16 
BULLETIN 1480, U. S. DEPARTMENT OP AGRICULTURE 
ence of 0.7 cent. The early months of the 1922-23 season were the 
low-priced months (see Table 1), and the marketings were heavier 
than usual during these months ; therefore, using the constant weights 
based on the usual rate of marketing would give a higher annual 
price average than would using as weights the actual marketings 
for that year. Table 3 gives a comparison of the constant per- 
centage weights and the current marketings and percentage equiva- 
lents for the two crop years when the differences between the annual 
average prices were greatest, 1920-21 and 1922-23. 
Table 3. — Comparison of constant and current monthly cotton weights 
Month 
Constant 
weights, 1 
all years 
Crop year 1920—21 
Monthly 
market- 
ings 
Equiva- 
lent per- 
centage 
Crop year 1922—23 
Monthly 
market- 
ings 
Equiva- 
lent per- 
centage 
August 
September 
October... 
November 
December. 
January... 
February. . 
March 
April 
May 
June 
July 
Per cent 
3 
12 
20 
20 
14 
7 
5 
5 
4 
4 
1,000 bales 
403 
1,296 
2,152 
2,092 
1,464 
847 
745 
795 
885 
925 
880 
763 
Per cent 
3 
10 
16 
16 
11 
1,000 bales 
502 
1,657 
2,472 
1,916 
1,207 
571 
423 
360 
200 
94 
142 
151 
Per cent 
5 
17 
25 
20 
13 
6 
4 
4 
2 
1 
1 
2 
1 Based on usual or average percentage of the cotton crop marketed in each month. 
In the two years when the differences between the A and C 
methods of determining the United States annual average prices 
were the greatest the actual rate of marketing did not correspond 
with the usual or average rate, which is used as a basis for the con- 
stant weights in method A. If the monthly United States prices 
obtained by method A are weighted by the actual marketing rates 
used in method C — the procedure called method D in Table 2 — the 
difference of 1.4 cents between the A method and the C method 
averages in 1920 is reduced to 0.1 cent, and the difference of 0.7 cent 
in 1922 is reduced to 0.1 cent. In Table 2 the comparison of the 
results obtained by using methods C and D show that for 9 of the 
15 years included in the table the results are identical, and in no 
year did the difference exceed 0.1 cent. 
Apparently the difference between the results obtained with 
method A, the method now generally used with farm prices, and 
method C, which is a much more logical and refined method, arises 
from the method of constructing the monthly marketing rates rather 
than from the State weights used in determining the United States 
monthly price. The small differences between the. monthly prices 
as determined by the two methods, as shown in Table 2, seem to be 
largely compensating differences when the monthly prices are com- 
bined to obtain the annual price for the United States, when current 
rates of marketing month b}^ month are used rather than constant 
weights based on the usual or average rate of marketing. 
