Here’s another look at the fates of the public and private sectors during the recession. Since the overall scale of private sector employment is much larger than the government sector (and rightly so), it’s difficult to get a chart that shows anything if you look at the raw numbers. So instead, this lines indexes both sectors to where they were in January 2009 when Obama took over:
What you see is that if you look at the private sector you see a fairly “normal,” albeit severe, recession. Private sector employment tumbles precipitously for a while, then hits bottom, and then starts growing consistently. And for a while, the public sector also behaves “normally,” growing at the same slow and steady trend it was growing at before the recession and providing a kind of countercyclical stabilizer. Thus, during the final year of George W Bush’s administration we actually saw a substantial rebalancing away from private sector work and toward public sector work. But since sometime in early 2009, the trend level of government employment has been sharply downward. That’s a trend that was interrupted by the census, but now looks very clear. It reflects the fact that the federal government hasn’t provided adequate aid to states and localities and the fact that conservative governors have preferred layoffs to tax increases. But even if the governors had different policy preferences, the macroeconomic impact of state/local policy would still have had to have been toward contraction and austerity.
But the federal government could have avoided this. Many policymakers seem to be consoling themselves with the Reinhardt/Rogoff finding that recovery from severe financial crises are typically very slow. But I’ve always thought the correct interpretation of that finding is that the policy response to severe financial crisis is typically inadequate. State and local governments are engaged in austerity budgeting because they’re not getting the federal aid they need. But the federal government isn’t running in to some kind of objective borrowing constraint.