Timothy Sinclair raises several interesting points in his review. First, he criticizes us for not focusing on “more proximate indicators of looming financial crisis” (such as credit quality) than our key explanatory variables: capital inflows and financial market structure. To be sure, more “proximate” factors help explain some crises, as numerous books already highlight. Yet these variables dominate the existing literature and frequently miss the forest for the trees. We show that market structure and capital inflows interact to form a dangerous cocktail that has contributed to banking crises in industrialized countries since the 1970s. Furthermore, as we show in our historical case studies of Canada and Germany, today’s financial market structure is the result of deeply contentious political battles over the very long run. “Proximate” factors also matter, but past work has overlooked the nonproximate deeper causes of financial instability.
Sinclair asks why, in focusing on the size of securities markets relative to traditional commercial banking, we “do not take the next step…and examine the shift away from relatively expensive bank loans to cheaper bond issuance.” Our primary metric in the statistical analysis is the ratio of stock market size to private credit from the banking sector, but this is but one of many proxies capturing a more general trend: the rise in the relative size of securities markets. Bond issuance is part of this trend, and bond market size is highly correlated with stock market size. Indeed, using the size of bond markets in our models yields substantively identical results. Ultimately, the relative size of traditional banks/banking to nonbank financial activity/intermediaries is what matters, not the precise form of nonbank activity itself.
Sinclair also suggests that our focus on capital inflows was not useful. I confess that I find this baffling. The exponential growth in cross-border capital flows is the single most important development in global finance in the last 50 years. The literature is clear that capital inflows are a major correlate of banking crises. Yet they only trigger crises in some cases and not in others. Explaining this puzzle and the effects of these massive changes in global finance is vital. Other variables surely matter, but our findings show clearly that deeper structural factors also do, and they evolve through complex, long-run political processes.
Sinclair suggests our German case study would have been better told with more fieldwork and interviews. Of course—yet there are trade-offs in everything. Our puzzle was “Why do large capital inflows cause financial crises in some cases and not in others?” Our answer involved testing the impact of capital inflows cross-nationally and exploring the politics of long-run historical transformation in financial markets in two countries over 150 years. A more detailed contemporary German case study would indeed be interesting. It was beyond the scope of our book.
Finally, Sinclair finds our policy implications disappointing. Because we provide a wealth of evidence demonstrating that large capital inflows and declining bank capital are the central determinants of banking crises, it is not surprising that our recommendation is to require banks to hold significantly more capital. We also clearly discuss why regulating bank activities, size, and other dimensions is likely to be far less effective in ensuring financial stability. Here, as with earlier parts of the book, we clearly disagree with Sinclair about which factors are the most important determinants of crises and why.
I thank Sinclair for his review. I wish it had engaged more directly with the theory and empirical evidence of the book we wrote, rather than the very different one he wished we had written.
