This project involves the creation of a new database of international capital flows for the heydays of financial globalization during the 19th and early 20th centuries. The PI will use this database jointly with modern data on capital flows to study financial contagion, to construct an anatomy of capital flow bonanzas, to examine the dynamics and scope of crises that start in the periphery and those that start in the financial center, and to shed more light on the relationship between capital flow bonanzas, sovereign defaults, and growth.
Empirical studies on the characteristics and effects of international capital flows examine mostly the experience of the last thirty years because there are data available only for this period. While these data allow researchers to untangle the cross-country differences in international capital flows and the links to country fundamentals and macropolicies, they cannot help to examine the effects of sporadic shocks in international capital markets and overall timedependence. For example, since 1970 most of the currency crises and the capital flow reversals they triggered started in countries in the periphery, Mexico (1994), Finland (1992), Thailand (1997), Russia (1998). In contrast, the ongoing global financial crisis started in the financial center, similarly to the crises in 1824, 1873, and 1929. Only by comparing the current crisis to those of the earlier period, will we be able to answer which are the mechanisms of contagion when the financial center is the epicenter or assess the future scope of the current crisis. A database on capital flows that spans two centuries can help in this endeavour.
This project will also deepen our understanding of capital flow cycles and channels of contagion as well as the elusive link between financial integration and growth. For example, until the mid 1990s most of international capital flows to developing economies took the form of bank loans. Only very recently did these countries gain access to international bond markets. In contrast, in the 19th century, most portfolio flows were bond flows. Are sudden stops more pronounced when capital flows take the form of bank loans as in the 1970s? Or is the bond market during the 19th century and nowadays (with its many uninformed and dispersed investors) what triggers more pronounced reversals, herding behavior, and financial contagion? Also, what are the links between trade openness, the composition of capital flows, sovereign defaults, and growth? re developing countries serial defaulters that only over time "graduate" and stop defaulting? Or are there reversals in their commitment not to default as is the case of Argentina who was only in default 3 percent of the years from 1858 to 1931 but was in default 41 percent of the years in the most recent period? Did the type of capital flows in the 19th century trigger higher commitment?
Broader impact: A central contribution of this project is to create a new database of gross primary issuance of bonds, equities, and loans floated in London, Paris, Berlin and Frankfurt, and New York, the financial centers of the 19th and early 20th centuries. The data to be collected include all individual issues floated in these financial centers. Using this new database, we can decompose capital flows into sovereign and private borrowing and by maturity and purpose of the issue. We can also identify the sectoral composition of the firms that are tapping international capital markets (say, railways, meat processing plants, petroleum).
