out-migration) rather than net-migration. However, because of the necessities for comparison with other migration studies and testing fitness of the model with a l l of the three independent variables, a summary net-migration results w i l l be explained below.  The following table  shows that correlation coefficient and partial correlations of independent variables with net-migration i n B.C. during 1966-70.  * The Partial Correlation Matrix of Out-Migration from B.C. i s shown below.  Table 13: Partial Correlations Matrix of Out-Migration from B.C.  VARIABLE  OUT-MIG  OUT-MIG  1.00  DWELST  0.08  1.00  CLIMTE  -0.032  0.526  1.00  UNEMPR  0.82  0.109  -0.35  DISTCE  -0.98  -0.244  0.133  0.542  1.00  INCOME  -0.085  0.082  0.256  -0.788  0.78  1.00  LFPART  0.55  0.014  0.064  -0.43  0.195  0.585  1.00  IMMIGR  -0.184  0.31  -0.141  -0.046  -0.169  0.387  0.007  FRIEND  0.859  0.049  -0.851  0.096  -0.817  0.55  0.76  DWELST  CLIMTE  UNEMPR  DISTCE  INCOME  LFPART  IMMIGR  1.00  1.00 -0.15  FRIEND  68  Table 14: Coefficient Correlation and Partial Correlations for Net-Migration i n B.C.  Algebraic Name  Variable Name  Coefficient Correlation  Partial Correlation  Y  Net-Migration  1.00  1.00  *1  Dwelling Starts  0.126  -0.186  X  2  Climate Index  -0.022  0.703  X  3  Unemployment Rate  -0.661  0.053  X  4  Distance  -0.84  -0.81  5  Income Differentials  0.117  -0.233  6  Labour Force Participation Rate  0.243  -0.132  X  7  Immigration Rate  0.34  &. 304  X  8  Friend (Information)  0.63  0.666  X  X  As Table 14 shows, the direction of variables are„same as was hypothesized i n Table 6. Distance (81%), Climate (70%), Friend (67%) and Immigration rate (30%) are the most significant variables i n explaining net-migration i n B.C.  The results of regression coefficients, R and standard error of estimates of net-migration (S.E.^.) have been summarized i n the following Table 15.  Table 15; Multiple Regression Coefficients for Net-Migration In B.C. during 1966-70  Coefficient for: Equation N o  Intercept  -  1  -0.019*  X  j  ^  ^  ^  ^  Standard X  g  X  ?  X  g  Error of  2  DWELST  CLIMTE  UNEMPR  DISTCE  INCOME  LFPART  IMMIGR  FRIEND  R  Estimate  -0.01 (2.29)  0.13 (7.69)  -0.085 (1.23)  -0.1 (13.12)  -0.46 (2.51)  0.4 (.95)  0.004 (4.92)  0.76 (3.94)  .898  .00054  -0.05 (0.56)  -0.11 (17.32)  -0.27 (1.58)  0.0034 (3.998)  0.92 (6.87)  .894  .00054  -0.18 (1.05)  0.0031 (3.61)  (1.05) .891 (12.09)  .00053  2  0.17  -0.0096 (2.14)  0.13 (7.84)  3  0.012  -0.01 (2.39)  0.16 (17.62)  -0.12 (26.83)  -0.01 (2.48)  0.16 (20.81)  -0.14 (56.7)  0.0033 (4.16)  0.81 .886 (18.34)  .00053  0.0026 (2.55)  0.83 (18.5)  .875  .00055  0.89 .862 (20.78)  .00057  4  -0.09  5  -0.13  0.17 (21.71)  -0.13 (51.21)  6  -0.149  0.18 (25.37)  -0.136 (49.73)  18 - A l l of the variables are significant at .01 level. - Degree of Freedom i s 29. *  Bracketed figures beneath the regression coefficients are F-ratio.  A l l of the regression coefficients have been multiplied by 100.  70 Using equation number 1, the regression equation for net-migration would be:  Y = -0.00019 - 0.0001X. + 0.0013X - 0.00085X, - 0.001X. - 0.0046X- + 0.004X, + 1 2 3 4 5 6 0.00004X + 0.0076X R = .898, o<= .01 o  2  7  g  R  2  = .87  In each iteration step-wise multiple regression tests a new independent variable and i t s contribution to explaining the dependent variable.  In comparison with  others, i f i t i s non-significant, i t w i l l be dropped from the equation.  Using  this technique the best equation was:  Y - -0.0015 + 0.0018X - 0.0014X + 0.0089Xg 2  4  R  2  = 0.862,  R  2  - .85  <* = .01  Using 8 independent variables leads to the problem of multi-collinearity.  The  simpler model with less variables reduces this problem.  b.  The Simpler Model:  In this model, the results of the complete model have been used.  In addition  19 the definition of friend and relatives have been changed.  For brevity, only  the equations and their specification for in-migration, and out-migration appear here. 19.  In the previous model, the operational definition of the independent variable "Friend" for time t i n j was the number of migrants i n time t-1 from i who were living i n j . In this model, this definition has changed to the multiplication of 5 i f there i s only 1 year data, by 4 i f there are 2 years data available plus the previous year data and so on. For more explanation please refer to Appendix B.  71 At the significanat level of 1 percent and with 30 observations (degree of freedom 30 - 1 = 29) and substituting a T-test for the F-test the results of in-migration equation to B.C. during 1966-70 with using 5 independent variables are:  )  L  In-Mig Ratio - 9.0437 + 1.322 (CLIMTE) - 0.5675 (UNEMPR) - 5.5076 (DISTCE) (4.46) (5.67) (4.99) + 0.0832 (IMMGR) + 11.2766 (FRIEND) (0.61)  (6.44)  „ R = .953, S.E.Y. = 1.124  (see footnote #20.j I  The above equation can be substituted by a simpler equation with less variables by including 3 independent variables which seems are good enough to explain the pattern of in-migration flow.  