Research Discussion Paper – RDP 2016-10 The Effect of Consumer Sentiment on Consumption Abstract
This paper seeks to identify whether changes in consumer sentiment have a direct effect on consumption. In order to demonstrate a causal effect running from sentiment to consumption we need to identify changes in sentiment that are likely to be unrelated to other factors simultaneously affecting sentiment and consumption. To do this, we take advantage of the fact that immediately after elections at which there is a change of government consumers supporting the winning party report substantially more optimistic expectations about both personal and general economic conditions than supporters of the losing party. Following a change of government, we find robust evidence that supporters of the winning party report higher spending intentions than supporters of the losing party, providing evidence that consumer sentiment has a causal effect on consumption. We also find evidence that, following changes of government, motor vehicle purchases increased by relatively more in postcodes with a greater share of votes for the winning party. This provides evidence that self-reported spending intentions are indicative of actual consumption behaviour. Because the share of supporters for the government and the opposition is similar, the variation in sentiment that we use for identification is not evident at the national level. Thus, our results do not imply that changes of government have a noticeable effect on aggregate consumption. However, they do imply a causal effect that can run from sentiment to consumption.