INTRODUCTION

The 1950 Census of Population obtained information on the money income of individuals and families in 1949 on the basis of a 20-percent sample. The tabulations of these data as published in regular reports of the Census of Population show separate income distributions for families and unrelated individuals in the urban, rural -non farm, and rural- farm populations. But farm operators and their families, defined in terms of the definition of a farm established for the 1950 Census of Agriculture, could not in many cases be identified as such from information available from the Population Census alone. This identification could be made and an income distribution established for all farm-operator families only by combining information obtained in the Agriculture and Population Censuses.

In the Population Census, information was requested of persons 14 years of age and over in the sample on the following income categories: (1) The amount of money wages or salary received in 1949; (2) the amount of net money income received from self-employment in 1949 (farm, business, or professional practice); and (3) the amount of other income received in 1949, such as interest, dividends, veterans' allowances, pensions, or rents. If the person was the head of a family, these three questions were repeated for the other family members as a group to obtain the income of the whole family.

By identifying heads of families in the 20-percent sample of the Population Census who were operators of farms included in the Census of Agriculture, and by combining information from the the two Censuses, family-income data from the Population Census were tabulated for farm-operator families in relation to the economic classification of farms determined from the Agriculture Census.

LIMITATIONS OF THE DATA

The figures on income, as indicated in regular reports of the Census of Population, are subject to errors of response and nonreporting which probably lead to some understatement of total income. The schedule entries are generally based not on records but on memory. The memory factor in data derived from field surveys of income probably produces underestimates because the tendency is to forget minor or irregular sources of income. Other errors of reporting are due to misunderstanding of the income questions or to misrepresentation.

Another possible source of understatement is the assumption made in the editing process that there was no other income in the family when only the income of the head of the family was reported, This assumption was adopted for about 5 percent of the families in order to make maximum use of the information obtained. In most of the fully reported cases, the head's income constituted all or most of the total family income.

These qualifications apply to all the income data collected in the 1950 Census of Population, including those here presented for farm-operator families. There are three additional known limitations in the data for farm-operator families. First, there is evidence of some nonreporting of farm income, total family income having been recorded in a significant number of cases clusive of income received from the farm operated. The evidence is given in table 3, and rough adjustments have been made in tables 4 and 5 to allow for such nonreporting. However, no adjustments were feasible in the straight family income distributions of tables 1 and 2.

ex

The second limitation arose in the process of matching agriculture reports with their counterparts among the population reports. In a few cases the operator of a farm determined from the Agriculture Census, whose name appeared on a sample line of the Population Census, turned out not to be the family head. Total family income in such cases was not available, but the personal income of operators who were not family heads was counted as total family income in the income tabulations. In some of these cases,

the operator's income may have been the same as or a close approximation of total family income, but in other cases it most probably was not.

Separate analysis of this group shows a distribution of income considerably different from that of family heads and an estimated average reported income about 20 percent lower than the average income of families with operator heads. However, the number involved is so small-only 3 percent of the total - that their inclusion is not believed to have had any significant effect on the results. For example, the estimated average of all family incomes would not have been increased by more than one percent if operators who were not family heads had been excluded. The third limitation, also a minor one, arose from the fact that a small segment of the agriculture reports could not be matched with corresponding population reports. The expanded agriculture sample covered 5,380,000 farms, the population sample only 5, 341,000. The difference of 39,000 is primarily farms where the operator's residence was not in the same enumeration district as his farm and the matching population report could not be located. In non-resident cases for which the schedules were matched- that is, where the operator and his family lived in the same enumeration district though not on the farm itself- family incomes averaged from 20 to 25 percent higher than the incomes of farm-resident operator families. There is a strong probability, therefore, that family incomes for the unmatched cases were on the whole quite a bit higher than the general run of farm-operator family incomes. Again, however, the number involved is too small-less than one percent of the total-to have any significant effect on the over-all results.

SUMMARY DISTRIBUTIONS OF FAMILY INCOME

Subject, therefore, to the general qualification that the degree of understatement of income may be slightly greater for farm-operator families than for other families, the distribution of all farm-operator families in the United States by total money income is given in table 1 and compared with national distributions for the rural-farm and total populations taken from regular reports of the Census of Population. The latter also show income distributions of persons 14 years of age and over by major occupation group, including farmers and farm managers as one group. These represent persons whose principal occupation in the week preceding the census was farmer or farm manager, a less inclusive definition than that of farm operator in the Census of Agriculture. These income distributions, moreover, reflect the personal income of those farm operators included therein, not the total income of their families, and thus cannot be compared with the distribution of farm-operator families by total family income given in the first column of table 1.

