October 29, 2002, started out much like any other Tuesday on Wall Street. The overnight news from around the world was relatively quiet, and the stock market drifted along in the moments after the 9:30 a.m. opening bell. Twenty minutes later, the Dow Jones Industrial Average stood at 8,351 and seemed headed for a calm morning.
But at 10 a.m. sharp, something odd happened â€” a â€œmarket hiccup,â€? in Street slang. The Dow began to fall. By 10:05 a.m. it had skidded 65 points, to 8,286. By 10:26 a.m. the Dow had sunk 113 points. Within the hour, it bottomed out at 8,116, a 170-point swoon.
What had caused the precipitous 2 percent plunge in the Dow? Had an angry OPEC minister threatened to increase oil prices? Had there been a terrorist bombing in the West Bank? Nothing so sinister. At 10 a.m., the Conference Board, a New York City-based nonprofit business research organization, announced its Consumer Confidence Index (CCI), as it does on the last Tuesday of every month. Analysts expected the figure to come in at around 93, which would have reflected the economy's rise out of recession. Instead, the number was 79.4, a 15 percent drop and the index's lowest level in nine years. Stock traders dumped shares. â€œBye, Bye Consumerâ€? blared a headline on CNBC.
â€œConsumers were pessimistic because of a weak labor market, the threat of military action in Iraq and the continued decline in the financial markets,â€? says Lynn Franco, director of the Conference Board's Consumer Research Center. â€œThe message was: An already weak economy could become weaker.â€?
Such a dramatic response to the CCI is no longer unusual. American consumers now spend a whopping $4 trillion a year, accounting for two-thirds of the U.S. economy. For retailers and economists alike, the two major consumer confidence surveys published each month â€” the CCI and the Index of Consumer Sentiment (ICS), managed by the Survey Research Center at the University of Michigan â€” have become must-see windows into the mood of the buying public. Are consumers feeling prosperous or poor? Is the gross domestic product (GDP) heading for the clouds or into the drink? Consumer confidence surveys usually have the answers. The University of Michigan's index is so accurate a predictor that the federal government has included it among its leading economic indicators since 1989. For a half-century, it has consistently outperformed other measures in anticipating changes in the GDP six to nine months before they occur.
â€œConsumer confidence figures are really a measure of how we feel about ourselves,â€? says David Wyss, chief economist for Standard & Poor's, the financial rating company in New York City. â€œIf consumers are worried, Main Street retailers better get worried too.â€?
But the indexes are more than public opinion polls: They are political, economic and social forces of their own. Politicians in Washington, D.C., debate whether a mixed reading means the economy is moving toward recession or recovery. The figures are even watched closely overseas, where business and investment decisions in Europe and Asia are affected by the whims of American shoppers. â€œConsumer confidence is now part of popular culture,â€? says Richard Curtin, PhD, director of the University of Michigan's Surveys of Consumers. In the global economy, the consumer confidence surveys provide a jolt heard round the world.
HOW THEY WORK
Dozens of surveys investigate consumer confidence, but economists trust the CCI and the ICS the most. They each measure attitudes in different ways, but their results tend to be in tandem. The two gauges are so similar, in fact, that many people confuse one with the other â€” much to the chagrin of their organizers, who think their survey is superior to all others.
The Conference Board launched its survey in 1967, when demographer Fabian Linden theorized that consumers offer clues to the economy's outlook faster than indicators like retail sales or the GDP, which require huge surveys and lots of time to analyze. The survey itself consists of just five questions. The first two assess attitudes toward the current state of business and employment and yield the Present Situation Index. The other three questions ask consumers to consider conditions six months into the future, including how much money they expect to make, in order to come up with the so-called Expectations Index.
The Conference Board's monthly readings get reported as three indexes, though they don't always move in lockstep. After the peak of the tech craze in May 2000, the Expectations Index began a steady decline while the Present Situation Index remained flat. â€œPeople were still in the malls, but they didn't feel good about the future,â€? says the Conference Board's Franco. Gradually, consumers began cutting back, leading to a vicious cycle: lower corporate earnings, increased job cuts and more pessimism. That was basically the state of the economy when the news hit that Tuesday morning in October. The Present Index declined 12 percent, and the Expectations Index slumped 17 percent.
