RELIABILITY AND ADEQUACY OF FARM-PRICE DATA 59 
From the standpoint of the trend of the gross value of all crop 
production for the United States, it makes very little difference 
which price is used. In the case of certain individual crops the 
changes from year to year may differ more if one method of comput- 
ing values is used rather than the other. In computing the gross 
value of all crop production the individual variations are more or 
less compensating. 
Prices on December 1 are probably as satisfactory as those on any 
other single date for estimating the gross value of all crop produc- 
tion, as it is a satisfactory date for some of the most important crops. 
The fact should not be overlooked, however, that for some commodi- 
ties and groups of commodities other dates may be more satisfactory. 
Considering only the question of the value of the crop at harvest 
time, the logical point at which to measure that value is when the 
crop has just been harvested. The practice of the Bureau of Agri- 
cultural Economics in the last few years has been to add to the 
December 1 valuation of some of the crops the values of other crops 
completely harvested earlier in the season at prices that prevailed in 
their respective harvest seasons. The value of crops at prices pre- 
vailing at time of harvest may naturally be expected to be lower than 
the value on the basis of the monthly prices weighted by monthly 
marketings throughout the season. As is shown in Table 27, the 
December 1 prices give lower gross values than the weighted-average 
prices for the season. 
After the December 1 farm prices are available, gross value of 
production may be computed on the basis of these prices, on the 
basis of the monthly prices of the season to date in comparison with 
past years, or upon the basis of prices to date and estimates of prices 
for the remainder of the year. Becent developments in statistical 
technique in the analysis of prices are preparing the way for using the 
last method with a fair degree of accuracy. In case it is desired to 
use a weighted-average price to compute probable income for the 
season, it is necessary not only to estimate the prices in advance but 
also to estimate the marketings monthly through the season. 
As a basis for comparisons between States at a given date, the 
December 1 prices have the advantage of being more fully represent- 
ative of all localities in the country than the monthly prices, and 
since they are based on a much larger sample the State average 
obtained is also much more reliable. For immediate use for a given 
State or as a basis for a comparison between States they are now 
available by States, while the monthly prices would have to be 
weighted by States. This would be no small task, as at least 1,700 
weighted annual averages would have to be computed. The depart- 
ment has already substituted for the December 1 prices of such 
commodities as are not being sold on December 1, such as some vege- 
tables and fruits, an average price or value per unit by States of 
commercial production based on prices received by growers during 
the harvest season only. 
Another criticism of the use of the December 1 price in calculating 
the total value of staple crops is that often it does not represent the 
average price at the time the given crops were sold. For example, 
in a season when farmers obtain an average price of 70 cents per 
bushel for all the corn that they sell, the average December 1 price 
