Salary Income Research Articles

Quantitative Determinants of Income with Some Implications to Insurance

By use of 1960 census data on the characteristics of the United States population, this study attempts to establish the significant variables which may be used to determine income, and to calculate the extent of the relationship of these variables to the level of income. The methods used involve standard factor and correlation analysis. After the have been isolated and variance quantified, an attempt is made to construct an econometric model of income. Application of the results to various fields of endeavor related to Risk and Insurance are suggested. Quantitative of are those variables which have a measurable relationship to the level of income. Alfred E. Hofflander, Jr., Ph.D., C.L.U., C.P.C.U., is Assistant Professor of Insurance in the University of California at Los Angeles. His earlier teaching experience included service at Florida State University and the University of Texas. He is Assistant Editor of this Journal, is Associate Editor of the CPCU Annals, and was formerly Editor of the North Florida Business Review. Dr. Hofflander was a Fellow in the Huebner Foundation for Insurance Education. This paper was presented at the 1966 Risk Theory Seminar and was sponsored in part by the UCLA Division of Research and The University of Texas, College of Business Administration Research Foundation. In addition, the author wishes to acknowledge the support of the Cooperative League of the United States of America, including the following companies: Desjardins Mutual Life Insurance Company, Mutual Service Insurance Company, Co-operators Insurance Association, Co-operative Insurance Services, Ltd., La Societe d'Assurance des Caisses Populaires, Nationwide Insurance Company and League Life Insurance Company. Certain data used in this paper were derived by the author from a tape file furnished under a joint project sponsored by the United States Bureau of the Census and the Population Council and containing selected 1960 census information for a 0.1 per cent sample of the population of the United States. Neither the Census Bureau nor the Population Council assumes any responsibility for the validity of any figures or interpretations of the figures published herein based on this material. In words, the extent of the relationship is quantitative. No implication is involved conceming causality which might exist between and the various determinants. The word determinants is used in a descriptive sense, in that these factors do not necessarily cause to be what it is, but that the analyst can, within a certain range, determine a hypothetical level of by knowing the significant variables and their relationships. The fact that variables are quantitative presupposes the ability of the analyst to perceive changes in the variables and to measure the related variance. The fact that variables are determinants means that they may be used to estimate levels of income. The term income refers to wage and salary income, self-employment income, and/or other received over the period of a year.' The size of an individual's is the result of the interaction of many factors, of which some are measurable and others are, for practical purposes, non-measurable. Age, education, occupation, number of weeks worked per year, geographic loI Miller, Herman P., Income of the American People. Social Science Research Council. (New York: John Wiley & Sons, Inc., 1955), p. 13.

A statistical analysis of the distribution of wage and salary income in New Zealand

A statistical analysis of the distribution of wage and salary income in New Zealand

An Application of the Durand Method for Estimating the Size Distribution of A Given Aggregate Income