To conduct the Conference Board's survey, NFO WorldGroup sends out a questionnaire on the first of the month to a representative sample of 5,000 Americans. There's no incentive to take part in the poll, notes Efrain Ribeiro, NFO's senior vice president for panel services. â€œThe appeal for participants is that this is an opportunity to provide their opinion about the economy,â€? he says. After compiling about 3,500 replies, NFO cuts off the poll, usually around the 20th of the month. The Conference Board then analyzes the data.
The University of Michigan survey queries only 500 people, but some would argue it is even more powerful because of the 200 industry-specific questions that are also part of the poll. Like the CCI, the ICS comes out monthly and is based on five questions used to determine if consumers are confident enough to take on debt for big-ticket purchases. Unlike the CCI, the ICS is drawn from 30-minute phone interviews, conducted by 30 students who ask respondents about business conditions, their personal finances and their spending â€” whether they plan to buy a new car or a refrigerator, for example, in the near future. Business strategists use the figures to forecast sales and plan production levels.
The older of the two polls, the ICS was launched in 1946 by George Katona, an economist who believed consumer confidence could be used as a broad measure of expected changes in income. For the past 25 years, the director has been Richard Curtin, an economist who studied at Michigan under Katona. The survey set 1966 as its baseline of 100, a flush year as it turned out. For years afterward, the index languished below the century mark as the economy slowed and inflation ravaged consumers' buying power. Indeed, Curtin thought the index would never reach its own baseline again. It took the â€œirrational exuberanceâ€? of the tech craze to prove him wrong in January 2000.
â€œThe surprise was not that it happened but that we established the same kind of optimism that we had in the 1960s,â€? Curtin says. â€œIn the '60s, the country had unemployment below 4 percent, record low inflation rates and growth in the GDP and employment. And it took 20 years for people to accept that the economy couldn't perform at that level. But consumers now have to reset the yardstick in terms of the performance of the national economy. We're again in an age of diminished expectations.â€?
The nation is experiencing a kind of Golden Age for consumer confidence surveys. In recent years, the media has reported every monthly reading within seconds of its announcement, and the Internet carries the news instantaneously around the world. Part of the reason is that the major indexes have been around for decades, and familiarity breeds acceptance. But a bigger factor is the increasing recognition of the importance of the consumer in keeping the economy humming. â€œConsumer spending is the key difference between the mild recession we're experiencing and something a lot more serious,â€? says Curtin. â€œSo people have a great interest in knowing how consumers feel. In America, it's one of our social goals to produce confident consumers.â€?
Every month, both at the Conference Board and at the University of Michigan, drama surrounds the release of the latest confidence numbers. Because the indexes move markets, the two groups withhold the results until their official announcements. The Conference Board releases its data precisely at 10 a.m. on the last Tuesday of each month, dispatching its figures to some 3,000 news reporters in a flurry of e-mails, faxes and wire reports. Michigan goes public three days later, also at 10 a.m., but first to a sponsoring consortium of academic groups, government organizations and private companies.
When the clock finally strikes 10 a.m., chaos follows. The stock market does flips. The phones at the Conference Board start ringing with reporters seeking comment. On release days, Franco alone typically logs 100 phone calls and just as many e-mails. She encourages everyone to take the long view when analyzing a monthly index. â€œOne month's drop does not a trend make,â€? she says. â€œYou have to follow the index over several months to see if there's a pattern.â€? Indeed, a comparison of the changes in Michigan's ICS versus the GDP shows consumer feelings shifting more erratically than GDP numbers. (See chart, page 27.) An impatient investor, noting the jump in the confidence indexes last September, may have been fooled into thinking the recession was over; later indexes revealed that to be only a temporary pause on a sinking trendline.
WHAT THE SURVEYS SHOW
The American consumer is a quirky animal â€” moody, unpredictable, a creature as changeable as the weather. Still, the confidence surveys have proven to be pretty accurate economic forecasters.
