OH. 5 PLURALITY OF SECONDARY PRICE-LEVELS 75

time of Great Britain’s return to the Gold Stand-ard the Treasury and the Bank of England were ledto the false conclusion that, because the WholesaleIndex, which was almost the same thing as the BritishInternational Index, was moving rapidly into adjust-ment—as an International Index necessarily must—with the movement of the gold exchanges, there-fore the same thing was true of “ prices generallyMoreover, students of the Theory of the Credit Cycle,and, indeed, of all those parts of Economic Theory which deal with short-period phenomena, have some-times, by overlooking the temporary divergences be-tween price-levels which in the long run are likely tomove together, assumed away the very facts whichit is the task of such a theory to investigate.

Nevertheless the enormous increase since the Warin the variety of price-levels for which index-numbersare being compiled, and the greater adequacy of thestatistical data on which they are based, are graduallydestroying the popular conception that an Index-Number of Prices, such as Sauerbeck’s Index or theEconomist’s Index, can be used to furnish more orless satisfactorily an index of the value of money inall sorts of contexts, different from that for which itwas expressly compiled. The duty of official StatisticalDepartments should be, I think, firstly to prepare areally good Index of the Purchasing Power of Money,and secondly to multiply the number and variety ofthe specialised secondary and sub-indexes which theycompile and publish (preferably with the assistanceof trade bodies and trade experts), so as to render itas easy as possible to build up by various combinationsof these sub-indexes more complex indexes appropriateto the particular purpose or inquiry on hand.