The regression equation of these 3 independent  variables to explain in-migration to B.C. can be written as:  In-Mig Ratio = 11.8729 - 5.7353 (DISTCE) + 0.0867 (IMMIGR) + 8.1414 (FRIEND) (4.99) (0.614) (6.44) R  2  = .948, S.E.Y. = 1.1438  The simpler out-migration equation i n the same condition as in-migration model (d = .01, DF = 29) with 5 independent variables i s :  Out-Mig. Ratio = 0.9325 + 2.2914 (UNEMPR) - 3.3 (DISTCE) + 3.75 (INCOME) (3.9) (2.64) (3.9) 0.0187 (IMMIGR) + 8.0394 (FRIEND) (-1.69) (5.35)  20.  R  = 0.976, S.E.Y. = 0.43  The bracketed number below coefficients are the T-value. S.E.Y. = Standard Error of Estimate of Y (dependent variable).  72  If we exclude the two variables "income" and "immigration" for possible multicollinearity the f i n a l equation of out-migration from B.C. to 6 other Canadian regions i n 1966-70 would be:  Out-Mig Ratio = 3.7124 + 1.567 (UNEMPR) - 2.944 (DISTCE) + 10.233 (FRIEND) (3.9) (2.64) (5.35) R  2  - 0.972, S.E.Y. = 0.4489  73 Theoretical problems and deficiencies of the model; Choosing the multiple regression method to test the model raises a few questions which make the analysis d i f f i c u l t .  1.  The major deficiencies of this method are:  Linear assumption of the model i s not r e a l i s t i c .  Some factors should be  considered jointly since they are inextricably linked i n determining migration.  2.  Multiple regression, as a technique, i s much less useful i n accounting for migration streams between areas when tempting to consider a l l streams simultaneously.  3.  Between the independent variables there are considerable degree of interdependency and multi-collinearity which the method has ignored.  4.  The data have not been tested for normality and try to be standardized. Comparing the data and the results requires data sources which have been measured identically with the same degree of importance.  The assumption of  normality of the data requires more research.  5.  Usually, multiple regression equations are used for prediction purposes. These equations assume the future trend of migration and their characteristics are the same as have been i n the past.  This assumption i s simplistic and  cannot be guaranteed to be continued i n the future.  74 Conclusion: An individual migrant i s influenced by different factors in making his decision. These factors may be at the origin or destination. or intervening obstacles.  They may be personal factors  A complete model i s assumed to take into account a l l  of these phenomena. With reference to "push" and "pull" factors at origin and destination and general characteristics of a migrant, a theoretical hypothesis can be made to indicate the future trend and pattern of migration flow.  To test  these hypotheses a model has been built based on past experiences and linear multiple regression analysis.  Two operational tests have been conducted: "the  complete" and the "simpler" one.  The complete model i s an elaborate model which  gives a general equation with 8 independent variables on i n - , out-, and netmigration in B.C.,  separately.  Their correlation coefficients, partial corre-  lations and the regression coefficients are stated.  The simpler model on the  other hand, ignores some of the independent variables which seem non-significant in explanation of the dependent variables.  The method of multiple regression  analysis, though i t i s inadequate to explain a l l aspects of migration flow, i s well-known, acceptable and allows easy interpretation of the results. Among a l l the independent variables, distance, unemployment rate, climate and the potential number of friends or relatives found to be the most important explanatory variables while dwelling start rates was not a significant variable.  75 Area for Further Research and Recommendations on Data Improvements: Two areas for further research and four suggestions about improvements in the quality of the data sources w i l l be recommended here.  1.  Analysis in internal migration requires an extensive data base. present study, data was only available for a five-year period.  In the Since more  Income Tax Filers Data w i l l be available i n the future, this w i l l allow an update of this study.  Furthermore, intra-provincial analysis of migration  i s possible with using those income tax data.  