Bank Competition and Risk Taking in a Zero Interest Environment

By Robin Döttling - November 26, 2017 - , -
print this

How do near-zero interest rates affect bank competition, risk taking and financial regulation? To answer these questions, I develop a dynamic general equilibrium model in which forward-looking banks compete imperfectly for deposits, and deposit insurance may induce excessive risk taking. Banks choose the riskiness of their loans, trading off the gains from shifting risk on the deposit insurance against the risk of loss of franchise value.

In contrast to other work in the “search for yield” literature, I find that a reduction in interest rates leads to stronger risk-taking incentives only if deposit rates are constrained by the zero lower bound (ZLB), or expected to be constrained in the future. Intuitively, as long as deposit rates are not constrained, market power allows banks to pass along low interest rates to depositors, guaranteeing stable interest margins and profitability. The dynamic perspective shows that risk-shifting incentives rise particularly strongly if the yield curve is flat and the ZLB is likely to bind for a long time.

The model delivers a novel rationale for counter-cyclical capital regulation. Calibrating the model to US data, I find an optimal level of around 7-8% when the ZLB remains slack. In contrast, the optimal requirement is lower when the ZLB binds. The intuition is that when banks cannot pass on the cost of capital, tight capital requirements erode franchise value, countervailing the usual “skin in the game” effect. For that reason, capital regulation is less effective at curbing risk-shifting incentives at the ZLB. Very low interest rates may therefore motivate weaker capital regulation, despite overall higher risk. Complementing existing regulation with policy tools that directly support the funding cost of banks may improve welfare.

Job Market Paper

Robin Döttling, 2017. ‘Bank Competition and Risk Taking in a Zero Interest Environment,’ University of Amsterdam and Tinbergen Institute.

Personal website of Robin Döttling.

Follow TI on Twitter

Our new study shows that incidental aversive affect suppresses trust and, concordantly, TPJ activity and connectivity during trust decisions. We also identify a "trust network": regions whose connectivity strength correlates with behavioral trust. https://t.co/R0wZHBR6Yq

Next Monday 18 March @EmilyNix100 presents "What Causes the Child Penalty? Evidence from Same Sex Couples and Policy Reforms" here @ErasmusESE @ResearchTI (Polak Building, Room 1-23; 12:00-13:00).
https://t.co/Gwjf8y8DkZ