Oskar Morgenstern, Aufgaben und Grenzen der Institute für Konjunkturforschung (1928)

Oskar Morgenstern’s scholarly assessment examines whether, and in what respects, practical business-cycle research institutes advance economic theory. This separately titled contribution to an edited volume proceeds through three sections: institutional types, institutes’ self-assigned tasks, and their dependence on economic theory. Its governing distinction is between the useful organization of information and empirical investigation, on the one hand, and the creation of theoretical knowledge, on the other. Morgenstern welcomes institutional assistance but rejects the claim that statistical machinery, collective research, or forecasting services inaugurate a scientific revolution.

The first section distinguishes functions rather than rigid organizational categories. Research institutes coordinate related individual studies; information agencies supply subscribers with regular reports, sometimes accompanied by forecasts. A single institution can perform both functions, but its supplementary research should not be confused with its reporting service. Organized research can relieve scholars of routine work and secure attention for their publications. Yet institutional prestige and funding can also divert researchers from their own projects toward a director’s programme. Morgenstern places these benefits and risks against the aspiration to emulate natural science: impressive equipment and extensive statistical archives may generate public respect without establishing theoretical achievement.

The second section tests four proposed tasks: information and diagnosis, forecasting, policy recommendations, and theoretical research. Reporting economic events differs fundamentally from locating an economy within a business cycle. The latter requires a theory of cyclical development and an account of how its phases can be identified. Statistical processing alone cannot supply either. Nor does employing scientifically trained staff make an information service itself a scientific enterprise. Nevertheless, information has genuine economic importance because it enters the circumstances on which entrepreneurs base their decisions.

Forecasting sharpens this connection between knowledge and conduct. Morgenstern separates two questions often conflated: whether prediction is possible and whether, if possible, it would stabilize economic activity. He rejects confident prediction because the relevant observations are results of human action, not manifestations of sufficiently recurrent regularities. Even granting predictive possibility, publication changes the object predicted. Entrepreneurs who trust a forecast will try to avoid the losses it announces; extending its audience therefore undermines its accuracy. Universal acceptance would reach the limiting case:

Dies ist genau der Fall, wenn die Prognose sich selbst aufhebt. Es ist ein vollkommener, sich immer wiederholender circulus vitiosus.

English translation: This is precisely the case in which the forecast cancels itself out. It is a complete, perpetually recurring vicious circle.

This argument concerns more than inadequate statistical technique: prediction becomes an input into the behaviour whose outcome it claims to anticipate. Morgenstern refers readers to his separate study for the full demonstration, but his institutional recommendation is categorical: abandon forecasting, even at the cost of popularity, and improve the dissemination of information.

His treatment of policy studies is considerably more affirmative. Sustained, impartial investigation could improve responses to currency, reparations, and other practical problems. Permanent institutes potentially possess greater continuity than temporary commissions. Such studies may also produce theoretical insights through unfamiliar materials, new arrangements of evidence, and scrutiny of the theorems employed. These gains remain incidental to practical inquiries, however, and institutional endorsement cannot replace the judgement of their individual authors.

The discussion of deliberately organized theoretical research makes this limitation explicit. Group projects can divide empirical work, answer specified factual questions, and undertake investigations exceeding one scholar’s practical capacity. Their statistical results do not by themselves resolve the methodological problem of generalizing from particular observations to causal knowledge:

Andererseits liegt eine Notwendigkeit in diesem Verhältnis; eine Theorie verlangt von einem zuerst gedacht und geschaffen zu werden, es gibt kein gruppenmäßiges Denken.

English translation: On the other hand, there is a necessity inherent in this relationship; a theory requires first to be thought out and created by an individual; there is no group thinking.

Morgenstern accordingly distinguishes the National Bureau’s collective investigations from Wesley C. Mitchell’s personal theoretical interpretation of them. Another economist could draw different conclusions from the same materials. Collective inquiry supplies indispensable preparatory work without becoming a collective theoretical mind.

The final section establishes the priority of theory even within apparently empirical undertakings:

Es sind also rein theoretische Entscheidungen, die gefällt werden müssen, bevor irgendeine solche Arbeit unternommen werden kann.

English translation: Thus purely theoretical decisions must be made before any such work can be undertaken.

What to measure, which relationships matter, and whether additional precision justifies its expense are questions that measurement cannot settle for itself. Mitchell’s investigation of price relationships illustrates the possibility that extensive empirical work will verify existing insights rather than generate new theory. Morgenstern nevertheless values continuous data collection as a compressed economic history, while emphasizing that statistics cannot capture every causally significant qualitative phenomenon.

His conclusion preserves this measured endorsement:

Die Bedeutung der Institute für die Theorie ist demnach nicht unbeträchtlich, jedoch kaum überragend; sie sind gewiß sehr willkommene Bereicherungen unserer Arbeitsmöglichkeiten.

English translation: The importance of the institutes for theory is therefore not inconsiderable, yet hardly outstanding; they are certainly very welcome additions to our means of working.

The closing qualification shifts from method to institutional power. Funding, publicity, state support, and publication subsidies can give institutes influence exceeding their intellectual warrant. The essay’s enduring relevance lies in connecting limits of prediction and measurement with the organization of knowledge: institutions can enlarge research capacity while also directing scholars, allocating attention, and making particular programmes authoritative.