Restating the '90s

We now have enough perspective to look back at the last decade and assess what was real--and what wasn't

Conjure up the economic gains of the 1990s, and what comes to mind? Perhaps it was how the stock market ruled: All those initial public offerings that raked in unprecedented billions for venture capitalists. Or the dramatic rise in 401(k)s and mutual funds. Or the growing ranks of the Investor Class who cashed in big-time as the Standard & Poor's 500-stock index quadrupled.

And wasn't it a great time to be a top manager, with productivity gains boosting the bottom line and igniting executive pay? While it was going on, venture capitalist L. John Doerr called the boom the "largest single legal creation of wealth in history." For both investors and managers, it seemed like nirvana.

Well, yes and no. With the recession apparently over, it's now possible to make a more realistic assessment of the entire business cycle of the 1990s: The sluggish recovery that started in March, 1991, the extraordinary boom, the tech bust, and the downturn of 2001. And guess what? A lot of things happened that defy the conventional beliefs about the decade.

For starters, over this 10-year period, productivity rose at a 2.2% annual rate, roughly half a percentage point faster than in the 1980s--a significant gain. But the real stunner is this: The biggest winners from the faster productivity growth of the 1990s were workers, not investors. In the end, workers reaped most of the gains from the added output generated by the New Economy productivity speedup. This revelation helps explain why consumer spending stayed so strong in the recession--and why businesses may struggle in the months ahead.

The key is that wage growth accelerated dramatically for most American workers in the 1990s business cycle. Real wage gains for private-sector workers averaged 1.3% a year, from the beginning of the expansion in March, 1991, to the apparent end of the recession in December, 2001. That's far better than the 0.2% annual wage gain in the 1980s business cycle, from November, 1982, to March, 1991. The gains were also better distributed than in the previous decade. Falling unemployment put many more people to work and swelled salaries across the board: Everyone from top managers to factory workers to hairdressers benefited. Indeed, the past few years have been "the best period of wage growth at the bottom in the last 30 years," says Lawrence F. Katz, a labor economist at Harvard University.

By contrast, the return on the stock market in the 1990s business cycle was actually lower than it was in the business cycle of the '80s. Adjusted for inflation and including dividends, average annual returns on the S&P-500 index from March, 1991, to the end of 2001 were 11.1%, compared with 12.8% in the previous business cycle. Bondholders and small savers saw their returns drop even more in the '90s. The real return on six-month certificates of deposit, for example, was only 3.1% over the past decade, compared with 4.7% in the '80s.

Overall, BusinessWeek calculates that workers received 99% of the gains from faster productivity growth in the 1990s at nonfinancial corporations. Corporate profits did rise sharply, but much of that gain was fueled by lower interest rates rather than increased productivity.

Why did workers fare so well in the 1990s? The education level of many Americans made an impressive leap in the '90s, putting them in a better position to qualify for the sorts of jobs that the New Economy created. Low unemployment rates drove up wages. And a torrent of foreign money coming into the U.S. created new jobs and financed productivity-enhancing equipment investment.

Meanwhile, U.S. corporations were hit by a one-two punch: an economic slowdown overseas following the 1997 Asia financial crisis and the tech bust at home in 2000. To the dismay of tech investors, the hundreds of billions poured into Internet ventures and new telecom equipment ended up lowering prices for users, not raising profits for corporations. "We convinced ourselves we had discovered some magic elixir of productivity that would elevate corporate profits far above historical standards," says Gary Hamel, head of consulting firm Strategos and author of the 1994 best-seller Competing for the Future. "But most of the productivity gains that are made possible by e-business will never go to the bottom line. They will all go to customers."

The fact that workers reaped the bulk of the benefits of New Economy productivity gains helps explain why consumer spending and the housing market stayed strong during the 2001 recession. Heftier wages have "sustained consumption at levels higher than we would have expected," says Barry Bluestone, an economist at Northeastern University.