SERIES of articles have appeared in this L REVIEW on methods of adjusting a given size distribution of income to differing aggregate or per capita income levels.2 If the distribution of income remains reasonably constant and price levels keep changing, a method of adjustment for changes in per capita income will no doubt be a useful statistical tool. I do not wish to enter into the controversy between Dr. Durand and Messrs. Clark and Fishman; the simpler Durand method was used in this particular application because over thirty transformations were originally made and because the usual class intervals of income were maintained. In trying to find a satisfactory income distribution for Metropolitan Boston, a number of income curves for different years and for different social groups were adjusted to I 94I income levels. These families of curves are interesting for themselves, apart from the Boston data, and they suggest the possibility of finding such pattern curves for estimating postwar distributions of income for various social groups. The original purpose of the study was to find an income distribution for Metropolitan Boston. Since a sample study could not be made, available statistics had to be utilized for what they were worth. The Census of I940 contained distributions of income, based on wage and salary income alone, for the year I939 for Metropolitan Boston, for Massachusetts, and for New England; also, the Census provided data on individual workers for Boston alone. Income tax returns for Massachusetts for I942 were also available. Questions immediately arose concerning the usefulness of the fragmentary data relevant to Boston. Do distributions of wage and salary income alone resemble distributions of total income? Apart from their shape, would such distributions be too high or too low? If there were differences in shape or level, could these distributions still be salvaged by comparing similar Census distributions for the nation with other recognized national distributions? How do the Boston or national distributions for individual wage earners compare with desired consumer unit or family distributions of income? Clearly it was desirable to compare the fragmentary Boston data with other data on the distribution of income. Such comparisons of income distribution (retaining dollar income classes) cannot be made until allowance is made for differences in per capita income levels of different years. It was necessary, therefore, for purposes of comparison, to adjust all the relevant distributions to a base year. The year 1941 was chosen and the Durand method was then used to adjust to that base year the Consumer Purchases Study of I935-36, the various 1939 distributions from the Census of I940, the I94I and I942 income studies made by the Bureau of Labor Statistics, the I945 and I946 surveys for the Board of Governors of the Federal Reserve System, and the income tax returns for I942.3 'I am indebted to the Housing Association of Metropolitan Boston and to Mr. William Loring for assistance in carrying out the study on which this article is based. fuller description may be found in Some Economic Aspects of Urban Housing, with Special Reference to Metropolitan Boston (Ph.D. dissertation, Harvard University, I948). 2 Edward Ames, A Method for Estimating the Size Distribution of a Given Aggregate Income, this REVIEW, xxiv (November 1942), I84-89; David Durand, A Simple Method for Estimating the Size Distribution of a Given Aggregate Income, this REVIEW, xxv (November 1943), 227-30; Eugene Clark and Leo Fishman, of Methods for Estimating the Size Distribution of a Given Aggregate Income, this REVIEW, XXIX (February 1947), 43-46; David Durand, An Appraisal of the Errors Involved in Estimating the Size Distribution of a Given Aggregate Income, this REVIEW, XXX (February 1948), 63-68; Eugene Clark and Leo Fishman, Further Remarks on Estimating the Size Distribution of a Given Aggregate Income, this REVIEW, XXXII (February 1950), IO6-07. 'Differences in per capita incomes were used to transform the distributions. The statistician's judgment enters at every step; a fuller discussion of the problems and errors involved may be found in the dissertation, op. cit., ch. ii. Appendices i to v of the dissertation contain the data and transformations.

A Note on the Supply of Female Labor

THIS note 1 deals with the measurement of the supply curve of female for the economy as a whole. It is not to be interpreted as a study of the supply for particular industries or occupations. In I940 for the first time the Bureau of the Census included data on wages and salary incomes in its Census of Population. The concept of labor was substituted for Earnings data were classified by sex. These changes in the presentation of population statistics made the measurement of the supply curve of easier and less open to doubt on the grounds of inadequacy of the basic data than was the case in the work of Paul H. Douglas.2 Both in his book and in his and Mrs. Schoenberg's article, the series representing wages was based on earnings in manufacturing. The average annual earnings in manufacturing in 4I cities were correlated with the proportion of the population gainfully employed. This procedure assumed that earnings in other industries bore the same ratio to those in manufacturing in all of the cities. With the Census now reporting frequency distributions of wages classified by size for each of the major cities, the assumption is no longer necessary. Douglas and Schoenberg found a negative relationship between numbers of women and children employed and average earnings. Since it was quite conceivable that instead of women seeking employment due to low incomes, the average earnings were low because a large number of women were employed, the authors adjusted the earnings series to measure the wages an equivalent adult male would receive. The correction was based on the ratio of men's to women's wages in each state as reported in a number of industries. They also found in each state the proportion of the employed who were women. The manipulation of the series in this way is no longer necessary, for the earnings data are classified by sex. Some bias in the earnings figures are still present because of the inclusion of those of children, but the number of children employed was not large enough to reduce the accuracy of my findings significantly. In previous Censuses, persons gainfully employed were defined as having an occupation, regardless of whether they were currently employed, available for employment, or neither of these. In I940 the force was defined by the Bureau as consisting of persons at work, seeking work, or on public emergency work in the week of March 24 to 30, I939. The figures excluded workers unemployed due to seasonal layoffs, who were not seeking work. Retired persons were also excluded. Neither of these groups at the time of the Cefisus was effectively a part of the supply. Under the former classification, however, many of them would have been reported as gainfully employed, since they would have given an occupation in answer to the questionnaire. The latest Census also excluded inmates of penal and mental institutions, who previously would have been considered gainfully employed because they worked regularly. For my purpose, this is an improvement, inasmuch as it is difficult to conceive of such inmates being available for other employment.3 Included in the force are new workers, without previous experience, seeking employment. In I930 new workers would probably not have reported occupations and thus would have been omitted. Entrepreneurs were included both in the I930 and the I940 series on labor. Since most of 1 I am indebted to Professors Milton Friedman and George J. Stigler for discussions on the methods used in this paper. 2Paul H. Douglas, The Theory of Wages (New York, I934); Erika H. Schoenberg and Paul H. Douglas, Studies in the Supply Curve of Labor: The Relation in I929 between Average Earnings in American Cities and the Proportions Seeking Employment, The Journal of Political Economy, February I937. Sixteenth Census of the United States (I940), Population, Volume iII, Part i, p. 3.

