Michael Mandel's response to Dean Baker's article, "Business Week Restates the Nineties" in Challenge Magazine July/August. 

    Economists—a stubborn bunch—don’t easily give up cherished beliefs. In the 1990s, economists resisted the notion that investment in information technology could boost productivity growth until the evidence was overwhelming. Similarly, after two decades of little or no real wage growth, economists are reluctant to accept that the 1990s and the early 2000s--and especially the last five years--have been good for most  American workers.

      Here are the facts. It is unambiguous that real wages rose much faster in the 1990s than they did in the 1970s and 1980s.  For example, from the fourth quarter of 1990 to the fourth quarter of 2000,  real average hourly wages for production and nonsupervisory workers rose at a 1.1% annual rate (adjusting for inflation using the personal consumption deflator), for a total wage gain of 11%. For the previous two decades, real average hourly wages rose at a meager 0.02% annual rate—virtually no gain at all.

     Similar results—decent wage growth in the 1990s, and little or no wage growth in the 1970s and 1980s-- hold no matter which measure of real wages we look at. For example, real wage and salary income per worker, as derived from the national income accounts, rose at a 1.7% annual rate from 1990 to 2000, compared to 0.3% growth over the previous two decades. The real median hourly wage, calculated by the Economic Policy Institute, rises at a 0.5% rate from 1990 to 2000, compared to no change in the period from 1973 to 1990 (where 1973 is the first year available).  

     Moreover, it doesn’t matter much which years we include in the 1990s. Real wages rise strongly whether we look at the ‘pure 90s’, which runs from 1990 to 2000; the 'peak to peak' business cycle, which runs from the third quarter of 1990 to the first quarter of 2001; or the 'trough to trough' business cycle, which runs from start of the expansion in the first quarter of 1991 to the fourth quarter of 2001. The last one is my preferred definition of the 1990s business cycle, because it includes both the boom of the late '90s and the bust of 2000-2001, which logically go together.

   Indeed,  2001 was an especially interesting year. Historically, real wages fall in recessions—but not this time. Over the last year, real hourly wages have risen by 2.4%-3.1%, depending on which measure you use. By contrast, real hourly wages fell at a 0.7%-2.0% rate in the recession of 1990-91.

    The data about rising real wages fits with anecdotal evidence on worker prosperity. Consumers seem to have plenty of money to buy cars and other goods, and homeownership rates have climbed to record levels. Meanwhile mortgage delinquencies are actually running below the average of the 1990s, and well below where  they were in 1991.

     The BusinessWeek article also argues that investors had lower returns in the 1990s than they did in the 1980s.  This result, unlike the one about rising wages, is sensitive to the period  over which investment returns are measured. In my article, I use the period which runs from the 1991 trough to the 2001 trough. Annual real total return on the S&P 500 in the business cycle of the 1990s was below the comparable measure in the business cycle of the 1980s.

    Baker objects to including 2001,  arguing that as a recession year it is atypical. All of his charts stop with the year 2000. But it simply makes no sense to include the boom without including the bust. At the end of 2000, Enron still had a stock price of $80 a share, and the Nasdaq was about double the level of spring 2002. I’m sure that if I had written a laudatory article about the stock market that stopped with 2000,  I would have gotten a flurry of letters objecting to it.

    Omitting 2001, as Baker does, also leads to the wrong conclusion about labor's share of corporate income. Baker devotes most of the space in his article to arguing that the labor share of income fell in the 1990s. But the only way he can come to this conclusion is by leaving out 2001.  In that year, the labor share of income at nonfinancial corporations reached 85.7%, by far its highest level in the post-war era. That more than compensates for the lower labor share earlier in the decade. Indeed, the average labor share at  non-financial corporations in the 1991-2001 period is higher than the previous two decades.

   Of course, current performance is no guarantee of future returns. As my article points out, a squeeze on profits may lead companies to start another round of layoffs, which would push down wages and send the pendulum swinging back again to capital. But there is no doubt that the 1990s was the best decade for workers in a long time.