nity basis are now tabulated with separate returns of husbands and wives. Separate returns of husbands and wives do not include joint returns, Form 1040A, on which the collector determined the minimum tax on the basis of separate incomes of husband and wife. Unequal numbers of returns for men and for women result from insufficient information to identify the marital status and sex of taxpayers and from the use of samples as a basis for compiling statistical data.

Returns of single persons are returns of unmarried individuals, including individuals divorced or legally separated on or before the close of the year.

EXEMPTIONS

Exemptions are allowed as a credit against income for the purpose of computing both normal tax and surtax. A per capita exemption of $600 is allowed for the taxpayer, his spouse, and each closely related dependent (specified below) who received more than one-half of his support from the taxpayer and who had less than $600 of gross income for the year, together with the additional exemptions of $600 for blindness and $600 for age 65 or over of the taxpayer and/or his spouse.

Exemption for dependents is allowable for the following close relatives: son or daughter (including legally adopted children) or descendent of either, stepchild; brother, sister, stepbrother, stepsister, half brother, half sister; parent, grandparent, or other direct ancestor; stepfather or stepmother; father-in-law, mother-in-law, brother-in-law, sister-in-law, son-inlaw, or daughter-in-law; and uncle, aunt, nephew, or niece if related by blood; provided that the dependent is a citizen or resident of the United States, or a resident of Canada or Mexico. Dependents meeting the above requirements need not be under 18 years of age.

The amount of exemption claimed is tabulated in table 2 by adjusted gross income classes, in table 6 by types of tax, and in table 8 by marital status of taxpayer. Table 9 shows the total number of exemptions claimed, a frequency of returns showing additional exemptions for age and blindness together with the number of additional exemptions taken, and the number of exemptions other than age or blindness; also, there is a frequency distribution of returns by number of exemptions other than age and blindness.

Both the number and the amount of exemptions tabulated include the exemptions automatically allowed through use of the optional tax table, on returns Form 1040A and short-form 1040, as well as exemptions on returns of taxpayers who compute their tax liability. Slight duplication of exemptions exists because of dependents who earned less than $600 of wages subject to withholding of income tax and who filed a return as the most convenient way of claiming the tax refund; such wages are not taxable to the dependent, nor do they constitute a part of the income of the taxpayer claiming the dependent. Exemptions from both returns are

NET GAIN OR LOSS FROM SALES OR EXCHANGES OF CAPITAL ASSETS

Net gain or loss from sales of capital assets is derived from sales or exchanges of property defined in section 117 of the 1939 Code as capital assets. The term, capital asset, means property held by the taxpayer (whether or not connected with his trade or business), but does not include (1) stock in trade or other property which would properly be included in inventory if on hand at the close of the income year, or property held primarily for sale to customers in the ordinary course of trade or business, (2) property used in trade or business of a character which is subject to the allowance for depreciation, or real property used in trade or business, (3) an obligation of the United States or any possession thereof, or of a State or Territory or political subdivision thereof, or the District of Columbia, issued on a discount basis and payable without interest at a fixed maturity date not exceeding 1 year from date of issue, or (4) a copyright, a literary, musical, or artistic composition, or similar property created by the tax

payer.

If bonds in registered or coupon form and corporate stocks become worthless during the year and are capital assets, the loss therefrom is considered a loss from the sale of capital assets; also, nonbusiness bad debts which become totally worthless within the year are considered a loss from sale of capital assets held not more than 6 months; and certain distributions under employees' trust plans, as specified under section 165 of the 1939 Code, to the extent that the distributions exceed the amount contributed by the employee, are considered a gain from sale of capital assets held more than 6 months.

For the purpose of computing net gain or loss from sales or exchanges of capital assets, distinction is made between short- and long-term capital gain and loss and different rules are applied for the treatment of each. The distinction between short- and long-term gain and loss is based on the length of time that the asset is held before the sale or exchange. Short-term applies to the gain or loss resulting from the sale or exchange of a capital asset held for not more than 6 months, and 100 percent of the recognized gain or loss thereon is taken into account in computing net short-term capital gain or loss. Long-term applies to the gain or loss resulting from the sale or exchange of a capital asset held for more than 6 months, and 50 percent of the recognized gain or loss is taken into account in computing net long-term capital gain or loss. The net short- and longterm capital gain or loss include the net short- and long-term capital gain and loss received through partnerships.

