The Role of Banks: What Can We Learn from Behavioural Macroeconomics?

Yuemei Ji & Paul De Grauwe

What were the causes of the global financial crisis in 2007-08? What economic policies should economists and policy makers design to prevent such crisis that have damaged our financial and economic system from happening again? Conventional macroeconomic models (even the most recent ones) do not offer good answers. We are told that the economy was hit by an unprecedented external shock and nothing could have be done about it.

In this Rebuilding Macroeconomics project we want provide better answers to these questions. We will develop a behavioural macroeconomic model that includes a banking sector. We introduce two novelties allowing us to fundamentally depart from the conventional macroeconomic modelling approach.

The first novelty is that we will allow agents to have cognitive limitations in making forecasts. If there is anything we have learned from the financial crisis, it is that this crisis was made possible by the fact that human beings do not understand the complexity of the world in which they live, and in particular the nature of the risks. Instead, their cognitive abilities are limited. This has not prevented conventional macroeconomic modellers from maintaining the assumption of rational expectations, i.e. an assumption that disregards these cognitive limitations of individuals. This conventional way of modelling expectations has constrained us in developing new insights on how boom–bust business cycles are generated in the context of a banking system in which agents may take excessive risk.

In our model, we will assume agents use simple rules (heuristics) to guide their behaviour. Agents also evaluate the performance of these rules against alternative rules and are willing to switch to the better performing one. This modelling approach is based on insights obtained from other disciplines such as psychology and brain science.

The second novelty is that we do not regard banks as pure financial intermediaries, where they are assumed to only collect money from savers and lend it to borrowing firms and households. In such a setting, there is always an upper limit constraining how much money banks can lend to the borrowers (firms and households) and hence the impact of banks on the macroeconomy is perceived to be very limited.

In our model, banks are assumed to be money creation and destruction institutions. When banks extend a new loan they at the same time create a new deposit. As a result, banks’ loan supply is not constrained by the total saving of the economy. It will be shown that this adds an important pro-cyclical feature into our model: during a boom when loan demand is high, banks accommodate this higher demand by increasing loan supply and in so doing create more money which in turn tends to intensify the boom. The opposite will be shown to occur during a recession.

One important implication of this modeling choice will be that there is heterogeneity among individuals. These individuals do not all use the same rules of behaviour, but they influence each other, and as a result collective movements of “optimism” and “pessimism” (described by John Maynard Keynes as “animal spirits”) can emerge endogenously, driving the boom-bust business cycle.

This departure from the conventional way of modelling will not only allow us to develop insights on how business cycles are generated in the context of a banking system, but also guide monetary policies that ensure financial and economic stability. In mainstream macroeconomic models, the role of the central bank is a minimalist one: the central bank limits itself to keeping the price level stable and after some exogenous shock it should help the market system to smoothly find its equilibrium. We will analyse whether this minimalist view of the central bank’s responsibility is appropriate and whether there is a need to “stabilize an otherwise unstable system”, through both conventional and non-conventional policies.