Alternative roles of consumer confidence in forecasting consumption: Evidence from European countries

Abstract

The predictive power of the consumer sentiment indicator to forecast consumption spending has already been accepted, but the theoretical explanation is still vague. Moreover, some studies have shown that, after adding other macroeconomic variables to the forecasting model, the predictive power of the consumer confidence indicator becomes limited. This paper proposes to distinguish between two predictive channels of confidence data for consumption spending. The first, the income channel, should represent the ability of confidence data to predict the expected future income of consumers. The predictive power of confidence through the second channel should serve for the approximation of consumers’ intentions to consume. According to this distinction, I propose to use the data from initial survey questions separately rather than using one composite consumer confidence indicator. In this way the predictive power of confidence data can be better clarified. I use the data from the European Commission Consumer Survey and, using a sample of 27 countries, confirm that higher predictive power of confidence data could be achieved by using separate confidence data rather than a composite indicator.