Such a comprehensive analysis  hasn't been done for British Columbia as the published available sources indicates.  2.  "Path Analysis" and "Canonical Analysis" should be included i n the multiple regression analysis.  Path analysis i s a method of determining the relative  . magnitudes of effects of a given change i n a certain network or a controllable variable. .Canonical analysis allows variables to interact simultaneously. Geometrically, the canonical correlation can be considered as a measure of the extent to which individuals occupy the same" relative positions i n the P-dimensional space as they do in the q-dimensional space.  Finally, i t i s  important to better standardize the data in using multiple regression analysis.  3.  Since no complete data sources on internal migrations i n B.C. are available, there are a few possible solutions to this important deficiency, namely: a)  Population Registers are the best sources of internal migration flow. A few modifications and additions to the Driver's Certificate Application or the like would make them as equally valuable source for intra provincial migration.  76 b)  The Hospital Insurance Benefits and the B.C. Medical Plan are the second best sources of data; however, they are not presently accessible and are not regularly published.  c)  B.C. Tel Data could be very useful i f i t included a few other questions on their request form from the clients and provided an attempt to tabulate them i n such a way to be useful i n the migration research.  d)  Census definitions are not comparable between census periods. of them historically i s misleading.  Any use  Comparative study of the different  definitions (e.g. Census Divisions, Regional Districts, etc.) are desirable.  And the f i n a l word, survey of migration with separate sampling when needed i s very costly and inefficient which makes any of the above four suggestions more economically feasible.  BIBLIOGRAPHY  1.  Anderson, I.B. Internal Migration In Canada:  1921 - 61 - Study No. 13,  Economy Council of Canada, Ottawa, Queen's Printer, 1966. 2.  Antonova, I.F. "The Effect of Migration on Changes, Composition, and Distribution of Canada's Population", Vestnik Moskovskogo Universiteta, Geografiya, 1969, No. 5, pages 35-40.  3.  Beshers, J.M., and Nishiura, E.N. "A Theory of Internal Migration Differentials", Social Forces.  4.  Bogue, D.J., H.S. Shryock, Jr., and S.A. Hoermann, 1957. Migration i n the U.S., 1935 - 40:  Subregional  Streams of Migration between  Subregions, Oxford, Ohio: Scripps Foundation, Miami University. 5.  British Columbia, Province of: Department of Industrial Development, Trade and Commerce:  Economics and Statistics Branch. Forecasts of  Population i n B.C. to the year 2000, Victoria, B.C., 1971. 6. 7.  , Net Migration to British Columbia, Victoria, B.C., 1971. British Columbia, "Population Growths i n British Columbia", Manpower Review Pacific Regions, March - April 1970.  8.  Canada, Ministry of Finance:  Economic Review, Ottawa: April 1974.  9.  Canada, Canada Year Book, 1972, Ottawa: Queen's Printer, 1973.  10.  Canada, Statistics Canada.  Migration Data, 1971 Census, Ottawa, 1974.  11.  Canada, Statistics Canada.  Population Projections for Canada and the  Provinces 1972 - 2001, Ottawa: 12.  , Statistics Canada. Development. Ottawa:  13.  Information Canada 1974. Regional and Urban Research  Intercountry Migration Data Base Manual: Users Manual,  1973.  Cebula, R.J. and R.K. Vedder. "A Note on Migration, Economic Opportunity, and the Quality of Life", Journal of Regional Science, 1973, Vol. 13, No. 2.  78 14.  Courchene, T.J. "Interprovincial Migration and Economic Adjustment", Canadian Journal of Economics, 3 (1970), 550-576.  15.  De Voretz, D.J. Internal Migration i n Canada: Some Conjectures and Hypotheses Tested for 1901 - 1921, Discussion Paper 73-5-2, British Columbia:  16.  Duncan, O.D.  Department of Economics and Commerce S.F.U., n.d.  "Path Analysis:  Sociological Examples", American Journal  of Sociology, 72 (1966). 17.  Fabricant, R.A.  "An Expectational Model of Migrations", Journal of  Regional Science, 10 (1970), 13-24. 18.  Farrar, D.E.,  and Glauber, R.R.  "Multicollinearity i n Regressions  and  Analysis", The Review of Economics and Statistics, 49 (Feb. 1967), pages 92-107. 4  19.  Fynn, S.W.  Canadian Interprovincial Migration:  An Empirical Analysis,  Unpublished Master Thesis in S.F.U., 1972. 20.  George, M.V.  "Estimation of Interprovincial Migration for Canada From  Place of Birth by Residence Data, 1951 -  1961", Demography Vol. 8,  No. 1 (Feb. 1971) pages 123-139. 21.  , Internal Migration in Canada: Demographic Analyses: One of a series of 1961 Census Monographs, Ottawa, 1970.  22.  