Haberler’s journal article examines the economic meaning of Carl Snyder’s broad index of the general price level. Its seven sections move from an appreciative account of Snyder’s statistical achievement through tests of its usefulness for real-income comparisons, deferred payments, production, and business-cycle analysis, before identifying a narrower application in monetary research. Snyder combines retail and wholesale prices with rents, wages, securities, real estate, and equipment prices. Haberler’s question is whether this statistical breadth yields an economically coherent measure. His central argument is that an index must be defined by the purpose it serves: comprehensiveness alone cannot establish its significance.
The general price level is not a given, self-evident fact, but a theoretical abstraction.
This formulation shifts the discussion from statistical construction to conceptual justification. Real-income measurement, debt adjustment, and monetary policy need not require the same price index. Different measures might coincide empirically, but such agreement would not establish their theoretical identity. Haberler accordingly assesses Snyder’s index against particular uses rather than judging it by an abstract ideal of complete coverage.
For real-income comparisons, relevant prices are those of the goods an individual consumes. A decline in luxury prices does not necessarily improve the position of the poor; tobacco prices do not directly matter to a nonsmoker. Strictly, each person requires a distinct index, though similar movements across individual indices can justify practical generalization. Haberler also considers the difficulty created by changing consumption baskets. Neither the Laspeyres nor the Paasche formula deserves preference merely on formal grounds. Averaging them becomes defensible if they bound the relevant price change: under assumptions not fully developed here, Laspeyres supplies an upper limit and Paasche a lower one. Their usual proximity supports approximation, but their expected ordering depends on empirical relationships between prices and quantities, not an unconditional mathematical necessity.
A real-wage index has to consist only of prices of consumables.
Snyder’s inclusion of producers’ goods, assets, and wages therefore makes his index unsuitable for measuring real income. Wages are especially revealing: in an advancing economy, unchanged consumption prices can accompany rising wages. A composite containing both would register an increase even though a given money income still buys the same consumption. Haberler allows wholesale prices to approximate retail consumption prices over short periods, but this practical concession does not justify mixing fundamentally different categories. He also expressly notes that Snyder does not claim his index is suitable for real-income calculations.
The discussion of deferred payments exposes the distributive choices behind apparently technical standards. A commodity or tabular standard preserves debts’ purchasing power over consumption goods and thus calls for the same kind of index used in real-income comparisons. A labor standard instead preserves command over labor or productive resources. In an advancing society, stable purchasing-power debts allow debtors to benefit from progress, whereas a labor standard increases creditors’ purchasing power as productivity rises. Intermediate standards divide those gains. Snyder’s composite corresponds clearly to neither principle, making its consequences difficult to determine. The objection concerns conceptual fitness, not a documented proposal by Snyder to index debts this way.
Turning to production, Haberler accepts that living costs alone cannot characterize business conditions. Firms depend on the relationship between the prices of their products and those of labor, equipment, and materials. Yet combining these prices into one average suppresses the relationships that matter.
Monetary policy can stabilize the price level, but not the relative position of the individual prices.
This distinction separates changes in an aggregate level from changes in the price structure. Producers’ prices may be essential evidence about industrial conditions without therefore belonging in a universal price-level measure. Neither debt adjustment nor stabilization of a production-price index can eliminate disturbances arising from divergent movements among particular prices.
Business-cycle analysis introduces a further conceptual shift. Wholesale prices may respond earlier than retail prices, but their usefulness as indicators puts them alongside credit, production, discount rates, and other symptoms. The relationship between an index and purchasing power is definitional; its relationship to prosperity or depression is empirical and potentially variable.
Now, however, the price level is only one symptom among others.
Haberler describes a movement from older theories identifying prosperity with rising prices toward analysis of the timing and interrelations of distinct series. The Harvard Economic Service exemplifies this approach. A broad composite obscures the sequence of movements that forecasting needs to distinguish; its inclusiveness becomes a liability rather than an advantage.
The final section nevertheless identifies a constructive use for Snyder’s work. Haberler questions the attempt to verify the equation of exchange statistically: in its strict form, the equation rearranges the same payments into equivalent totals.
The equation of exchange needs no more statistical verification than the proposition that (a + b) + (c + d) is equal to (b + c) + (a + d).
An accounting identity does not itself explain economic relationships. Yet a broad price index, paired with a corresponding quantity index, might estimate total circulation and permit indirect calculation of otherwise inaccessible monetary magnitudes, including velocity. Haberler leaves this possibility provisional pending Snyder’s completed results. The article’s lasting conceptual contribution is its separation of purchasing-power measurement, distributive standards, cyclical diagnosis, and monetary accounting: the same statistical aggregate cannot acquire authority across these tasks merely by including more prices.
This work was divided into 8 sections when it entered the library's research corpus—an apparatus for search and citation, not necessarily the author's own table of contents. Each title opens its summary.
Put a question to this work; the Librarian answers from its 8 sections and cites the passage.
Ask the Librarian