Academic journal article Economic Inquiry

Consumption, Sentiment, and Economic News

Academic journal article Economic Inquiry

Consumption, Sentiment, and Economic News

Article excerpt

I. INTRODUCTION

Since the introduction of surveys of consumer sentiment after World War II, it has been a puzzle why measures of consumer sentiment have predictive value for changes in consumer spending. Because sentiment measures are based on questions about households' expectations for changes in business conditions and their personal finances, they may reflect emergent shifts in fundamental determinants of spending, like income, not yet apparent in economic data. It is also possible that they reflect shifts in outlook or mood exogenous to changes in fundamentals, but that influence people's willingness to spend. An important study by Carroll, Fuhrer, and Wilcox (1994) found that, after controlling for fundamental determinants of spending, sentiment still has a small but significant value for predicting future changes in spending. (1) The finding that "nonfundamental" changes in consumer sentiment affect consumption is consistent with recent developments in the real business cycle (RBC) literature, in which it has been shown that models with nonfundamental demand shocks and indeterminate equilibria can match observed patterns in macroeconomic data relatively well (Benhabib and Farmer 1999; Benhabib and Wen 2004; Farmer and Guo 1994).

It remains poorly understood, however, what "nonfundamental" fluctuations in consumer sentiment reflect. The traditional interpretation is that they capture intangible psychological factors that vary over time and have independent causal effects on consumers' willingness to spend. (2) Such factors are variously described as "animal spirits," optimism or pessimism, outlook, mood, or emotional response to uncertainty. However, there is limited information on whether all nonfundamental fluctuations in consumer sentiment are "psychological" in this sense or whether other types of intangible factors are also involved. An important possibility is that some part of nonfundamental fluctuations in sentiment reflect discrepancies between what people think is happening in terms of economic fundamentals and actual changes in conditions. Notably, it is commonly questioned whether the news media cover changes in economic conditions in ways that accurately represent incoming information on them, which is important because the media are the public's primary source of information on the economy. (3) For one, pressures to confine news coverage to stories with demonstrable "news value" mean that only relatively dramatic economic news tends to get covered. (4) For another, the probability that given economic news gets covered depends on what else is going on in the news landscape. In addition, the salience of economic news to the public may vary over time; in particular, people are known to pay more attention to the state of the economy in the months before presidential elections, so that a given news item may become more widely known then than at other times. (5)

If media representations of changes in economic conditions are not consistently in line with reality, then the perceptions of economic conditions with which businesses and consumers operate may also at times be misaligned with economic fundamentals. An example given in this respect concerns the recession of the early 1990s; although it was by all measures a relatively mild downturn, some analysts thought that unduly negative news stories created anxiety among consumers, which in turn depressed spending and delayed the onset of recovery (Blanchard 1993; Doms and Morin 2004; Kurtz 1990). Conceivably, then, at times when media coverage makes economic conditions seem too strong or too weak relative to incoming data on economic fundamentals, it is possible that consumers may spend more or less than one would expect them to, given the "hard data" on economic fundamentals--but as a result of their information sets being misaligned with fundamentals, not because of exogenous changes in "mood." This indicates the problem for inferences that nonfundamental causes of aggregate fluctuations are "psychological" in nature, because they may instead or also reflect variations in the correspondence between people's perceptions of economic conditions and what they actually are. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.