Glantz, F.B. The Determinants of the Interregional Migration of the Economically Disadvantages, Federal Reserve Bank of Boston, 1973.  23.  Gnanasekaran, K.S. Migration Projections for Canada, 1969-84, Analytical and Technical Memorandum No. 6 of Census Division, DBS, Ottawa, 1970.  24.  Greenwood, N.M.  "Lagged Response i n the Decision to Migrate",  Journal of Regional Science, 10: 25.  1972.  pages 375-384.  , and Sweetland D, "The Determinants of Migration between Standard Metropolitan Statistical Areas", Demography, Vol. 9, No. 4 (Nov. 1972).  79 Hamilton, CH.  "Practical and Mathematical Considerations in the  Formulation and Selection of Migration Rates", Demography, No. 2 (1965). Heide Ter. H. "Migration Models and their Significance for Population Forecasts", The Milbank Memorial Fund Quarterly, 16 (Jan. 1963), pages 56-76. Hightower, H.C.  "Population Studies", Ch.3, in W.I. Goodman and E.C. Freund  (eds.) Principles and Practice of Urban Planning, Washington,  D.C.:  The International City Managers' Association, 1968 pages 51-75. Inter-Institutional Policy Simulation Group, Population Submodel, University of British Columbia, Mimeographed documentation and computer printouts. Isard, W. Methods of Regional Analysis:  An Introduction to Regional Science,  Cambridge, Mass.; The Massachusetts Institute of Technology, 1960. Kerlinger, F.N. and Pedhazur, E.J. Multiple Regression in Behavioural Research, New York, Holt, Rinchart and Winston, Inc., Laber, G. and Chase, R.X.  1973.  "Interprovincial Migration in Canada as a Human  Capital Decision", Journal of P o l i t i c a l Economy, 79 (1971), pages 795-804. Lee, E.S.  "A Theory of Migration", Demography, 3 (1966), pages 47-57.  Lianos, T.P. "The Migration Process and Time Lags", Journal of Regional Science, Dec. Lithwick, H.  1972.  Urban Canada: Problems and Prospects, Ottawa:  C.M.H.C, 1970.  Lowry, l.S. "A Short Course in Model Design", American Institute of Planners Journal, 31 (1965), pages 158-166. Lowry, l.S. Migration and Metropolitan Growth: San Francisco: Lycan, R.  Two Analytical Models,  Chandler Publishing Company, 1966.  "Interprovincial Migration i n Canada: The Role of Spatial and  Economic factors", Canadian Geographer, 13 (1969), pages 237-254.  80 Mclnnis, R.M.  "Provincial Migration and Differential Economic Opportunity"  in L.O. Stores (ed.). Migration i n Canada. Ottawa:  Statistics  Canada, 1969. , "Age, Education and Occupation Differentials i n International Migration:  Some Evidence for Canada", Demography, 8 (1971),  pages 195-204. Meredith, J.R. and Knight, G.R. British Columbia Population Projection to 1975 i n Inventory of the Natural Resources of B.C., published by the B.C. Natural Resources Conference, 1964. Montague, J.T. and Vanderkamp, J . A Study in Labour Market Adjustment, Vancouver: .Institute of Industrial Relations, U.B.C., 1966. Morrison, P.A.  "Theoretical Issues i n the Design of Population Mobility  Models", Environment and Planning, 5 (1973)', pages 125-134. Nelson, P.  "Migration, Real Income and Information", Journal of Regional  Science, 1 (1959), pages 43-74. Officer, L.H. and Andersen, P.R.  "Labour Force Participation i n Canada",  Canadian Journal of Economics, No. 2 (May 1969). Pack, J.R. "Determinants of Migration to Central Cities", Journal of Regional Science, Aug. 1973, 13(2), pages 249-260. Rogers, A. An Analysis of Interregional Migration i n California, Berkeley: California State office of Planning, Dec. 1965. _, "A Regressions Analysis of Interregional Migration i n California", The Review of Economics and Statistics, 49 (May 1967), pages 262-267. Routledge, J.A. B.C. Demographic Forecasts, Vancouver:  B.C. Telephone  Company, Economic Studies, 1973. Schwartz, A. "Interpreting the Effect of Distance on Migrations", Journal of P o l i t i c a l Economy, 1971.  81 51.  Shaw, P.R.  "Migration Theory and Fact;  A Review and Bibliography of  Current Literature", mimeograph, 1971. 52.  Shaw, P.R.  and Guild, C.  "Elements of the Population Submodel", mimeo-  graph, Vancouver, HPS, 53.  Sjaastad, L.A.  Nov.  1972.  "The Costs and Returns of Human Migration", Journal of  P o l i t i c a l Economy, 70 (Oct. 1962), pages 80-93. 54.  Skene, P.A.T. Projection of Population and Industrial Employment In the Vancouver CM.A., Unpublished Master Thesis, U.B.C, The Department of Community and Regional Planning,  55.  Stone, L.O.  1974.  Migration i n Canada: Regional Aspects Ottawa:  Statistics  Canada, 1969. 56.  Stott, A.G.E.  Influencing Interregional Migration, Unpublished Masters  Thesis, U.B.C, The Department of Community and Regional Planning, 57.  Stouffer, S.A.  "Intervening Opportunities:  1974.  a theory relating mobility  and distance", American Sociological Review, 5 (845-867). 