Oskar Morgenstern’s economic discussion essay, published on 8 January 1927, examines the scientific claims and practical expectations surrounding business-cycle research. Responding to F. A. Hayek’s advocacy of an Austrian research institute, Morgenstern argues that the case for establishing one must be distinguished from confidence in existing forecasting methods. His intervention proceeds from American research practice through statistical and predictive weaknesses to the conditions for productive work in Vienna. He supports Hayek’s institutional proposal while challenging his favorable assessment of the field: research deserves encouragement precisely because important problems remain unresolved, not because reliable economic prediction has already been achieved.
America supplies Morgenstern’s principal evidence. Its numerous commercial, university-connected, banking, and governmental research bodies have generated an impressive volume of publications, but visible activity does not establish either scientific progress or practical influence. Entrepreneurs may subscribe to reports as a fashion, while smaller businesses lack the resources to interpret their diagrams. Large banks and corporations often employ economists to investigate their own specialized problems. Morgenstern thus separates the public prestige of business-cycle charts from their effective use, and the commercial authority of forecasting services from the quality of their work.
His deeper objection concerns the relationship between accumulating observations and advancing economic explanation. Despite refinements in statistical technique, he doubts that American research has decisively surpassed Mitchell’s foundational work of 1913:
Man hat im Gegenteil den theoretischen Rückhalt verloren; Einzeluntersuchungen über meist sehr enge Gebiete häufen sich und das Gesamtbild ist das einer ziemlichen Planlosigkeit.
English translation: On the contrary, theoretical grounding has been lost; individual investigations into mostly very narrow fields accumulate, and the overall picture is one of considerable lack of direction.
This criticism allows for valuable findings, particularly about banking and the New York money market. The problem is their explanatory reach: results concerning specifically American institutions cannot automatically acquire general validity. Nor does removing seasonal fluctuations and trends ensure economically meaningful analysis. Mathematical refinement can become computational play when researchers fail to establish what their procedures explain.
Morgenstern’s central methodological distinction is between calculating relationships and interpreting them. Correlation coefficients do not supply their own economic meaning. Comparing pig-iron prices with interest rates may accomplish little; a relationship between call-money rates and bank reserves becomes intelligible through the particular American law governing reserves. Institutional knowledge and theory must therefore guide both the selection of series and the interpretation of results. He also calls for greater clarity about statistical probability and a critical examination of research materials and methods, rather than treating technical sophistication as evidence of secure knowledge.
The discussion then turns from retrospective measurement to prediction. No cycle of determinate duration has been demonstrated across a sufficiently long period, and the necessity of one cycle following another remains unexplained. The observed succession of the Harvard Index’s A, B, and C curves offers too narrow a foundation for far-reaching conclusions, particularly when forecasts inform central-bank policy or schemes of economic stabilization. Composite indices introduce an additional difficulty:
Indices dieser Art bedeuten eine Häufung der Fehler, die die als Ausgangspunkt verwendeten Indices (Preis, physisches Volumen der Produktion usw.) bereits enthalten, und sind schon aus diesem Grunde mit Vorsicht zu genießen, geschweige denn bei Errichtung von Vorhersagen auf ihrer Grundlage.
English translation: Indices of this kind entail an accumulation of the errors already contained in the indices used as their starting point (prices, physical volume of production, etc.), and for this reason alone must be treated with caution, all the more so when forecasts are constructed on their basis.
Calling these instruments “barometers” obscures the distinction between recording conditions and anticipating developments. Morgenstern also raises the unresolved contradiction involved in predicting human actions while treating them as inevitable and law-governed. His comparison of Harvard’s forecast for 1927 with an experienced Boston banker’s spontaneous judgment makes a narrower practical point: the statistical forecast differed chiefly in numerical precision and degree of rationalization, not fundamentally in kind. Institutions should not be credited with miraculous predictive powers.
These objections lead to a constructive institutional argument. Vienna should establish a research center, but its supporters must understand that they are creating working conditions for scientists pursuing durable results, not purchasing an immediate remedy for Austria’s economic distress.
Die Wissenschaft ist es sich selbst schuldig, Zweifel nicht zu verschweigen und eine ungerechtfertigte Hoffnungsfreudigkeit abzubremsen, denn sonst sind Enttäuschungen unausbleiblich; die Schuld aber wird der Wissenschaft zugeschrieben, wenngleich sie lediglich mißbräuchlich angewandt wird.
English translation: Science owes it to itself not to conceal doubts and to curb unjustified optimism, for otherwise disappointments are inevitable; yet the blame is attributed to science, even though it is merely being misapplied.
The analogy with a physics laboratory clarifies his position: useful discoveries may emerge, but immediate practical inventions are not an appropriate founding expectation. Vienna’s theoretical tradition and capable economists provide favorable conditions. They must be complemented by faster, richer federal statistics, greater financial latitude for the statistical office, and substantially expanded university instruction in statistics. His complaint that Austria lacks a full professorship in statistics makes clear that criticism of statistical dominance is compatible with demanding stronger statistical training.
The concluding conceptual move assigns responsibility for research design and inference:
Daher braucht er den Statistiker; aber der Wirtschaftstheoretiker bestimmt die Wahl der Methode und er zieht die Schlüsse daraus.
English translation: He therefore needs the statistician; but the economic theorist determines the choice of method and draws the conclusions from it.
Numerical evidence is indispensable, but it remains material for economic inquiry rather than an autonomous substitute for explanation. Morgenstern accordingly wants Vienna’s institute directed by an economic theorist. The essay’s significance lies in joining methodological restraint to institutional ambition: acknowledging uncertainty protects scientific credibility, while theoretical direction, improved data, and statistical competence make future practical gains possible. His warning against turning a difficult research field into a fashion is ultimately an argument for supporting it on intellectually defensible terms.
This work was divided into 1 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 1 sections and cites the passage.
Ask the Librarian