Occupational Mobility in the United States 1930–1960 9780231887397

Citation preview

OCCUPATIONAL MOBILITY IN THE U N I T E D STATES 1930-1960

OCCUPATIONAL MOBILITY IN THE UNITED STATES 1930-1960 A. J. R. 0.

JAFFE CARLETON

With a Preface by Seymour L.

KING'S CROWN Columbia University,

Wolfbein

PRESS

New York, 1954

Copyright 195^ fcy COLUMBIA UKIYEFSITT PRESS

King's Crcvn Prese i s an imprint established by Columbia University Press f o r the purpose of making c e r t a i n scholarly material a v a i l able a t minimum c o s t . Toward t h a t end, the publishers have used standardized formats Incorporating every reasonable economy that does not i n t e r f e r e with l e g i b i l i t y . The author has assumed complete responsibility for editorial s t y l e and f o r proofreading.

Library of Congress Catalog Card Number: 5^-11813

Published in Great B r i t a i n , Canada, I n d i a , and Pakistan by Geoffrey Cumberlege, Oxford University Press London, Toronto, Bombay, and Karachi MANUFACTURED IN THE UNITED STATES OF AMERICA

PREFACE Considering the significant needs for information on patterns of working life - for the working population as a whole as well as for individual occupations - surprisingly little research has been carried on in this field. To a great extent our thinking on this subject has consisted of an uneasy composite of two contradictory legends or stereotypes: first, the assumption (all too common in vocational guidance) that people choose an occupation, train for it, enter it, progress in it and in due time retire from it; second, the Horatio Alger tradition - from shoeshine boy to bank president in regulation time. The realities behind these legends are relevant to a number of problem areas in addition tc the general light they may shed on the sociology of work. In the appraisal of manpower requirements and supply for a mobilization economy it is essential to be able to assess the flexibility and occupational mobility of workers. In estimating the long-term enpliyment outlcok as a basis for determining the size of vocational training or educational programs, data are necessary on the prospective losses to each occupation resulting from retirement or the movement of workers tc other occupations. In reviewing the content and goals of educational programs and providing a context for vocational guidance, an understanding of the patterns of working life and the kinds of occupational changes people make are extremely valuable. Dr. Jaffe's and l£r. Carleton's contribution to the recently growing literature in this field is especially significant because it is a pioneer attempt to study mobility and patterns of working life by applying cohort analysis to the data from successive decennial censuses of population, fthat they have accomplished on the study of major occupation groups points a clear way for further analysis of individual occupations. Finally not the least of the contributions of this study is the ingenious and painstaking work required in adjusting the Census data to make them comparable frcm one census to another for purposes of analysis. Everyone interested in the varied aspects of working life patterns owes Dr. Jaffe and lir. Carleton a debt of gratitude for this volume.

Seymour L. Wolfbein

FOREWORD This project was undertaken with a very specific goal in mind, namely, the construction of models and procedures for estimating possible future manpower supply by occupation in the United States. 1 These were the terms of reference laid down in the original study outline and adhered to throughout the stucty. During the process of actually building these models and applying them to data, considerable additional information was obtained on certain aspects of mobility directly associated with changes in occupational distribution. It is believed that these data are significant enough in their own right to warrant separate presentation, and accordingly this is being done. However, in light of the purpose of this project primary emphasis in our report is being placed upon the models and the data relating to the projection of manpower supply by occupation} such projections were calculated for I960 for the total male population of the United States and are shown in Part I. Secondarily, we are presenting such additional information and derivative analyses as were obtained bearing on: (a) the rate of new entries into major occupation groups, (b) retirements by occupations, (c) net mobility among the various occupations, and (d) patterns of working life. These analyses are presented in Part II. Cur methodology is described in Part III. that:

Furthermore, in view of the main purpose of the project it should be emphasized

1. We are concerned with occupational mobility exclusively; all other types of labor mobility receive secondary, if any, consideration in this report; 2.

Projection rather than historical analysis was the dominant consideration.

As will be developed in more detail subsequently, any changes in the basic purpose of the study would have resulted in changes in the models and procedures which would have led ultimately to a study very different from that presented here. Occupational mobility is but one aspect of labor mobility. Other aspects include mobility among industries, mobility among employers, geographic mobility, change between status of employed and unemployed, etc. Perhaps the most comprehensive definition of labor mobility would involve the sum totality of all factors which describe how a person earns his livelihood throughout his lifetime.2 Obviously the occupation which he enters or his movements from one occupation to another or the occupation from which he retires are but a part of the story. Other parts have been studied and described in various other publications to which the reader is referred. Ctte such volume is Herbert S. Parnes, "Research on Labor Mobility: An Appraisal of Research Findings in the United States," Social Science Research Council Bulletin 65, 195Uj another is Gladys L. Palmer, Labor Mobility in Six Cities, Social Science Research Council, 1951*; still another is Mobility and Economic Opportunity, Technology Press and John Wiley and Sons, 195U. Our study on occupational mobility and the above-mentioned studies supplement each other to a considerable extent. We have gone into certain aspects of occupational 1. "This research was supported in whole or in part by The United States Air Force under Contract No. AF 18(600)-65U, monitored by Deputy Chief of Staff, Comptroller, Headquarters, USAF, AFAPA-3E, Washington 25, D. C.n 2. See Gladys Palmer, Labor Mobility in Six Cities, Social Science Research Council, New York, 195k, Chapter 1, "Why Study Mobility." See also Herbert S. Fames, "Research on Labor Mobility," Social Science Research Council Bulletin 65, New York, 195Ü, Chapter 2, "Some Conceptual and Methodological Problems."

