|
|
Banks have substantially expanded their activities across borders over the past two decades in several ways. They have invested more domestic capital in foreign countries, a cross-border operation defined as international banking. Moreover, they have, to a much larger extent, intermediated capital locally through branches and subsidiaries in foreign markets, an activity called global banking. While growth in banking across borders has been similar in many countries, there has been significant heterogeneity in the extent to which banking sectors engage in international versus global banking. The reasons for these differences have remained largely unexplored.
In this paper, the Fed's Friederike Niepmann develops a model of banking across borders that can explain these facts and replicate major patterns in the data. In the model, countries differ in relative factor endowments and in banking sector efficiencies. These differences lead to trade in banking services when countries become integrated. The model shows that international banking is driven by differences in factor endowments, whereas global banking arises from differences in banking sector efficiencies. Banks’ foreign asset and liability positions are a result of the two driving forces working together.
In the empirical part of the paper, Niepmann tests key implications of the theory concerning the cross-country pattern of banks’ foreign positions. To that end, he draws on data on foreign asset and liability holdings from the Bank for International Settlements that vary across source countries and recipient countries. In addition, data from Deutsche Bundesbank is used that provide details on the foreign positions of German banks in a large number of recipient countries. Evidence from both sources strongly supports the model.
Specifically, the results indicate that more efficient banking sectors lend more and borrow more in countries that host less efficient banking sectors. In addition, banks from capital-abundant countries invest more in the private sector in capital-scarcer countries. Thus, the cross-country pattern of international and global banking appears to be driven by real factors with potential benefits to home and host countries.