Most experts agree that that the key factor affecting consumer sentiment is employment. Perceptions are relative, however. The current unemployment rate of 5.7 percent, for example, is low by historical standards. But just two years ago, the figure hovered around 4 percent. The rapid deterioration of the job market leaves consumers feeling insecure about keeping their jobs. And when consumers start to worry about their jobs, they tend to slam their wallets shut. Retail sales drop. Savings levels rise. Over the next year, the manufacturers and sellers of big-ticket items â€” such as cars, homes and major appliances â€” start feeling the pinch. â€œAs the labor market goes, so goes the Confidence Index,â€? says Franco. â€œThe three keys to consumer confidence are jobs, jobs and jobs.â€?
In the past, stock market volatility depressed consumer confidence. But that connection has faded. As the tech bubble grew, as well as after it popped, the Dow's daily upheavals of 200 points or more became commonplace. Americans are no longer scared by it. â€œConsumers have learned to adapt,â€? says Franco. â€œThe shock value has worn off.â€? Geopolitical unrest is another story. After the U.S. invaded Iraq in 1991 during the Gulf War, the CCI fell 17 points in one month. In the month following Sept. 11, 2001, the index again fell 17 points. Delos Smith, a senior business analyst at the Conference Board, suspects that the low index in October 2002 reflected response to the sniper attacks around Washington, D.C., the nightclub explosion in Bali and the hostage crisis in Moscow. â€œThe Consumer Confidence Index is an emotional indicator,â€? says Smith. â€œAnd this is scary stuff. People are jumpy.â€?
Naturally, not all consumers behave the same way, and survey sponsors break down their results by age and income to highlight demographic differences. Not surprisingly, younger people are more optimistic than those who are middle-aged. And middle-aged consumers, in turn, are more optimistic than older people. â€œIt moves in a linear fashion,â€? says economist Curtin, â€œbecause real income tapers off as Americans age.â€? Younger people also are optimistic because they are more likely to change their jobs to increase their income. Apparently, people who are footloose make for confident consumers.
Economic cycles also influence consumer confidence. During prosperous times, research shows there's only a small gap in confidence between young and old Americans â€” and between rich and poor ones, for that matter. During a recession, however, the gaps widen. â€œThose at the lower end of the income spectrum and in the higher age brackets become more pessimistic,â€? explains Franco. â€œThey'll spend less. And depending on the severity of their cutbacks, they'll slow down the economy.â€?
Neither survey reports its findings by race or ethnicity. The Michigan researchers, however, did look at racial differences in previous studies. Their findings: African Americans and Hispanics are less optimistic than whites, though the difference is not as great when levels of affluence are factored in. Again, the gap widens during times of recession.
Both surveys do a better job analyzing their findings by geography. Each one breaks out confidence ratings for the nine census regions. Sometimes the ratings vary a lot by region. During the last Asian economic crisis in 1997, for instance, consumers in the Pacific region of the U.S. felt more pessimistic than those elsewhere. Ever since the securities industry collapsed following the World Trade Center attacks, the mid-Atlantic region has been in a greater funk than other parts of the country. The regional index took a further hit when the snipers around Washington, D.C., scared shoppers in that area out of the malls.
The bear market on Wall Street is also shaking the confidence of one demographic group in particular: Baby Boomers. According to Michigan's Curtin, many Boomers had put too much of their retirement money into stocks, and the recent market crash wiped out big chunks of their savings. As Boomers learn that they'll have to save more and spend less, Curtin estimates the U.S. savings rate will increase by .75 percent.
If there's one thing that the confidence surveys have shown over the years, it's that American consumers bounce back from tough times. Even after the Dow dropped 30 percent in October 1987 and the CCI plummeted 12 percent, the index rebounded two months later. A similar rebound occurred after the Gulf War. When Iraq invaded Kuwait in August 1990, the CCI dropped to 85 from 102. It skidded further, to 55, as Americans mobilized for war. Once U.S. troops won the war, however, confidence rose. The index jumped to 81 from 59 between February and March 1991.
