Should Central Banks Take On Credit-Risk?

Central banks have a long tradition of minimizing their exposure to credit-risk. The Federal Reserve’s response to the recent financial crisis, which entailed greater risk-taking, has raised the question of whether such ‘unusual’ practices are desirable. This paper addresses the vacuum in the literature with a highly simplified model that has nevertheless the characteristics missing in the literature: it is a monetary dynamic stochastic general equilibrium model with an inflation-targeting central bank, aggregate risk, bankruptcy, and it is tractable. The main contribution is showing that, in an economy with bankruptcy rights and considerably indebted households, a Central Bank that operates exclusively with risk-free assets effects important distortions; in particular, it benefits the failure-free industries and punishes the failure-prone ones. Thus, on average, it takes longer for the economy to recover and risk-taking behavior is pro-cyclical. This is even with complete financial markets, perfect competition, both well-behaved production technologies and preferences, as well as flexible prices. Other results include conditions under which the behavior of the Credit-Spread is pro-cyclical or counter-cyclical (despite of a constant probability of failure). Proposals for avoiding the aforementioned distortions, and implications for the solution of the Zero Lower Bound problem, are provided at the end.