Risky banks and macro-prudential policy for emerging economies

Abstract

We develop a two-country DSGE model with global banks to analyze the role of crossborder banking flows on the transmission of a quality of capital shock in the United States to emerging market economies (EMEs). Banks face a moral hazard problem for borrowing from households. EME's banks might be risky: they can also be constrained to borrow from U.S. banks. A negative quality of capital shock in the United States generates a global financial crisis. EME's macroprudential policy that targets non-core liabilities makes the domestic economy resilient to the volatility of cross-border banking flows and makes EME's households better-off.

Assessing the Gains from International Macroprudential Policy Cooperation

“…They find that it is generally optimal to use both. Kollmann et al (2011), Kollmann (2013, and Cuadra and Nuguer (2018) have considered more general coreperiphery models where global banks play a key role in the international transmission of shocks. 10 These contributions have considered a number of MaP instruments, often in the form of simple countercyclical rules.…”

Determinants of Non-Core Liabilities of Banks in Emerging Markets in the Post-Crisis Era

“…They also suggest macroprudential policies such as a levy on wholesale funding, short-term debt or foreign flows would be effective to dampen these risks. Corroborating with this, Cuadra and Nuguer (2018) DETERMINANTS OF NON-CORE LIABILITIES OF BANKS IN EMERGING MARKETS IN THE POST-CRISIS ERA suggest that if the credit growth is faster than that of bank deposits in emerging markets, a levy on non-core liabilities would smooth the transmission of financial shocks from advanced economies.…”

Robust Optimal Macroprudential Policy

“…However, in order to obtain one average measure of gains in the long-run based on simulations, we require that these simulations be run based on shocks that are either realizations from the benchmark distribution or from the worst-case probabilities, corresponding to each case. For example, if we want to find an average welfare gain from implementing a robust planner's allocation compared to the allocation from a robust household, and we want to compute utility expectations from the perspective of the household, then we use the simulation of shocks that are derived from the worst-case distribution that the household has in mind 8 . Note: The numbers represent the mean welfare gain over the periods in the simulation when a tax is needed to decentralize the planner's allocation.…”

“…This was not evident a priori: a robust planner will choose an allocation that generates a lower bound of lifetime discounted utility, in the sense that implementing that allocation will result in larger welfare if shocks are actually drawn from the benchmark model rather than the worst case distribution. However, if indeed shocks follow the 8 One caveat in our computations is that, in a case like this, the value function associated to the robust planner's allocation was computed using expectations that reflect the planner's worst-case scenario, which differs slightly from the household's worst-case scenario -recall that these worst-case transition probabilities are dependent on the level of current debt and take into account an agent's optimal debt choice. Ideally, we should re-compute these value functions under expectations that consider a household's worst-case distribution.…”

“…Andreasen et al (2019) study a general equilibrium model with heterogenous firms and collateral constraints, and find that capital controls reduce aggregate production and investment while increasing exports, the share of exporters and TFP. Cuadra and Nuguer (2018) show that implementing macroprudential tools that target crossborder bank flows is welfare improving in emerging economies. Farhi and Werning (2016) build a small-scale model with nominal rigidities in which macroprudential policies correct for aggregate demand externalities.…”