This paper studies the growth and welfare effects of macroprudential regulation in an overlapping generations model of endogenous growth with banking and agency costs. Indivisible investment projects combine with informational imperfections to create a double moral hazard problem à la Holmström–Tirole and a role for bank monitoring. When the optimal monitoring intensity is endogenously determined, an increase in the required reserve ratio (motivated by systemic risk considerations) has conflicting effects on investment and growth. On one hand, requiring banks to put away a fraction of the deposits that they receive reduces the supply of loanable funds. On the other, a higher required ratio raises incentives to save and mitigates banks' incentives to monitor, thereby lowering monitoring costs and freeing up resources to increase lending. In addition, it may mitigate the systemic risk externality associated with excessive leverage. This trade-off can be internalized by choosing the required reserve ratio that maximizes growth and welfare. However, the risk of disintermediation means that in practice financial supervision may also need to be strengthened, and the perimeter of regulation broadened, if the optimal ratio is relatively high.

This article reinterprets the origin and evolution of the Basel Accords. We argue that the Basel I paradigm was very different from the regulatory approaches that had been applied successfully in most European countries since the Second World War. Banking systems relied on a multitude of tools including entry restrictions, liquidity rules, reserve requirements, deposit rate ceilings, lending and investment restrictions, combined with hands-on supervision and discretional interventions. By focusing exclusively on capital adequacy and credit risk, Basel I shifted attention in a very different and somewhat unexpected direction. The Basel regulations are often understood as a reaction to the bank failures of the 1970s and 1980s, but in fact their capital adequacy rules would not have prevented these failures. Indeed, even today, several of these risks are still not addressed by Basel updates, suggesting that the original and current proposals have a rather different raison d’être, placating political constituencies and banking interests.

This article proposes to insert a bank capital channel (BCC) in the theoretical framework proposed by Bernanke et al. (1999) (BGG), in order 1) to evaluate the mean features of the BCC and 2) to assess the benefits of a countercyclical prudential measure. In a costly state verification framework, banks are charged a financing premium on their external funds which depends on the structure of their balance sheets and which then fluctuates in a counter-cyclical way. This premium is lastly passed on to firms' credit conditions. Simulations confirm that shocks are more amplified than in the BGG model, and confirm the importance of the financial context for macroeconomic volatility. Given the effects of this channel and the ineffectiveness of monetary policy in a capital crunch context, the model finally allows a conclusive evaluation of the advantages associated with a preemptive and counter-cyclical prudential measure in a DSGE model to be made.

The 2007–8 banking crisis in the advanced economies has exposed deficiencies in risk management and prudential regulation approaches that rely too heavily on mechanical, albeit sophisticated, risk management models. These have aggravated private and economic losses. While fiscal costs were at first limited, it remains to be seen to what extent the taxpayer will be protected. Policymakers and bankers need to recognise the limitations of rules-based regulation and restore a more discretionary and holistic approach to risk management.