Working Paper: NBER ID: w26793
Authors: Graciela L. Kaminsky; Leandro Medina; Shiyi Wang
Abstract: With a novel database, we examine the evolution of capital flows to the periphery since the collapse of the Bretton Woods System in the early 1970s. We decompose capital flows into global, regional, and idiosyncratic factors. In contrast to previous findings, which mostly use data from the 2000s, we find that booms and busts in capital flows are mainly explained by regional factors and not the global factor. We then ask, what drives these regional factors. Is it the leverage cycle in the financial center? What triggers the leverage cycle in the financial center? Is it a change in global investors’ risk appetite? Or, is it a change in the demand for capital in the periphery? We link leverage in the financial center to regional capital flows and the cost of borrowing in international capital markets to answer these questions. Our estimations indicate that regional capital flows are driven by supply shocks. Interestingly, we find that the leverage in the financial center has a time-varying behavior, with a movement away from lending to the emerging periphery in the 1970s to the 1990s towards lending to the advanced periphery in the 2000s.
Keywords: capital flows; leverage cycle; financial center; regional factors
JEL Codes: F30; F34; F65
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| leverage in the financial center (G19) | regional capital flows (F32) |
| leverage in the financial center (G19) | cost of borrowing in international capital markets (F34) |
| shocks to leverage in the financial center (F65) | capital flows to emerging periphery (F32) |
| shocks to leverage in the financial center (F65) | capital flows to advanced periphery (F32) |
| shocks to leverage in the financial center (F65) | regional factors for Asia and Latin America (N95) |
| shocks to leverage in the financial center (F65) | regional factor for Europe (O52) |