What Scented Candles Say to an Economist

ECONOMISTS pay too little attention to what is happening around them. Our record at forecasting economic growth has not recovered from the failure to predict the 2008 crash, yet the profession has not done enough to improve things.

Growth forecasts for gross domestic product in the United States at the end of this year vary from about 1.75 percent to 3 percent — a good measure of the lack of consensus. We need a wider variety of indicators to help us take a more accurate reading of the economy. Some of these might seem frivolous, but paying close attention to worldly detail could make forecasting more reliable.

Take women’s fashions, for instance. One playful theory, born in the days of flapper dresses in the 1920s, is that the height of hemlines provides a surprisingly good indicator of the state of the economy. Long skirts are associated with hard times; but when the economy booms, as in the ’20s or ’60s, skirts get shorter. So if you see skirts sweeping the floor next season, it might be time to start worrying — assuming there’s a discernible fashion trend.

The hemline is only one of myriad indicators that economists have used to divine prospects for growth. Another anecdotal measure is the number of cranes visible on the skyline, their stately dance a prominent visual signal of the underlying pace of construction activity.

Spending on luxury items is another example. During a boom, sales of fast cars, expensive paintings, prime real estate and diamond necklaces all soar, as do their prices. A chart of prices realized on modern art at auctions looks like the stock market. One well-known index, the Modern Art 100, rose by two and a half times in the two years before the collapse of Lehman Brothers in 2008; it then lost half its value in 18 months.

Less obvious are trends in retailing. When the good times roll, people decide that their great idea for a specialty store is viable. Thus booms bring all those boutiques selling just one type of good: socks or scented candles or freshly squeezed juices. But like flowers that display the behavior known as nyctinasty — opening to the sun’s light and warmth — they close as soon as the skies darken and things start to cool.

Anarchists in East London were onto the symbolism of such retailers when they recently attacked a cafe selling solely breakfast cereal varieties for about $5 a bowl. No matter that it was a struggling small business, to the anti-gentrification protesters it was catering to the decadent tastes of the 1 percent.

There are other indicators we’re all familiar with, such as how easy, or otherwise, it is to get restaurant reservations or tickets for shows (box office grosses for Broadway shows have been climbing steadily since 2008). It used to be that the difficulty of finding a cab was another sign — but one whose usefulness may be being eroded by competition from Uber; perhaps the prevalence of surge pricing is its modern counterpart.

One of my favorites is how many “help wanted” signs appear in the windows of stores and restaurants. While nobody collects these statistics, the impressionistic evidence can amplify formal data about vacancies. These have been on an upward trend since early 2010, pointing toward employers’ being so desperate for workers that, even in these times of online job search, they’re willing to hire someone who walks in off the street.

All these signs, statistical or anecdotal, tend to be clearest when the economy is growing fast, because they reflect spending that is sensitive to the state of the business cycle. Economists working in governments and central banks need indicators that give prompt warning of the trends, supplementing these with informal signs that can act as early alarm signals. By the time official stats are available, the last wave of specialty stores will have closed down — then it’s too late either to cool the boom or to temper the recession quickly by adjusting interest rates or using fiscal measures.

Better yet would be signals that can foretell the future of the economy, but there are fewer of these “leading indicators.” One of the earliest attempts to predict the business cycle looked at the pattern of sunspots. The British economist William Stanley Jevons argued in 1875 that sunspot activity was linked to subsequent weather patterns, and this in turn affected corn harvests. That sounds bizarre now, and was never a successful forecasting tool, but it was a not unreasonable attempt in an economy far more reliant on agriculture than ours is today.

Since then, stock markets have been the classic leading indicator, as investors are supposed to foresee the economic trends — with every incentive since they’re betting on them. But these days, greater skepticism about the rationality and efficiency of financial markets has undermined confidence in their predictive power.

Still, another standard financial sign of an impending downturn is the relationship between short- and long-term interest rates, known as the yield curve. Normally, this has an upward slope because investors need a higher return the longer their money is tied up. However, if they expect the economy to weaken soon, and the Federal Reserve to respond by cutting interest rates, the yield curve will slope down.

This held true before the 2008 financial crisis: There was a negative yield curve for months. The problem was that hardly anybody paid attention to it. (At present, the yield curve bends reassuringly upward.)

Of course, the one figure that everyone takes seriously as a key economic barometer is the quarterly change in G.D.P. Yet it is one of the least useful. It lags events, is frequently revised and provides no meaningful detail.

It cannot tell us, for example, how people on low incomes, as well as high ones, are doing; nor how Nevada is growing compared to Massachusetts. G.D.P. almost certainly fails to capture newer areas of economic activity, such as today’s digital innovation — so other sources of information are needed to fill the gap. Commentators sometimes say that steering the economy using G.D.P. data is like driving a car using only the rearview mirror.

So economic policy makers usually scrutinize tens, or even hundreds, of indicators, covering different industries and assets, different parts of the country, different groups of people. They monitor jobs reports, advertising rates, wage settlements, the cost of shipping freight, asset prices, sales of consumer durables and much, much more. The White House’s Council of Economic Advisers, for example, uses about 200 indicators.

No forecasters have a consistently good track record, but the best supplement their statistics and computer models of growth with informal evidence. To get an accurate picture of the economy, the wise economist uses something like an artist’s pointillist technique — individually, all those tiny dots make little sense, but together they provide a decent portrait of the whole.

That’s why it can pay for economists to check the skyline and skirt lengths as they stroll around the city.

Diane Coyle, a professor of economics at the University of Manchester, is the author, most recently, of “GDP: A Brief But Affectionate History.”