views updated


I. Consumption Levels and StandardsMargaret G. Reid


II. Consumer AssetsRobert Ferber


III. Consumer BehaviorHerbert E. Krugman



Consumption is defined as the use of goods and services. Differences in consumption among individuals at a given time and between points in time are called differences in consumption levels. Consumption standards are goals to be striven for, maintained, or regained. These standards influence the ways in which families spend their income, that is, consumption patterns. Standards also can cause feelings of discontent that may result in action when consumption is below the standard— when, for example, people consider that the dwelling they occupy is overcrowded, the food they eat is inadequate, or the quality of schooling available for their children is poor. Consumption standards are part of a larger set of standards pertaining to work, leisure, social relationships, and living in general. Consumption is influenced by standards and by income, and income in turn is influenced by standards of living insofar as standards affect incentives (Davis 1945; Pipping 1953).

The goals of life are many, so that choices must be made among them. These choices are influenced by the hierarchy of preferences expressed in the standard of living and by the costs of the things desired. The availability of resources of several types acts as a constraint on total achievement. Earning additional income, obtaining more unpaid services from family members, and making wiser choices among products are all important paths to higher consumption. A consumption standard that is low relative to current money income tends to weaken the motives to earn and to economize in the use of income.

Consumption is overt behavior, and standards are among the forces shaping it. Consumption data, therefore, indirectly provide some information about standards; the strength of preferences is implied by what is chosen. Attempts have also been made to gauge more directly the intensity of desire for things not currently being consumed. The notions of “poverty,” “comfort,” and “affluence” are found to be related to earlier experiences, especially to recent changes. However, our knowledge of standards comes largely from observing behavior under varying conditions.

Differences within and among countries

About a century ago Engel formulated a very important law of consumption, namely, that as total consumption increases, the percentage of expenditures going to food tends to decrease (Stigler 1954). This percentage is referred to here as the food rate. Historically, Engel's law seems to apply even though many changes have accompanied the increase in total consumption. For example, in the United States the food rate of urban families of two or more persons whose head was a wage earner or clerical worker was 41 per cent in 1888–1891 (Williams 1958); 38 per cent in 1918–1919 (U.S. Bureau of Labor Statistics 1924); and 35 per cent in 1934–1936 (U.S. Bureau of Labor Statistics 1941). For all urban consumer units of two or more persons the food rate was 24 per cent in 1960–1961 (U.S. Bureau of Labor Statistics 1964). Changes in prices as well as in incomes may have had an effect on this rate. The Consumer Price Index shows that between 1913 and 1957 rents in large cities increased 77 per cent and food prices increased 187 per cent (U.S. Bureau of the Census 1960).

Relationships between the food rate and income similar to those indicated by the secular change in the food rate of urban families in the United States are apparent in comparisons among countries that obviously differ in real income as well as in other ways (Bennett 1951; Williams 1958). For example, in 1962 the food rate of countries ranged from 51 and 50 per cent for Ghana and Ceylon to 22 and 21 per cent for Canada and the United States, respectively (United Nations 1964). It seems probable that among countries the food rate is a suitable proxy for total consumption: the lower this rate, the higher is total consumption.

The food rate is correlated with a number of conditions closely related to total consumption. Among 24 countries investigated around 1960, the coefficient of correlation between the food rate and the infant mortality rate was approximately 0.74. At the same time among 23 countries the coefficient of correlation between the food rate and the percentage of households with 1.5 or more persons per room was 0.86. Similar relationships have been observed among states and cities of the United States (U.S. Bureau of Labor Statistics 1964; U.S. Bureau of the Census 1963; 1964).

Expenditures as measures of consumption

Expenditures may be viewed as numbers of units of consumer goods and services multiplied by their prices. The prices of products of specified quality differ from time to time and place to place, and differences in expenditures may therefore not represent differences in consumption. Many countries have developed price indexes that permit at least crude intertemporal comparison of real consumption, that is, of the quantity of consumer products measured in dollars of constant purchasing power. Differences in expenditures among places within countries may also reflect differences in prices as well as in the quantity of products purchased. Food prices probably vary with the distance between consumers and primary producers; in other words, they are a function of the regional specialization of agriculture. The price of residential housing probably increases with urban congestion. The costs of services are much influenced by the availability of job opportunities outside the service industries and hence tend to be positively correlated with local incomes. Little progress has so far been made in providing interplace indexes of consumer prices (Brady & Hurwitz 1957). If consumer prices are correlated with regional and local income, as seems probable, then the variation in expenditures among the consumers of a country overstates the variation in consumption.

Expenditure data seldom distinguish precisely between consumption expenditure and outlays made to provide income. With the growth of cities and suburbs and the increased mobility of the labor force, job expenses have increased. In the United States from 1940 to 1963 the share of transportation in total expenditure increased from 9.7 to 12.6 per cent (U.S. Bureau of the Census 1964). Some of this increase was caused by the greater distances traveled to work and thus represents an increase in the cost of production rather than an increase in the consumption of transportation service for personal or recreational purposes. There are still few estimates of job expenses, and even these are limited in their coverage (Holmes 1962). It seems reasonable to expect that such expenses tend to increase with earnings.

The boundaries between consumption and savings are also blurred. Expenditures for term life insurance are often classified as savings, although outlays for other forms of current insurance against risk are not. Capital investments in owner-occupied housing are classified as savings, whereas the purchase of an automobile is not. The purchase of land to establish a son in farming is counted as savings, but expenditures for the professional education of a son are not, although both are investments designed to augment future income.

Expenditure data understate consumption if some products or services consumed are excluded. Among the items usually not represented in survey data are (a) products withheld from sale by producers for their own use, such as food and fuel produced and used by the same farm families; (b) the services of owner-occupied housing in excess of current expenditures for them; (c) the services of the stock of consumer durables, such as automobiles and household equipment, in excess of expenditure for its maintenance; (d) nonmarket services used, such as those of family members; and (e) products or services provided from public funds.

Income in kind. Consumer products used but not purchased are commonly referred to as income in kind. Some income in kind, notably items (a) and (b) above, are included in national estimates of personal consumption expenditures and occasionally in survey data. The other types of income in kind are seldom if ever included in data on personal consumption. Many types of income in kind are negatively correlated with the money income of consumers, and their omission therefore tends to increase the inequality of consumption as it is inferred from expenditures.

Investment in owner-occupied housing is an important source of income in kind. At about 1950 in the United States such income was approximately $6 for every $100 of expenditures of nonfarm consumers. Income in kind from this source tends to increase with the age of the head of the consumer unit, since older heads are more likely to have full equity in their homes (U.S. Bureau of Labor Statistics 1964; Reid 1962; U.S. Bureau of the Census 1963). In addition, it tends to increase with average lifetime income, and in this respect differs from many other types of income in kind.

The omission of income in kind results in considerable understatement of the consumption of farm families. For example, the income in kind of farm families, largely in the form of food and housing, in a low-income area of the United States during 1957 was $52 for every $100 of expenditures (Pennock 1964). In 1955, despite a general decline in subsistence farming, food produced on the farm and valued at the prices that would have been paid had it been purchased accounted for about 40 per cent of the food consumption of all farm households (U.S. Agricultural … 1956–1961).

Current expenditure overstates consumption when a stock of consumer durables is being increased. A stock of consumer capital, such as furniture, equipment, an automobile, and clothing, is a source of income in kind. The time of purchase of these products tends to differ with the age of the family head. In the United States in 1960–1961, expenditures for furnishings, equipment, and transportation were 25 per cent of consumption expenditures for consumer units whose head was under 25 years of age and 11 per cent for those whose head was 75 years of age or more (U.S. Bureau of Labor Statistics 1964). This difference reflects different stages in the cycle of building up consumer capital and allowing it to depreciate, as well as some differences in income and consumption.

Expenditure data do not include the value of services to the household by its own members. There has been a downward trend in the percentage of households with a full-time homemaker. Consequently, time series of expenditure data are likely to overstate the increase in consumption. In addition, consumption seems likely to exceed expenditure by larger amounts where there is a full-time rather than a part-time homemaker (Clark 1958; Holmes 1962; Rowntree & Lavers 1951). Since earning by wives tends to increase with family income, the higher the family income, the less important, in general, are the unpaid services supplementing expenditure.

Consumption wholly or partly financed by public funds has become an increasing proportion of total consumption. In the United States public expenditure for health, education, recreation, food, and housing was $7 per $100 of personal expenditure in 1950 and $9 in 1962. Public expenditure differs by type of product. During 1962 private expenditure for education and related research in the United States was $1.46 per $100 of total private expenditure. When public expenditure for education is included, this relation increases to $7.50. However, private expenditure for medical care was $6.20 per $100 of total private expenditure, and when public expenditure for “health and medical services” is included, this relation increases only to $7.63 (Social Security Bulletin . . . 1961- U.S. Bureau of the Census 1964).

Publicly provided products or services available only to low-income families, such as public housing, subsidized food, medical care, and legal services, tend to decrease the inequality of consumer real incomes. This also occurs if the product publicly provided is one much used by low-income families or if its quality is too poor to be acceptable to high-income consumers. In some communities of the United States expenditures for tuition in private high schools increase markedly with income, although publicly financed high school education is available without charge to all persons. Public financing of colleges probably benefits families of moderate to high incomes more than those of low incomes; this is probably also true of subsidies to encourage owner occupancy of housing. The extent to which the inequality of real income is affected by public financing of consumption is almost a virgin field of inquiry (Peacock 1954).