Moreover, it is now easier to understand why corporate executives remain relatively bleak about the future despite the apparent recovery. Labor costs now absorb almost 87% of the output of nonfinancial corporations, the highest level ever, and way above what companies were paying out at the end of the last recession. When the economy improves, companies are likely to face a lethal combination of rising interest rates and rising wages. "If the recovery is taking hold, workers will be in a position to bargain for wage gains," says George Magnus, chief economist at UBS Warburg. "The benefits of productivity growth are being drained away by labor."

There are two possible outcomes for this kind of profit squeeze. In the positive scenario, companies can increase productivity fast enough to fund both higher wages and decent profits for shareholders. The darker possibility is a double-dip recession, with rising wages depressing corporate profits even as the economy recovers. That would mean lower levels of business investment and slower growth rates. Eventually, companies would resort to wholesale layoffs to try to eke out profits--leading to another downturn.

Whatever the outcome, there's little doubt that the productivity gains of the 1990s are real and sustainable. They have been tested during this recession and have remained strong, breaking the historic pattern of sagging during downturns. "You can look at Enron and the dot-com bust and wring your hands," says Martin N. Baily, who served as chairman of the Council of Economic Advisers under President Bill Clinton. "But I wouldn't be surprised to see 2.5%" annual productivity growth in the coming years.

The real issue is determining who benefits most. When the productivity revolution started in 1995, it seemed that corporations, not workers, were going to be the big winners. As late as mid-1997, real wages were still growing slowly, while profits soared.

But historically, wage gains have trailed productivity increases by a year or two, and that's exactly what happened this time. As productivity continued to stay strong and unemployment fell below 5%, wage gains took off. From mid-1997 to 2001, real wages accelerated at a 2.1% clip. Real compensation per hour--which includes benefits--in the nonfarm business sector rose at an incredible 3.1% rate. The last time compensation rose that fast for a four-year stretch was in the 1950s.

The wage gains in the past few years were so momentous that they made up for the slow growth in the early part of the 1990s. All told, real wages for the average private-sector worker rose by about 14% in the 1990s business cycle, measured by the Labor Dept.'s employment cost index. That's compared with a slim 1.4% gain in the previous decade.

What's more, workers with a wide range of skills and occupations thrived over the past decade. In the '80s business cycle, real wages of blue-collar and service workers fell substantially. Blue-collar wages, for example, declined by 3.5% from 1982 to 1991. But in the '90s, real wages for these less-skilled jobs rose by 12%. Full-time cashiers saw their median weekly earnings jump by 11% (adjusted for inflation), while auto mechanics' pay went up by 14%, after falling sharply in the 1980s. Hairdressers got an almost 18% boost. That's despite Clinton-era welfare reform and a huge influx of immigrants, both of which were expected to hold down wages at the bottom.

That said, the income gap between rich and poor did continue to widen, but not as fast as it did in the 1980s. One reason: The 20% increase in the minimum wage in the mid-1990s, which immediately benefited people at the bottom. Equally important, low unemployment rates forced employers to reach deeper into the pool of low-skill workers, hiring and training people who would otherwise have been left out in the cold. "Tight labor markets for a long-lasting period do more good at the bottom than we would have predicted," says Katz.

In many ways, the most tangible sign of worker gains in the 1990s was the home-buying boom. Homeownership has always been a critical part of the American dream. During the 1980s, that dream seemed elusive, as the percentage of households owning their homes fell slightly from 1982 to 1991. By contrast, homeownership rates over the past decade rose from 64% in 1991 to 68% in 2001, the highest level ever.

Even the economic slowdown of 2001 and the events of September 11 have failed to put much of a dent in wage growth. When William M. Mercer Co., a human-resources consulting firm, surveyed corporations in January, projected pay increases in 2002 had come down a bit since last summer. Nevertheless, expected wage increases are still running well ahead of inflation. The reason? Fear of being caught without enough workers in the recovery, says Steven Gross, principal at Mercer. "Corporations worry [that] if they disenfranchise core workers too much, they'll leave."

