One by one, the old cliches about the service sector are exploding. Growth there, once thought to be recession-proof, slowed to a halt last year. Then, supposedly safe service jobs began disappearing as bankers and assistant buyers were laid off in droves.

Now, new evidence is beginning to call into doubt the widely held assumption that service industries have not seen significant productivity growth. Using recently published Commerce Dept. estimates of the sector's output, BUSINESS WEEK has put together a list of productivity winners and losers (table). The data clearly show healthy productivity growth for certain service industries. Most of the shortfall in the sector's measured productivity stems from just five industries: health care, legal services, hotels, repair services, and grocery stores. And some important service businesses are boosting output per worker at least as fast as most manufacturers.

The good news doesn't end there. The low productivity growth noted for some industries may actually understate the improvement in service that these businesses have achieved in recent years. From the corner grocery store to the largest accounting firms, service businesses have spent most of the 1980s trying to deliver new types of services. And many economists believe that government statistics have not kept pace with these changes. Notes University of North Carolina economist James F. Smith, a former director of credit research at Sears, Roebuck & Co.: "The bulk of the overall productivity shortfall is a measurement problem, and the biggest area where we have a problem is the service sector."

SUSPECT STATS. In its recent report Commerce appeared to reach some very negative conclusions about service sector productivity. It reported that manufacturers boosted productivity at an annual rate of 3.9% from 1979 to 1988, while the service sector could manage an annual increase of only 0.4% during that period.

But, in that same report, Commerce published its first real estimates of output for such industries as legal services and retailing. Until this data became available, there was no way to determine the productivity growth for most individual service industries. That's still true for banking, real estate, and business and professional services. For the rest of the service sector, however, a close look at the improved government numbers shows some surprising results.

After excluding those industries whose productivity Commerce was unable to quantify, about one-quarter of the service sector--measured by output--shows annual productivity improvements of 3.0% or more from 1979 to 1988. The top performer was the communications industry, which now handles many more telephone calls with about 15% fewer workers. Also showing stellar results was the securities business, which has used automation to cope with a fivefold increase in stock market trades. Even department stores, many of which are now skirting bankruptcy, saw their sales-per-employee-hour, adjusted for inflation, rise by 32% over the past 10 years.

CHOICE CUTS. Indeed, the productivity leaders in the service sector are doing quite well compared with the manufacturing sector. Much of manufacturing's gains came from the machinery industry, which includes computer makers. Largely because they could pack more and more computer power into the same box, the output of the machinery industry doubled from 1979 to 1988. Taking out this single industry, the productivity growth rate for manufacturing drops to just 2.9%.

At the other end of the spectrum, the measured productivity losses in services also are concentrated in just a few industries. BUSINESS WEEK's analysis shows health and legal services alone to be responsible for about 40% of the total productivity shortfall, while grocery stores account for 12% of the drop.

And there's good reason to be suspicious of these declines. Consider the supermarket business, for example. Economist Thierry Noyelle of Columbia University compared supermarkets in the U. S., where productivity growth has been negative, with those in France, where reported gains have been fairly strong. He found that French productivity was rising, in part, because mom-and-pop stores are being replaced by large self-service markets, a shift made years ago in the U. S. At the same time, American supermarkets are opening up such labor-intensive services as deli counters and using computers to manage an ever-increasing variety of goods. According to an annual survey by Progressive Grocer, an industry trade journal, supermarkets now stock almost 19,000 different items, up from 14,000 in 1985. Such an increase in consumer choice, usually deemed to reflect an increase in quality of service, isn't picked up in the government's productivity numbers. "The negative productivity growth does not reflect deterioration," says Noyelle. "It's a change to a very different mode of supermarket."

The rapid growth of new brands and products also helps explain the poor productivity performance of the repair industry. More makes of imported cars mean more specialized repair shops. Similarly, the rapid growth of computers and other electronic equipment has created an entire repair industry that didn't exist before. Noyelle's study of the health care industry shows that there, too, government statistics don't capture the gains to consumers from advances in the quality of care.

The measurement problem is so bad in the case of banks that Commerce still cannot provide even a rudimentary estimate of productivity. According to the Labor Dept., which measures such things as the number of checks cleared and loans made per employee, commercial banking shows a respectable 2.3% annual increase in productivity over the past 10 years. But even that may understate the gains in customer convenience from 24-hour cash machines. "They don't have a clue about how to measure the output of a financial intermediary," says Jerry L. Jordan, senior vice-president and economist at First Interstate Bancorp in Los Angeles.

Unfortunately, for most of these service businesses, the improvements in quality, if real, have not yet shown up in profits. The reason: overcapacity. There are too many stores, too many banks, and, in some parts of the country, too many fast-food restaurants.

LEAN AND MEAN. Elusive profits, however, didn't halt growth in the service sector as a whole until recently. Even as one part would cut back, another part would absorb the extra workers. As a result, overall productivity in the service sector actually dropped from mid-1988 until the fourth quarter of 1990.

But the current recession has started to squeeze a lot of the overcapacity out of the service sector, leading to a rebound in productivity in the final quarter of 1990 (chart). The list of companies in Chapter 11 lengthens each week--Carter Hawley Hale Stores Inc. and Pan American World Airways Inc. are just the latest--and companies are actively cost-cutting. Already in 1991, banks, retailers, and airlines have announced permanent reductions totaling more than 60,000 jobs, according to labor consultant Dan Lacey. This could lead to larger productivity gains later.

What happens to all those workers, as the service sector slims down, could be a problem. But to most economists, an upturn in service productivity can only be good news. Services would become cheaper to both consumers and manufacturers, making their goods less expensive as well. The trick will be to achieve these gains without sacrificing the quality of services in the process.

Before it's here, it's on the Bloomberg Terminal. LEARN MORE