Expenditure patterns

Three sets of conditions cause consumption to differ among families: needs and preferences for various consumer products, the resources available, and the cost of things desired. Each of these sets includes a vast array of factors. Intercorrelations among them and incomplete information about them have been formidable obstacles to obtaining definitive estimates of the effect of any one factor. A large stock of survey data describes expenditure in relation to income, the size and composition of consumer units, the occupation of the household head, tenure of dwelling, place of residence, and other conditions. The effect of any one factor, such as prices, income, or the age of the head, appears to differ among sets of data (Friedman 1957; Houthakker 1957; Reid 1962; Reid & Dunsing 1956). Understanding the cause of such differences among survey findings is essential to knowledge of the structure of consumption, and inquiry along many lines is under way.

Engel curves and permanent income theory. The allocation of increased income between expenditure and saving and among the various types of products has important implications for economic development. Engel curves describe the average expenditures of consumers by income levels. The ratio of the rate of increase in expenditure to the rate of increase in income is called the coefficient of income elasticity of expenditures. It is common to describe as necessities products with an income elasticity of less than 1.0 and as luxuries those with an income elasticity greater than 1.0. In comparing the Engel curves of broad consumption categories such as food, clothing, housing, and transportation this distinction has little merit, since income elasticities as great as 1.0 are seldom observed. Each of these categories includes some necessities and some luxuries.

As the stock of survey data has grown, it has become obvious that the slopes of Engel curves differ from group to group and time to time. Such differences have been observed for total expenditures (Friedman 1957; Houthakker 1957) and for food and housing (Reid 1962; Reid & Dunsing 1956). The permanent income theory of consumption has done much to rationalize these differences. It views annual incomes as having two components, one positive component representing expected income and a second, which may be negative or positive, representing deviations from this expected income. These deviations are referred to as transitory income [seeConsumption function]. The theory postulates that consumption is highly correlated with expected income but not with transitory income. Although the theory is by no means fully tested, there seems to be no doubt that it explains much of the variation in Engel curves for a single type of product and that transitory income is more highly correlated with expenditures for durable goods than with expenditures for housing and food.

The mixture of transitory and expected income represented by income distributions is not constant. With the decrease in the proportion of consumer units whose head is self-employed (a group for which transitory income is common) and with the increase in income made available through transfer payments during periods of unemployment, sickness, and disability, the relative size of the transitory component of consumer income seems likely to decrease. Such a decrease will tend to increase the slope of Engel curves. On the other hand, an increase in the progressive taxation of incomes, which makes the normal income of consumer units less unequal, is likely to increase the proportion of the income variation that represents transitory income. This will tend to reduce the slope of Engel curves.

Engel curves are useful chiefly for indicating the short-run response to changes in income. They do not describe the differences in consumption between the rich and the poor or the effect on consumption of secular changes in real income. Observations describing the average consumption and income of subsets of consumers by the occupation of the head have considerable merit for this purpose. The designing of consumer surveys suitable for testing the effect on consumption of differences in permanent, or normal, income is still in a preliminary stage.

Role of assets. The growth of data on assets has lagged behind that of data on expenditures and income. A comprehensive survey of the assets of consumer units in the United States was made for 1962 (Survey of the Financial … 1964). Marked increases in the value of assets were reported with increases in the age of the head of consumer units and with increases in income. Even so, units with incomes under $3,000 had a mean net worth of about $9,000. This figure reflects in part a concentration at incomes under $3,000 of consumer units with an elderly head. Short-run declines in income also result in some large holdings of assets being reported among units with low current incomes. Better data on assets will aid greatly in explaining the deficit spending observed at low annual incomes.

Role of time. The role of time in decisions concerning earning and consumption is receiving increasing attention (Clark 1958; Kriazhev & Markovich 1959; Scientific Conference … 1962; Moscow 1959–1962). Time must be allocated among earnings to provide current income, formal training to provide future income, unpaid productive services to supplement expenditure, and leisure or consumption activities as such. Mincer (1963), in his consideration of the gainful employment of women, notes that the time costs incurred by individuals in earning their money income must be considered if the effects of money income and prices of products on consumption are not to be misunderstood. Income foregone while attending college is widely recognized as a cost of college education. Becker (1965) extends the consideration of time cost to the purchase of consumer products in general and to use of leisure time. For example, the cost of attending the theater includes the time entailed as well as the price of tickets and of transportation. The cost of commuting to work involves time as well as money. In this frame of reference leisure is assessed in terms of consumption foregone because of reduced income.

Household composition. The economic aspects of household membership are also receiving considerable attention (Brady 1958). The rise in real income and the growth of programs providing retirement income have been accompanied by a reduction in the number of three-generation households. At what income does preference for separate households offset the economies of scale of households of three generations? How does this un-doubling change the correlations between income and household size and composition, and hence the observed expenditure-income patterns? It seems likely, for example, that undoubling increases the proportion of low-income consumer units that have an elderly head.

Expenditure data tell us a great deal about consumption. Their usefulness for this purpose would be increased if allowance were more fully made for differences in the cost of living due to inter-spatial differences in prices and in climate, for job expenses, and for income in kind. Without such allowance, the measurement of consumption is still in a very elementary stage.

Consumer budgets

Budgeting by consumers for the most part appears to be very informal. What are commonly referred to as consumer budgets are plans prepared by specialists used for consumer education and for the administration of public programs. Among the best known consumer budgets are the low-cost and moderate-cost food plans of the U.S. Department of Agriculture (Cofer et al. 1962) and the City Worker's Budget of the U.S. Department of Labor (Lamale & Stotz 1960). These budgets combine data on the purchases of consumers as reported in occasional surveys with the recommendations of specialists as to physical and social needs. For example, the recommended dietary allowance of the Food and Nutrition Board (Cofer et al. 1962) are intended to represent physical needs, and the standard that there be at least one room per person in the family is intended to represent social needs. Budgets are developed for consumer units of specified size and composition, and scales have also been developed that permit an adjustment in the cost of the budget to the size of the consumer unit.

The recommendations of specialists involve arbitrary judgments, as does the use of data on purchases. The low-cost food plan of the U.S. Department of Agriculture tends to represent the dollar value of the food consumption of families at the first quartile of the income distribution of the group referred to in preparing the plan; the moderate-cost plan tends to represent that of families at the median income. A new national survey is followed by a revision of the food plans with purchase patterns reflecting a new income distribution. Hence, secular change in the food plans tends to approximate secular change in the expenditure for food at the first quartile and the median of the income distributions, when account is taken of the size and composition of consumer units. A lag exists between the consumption represented by the food plans and that of families surveyed because the collection, tabulation, and analysis of the data are time consuming.

The City Worker's Budget purports to represent a modest but adequate level of consumption. It is less explicitly related to a particular position in the income distribution than are the food plans discussed above. The recommendations of many sets of specialists are utilized. When purchase data from surveys are used, they tend to represent the expenditure of median-income families with a husband, a wife who is not in the labor force, and two children living at home. The mean City Worker's Budget for 20 large cities for the autumn of 1959 was $6,083. The median income in 1959 for urban families with a husband who worked and a wife who did not was $6,283 (U.S. Bureau of the Census 1961-). Thus it seems reasonable to conclude that the City Worker's Budget in 1959 tended to approximate the consumption of median-income families. Successive revisions of this budget have brought it to higher levels. For 20 large cities the mean budget for 1959 exceeded the mean budget for 1947 by 88 per cent. Over the same period national per capita personal income increased only 65 per cent. The “modest but adequate” City Worker's Budget has at times been interpreted as a minimum budget. In such terms, about half of the population is below the minimum, and the secular rise in the budget makes any change in this proportion unlikely.

Inequality and poverty

In the antipoverty program launched by the federal government of the United States in 1964, an annual money income of $3,000 was used as a dividing point between poverty and not-poverty (U.S. President 1964). According to this definition, a startling amount of poverty was found to exist in the midst of general affluence. The use of an annual income of $3,000 as the test of poverty has been much debated. What the criterion should be is not an issue that scientific procedure can resolve. It is, however, obvious that a yardstick of this type, with certain modifications, can serve useful purposes. For example, it can be used to determine what progress has been made in reducing the frequency of income that is low in relation to need.

In 1961 urban consumer units with a money income of $3,000 had a per capita income of $1,224 (U.S. Bureau of Labor Statistics 1964). Thirteen per cent of the urban population was in consumer units with incomes of less than $3,000. A per capita income of $1,224 in 1961 had the same purchasing power as $984 in 1950, and 24 per cent of the urban population of 1950 was in consumer units with incomes below this level (Study of Consumer Expenditures . . . 1957). An income of $1,224 in 1961 had the same purchasing power as $602 in 1941, and 40 per cent of the urban population of 1941 was in consumer units with incomes below this level (Hanson 1945). These changes imply a secular decline of more than one per cent a year in the percentage of the urban population receiving purchasing power of less than 1,224 1960 dollars per capita.

