The Consumer Confidence Thing
David Andolfatto
October 3, 2001

“We do have to fear fear itself, it matters, it affects consumer spending…if people aren’t confident, people don’t spend and the economy is damaged by  that.” Bill Cheney, chief economist at John Hancock Financial Services. Quoted in The Globe and Mail, September 26, 2001, pg. B11.

Consumers Still Hold Recovery in Their Hands: North American Economy Relies on Shoppers' Psyche
 Headline: National Post, September 29, 2001, pg. IN1.

Mr. Cheney’s statement encompasses three commonly held (if ill-defined) ideas: (1) Our economic well-being depends on consumer spending; (2) Consumer spending depends on consumer confidence; and (3) Confidence depends on self-fulfilling beliefs (if we all just believe hard enough, it will come true). What are the foundations for these widely held sentiments?

The first sentiment is what many of us learned in our undergraduate macroeconomics course many years ago. (Or, if we did not learn it directly in university, we learned it from those who did, and subsequently parroted the idea ad nauseam). It is an hypothesis that has evolved from ideas first put forth by the great economist John Maynard Keynes as part of his attempt to explain the Great Depression. Unfortunately, this radical and provocative idea (so worthy of academic consideration) has become nothing more than an uncritically accepted truth by many people of influence (in particular, journalists, business analysts, and policymakers). The idea has become so entrenched in the minds of people outside of academia that even to criticize it seems tantamount to criticizing the obvious. (It is interesting to note, however, that it wasn’t so obvious until an academic first proposed it as an hypothesis).

For the uninitiated, the idea goes something like this: If consumers would only spend more of their incomes, then firms would hire more workers to meet this added demand. These newly employed workers would then spend their newly acquired earnings, leading to yet greater demand and yet higher incomes all around, including the consumers who initially increased their spending. (Note: It is not considered polite to ask whether this process can continue indefinitely or to point out its resemblance to a Ponzi scheme).

One problem with this argument is what it implicitly assumes about the income not spent by consumers. In particular, it assumes that savings somehow represent “idle” resources (perhaps destined to be stuffed under the mattresses of inexplicably frightened consumers). In fact, household savings are channeled through financial markets in the form of business loans and equity, which are then spent on new capital expenditures (investment). It is difficult to see why a dollar spent in the consumer goods sector should be any more “stimulative” than the same dollar spent in the capital goods sector (should it matter whether a service worker or construction worker gets paid this dollar?). Consider the following argument: If consumers would only spend a smaller fraction of their incomes, then more resources would be available for firms interested in expanding their capacity. To add capacity, firms must hire workers; this added payroll is financed by the cheaper loans secured by firms (thanks to the added supply of saving). These newly employed workers then go out and spend their earnings (prudently saving a fraction of it, of course), which leads to higher demand, which leads to higher earnings, and so on. Furthermore, the added capital expenditure has the virtue of increasing the economy’s productive capacity, unlike an increase in consumer spending, which serves only to deplete it.

When presented with this argument, most people stop and scratch their heads. What is confusing here is that while these two arguments contradict each other, they nevertheless sound equally plausible. I am not claiming that either argument is correct. My point here is simply to demonstrate that Mr. Cheney’s first sentiment is not something that is as obvious as it first sounds and that we should keep our minds open to alternative explanations.

Let us now consider Mr. Cheney’s second sentiment, which is that consumer spending depends on consumer confidence. The main problem with this statement is that no one really knows what is meant by the term “confidence” in this context. Does increased confidence mean less uncertainty? And if it does, then less uncertainty over what? (Consider the statement: I am now confident that the economy will collapse). Or does confidence refer to a consensus consumer forecast of future earnings levels? (And if it does, why not just say so?). This latter notion seems consistent with what many surveys of “consumer confidence” are trying to measure, so let’s run with this. (See:

The second sentiment can thus be rephrased as follows: If consumers expect higher earnings in the near future, then they are likely to increase their spending today. I am not aware of any economic theory or evidence that would contradict this statement, so I will not dispute it. The key issue here is not whether an increase in consumer confidence leads to more spending, but whether the level of confidence is in some sense justified by “fundamentals” or whether confidence depends on “unsubstantiated” (but possibly self-fulfilling) beliefs. Mr. Cheney’s statement, that we must fear fear itself, suggests that he must believe in the latter hypothesis.

Taken together then, Mr. Cheney’s argument runs as follows. Suppose that consumers suddenly (and for some unexplained reason) believe that their future earnings are to fall (i.e., there is a decline in consumer confidence). Then consumer demand will fall, which leads to lower sales, lower production, lower employment, and hence lower earnings. The lower earnings that result are fully consistent with what was expected; in this sense, a decline in consumer confidence can become “self-fulfilling”.

The main weakness with this argument lies in what it leaves unexplained: Why should consumer confidence fall in the first place? One answer to this question is that there is no explanation. This is the view that Keynes took when he referred to such unexplained fluctuations in consumer and investor sentiment as “animal spirits”. Such an hypothesis is attractive to many people for many different reasons. For example, some people feel comfortable with interpreting the economic behavior of others (or even themselves) as being driven by emotion rather than logic. Others who enjoy imposing their own preferences on their neighbors also find the hypothesis attractive, since it provides them a justification for doing so. (E.g., I think that the government should tax everyone and spend it on things that I like…this will obviously benefit all consumers because it will increase their confidence and hence their incomes).

The other way to interpret a decline in consumer confidence is that the change reflects new information concerning the evolution of market fundamentals, such as the arrival and spread of new technologies, together with information concerning their profitability. For example, the late 1990s witnessed a tremendous boom in productivity and earnings, driven largely by the IT revolution. As the spread of these technologies slowed, so did productivity and earnings. Consumers began to revise downward their forecasts of earnings growth, not for some unexplained reason, but because it was the sensible thing to do in light of these new developments. Over time, it became apparent that many IT investments did not fulfill their initial promise (Memo: there is no God-given rule stating that investments must always succeed). When this happens, labor and capital need to be reallocated to alternative uses, which is a process that takes some time. Individuals affected by these adjustments are naturally going to revise downward their earnings forecasts. Again, this decline in confidence is not driven by emotion, but by the reality of the adjustments that take place every day in dynamic economies. This is not to say that the prospect of losing one's job can or should not elicit an emotional response; what I simply mean is that being emotional is not going to help avoid what is inevitable. Consequently, the calls to make people spend when the reality calls for prudence is likely to make things worse, not better. Consider, for example, what happened in Ontario during the deep 1991 recession. The provincial government (Bob Rae's NDP government) undertook a large expenditure program designed precisely with the intent of boosting demand in order to "kickstart" the economy. The effect of the program was to divert significant resources to projects of questionable value, all of which was financed by running up the provincial debt to all-time highs. The workers (and political friends) who were paid these monies were no doubt made somewhat better off, but one cannot seriously claim that the provincial economy as a whole benefited.

So which of these two interpretations is correct? Unfortunately, it is impossible to say for sure. In all likelihood, they both contain an element of truth. But my object here is not to persuade you of the plausibility of one hypothesis over another. Rather, my goal is to point out that what currently passes for “conventional wisdom” in the financial pages of newspapers should not be accepted uncritically by intelligent readers. Conventional wisdom simply reflects a consensus interpretation of how the world works; it does not necessarily reflect “truth”.
  Hits Since October 24, 2001:
FastCounter by bCentral