RELATION OF LAND INCOME TO LAND VALUE. 31 
Group 25 shows a much higher anticipated increase in income than 
group 26. which lies next to it. This is explained by the fact that the 
counties in group 25 lie closely around San Francisco. Furthermore, 
a majority of the cash-rented farms will probably be found in those 
sections of those counties that lie closest to the city. There is a 
potential residential value attached to these lands which does not 
attach to those lying farther out. 
Anticipated increase in income is shown to be very low in group 37, 
and in groups 42 and 46 i, as calculated, is probably smaller than it 
should be. This seems to be true, because land values have been 
advancing in these southern States at a much greater rate than in 
any other part of the country. The antk-ipated increases as computed 
by the formula are probably lower than they should be. because the 
average rent obtained for these areas is not a true net rent. Because 
the cash rents in the South, on which these averages are based, 
contain payments for the risk borne by the landlord, a is too high 
and the computed i is too low. But this will not account for the 
very low i in group 37. Here the ratio of rent to value is almost 
equal to the mortgage rate of interest, and the anticipated increase 
in income only 3 cents a year. Offhand, it appears that the land 
in this area was relatively undervalued, but sufficient data are not 
available to enable one to make this statement with certainty. It 
may be that this land appears to be undervalued, because all the 
capital which can be obtained is absorbed in clearing and draining the 
land. That is, those who can obtain capital find it more advantageous 
to use it in clearing; and improving land than to use it in buying more 
land and thus bidding up the price. 
It is clear that a low ratio of rent to value, or, in other words, a 
low per cent of return on the valuation of land at a given year, is to 
be explained for the most part by a relatively high anticipated increase 
in income. It must not be thought, however, that two regions hav- 
ing the same mortgage rates of interest and the same anticipated 
increases in income will necessarily have the same ratio of rent to 
value. Compare, for instance, groups 7 and 14. The mortgage rate 
of interest is 5.8 per cent in each of these areas. The anticipated 
increase in income is SO. 23 per acre in group 7 and 80. 2S in group 14 — 
that is, 80.05 higher in the latter area than in the former. But it is 
not enough higher to account for the fact that the ratio of rent to 
value is 1.2 per cent lower in group 14 than in group 7. 
There is no reason to expect that it should be. If the value of i 
were the same in both areas, the ratio of rent to value would still be 
much lower in the latter group than in the former. The point i< that 
in interpreting the effect of i on the ratio of rent to value it must be 
considered in relation to a, the average rent. In group 14. i is about 
twice as great relative to a as it is in group 7. Stated in other terms, 
the capitalized value of i constitutes 62 per cent of the value of land 
per acre in group 14, but only 42 per cent of the value per acre in 
group 7. The greater the proportion of the value of the land which 
is based upon anticipated increases in rent, the lower the ratio of rent 
to value in any given year. 
The calculated i s in Table 10 must be what they arc. Given the 
assumption that the rate of capitalization is the mortgage rate of in- 
terest then the i's result as a matter of mathematical necessity from 