58.  Systems Research Group.  Canada;  Economic Projections to the year 2000.  Toronto, 1970. 59.  Tarver, J.D.  "Predicting Migration", Social Forces, 39 (1961), pages  207-213. 60.  United Nations, Department of Economic and Social Affairs.  Methods of  Measuring Internal Migration, Manual IV, Population Studies, No. 47 (1961). 61.  Vanderkamp, J.  "Interregional Mobility in Canada: A study of the Time  Pattern of Migration", Canadian Journal of Economics, 1 (1968), pages 595-609. 62.  , "Migration Flows, their Determinants and the Effects of Return Migration", Journal of P o l i t i c a l Economy, 79 (1971), pages 1012-1032.  82 63.  , Mobility Behaviour i n the Canadian Labour Force. Ottawa: Information Canada, Special Study No. 16, Economic Council of Canada, 1973.  64.  Wills, K.G.  "The Influence of Spatial Structure and Socio-Economic Factors  on Migration Rates: A Cast Study:  Tyneside 1961 - 66", Regional  Studies, Vol. 6 (March 1972), pages 69-82. 65.  Wolpert, J. "Behavioral Aspects of the Decision to Migrate", Papers of the Regional Science Association, 14 (1965), pages 159-169.  66.  Zipf, G.K.  "The P ^  Hypothesis: On the Inter-city movement of Person",  ~D American Sociological Review (Dec. 1946), pages 677-688.  /  83  APPENDIX A  In-Migration, Out-Migration, Immigration and Emigration Flows Between the 10 Census Divisions \  i n B.C.,  other Canadian Provinces and the Foreign  Countries from 1966 to 1970.  84 Sources of the data on this appendix are based on the Income Tax f i l e r s data.  The ten census divisions i n B.C. were established i n 1929 for the 1931  census and they were partially modified for the 1956 and 1961 census.  For  the 1971 census they were totally replaced by the 29 newly created Regional Districts. Data for the foreign countries refers to the total immigration and emigration flows i n that period between each Division and other countries. In order to obtain net-migration flows between the specified areas, i t i s necessary to subtract each in-migration data from i t s corresponding outmigration data. The usefulness of these data i n terms of their accuracy and r e l i a b i l i t y has been discussed i n the main text.  Table A - l :  Total In-Mlgratlon and Immigration Flows between each Census Division In B.C and other Canadian Provinces and Foreign Countries i n 1966-70  From:  To: Year  Alberta Total  Atlantic Total  Foreign Countries (Immigration) Total  Division 1  D.2  D. 3  1966-•67 1967-•68 1968-•69 1969-•70 1970-•71 1966-•71  693 486 738 1,192 1,500 4,610  887 769 773 1,027 808 4,264  1966-•67 1967-•68 1968-•69 1969-•70 1970-•71 1966-•71  43 8 51 102 68 272  63 50 55 66 37 270  22 80 104 79 68 353  1966- 67 1967- 68 1968- 69 1969-•70 1970- 71  174 204 262 251 370  541 637 498 433 497  1966- 71  1,261  2,606  D.4  8,372 2, 121 7,012 2,300 7,947 2,666 9,520 2,839 8,102 2,351 12, 277 40,953  D.5  D.6  D.7  D.8  D.9  D.10  Total B  768 365 422 673 532 2,760  1,432 1,013 1,236 1,380 1,124 6,186  19,695 16,721 18,614 23,071 19,804 97,907  2,903 2,705 2,581 3,275 2,742 14,206  987 996 1,048 1,522 1,422 5,975  899 932 1,103 1,481 966 5,380  613 452 660 777 655 3,157  52 43 82 119 98 395  31 23 12 68 101 235  141 162 179 262 194 938  184 151 115 241 212 904  22 16 20 14 12 83  2,070 1,917 2,381 3,209 2,411 11,987  522 728 549 653 687  16,851 19,965 16,208 16,405 14,722  2,425 3,063 2,690 3,103 2,838  577 577 596 631 607  299 215 132 111 110  1,248 1,503 1,239 1,404 1,356  914 963 631 700 988  618 400 252 238 228  24,169 28,255 23,057 23,929 22,403  3,139  84,151  14,119  * 2,988  867  6,750  4,196  1,736  121,813  126 1,406 45 1,030 42 1,161 54 1,589 30 . 1,193 297 6,379  Table A - l : (Continued) Year  Manitoba Total  Northern: Total  Ontario: Total  Quebec: Total  Saskatchewan Total  Division 1  D.2  1966--67 1967--68 1968--69 1969--70 1970--71 1966--71  126 57 67 183 166 599  225 123 106 165 111 730  1966--67 1967--68 1968--69 1969--70 1970--71 1966--71  73 10 73 57 60 272  40 33 80 63 112 329  1966-•67 1967--68 1968--69 1969--70 1970--71 1966-•71  140 96 157 308 292 993  322 264 264 290 312 1,451  1966-•67 1967-•68 1968-•69 1969-•70 1970-•71 1966-•71  23 24 69 171 108 395  86 44 48 107 86 370  1966-•67 1967-•68 1968- 69 1969- 70 1970- 71 1966- 71  158 135 229 764 871 2,157  323 209 365 435 457 1,788  D.3  D.4  596 5,,670 605 4,,525 724 4,,226 785 5,,197 572 4,,320 ,282 23, ,939 3, 55 97 89 70 122 434  491 525 619 709 643 2,987  422 8,819 627 8,648 777 10, 509 912 U, 569 858 10, 321 3,,596 49, 866 75 82 163 209 176 705  2,714 2,971 3,627 4,690 4,082 18, 085  819 3,207 043. 