یک الگوی اقتصاد سنجی ماهانه از انتقال رکود اقتصادی بزرگ بین اقتصادهای عمده صنعتی

This article describes and estimates, with monthly data, a model of the economic interactions between the United States, the United Kingdom, France and Germany over the years 1927–1936. Despite the radically different economic environment, the model shows broadly similar qualitative and dynamic responses to policy instruments and other changes to those of multi-country models estimated on more recent data. The model is simulated to assess the causes of the Great Depression and the particular contribution of European and American policies to the slump. Optimum strategic policy equilibria are then computed. They point to the mismanagement of the US economy as the principal cause of the depression, although French and German policies were also harmful. British policymakers performed rather well, but their economy suffered because of the other countries’ policy errors.

There are several recent multi-country models available for simulating national policy spillovers and international co-ordination strategies (Whitley, 1992 and Douvens and Plasmans, 1996). All of them are concerned with the rather benign international economy after 1960. The years between the world wars offer an altogether different test-bed for econometric models and a far more traumatic environment for international economic policy. The Great Depression, in particular, ‘continues to be a challenge to economic doctrine’ (Rothermund, 1996, p. 1). Estimation of structures similar to more recent international models over the inter-war period therefore offers insights into their properties and those of co-ordination exercises conducted with them. As early as 1953, Neisser and Modigliani (1953) constructed the first multi-country model for these years, and there have been other excursions subsequently (for example Newell and Symons, 1988). Such econometric attempts to analyse this period have typically been hampered by the small number of annual observations. To increase the available degrees of freedom, the present model of the interplay between several economies utilises the abundant monthly data of the time. An additional advantage of these observations stems from the international financial crises that began with the onset of the Great Depression in 1929. Financial markets change rapidly and therefore, high frequency data may be valuable in understanding both their operation and their pathologies. The present paper models the four principal industrial economies of the period: the United States, the United Kingdom, France, and Germany (G4), and their interactions between 1927 and 1936. Section 2 discusses their salient characteristics and Section 3 expounds the general specification of the national models. Section 4 outlines some implications of the individual equation estimates. Section 5 simulates the model to derive some properties and to investigate the causes of the Great Depression. Section 6 briefly outlines a number of policy co-ordination games simulated with the model. Appendices (available at http://scottie.stir.ac.uk/∼yma01/interwar) describe the monthly national income estimates, the model equations in four national groups, a trade block, a capital movement and an exchange rate block, and some equations employed in an earlier version of the model (Foreman-Peck et al., 1992).

Multi-country models have typically not been used to simulate shocks of the size of the Great Depression or the appropriate policy responses. They have also been estimated with data generated in times less turbulent than the years between the world wars. It is, therefore, reassuring that broadly similar qualitative and dynamic responses to policy instruments and other changes are found in the present study. Equally satisfactory is that spillovers between national economies were estimated to be within the bounds found previously by these models. The structures adopted for more contemporary interactions between national economies appear robust to the elapse of considerable periods of time and to different international institutions. The present model provides parameter estimates from monthly data of the simplest set of equations consistent with explaining the greatest international economic crisis. Simulations suggest that a sharp, but not unprecedented, cyclical downturn in economic activity in the United States had unusually long-lasting and severe effects, both in the US and in continental Europe. The cause was failure of national monetary policies which permitted financial collapses from l93l, just before a cyclical upturn might have been expected. British monetary policy saved Britain from the full force of the depression but a relatively successful story has been obscured by the superimposition of the world crisis on the poorly performing economy of the l920s. This in turn appears to have been a consequence of deflationary tendencies in US policy before l929. Cooperation, leadership or non-co-operation, over monetary (or fiscal) policy, was largely irrelevant so long as policy-makers optimised with conventional economic targets. With the benefit of historical hindsight, the policy problem that really did require co-operation (or leadership) was to avoid enlarging the instrument set to include devaluation, exchange controls and trade quotas. For Europe, but not for the much larger US, this might have been achieved by official international reserve lending sufficiently promptly and abundantly. But political constraints ruled out this solution. The full model structure is posted in the Macro Models Repository on the homepage of Economic Modelling. To access the Repository, please go to www.elsevier.nl/locate/ECONbase and then click on ‘Economic Modelling’.