vili

Foreword

mobility to a considerably greater extent than have these other studies. On the other hand, these other studies have investigated other aspects of labor mobility - aspects which we ignored almost entirely in our study. Since the main purpose of this study was to project the occupational composition of the male civilian adult population of the United States, all of the models ana the calculations were designed and carried through for projection purp°ses. This means in turn that they are less useful for describing the p^ely historical picture. If the aim of cur study had been tc ascertain historically what happened in the United States and without making any attempt to project into the future, we would have built somewhat different models and would have carried out different types cf calculations. As will be seen in the discussion of the methodology, Part III, we adopted these procedures which were most pertinent for providing an estimate of the future occupational distribution. Subsequently in attempting to analyze the derived information for substantive findings of a historical nature, we must confess that such history has not always been clearly demarked. In various situations, perhaps mcst obviously in the case of new entries, a better historical picture would have resulted from the use of different procedures. As ment.1 m ^ d in the methodological discussion, however, such other procedures as might have been used would not h;;ve permitted making projections of the new entries in the decade 1950 to I960. We also were not directly concerned with an analysis of the factors in the society - social, economic, demographic, cultural, technological and other factors - which influence the changes in occupational composition. Considerable discussion of the historical roles of many of these factors will be found in the volume Manpower Resources and Utilization by A. J. Jaffe and Charles D. Stewart. The present study makes some use of the information available about the changes in the social scene which have accompanied the observed occupational changes. We have not, however, tried to bring in new evidence on these interrelationships nor to investigate them in any great detail. In order to obtain more information on life patterns of occupational mobility than was possible from the analysis of census data alone, additional materials from the Six City Study were utilized. A special tabulation of the original cards was designed, and this tabulation was made by the United States Census Bureau. On the basis of these data we calculated working life patterns, showing the occupations which youtn entered at the beginning of their working careers cross-classified by the occupations they held at the end of their careers. The results show the extent of movement up and down the occupational ladder. These life patterns together with related materials are presented in Chapter 7. The authors wish to thank the following mec.bers of the Committee which helped set the framework for this study: David Kaplan, Margaret Martin, Gladys Palmer, Robert Steffes, Col. Watson and Seymour L. Wolfbein. We also wish to express our appreciation to the United States Census Bureau for guidance on the problems of reclassifying census data and for making available tc us advance tabulations from the 1950 Census. To Miss Edith Donald we express our thanks for editing and typing the manuscript. Although we are happy to acknowledge the advice and aid received from friends and colleagues - including those whose names we may have inadvertently omitted - we alone assume full responsibility for the quality and quantity of this investigation. Bureau of Applied Social Research Columbia University June, 19Sh

A. J. Jaffe R. C. Carl e ten

TABLE OF CONTENTS Part I.

Problem and Summary of Findings

1»

Introduction Aim of Project Development of models Application of models General Approach Cohort analysis Component analysis Net changes Use of total working force Basic data utilized Basic Assumptions Used in Projecting Continuity of the occupational structure Economic and labor market conditions

2.

The Projections for I960 Introduction Projected Numbers of Workers Projected Occupational Distribution Variations attributable to economic conditions Variations attributable to the size of the military forces Changes in Age Composition Part II.

1 1 1 2 2 3 3 5 5 5 6 6 7 9 9 10 12 13 1U 15

The Components of Change

3.

Summary of Components of Change The Entire Working Force Changes in the decade 1930-1*0 Changes in the decade 19U0-50 Comparison of the changes in the two decades Possible changes in the decade 1950-60 Components of Change Among the Various Occupations Accessions to occupations Withdrawals from occupations Components of change and occupational growth Summary

19 19 19 19 20 21 22 23 26 26 27

U.

New Entries RateB of New Entries Into the Civilian Working Force New Entries by Occupation Occupational Distribution of New Entries in the 1930s as Compared With the 19l*0s Factors Associated With the Distribution of New Entries

28 28 30 32 33

5.

Net Mobility Definition of Mobility Rates Total Mobility Rates Mobility Rates by Occupation Comparison of the Two Decades Factors Associated With Mobility Levels

35 35 36 38 39 39

6.

Retirements Rates of Retirement Fran the Civilian Working Force Retirement Rates by Occupation The Retirement of Farmers ^ Factors Associated With Occupational Differences in Retiraient Rates

^2

1+2

M 1+6

z 7«

Table of Contents P a t t e r n s of Working L i f e Introduction Average Length of Working L i f e The Occupational Ladder S e l e c t e d Working L i f e H i s t o r i e s P r o f e s s i o n a l , t e c h n i c a l and kindred workers Managers, o f f i c i a l s and p r o p r i e t o r s , e x c e p t farm C l e r i c a l and s a l e s workers Summary o f the white c o l l a r occupations Craftsmen, foremen and kindred workers Operatives and kindred workers S e r v i c e workers L a b o r e r s , e x c e p t farm and mine Sunmary of the manual occupations Summary of a l l occupations Occupational M o b i l i t y P r i o r t o 1930 Summary Part H I .

1+8 1+9 50 53 51+ 51+ 5U 55 55 55 56 56 56 56 57 58

Methodology

8.

Introduction Resume of t h e Methodological Problems Cbjectives D e s c r i p t i o n of u n i v e r s e studied Major methodological problems R e s u l t i n g By-Product S t a t i s t i c s Primary Sources of Data

60 60 60 60 60 62 62

9.

Comparability of Data Occupational Comparability Comparability With Respect t o Coverage Comparability of Age Reporting

6k 61+ 65 67

1 0 . Census S u r v i v a l Rates Reasons f o r Using Census S u r v i v a l Rates S u r v i v a l D i f f e r e n t i a l s by Occupation D e t a i l e d Explanation of Census S u r v i v a l Rate Construction The g e n e r a l computational procedures Computing urban and r u r a l s u r v i v a l r a t e s f o r the 1930s

69 69 69 70 7c 70

1 1 . New E n t r i e s New E n t r i e s as of the Census Year Procedures Underlying assumptions Age l i m i t s f o r new e n t r i e s New E n t r i e s During a Decade General procedures Applied procedures Considerations underlying these procedures

7U 71+ 7I1 71+ 76 77 77 77 ¿0

12.