3,041 1, 3,394' 1,211 4,706 1,847 3,654 1,297 6,217 18, 002  D.5  D.6  D.7  D.8  D.9  D.10  Total  1,612 1,306 1,210 1,196 1,307 6,631  380 260 293 395 284 1,612  53 76 17 55 42 243  385 297 279 383 240 1,584  266 170 223 207 127 994  291 90 94 85 75 635  9,577 7,509 7,239 8,651 7,244 40,220  262 167 220 175 133 958  63 27 67 64 79 300  6 8 1 7  22  -  128 99 115 137 142 622  134 81 93 94 128 529  183 120 124 136 158 720  1,435 1,167 1,482 1,512 1,580 7,176  135 85 92 108 128 548  661 568 628 765 679 3,302  658 521 390 587 506 2,661  258 135 108 114 112 727  14,466 13,723 15,861 18,189 16,731 78,968  81 52 90 183 165 571  93 50 50 43 32 269  102 123 99 209 177 709  131 175 174 221 248 949  79 48 14 6 31 178  3,787 4,039 4,943 6,486 5,738 24,994  1,373 431 1,260 430 1,441 489 1,994 967 1,532 705 7,600 3,022  40 13 22 57 47 179  699 479 561 1,128 773 3,640  237 174 173 505 402 1,491  354 235 282 303 268 1,442  7,641 7,019 8,167 12,706 10,006 45,538  2,587 464 2,325 454 2,499 437 2,957 579 3,057 466 13,424 2,400 403 470 609 647 633 2,763  '  Table A-2:  Total Out-Migration and Emigration Flows between the Canadian Provinces, Foreign Countries and each B.C.'s Census Division i n 1966-70  To;  From: Year  Alberta Total  Atlantic Total  Division 1  D.2  D.3  D.4  D.5  D.6  1966- •67 1967- •68 1968- •69 1969- •70 1970- •71 1966- •71  606 545 582 576 731 3,039  797 868 743 614 500 3,522  719 934 1,029 1,089 1,097 4,868  '4,131 4,590 4,844 4,911 5,403 23,879  1,162 1,254 1,046 1,500 1,442 6,404  575 534 564 627 668 2,969  1966- •67 1967- •68 1968- 69 1969- •70 1970- •71 1966- 71  20 23 19 28 39 129  56 48 40 70 62 275  17 28 39 60 57 201  416 492 596 637 926 3,066  313 345 240 428 365 1,692  32 32 16 44 94 218  I  D.7  D.8  D.9  67 1,190 68 935 50 • 1,029 50 909 34 1,057 269 5,120  472 454 459 412 487 2,285  1,301 1,368 1,541 1,118 1,089 6,416  11,020 11,550 11,890 11,806 12,508 58,774  85 55 72 49 130 390  19 26 50 24 30 150  1,057 1,123 1,146 1,420 1,799 6,545  25 5 17 3 14 64  74 69 57 77 82 360  D.10  Total (from B  Table A-2:  (Continued)  To:  Manitoba Total  Northern Total  Ontario Total  From: Year  Division 1  D.2  D.3  D.4  1966-•67 1967-•68 1968-•69 1969-•70 1970-•71 1966-•71  30 51 41 28 59 209  97 109 134 59 73 472  48 133 198 108 107 594  1966-•67 1967-•68 1968-•69 1969-•70 1970-•71 1966-•71  56 51 28 66 59 261  79 107 68 140 105 499  1966-•67 1967-•68 1968- 69 1969-•70 1970-•71 1966-•71  176 107 83 89 136 592  198 322 323 190 230 1,264  D.5  D.6  D.7  D.8  D.9  1,485 1,743 1,897 1.479 1,727 8,330  309 429 320 391 467 1.916  103 112 107 86 94 503  42 27 11 13 42 135  161 179 166 133 217 • 856  77 118 80 110 178 5,562  73 117 179 53 34 456  2,425 3,018 3,133 2,460 2,998 14,033  31 40 47 48 108 274  422 443 662 802 720 3,048  128 150 156 189 187 810  32 52 83 79 114 361  9 5 7 21 18 60  133 133 187 187 199 838  109 141 149 150 130 679  156 200 252 225 229 1,062  1,155 1,322 1,639 1,907 1,869 7,892  258 278 275 308 339 1,458  4,912 5,727 6,365 6,249 7,037 30, 289  1,262 1,178 1,036 1,463 1,327 6,266  211 313 304 207 247 1,283  61 100 77 74 51 363  403 381 359 413 488 2,044  354 374 393 412 430 1,963  161 180 179 84 78 682  7,996 8,960 9,394 9,489 10,363 46,202  D.10 Total (from  Table A-2:  (Continued)  To:  From: Year .  Division 1  D.2  D.3  D.4  D.5  D.6  D.7  D.9  D.10  98 71 61 76 96 403  87 87 95 110 180 559  17 38 41 12 4 112  2,211 2,012 2,010 2,150 2,445 10,828  1966- •67 1967- •68 1968- •69 1969- •70 1970- •71 1966- •71  45 29 18 24 27 144  73 78 50 52 55 309  25 33 48 24 43 174  1,557 1,423 1,467 1,556 1,674 7,676  246 195 188 244 271 1,144  37 38 17 38 81 210  Total  1966- •67 1967- •68 1968- •69 1969- •70 1970- •71 1966- •71  146 113 71 64 111 505  162 166 142 103 136 709  199 212 196 166 • 231 1,003  1,122 1,243 1,368 952 1,315 6,000  420 436 301 399 438 1,994  111 207 199 139 182 837  15 44 18 7 16 100  517 344 310 162 289 1,622  149 201 183 114 129 775  145 141 101 108 90 584  2,986 3,107 2,889 2,214 2,937 14,129  Foreign Countries (Emigration) Total  1966- •67 1967- •68 1968- •69 1969- •70 1970-71 1966- •71  52 67 63 94 116 392  145 204 194 198 232 973  120 3,203 157 5,164 232 5,939 300 6,513 256 6,384 1,065 27,204  673 885 1,089 1,172 1,169 4,988  115 167 193 210 273 958  77 56 55 55 45 289  213 242 • 315 370 437 1,577  163 208 273 329 311 1,284  58 148 193 136 81 615  4,819 7,298 8,546 9,377 9,304 39,345  Quebec Total  Saskatchewan  26 20 25 14 12 97 .  Total (from  D.8  90  APPENDIX B  Specifications and Definitions of the Variables and their Sources  91  In this appendix, attempt has been made to point out the data sources which have been used In the model.  Definitions of each dependent and i n -  dependent variable and the way which each variable i n the model has been calculated w i l l also be explained.  1.  