After a study in Great Britain of the income of wage earners and families receiving assistance in 1936 and 1953, Schulz concluded that “poverty, in the sense in which the term was understood before the war, has been abolished for the families of the type considered” (1955, p. 232). Rowntree identified five comparable consumption standards for 1936 and 1950, and for the city of York, England, found 18 per cent of the population in the two lowest classes in 1936 and 2 per cent in 1950 (Rowntree & Lavers 1951).

A yardstick of poverty can also be used to discover the people who are most likely to be poor and the conditions that contribute to their poverty. To be suitable for this purpose a yardstick must represent the same level of consumption wherever it is applied. To consider only money income is to ignore income in kind and important interspatial differences in prices and other conditions that affect the cost of living, such as climate. How the size and composition of consumer units affect needs is also ignored when a single yardstick is used for all units. However, many descriptions of “who are the poor” have taken into account variation in need that depends on the number of persons per consumer unit. The other shortcomings of the single money measure of poverty have tended to be neglected. As a result, descriptions of poverty in the United States have tended to overstate the frequency of low consumption in farm and other rural populations and in warmer regions where the cost of shelter is low; they have tended to understate the frequency of poverty in large cities where the costs of housing and of transportation to work are high.

Relative as well as absolute poverty is a matter of considerable importance. [SeeIncome distribution, article onsize.] Data on the size distribution of money income have been the most important basis for estimates of the degree of inequality. For industrialized countries, a secular decrease in inequality is indicated (Kuznets 1965). However, the biases of money income as an index of consumption and the failure to consider the changing membership of consumer units greatly limit the reliability of such estimates. These biases have not been constant through time and differ greatly among populations. The limiting conditions include income taxes, job expenses, correlations between current income and prices, income in kind, the characteristics of consumer units, and the extent to which income distributions represent transitory conditions. Systematic exploration of many of these biases has yet to be undertaken.

Data available leave no doubt that the inequality of expenditures is appreciably less than the inequality of money incomes (Snyder 1960). When urban consumer units of two or more persons are ranked by 1961 income, those in the lowest fifth of the income distribution are found to have received 6.9 per cent of the money income and 7.6 per cent of the disposable income and to have made 9.5 per cent of all consumer expenditures. Measurement in terms of the number of persons in the consumer units further reduces the apparent inequality of consumption. The lowest fifth of the urban population (ranked by the income of consumer units) received 8.8 per cent of the money income and 9.6 per cent of the disposable income and made 11.9 per cent of all consumer expenditures. So far expenditure data have been little used to provide insights into the inequality of consumption and of normal income.

Viewed historically, scientific discoveries and new technology stand out as the main factors bringing about changes in consumption. They have resulted in new products, greater efficiency of production, higher real income, and changes in the relative cost of various products. During much of the twentieth century the tempo of change has been very rapid—within a few decades transforming rural, rather static, societies into urban, highly dynamic societies. In general, current changes sustain an expectation of further increases in consumption, year-by-year improvement in old products, and a continuing flow of new products.

Changes in the structure of market economies, in the relation of private to public economies, and in the welfare objective of societies create a need for increased knowledge of consumption and of the forces shaping it and for a greater understanding of its implications for social goals. Considerable progress has been made in measuring the total output of consumer products of the market economy. Only moderate advances have been made in understanding the response to changes in costs as reflected in money outlays, in consumer or home production, and in the use of the products of the public economy. Nor has much progress been made in measuring changes in the inequality of consumption.

Margaret G. Reid

[See also the biography ofEngel.]


Becker, Gary 1965 A Theory of the Allocation of Time. Economic Journal 75:493–517.

Bennett, Merrill K. 1951 International Disparities in Consumption Levels. American Economic Review 41:632–649.

Brady, Dorothy S. 1958 Individual Incomes and the Structure of Consumer Units. American Economic Review 48, no. 2:269–278.

Brady, Dorothy S.; and Hurwitz, Abner 1957 Measuring Comparative Purchasing Power. Conference on Research in Income and Wealth, Studies in Income and Wealth 20:301–347.

Clark, Colin 1958 The Economics of House-work. Oxford University Institute of Statistics, Bulletin 20:205–211.

Cofer, Eloise; Grossman, Evelyn; and Clark, Faith 1962 Family Food Plans and Food Costs: For Nutritionists and Other Leaders Who Develop or Use Food Plans. U.S. Department of Agriculture, Home Economics Research Report No. 20. Washington: Government Printing Office.

Davis, Joseph S. 1945 Standards and Content of Living. American Economic Review 35:1–15.

Friedman, Milton 1957 A Theory of the Consumption Function. National Bureau of Economic Research, General Series, No. 63. Princeton Univ. Press.

Hanson, Alice C. 1945 Family Spending and Saving in Wartime. U.S. Department of Labor, Bulletin No. 822. Washington: Government Printing Office.

Holmes, Emma G. 1962 Job-related Expenditures and Management Practices of Gainfully Employed Wives in Four Georgia Cities. U.S. Department of Agriculture, Home Economics Research Report No. 15. Washington: Government Printing Office.

Houthakker, Hendrik S. 1957 An International Comparison of Household Expenditure Patterns, Commemorating the Centenary of Engel's Law. Econometrica 25:532–551.

Kriazhev, V.; and Markovich, M. 1959 The Time Budget and Measures to Improve the Living Conditions of the Working People. Problems of Economics [1959] no. 8:49–53.

Kuznets, Simon 1965 Economic Growth and Structure. New York: Norton.

Lamale, Helen H.; and Stotz, Margaret S. 1960 The Interim City Worker's Family Budget. U.S. Bureau of Labor Statistics, Monthly Labor Review 83:785–808.

Mincer, Jacob 1963 Market Prices, Opportunity Costs and Income Effects. Pages 67–82 in Measurement in Economics: Studies in Mathematical Economics and Econometrics in Memory of Yehuda Grunfeld. Stanford Univ. Press.

Moscow, Nauchno-issledovatel'skii Institut Truda 1959–1962 Metodologicheskie voprosy izucheniia urovnia zhizni trudiashchikhsia (Methodological Problems in the Study of the Standard of Living of Workers). 2 vols. Moscow: Izdatel'stvo Social'no-ekonomicheskoi Literatury.

Peacock, Alan T. 1954 Income Redistribution and Social Policy: A Set of Studies. London: Cape.

Pennock, J. L. 1964 Rural Family Spending and Consumption in a Low-income Area in Kentucky. U.S. Department of Agriculture, Home Economics Research Report No. 26. Washington: Government Printing Office.

Pipping, Hugo E. 1953 Standard of Living: The Concept and Its Place in Economics. Societas Scientiarum Fennica, Commentationes humanarum litterarum, 18, No. 4. Helsingfors (Finland): Societas Scientiarum Fennica.

Reid, Margaret G. 1962 Housing and Income. Univ. of Chicago Press.

Rid, Margaret G.; and Dunsing, Marilyn 1956 Effect of Variability of Incomes on Level of Income-Expenditure Curves of Farm Families. Review of Economics and Statistics 38:90–95.

Rowntree, Benjamin S.; and Lavers, G. R. 1951 Poverty and the Welfare State: A Third Social Survey of York Dealing Only With Economic Questions. London: Longmans.

Schulz, T. 1955 The Means of Subsistence: Income From Earnings and From Assistance, 1935–1953. Oxford University Institute of Statistics, Bulletin 17:215–238.

Scientific Conference on Statistical Problems, Budapest, 1961 1962 The Standard of Living: Some Problems of Analysis and of International Comparison. Edited by M. Mod et al. Budapest: Akademiai Kiadé.

Snyder, Eleanor M. 1960 Low Income in Urban Areas. American Economic Review 50, no. 2:243–250.

Social Security Bulletin: Annual Statistical Supplement, 1961.→ Published since 1956.

Stigler, George J. 1954 The Early History of Empirical Studies of Consumer Behavior. Journal of Political Economy 62:95–113.

Study of Consumer Expenditures, Incomes, and Savings; Statistical Tables: Urban U.S.-1950. Vol. 18. 1957 Philadelphia: Univ. of Pennsylvania. → Tables tabulated by the Bureau of Labor Statistics for the Wharton School of Finance and Commerce, University of Pennsylvania.

Survey of the Financial Characteristics of Consumers. 1964 [U.S.] Board of Governors of the Federal Reserve System, Federal Reserve Bulletin 50:285–293.

United Nations, Statistical Office 1964 Statistical Yearbook, 1963. New York: United Nations. → Published since 1948.

U.S. Agricultural Research Service, Household Economics Research Branch 1956–1961 Household Food Consumption Survey, 1955: Report. 16 pts. Washington: U.S. Department of Agriculture.

U.S. Bureau of Labor Statistics 1924 Cost of Living in the United States. Bulletin No. 357. Washington: Government Printing Office.

U.S. Bureau of Labor Statistics 1941 Money Disbursements of Wage Earners and Clerical Workers: 1934–1936: Summary Volume. Bulletin No. 638. Washington: Government Printing Office.

U.S. Bureau of Labor Statistics 1964 Consumer Expenditures and Income: Urban United States 1960–61. BLS Report 237. Washington: Government Printing Office.

U.S. Bureau of the Census 1960 Historical Statistics of the United States; Colonial Times to 1957: A Statistical Abstract Supplement. Washington: Government Printing Office.

U.S. Bureau of the Census 1961— U.S. Census of Population: 1960. Washington: Government Printing Office.