â€œThat's emotion. That's psychology,â€? says the Conference Board's Smith. â€œThat's why the Consumer Confidence Index is more than an economic number.â€?
HOW THE SURVEYS ARE USED
As the purchasing power of American consumers grows, the audience for the confidence indexes is expanding as well. Today, it's not just stock pickers who are following the numbers. Businesses also analyze the data closely for their own economic forecasting, marketing efforts and investment decisions.
Few companies watch the numbers as closely as Standard & Poor's. On that morning last October when the confidence rating plunged, for example, S&P analysts began calculating what the bad news meant for the companies whose debt they rate. Reading the survey as a forecast for a slow Christmas season, the S&P team lowered its revenue expectations for retailers. S&P also revisited its forecast for banks whose credit card business might be hurt by consumers turning frugal. And the moves proved prescient: The 2002 holiday season was one of the weakest in 30 years, according to early retail receipts. Michigan's Curtin began forecasting a weak Christmas last summer.
Some businesses are more sensitive to consumer psychology than others. Manufacturers and retailers of durable goods â€” cars, furniture and major electrical appliances â€” are particularly vulnerable to swings in consumer confidence. The banking and housing sectors are also tied to consumer sentiment. â€œWhen the number drops, it's bad news for the companies that produce and sell big-screen TVs and SUVs,â€? says Standard & Poor's economist Wyss. â€œBut it's also tough on banks because consumers will borrow less money on their credit cards to pay for everything. And consumers will be worrying whether they can pay off their existing credit card debt.â€?
If polls aren't perfect economic forecasters, however, they've still performed impressively over the years. The Michigan index declined sharply in 1979, anticipating the 1980 recession. It dropped again just before the 1981-1982 recession. As Curtin explains, the index tends to signal trends in the GDP six months to a year down the road. â€œIf you plan to buy a car and you decide that your economic future isn't as good as it once was, you'd still buy the car,â€? he says. â€œYour current spending won't change. But as you get more information that your income won't be as high as you expected, then you start cutting back on your spending plans. And that takes time to work into the full economy.â€?
Does this mean the general public is smarter about the nation's financial health than a roomful of pointy-headed economists? Perhaps, says the Conference Board's Franco. â€œConsumers are quick to pick up what's happening in employment, from hearing whether Joe's brother Bob has been laid off. So you get front-line information. And you don't have to wait. If you want to know what's happening in November, we'll know in October.â€?
SURVEY STRENGTHS AND WEAKNESSES
Despite their ability to predict where the economy is headed, the two leading consumer indexes do have their critics. C. Britt Beemer, president of America's Research Group, a strategic marketing company in Charleston, S.C., is one of them. Beemer conducts his own monthly Consumer Mind Reader survey, which asks consumers where they shopped and how much they spent in the previous month. And he agrees with Larry Wachtel, chief market analyst at Prudential Securities in New York, that confidence doesn't always drive retail sales. In fact, Beemer says, some evidence points to the contrary. Take furniture sales: In the past few years, even with high consumer confidence figures, furniture sales grew by a paltry 3 percent annually.
â€œNew mortgages and home refinances are a better gauge of what drives retail sales,â€? asserts Beemer, who's been polling consumers for 23 years. â€œPeople who refinance, so their mortgage drops from $1,200 to $900 a month, are spending that differential to buy something nice for themselves or their home. They're the ones going to Circuit City or Best Buy.â€?
What troubles Beemer most is the type of questions in the consumer confidence surveys: They're not open-ended. â€œPeople say they're feeling pessimistic, but they're not allowed to say why,â€? he adds. â€œIt may be that they don't like the economy or their health insurance went up and they'll have less money to spend. You end up missing the tremors before the earthquake.â€?
Other critics, who did not want to be quoted for attribution, complain about the surveys' sampling techniques. One notes that Michigan's phone surveys miss the Americans in the highest and lowest income brackets. (Many rich people have unlisted phone numbers, while some poor people don't have phones.) Another says that the Conference Board's NFO survey relies on a self-selected pool of people who are motivated to fill out and return their response cards. Says Standard & Poor's Wyss, â€œPeople tend to respond when they've got an ax to grind, which makes them more pessimistic.â€?
