Get the latest updates as we post them — right on your browser

. Last Updated: 07/27/2016

Stats Fall Short in New Economy

LONDON -- There are three kinds of lies, said Benjamin Disraeli, a 19th-century British prime minister: lies, damned lies and statistics.

Many economists would agree, and the problem is getting worse as economies shift rapidly into the service sector. The data simply aren't keeping up.

"One of the issues in the U.S. undoubtedly is that the economy is largely now a service-sector economy, and most of the data are on manufacturing," said Robin Marshall, chief economist at Chase in London.

At least the United States produces data in a timely fashion. European countries, particularly Germany, still struggle to present even basic statistics.

Some frustrated economists now are creating their own measures to get a truer economic picture -- and little wonder. Consider some statistical anomalies:

?The U.S. employment figure, among the world's most influential economic statistics, soars by 624,000 in February after a drop of 188,000 in January, shows growth of around 200,000 in most of the summer months, then drops like a stone in September with a decline of 40,000.

?Belgium no longer counts people looking for work if they are over 50 and have received unemployment benefits for a year. As a result, only 14,970 of the 39,812 decrease in unemployment during the first six months of the year was attributable to job creation.

?Eurostat, the European Union's statistical agency, reported German unemployment at 8.9 percent in July and Spanish unemployment at 21.3 percent. Those two countries themselves reported rates of 10.2 percent and 13.7 percent.

Analysts say that while markets have learned to live with such statistical nightmares, there is a growing problem with the relevance of current statistics.

Data such as producer prices, wholesale prices, industrial production and business inventories are important, but do not give a clear view of an economy that employs more people in fast-food outlets than on factory floors.

"The problem is, specific service-sector data is harder to come by," said Stephen Lewis, director of research at London Bond Broking. "It is very easy to count tons of steel produced, but very difficult to measure the provision of a service."

U.S. payroll data have taken on huge significance during the 1990s, sending markets into fits every month when the figures are released.

Such obsession is justified to an extent, said Michael O'Hanlon, chief international economist at PaineWebber in London.

Labor makes up around three-quarters of a company's costs, and payrolls can indicate inflationary pressures across sectors, he said.

What the market initially fails to note, he said is that "the rise in payrolls may not be a problem because the nature of the labor market may be changing."

As wages rise, for example, benefits may shrink as companies use more outside contractors, consultants and part-time workers -- a trend evident in the burgeoning computer industry.

Meanwhile, surveys from agencies outside the government, such as the U.S. National Association of Purchasing Management, are becoming more prevalent -- and more relevant.

Federal Reserve Chairman Alan Greenspan in particular has assiduously turned to a broader range of data, analysts say. That is why he kept interest rates steady throughout the spring and summer, while markets feared inflation because of employment growth.

"He is right in his judgment that the U.S. economy is not behaving according to the pattern which it has sewn in the past, and he needs to exercise a lot of discretion in judging what each piece of data means," said Lewis at London Bond Broking.

This idea is gradually dawning on markets, and like U.S. money supply in the 1980s and British trade data in the 1970s, payrolls may eventually lose their predominance.

Some European countries still have trouble providing even minimal raw data, analysts say.

In Germany, for example, there is the barest schedule of releases, and even vital numbers like M3 money supply arrive with little or no notice.

Economists say astute economic fortune-tellers rely heavily on their own pool of indices.

O'Hanlon said PaineWebber has recently come up with a "Phindex," or "policy happiness index," which measures the state of a bond market's "contentment with policy." The index rises as unemployment goes up and inflation falls.

He also takes a more pragmatic approach.

"I know it sounds daft, but you can check with a London cabbie," he said. "In the recession of 1991 to 1992, they were doing little business, but it's pretty buoyant at the moment."