Generating Cycles in a Century of Prices 55 



and 7.38 years have been fitted to the data separately. 

 The curve of 8.73 years, as we should expect from the 

 greater value of A- in Table II, gives a better descrip- 

 tion of the residuals than the curve of 7.38 years. 



Crop Cycles as Generating Cycles 



No one familiar with the theory of prices and with 

 the multitudinous causes of change would expect the 

 record of general wholesale prices to show an exact 

 mathematical precision in the working out of any one 

 cause. If there were a predominant cause one would 

 feel that, at best, the nature of its effect would be 

 revealed only in the average of a fairly long record. One 

 would be prepared for a marked deviation from the 

 average in any particular instance. On the other hand, 

 if in the average of a fairly long record there should 

 be evidence of a persistent cycle, no one acquainted with 

 the statistical theory of cycles would fail to suspect the 

 presence of a predominant periodic cause. 



The analysis of a century of wholesale prices has 

 revealed the existence of a cycle of about eight years 

 in length. What is its cause? 



I have shown that in the United States the yield 

 per acre of the leading crops between 1882 and 1918 

 passed through cycles of about eight years with maxmia 

 at about 1882, 1890, 1898, 1906, and 1914. These 

 eight-year cycles in the yield of the crops generated 

 eight-year cycles in the prices of organic raw materials 

 of manufactures which, according to the law of competi- 

 tive price, were followed by corresponding cycles in the 

 prices of manufactured commodities. In two earlier 