82 82 82

Retirements Retirements as of the Census Year Retirements During a Decade

Table of Contents

xi

13. Net Mobility Computing Net Mobility for Cohorts Under Age 25 Computing Net Mobility for Cohorts Age 25 and Over Operational Definition of Net Mobility

85 85 89 89

Hi. Occupational Change Patterns Summarized

91

15. The I960 Projections Rates of Change for Each of the Components Deaths New entries Retirements Net mobility The Projection Calculations Deaths New entries Retirements Net mobility Age Group Projections Revised Cohort age 15 to 19 in I960 Cohort age 20 to 2h in I960 Cohort age 25 to 29 in I960 Cohort age 30 to 3h in I960 Cohort age 35 to 39 in I960 Cohort age UO to Wi in I960 Cohort age U5 and over in I960

91* 9U 9U 96 96 99 99 99 99 99 102 102 103 103 103 103 103 103

16. Comparison of the Six City Study With This Stucty Universe Coverage Occupational distribution Method of observation

10U 10U 10U 10U 101*

TABLES Table 1.

Male Working Force (Aged 15-7U Tears) By Occupation, 1930, 19Ü0, 19$0 and I960

11

Table 2.

Percentage Distribution of Male Working Force By Occupation, 1930, 19i*0 1950 and I960

11

Table 3.

Index Numbers of Changes in Size of Male Working Force By Occupation,

1930, 19hO, 1950 and i960

16

Table h.

Median Age of Male Working Force, By Occupation, 1950 and I960

18

Table 5.

the Male Working Force19U0-50 (Aged and 1950-60 Tears as of Components of Change in the Beginning of the Decade) 1930-^0,

18

Table 6.

Components of Change in Male Working Force (Aged h-7h Tears as of Beginning of Decade) By Occupation, 1930-1*0, 191*0-50 and 1950-60

21*

Table 7 .

Rates of New Entries Into Male Working Force (Age3 li* to 21*), 1950

29

Table 8.

Percentage Distribution of New By Entries IntoOccupation, Male Working Force and (Age19U0-50 as of Beginning of Decade) Age and 1930-1*0

29

Table 9 .

Total Mobility Rates For All Occupation Groups Conti ined (Age as of Beginning of Decade) 1930-1*0 and 19Ü0-50

37

Table 10, Retirement Rates From Male Working Force By Age, 1930, 19U0 and 19 >j Table 11. Standardized Retirement Rates of Male Working Force By Occupation 1930—1*0 and 19U0-50

1*3

Table 12. Occupational Hierarchy of Male Working Force, 1930 and I960

52

Table 13. Men Classified Career 3y Major Occupation Group at Beginning and End of Working Table ll*. Census Survival Rates (Probability of Five-Tear Age Cohort Surviving From One Census Date to the Next) By Urban and Rural for 1930 to 19U0

52

Table 15. Computation of Rural-Urban Life Table Survival D i f f e r e n t i a l 1930 to 191*0

71

Table 16. Computation of Urban and Rural Census Survival Rate3, 1930 to 19i*0

73

Table 17. Computation of New Entries Into Male Working Force By Occupation, 1939, Cohort Age 18 in 1939

75

Table 18. Making Adjustments For New Entries Into Male Working Force By Occupation, 1^39, Cohort Age 23 in 1939

75

Table 19. Computation of Unsurvived New Entries Into Male Working Force By Occupation, 19UO-50, Cohort Age 7 in 19h0

78

Table 20. Computation of Retirements From Male Working Force By Occupation, 1939, Cohort Age 61* in 1939

83

Table 21. Computation of Net Mobility of Male Working Force By Occupation, 1930^0 Cohort Age 15 to 19 in 1930

86

71

TABI£S

xiv

Table 22. Computation of Net Mobility of Hale Working Force By Occupation, 191*0-50 Cohort Age 30 to 3h in 191*0

86

Table 23. Computation of Net Mobility of Male Working Force By Occupation, 191*0-50 Cohort Age 55 to 59 in 191*0

88

Table 21*. Components of Occupational Change in Male Working Force, 191*0-50, Cohort Age 15 to 19 in 19 1*0

92

Table 25. Components of Occupational Change in Male Working Force, 1930-1*0, Craftsmen, Foremen and Kindred Workers

92

Table 26. Census Survival Rates, 1950-60

95

Table 27. Computation of New Entry Rates Into Male Working Force, 1910-50, C.ohort Age 15 to 19 in 191*0

95

Table 28. Computation of Retirement Rates From Male Working Force, 191*0-50, Cohort Age 50 to 51* in 191*0

97

Table 29. Computation of Net Mobility Rates For Male Working Force, 191*0-50, Cciiort Age 15 to 19 in 191*0

97

Table 30. I960 Projection Computations (Based on Experience of 19l*0s) For Male Working Force, Cohort Age 15 to 19 in i960

100

Table 31. Male Working Force Participation Rates, By Age, 1930, 191*0, 1950 and i960

100

Appendix Table 1. I960 Male Working Force, By Age and Occupation, 1930, 19U0, 1950 and

106

Appendix Table 2. New Entry Rates Into Male Working Force (Age as of Beginning of Decade) By Occupation and Age, 1930-1*0 and 191*0-50

112

Appendix Table 3. Retirement Rates From Male Working Force (Age as of Beginning of Decade) By Occupation and Age, 1930-1*0 and 191*0-50

112

CHARTS Chart 1 .

Occupational Mobility Rates (All Occupations Combined), by Age, 1930-1*0 and 191*0-50

Chart 2.

Occupational Mobility Rates for Male Working Force (Age as of Beginning of Decade), by Occupation and Age, 1930-1*0 and 191*0-50