Southern Children and Family Security

CHILDERN occupy a much more important place in southern families than in other regions (Table 1). In absolute numbers more than one-third of the nation's children are in southern homes. In later life migration distributes many of these children throughout the country, emphasizing the national stake in their development. Western States average only about 0.9 of a child per family as against nearly 1.5 in Southern States. The Northeast and North Central States average almost exactly 1 child per family. The average children per family is about the same for whites and Negroes within the urban and farm population, but the Negro total is higher because they are more rural. On the other hand, nonfarm family wage or salary incomes in the South ($965) are only about two-thirds of the national average ($1,380). The support of a child population one-third larger than that of other regions by family incomes which are one-third smaller raises many problems of family economics and has its effect both upon the level of living of the parents and of the children. These problems constitute a challenge not only to families but also to the planners and legislators of States and communities and students of social policy who are concerned with the maximum equalization of opportunity. The standard of living is not always maintained by the earnings of a single wage earner who provides the sole support of a family. The inadequacy of the individual earnings of many workers for the full satisfaction of the needs of their family is indicated by the number of supplementary earners in families-wives, sons, daughters, and other relatives who contribute to the common budget. The proportion of families with supplementary earners and the differential incomes of such families is shown in Table 2. The South as a whole has a higher proportion of families with more than one earner than the national proportion, and within the South 42 percent of the Negro families have more than one earner as against 31 percent of the white families. When all wage or salary workers, farm and nonfarm, are classified by relationship to the family head, the distribution is: head of family 57 percent; wives 10 percent; and other relatives of the family head 33 percent. The social consequences of the need for the earnings of members of the family other than the head are of very widesignificance. To mention only the most obvious implications: Child labor with its lack of educational opportunity results. Families with working wives usually have few children, and when wives or widows with young children do work, the home life of the children is often neglected. Adult offspring, who remain to contribute to family support, usually delay their marriages and start their own families later. The analyses presented herewith deal with family income (as of 1939) based upon the Census enumeration of 1940. Only occasionally in recent years has information been available as to the family incomes of a sufficiently large segment of the population to warrant safe generalizations. These occasions were the Consumer Purchase Study of 1935-36, The National Health Survey of 1935, and the Census enumeration of 1940.2 The latter is more comprehensive in coverage and lends itself to more accurate relation of the family income to family composition; that is, the census data can be more accurately reduced to per capita incomes for the various types of families. serious deficiency of the census figures, however, is that they include only wage or salary income and, therefore, exclude farm incomes and the incomes from business enterprises, professional fees, 'In this analysis the States included in the southern region are those included by the Census of 1940 in its regional division of the country. These analyses were computed from the following volumes of the U. S. Census of 1940: Population-Characteristics by Age; Population-Families-Wage or Salary 1939; Population-The Labor Force-Wage or Salary 1939; Population and Housing-FamiliesGeneral Characteristics; Population-Families-Types of Family; Population-Families-Sizes of Family and Age of Head. For a fuller discussion and methods of computation, see T. J. Woofter, A Method of Analysis of Family Composition and Income, Journal of the American Statistical Association, December 1944. 2 It is probable. that another cross section will be provided by the special census of 1945.