U.S. Bureau of the Census 1963 U.S. Census of Housing: 1960. Volume 1, Part 1: United States Summary. HC[1]-1. Washington: Government Printing Office.

U.S. Bureau of the Census 1964 Statistical Abstract of the United States. Washington: Government Printing Office. → Published since 1878.

U.S. President 1964 Economic Report of the President. Washington: Government Printing Office. → Published since 1947.

Williams, Faith M. 1958 International Comparisons of Patterns of Family Consumption. Pages 270–307 in Lincoln H. Clark (editor), Consumer Behavior: Research on Consumer Reactions. New York: Harper.

Williams, Faith M.; and Zimmerman, Carle C. 1935 Studies of Family Living in the United States and Other Countries: An Analysis of Material and Method. U.S. Department of Agriculture, Miscellaneous Publication No. 223. Washington: Government Printing Office.


In a general sense, consumer assets encompass all tangible goods or intangibles which consumers own or on which they have claims. This includes not only household goods but also most of the assets of business corporations and of government, as evidenced by consumer holdings of stocks and bonds.

It is customary to classify consumer assets in various ways, roughly analogous to the definition of national wealth [seeNational wealth]. Perhaps the most frequently encountered of such classifications are the following.

Tangibles vs. financial Tangibles include all material possessions owned directly by the consumer, e.g., household goods, cars, homes, jewelry, and ownership of a farm. Intangibles, or financial assets, include cash and claims on money or goods, such as checking and savings accounts, common and preferred stock, bonds, cash value of life insurance, mortgages held on property of others, and loans to others.

Financial assets may be subdivided into fixed-dollar and variable-dollar holdings. The former represent assets for which the consumer has a claim on a fixed number of dollars, such as cash, a checking account, or a mortgage. In contrast, variable-dollar holdings are those for which the return is dependent on the price level or on other economic conditions. Thus, the current value of 100 shares of common stock originally purchased at $25 per share will depend on the performance of the company since that time, and may be substantially more or less than $2,500. Other examples of variable-dollar holdings include debentures and the variable annuity.

Liquid vs. nonliquid. Liquid assets are those which are immediately convertible into cash, into a fixed number of dollars. Hence, a liquid asset is a fixed-dollar asset, but not all fixed-dollar assets are liquid assets. For example, U.S. government Series E bonds are fixed-dollar assets, immediately convertible, and therefore liquid assets. Series H bonds, however, which are also fixed-dollar assets, are not immediately convertible into cash and are therefore nonliquid assets. The principal forms of liquid assets in the United States, other than money itself, are savings accounts and U.S. government Series E bonds.

Gross vs. net. The total wealth of the consumer, without deduction for debts or liabilities, is known as his gross, or total, assets. Deducting his debts, or liabilities, from this total yields net assets. The gross assets of a consumer are equal to his net assets only if he possesses no debts.

Holdings of consumer assets

Accurate estimates of consumer assets, or wealth, are as yet obtainable only for national totals and only for certain aggregates and certain countries. The reason is that, with a few exceptions, such data can be obtained only from consumers, and consumers are notoriously unreliable reporters on such subjects. Financial holdings tend to be understated and distorted in various ways. Nonfinancial holdings, such as the value of one's appliances or of clothing, cannot be estimated with much reliability because of the difficulty of ascertaining such holdings and because of the very poor basis for estimating the resale value of such goods. The estimation of consumer financial holdings from institutional records is usually hardly feasible either, partly because of the difficulty of collating holdings in different institutions for the same consumer unit and partly because few institutions distinguish between consumer holdings and holdings of businesses or of institutions. The latter problem makes it difficult to estimate even aggregates of consumer assets from institutional data.

For these reasons, the consumer asset, or wealth, estimates that have been made should be treated as rough approximations and not as exact figures. For the United States, estimates of total household wealth have been made as part of the national wealth estimates of Raymond Goldsmith. These estimates are discussed and presented for the end of 1958 in the article National wealth. As shown in Table 1 of that article, household net wealth at that time in the United States amounted to just under $1.5 billion, or over 60 per cent of the national total. Roughly three-fifths of this house-hold wealth was in financial form, with the rest in physical assets.

Detailed estimates of the aggregate financial wealth of individuals are published by the U.S. Securities and Exchange Commission, based on data obtained from financial institutions, and are shown in Table 1 for selected years between 1950 and 1964. These data include not only consumers but also nonprofit institutions, and encompass such peripheral items as private insurance and pension reserves. Therefore, not all of these holdings can be attributed to consumers or to households. Nevertheless, these data suggest that the great bulk of consumer financial wealth is in the form of preferred and common stock, insurance, and savings deposits of one form or another. The table also brings out the dramatic increase in the postWorld War Ii period in mortgage and consumer debt, on the liability side, and in equities and savings shares (mostly in savings and loan associations), on the assets side.

Another source of estimates of the financial wealth of consumers is sample surveys of saving or of components of wealth made periodically by governmental and other agencies. The annual Survey of Consumer Finances of the University of Michigan Survey Research Center provides information on the liquid asset holdings of U.S. consumers as well as on ownership of homes, cars, and selected durable goods and on different types of consumer debt. The most comprehensive study of this nature so far is the “Survey of Financial Characteristics of Consumers” conducted for the Federal

Table 1Financial assets and liabilities of individuals. United States, selected years (in billions of dollars as of end of year)a
a. Includes Alaska and Hawaii.
b. Not available
c.Beginning 1955, includes nonguaranteed federal agency issues. In 1950, such issues were included with corporate and other bonds.
d. Rough estimates of market value.
Source: U.S. Securities and Exchange Commission, Annual Report.
Financial assets:    
Currency and deposits130154181243
Currency and demand deposits(73)(80)(80)(97)
Time and savings deposits(57)(75)(101)(146)
Savings shares153466109
U.S. savings bonds50(50)(46)(49)
Other U.S. governmentc18(20)(27)(29)
State and local government13(23)(31)(36)
Corporate and otherdb(304)(401)(615)
Bonds and notesb(21)(22)(22)
Investment company sharesb(12)(23)(39)
Other preferred and common sharesb(271)(357)(554)
Private insurance and pension reserves68106152206
Insurance reserves(57)(77)(96)(118)
Insured pension reserves(6)(11)(19)(25)
Noninsured pension reserves(6)(18)(37)(63)
Government insurance and pension reserves40577184
Total assetsb7499751,372
Mortgage debt3878130187
Securities loans3558
Total liabilities58117185265
Net equity (assets – liabilities)b6327901,107

Reserve Board by the U.S. Bureau of the Census in the spring and summer of 1963 (Projector 1964). This study sought to obtain from a national sample of roughly 3,600 families information on ownership of liquid assets, common and preferred stock, marketable bonds, life insurance, annuities and investment in retirement plans, the value of ownership of a business (both farm and nonfarm), ownership of homes and automobiles, and ownership of different forms of debt. Reinterviews with this sample were conducted in 1964.

The survey-based estimates of consumer financial wealth tend to complement the aggregative estimates, because only the survey data provide extensive information on the extent to which asset holdings are distributed among different types of consumer units. These and other data indicate that consumer wealth is highly concentrated [seeNational wealth, article ondistribution]. Some idea of the extent to which wealth is concentrated among consumer units is given by Table 2.

Table 2 — Estimated distribution of U.S. families in 1963 by net worth and the average net worth of the members of each group
Net worth (in dollars)Per cent of familiesaAverage net worth (in dollars)
a. Per cents do not add to 100 because of rounding.
b. Less than half of one per cent.
Source: Projector 1964, pp. 289–293.
Negative 0–9998–538
500,000 and overb1,176,281

As a rule, net worth, as well as total assets, tend to be higher for older families, for families with above-average incomes and savings rates, and for the self-employed.

Perhaps equally striking is the extent to which average holdings of different types of assets vary by family characteristics, as is shown in Table 3. Lower-income families have most of their assets in tangible goods or liquid form (mostly checking and savings accounts); this is much more so than the table indicates because most widely owned durable goods are not included. At higher income levels, and also among older people, holdings tend to be concentrated increasingly in a business and in investment assets (primarily common and preferred stock). Particularly noteworthy is the fact that among the wealthiest families in the United States in 1962—the fraction of 1 per cent having a net worth of $500,000 or more—over 90 per cent of these assets were in variable-dollar form (mostly a business, stocks, and marketable bonds), and only a very small fraction was in liquid assets or in other fixed-dollar investments.

These data have to be treated as approximations only, because consumer financial survey data are subject to substantial biases. These biases may not only lead to substantial underestimates of the aggregative value of consumer financial holdings but may also distort the pattern of these holdings. In particular, tendencies exist for small holdings to be overstated, for large holdings to be understated, and for the degree of understatement to increase with the sensitivity of the respondent about the asset.

Consumer durables

Much less is known about consumer holdings of durable, or tangible, goods than of intangibles because of the many problems involved in locating and valuing such goods. For the United States, the Goldsmith estimates indicate that roughly a fourth of consumer wealth is in land and housing, and about 10 per cent is in other consumer durables.