^

PART I PR1BLEM AND SUMMARY OF FINDINGS CHAPTER 1 INTRODUCTION Aim of Project From time to time research efforts have been directed at attempts to estimate manpower requirements in the United States under hypothesized conditions. Simultaneously it has been recognized that such possible requirements cannot be understood or evaluated without reference to possible supply. Furthermore it is recognized that neither requirements nor supply are uniquely determinable for the simple reason that under varying conditions both can be increased or decreased. For example, even the total United States population does not set an ultimate ceiling on manpower supply since under certain conditions (theoretically, if not actually, conceivable) the United States can draw upon the manpower resources of other parts of the Western Hemisphere. This range of variation in possible size of manpower supply has long been recognized by students of the subject. Accordingly efforts have been directed at trying tc estimate the manpower which might be required and which might be available under certain specified conditions. When specifying the conditions, it then becomes possible to speak in much more definite terms than would be the case if we simply tried to speak of manpower in some broad generalized manner. This particular study is concerned with the manpower supply situation in the United States. It is limited to males and covers the decades 1930 to I960. Hie basic question considered here is» on the basis of our knowledge of changes in male manpower supply in the decades 1930 to 1950, how might we go about extrapolating this past history into the future - say a decade hence into I960. The answer to this question as it is developed in this monograph involves specific assumptions regarding a whole host of conditions which might be encountered during the 1950s. The most basic assumption necessary, which must be emphasized at the outset, is that conditions in the 1950s will be substantially "normal" in the sense that there will be no repetition of World War II in the years ahead to I960. We are extrapolating "normal" past history into a "normal" future in an effort to estimate the characteristics of the "normal" supply of civilian male workers which migfrt be available. If highly "abnormal" conditions are to be posited, such as those involved in large-scale warfare, then the characteristics of this labor supply may be altered beyond recognition. In such an event the extrapolation of past history becomes almost a game of chance. Development of models. "Model cohort work experience tables" were designed for the major occupational groups as presented in the United States decennial censuses of population for the periods 1930 to 1950. The primary purpose of -these models was to devise a framework which would permit uncovering: (l) the net occupational mobility which actually occurred during these twenty years; (2) the occupational distribution of new entrants into the labor force during the 1930s and during the 19U0sj and (3) the occupational distribution of retirements fl*om the labor force in each of these decades. The actual models were so constructed that they could be applied to any number of occupations and any type of occupational classification scheme which might be available or desired, these models were built to be applied to males only, mainly because of methodological considerations. It is believed that similar models cculd be developed for women also but that of necessity they would have to be much more complex than the ones for men.

2

Introduction

In developing t h e models, t h e major emphasis was placed on p r a c t i c a l i t y * The i n t e n t was t o develop models which could be a p p l i e d meaningfully and p r a c t i c a l l y and which a t the same time would be l o g i c a l l y and t e c h n i c a l l y sound. No attempt was made t o t r y t o develop h i g h l y complex models r e q u i r i n g types of data which a r e a t p r e s e n t u n a v a i l a b l e and which a r e l i k e l y t o be u n a v a i l a b l e i n the immediate f u t u r e . In s h o r t a l l t h e emphasis was placed on b u i l d i n g models which could be put i n t o use a t p r e s e n t . A p p l i c a t i o n of t h e models. As p a r t of t h e process of developing t h e s e models and i n s u r i n g t h e i r p r a c t i c a l i t y , we t e s t e d them with d a t a c o n t i n u o u s l y and modified them as n e c e s s a r y i n view of t h e l i m i t a t i o n s of t h e a v a i l a b l e d a t a . Consequently t h e r e s u l t s p r e s e n t e d i n t h i s volume a r e t h e j o i n t product of t h e o r e t i c a l model-building and a p p l i c a t i o n t o e x i s t i n g d a t a . In developing and t e s t i n g t h e s e models, they were a p p l i e d t o t h e United S t a t e s p o p u l a t i o n census d a t a on major o c c u p a t i o n a l d i s t r i b u t i o n ( i . e . , ten major groups) f o r males, f o r t h e censuses of 1930, 19U0 and 1950. We then made p r o j e c t i o n s of t h e t e n major o c c u p a t i o n a l groups t o I960, f o r t h e males. These p r o j e c t i o n s involve e s t i m a t e s of t h e numbers and o c c u p a t i o n a l d i s t r i b u t i o n of new e n t r i e s i n t o t h e l a b o r f o r c e i n the decade 1950 t o I960, e s t i m a t e s of n e t o c c u p a t i o n a l m o b i l i t y , e s t i m a t e s of t h e numb e r s of r e t i r e m e n t s from t h e l a b o r f o r c e by occupation during t h i s decade, and e s t i mates of t h e amount of m o r t a l i t y which might be expected. A l t o g e t h e r f o u r s e t s of e s t i m a t e s were prepared. Methodologically and from the viewpoint of the b a s i c models t h e s e f c u r a r e a l l i d e n t i c a l . The d i f f e r e n c e s among them, as w i l l be d e s c r i b e d s h o r t l y , a r e based on varying assumptions r e g a r d i n g economic c o n d i t i o n s during t h e 1950s and assumptions r e g a r d i n g t h e s i z e of t h e m i l i t a r y f o r c e in

I960.

Although t h e s e p r o j e c t i o n s t o I960 were made p r i m a r i l y i n order t o t e s t t h e f e a s i b i l i t y of the t h e o r e t i c a l models, i t i s b e l i e v e d t h a t t h e p r o j e c t i o n s i n thems e l v e s w i l l have some p r a c t i c a l a p p l i c a t i o n f o r c e r t a i n types of manpower a n a l y s e s . Furthermore i t i s hoped t h a t the r e s u l t s of t h i s study w i l l demonstrate t h e advantages t o be gained by extending t h e a p p l i c a t i o n of the models t o more u s e f u l and a t t h e same time more d i f f i c u l t d a t a . P r o j e c t i o n with r e f e r e n c e only t o the major o c c u p a t i o n a l c a t e g o r i e s i s of somewhat l i m i t e d u t i l i t y both because of t h e heterogeneous c h a r a c t e r of t h e s e broad groupings and the need f o r i n f o r m a t i o n about h i g h l y s p e c i f i c occupat i o n s . However, while t h e models now appear t o be a p p r o p r i a t e f o r making p r o j e c t i o n s f o r more d e t a i l e d o c c u p a t i o n a l c l a s s i f i c a t i o n s , t h i s was not apparent a t t h e o u t s e t . To have launched a l a r g e p r o j e c t i n i t i a l l y , i n v o l v i n g hundreds of occupations, was thought n o t t o be a d v i s a b l e u n t i l the f e a s i b i l i t y of t h e models was e s t a b l i s h e d f i r s t of a l l on a small s c a l e . This purpose was adequately served by t h e use of t h e c l a s s i f i c a t i o n of t e n major occupational groups. General Approach B a s i c a l l y the methodology c o n s i s t s of a combination of cohort and component a n a l y s e s . In t h e f o l l o w i n g paragraphs t h e s e procedures are b e i n g p r e s e n t e d as s e p a r a t e forms of a n a l y s i s . A c t u a l l y as w i l l be seen when t h e methodology i s s t u d i e d i n d e t a i l , t h e s e a r e n o t two t o t a l l y s e p a r a t e approaches; r a t h e r they a r e two avenues f o r viewing the process of change over time - i n t h i s case, occupational change. The approaches a r e i n t r i n s i c a l l y r e l a t e d one t o a n o t h e r , and i n t h e i r use they overlap and ccmplement each o t h e r . For the sake of a b e t t e r comprehension of the models and p r o j e c t i o n s , howe v e r , and i n order t o make c l e a r whatever assumptions and hypotheses a r e involved, we s h a l l examine these two b a s i c frameworks s e p a r a t e l y .

