Activity-based travel demand models have become state-of-the-art over the last decade. A crucial assumption of these models is that travel is to be understood in the context of participation in activities. As a consequence, activity participation and time use have become important issues, resulting in the development of models in these respective domains. An issue that has received limited attention is how and to what extent monetary budgets impact on the decision to engage in activities and the time spent in them.

Activity generation models include household income in the explanatory variables, but do not describe how trade-offs in the allocation of monetary household budgets lead to prioritizing one activity over another. Micro-economic models of time allocation account for monetary budget constraints, but make limiting assumptions about the monetary input per unit time allocation.

There are several reasons for trying to describe the impact of monetary budgets on activity participation and time use in a more comprehensive way. First, changes in household income are probably the key factor underlying the structural changes in travel behaviour we have witnessed over the past decades. Hence, insights in how additional income is spent on various purposes can help us to predict structural changes in travel behaviour due to expected changes in income per capita. Second, increasing diversification in lifestyles implies that assuming fixed expenditure rates associated with activities is a too limiting approach, and that instead the expenditure rate is a deliberate household decision.

To deal with the above issues, this paper proposes a random utility maximisation model of the allocation of time and money (household income) on the household level. It assumes that households aim at allocating time to activities such that the overall utility is maximised. The utility derived from activity participation is, however, affected by the allocation of money. Monetary inputs may have two effects: first, input of money may influence the quality of activities (e.g. eating in a exclusive restaurant vs. eating in a cheap restaurant). In addition, monetary inputs may impact on the efficiency with which an activity is carried out. For instance, one may choose to pay more for a quicker service (transport, banking) or invest in time saving equipment. With respect to the latter, the model distinguishes between direct input of money into activities and indirect input of money via facilities (e.g. equipment, transport modes, dwelling). A limiting factor, though, is that the aggregated money inputs may not exceed the household income.

The paper proposes various model specifications and demonstrates their predicted impact of changes in monetary budgets and prices in numerical examples. In addition, a simplified version of the model is tested on a 1998 Dutch database containing data about households time use and expenditure patterns.