IS RHODE ISLAND'S RECOVERY NOW ON?

The Providence Journal Editorial Page News, July, 1996

In an election year such as this, a number of economic issues will be intensely debated. One of these issues, I will surmise, will be the question of whether Rhode Island’s economy has recovered from the last recession. Answering this question seems like a fairly straightforward exercise: contrast the number of jobs at this stage of our recovery with the number during the last business cycle. This certainly sounds simple enough. Unfortunately, the altered structure of Rhode Island’s economy has made the task of answering this and so many other important questions much more involved than it would have been in past years.

The first issue that must be dealt with is which part of the last business cycle we should contrast current employment with. Should we base our comparison on peak employment or the number of jobs at the trough of the last business cycle? This is akin to the question of whether the glass is half-empty or half-full. There is no single and universally-accepted answer to this question, so judgment, the intended purposes of such analysis and the persons to whom this is directed become relevant considerations.

If we contrast current employment with that at the peak of the last business cycle, we observe that peak employment, which occurred in June of 1989, was 471,500, versus current employment (for May, 1996) of 446,200. The difference between these, 25,300, is 5.4 percent. Next, compare current with trough employment. The employment trough occurred in January of 1992 at 414,100. Contrasting this with May, 1996 employment, we see a gain of 32,100 jobs, or 7.8 percent. The possibilities for "strategic" manipulation of these numbers should be obvious. But, this aside, can you see any problem with using either of the two numbers above?

Both of these changes involve the comparison of employment levels for different months. Since labor market data embody a number of seasonal factors, comparing May with either January or June is misleading. For example, payroll employment almost always falls in January as many holiday employment jobs (seasonal employment) are eliminated after Christmas. Certain other months tend to be above-average for the year. So, contrasting employment for different months without controlling for the effects of seasonality is not valid.

This is problem easily fixed. Data for different months can be compared using seasonal adjustment. Based on seasonally-adjusted employment data, the employment peak, which remains in June, 1989, is 466,700. Trough employment now occurs in December of 1991, at 413,200, while for May, 1996, seasonally-adjusted employment becomes 443,600. Relative to peak employment, our current job situation thus entails a decline of 23,100 jobs or 4.9 percent. Compared to the trough, our employment gain is 30,400, or 7.4 percent.

The use of seasonally-adjusted data did not reverse the earlier results, it merely refined them. Now, even if one decides which of these two numbers to utilize, does this end the analysis of where we currently stand? I will argue that it does not.

As Rhode Island continues its transformation from a manufacturing to a service-based economy, both labor market trends and the nature of the jobs that are created have changed. Unlike the last recovery, the combination of labor-saving technology, defense-related cutbacks and corporate downsizing has resulted in the persistence of job loss throughout this recovery. Since employment change now understates the number of jobs created, the number of jobs created since December of 1991 exceeds the employment change value of 30,400. Second, throughout this entire recovery, Rhode Island’s job growth has been concentrated in the service sector as manufacturing employment has continually declined. A "job" today often entails part-time employment, perhaps without fringe benefits. So while this recovery has created more jobs than the employment numbers indicate, they are generally different types of jobs than in past recoveries, both in terms of their industries of origin and remuneration. And, to complicate matters further, a great deal of the job loss during this recovery has involved white collar employees. Thus, the loss of relatively high paying jobs in Rhode Island is not restricted to manufacturing; it extends across industry boundaries to white collar jobs as well.

To many Rhode Islanders, these trends indicate a deterioration in our standard of living that is certain to continue. In spite of the fact that we have been in a recovery for four and a half years, they have only abandoned the notion that we are in a recession in the last month or two. For them, the question of whether the glass is half-full or half-empty has always been irrelevant: they have chosen to presume, in spite of ample evidence to the contrary, that the glass is broken. Perhaps they should consider the following statistic: as of 1994, median income for a family of four in Rhode Island was 10 percent higher than the national average. Furthermore, median income growth for the entire 1989-1994 period was 15.3 percent for the nation versus 19.5 percent for Rhode Island.

Has Rhode Island recovered from the last recession? By traditional measures it has not. But simply considering the change in employment from peak or trough to today is as meaningless as attempting to add apples and oranges. If you need further proof of this, consider the following fact: with its current unemployment rate of 4.6 percent, Rhode Island is not terribly far from full employment!

by Leonard Lardaro