Visual Analysis—A New Method in Market Research

M ANY marketing men are familiar with the published material of the Census. Few people, however, are familiar with all of it, and fewer still know much about the enormous amount of unpublished material that can be made available by paying the cost of processing. This unpublished material ranges all the way from data already prepared in tabular form and available to interested persons for the bare cost of photostatting, to the individual reports to the Census, which, of course, are not available as such to any person outside of the Census. The individual reports can be used in special studies made by the Census. These reports while not directly available to persons outside of the Census, nevertheless, constitute a mine of information in research conducted for special purposes, provided always that no information concerning individual operators is disclosed in the process. Much of the material can be retabulated directly from cards already punched, or from samples of these cards. Special studies made for a specific purpose, in order to be of greatest value must be planned with certain definite objectives to be attained. In studies using Census data the objectives are best known to the agency desiring the study and the material and processes best known to the Census. Bringing these two sources of specialized knowledge to bear on the study has in the past constituted a serious obstacle. To overcome this difficulty a technique has been developed in the Census whereby, given a mass of material, the researcher is enabled to explore this material much more rapidly and completely than has been possible heretofore. With this technique he can do this even without a wide knowledge of the field in which the material is relevant. Thus, marketing men in bringing their problems to the Census are not compelled to learn various processes and procedures in the Census before they can obtain the results they need. The Census can make the study and furnish to the marketing man a thorough analysis of the material used. The technique used is called Visual Analysis, because of the fact that relationships are visible. These relationships are actually visible to the eye, and not dependent on the interpretation of figures. Visual analysis is essentially an exploratory technique in that figures are not usually obtained during the preliminary analysis. The results obtained, however, are fully as conclusive, and in many cases more so, than the conventional tabulations made in the statistical presentation of the findings. By means of Visual Analysis, material from several different sources may be used to encompass insofar as possible all significant factors of influence. In one special job information from two other sources was added to Census material collected in regular enumerations. The Census material consisted of information concerning the education, occupations, and salary incomes in families, the living conditions as indicated by dwellings and facilities enjoyed, and farm information

The Illinois Faculty Dollar

O NE basic use of a cost of living index is to measure changes in the cost of living relative to change in income available for living. Thus under ideal conditions increases in costs would be matched by increases in income, and if cost of living declined, income should not be curtailed by more than the reduction in costs. This would be assuming a static relationship between income and cost of living. Such a situation is academic and not likely to be achieved. Furthermore, under our economic system, which is growing constantly more productive, there is sound reason why income should normally outstrip cost of living leaving an increasing surplus for enjoyment of life over the minimum essentials for living. Otherwise there would be no economic progress. The purpose of this article is to see what the relationship has been between the cost of living and the salary income of the full-time instructional, research, and extension staff at the University of Illinois. Although these figures are restricted in their scope and applicability elsewhere, they suggest some highly illuminating circumstances which no doubt are widespread in other institutions throughout the country. It is primarily to bring out certain suggestions and implications that the University of Illinois data have been employed. Salary data for any other institution might be similarly employed. Three series of data are assembled in Table I. The first of these given in Column 2 is the index of cost of living for a faculty of state universities and colleges from I9I3 to I942 using the average of years I935 to I939 as a base of reference. This series shows the great increases in living costs from an index of 68.i in I9I3 to I35.7 in i920, the subsequent decline to 93.6 in I933 and the present resurgence to II5.3 in