Only for the two most expensive such goods— houses and automobiles—are many data available. The Federal Reserve survey data cited earlier indicate that at the beginning of 1963 roughly three-fifths of all U.S. families owned the house in which they lived. Summary data on the characteristics of these homes are shown in Table 4, relating to one-family homes with Federal Housing Authority insured mortgages. The table shows how the value of these homes rose between 1950 and 1964 and suggests that further increases are in store because of the higher costs and values in 1964 of new homes than of existing homes. The table also illustrates the relative growth during this period of mortgages in relation to home values and of land in relation to building costs.

Automobiles were owned by nearly four-fifths of all U.S. families in 1965, as shown in Table 5. Nearly a fourth of the families owned two or more automobiles. The table also brings out the fact that just under a third of all cars in 1965 were eight years old or more and slightly more than another third were less than four years old, and that most cars, especially new ones, have been purchased through the use of credit.

Extensive data on the value and characteristics of other durable goods owned by U.S. consumers, as well as of houses and cars, were obtained as part of the 1950 and 1960–1961 Consumer Expenditures

Table 3 — Average family holdings of specific types of net worth, in dollars, by selected characteristics, December 31, 1962°
          Investment assets
a. Note: All data are preliminary and are jbject to revision. Details may not add to totals because of rounding.
b. No cases reported.
Source: Projector 1964, p. 293.
All families22,5886,6125,9756373,9131,3769,6422,5797,0634,0724562,5351,528483
Size of net worth:
5,000-9,9997,3054,536 5404049271,6651,2683971508240197374
10,000-24,99916,2819,4223,9968,63 47891,6561,5113,9802,2661,71556781,140241529
25,000-49,99935,30914,95613,721 5,283 11,8745,9615,9142,1322723,510795225
50,000-99,99967,04215,74814,4291,2361,31 915,7012,6254,34 2 9,512 9,659461 1,181490
100,000-199,999129,95826,960 1,745 5,31230,56073,06 814,45421,04858,61 438,301 10,92818,11 13,7951,6621,65
200,000-499,999293,655 25,21524,69 12,51922,48465,83 28,803182,00619,151162,855105,1602,2024,24 953,44511,4649
500,000 and over1,176,28127,20954,00 651,4522,554248,81118,677590,16040,973549,187363,20879,023106,956273,2728,646
1962 income:
0-2,9998,875  1491,418190  2,1281,480201448113205
3,000-4,99910,9143,9013,9563,7523,5444121,9026353,4584,66 31,3301,73 82,925818192,088137378
5,000-7,49915,112 4,9736432,0501,135 1,7163,7102,365181,3261,339453
7,500-9,99921,2435,6158,367 8682,5771,8795,4267,50 0  1,476443,2581,632712
10,000-14,99930,389 7,4999,52 7 5,1742,97511,2022,7224,23 34,7796,96 93,7613162,893749584
15,000-24,99974,32910,87317,00 415,1881,3461,81 69,0885,19639,8809,241 18,7331,44510,4603,664502
25,000-49,999267,996   66,144 111,76119,09830,63892,66 358,1114,74229,81048,736 
50,000-99,999789,58235,09048,76432,21545,96 12,8752,80 3251,97710,81919,55 9387,573 345,728204,66571,97169,09286,3134,5534,604
100,000 and over1,554,15289,64585,6344,011288,91532,3091,058,67241,84554,4261,004,246758,253121,985124,00896,87912,268
Age of family head:              
Under 257625442482973612538125612544b81169493
35-4419,442  708 1,4966,0611,5564,505 1951,9532,541546
45-5425,4595,9528,55 75,2447,64 59123,9395,77 62,2418,1442,5635,5812,3562,83 42722,4751,472730
65 and over30,7187,8467,4743723,26787318,4524,67013,7828,3491,2344,198535256
Employment–housing status:
Nonfarm homeowner31,478  8484,4411,82712,7783,3019,4775,4535653,4592,013576
Self-employed96,38510,99617,69 510,14816,40 31,29234,3673,883 6,71030,43814,3882,64213,4084,6461,355
Employed by other s22,0269,9028,9749287021,88437,1488,067 5,6783,5861951,8972,010539
Retired29,75211,28710,95233574860316,9912,3885,02 311,9697,3616783,930421299
Nonfarm renter8,092383483351,1677535,2391,5863,6542,411303940903354
Employed by others5,268395283674208033,8131,2442,5691,60090879150312
Farm operator43,9736,1825,50168125,7671,27810,1382,3097,8291,3545355,9401,095486
Northeast23,9807,1416,6115303,0261,70810,8333,400 5,5816821,1711,783512
North Central23,6327,4546,728726 1,312 2,6267,4346,52 73,0063353,1861,245486
South18,318 4,5715974,9543,40 9 9,1538,11 21,9156,197 5132,426929429
West26,1925,1686,94 16,2197234,4231,1281,40 811,3002,4158,8853,2584,94 92123,7252,647529
Table 4 — Median cost and value of new and existing one-family homes in the United States with Federal Housing Authority insured mortgages, selected years (in dollars)
 New HomesExisting Homes
* Not available
Source: U.S. Housing and Home Finance Agancy 1964.
Estimated Value8,28614,60715,15116,0638,86513,04314,08214,614
Mortgage amount7,10113,56914,19515,1186,80111,97813,10013,725
Ratio: Mortgage to value88.093.594.494.577.892.694.494.8
Price of site*2,4042,6492,990*2,2852,6532,824
Ratio: Site to value**16.818.118.7

penditures Studies of the Bureau of Labor Statistics. More recent estimates of the frequency of ownership of different appliances are shown for the U.S. in Table 6 for 1963. It is interesting to note that by that year only one appliance (electric refrigerators) was owned more widely than television sets.

Assets and consumer behavior

A number of studies have been made of the factors that influence consumers to hold particular combinations of assets, some dealing with financial assets and others with durable goods. Still other studies have sought to ascertain the effect of asset holdings on consumer spending and saving.

Influencing factors. Reasons for consumer holding of financial assets have been discussed in the theoretical literature for some time (for example, Keynes 1936, chapter 16; Katona 1951, pp. 98–107). However, relatively few empirical studies have been made on this subject, to a large extent, no doubt, because of lack of data. The findings of the principal such studies may be summarized, briefly, as follows:

(1) The composition of consumer portfolios is influenced by numerous social, economic, and environmental variables (Watts & Tobin 1960). In particular, holding other variables constant in each case:

(a) Financial assets rise in importance and the value of durables declines as a percent of the total as household income rises.

Table 5 — U.S. automobile ownership, age, and financing: selected yearsa
a. Excludes Alaska and Hawaii.
b. Not available.
c. Percentages may add to less than 100 per cent because of not ascertained cases.
Source: Survey of Consumer Finances.
Total number of families in U.S.millions45.249.353.454.254.956.556.858.4
Total owning automobilesper cent5970777674807879
Owning 1 automobileper cent5260625857585655
Owning 2 or more automobilesper cent710151817222224
Automobiles owned, by age:
Less than 2 years oldper cent1712141312131316
2 and 3 years oldper cent1922201721182021
4, 5, 6 and 7 years oldper cent643414238363333
8 years old and overper cent5823252829333430
Total automobiles purchasedmillionsb13.614.316.415.618.917.318.3
New carmillionsb4.
Average price paiddollarsb2,7303,1403,0102,8302,9903,1303,140
Used carmillionsb9.
Average price paiddollarsb780980800800840920920
Method of financing purchases-.0
All passenger cars:
Full cash (including trade-in allowance)per cent5137384743454748
Installment credit and other borrowingper cent4961615256555352
New passenger cars:
Full cash (including trade-in allowance)per cent5638333937383940
Installment credit and other borrowingper cent4461666162616060
Used passenger cars:
Full cash (including trade-in allowance)per cent4737415244485153
Installment credit and other borrowingper cent5362574753504444
Table 6 — Number and per cent of U.S. households with specified household appliances, 7963°
a. Excludes Alaska and Hawaii.
b. Total number of households = 55,050,000.
Source: Look Magazine 1963.
 Number, in thousandsPer cent
Major appliances:
Air conditioner:
Room unit7,24013.2
Clothes dryer:
Combination washer–dryer1,1192.0
Garbage disposer4,7898.7
Range, electric19,04734.6
Range, gas33,91461.6
Refrigerator, electric53,06396.4
Refrigerator, gas1,3132.4
Sewing machine, electric25,40846.2
Black and white50,00190.8
Washing machine:
Vacuum cleaner39,80372.3
Small electrical appliances:
Broom, upright1,0451.9
Can opener5,97310.9
Coffeemaker, automatic27,16549.4
Clock (excluding radio)34,96863.5
Floor polisher5,0769.2
Frying pan21,67339.4
Hair dryer (hat box)8,71915.8
Radio (including clock)26,66548.4
Radio, pocket/purse15,08927.4
Radio, other portable21,24738.6
Steam iron35,75465.0

(b) Financial assets rise in importance and debt declines with increased education of the household head.

(c) Older households have fewer holdings in durables and less debt.

(d) Larger households tend to have more of their holdings in durables and fewer in liquid assets.

(e) Pronounced, though erratic, differences in asset holdings exist by occupation, by region, and by city size.

(2) Holdings of variable-dollar assets are also influenced by numerous factors, principally age, home ownership, total assets, and occupation (Claycamp 1963).

