Consumption expenditures in economic impact studies : an application to university students

We raise the issue of how appropriately to attribute economic impact to consumption expenditures. Despite the salience of the topic for applied economics it has not received much explicit attention in recent literature. In Input-Output analysis consumption expenditures are either treated as wholly endogenous or wholly exogenous. For many applications, such as those focusing on the impacts of tourism, benefits systems or student expenditures, these binomial assumptions are not satisfactory. Therefore practical necessity has resulted in conventions ("rules of thumb") for dealing with such cases. We argue that consumption is neither wholly endogenous nor wholly exogenous but that the degree of this distinction is rather an empirical matter. To deal with this issue we set out a general model for the treatment of consumption expenditures. We illustrate its application through the case of university students in Scotland and determine the exogeneity of their expenditures drawing on survey evidence. Students are a particularly useful example as we can analyse the difference in impacts between individual student groups and how the impacts of students at particular institutions are influenced by the composition of the student body. Furthermore we take into account the binding budget constraint of public expenditures (as is the case for devolved regions in the UK) and examine how this affects the impact attributed to students' consumption expenditures.