(This post was written jointly with Larry Mishel, president of EPI)

Allow us to quickly offer one more wrinkle—a really important one—to the ongoing debate about economic stimulus: we all want a quick, timely package to offset what, with each new data release, looks like a recession. But even if the process takes awhile, a stimulus package will still be very much worth pursuing.

There have been many statements in the press contradicting this point, i.e., asserting that unless we can get a stimulus package into the economy quickly, there’s little point in pursuing it. One economist told Bloomberg News, “Timing is extremely important…[r]ecessions typically last less than a year, so unless you can be pretty quick, it's not worth doing.”

The reason this statement is wrong is because it is based exclusively on gross domestic product, as if mitigating the fall in overall growth is the sole focus of an anti-recession package. In fact, our efforts should also target insufficient job growth, rising unemployment, and the resulting wage and income losses for many families, including those who don’t lose their jobs.

This insight is critical, because in both of the last two recessions—1990-91 and 2001—job market problems far outlasted the GDP contraction.

One reason for this is definitional: the officials who date recessions (and it’s not easy for recessions to get dates…) tend to weight the growth of the overall economy, as measured by real GDP, most heavily, and conversely, give less consideration to unemployment and job growth.

The 1990-91 recession officially ended in March 1991, but unemployment kept rising for another fifteen months, until June 1992. That was also the birth of the jobless recovery, as employment growth stagnated over this same period. But that was nothing compared to the jobless recovery following the 2001 recession: between the official trough (recession end-date) and August 2003, we lost another 1.1 million jobs (in addition to the 1.7 million lost over the recession).

Since most families depend on their paychecks, not their stock portfolios, the weak job market translated into income losses for many in the middle-class on down. In both of the last two cases, real median family incomes fell not just in the recessions, but for the first three to four years of the recoveries.

Now, we honor GDP growth as much as the next economists, but it’s jobs and incomes that matter most to most people. An effective stimulus package, even one that’s late, can boost these important variables, so we don’t want to give up on it if we can’t get it out the door in the next few months.

But will the pattern of the last two recessions show up in the next one? That’s unknowable, and we’ll keep our statistical eyes peeled for precisely this type of GDP up, jobs/incomes-down dynamic in coming quarters. At this point, many forecasters are calling for a relatively mild recession, but again, that’s ‘mild’ judged purely in GDP terms. Goldman Sachs analysts forecast unemployment climbing through 2009, hitting 6.5 percent by the last quarter. According to our calculations, based on the historical relationship between unemployment and middle-class incomes, an increase in joblessness of that magnitude would lower the real incomes of such families by $2,400 over 2008 and 2009.