Under certain circumstances, gain or loss from the sale of property which is not a capital asset may be treated as gain or loss from the sale of capital assets. Such gain or loss includes that from the sale of land and

ting of timber or the disposal of timber or coal under contract, if held more than 6 months, and from the sale of certain livestock held 12 months or more and from unharvested crops sold with land which has been held more than 6 months. If the recognized gains upon sales or exchanges of such property plus the recognized gains from compulsory conversion (through seizure, condemnation, destruction, fire, or theft) of property used in business and of capital assets held more than 6 months exceed the recognized losses from such sales, exchanges, and conversions, then such gains and losses are considered as gains and losses from sales of capital assets held more than 6 months. However, if such gains do not exceed such losses, then such gains and losses are not considered as gains and losses from sales of capital assets; but each gain is fully reported and each loss, if allowable at all, is déductible in full.

The capital loss carryover provision allows the amount of “net capital loss" of any income year to be carried forward as a short-term capital loss in each of the 5 succeeding years to the extent that it exceeds any "net capital gains" of subsequent years intervening between the year in which the net capital loss is sustained and the year to which it is carried. If there are net capital losses carried over from more than 1 year, they are allowed in the order in which they arose. The net capital loss for any year, to be used as a capital loss carryover, is the excess of (1) current year losses from sales of capital assets over (2) the sum of current year gains from sales of capital assets and the smaller of (a) $1,000 or (b) net income (adjusted gross income, if tax is determined from tax table) computed without regard to gains or losses from sales of capital assets. For the purpose of computing the net capital loss carryover, net capital gain for any year is the excess of (1) current year gains from sales of capital assets plus the smaller of (a) $1,000 or (b) net income (adjusted gross income, if tax is determined from tax table) computed without regard to capital gains or losses over (2) current year losses from sales of capital assets.

Capital loss carryover reported on the 1951 returns is a combination of the 1950 net capital loss and the remaining capital loss carryovers from 1946-49 not eliminated by net capital gains of the succeeding years 1947-50. On returns showing a net gain from sales of capital assets in gross income, the carryover is entirely offset by current year capital gains. On returns showing deduction for net loss from sales of capital assets, the carryover is, in some cases, partially offset and, in others, wholly offset by current year capital gains and/or the allowable deduction not exceeding $1,000; but in cases where the current year capital loss exceeds the $1,000 deduction, no part of the carryover is offset.

Net gain from sales of capital assets reported for the computation of adjusted gross income for 1951 is a combination of the net short-term capital gain or loss

gain or loss together with the capital loss carryovers from 1946-50, inclusive.

Net loss from sales of capital assets reported as a deduction from gross income is the statutory amount deductible. The statutory deduction for a net loss from sales of capital assets is limited to the amount of such loss, or to the net income (adjusted gross income, if tax is determined from tax table) computed without regard to capital gains or losses, or to $1,000, whichever is smallest. Net loss from sales of capital assets before application of the limitation is a combination of net short-term capital gain or loss and 50 percent of the recognized net long-term capital gain or loss together with the capital loss carryovers from 1946-50, inclusive.

The amount of net gain from sales of capital assets and the statutory deduction for net loss from such sales are tabulated in table 2, among the sources comprising adjusted gross income, along with their frequency of

occurrence.

Other significant data, reported in schedule D(1), concerning sales of capital assets are to be found in table 11, wherein the returns with statutory deduction for a net loss from sales of capital assets are tabulated separately from returns with net gain from such sales. Here are shown the net short-term capital gain, net short-term capital loss, net long-term capital gain, net long-term capital loss, and the capital loss carryover from 1946-50. Inasmuch as the carryover is reported independently from the short-term gain or loss in schedule D, the net short-term gain and net short-term loss are the results of 1951 sales. The net long-term gain and net long-term loss are the amounts to be taken into account, that is, reduced 50 percent. Also, for returns with net loss from sales of capital assets, there is shown the net loss from sales of capital assets before the statutory limitation.

Returns with net gain from sales of capital assets are subdivided to show returns with alternative tax apart from returns with normal tax and surtax. Returns with self-employment tax only are not shown separately but are included in the total for returns with net. gain from sales of capital assets. For returns with alternative tax, the approximate amount of the excess of net long-term capital gain over net short-term capital loss, which is taxed at the 50 percent alternative rate, is computed from data in the table by combining the net long-term capital gain and the net short-term capital loss. However, because the carryover is not merged with short-term gain or loss, this arbitrary method results in a slight overstatement of the amount subject to the alternative rate on returns where a carryover was combined with a short-term loss to determine the excess of long-term gain, or where a carryover exceeded the short-term gain resulting in a short-term loss which was used to determine the excess, or where there was no short-term gain or loss but a carryover

DATA FOR STATES AND TERRITORIES

Although selected data for individual returns are tabulated by States and Territories, the distribution does not represent a precise geographic distribution of the income reported. There is no way to determine from the income tax return the amount of income originating in a particular State or Territory. An individual may file his return in the collection district in which he resides or in the district in which his principal place of business is located, but the income reported may originate outside the State or Territory in which the district is located. The segregation of returns by States and Territories is based entirely on the location of the collection district in which the return is filed and does not necessarily indicate the area from which the income was procured.

