CORN AND HOG CORRELATIONS 59 
Table 13. — Multiple regression equations involving contemporary variables — Con. 
Total Western Pack 
Per cent 
Summer and winter pack J= +0.61 F +0.53 G_. 100 
Total Eastern Pack 
Western pack and hog prices K= +0.24 J -0.22 W -0.25 S — 0.20 W 56 
Farm Price January 1 
Packers' prices L= +0.70 W +0.29 S 85 
SUMMARY 
An attempt to analyze the play of interacting factors responsible 
for the annual fluctuations in production and price of hogs during 
the period of relatively stable conditions between the Civil War and 
the World War is made in this bulletin. It is shown that variations 
in the corn crop and certain interrelations among the hog variables 
themselves determine from 75 to 85 per cent of the variations of the 
latter. The factors dealt w^ith are listed in the next paragraph. It 
is recognized that there are important external factors other than the 
corn situation which are not dealt with. 
The annual fluctuations about the trend during the period of years 
from 1871 to 1915, inclusive (so far as data were available), have been 
found for corn acreage, yield, crop, and price and for the western and 
eastern wholesale hog packs and the farm price of hogs. The fluctua- 
tions for the summer and winter seasons separately have been found 
for the western wholesale hog pack, and the corresponding average 
live weight, pork production, and price. 
All the possible correlations among these variables have been cal- 
culated (Tables 3 to 6, figs. 16 to 25, inclusive). 
An attempt has been made to find the system of causal relations 
that would account best for the observed correlations. Path coeffi- 
cients have been calculated for each path of influence in this system, 
from which the expected value of each of the 510 correlations has 
been found. 
A series of formulas is given for predicting the fluctuations of each 
variable, with discussions of the limitations. The following general 
conclusions are reached: 
The dominating features of the hog situation are certain effects of 
the corn crop and price and an innate tendency to fall into a cycle of 
successive overproduction and underproduction, two years from one 
extreme to the other. 
The amount of breeding at a given time is determined largely by 
the profits from hog raising during the preceding year, which depends 
on the ratio between the price the packers pay for the hogs and the 
price of corn during the period in which they were raised and finished.' 
The peak effect on breeding occurs from half a year to a year after 
the hog-corn price ratio has reached its peak and begun to decline. 
The result is a surplus of hogs a year or more later, low prices, and 
losses. Breeding, however, does not reach its low point until some 
time after the hog-corn price ratio has begun to rise. A swing to 
underproduction is the consequence. Thus any cause which leads 
to unusual profits or losses tends to set up oscillations in the hog 
population, four years from crest to crest. In order to eliminate 
