An excellent chart from Jacob Goldstein shows where jobs were lost during the recession and what kind of rebound we’ve had:

So this shows us a few things. One is to underscore the oddness of Naryan Kocherlakota’s observation that “the Fed does not have a means to transform construction workers into manufacturing workers.” It’s true, they don’t. But the biggest source of job losses is that we’ve transformed manufacturing workers into unemployed people, and the Fed really should have a means to transform the vast majority of unemployed manufacturing workers into employed manufacturing workers.

And this is how it looks across the board. You do have a big hit to finance and a bigger hit to construction, but the combined finance & construction share of the job losses is less than 50 percent. The manufacturing, retail, administration, leisure & hospitality, professional services, and transportation sectors all went down when aggregate spending went down and have recovered weakly as aggregate demand recovered weakly.

Policies that boosted aggregate would have majorly mitigated these losses.

The construction and state/local pictures, meanwhile, should be separated from finance. You could have eliminated the state/local losses via revenue sharing, or you might have deemed that to be an undesirable policy. Similarly with construction. Many, though not all, of these construction workers who’ve been sitting around collecting unemployment checks could instead have been paid to construct things. That could mean weatherization retrofits, it could mean mean train tracks, it could mean large obelisks. Spending money on construction projects at the pace and scale needed to put half of these people to work would have in practice entailed undertaking some low-social-value projects. But would the social value have been lower than the social value of having the unemployed workers do nothing? I don’t see it.

We’re experiencing a massive and largely avoidable plague of idless in which the country is producing far fewer goods and services than it is capable of producing.

Aaron Carroll delivers on my suspicion that the flatlining of health care spending as a percent of GDP in the 1990s is mostly about rapid GDP growth during that decade:

There was a moderation of the rate of increase in the nineties, but clearly growth as such is playing a big role here. This is sort of an obvious point, but the implication is that the debate over “health care costs” is a little bit misguided. Making it easier for English-speaking college graduates to move to the United States would reduce health spending as a share of GDP, but it’s not something a blue-ribbon commission on health care costs is going to come up with.

An except from the President’s statement on the Simpson/Bowles Commission:

This morning, my budget director, Jack Lew, spoke with Chairman Bowles and invited the entire Commission in to meet with him and Secretary Geithner to discuss the Commission’s proposals. Overall, my goal is to build on the steps we’ve already taken to reduce our deficit, like slowing the growth of health care costs, proposing a three-year freeze in non-security discretionary spending and a two-year pay freeze for federal civilian workers, and restoring the rule that we pay for all of our priorities.

I have various disagreements with this, but I think this also illustrates the extent to which the former-Senator President his senior staff full of former legislative aids have become mental prisoners of the legislative process. “Restoring the rule that we pay for all of our priorities” is a references to re-adopting statutory PAYGO rules in the congressional process. That might or might not be a good idea, but if the President does think it’s a good idea he can adopt the same thing unilaterally. He just needs to say “I will veto any bill that increases the deficit relative to current law.”

He needs to really, really mean it. When people say “what about the AMT patch?” he could say “I’m for AMT patches, but only if they’re paid for.” When people say “what about the Bush tax cuts?” he could say “I’m for middle class tax cuts, but only if they’re paid for.” When people say “whata bout the doc fix?” he could say “I’m against cutting Medicare reimbursement rates, but only if it’s paid for.” Repeat that enough times and suddenly it becomes congress’ problem. Congress wants an AMT patch? Fine, then congress needs to pay for it. There are lots of things the President can’t do in the legislative process, but refusing to sign deficit-increasing bills is something he definitely can do.

There’s been a fair amount of blogospheric bemoaning of the failure of policymakers to adequately embrace the idea of fiscal stimulus as a macroeconomic stabilization policy. And as readers know, I’m very sympathetic to that point of view. But at some point when your side of the argument fails to carry the day, you do need to start thinking about why you’re not persuading people.

And in this case, I think the problem is pretty clear: In the absence of broad political consensus about the appropriate size and scope of the public sector, efforts at fiscal expansion will necessarily get bound up with these debates. For example, many liberals view the idea of a payroll tax holiday with suspicion because they suspect the revenue losses will provide further grist for the argument that Social Security is “bankrupt” and benefits need to be cut. Conversely, conservatives are reluctant to support “temporary” increases in the generosity of social welfare programs out of fear that in the future allowing those increases to expire will be characterized as a “cut.”

Or at the state and local level, anyone who’s thinking seriously about the issue ought to see that the middle of a recession is a terrible time to implement major cutbacks in public spending. But at the same time, people who believe in good faith that state and local spending is above the optimal level will understandably agree with Rahm Emannuel that you don’t want to let a good crisis go to waste. After all, were center-left Keynesian economics bloggers issuing table-thumping condemnations of state and local spending increases in 2004-2007? Bemoaning the fact that we’d entered a new “dark age” of macroeconomic understanding where policymakers didn’t realize that under the circumstances states and municipalities ought to be trimming its workforce and accumulating huge surpluses? I think I missed that.

It seems to me that Germany and Sweden entered the recession with less Keynesian political cultures, but more consensus about the long-term equilibrium level of public spending. And not coincidentally, both of those countries did a much better job of executing the “stimulus in the downturn, then move to consolidation when growth returns” move than America has.

So the practical problem facing proponents of a large fiscal role in stabilization is to come up with ideas that work given this setting. That means, I think, doing something to stop state and local fiscal policy from being pro-cyclical and it means doing something to elaborate rule-based monetary policy measures for when nominal interest rates are close to zero.