INTRODUCTION
This chapter discusses work which applies the methods of stochastic dynamic programming (SDP) to the explanation of consumption and saving behaviour. The emphasis is on the intertemporal consumption and saving choices of individual decision-makers, which I will normally label as ‘households’. There are at least two reasons why it is important to try to explain such choices: first, it is intrinsically interesting; and, second, it is useful as a means of understanding, and potentially forecasting, movements in aggregate consumption, and thus contributing to understanding and/or forecasts of aggregate economic fluctuations. The latter motivation needs no further justification, given the priority which policy-makers attach to trying to prevent fluctuations in economic activity. The former motivation – intrinsic interest – is less often stressed by economists, but it is hard to see why: it is surely worthwhile for humankind to improve its understanding of human behaviour, and economists, along with other social scientists, have much to contribute here.
The application of SDP to household consumption behaviour is very recent, with the first published papers appearing only at the end of the 1980s. Young though it is, this research programme has already changed significantly the way in which economists now analyse consumption choice, and has overturned a number of previously widely held views about consumption behaviour. Any research programme which achieves such outcomes so quickly would normally be judged a success, and in many respects this is an appropriate judgement here. The judgement needs to be qualified, however, on at least two counts. First, some of the ideas which the SDP research programme has overturned, although previously widely believed by mainstream economists, were never subscribed to by those working outside the mainstream. Non-mainstream economists might argue that the SDP programme has simply allowed the mainstream to catch up with their own thinking. Second, there is room for doubt that SDP methods really capture at all well the ways in which humans actually make decisions.
These issues are considered in the rest of this chapter. Section 2 reviews the development of economists’ thinking about consumption behaviour since the time of Keynes, and places the SDP programme in this longer term context. Section 3 looks in more detail at some of the most prominent contributions to the SDP research programme. Section 4 considers criticisms of the SDP programme, and looks at other possible approaches to modelling consumption behaviour.