Consumer Confidence Index

The Consumer Confidence Index is a monthly index summarizing U.S. consumers' assessment of the economy. Its publisher, The Conference Board, opens its own technical note with the purpose: the index "is a barometer of the health of the U.S. economy from the perspective of the consumer," based on "consumers' perceptions of current business and employment conditions, as well as their expectations for six months hence regarding business conditions, employment, and income."

The naming is worth getting straight, because three related names are in circulation. The Consumer Confidence Survey is the instrument, the Consumer Confidence Index is the headline number computed from it, and the University of Michigan's Index of Consumer Sentiment is a separate product from a separate organization. Only the first two belong to The Conference Board. A story reporting that "consumer confidence fell" may be describing either publisher's series, and the two can move differently in the same month because they ask different questions of different samples.

The instrument is small and completely specified, which is unusual and makes the index easy to reason about. The publisher lists the five questions behind it as respondents' appraisal of current business conditions; their appraisal of current employment conditions; their expectations regarding business conditions six months hence; their expectations regarding employment conditions six months hence; and their expectations regarding their total family income six months hence. Note what is absent: no question asks about prices, interest rates, the stock market, or the respondent's own spending plans. The index measures perceptions of business conditions, jobs and household income, and nothing else.

How those answers become a number explains why the level should be read with suspicion. Each question offers three responses, positive, negative or neutral. For each question the positive count is divided by the sum of the positive and negative counts to give what the publisher calls a "relative" value, and then, in its words, "the average relative value for the calendar year 1985 is then used as a benchmark to yield the index value for that question." Nineteen eighty-five was chosen as a base year, not as a neutral or normal condition, so an index of 100 means "consumers answered about as positively as they did on average in 1985" and no more than that.

The publisher also quantifies its own break in the series, which is the kind of disclosure most survey producers do not make so precisely. The survey moved online and changed providers in May 2021, having run by mail through TNS to November 2010 and then through Nielsen. To handle the transition the publisher restated a window of months and reports the size of the shift: "The transition effect of changing the survey provider for the national CCI series is a decrease of 1.8 points for the aggregate index for that month," using January 2021 as the transition month, and it states that "the historical series before January 2021 remains unchanged." A reader comparing a reading from before that point with one from after is comparing across a documented discontinuity of known size.

On sample and cadence, the details matter for how much weight a single month deserves. The survey has run since 1967, first every two months by mail, and monthly since June 1977. The publisher reports that the online transition increased the number of complete responses "to 3,000 per month," collected in roughly four weekly waves, with the sample closing for the preliminary estimate about a week before release and later responses folded into a final estimate published alongside the following month's data. So the first number reported each month is preliminary by design, and it is a count of completed responses from a panel rather than a count of households contacted.

Consumer expectations also reach the business-cycle indicator family, which is why this survey turns up in discussions of leading economic indicators. The Conference Board's leading composite carries ten components, and one of them is listed as "average consumer expectations for business conditions." Survey answers about the next six months are therefore an input to a turning-point index as well as a headline in their own right.

The arithmetic is stated by the publisher in three lines: the Consumer Confidence Index is the "average of all five indexes," the Present Situation Index is the "average of indexes for questions 1 and 2," and the Expectations Index is the "average of indexes for questions 3, 4, and 5." Because the headline is a simple average of five equally weighted question indexes, three fifths of its weight sits on expectations about the next six months and two fifths on current conditions.

A hypothetical shows why the split is worth watching. Suppose the two current-conditions question indexes come in at 130 and 120, and the three expectations indexes at 70, 65 and 60. The Present Situation Index is (130 + 120) ÷ 2 = 125. The Expectations Index is (70 + 65 + 60) ÷ 3 = 65. The headline is (130 + 120 + 70 + 65 + 60) ÷ 5 = 89. That single number, 89, sits much closer to the pessimistic half, because the majority of the weight is there, and it conceals a household view that current conditions are good and the outlook is poor. Reporting the headline alone loses the most informative thing the survey found.

What it is good for

• It arrives early. The publisher describes its series as "among the earliest sets of economic indicators available each month."
• The instrument is fully published: five questions, three response options, a stated averaging rule.
• The split between present situation and expectations lets a reader see whether households are unhappy about now or about what is coming.
• The publisher discloses and quantifies its own methodology break rather than smoothing over it.

What it does not tell you

• The level means nothing on its own, because the base is an arbitrary calendar year rather than a neutral condition.
• It asks nothing about prices, rates, markets or spending intentions, so it is not a read on any of those.
• The first monthly figure is preliminary and is revised into a final estimate published a month later.
• It records what people say about business conditions, not what they do with their money, and the two need not move together.
• It is one of at least two widely reported consumer surveys, and a headline that names neither publisher cannot be checked against either.