Consumption, leisure and the demand for money and money substitutes

The purpose of this research is to develop and test a model of the demand for money within a general optimising model of household behaviour. The framework adopted is the direct utility approach. The services of money and money substitutes, along with the services of consumption goods (durable and non durable) and leisure are assumed to enter as arguments in the representative household's utility function. The theoretical part of the thesis consists of applying the tools of modern utility theory to the particular problem of the demand for money. The development and solution of the model provides a clear basis for interpreting the demand equations used in estimation, and also makes explicit various assumptions implicit in previous empirical models in this area. In particular, derivation of the rental price of money and money substitutes serves to clarify the role of expectations and the relationship between the rental prices of money and goods within the direct utility model. The major part of the thesis consists of applying the model to annual Canadian data for the period 1947-1974. A substantial portion of the empirical contribution is the construction of data series consistent with the theoretical framework of the model. We differ from other researchers in this area in using the ARIMA model to take expected capital gains into account when constructing rental price series for durable goods. Three different groups of models are examined empirically. The first group contains only consumption goods and leisure. The second group includes aggregate 'money' and aggregate 'near money' along with consumption goods and leisure, while the third group contains only 'liquid assets', i.e., disaggregated components of 'money' and 'near money.' The demand equations for each model are derived from a Gorman polar form representation of the indirect utility function, and are evaluated using a constrained estimation technique. The presence of autocorrelation is explored, and the model tested for parametric stability over time. Tests of the restrictions implied by the theory of utility maximising behaviour and of homotheticity are performed. The estimated models were found generally to be consistent with the underlying theory, and also provided some useful information. Money has an expenditure elasticity less than one, while near money is a luxury good. There is no evidence of substitutability between aggregate money and aggregate near money; however, some substitutability is reported between chartered bank personal savings deposits, and trust and loan company savings deposits.Arts, Faculty ofVancouver School of EconomicsGraduat