Introduction

3

Cohort analysis« One of our basic propositions is that there is an underlying continuity within a group of individuals which carries over time; this continuity applies to a number of aspects of their lives, but here we are concerned only with their occupational composition. The occupational pattern of a specific group of men - a cohort - at any one moment of time is, in large measure, a function of their past occupational experiences. Hence their future occupational experiences to a considerable extent will be a function of their historical past and present occupational experie n c e s . F o r example, if we know the occupational distribution of a group of men, say aged 35 to 39 years in 1950, and their occupations in 19U0 when they were 25 to 29 years old, and in 1930 when they were 15 to 19 years of age, we know one of the components needed for estimating their occupational distribution in I960 when they will be US to Ii9 years of age. In short it is held that after tracing through the occupational history of a cohort of men, we then have a basis for projecting this history into the not too distant future with a minimum of error. Each age cohort has its own historical pattern of occupational change which will influence its I960 occupational distribution. The occupational composition of men aged 55 to 59 years in I960, for example, will be different from that of men aged US to U9 in I960, not only because of the differences in age, but also because the two cohorts have had different occupational histories. These differences in occupational history can be traced back to the period in which they first entered the working force. Men aged 55 to 59 years in i960 for the most part entered the working force in the period around World War I. The cohort ten years younger in i960 entered the working force during the boom of the later 1920s and the early part of the depression of the 1930s. Having entered at different periods, they were confronted by varying types of job opportunities and thus entered various occupations. Once having entered the working force, their subsequent careers were variously affected by prosperity and depression, by peace and war. Each of these influences affected their occupational distribution and added to the history which helped mold their subsequent occupational experiences. Each age cohort began its working life under somewhat different conditions from every other age cohort and was differently affected by succeeding historical events. No two cohorts had identical working force experiences in the past, and therefore we assume that their future working force experiences will also be different. Hence our first line of investigation involves that of tracing through specific age cohorts historically prior to projecting these specific age cohorts to i960. Component analysis. Projection accuracy can also be improved by separate analysis and projection of some or all of the important components of past change. Since component analysis is so important to our model, let us review what it consists of and wherein lies its value. Cbviously it is simple to project or extrapolate totals or percentage distributions without reference to the separate components which contribute to any observed changes over time. For example, between 1930 and 19U0 the numbers of nonagricultural laborers decreased by 1.1$. One assumption we could make is that if the 1950s recapitulated the economic picture of the 1930s the number of such laborers might decrease again by l.U£. Component analysis, however, provides an estimated increase of 13.k%* This reversal in direction of change results from the fact of population growth during the 1950s; since there will be more men - particularly younger men very many of whom enter the working force as unskilled workers - the number of unskilled workers can be expected to increase. 1. See also below, Chapter 7> "Patterns of Working Life."

Introduction

u

We might t r y t o p r o j e c t t h e percentage d i s t r i b u t i o n r a t h e r than the percentage change. In 1930, 13.3 percent of t h e men i n t h e working f o r c e were n o n a g r i c u l t u r a l l a b o r e r s ; i n 19i «H à a, a) H 03 A H

G 0} X) e c 0) 3 u o B «M G «0 TJ a -3 CM UCO 1 -t & O s s uco &I rH O rH O f 1 •P LA >c aH >l o H Os O W rH O OH s uco r i s)a) se o H ÏM o o

CO

TD 03 r—I i-l O ^

•P a) >-i l ^• V A s O•

Os • •

CO • •Cm •s O • N v• O •

4

N O J

N H

H O

C N f t ^

1

8.

^

T3 -S ti O &

•P c 0) o u

& Os

a

^•

£

£

fc.
i A I >H

Os • • ("AsO

-í CM IÖ

p U a) a)

s O• c o• rH VA

o - J Os

Os

m

s O• V A• sO O rH

VTNsO ^O

CM • V A• r r \ 1_A

cm r — s o

O 7 ä - P m \ 0) r H >H O rA 00 Os & a>

fc. o o s ss 55

fc O

CO C U o

w

8.

55

CQ

c 55

B

oO

a w o

a>

N N r l f - J O O O H H H W r I

S

f"