An income distribution of families whose heads reported the major occupation of farmer or farm manager would be of some TABLE 1.-DISTRIBUTION OF FARE-OPERATOR FAMILIES, RURAL-FARH FANILIES, AND ALL FAMILIES BY TOTAL MONEY Income In 1949, FOR THE UNITED STATES

comparative interest, but such a distribution has not been tabulated from the 1950 Census data. Consequently, of the 1949 income distributions available from published Census reports, that for rural-farm families, given in the second column of table 1, seems to be most nearly comparable with the distribution for farm-operator families developed in this study. Most farmoperator families are rural-farm families. However, a few live on urban farms and a few others do not live on the farms they On the other hand, rural-fam operate or on any other farm. families would include some farm-laborer and other families who live on rural farms but do not operate them.

Despite these differences, farm-operator families and rural-farm families overlap to a considerable extent. As might be expected, therefore, the income distributions of the two groups are quite similar. Also as expected, because rural-farm families include some farm-laborer families, the median income of farm-operator families in 1949 was a little larger than that for rural-farm families-$1,867 as compared with $1,729.

The median income of operator families was only 61 percent as large as that for all families in the United States. This comparison reflects money income only, and nonmoney income is relatively much more important for farm-operator families than for others. But while the inclusion of nonmoney income would certainly have reduced the difference just cited, it would probably not have eliminated it.

It should be noted that about 4 percent of all farm- operator families as shown by this report for the United States are singleperson families-"unrelated individuals" in the usual terminology of income distributions. There is, therefore, some question as to whether income distributions for operator families should be compared with other distributions for families alone or for families plus un related individuals. However, the 4 percent of unrelated individuals included in the operator distribution compares with 11 percent for the total of rural-farm families and unrelated individuals and 22 percent for all families and unrelated individuals in the United States. In view of these wide differences in the relative importance of unrelated individuals, and considering the fact that their incomes are generally much lower than those of families, the more reasonable comparison would seem to be with income distributions for families alone.

The highest regional median income for farm-operator families in 1949 was $2,682 in the Western region (see table 2). At $2,438 in the North, it was not much lower. But the median income for operator families in the South was only $1, 284, about half that of the other two regions. The same regional order, and about the same relative differences, are evident in the median incomes for all families in the three regions- and also in the ratios of operator-family to all family incomes. The median income of operator families was 78 percent as large as the median for all families in the West, 73 percent in the North and 57 percent in the South. The income distribution for operator families in the South includes cropper families.

ANALYSIS OF FAMILY INCOME BY ECONOMIC CLASS OF FARM Tables 3-5 summarize the results of an analysis, for the United States as a whole, of the data on farm-operator family incomes in 1949 by economic class of farm. The various economic classes of farms as established in the 1950 Census of Agriculture are fully defined in terms of upper and lower limits for value of farm products sold (see table 4), except for class VI and parttime farms which are distinguished on the basis of criteria set forth in footnote 6 of table 4. This is the first time that detailed income data have been obtained for farm-operator families associated specifically with each economic class of farms.

Class I and class II farms are shown separately in tables 3-5 in order to bring out the marked differences in their income characteristics. They are lumped together in other tabulations in this report because for most other characteristics they are not significantly different and also because the size of the sample on a regional basis did not in some cases permit separate tabulations for class I farms.

"Abnormal" farms, lumped with part-time farms elsewhere in this report, are excluded entirely from tables 3-5. They include such categories as public and private institutional farms, community enterprises, experiment station farms, and grazing associations. Information was obtained in some cases on the wages or

salaries received by hired managers, but, in general, familyincome data are not applicable to farms of this type.

Table 3 shows the total number of farms in each economic class, the number for which total family income was recorded in the Population Census, and the percentages reporting each of the six sources of family income. Total family income was recorded for 90 percent of all farms, and this ratio did not vary materially among the different economic classes. On the other hand, there were significant variations in the percentages reporting each of the specified sources of income.

A third of all farm operators reported income from wages or salaries. However, only a fifth of the commercial operators reported such income whereas three-fifths of the noncommercial operators reported it. Within the commercial group, the percentage of operators reporting wages or salaries increased uniformly from class I to class VI with two exceptions: (1) The percentage for class I operators was greater than that for either class II or class III operators; and (2) the percentage for class VI operators was smaller than that for class V operators. By definition, a class VI operator did not work more than 99 days off his farm in 1949. In view of this limitation, it is worth noting that a fourth of all class VI operators reported some wage or salary income in 1949, a higher fraction than that for