Then, of course, there are pollsters who maintain that their own consumer surveys are superior. Gary Langer, director of polling for ABC News, is understandably partial to the ABC News/Money Consumer Comfort Index, based on an ongoing weekly survey of 250 Americans interviewed randomly by phone. Launched in 1985, this index is based on three questions. Respondents are asked to rate the condition of the national economy, their personal finances and the buying climate. Langer distinguishes the ABC/Money poll from the Conference Board's and Michigan's by the fact that his measures only current â€” not future â€” trends. Still, he notes that the ABC/Money poll correlates strongly with the other two surveys: In a 16-year comparison, he found a .89 correlation between ABC and Michigan and .92 between ABC and the Conference Board.
Even Curtin says the Michigan index may become outdated for 21st- century America. Consumers today are better educated than ever before â€” they may even be too sophisticated for such simplified surveys. â€œConsumers used to have positive or negative views about all aspects of the economy, so you could talk about a universal optimism or pessimism reading,â€? he says. â€œBut today they analyze specific factors of the economy. They say that mortgage rates are the lowest they can remember or that businesses aren't investing enough to create new jobs. They're better able to make nuanced judgments about how the economy is faring.â€? In the future, Curtin believes that consumers may need multiple indexes to truly reflect their complex feelings about the economy and their place in it.
Still, it's hard to discount proven winners. The confidence indexes of both the Conference Board and the University of Michigan should continue to help retailers, marketers, bankers and investors make sense of the enigmatic American consumer. And that's no small challenge in these uncertain times. In the month after that sour October morning, both indexes improved by more than 4 percent. Economists pointed to a slow but steady rise in the stock market, and recent interest rate cuts by the Federal Reserve Board, as reasons for Americans' better spirits. Further, they reasoned that consumers would continue spending to keep the economic recovery from fizzling out.
But what do the flip-flopping reports really say about where American consumers are heading? Time will tell. However, if the history of consumer confidence surveys is any guide, investors and retailers better fasten their safety belts â€” 2003 is going to be a bumpy ride.
Michael J. Weiss, a Contributing Editor at American Demographics, is the author, most recently, of The Clustered World.
TALE OF TWO SURVEYS
The two leading consumer confidence surveys, the Consumer Confidence Index conducted by the Conference Board and the Index of Consumer Sentiment conducted by the University of Michigan, provide a monthly gauge of whether Americans are feeling giddy or gloomy â€” mood swings that influence economic activity. Although both surveys are conducted monthly and ask five questions, there are many differences in how they work.
|Sponsor||Conference Board||University of Michigan|
|Number of respondents||5,000 surveyed; about 3,500 analyzed||500|
|Survey method||Mail panel||Random phoning|
|Focus||Job security||Accepting debt for big-ticket purchases|
|Time frame||Present; six months into the future||Present; one year into the future|
|Release date||10 a.m., last Tuesday of the month||10 a.m., last Friday of the month|
|Questions||1. An appraisal of current business conditions*||1. We are interested in how people are getting along financially these days. Would you say that you (and your family living there) are better off or worse off financially than you were a year ago?|
|2. Expectations regarding business conditions in six months||2. Looking ahead, do you think that a year from now you (and your family living there) will be better off financially, worse off or just about the same as now?|
|3. An appraisal of current employment conditions||3. Turning to business conditions in the country as a whole, do you think that during the next 12 months we'll have good times financially, or bad times or what?|
|4. Expectations regarding employment conditions in six months||4. Looking ahead, which would you say is more likely â€” that in the country as a whole we'll have continuous good times during the next five years or so, or that we will have periods of widespread unemployment, or depression or what?|
|5. Expectations regarding total family income in six months||5. About the big things people buy for their homes â€” such as furniture, a refrigerator, television, and things like that â€” generally speaking, do you think now is a good or a bad time to buy major household items?|
|*Note: For each of the five questions, there are three response options: positive, negative and neutral.||Sources: Conference Board; University of Michigan|