The number of returns, amount of salaries and wages, dividends, interest, adjusted gross income, income tax liability, and self-employment tax are tabulated in table 12, in aggregate for each State and Territory. The number of returns, adjusted gross income, and income tax liability for each State and Territory are shown in table 13, by adjusted gross income classes, the intervals of which are established especially for this table. Taxable and nontaxable returns are combined for these two tabulations but the returns with no adjusted gross income are not included. Data for returns with a District of Columbia address are tabulated separately although filed in Maryland. Data for returns from Alaska are included in the data for Washington. See the discussion on pages 19-21, for description of samples and method used in compiling statistical data and for an explanation of the variation between data in these two tables and that in the national distributions.

SOLE PROPRIETORSHIPS

Data concerning the 1951 business and professional income of taxpayers each of whom owns his business are tabulated in tables 14, 15, and 16. Sole proprietorship data are tabulated from schedule C accompanying Form 1040, from Form 1040F, and from any other business schedule submitted by the taxpayer to support the business profit or loss reported in his income. Business data for returns with adjusted gross income under $50,000 are estimated from data contained in the business schedules attached to returns in the samples selected from the various strata. The samples are explained in the description of the sample on pages 19–21. This commentary on the sample and limitation of data give important information respecting sole proprietorship data and should be read in connection with the use of these tables.

Table 14 shows by selected industrial groups the total number of businesses and the total receipts for all businesses and, also, the number, receipts, and net profit for businesses showing a net profit. Table 15 presents the same three items by size of total receipts and, for

industrial groups but the businesses with net loss are in aggregate. In table 16, there is a frequency distribution of the number of businesses with net profit by the size of net profit.

Industrial groups.-The classification of business activity by industrial groups is based on the nature of business as stated by the sole proprietor in his business schedules. Where the taxpayer owns two or more businesses of the same kind and reports them on separate schedules, these like businesses are combined and considered one business for classification and tabulation. If the taxpayer is engaged in more than one kind of business, each kind of business is classified independently for its respective industrial activity. There are nine basic industrial groups-agriculture, mining, construction, manufacturing, public utilities, trade, finance, service, and business not allocable. Each basic group has subgroups, but only selected subgroups are presented in the tables.

Total receipts.-These receipts comprise all income from profession or business activity of the owner and include miscellaneous business income as well as gross receipts from sales and services. Details as to the sources of business receipts are not required in the business schedule. If the sole proprietor fails to submit a business schedule, or the schedule is lacking for any reason, the amount of total receipts is not available; thus total receipts are understated by an indeterminable amount. There are 47,165 businesses with net loss for which total receipts are not available. The number of businesses with net profit for which total receipts are lacking is not known, inasmuch as the amount of net profit is substituted for total receipts for purposes of these tabulations. The size of total receipts is based on the amount of receipts tabulated for each kind of business reported.

Current year net profit or loss.-The net profit or net loss is the net result of all transactions from each kind of business owned. If different kinds of businesses are reported by the taxpayer, the net profit or the net loss from each kind of business is tabulated separately. If two or more businesses of the same kind are owned, they are combined as one business and only the net result, either net profit or net loss, is tabulated. The size of net profit is based on the net profit tabulated for each kind of business owned by the sole proprietor.

The number of businesses reported by individuals who are sole proprietors, the total receipts from these business activities, and the net profit or loss derived therefrom are:

119, 897, 053 18, 377, 038

As shown above, data are tabulated for 7,339,811 businesses which are owned by individuals as sole proprietors. Each kind of business which the taxpayer owns is counted and classified independently. If more than one business of the same kind is owned, data are combined and the activity counted and classified as one business. In case of a business profit or loss divided between spouses on a community income basis, each kind of business activity is considered as a whole, without regard to the community division, and is counted and classified as one business.

The total number of businesses owned differs from the frequency distribution of returns showing a business profit or a business loss in table 2. In this table, there is a total of 7,175,342 returns on which a net profit or a net loss from business occurs. For this frequency distribution of returns, the net profit or net loss reported by the taxpayer who owns more than one kind of business is the net result of his multiple business activities and is a frequency of one; but, in the case of a net profit or net loss from business divided between spouses on a community income basis, each spouse reports on his respective return his share of the net result from all business activities and, consequently, the frequency is two.

Since the number of businesses owned by sole proprietors is greater than the frequency of returns showing business profit or loss, this indicates that there are a greater number of taxpayers with multiple businesses than there are of taxpayers who divide business income on a community income basis. No data are available regarding the number of instances in which multiple businesses occur nor the number of community property returns on which the business is divided between spouses.

