Expectations and the stability of stock-flow consistent models

Expectations are usually introduced in macroeconomic stock-flow consistent models (SFC-models from hereon) in an ad hoc way, without much motivation. Moreover, these are usually very simple forms of expectations, and certainly not some form of rational expectations. The implicit assumption is that expectations do not matter very much in these models. However, the way expectations are modelled in SFC-models is very important for two reasons. The first reason is that expectations are very important in understanding the way the economy reacts to a shock, since the stability of the economy is dependent on the nature of expectations. We show for instance that the more backward-looking expectations are, the more stable the economy tends to become. The second reason is that expectations themselves can also be a source of shocks. We show how under certain circumstances optimism or pessimism in expectations can lead to self-fulfilling prophesies. To illustrate the impact of expectations on the stability of an economy we use a simple model, based on the models in Godley & Lavoie, 2007. The model includes a financial sector and government, since we are convinced that the notion of a monetary economy is crucial to understand the impact of expectations on an economy. We also introduce a foreign sector in a very simple way to allow for a better understanding of the multiplier impact of shocks and of foreign reserves on the economy. First we analyse the stationary state solution and analyse its properties. We show that this model is only stable when either the tax rate or government debt is not too high. We also point out the self-fulfilling properties of optimism and pessimism in expectations in this model. Next to that, we show that under "perfect foresight" the model becomes less stable – more restrictions on taxes and government debt are necessary to guarantee stability of the model. However, under naïve expectations the model becomes more stable – there are less restrictions necessary to guarantee stability of the model (due to path dependency). Finally, we introduce the notion of fundamentalist expectations and show how these affect the stability of the model in an intermediate way. In order to introduce adaptive expectations, we conclude our model with some simulation results – analytical solutions cannot be found. We show how adaptive expectations also require an intermediate reaction of fiscal policy to keep the economy stable.