A key reason many Americans could take advantage of the New Economy is that they absorbed the big lesson of the 1980s: Education pays, especially in an information-based economy. The latest numbers show that 51% of the adult population now has at least some college education, up sharply from 40% in 1991 and 33% in 1982. Among the critical 25- to 34-year-old age group, the percentage with some college education has risen from 45% in 1991 to 58% in 2000.

Particularly encouraging was the rising participation of minorities in higher education, especially among blacks and hispanics in their 20s, groups that traditionally had lagged behind. For blacks age 22 to 24, the percentage enrolled in school rose from 19.7% in 1990 to 24% in 2000, nearing the 24.9% level for non-Hispanic whites. The improvement among Hispanics age 22 to 24 was even more dramatic, almost doubling from only 10% school enrollment in 1990 to 18% by 2000.

And contrary to the conventional wisdom, U.S. workers were the big beneficiaries of globalization. Many expected that globalization meant U.S. corporations would shift their capital spending abroad, building factories and back-office operations in low-wage countries and shifting jobs overseas. Competition from low-wage foreign workers would then drive down U.S. pay--what economists call factor-price equalization.

Guess what? The jobs went the other way. True, U.S. companies did boost their direct foreign investment abroad in the 1990s, with $1.2 trillion flowing out of the country between 1991 and the end of 2001. But foreign companies invested even more--$1.3 trillion--in U.S. factories and businesses, creating new jobs and raising demand for labor.

More important, the foreign money that flooded the U.S. stock and bond markets during the 1990s financed a big chunk of the New Economy productivity gains. From 1991 to 2001, total foreign investment in U.S. financial markets reached $2.3 trillion more than U.S. investment abroad. This inflow provided the resources for much of the $3.4 trillion spent by businesses on information-technology equipment and software over the decade. Without that foreign money, it would have been a lot more expensive for companies to make the investments that boosted productivity--and wages--in the 1990s.

Despite fears that U.S. incomes would be dragged down by foreign competition, the wage gap between U.S. manufacturing jobs and those in the rest of the world actually widened in the 1990s, according to government data. In 1991, the hourly compensation for U.S. factory workers was 17% higher than the average for foreign workers, measured in U.S. dollars. By 2000, the difference had increased to 31%, as high-tech, high-wage manufacturing industries expanded and low-wage industries shrank. "Factor-price equalization is out the window," says Donald R. Davis, a trade expert who is chairman of the Columbia University economics department. "It isn't happening, in any form."

It's not just the impact of trade that needs to be restated. Technology, too, had a very different effect on wages than most people anticipated. The common belief was that technology eliminated many low-skill jobs and depressed wages for the rest. That's certainly what happened in the 1980s. But new research suggests that technology has a much more selective impact. Jobs that can be boiled down to a set routine--such as making a loan, assembling an engine, or processing a bill--are prime targets for computerization, whether they are done by low-education or high-education workers. Computers are good at "rules-based" tasks, argues Frank Levy, an economist at Massachusetts Institute of Technology. Indeed, the amount of work using routine skills, both manual and cognitive, plunged in the 1990s, according to a new paper by Levy and two other economists, David H. Autor of MIT and Richard J. Murnane of Harvard.

But there are plenty of nonroutine tasks that cannot be easily replaced by technology--and those were the ones that boomed in the 1990s. They span a wide range of skill and education levels and include such jobs as sales, truck driving, and network installation.

But this litany of good news raises a question: If workers prospered in the 1990s, why didn't investors get their fair share of the gains? After all, corporate profits did rise substantially in the 1990s. Over the entire business cycle, inflation-adjusted earnings per share for S&P 500 companies averaged $36, measured in 2001 dollars. That's 40% above the '80s average. And this gain is not the result of squirrely accounting: An even bigger increase shows up in the government's numbers for operating profits, which, unlike S&P earnings, deduct the cost of exercised stock options.

