Gottfried Haberler, Der Sinn der Indexzahlen (1927)

Haberler’s study is an attempt to rescue index numbers from both uncritical statistical realism and total skepticism. He grants that economists and policymakers cannot avoid saying that prices have risen, money has lost purchasing power, or wages have changed “in real terms.” But he denies that such statements are self-evidently measurable by a single formula. The opening problem is therefore conceptual, not technical:

Ein Sinn muß doch bei dem Worte sein! Aber welcher?

English translation: There must be some meaning in the word! But which one?

The first, mathematical-statistical part reconstructs the major index formulas from within the standpoint of the “index believers.” Haberler surveys arithmetic, harmonic, geometric, median, and modal averages; weighted and unweighted methods; Laspeyres, Paasche, Drobisch, Fisher, Walsh, Lehr, chain indices, circular tests, time reversal, and factor reversal. His aim is not to find the most elegant formula but to show that purely formal criteria cannot determine economic meaning. The critical sentence in the preface states the diagnosis:

Wo die Begriffe fehlen, da stellt sich zur rechten Zeit — ein Durchschnitt ein.

English translation: Where concepts are lacking, an average conveniently presents itself at just the right moment.

Against Fisher’s formalism, Haberler insists that an index number is intelligible only when it can be interpreted as comparing the money cost of a definite bundle of goods across periods. This is his central mathematical move: formulas must be translated back into price sums of determinate commodity combinations. Laspeyres compares the old basket at old and new prices; Paasche compares the new basket at old and new prices. Compromise formulas, including Fisher’s “ideal” index, are therefore not revelations of a true price level but compromises between two economically interpretable boundary cases.

Verständlich und sinnvoll sind bloß jene Indexzahlen, die sich irgendwie auf Verhältnisse von Preissummen zurückführen lassen, und zwar Preissummen einer und derselben Güterkombination.

English translation: Only those index numbers are intelligible and meaningful which can somehow be traced back to ratios of price sums, and specifically to price sums of one and the same combination of goods.

The second, economic part supplies the missing foundation. Haberler adopts a Weberian-Austrian methodological individualism: economics must reduce collective expressions such as price level, income, and money value to the acts and choices of individuals. This is why the “national” price level cannot be treated as an independent object waiting to be measured.

Die Volkswirtschaftslehre hat es mit menschlichen Handlungen, mit Wirtschaftsakten zu tun

English translation: Economics deals with human actions, with economic acts.

The decisive application is the income index. Haberler distinguishes money income, natural income, and “real” income, but rejects the idea that statistics can directly measure psychic income or utility. What can be known is not an objective quantity of satisfaction but the choices persons make between possible budgets. From this he derives his most important constructive result: for a given person or homogeneous group, Laspeyres gives one boundary and Paasche the other for the change in that person’s price level. If the two are close, a compromise index is practically acceptable; if not, no formula can create precision.

This yields Haberler’s pluralist thesis. There is no single purchasing power of money valid for everyone. A price level exists only relative to a person, group, budget, and even income situation. The same price changes may raise the cost of living for one class and lower it for another.

Ein volkswirtschaftliches Preisniveau, eine volkswirtschaftliche Kaufkraft, einen objektiven, überpersönlichen Tauschwert des Geldes gibt es nicht.

English translation: There is no such thing as an economy-wide price level, an economy-wide purchasing power, or an objective, supra-personal exchange value of money.

Haberler does not conclude that all index use is meaningless. Rather, he relocates its meaning in purpose: income comparison, debt adjustment, wage policy, monetary stabilization, or business-cycle diagnosis require different constructions. For deferred payments, no purely scientific index can settle conflicts between debtor and creditor if their relevant price levels diverge. For monetary policy, the “national” price level is only an average of individual indices and is usable because, in practice, price movements often show enough coherence.

Das volkswirtschaftliche Preisniveau — ein statistischer Zufall.

English translation: The economy-wide price level — a statistical accident.

The book’s relevance lies in this disciplined middle position. Haberler rejects the fantasy of exact measurement, but also the sterile denial that index numbers have any use. Their usefulness depends on limited dispersion, stable economic relations, and a clearly specified question. In the final discussion of business cycles, he notes that the general price index loses centrality: modern conjunctural analysis uses special price series and other indicators as symptoms, not as direct measures of money’s value. The work ends, in effect, by denying that one index can satisfy all social, monetary, and analytical purposes.

Eine allseitige Stabilisierung, die es allen recht macht, die den Schuldnern und Gläubigern Gerechtigkeit widerfahren, alle Geldeinkommen unverändert läßt und zugleich die zyklischen Schwankungen ausschaltet, ist eben ein unerreichtbares Phantom.

English translation: A comprehensive stabilization that satisfies everyone, does justice to debtors and creditors alike, leaves all monetary incomes unchanged, and at the same time eliminates cyclical fluctuations, is simply an unattainable phantom.