DESCRIPTION OF THE SAMPLE AND LIMITATIONS OF DATA

The statistical program for individual income tax returns filed for the tax year 1951 called for the usual extensive tabulation of data on a nationwide basis and for limited data on a statewide basis. In addition it called for a series of distributions relating to sole proprietorship data reported in Schedule C (or Form 1040F) of the Form 1040 return. A probability sample of individual income tax returns was employed which provided reliable estimates of data without experiencing the cost and delay of tabulating all returns.

A stratified sample was prescribed based on the method of segregating returns used by the internal revenue collectors' offices in their administrative processing of returns. This segregation was by type of return and by size of adjusted gross income reported, which provided an effective basis for sample stratification. Once the strata were established, the same sample rates for each stratum were used to select returns in each collectors' office. The method of sample selection approx

Sample selection.-Before sample selection, returns in each sampling stratum were blocked in units of 100 returns and assigned consecutive serial numbers from 00 to 99 within each unit. All returns with adjusted gross income under $25,000 were sampled in the field on the basis of the serial numbers assigned; all returns with adjusted gross income $25,000 or more were sent in to Washington for sampling.

Form 1040A returns were sampled at the rate of 0.3 percent, by withdrawing the first return from 3 specified blocks of every successive 10. Collector returns, Form 1040, were sampled in the field at the rate of 0.9 percent; the first 3 returns from 3 specified blocks of every successive 10 were selected in the sample. This 0.9 percent sample was reduced to a 0.3 percent sample in Washington by eliminating, for purposes of the national and State distributions, the second and third returns obtained from each sample block. The latter returns, however, were utilized for purposes of the sole proprietorship distributions.

A 10-percent sample of Form 1040 returns with adjusted gross income under $25,000, was selected in the field by withdrawing the first 10 returns from each

All blocks of agent returns, Form 1040, with adjusted gross income $25,000 or more were received in Washington. Twenty-five serial numbers, appropriately spaced, were designated and all returns bearing such serial numbers and having adjusted gross income $25,000 to $50,000 were withdrawn from each block for the sample. A count of the remaining 75 percent of returns excluded from the sample with adjusted gross income $25,000 to $50,000 was made to provide an independent population for weighting purposes.

A 100-percent sample of returns with adjusted gross income $50,000 or more, was accomplished by selecting all such returns in above-mentioned blocks of agent returns, Form 1040.

Population sizes and weighting procedures.-The primary sources of population data for 1951 were statements submitted by the 64 collectors' offices showing the numbers of Form 1040A, collector returns, Form 1040, and agent returns, Form 1040, filed. Since all agent returns, Form 1040, with adjusted gross income $25,000 and over were received and counted in Washington, these were subtracted from the number of agent returns submitted by each collector to derive the populations with respect to Form 1040 returns with adjusted gross income under $25,000.

Separate systems of weighting were used for the national tabulations and for the State tabulations. The weights for the national tabulations were based on nationwide stratum populations obtained by summing the stratum populations reported by the 64 collection districts. The separate collection office stratum populations provided the basis for independent collection office weights for the State tabulations.

Although the sampling pattern for 1951 called for many distinct strata for sample selection and population determination purposes, it was possible to achieve a substantial degree of simplification in the tabulating and weighting operations by combining multiple strata, into the following five groups: (1) Form 1040A, (2) collector returns, Form 1040, (3) agent returns, Form

If the number of returns in a cell of a table is

1040, with adjusted gross income under $25,000 (4) agent returns, Form 1040, with adjusted gross income from $25,000 to $50,000; and (5) agent returns, Form 1040, with adjusted gross income $50,000 or more.

Table A presents, for each of the five estimating strata, the number of returns in the population, the number of returns in the sample, the prescribed sampling rate, and the actual sampling rate.

Table A.-INDIVIDUAL RETURNS FOR 1951: NUMBER OF RETURNS FILED AND NUMBER OF RETURNS IN SAMPLE BY ESTIMATING STRATA

Sampling variability.-Because the data in this volume are tabulated from a sample of returns filed, they are subject to sampling variability. The actual and relative amounts of sampling variability applicable to specific frequencies for each of three levels of adjusted gross income are shown in table B. No sampling variabilities are shown for money amounts. Each of the various income areas constituting an independent estimating stratum has its own variability pattern; accordingly, the relative errors are presented separately for three distinct income areas. Returns in the adjusted gross income area under $8,000 are a composite group of collector returns contributing more than 99 percent of the total population and agent returns contributing less than 1 percent. Relative errors for this group are based on the collector returns component since the effect on relative error of agent returns with adjusted gross income under $8,000 is almost negligible.