Dependent Variables; Dependent variables i n this model are in-migration, out-migration and  net-migration rates. In-migration rate has been defined as the ratio between In-migrants from each 6 regions to B.C. between 1966-70, to the total population of each sending area.  Symbolically: IN-MIG. AB  IN-MIG.^ P. A  =  Out-migration rate i s the ratio between the total number of out-migrants from B.C. to each 6 regions to the total population of B.C. i n 1966-70. Symbolically: OUT-MIG. =  0 U T  -  M I G  P  -BA  B.  Net-migration rate i s the ratio between the difference between In- and Out-migrants and the multiplication of both the sending and receiving regions. Symbolically: NET-MIG. = < - -AB) IN  ^'^'B^  MIG  P  A« B P  Data sources for dependent variables are Statistics Canada: and Urban Research & Development and The Census publications.  Regional  92  2.  Independent Variables To run the model, two sets of ordinary and stepwise multiple regression  analyses have been used.  Altogether, 8 independent variables have been  considered: 1.  Dwelling Starts Rate (DWELST): Dwelling start rate i s the ratio between the annual percentage change i n dwelling starts from each region vis-a-vis B.C.  Further  one-year time lags i n the process of decision-making have been assumed.  If we assume that:  D s (A) = Dwelling Starts i n time t i n Region A, and; t  D S  ^(A)  =  Dwelling Starts i n time t-1 i n Region A, and;  D S (B), D S _ (B) = Dwelling Starts i n time t-1 i n B respectively. t  fc  1  The formula for calculating the rate of change i n dwelling starts would be as follows:  DWELST Rate  =  D S (A) - D S ._ (A) t  (  1  . D S (B) - D t  S^B)  This formula i s similar to the calculation of elasticity i n economics which measures the rate of change rather than the magnitude of absolute number. The data hve been obtained from the Central Mortgage and Housing Corporation:  Canadian Housing Statistics, 1973, Tablel2.  93  2.  Climate (CLIMTE): In a country as large as Canada, measuring climate as a degree of performance for a potential migrant i s very d i f f i c u l t .  Mean daily temperature  and i t s maximum and minimum, the number of frost free days, darkness, wind-chill, and precipitation can a l l be important i n decision-making. In this regard, a Climatological Index Map for the use of Treasury Board Officials which evaluates the relative advantages and disadvantages of various isolated posts i n Canada has been used.  On the basis of this  index map, the following climate index has been obtained: Region  Index  British Columbia .  100  Alberta  123  Saskatchewan  200  Manitoba  321  Ontario  162  Quebec  192  Atlantic  215  Northern  262  As these indices show, the greater the percentage of difference, the higher the probability to move.  1  1.  The climate information and the index map were kindly given to me by Mr. Phil Suckling, Ph.D., Department of Geography, U.B.C.  94 Unemployment Rate (UNEMPR): The unemployment rate i s the number of unemployed persons expressed as a percentage of labour force (total employed and unemployed persons) 14 years of age and over. In the model, the ratio of unemployment rate between B.C. and other regions has been used.  Statistics Canada "The Labour Force" i s the data source for this variable.  Distance Ratio (DISTCE): As an indicator of the distance variable, the highway mileage between chief metropolitan centres of the Canadian provinces has been used.  In  the model, the distance ratio i s the highway mileage between Vancouver and the major city of another region to the average distance of that city centre to a l l other regions. DISTCE DISTCE = V  Symbolically, the formula i s : >A  DISTCE i = i  1  —* / A  5  The data source of this variable i s the "Canada Yearbook 1972" by Statistics Canada.  Income Differentials (INCOME): With the assumption of a one-year time lag, this variable i s actually Real Disposable Personal Per Capita Income which has been calculated as follows:  Total personal disposable income i n any region has been divided by the population of that region i n the same year to get per capita income.  95 This figure i s divided again by the Consumer Price Index to obtain the real income figure.  In the model this obtained value has been used as  the income variable.  Statistics Canada "The Economic Review, A General  Review of Recent Economic Developments" by John Turner, i n April 1974, i s the major source of this variable.  Labour Force Participation Rate (LFPART): The labour force participation rate i s the total male and female labour force (as a percentage of the population 14 years of age and over) expressed as a percentage of the population of 14 years of age and over. In the model, the ratio of labour force participation rate in each region to the equivalent rate for B.C. has been used.  