What Has Been the Lasting Damage?
Op-ed in Eurointelligence
Adam Posen is Senior Fellow at the Peterson Institute for International Economics and a member of the Monetary Policy Committee of the Bank of England. The views expressed here are solely his own, and not those of the Bank, the MPC, or any of its staff.
The major economies seem to be on their way to economic recovery. But how much of and how long a recovery should we expect? The question is key to setting monetary policy as well as wage and price expectations going forward. This comes down to the assessment of economic potential, the average rate at which we can expect a national economy to grow in real terms sustainably without causing inflation or deflation. Economic potential is difficult to measure even in normal times. Coming out of the worst financial shock in 70 years complicates the assessment considerably.
We have to assess how much lasting damage the crisis has done to our economies. The shock was primarily demand driven, resulting from the financial panic and increase in savings-that is why the policy response of fiscal and monetary stimulus was both appropriate and effective. But the shock still could have done lasting damage both to aggregate supply, the base level of capacity from which our economies will grow once the slack is taken up, and to the trend rate of growth going forward. I fear that it has indeed done such damage, particularly to the latter.
As on many things to do with macroeconomic policy, the United States and the European Union are divided on how to think about this. At a conference I co-organized at the Bank of Italy this month, two American economists gave rather optimistic assessments. They saw the US economy as quite flexible, the share of the financial sector in the US economy to be reallocated to other uses quite small, and the functionality of the financial system for allocating real investment to be relatively unimpeded. Hence, they thought there was plenty of room for demand to run hot, and even for greater stimulus policy, before the slack in the economy was taken up-they also thought that there would be no lasting damage to US trend growth.
A number of European participants in the conference disagreed. Even in the continental economies where the financial sector had not grown so large, there was concern that aggregate supply had shrunk. This would mean less room for catch-up growth (and for stimulative policies) to run and slower growth going forward. The need to reallocate labor and capital across sectors-say from construction to exports in Spain and the United Kingdom, but also away from autos in Germany and France-loomed large. So did fears of what distortions and costs rising government deficits would impose.
This difference arises partly from experience, as my colleague Jean Pisani-Ferry points out. In Europe, growth shocks have tended to be persistent, meaning that recessions had lasting negative effects, whereas in the United States, the economy has tended to bounce back fully after recessions. While that certainly is true of the past, I am not so sure that the difference will be as stark this time. In fact, the convergence across the Atlantic may arise as much out of the United States showing more persistent difficulties coming out of this recession than it did coming out of past recessions, as it will from Europe rebounding better.
On the positive side of the convergence towards full rebound, labor flexibility has certainly increased meaningfully in most of Western Europe. That should speed reallocation across sectors and lead to lower increases in long-term unemployment. The clear commitment by all European governments to reverse temporary fiscal measures, as well as the well-anchored inflation expectations throughout the European Union, is a major improvement over the past. This will bring rewards in terms of limiting the lasting damage to European potential.
The factors that are likely to lead to slower growth as a result of the crisis are daunting, however, and are relatively more applicable to the United States than in past recessions. First, a lot of human and financial capital has been lost or foregone due to the crisis. Some of this investment was specific to sectors-residential construction, financial services, auto production-that cannot be expected to return to the share in the economy or the rapid growth they displayed earlier this decade. People will have to adjust to working in new fields, and find employment in those fields. The capital investment available in those fields to make those people more productive will lag as well.
Second, the financial sector is unlikely to do a terribly good job of allocating financing to new uses and newer businesses for a while. Not only will banks be rebuilding their balance sheets and raising capital, which will restrict lending, but also banks and other funders are likely to make poor choices at the margin about whom to lend to until their government guarantees are reduced and their capital restored. Appetite for riskier investments will be low, putting some brake on innovation. And the sea change in public debt levels for the United States as well as the greater awareness of coming health care costs will crowd out some private investment for at least a few years.
It is worth noting that US potential output growth was already noticeably declining by 2007, prior to the crisis. Fewer women and immigrants were joining the labor force, and productivity growth had slowed, perhaps due to the diminishing returns from applying IT in the economy with nothing to take its place. Europe was doing no better, its productivity had gone up significantly only in privatized industries during the past decade, according to European Commission and European Central Bank analyses. The idea that anything coming out of this would reverse those trends seems to be a stretch. That the crisis will worsen the strains on employment and postpone our next technological leap seems much more likely. I fear the European views about the lasting damage from the crisis will prove right on both sides of the Atlantic.