O

Ph

s

|0O 3

o•000 more "managers, proprietors and o f f i c i a l s , " f o r example, according to the CPS estimate - a d i f f e r e n c e almost three times as l a r g e as the sampling v a r i a b i l i t y . Therefore i t was f e l t t h a t an a n a l y s i s of occupat i o n a l changes shown by t h e CPS f i g u r e s between A p r i l , 1950, and A p r i l , 1953, would be of l i t t l e help i n estimating changes subsequent to 1950 on a b a s i s comparable t o the 1950 census data. Comparability o f Age Reporting Census age data are i n general s u b j e c t t o two d i f f e r e n t kinds of e r r o r s ! diff e r e n t i a l completeness of enumeration by age and misreporting of age by those enumerated. A b r i e f discussion of t h e s e kinds of e r r o r s in the 19U0 and 1950 censuses nay be found in 19it0 Census of Population, Vol» IV f C h a r a c t e r i s t i c s by Age, Part 1, page 3, and 1950 Census of Population, P-Cl, page X I I I . Their relevance f o r the a p p l i c a t i o n of cur model was f e l t in two d i f f e r e n t ways: 1 . Accurate f i v e - y e a r age cohort a n a l y s i s i s presented i f persons a l i v e a t two successive censuses are e i t h e r not enumerated a t both censuses or are not reported in t h e same cohort ( i . e . , the same age group aged by t e n years) a t each census. In 191*0 and 1950, p o s s i b l y because of the Old Age and Survivors Insurance l e g i s l a t i o n , there appeared a d e f i c i t of persons r e l a t i v e to what might be expected in the age range 55 t o 62*; and correspondingly, the numbers reported as 65 and over were unexpectedly l a r g e . Incomparab i l i t i e s occur i n other cohorts a l s o . For i n s t a n c e , "males between the ages of 18 and 3h a l s o appear to have been r e l a t i v e l y underenumerated" in 1950 ( P - C l , page X I I I ) . Age cohort incomparability, as w i l l be brought out «ore c l e a r l y in subsequent chapters, n e c e s s i t a t e d adjustments in our techniques f o r estimating the components of occupational change from one decade t o the n e x t . 2 . S t a t i s t i c s f o r s i n g l e years of age are not smooth p a r t l y because of i r r e g u l a r i t i e s in the annual numbers of b i r t h s and p a r t l y because of a tendency towards the U. iil,U92,000 according t o Current Population Reports Labor Force, S e r i e s P-5U, No. 9U, page 11, and l*0,5l9,U62 according to page 1 - 1 0 1 of Vol. P-Bl of the 1950 United S t a t e s of Population.

68

Comparability of Data

o v e r - r e p o r t i n g of ages ending in 0, 2, 5 and 8. For t h i s reason, t h e s i n g l e year of age d a t a f o r t h e male population as r e p o r t e d by the censuses, could not be used in c a l c u l a t ing the working f o r c e p a r t i c i p a t i o n r a t e s by single years of age, which were needed f o r e s t i m a t i n g new e n t r i e s and r e t i r e m e n t s . Furthermore the working f o r c e data which provide t h e numerators f o r the p a r t i c i p a t i o n r a t e s were a v a i l a b l e only by f i v e - y e a r group®. Therefore in order t o have working f o r c e p a r t i c i p a t i o n by s i n g l e years of age, the numbers of workers in each s i n g l e y e a r of age had t o be estimated by employing m u l t i p l i e r s t o subdivide f o r f i v e - y e a r age groups. The s i n g l e year of age labor f o r c e data provided by t h e m u l t i p l i e r s were a u t o m a t i c a l l y smoothed; t h e r e f o r e they were not comparable with the i r r e g u l a r data f o r the general male population by s i n g l e years of age as provided by the censuses. Hence b e f o r e working f o r c e p a r t i c i p a t i o n r a t e s could be computed, comparability had t o be achieved by smoothing the s i n g l e year of age data f o r the population through the a p p l i c a t i o n of m u l t i p l i e r s t o the f i v e - y e a r age group data f o r the male p o p u l a t i o n .

CHAPTER 10 CENSUS SURVIVAL RATES Reasons For Using Census Survival Rates As outlined previously, one of the components to be measured was the number of deaths occurring in each occupation f o r each f i v e - y e a r age cohort. For t h i s purpose, decade survival r a t e s were necessary. Under i d e a l conditions i t would have been p r e f e r able to use i n t e r c e n s a l working force l i f e tables from itfiich decade survival r a t e s could have been computed. Aside from the f a c t that such t a b l e s were not a v a i l a b l e and t h e i r construction was beyond the scope of t h i s p r o j e c t , other considerations a l s o led to the decision that t h e i r use was not f e a s i b l e . F i r s t of a l l , age cohorts were not e x a c t l y comparable from census to census because of age misreporting and the r e l a t i v e underenumeration of some age groups. The estimated number of survivors obtained by application of l i f e table survival r a t e s would in the case of many age cohorts have been widely a t variance with the numbers a c t u a l l y observed a t the following census and l a r g e - s c a l e adjustments would have had to be made a r b i t r a r i l y . Net immigration, e s p e c i a l l y in the 19U0s, was a further complicating f a c t o r which would have thrown o f f our estimation of deaths i f decade survival r a t e s had been c a l c u l a t e d from l i f e t a b l e data. Between 19^0 and 19$0 net immigration was 879,000, of which i t i s estimated t h a t approximately 30u,000 were males age 16 and over. While net immigration of males of working force age was only about 0 . 5 percent of a l l males of working force age, i t i s probable that the Immigrants were concentrated in the younger ages rather than evenly distributed as t o age. Also l i t t l e r e l i a b l e information was a v a i l a b l e on the occupational d i s t r i b u t i o n of the immigrant working f o r c e . I f working force data had been tabulated according to n a t i v i t y , the e f f e c t of immigration could have been eliminated by confining our study to the native population. Since t h i s was not possible, the migration f a c t o r would in the case of some occupations and age cohorts have increased the discrepancy between expected and observed survivors i f l i f e table survival r a t e s had been used to estimate survivors. The decision was made therefore t o estimate i n t e r c e n s a l deaths by means of census survival r a t e s , i . e . , the numbers in the f i v e - y e a r age cohort a t the end of the decade divided by the numbers in the f i v e - y e a r age cohort ten years younger a t the b e ginning of the decade. The product of the census survival r a t e and the numbers in the working force for the corresponding age grcwp a t the beginning of the decade y i e l d s the estimated number of working force survivors a t the end of the decade. The assumption i m p l i c i t in the application of census survival r a t e s to working f o r c e data i s that those in and outside of the working force undergo the same survival experiences. This i s known not to be p r e c i s e l y the case. Nevertheless since a t most ages the great majority of men are in the working f o r c e , the survival rate for a l l men in any s p e c i f i e d cohort must be very similar to the survival r a t e f o r the men in the working force in t h a t cohort. I t should be r e a l i z e d of course t h a t deaths estimated by t h i s procedure are net deaths, i . e . , gross deaths plus or minus gains or l o s s e s due to migration, underenumeration and misreporting of age. Survival D i f f e r e n t i a l s by Occupation The application of the same survival r a t e s t o a l l occupations within a s p e c i f i e d cohort c a r r i e s with i t the f u r t h e r assumption that occupational survival d i f f e r e n t i a l s are n e g l i g i b l e . The question of using occupational d i f f e r e n t i a l s was i n v e s t i g a t e d , and i t was fcwnd t h a t the available l i t e r a t u r e was too fragmentary to be of any value. Accordingly i t was f i n a l l y decided that - considering available resources of time and labor - the most that could be done was to weight the census survival r a t e s f o r the decade 1930 to 19U0 by rural-urban d i f f e r e n t i a l s which would be applied r e s p e c t i v e l y to

