I posted this at Maximum Utility a few days ago:
Growth Policy versus Stabilization Policy, by Mark Thoma (with a few minor edits): In macroeconomics, there are two important policy questions, and our attention to one or the other changes with the economic events of each era. One question concerns stabilization policy -- keeping the economy as close as possible to the long-run growth path -- and the other is growth policy, i.e. policy that attempts to maximize the long-run growth rate. (There is also work on whether stability and growth are related. More stable economies could grow faster due to reduced uncertainty, but government intervention to stabilize the economy could also stifle growth according to some models, so the relationship is not clear a priori. In modern models, these are not strictly separable, but it is still a useful way to think about policy conceptually)
We could go back further than this, but let me pick the story up in the 1970s. A few economists were worried about growth at this time, but the main concern during the tumultuous 1970s and early 1980s was with how to do a better job of stabilizing the economy. The traditional Keynesian policies, which had not taken account of inflation or expectations in a satisfactory way, had failed to produce the desired stabilization. This led to the search for a new economic model that could provide better guidance. The result was the development of the New Classical model, replaced soon after by the New Keynesian model when the New Classical could not explain both the duration and magnitude of actual cycles, and it's implication that only unanticipated money matters appeared to be contradicted by actual data.
The New Keynesian model, and its new advice for stabilization policy concerning the use of interest rate rules, seemed to work and we entered into a period known as "The Great Moderation" (stated compactly, the new policy involved targeting an interest rate with a Taylor rule that responds to output and inflation, where the response to inflation was more than one to one). This period, which began in the early 1980s, saw low and stable inflation rates, and a fall in the variation in GDP of around 50 percent. The result was the emergence of the view that the stabilization problem had been solved. By using the correct monetary policy, policymakers had produced the Great Moderation, and that left other policy tools such as fiscal policy free to pursue the maximum growth objective (and the result was supply-side fiscal policies such as cutting capital gains and dividend taxes justified by arguments about their contribution to growth).
Because of this, the profession moved on to growth theory and policy. Stabilization had been solved with monetary policy, and growth was now the major question to be solved. If the economy was still as jittery as it had been in the past, then stabilization policy would have also been of concern to academic economists, but developing optimal monetary policy rules from the New Keynesian structure seemed to have solved that problem.
Of course, recent events show us in no uncertain terms that the stabilization problem has not been solved, and questions about how to stabilize the economy ought to be coming to the forefront again. And they are, to some extent, but I'd argue that our ability to stabilize the economy has been limited by those who still think growth is the only important consideration for evaluating policy. For example, because of this, the stimulus package that was put into place had to be justified by its ability to stimulate long-run growth when its main concern should have been with how to stabilize the economy. That led us to concentrate on tax cuts (because conservatives believe tax cuts increase economic growth) and infrastructure spending. However, tax cuts of the type that were implemented are mostly saved, and infrastructure spending takes much too long to put into place (and may not generate as much employment per dollar as other types of spending). These are not optimal stabilization policies. Other types of spending, the types that get money into people's hands and puts people to work right away, might have worked faster and had a greater benefit in terms of moving the economy closer to trend, but since these policies were harder to justify in terms of their contribution to long-run growth. Therefore, they could not find the support they needed.
I believe that stability is important to people (i.e. that utility is lower when there is more economic uncertainty), and because of this stabilization policy can be justified on its own terms, there's no reason to insist that stabilization policy maximize growth. The policies that maximize growth are different in some cases from the polices that stabilize the economy, and insistence that all policies can be justified by their contribution to long-run growth causes us to sacrifice economic stability. The policies we put into place should pay attention to both goals, but I believe we have paid far too much attention to growth in formulating recent policy, and not nearly enough to stability.
Hopefully, recent events will begin to shift our thinking away from the "growth above all else" policies we've pursued since the early 1980s, and that we will devote more attention to stabilization policy. We can put people back to work faster than we did this time around, and we can do a better job of increasing aggregate demand early in the recession (thereby reducing the fall in GDP and employment). But to do so we have to realize that stabilization is an important policy goal, and that it does not always lead to the same policies that are needed to maximize growth. People's lives, or at least their livelihoods, depend on it.