The bad news for investors is that much of these corporate gains were a result of companies paying lower interest rates for debt. At several points in the 1990s, long-term corporate interest rates plunged below 7%, enabling companies to borrow at low cost, helping the bottom line. At Gillette, for example, interest expenses have gone up less than 25% since 1991, even though debt has more than tripled. Indeed, adjusted for inflation, interest expenses have been roughly flat.

And remember that the average investor has a portfolio not only of stocks but also of bonds and money-market funds. These funds invest in corporate bonds and commercial paper. So when companies cut their interest costs, it shows up as lower interest payments to investors. Building up profits by cutting interest costs is like robbing Peter to pay Paul: It all comes out of the same investor pocket in the end.

It's important to step back and quantify how the productivity gains of the 1990s were distributed. Consider nonfinancial corporations, where annual productivity growth accelerated from less than 1.8% in the 1980s to 2.2% in the 1990s. Over the course of the 1990s business cycle, this increase in added productivity translated into $812 billion in additional output, measured in 2001 dollars. Out of that sum, an astounding $806 billion--or 99%--went to workers in the form of more jobs and higher compensation, including exercised stock options. In effect, not only did the economy speed up in the 1990s but the workers got a bigger share of the pie.

At the same time, faster productivity was not propelling corporate profits with the same intensity. Corporate profits went up by an extra $559 billion, a truly striking performance. But that was mainly because because companies paid lower interest rates on debt.

Look beyond nonfinancial corporations to the economy as a whole, and the results are similar. Faster productivity growth created $1.9 trillion in additional output in the 1990s business cycle, measured in 2001 dollars, and almost all of that gain went to workers and small-business owners. The combined gains for corporate profits, interest, and returns on real estate amounted to only about $50 billion.

To understand why investors failed to cash in from the productivity gains, it's important to recall the devastating impact of the tech bust. Enormous sums of money were thrown away in the dot-com mania and the telecom meltdown. Furthermore, globalization compounded investors' headaches. Increased trade and higher levels of investment abroad were supposed to pay off by getting U.S. companies access to fast-growing foreign markets. The result was expected to be a big shot of profits from abroad.

But that didn't materialize. Export growth over the decade fell short of expectations, and net profits from overseas actually grew only slightly faster than domestic profits. The reason: Foreign markets did not boom as expected, especially after the 1997 Asian financial crisis. Many countries, both industrialized and developing, actually posted slower growth in the 1990s business cycle than they did in the '80s. The one big exception was China--but even with that country included, global growth slowed from 3.6% in the 1980s to 3.3% in the 1990s business cycle.

Of course, there's no reason why the factors that depressed profits in the 1990s should be repeated in the coming decade. It's possible that as the U.S. moves into a new expansion, profits could bounce back quickly. Germany and France show signs of escaping the doldrums, while key exporters like Korea and Taiwan are picking up speed. Certainly a synchronized global recovery could help profits, just as the worldwide slowdown of the late '90s hurt them.

Moreover, technology investments by companies could produce even bigger productivity payoffs than in the 1990s. Productivity surged in the fourth quarter of 2001 at an amazing 5.2% rate. That level of productivity growth is not sustainable. Still, big enough productivity gains could enable companies to boost profits while paying high wages. That could start a new virtuous cycle, as it did in the 1990s when high levels of investment produced faster productivity gains, fueling growth and investment.

But it won't be as easy as the 1990s, because workers are starting with a bigger share of output and faster wage growth. To satisfy workers and build profits at the same time, companies are going to have to do better than the 2.2% productivity growth of the 1990s. And that will likely take place in an environment of rising interest rates, putting more pressure on profits.

As it turns out, our original perceptions of who benefited most from the productivity gains of the 1990s was flipped on its head. Looking ahead, the economic pie is growing bigger all the time, but it's still up for grabs who will get the largest piece in the future. "At a national level we may get substantial productivity gains, but the real question is how are they shared," says management consultant Hamel. And in the end, that's the real lesson of the 1990s.

By Michael J. Mandel

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