70

Census Survival Rates

agricultural occupations (farmers and farm managers, and farm laborers and foremen) and to nonagricultural occupations (the remaining eight major occupation categories). How these rural-Hirban differentials, which were based on 1939 rural and urban abridged life tables for the United States,1 were calculated will be explained in detail presently. Similar differentials were not employed for the decade 19 UO to 1950 both because urban and rural life tables for that decade could net readily be had and also because the narrowing of urban and rural mortality differentials made it seem that this differential would be much less important in the 19U0s than it had been in the 1930s, In summary therefore for the decade 1930 to 19U0 separate urban and rural census survival rates were used; for the 19U0s the same census survival rates were used for all occupations. Detailed Explanation of Census Survival Rate Construction The general computational procedures. We turn now to a step-by-step explanation of the procedures involved in the construction of the census survival ratest The first very simple step was the computation of total male census survival rates (without urban-rural differentials) for 1930-U0 and 19^0-50. These rates are shown in Table lU. For illustration purposes the cohort age UO to UU at the beginning of the decade will be used, for this cohort the computations were as follows! Males 50 to 5U in 19U0 = 3,752,750 Males UO to UU in 1930 U,139,911 ~ * 9 0 6 5 Males 50 to 5k in 1950 . U,128,658 . Males U0 to UU in 19U0 U,U19,135 ~

,931i3

It will be noted that .9065 and .93U3 are the census survival rates of Table lU« Since no urban-rural differentials were used for the 19U0s, the tions for this decade are now completed. The rest of this chapter is devoted process whereby the total census survival rate for 1930-U0 was split into the urban and rural census survival rates (also shown in Table lU).

on row (8) calculato the separate

Computing urban and rural survival rates for the 1930s. In general what was done was to convert the four 1939 abridged life tables (viz., white and non-white males, urban and rural) into separate ten-year life table survival rates for five-yoar age cohorts. Then the white and non-white urban survival rates were combined to give total urban rates for each five-year age cohort. The white and non-white rural rates were similarly combined. The ratio of the rural to the urban survival rates yield a ruralurban differential which was applied to the ten-year census survival rates, due consideration being given to the proportion of males in each cohort living in rural and in urban areas. Of course it would have been preferable to use 1930-U0 rural-urban life tables - or at least, 1935 life tables - but such tables were unavailable. The 1939 life tables were used as a reasonable approximation. Let us now follow through the actual details for the cohort age U0 to UU in 1930. 1. The first three rows of Table 15 show the number of white male survivors to ages U0, U5, 50 and 55 out of 10,000 births fori (a) cities of 100,000 Dopulation and over, (b) other urban, and (c) rural. Dividing column (3) by column (l), and column (U) by column (2) for each region gives the probability at age UO of surviving to age 50, 1. "U. S. Abridged Life Tables, 1939, Urban and Rural, By Regions, Color and Sex," FSA, Vol. 23, No. 15, June 30, 19U7.

TABLE 14 CENSUS SURVIVAL RATES (PROBABILITY OF FIVE-YEAR AGE CCHORT SURVIVING FRCM CNE CENSUS DATE TO THE NEXT)

BT URBAN AND RURAL FCR 1930 TO 1940

P r o b a b i l i t y of surviving throughcmt d e c a d e

Age a t beginning of decade

Total (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)

5 to 9 years 10 to 14 years 15 t o 19 years 20 to 24 years 25 to 29 years 30 t o 34 years 35 t o 39 years 40 t o 44 years 45 to 49 years 50 t o 54 years 55 t o 59 years 60 to 64 y e a r s

.9677 .9372 .9459 .9493 .9756 .9679 .8987 .9065 .8194 .7650 .7809 .6541

1930-1940

Urban

Rural

.9671 .9372 .9463 .9494 .9739 .9637 .8911 .89 35 .8016 .7406 .7440 .6123

.9683 .9373 .9454 .9490 .9780 .9741 .9105 .9270 .8471 .8012 .8308 .7065

1940-191 .9802 .9419 .9663 .9881 1.0123

1.0000 .9538 .9343 .8624 .8095 .8051 .6793

TABLE 15 COMPUTATION OF HJRAL-URBAN LIFE TABLE SURVIVAL DIFFERENTIAL, 1930 TO 1940 White Males Age X

(1)

45 (2)

(1) Number of (2) survivors (3) out of 10,000

Cities 100,000«. Other Urban Rural

.8763 .8548 .8651

.8457 .8272 .8430

(4) P r o b a b i l i t y a t age X (5) of surviving 10 years (6) column ( i ) f coluim ( i - 2 )

C i t i e s 100,000* Other Urban Rural

.9124 .9212 .9393

.8684 .8839 .9133

(7) Number of survivors out of 10,000 row (1) • row (2) 2

Urban

.8656

.8365

(8) P r o b a b i l i t y a t age X of surviving to Age X ^ 10

Urban

.9168

.8762

Urban Rural

.8968

row ( 4 ) » row ( 5 )

2

(9) P r o b a b i l i t y a t age I to I • 5 (10) of surviving 10 y e a r s (11) Rural-urban survival d i f f e r e n t i a l row (10) f row (9)

1.0331

Non-White Males (12) Probability a t age X t o X * 4 of surviving (13) 10 years (l4) Rural-urban s u r v i v a l d i f f e r e n t i a l

Urban

.7758

Rural

.8507 1.0965

io

(3)

55 (4)

.7995 .7874 .8126

.7344 .7312 .7699

.7935

.7328

72

Census Survival Rates

and a t age U5 of s u r v i v i n g t o age 55. These p r o b a b i l i t i e s are shovm in and ( 6 ) . For example, .8126 divided by .3651, the nunber of white male s u r v i v i n g t o age 50» divided by t h e number of r u r a l r e s i d e n t s s u r v i v i n g equal t o .9393, t h e p r o b a b i l i t y a t age 1*0 of a r u r a l r e s i d e n t surviving

rows (U), (5) rural residents t o age UO, i s t o age 50.