(3) Holdings of liquid assets are influenced primarily by total wealth rather than by income (Lydall 1958; Claycamp 1963; Butters et al. 1953). Liquid assets as a proportion of total assets tend to decline as the amount of wealth rises. Liquid assets in relation to income tend to rise with age, to fall as the size of the consumer unit rises, and to be higher for those with permanently depressed incomes (Guthrie 1960).

(4) Holdings of specific assets such as life insurance and common stock are also influenced by a large number of different variables, though income and occupation emerge frequently as dominant factors. Of particular interest is the finding that price expectations and attitudinal variables are related to ownership of common stock (Kreinin 1957; 1959; Kreinin et al. 1961).

Factors influencing the ownership of durable goods have been studied largely in terms of the purchase of such goods, but also in terms of the acquisition patterns of such goods. Thus, attempts have been made to derive so-called saturation curves, showing how ownership of particular goods spreads through a population. Such information has been used as a basis for forecasting durable goods purchases as well as for general analytical purposes (Brown, Buck, & Pyatt 1965; Pyatt 1964). Various sociological studies have explored the role of the innovator in durable goods purchases and the manner in which ownership of a good spreads through a population. These studies bring out the importance of the group, and of group leaders, in influencing such purchases (Bourne 1957; Katz & Lazarsfeld 1955; Whyte 1954). These innovators appear to be scattered through the population but, judging by one such study, are generally found among younger, better-educated people as well as among those who already own similar products and are financially optimistic (Mueller 1958).

Other studies have shown that many factors enter into the durable goods purchase decision, some economic, some sociological, and some psychological. The socioeconomic factors include income, income change, assets, age, family size, and occupation (Ferber 1955; Crockett & Friend 1960; Morgan 1958). The life cycle of the family has been found to be particularly relevant (Consumer Behavior 1955), with purchases increasing and accumulating as a family begins to raise children, and then declining as the children grow up and leave home. These studies also suggest that although many durable goods purchases are planned and carefully thought out in advance, many others are made almost on the spur of the moment (Ferber 1955; Juster 1964; Katona & Mueller 1954). Social psychologists have attempted to explain this decision process in terms of accumulation of desires for and against making a particular purchase (Bilkey 1957) and in terms of the interplay of different motivations and status intensities (Clawson 1961).

Effects of ownership. Increasing attention has been focused in recent years on the role of assets in determining consumer spending and saving behavior as well as on the factors influencing asset holdings. In the former connection, assets serve as the bridge for reconciling the cross-section tendency for saving to rise with income and the rising levels of income with the virtual constancy in the United States of the aggregate proportion of income saved over time. Assets play a central role in the so-called “permanent income” hypothesis of Milton Friedman and in the life-cycle hypothesis of Franco Modigliani and his associates; in each case, anticipated expenditure or saving is geared to a long-run perspective of the resources available to the consumer [seeConsumption function].

In a more pragmatic vein, the accumulation of assets seems to lead not to saturation but to the desire for still more assets, a phenomenon which appears to be characteristic of both tangibles and of financial assets (Katona 1964; 1965). Particularly evident is the tendency for those saving through pension funds to save at least as much in other forms as do those not in pension funds, though one may question which is cause and which is effect. At the same time, it is also clear that holdings of some assets influence holdings of others —that consumers try to “balance” their asset holdings (Watts & Tobin I960)—and that to a large extent purchases of durable goods serve as substitutes for financial saving (Klein 1955; Maynes 1959; Friend & Jones 1960).

Robert Ferber

[See also Nationalealth.]


Ando, Albert K.; and modigliani, franco 1963 The “Life Cycle” Hypothesis of Saving: Aggregate Implications and Tests. American Economic Review 53:55–84. → Includes a bibliography on pages 82–84.

bilkey, W. J. 1957 Consistency Test of Psychic Tension Ratings Involved in Consumer Purchasing Behavior. Journal of Social Psychology 45:81–91.

bourne, francis S. (1957) 1961 Group Influences in Marketing and Public Relations. Pages 207–257 in Rensis Likert and Samuel P. Hayes (editors), Some Applications of Behavioral Research. Paris: unesco.

brown, D. A.; buck, S. F.; and pyatt, F. G. 1965 Improving the Sales Forecast for Consumer Durables. Journal of Marketing Research 2:229–234.

butters, john K. et al. 1953 Effects of Taxation: Investments by Individuals. Boston: Harvard Univ., Graduate School of Business, Division of Research.

clawson, C. joseph 1961 Family Composition, Motivation, and Buying Decisions. Pages 200–217 in Consumer Behavior. Volume 4: Household Decisionmaking. Edited by Nelson N. Foote. New York Univ. Press.

claycamp, henry J. jr. 1963 The Composition of Consumer Savings Portfolios. Studies in Consumer Savings, No. 3. Urbana: Univ. of Illinois, Bureau of Economic and Business Research.

Consumer Behavior. Volume 2: The Life Cycle and Consumer Behavior. Edited by Lincoln Clark. 1955 New York Univ. Press.

crockett, jean; and friend, irwin 1960 A Complete Set of Consumer Demand Relationships. Volume 1, pages 1–92 in Conference on Consumption and Saving, University of Pennsylvania, 1959, Proceedings. Edited by Irwin Friend and Robert Jones. Philadelphia: Univ. of Pennsylvania Press.

falkprojectforeconomicresearchinisrael 1961 Survey of Family Savings 1957/1958 and 1958/1959. Jerusalem (Israel): The Project.

ferber, robert 1955 Factors Influencing Durable Goods Purchases. Urbana: Univ. of Illinois, Bureau of Economic and Business Research.

ferber, robert 1962 Research on Household Behavior. American Economic Review 52:19–63.

friedman, milton 1957 A Theory of the Consumption Function. National Bureau of Economic Research, General Series, No. 63. Princeton Univ. Press.

friend, irwin; and jones, robert 1960 The Concept of Saving. Volume 2, pages 336–359 in Conference on Consumption and Saving, University of Pennsylvania, 1959, Proceedings. Edited by Irwin Friend and Robert Jones. Philadelphia: Univ. of Pennsylvania Press.

friend, irwin; and natrella, vito 1954 Individuals' Saving: Volume and Composition. New York: Wiley.

friend, irwin; and schor, stanley 1959 Who Saves? Review of Economics and Statistics 41:213–248.

goldsmith, raymond W. 1955–1956 A Study of Saving in the United States. 3 vols. Princeton Univ. Press. → Volume 1: Introduction; Tables of Annual Estimates of Saving, 1897–1949. Volume 2: Nature and Derivation of Annual Estimates of Saving, 1897–1949. Volume 3: Special Studies. See especially Volume 3.

goldsmith, raymond W. 1962 The National Wealth of the United States in the Post War Period. National Bureau of Economic Research Studies in Capital Formation and Financing, Vol. 10. Princeton Univ. Press.

goldsmith, raymond W.; and saunders, christopher (editors) 1960 The Measurement of National Wealth. Income and Wealth Series, No. 8. Chicago: Quadrangle.

guthrie, harold W. 1960 Consumers' Propensities to Hold Liquid Assets. Journal of the American Statistical Association 55:469–490.

hill, T. P. 1955 Incomes, Savings and Net Worth: The Savings Survey of 1952–1954. Oxford University Institute of Statistics, Bulletin 17:129–172.

juster, francis T. 1964 Anticipations and Purchases: An Analysis of Consumer Behavior. Princeton Univ. Press.

katona, george 1951 Psychological Analysis of Economic Behavior. New York: McGraw-Hill. → A paperback edition was published in 1963.

katona, george 1964 The Mass Consumption Society. New York: McGraw-Hill.

katona, george 1965 Private Pensions and Individual Saving. Ann Arbor: Univ. of Michigan, Institute for Social Research, Survey Research Center.

katona, george; and mueller, eva 1954 A Study of Purchase Decisions. Pages 30–87 in Consumer Behavior. Volume 1: The Dynamics of Consumer Reaction. Edited by Lincoln Clark. New York Univ. Press.

katz, elihu; and lazarsfeld, paul F. 1955 Personal Influence: The Part Played by People in the Flow of Mass Communications. Glencoe, 111.: Free Press. → A paperback edition was published in 1964.

keynes, johnmaynard 1936 The General Theory of Employment, Interest and Money. London: Macmillan. → A paperback edition was published in 1965 by Har court.

klein, L. R. 1955 Major Consumer Expenditures and Ownership of Durable Goods. Oxford University Institute of Statistics, Bulletin 17:387–414.

kreinin, mordechai E. 1959 Factors Associated With Stock Ownership. Review of Economics and Statistics 41:12–23.

kreinin, mordechai E. 1961 Analysis of Liquid Asset Ownership. Review of Economics and Statistics 43:76–80.

kreinin, mordechai E.; lansing, J. B.; and morgan, J. N. 1957 Analysis of Life Insurance Premiums. Review of Economics and Statistics 39:46–54.

lookmagazine 1963 National Appliance Survey, 1963. 2 vols. New York: Cowles Magazine and Broadcasting, Inc. → Volume 1: Major Household Appliances. Volume 2: Portable Household Appliances.

lydall, harold F. 1958 Income, Assets, and the Demand for Money. Review of Economics and Statistics 40:1–14.

maynes, E. scott 1959 The Relationship Between Tangible Investment and Consumer Saving. Review of Economics and Statistics 41:287–294.

morgan, james 1958 Consumer Investment Expenditures. American Economic Review 48:874–902.

mueller, eva 1958 The Desire for Innovations in Household Goods. Pages 13–37 in Consumer Behavior. Volume 3: Research on Consumer Reactions. Edited by Lincoln Clark. New York: Harper.

projector, dorothy S. 1964 Survey of Financial Characteristics of Consumers. Federal Reserve Bulletin 50:285–293.

pyatt, F. graham 1964 Priority Patterns and the Demand for Household Durable Goods. Cambridge Univ. Press.

