An assessment of the contribution of consumer confidence towards household spending decisions using UK data

The European Commission’s consumer confidence indicator (CCI) is assembled from responses to four questions about individual and general economic prospects which form part of the EU’s Consumer Survey. However, concerns may be raised about whether the four components should be constrained to exerting the same influence in a forecasting model of household consumption. Also, in this context, it would seem to be appropriate to permit a role to other information that is obtained from the EU survey. Consequently, in this article, different regression functions are specified in order to assess whether there is any gain to be achieved in predictive accuracy from adopting a more flexible approach towards using the data from the EU questionnaire. With an emphasis upon parsimony, an econometric analysis is performed in conjunction with UK quarterly data on household consumption expenditure. For two categories of spending, it is discovered that the quality of forecasts benefits from having undertaken disaggregation involving survey data beyond those which contribute towards the calculation of the CCI. Indeed, the respective consumption variables (relating to non-durable goods and durable goods excluding vehicles) are seen to be associated with relatively volatile behaviour over the forecast interval, 2008–2013.