The data source is. "The  Labour Force" (71-201) by Statistics Canada.  Immigration Rate (IMMIGR): Immigration rate i s the ratio between the number of immigrants to each region and B.C. i n each year during 1966-70. The formula for calculating immigration i s the same as the elasticity formula which has been explained under Dwelling Starts.  Statistics Canada i s the data source of this  variable.  Information Rate (FRIEND): This independent variable i s the number of persons born i n " i " and living i n " j " at the beginning of the period. To calculate this variable, the income tax return data i s availabe for only 5 years (1966-70).  In the  f i r s t year (1966) FRIEND equals the number of i n - or out-migrants.  In  the second year (1967) FRIEND i s the average number of the f i r s t two years. In the third year FRIEND i s the average of the f i r s t three years, and so on.  96  The following two tables are the input data after f i n a l calculation and adjustment, as have been used in the model.  97  Table B-l: List of Input Data on the In-Migration to B.C. During 1966-70. Region  (1)  Period Dependent variable (2)  1966-67 67-68 Alberta 68-69 69-70 70-71  IN-MIG (3) 11.41 11.22 12.21 14.8 12.42  Independent Variables DWELST CLIMTE (4) (5)  UNIN- LFIMEMPR DISTCE COME PART MIGR FRIEND (6) (7) (8) (9) (10) (11)  1.12 .98 5.16 .74 2.16  1.23 1.23 1.23 1.23 1.23  .59 .41 .53 .41 .41 .59 .46 • .41 .86 .41  .87 .92 .91 .94 .92  1.09 1.07 1.06 1.06 1.06  .78 4.13 .67 6.45 6.48  .34 .34 .34 .33 .33  66-67 67-68 Saskat- 68-69 chewan 69-70 70-71  8.01 7.33 8.51 13.26 10.63  1.35 .75 -.97 -1.08 6.77  2.0 2.0 2.0 2.0 2.0  .59 .33 .35 .39 .62  .69 .69 .69 .69 .69  .80 .85 .74 .81 .77  .96 .90 .92 .94 .28 .93 .94 14.41 .94 30.12  .13 .13 .14 .15 .15  66-67 67-68 68-69 69-70 70-71  9.95 7.8 7.46 8.84 7.37  .69 .35 1.21 3.04 1.83  2.31 2.31 2.31 2.31 2.31  .59 .56 .69 .46 .90  1.08 1.08 1.08 1.08 1.08  .83 .82 .86 .89 .84  1.02 .90 1.00 6.09 .34 .99 1.05 12.70 1.05 7.33  .17 .16 .16 .14 .13  66-67 67-68 Ontario 68-69 69-70 70-71  2.08 1.93 2.18 2.46 2.22  1.3 .86 1.98 .07 .40  1.61 1.61 1.61 1.61 1.61  .59 .56 .61 .59 .62  2.08 2.08 2.08 2.08 2.08  .98 1.03 1.03 .99 1.03 .87 1.00 1.02 1.02 1.03 .99 4.28 1.02 .99 5.66  .25 .26 .26 .26 .26  66-67 67-68 68-69 69-70 70-71  .66 .69 .83 1.08 .95  1.14 .16 2.5 -.35 -.54  1.92 1.92 1.92 1.92 1.92  1.29 1.04 1.04 1.10 1.38  2.15 2.15 2.15 2.15 2.15  .79 .79 .81 .82 .81  .98 .88 .99 1.62 .98 1.33 .97 9.32 .96 15.60  .06 .07 .07 .08 .08  66-67 67-68 68-69 69-70 70-71  1.05 .96 1.19 1.58 1.18  .59 .15 3.29 1.14 .65  2.15 2.15 2.15 2.15 2.15  1.76 1.42 1.29 1.24 1.46  1.87 1.87 1.87 1.87 1.87  .62 .63 .65 .68 .67  .89 .79 .89 1.25 .87 .83 .86 -2.44 .85 10.27  .04 .04 .04 .04 .04  Manitoba  Quebec  Atlantic  98  Table B-2: L i s t of Input Data on the Out-Migration to B.C. During 1966-70. Region  ' (1)  4  1  Period Dependent variable (2)  Independent Variables  OUT-MIG DWELST (3) (4)  1966-67 67-68 Alberta 68-69 69-70 70-71  7.53 7.75 7.80 7.57 7.84  1.12 .98 5.16 .74 2.16  66-67 67-68 Saskat- 68-69 chewan 69-70 70-71  3.13 3.25 3.01 2.31 3.12  66-67 67-68 Manitoba68-69 69-70 70-71  CLIMTE (5)  UNIN- LFIMEMPR DISTCE COME PART MIGR FRIEND (6) (7) (8) (9) (10) (U)  1.23 1.23 1.23 1.23 1.23  .59 .53 .59 .46 .86  .41 .41 .41 .41 .41  .87 .92 .91 .94 .92  1.35 .75 -.97 -1.08 6.77  2.0 2.0 2.0 2.0 2.0  .59 .33 .35 .39 .62  .69 .69 .69 .69 .69  .80 .85 .74 .81 .77  .96 .90 .94 .92 .93 .28 .94 14.41 .94 30.12  2.52 3.13 3.23 2.51 3.05  .69 .35 1.21 3.04 1.83  2.31 2.31 2.31 2.31 2.31  .59 .56 .69 .46 .90  1.08 1.08 1.08 1.08 1.08  .83 .82 .86 .89 .84  1.02 .90 1.00 6.09 .99 .34 1.05 12.70 1.05 7.33  .09 .09 .09 .09 .09  66-67 67-68 Ontario 68-69 69-70 70-71  1.15 1.26 1.29 1.28 1.37  1.3 .86 1.98 .07 .40  1.61 1.61 1.61 1.61 1.61  .59 2.08 .56 2.08 .61 2.08 .59 2.08 .62 2.08  .98 1.03 1.03 .99 1.03 .87 1.00 1.02 1.02 1.03 .99 4.28 1.02 .99 5.66  .29 .29 .30 .30 .31  66-67 67-68 68-69 69-70 70-71  .38 .34 .34 .36 .41  1.14 .16 2.5 -.35 -.54  1.92 1.92 1.92 1.92 1.92  1.29 1.04 1.04 1.10 1.38  2.15 2.15 2.15 2.15 2.15  .79 .79 .81 .82 .81  .92 .88 .99 1.62 .98 1.33 .97 9.32 .96 15.60  .08 .08 .07 .07 .07  66-67 67-68 68-69 69-70 70-71  .54 .56 .57 .70 .88  .59 .15 3.29 1.14 .65  2.15 2.15 2.15 2.15 2.15  1.76 1.42 1.29 1.24 1.46  1.87 1.87 1.87 1.87 1.87  .61 .63 .65 .68 .67  .89 .79 .89 1.25 .87 .83 .86 -2.44 .85 10.27  .04 .04 .04 .04 .04  Quebec  Atlantic  I  1.09 1.07 1.06 1.06 1.06  .78 4.13 .67 6.45 6.48  .40 .40 .39 .39 .39 ' .11 .11 .10 .10 .09  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0099981/manifest

Comment

Related Items