ally no effect on the comparability of the figures. In particular, the ratio which the number of farms mortgaged bears to the total number should be almost exactly comparable from census to census.

By reference to Table 1 it may be noted that at the different censuses most of the farms operated by their owners were reported definitely in one of the two classes, "mortgaged," or "free from mortgage," while for a smaller number the mortgage status was not specified by the enumerator. It is believed that the failures to report as to mortgage indebtedness were due largely to two factors, more or less conflicting in their nature. In the first place, it seems likely that some of the instances of failure to report were due to the unwillingness of the farmer to disclose the fact that his farm was mortgaged. On the other hand, there were many cases where the enumerator apparently intended, by leaving the questions blank, to indicate that there was no mortgage debt, in spite of positive instructions calling for a definite answer either in the affirmative or in the negative.

In the reports of the 1910 census the number of mortgaged farms was presented both "as reported," and in the form of an estimated total, which included a percentage of the "unknown" farms equal to the percentage which mortgaged farms constituted of the total number with mortgage reports. By reason of the much larger proportion of farms for which no mortgage report was secured in 1920, however, it is believed that such estimates are hardly justified for that year. In the present report, therefore, all comparisons of numbers of farms mortgaged and free from mortgage are made on the basis of the mortgage status as reported.

Number of farms mortgaged: 1890 to 1920.-The number of farms operated by their owners in 1920 was 3,925,090. Of this number, 2,074,325, or 52.8 per cent, were definitely reported as free from mortgage indebtedness; 1,461,306, or 37.2 per cent, were definitely reported as mortgaged; and for the remaining 389,459 farms (9.9 per cent), the mortgage status was not specified by the enumerators. In 1910, 33.2 per cent of the owner-operated farms were reported as mortgaged; in 1900, 30 per cent; and in 1890, 27.8 per

cent.

The number of farms for which no mortgage report was secured was much greater in 1920 than at any of the preceding censuses, which fact may affect somewhat the comparability of the figures. An investigation of the farms tabulated as "unknown" for 1920 in a number of typical counties indicated that by far the greater part of these farms were actually free from mortgage. Assuming that these counties were fairly representative, it may be stated that the figures representing the number of mortgaged farms in 1920 are fairly comparable with those of the earlier censuses, in spite of the

The increase in the number of mortgaged farms in the United States and in the percentage of the total number of farms mortgaged does not necessarily represent any lack of prosperity, since the money is borrowed in many instances for profitable investment in improvements or for the purchase of land. In fact, a large fraction of the total debt doubtless represents the balance due on the purchase price from recent purchasers who have paid a part of the price of the farm and given a mortgage for the rest.

In general, the percentage of owner-operated farms mortgaged in 1920 was relatively low in all the Southern states except Oklahoma, and relatively high in a group of North Central and Western states, extending from Michigan, Wisconsin, Iowa, and Missouri on the east to the three Pacific Coast states on the west. The highest percentage of such farms mortgaged among the geographic divisions was 51.8 per cent in

PER CENT

3,925, 090

3,366, 510

558, 580

10

20

40

50

60

70

80

90

100

MAINE N. H.

Unknown (no report).

Not reporting amount of debt.

389,459

24, 187 304, 806

84,653

VT.

MASS.

R. I.

CONN.

N. Y.

N. J.

PA.

OHIO

IND.

ILL.

MICH.

WIS.

MINN.

IOWA

MO.

N. DAK.

S. DAK.

NEBR.

KANS.

DEL. MD.

D. C.

VA.

W.VA.

N. C.

S. C.

GA.

FLA.

KY.

TENN.

ALA.

MISS.

ARK.

LA.

OKLA

TEX.

MONT.

Amount of farm mortgage debt: 1890, 1910, and 1920. The farms for which data on the amount of the farm mortgage debt are presented for 1920 and 1910 include only those farms operated by full owners for which the amount of debt was reported; and the debt shown in the tables is the debt actually reported.

In the reports for the census of 1890, on the other hand, not only were the farms without mortgage report distributed and a part of them assumed to be mortgaged, but estimates were made for the amount of debt on those farms and also on farms for which the fact but not the amount of indebtedness was reported. The number of farms for which estimates were made, however, was relatively small, and the estimates should have but little effect on the comparability of the figures.

TABLE 3.-FARM MORTGAGE DEBT IN THE UNITED STATES, VALUE OF FARMS REPORTING AMOUNT OF DEBT, RATIO OF DEBT TO VALUE, AVERAGES PER FARM, AND INTEREST RATE: 1890, 1910, AND 1920.

[Figures for divisions and states in Tables 10, 11, and 13.]

Mortgage status of farms operated by full owners and part owners: 1920.-The subdivision of farm owners into full owners (those operating their own land only) and part owners (those hiring some additional land) has already been noted. The distribution of the farms operated by those two classes of owners according to mortgage status is shown in Table 2. Of the farms operated by full owners, 36.2 per cent were reported as mortgaged, while of the farms operated by part owners, 43.7 per cent were reported as mortgaged.

88735°-22-31

1 Figures include only farms consisting wholly of land owned by the operator and reporting amount of mortgage debt.

The figures for 1890 include estimates, since at that census the farm homes for which no mortgage report was secured were distributed between the two groups, "mortgaged" and "free from mortgage," and in addition the mortgage debt was estimated, both for the mortgaged homes thus added to the number actually reported and for other mortgaged farm homes failing to report the amount of debt. The inclusion of these estimates explains why the number of farm homes shown in this table is greater than the number of farm homes "reported as mortgaged," which appears in Tables 1 and 6.

1 Mortgage debt on farms consisting wholly of land owned by the operator and reporting amount of debt.

Farm mortgage debt classified according to value of farm: 1920.-Table 5 shows the farm mortgage debt reported for 1920 classified according to the value of the mortgaged farms, together with the ratio between debt and value and the average interest rate for each group. The most significant figures for the United States as a whole are those for the $5,000 groups, omitting the subdivisions of the first two groups. In studying the figures for some of the individual states, however, where the average value of farms is relatively low, the smaller value groups are needed.

TABLE 5.-FARM MORTGAGE DEBT IN THE UNITED STATES CLASSIFIED ACCORDING TO VALUE OF FARM, WITH RATIO OF DEBT TO VALUE AND AVERAGE INTEREST RATE: 1920.

[Figures for divisions and states in Tables 16, 17, and 18.]

GA.

FLA.

Tables for divisions and states. In Tables 6 to 18, inclusive, practically all of the data which have been summarized in Tables 1 to 5, above, are presented by geographic divisions and states. Whatever explanation or comment has been made with regard to the figures for the United States as a whole will also apply, in general, to the figures for divisions and states. A study of the figures here presented by divisions and states will, in many cases, suggest an explanation for a change or a tendency which appears simply as a stated fact in the summary for the United States as a whole. In order to expedite the publication of the figures themselves, however, it has been. thought best not to attempt any detailed analysis of these tables in the present report.