2. In row (7) we have used a simple a r i t h m e t i c average t o combine f o r use i n s t e p U below t h e c i t i e s 100,000 and over and t h e other urban l i f e t a b l e s u r v i v o r s . It was not thought n e c e s s a r y t o use a weighted average because a t the time of the 19U0 census the t o t a l U." S. urban population was d i s t r i b u t e d approximately h a l f and h a l f among c i t i e s 100,000 and over and o t h e r u r b a n . 3 . In row (8) of Table 15 the p r o b a b i l i t y of surviving ten y e a r s f o r t o t a l urban white males a t ages U0 and U5 was computed by t h e unweighted average of the c o r responding p r o b a b i l i t i e s f o r c i t i e s 100,000 and over and o t h e r urban i n rows (ii) and ( 5 ) . For example, one h a l f of t h e sum of »9121* and .9219 i s .9168 as shown in column ( l ) of row ( 8 ) . U. Rows (6) and (8) are r e s p e c t i v e l y the white male r u r a l and urban p r o b a b i l i t i e s a t ages UO and U5 of s u r v i v i n g an a d d i t i o n a l t e n y e a r s . In row (9) we have a v e r aged ( u s i n g as weights the number of urban survivors t o age U0 and age hS in row ( 7 ) ) the urban p r o b a b i l i t i e s a t ages ijO and U5 of surviving t e n years as shown in row (8) and c a l l e d t h i s average t h e p r o b a b i l i t y of the white male urban cohort age hO t o kh surviving ten y e a r s . This i s e q u i v a l e n t t o saying t h a t the p r o b a b i l i t y of a cohort age UO t o luU of surviving t e n s u r v i v i n g ten y e a r s approximately equals the p r o b a b i l i t y a t age years. In row (10) the same procedure was applied t o rows (6) and (3) t o g e t the t e n year r u r a l s u r v i v a l r a t e of a r u r a l cohort age Uo t o UU. 5. Row (11) - n o t a c t u a l l y made use of i n subsequent computations - i s the r u r a l - u r b a n t e n - y e a r s u r v i v a l d i f f e r e n t i a l derived by d i v i d i n g row (10) by row ( 9 ) . 6. Rows (12) t o (lii) in Table 15 show the l i f e t a b l e s u r v i v a l r a t e s and r u r a l urban d i f f e r e n t i a l obtained f o r non-white males when s t e p s 1 thrcwgh 5 were a p p l i e d t o the non-white male urban and r u r a l l i f e t a b l e s . 7. For subsequent use as weights the r a t i o of white t o non-white males age UO t o UU i n urban and r u r a l a r e a s in 19U0 was c a l c u l a t e d . These r a t i o s are shown i n rows (1) and ( 2 ) of Table 16. 8. In row (3) of Table 16 we have combined (by means of a weighted average, with 9 . 7 and 1 as w e i g h t s ) the urban white and non-white ten-year l i f e t a b l e s u r v i v a l r a t e s i n t o a t o t a l urban t e n - y e a r l i f e t a b l e s u r v i v a l r a t e . In row (U) we obtained a t o t a l r u r a l t e n - y e a r l i f e t a b l e s u r v i v a l r a t e in the same manner, i . e . , a weighted a v e r age of rows (10) and (13) from Table 15, with 8.90 and 1 as weights. 9 . Row (5) of Table 16 shows the r u r a l - u r b a n t e n - y e a r l i f e t a b l e s u r v i v a l d i f f e r e n t i a l obtained by d i v i d i n g the r u r a l t e n - y e a r s u r v i v a l r a t e in row (U) by the urban s u r v i v a l r a t e in row ( 3 ) . 10. Row (6) i s the r a t i o of r u r a l males age UO t o UU in the general population i n 19U0 t o t h e urban males age UO t o UU. 11. We now have a l l the necessary i n g r e d i e n t s f o r s p l i t t i n g the t e n - y e a r census s u r v i v a l r a t e i n row ( 7 ) (from row (8) of Table lU) i n t o s e p a r a t e urban and r u r a l t e n - y e a r census s u r v i v a l r a t e s . To take t h i s f i n a l s t e p , we note f i r s t the f o l l o w i n g two requirements:

Census Survival Rates

73

a . We want the r a t i o of the rural t o the urban t e n - y e a r census s u r v i v a l r a t e s t o be equal t o 1.0375, the rural-urban t e n - y e a r l i f e t a b l e s u r v i v a l d i f f e r e n t i a l f o r a cohort age h0 to Uli i n 1930. b . We want the t o t a l t e n - y e a r census s u r v i v a l r a t e t o be equal t o the average of the r u r a l and urban ten-year census s u r v i v a l r a t e s , each weightad according t o i t s r e l a t i v e s i z e in the t o t a l male population age iiO t o Ui, i . e . , .629 and 1 . 0 . These two requirements can be expressed by the f o l l o w i n g two e q u a t i o n s , the row numbers of Table 16 as symbols:

using

1 6 ) | 9) » 1 . 0 ( 8 ) s ( 7 ) U ) ¿) • 1 . 0 where: (8) i s the unknown urban census s u r v i v a l r a t e , (9) i s the unknown r u r a l census s u r v i v a l r a t e , and a l l the other terms are as l i s t e d i n Table 16. S o l v i n g the s e t of equations f o r ( 8 ) , the urban census s u r v i v a l r a t e , we g e t (8) = ( 7 ) / T 6 ) » 1 . 0 7 K } (5)(6J * 1.(5 The r u r a l census s u r v i v a l r a t e f o l l o w s immediately from ( 9 ) -