Survey of Consumer Finances. → Published since 1946 in the Federal Reserve Bulletin.

U.S. bureauofthecensusStatistical Abstract of the United States. → Published since 1878.

U.S. housingandhomefinanceagencyAnnual Report. → Published since 1947.

U.S. securitiesandexchangecommissionAnnual Report. → Published since 1934/1935.

U.S. securitiesandexchangecommissionStatistical Bulletin. → Published since 1942.

watts, harold W.; and tobin, james 1960 Consumer Expenditures and the Capital Account. Volume 2, pages 1–48 in Conference on Consumption and Saving, University of Pennsylvania, 1959, Proceedings. Edited by Irwin Friend and Robert Jones. Philadelphia: Univ. of Pennsylvania Press.

whyte, william H. jr. (1954) 1955 The Web of Word of Mouth. Pages 113–122 in Consumer Behavior. Volume 2: The Life Cycle and Consumer Behavior. Edited by Lincoln Clark. New York Univ. Press.


In recent decades large masses of consumers in the Western world have moved into relatively affluent and secure positions. Their increased purchasing power has given them the opportunity to embroider upon basic needs with a sense of individual taste and creativity, as they search for a style of life rather than for security. Style of life, in this case, refers to the conscious and carefully developed sets or patterns of individual preferences in personal consumer behavior. Increases in disposable income and leisure time have permitted types of personal explorations that have made possible the rise of huge new industries devoted to cultural, recreational, and sports products. At the same time political ideologies based on conflict between the economic classes have all but disappeared (Bell 1959), and minority groups are looked at with almost as much interest in their innovations in consumption (for example, how they dress and play) as in their political views (Kluckhohn 1958). Because the potential variation in consumer behavior is now quite suddenly at a level far over and above that which was once economically dictated, the consumer's search for a personalized style of life becomes mankind's first large-scale nongovernmental, spontaneous, and awkward groping with the problem of economic freedom. The idea of this type of freedom has been appreciated in many lands, even in politically unfree lands where affluence and security do not exist. While it is too early to say what political consequences may emanate from the new consumer ideal, the apparently vast possibilities open to consumer exploration now demand the serious attention of the social sciences in a way that they did not before.

Some social scientists who might otherwise be interested in studying the consumer have been held back by the “silly” effects of initial affluence, an affluence which may permit such strange imbalances in family spending patterns as “living on beans” in order to afford a Cadillac. Yet in more mature forms affluence permits wiser shopping, for example, the ability to shop at a greater number of supermarkets to take advantage of price cuts and sales without the hindrance of a slavish single-store ”loyalty” (Tate 1961), the opportunity generally to obtain better and fairer prices (Caplovitz 1963), and a lesser susceptibility to advertising (Smith 1965). The same selection factors apply to style of life. Here, too, early exploration may produce bizarre or “silly” effects and combinations, but ultimately, with experience, comes more mature behavior.

Other social scientists who would seriously study the consumer have been held back by a distaste for what seemed to be tendencies toward local conformity in style, not realizing, for example, that the ability to flock together into homogeneous and “conforming” communities may itself be an expression of economic freedom. Furthermore, as changes in style are required by further changes in the cycle of family development, these same consumers flock easily and freely into other but different types of again homogeneous communities. That certain types of life style are now made more visible by the greater freedom for similar people to congregate may suggest an appearance of strong group conformity that persists even in the face of the individual consumer's easy, though less noticed, ability to shift from one4 style to another.

It is not enough to say that people have more time and more choice. The striking thing is that they are tending not to buy more and more goods as a form of conspicuous consumption, but to buy time to experiment more with their choices. The most conspicuous form of time consumption is the vacation trip. At the same time employees seem eager to receive leisure hours as a substitute for wage dollars. They bid against their employers for their own time, and their time budgets are becoming more important and more worthy of study than their money budgets [seeWages].

While these tendencies have appeared primarily in those areas of Western affluence which initially made the new consumer innovation possible, their study is also of importance to the less affluent lands of the East. That consumer freedom happened first in the “rich” West does not make it irrelevant to the East. Hopefully, the spirit of that freedom may be built into economic advances in the East and still newer and equally stimulating styles of life may emerge to compete creatively with those of the West. If anticipated, it may not be necessary to attain Western levels of affluence before the equivalent ideas take hold. For example, it may be possible during the stage of first or early affluence to leapfrog, or at least to inhibit, the consumer tendencies to spend more in terms of class and status considerations and to arrive more quickly at a point where spending reflects the greater freedom of individual style. In the West, it has happened without forethought or warning and even while some poverty remains. The release from serious economic anxiety has permitted many consumers, although not swamped with wealth, to look around for the first time with a sense of real choice about how they might spend their time, what kind of life they would make for themselves, and what kind of people they might be.

Conceptions of the consumer

Professional or official definitions of the study of consumer behavior are somewhat less focused. For example, the division of consumer psychology of the American Psychological Association defines its purpose as the study of human behavior as it relates to the consumption of goods and the uses and acceptance of services. The definition is broad enough to serve as well for a sociological or anthropological association. This breadth may reflect the newness of the concern with professional study of the consumer. That is, behavioral scientists, especially psychologists, have entered into the study of matters related to commerce and industry only in this century, a century which has seen the economy of the United States move from an orientation based primarily on production to one that is based as well on sales and distribution, and finally, and still in the process, to one that is also concerned with consumer needs.

In between the two world wars there developed what might be called the older orientation to consumer psychology, one that was based primarily on an identification with the businessman, the seller of goods and services. With this identification came early emphasis on sales research and studies of advertising effectiveness. To be sure, the latter coincided well with the psychologist's special ability to conduct experimental research on the learning of information, and, by so doing, an enormous research literature developed in advertising psychology (Lucas & Britt 1950). Nevertheless, the new definition of consumer psychology is based not so much on an identification with the manufacturer (as is common in industrial psychology), not so much on an identification with the salesman (as is common in advertising psychology), but more on an identification with the changing consumer himself.

As the consumer begins to innovate in life style, a change is also taking place in how he is viewed by social scientists. The old way was to view him, even define him, predominantly through the eyes of economics. The term “consumer” was not then a social science term, but a focus for concern over efficiency and for a harmonious balance of forces in the free market place. In response to the new affluence the old way of thinking was reflected in a swing from concern with alleged consumer wastefulness to concern with advertising's alleged over-persuasiveness. However, once Packard (1957) and similarly inclined critics had clearly and widely raised the alarm that consumers were being made to buy more than they actually needed, the older type of thinking had to be abandoned as a serious approach, that is, it had no further comments to make or questions to ask about what was happening to consumers as people.

While proponents of the older view may acknowledge that the purchase of household appliances saves the time and attention of family members for performance of higher level activities and that consumers spend more on such living aids as they become more affluent, it comes as something of a surprise for them to discover that the proportion of consumer incomes which is spent after taxes, at least in the United States and except for a postWorld War buying spree, has remained relatively constant at 92 to 94 per cent (White 1961). Great changes in consumer behavior have occurred, therefore, during a period when, strictly speaking, changes in economic behavior did not. The more serious concern, then, of the social sciences is to chart the social consequences of consumer access to these new living aids and resources.

Theories of consumer behavior. Ordinarily, when an area of human behavior is marked off for study by social scientists, one may expect much employment of comparative or experimental analyses to uncover and identify regularities in behavior within the area. These then provide a basis for understanding, as well as for testing the understanding. The tests usually are set up in the form of prediction studies. This testing is efficiently accelerated by a clear outline and definition of the area to be studied. One of the functions of theory is to pose a persistent set of questions and thereby to define what is considered important. However, there is not now a general theory of consumer behavior, although Katona (1951) has developed a theory of the role of consumer expectations or confidence upon the levels of spending and saving in the economy. A general theory of consumer behavior has probably not existed since Veblen examined the psychological pressures of social class for their impact on the efficiency of human consumption of goods and resources. From Veblen to Packard, a half century saw no marked change in this economic-psychological approach, except that Packard identifies the cause of consumer inefficiencies not so much in social class pressures per se but in their exploitation and manipulation by advertising and business interests. Both Veblen and Packard are essentially social critics using psychological insights to advance a theory which is basically economic, wherein utility and efficiency are defined as the important qualities.

Any theory based on economics might have deserved to be considered quite seriously in a day when economic indicators were the best predictors of human behavior. Today, this is less so. Income is no longer the best predictor of consumer purchasing; for example, a time series correlation between the data of 17 University of Michigan consumer surveys from 1952 to 1960 and the level of national expenditures for durable goods showed that estimates based on both recent disposable income and attitudes approximate future durable goods expenditures much more closely than the estimates based on disposable income alone. An analysis of variance indicated that consumer demand was more a function of willingness to buy than of ability to buy. In effect, Katona (1960) has turned the tables on economics by showing how consumer attitudes of optimism or pessimism initiate changes in economic indicators. The causal flow has been somewhat reversed, therefore, first in observation and then in theoretical formulations. If this is so, if economic predictors are no longer all powerful, what new indicators should we be looking at more closely?

Predictors of consumer behavior

Perhaps stages in the family life cycle are the best predictors of consumer behavior (Clark & Foote 1955–1961). Is the family a childless couple? Is there a young baby in the family? Have the children grown and left? [SeeLife cycle.]

In addition there are the many household buying decisions made within one particular stage of the cycle. What predictors seem to work here for the shorter run? One set seems to involve the career roles brought into the marital relationship by the two partners. This is partially a matter of relative buying influence within the pair, as affected by such simple considerations as whether or not the wife earns income outside the home. More important however is the matter of what particular occupations and specific career commitments may be brought together within the pair. Every occupation, especially if it has the lifelong involvement of a career, sets forth certain style of life requirements based on where the work is performed, the standards for success, who are the co-workers (clients, colleagues, supervisors), etc. Next to family cycle it is probably industry itself which has the most pervasive, though less direct, influence on style of life. It is important to note, however, that in this sense occupation and profession are not economic but social variables, involving different types of networks of interpersonal relationships much more than different levels of income.

Family and jobs make good starting places for understanding consumers as whole people. Strangely, industry does not yet seek this whole view but tends to spend vast sums on the study of particular consumer functions (for example, food preparation) and closely related products. Yet some sociologists have looked at the new consumer life styles, have asked what products consumers are buying, and have come up with a whole view of the consumer which has relevance to all industry. Riesman and Roseborough (1955), for example, have said that today's young people learn to want a recognizable complex of goods and services which they call “the standard package.” This may include a college education, car, gadget-equipped home, and annual travel vacations as base elements. The package may come in a series of standard through de luxe grades, and with the aid of credit progress through those grades can be made easy. Such a view does not, as some have inferred, suggest increased conformity, for obtaining the elements of the package is intimately related to the acquisition of mobility and freedom, that is, greater job choice, reduction of household chores, and increased facilities for and use of transportation. It would seem, however, that as something like a standard package becomes universally wanted, especially for the freedom that is seen in it, then deprivation becomes less tolerable for those without jobs and/or education. On the level of individual protest it may lead to delinquency and crime, but on the level of mass protest it may lead to an insistence on educational and job rights for underprivileged or minority groups. Thus do matters of consumer style shade over into social and political issues.

The matter of social issues raises questions of intensity and amount. For example, to what extent are consumers motivated by the standard package? And how de luxe or how large may this package become? At present we can only cite Chinoy's (1955) study of the American automobile worker, which concludes that satisfactions are reported in terms of a small package of independence represented in home and car ownership, with aspirations for further advancement focused on the children and their chances for social mobility through education. While this modest version of the dream may apply only to industrial workers, the emphasis is more on having some of the good things of life and the time to use them and less on the ideologically distant goals. While it is true that in the United States a blue-collar majority is giving way rapidly to a “middle-class” white-collar majority, a similar evaluation has not yet been made for the latter group. This latter group, of course, is exposed to a different side of American industry and is more likely to work in office buildings than in factories. The standards for home ownership and decor of this group may therefore be different and possibly higher than that of the blue-collar group.

The question of what kind of houses people want and what they want in them is a good over-all way of assessing the consumer market. For example, the number of new homes for which construction has started each year is the best single indication of whether or not it will be a good year for business. The question of what kinds and numbers of homes to build, and where to build them, becomes a critical determiner of both national economic well-being and of the consumer's physical environment, complete with built-in limitations and opportunities for style development. Thus, the home architects, the appliance designers, the federal housing laws, and the bank credit policies all have significant effects on the potentialities of this environment.

It should also be noted that new houses reflect consumer demand and initiative, or put another way, consumer investment. As Katona (1960) emphasizes, the growth and expansion of the economy are dependent on consumer investment as well as business investment, and these are equally important forces in the extent to which they can stimulate or retard the economy.

Research approaches

With the consumer now seen as so “powerful,” it is to be expected that attempts to study him have become intensified. One reviewer (Halbert 1965) has identified six different general approaches:

(1) General studies of consumer behavior: the role of consumers and consumption, consumer spending as investments, consumption in economic development.

(2) Income studies: changes in income distribution, credit, and expenditures by various classes. This category accounts for a large part of past studies.

(3) Population studies: population trends and shifts, life-cycle influences, various demographic classifications.

(4) Life-style factors: studies of factors influencing styles of living, such as purchases, status symbols, social class, images, needs, leisure, convenience. This area is receiving increasing emphasis.

(5) Consumers as decision makers: studies of who makes what decisions, what decisions are planned, who is influential in various situations, the household as a decision unit.

(6) Specific purchase and consumption studies: studies of purchases and determinants of purchases for specific products such as coffee, cleansers, gasoline.

The first three of these approaches are dependent upon availability and analysis of mass data. While the federal and state governments have long made such data available, it is only recently that correspondingly massive tools of analysis have become available (for example, the computer). This means that research in these areas was until recently primarily reportorial. With so much detailed data to comprehend it was all that scholars could do to keep up with descriptive identifications of major historical trends. The computer brings a new type of inquiry to such data—an inquiry that seeks to build dynamic theory and is not daunted by complexities in the manipulation of data. One may expect therefore to hear more in future years, from investigators with a primarily methodological orientation, about the science of retailing, of marketing, and of other fields.

The fourth and fifth approaches seem to represent more clearly the opening up of consumer study to sociologists and psychologists. Traditionally, the former have studied the ills of society, such as divorce and unemployment. Psychologists, too, have been involved, if not with society's ills, then with the emotionally ill. However, as the current generation of social scientists has put behind it depression and World War II, and has also participated in the new affluence, there has gradually developed a new curiosity about the nature of the normal personality, a shift away from concern with irrational processes to rational thought, and an interest in theories about how various everyday decisions are made.

All of these areas of research involve a move away from descriptive or reportorial research to inquiry guided by the use of theory and undertaken for the purpose of further theory development. This means that the study of the consumer has now finally linked up with the main body of the behavioral sciences.

As for the consumer himself, we may expect further inroads on his ancient image of himself as a person in want and as one who is preoccupied with the allocation of scarce means to satisfy basic needs. Foote (1963) proposes that as the proportion of household expenditures devoted to food continues to fall, today's somewhat obese consumer may well develop wants that do not require production or consumption, such as stimulating conversation or the creation of music. The constraint on wants such as these will not be income but learning ability. In short, consumers may yet outgrow consumption.

Herbert E. Krugman

[See alsoAdvertising, article onAdvertising Research; marketresearch, article onconsumerresearch. Other relevant material may be found inconformity; identity, psychosocial; leisure.]


bell, daniel 1959 The End of Ideology. Glencoe, 111.:

Free Press. → A paperback edition was published in 1961 by Collier.

caplovitz, david 1963 The Poor Pay More: Consumer Practices of Low-income Families. New York: Free Press.

chinoy, ely 1955 Automobile Workers and the American Dream. Garden City, N.Y.: Doubleday.

clark, lincoln H.; and foote, nelson N. (editors)

1955–1961 Consumer Behavior. 4 vols. New York Univ. Press. → Volume 1: The Dynamics of Consumer Reaction, 1955. Volume 2: The Life Cycle and Consumer Behavior, 1955. Volume 3: Research on Consumer Reactions, 1958. Volume 4: Household Decisionmaking, 1961.

foote, nelson N. 1963 The Image of the Consumer in the Year 2000. Pages 13–18 in Boston Conference on Distribution, Thirty-fifth Annual, 1963. Chestnut Hill, Mass.: Boston College.

halbert, michael 1965 The Meaning and Sources of Marketing Theory. New York: McGraw-Hill.

katona, george 1951 Psychological Analysis of Economic Behavior. New York: McGraw-Hill. katona, george 1960 The Powerful Consumer: Psychological Studies of the American Economy. New York: McGraw-Hill.

kluckhohn, C. 1958 Have There Been Discernible

Shifts in American Values During the Past Generation? Pages 145–217 in Elting E. Morison (editor), The American Style: Essays in Value and Performance. A Report on the Dedham Conference of May 23–27, 1957. New York: Harper. → See especially page 197.

lucas, darrell B.; and britt, steuart H. 1950 Advertising Psychology and Research: An Introductory Book. New York: McGraw-Hill.

packard, vance O. 1957 The Hidden Persuaders. New

York: McKay. → A paperback edition was published in 1958 by Pocket Books.

riesman, david; and roseborough, howard 1955 Careers and Consumer Behavior. Pages 1–18 in Lincoln H. Clark and Nelson N. Foote (editors), Consumer Behavior. Volume 2: The Life Cycle and Consumer Behavior. New York Univ. Press.

smith, stewart A. 1965 Criteria for Media Comparisons: A Critique. Journal of Marketing Research 4: 364–369.

tate, russell S. 1961 The Supermarket Battle for

Store Loyalty. Journal of Marketing 25:8–13.

white, winston 1961 Beyond Conformity. New York:

Free Press. → See especially Chapter 8.