How do we pay for it?

Yesterday I posted a response to a Paul Krugman column that my sister had sent me. I agreed that the current emphasis on "austerity" was misguided, but wondered whether conventional stimulative policy would be as effective as our models suggest, because we're up against long-run energy constraints that weren't a major concern in earlier recessions. It's not that there isn't work that needs to be done to adapt to an energy-constrained future. Rather, our normal way of paying for these things is based on the premise of greater wealth in the future; if energy constraints make it hard for the economy to grow, then it's not clear how we do pay for the necessary work.

Between my sister and me, she's definitely the practical one, as shown by her immediately asking the sensible next question, "So how do we pay for it?" I don't know. That's a bone I've been worrying at for a few years now and I think I'm getting closer to an answer, though I don't quite have it. My closest approach so far has been in a presentation I made last month at a conference in Cooperstown on local energy. I'm getting the presentation "out" in the form of a series of blog posts, and I'm about halfway through; with any luck, I'll be able to finish them up in the next couple of weeks, but even once I do, my answer to my sister will probably still be, "I don't quite know ... quite yet."

My colleague Jason Antrosio pushed back harder than my sister. His first observation was that the "kinked supply curve" I described in my post could be less about geology and more about geopolitics--specifically, the removal of Iraq from the world oil market due to sanctions, then invasion, then civil war. His second point was that I sound awfully like a structuralist. I'll deal with that second concern first.


There are two common explanations for why the labor market has been in such bad shape since the recession took off in 2008. One school blames inadequate demand: people can't get jobs because households and firms aren't spending enough money to make firms want to hire people, and people aren't spending money because they don't have jobs. If the government would increase its spending (fixing bridges, building high-speed rail, (re)hiring teachers), then more people would have jobs, and they'd spend money, and firms would have more reason to hire people, and so more people would have jobs ... In other words, the classic "pump-priming" idea.

The other explanation is that the economy has "structural" problems. In this story, it's not that there's too little demand in general, it's that there's too little demand for certain kinds of work that got overextended during last decade's boom (construction and finance are too obvious candidates). If that's true, then what the economy needs is time--time for people to retrain from the professions that we had too much of, into ... other professions. And if that's what's needed, then increased government spending isn't particularly useful. In fact, it's harmful, because it'll just cause employment in the overextended sectors to get artificially pumped back up, blunting people's incentive to do the necessary work of getting into another sector.

So when I raise questions about whether standard demand stimulus can work, I can see how that sounds an awful lot like I'm a structuralist, but I certainly don't consider myself one.  First, there's the matter of the evidence. Krugman, and Mike Konczal, and myriad others have a very strong argument in the pattern of unemployment. If the problem were structural, rather than an overall lack of demand, then there would logically have to be plenty of demand in some part of the economy, or for some kind of employment. So there must be a sector of the economy or a kind of work where demand is strong and employment is rising. But there's simply no sign of that. Some sectors have been hurt worse than others, but there's no significant sector of the economy where employment is strong. It's hard to explain that in any way other than an overall lack of demand.

Second, there's the unprecedented long-term unemployment situation. The Bureau of Labor Statistics (BLS) tracks various measures of unemployment, including what's sometimes called the "headline" unemployment rate, their U-3 series. This number gets the most attention, and while it's been seriously bad during this recession, it's not stand-out bad. It peaked at 10% in October 2009, which didn't quite match the highs from the early 1980s, and since then it's made it's way unsteadily down to its May 2012 level of 8.2%. So it's really bad, but it's not unprecedented.

The long-run numbers are another story entirely. The BLS tracks the number of people unemployed more than 26 weeks. They also track the "labor force," defined as those people who have jobs, plus those people who are looking for work. If you take the long-term unemployed and divide it by the labor force, you get this picture:



Data from BLS, series LNS13008636, divided by series LNS11000000

 By this measure, the long-term unemployment problem peaked in early 2010, and while it's improving, it's still far worse than at any time since World War II--which is a polite way of saying "at any time since the Great Depression."

BLS also tracks the average and median duration of people's unemployment. The median data only pick up in July 1967 (the month I was born--Hmm, I wonder if there's anything to that ...), but if you graph the two of them, it looks like this:



Data from BLS, series LNS13008275 and LNS13008276

 The median, like the long-term rate shown above, peaks in early 2010 and then gets better, though its improvement hasn't been as pronounced. The average, on the other hand, perhaps peaked at the end of 2011, but has hardly come down since then. And with both of these series, the comparison to any other post-WWII experience is even bleaker than with the first chart. We've had other structural shifts since the war, but I haven't seen any structuralist explanation as to why this one is beating us up so much worse than any other.

(These charts are also a problem for those who want to blame the unemployment problem primarily on the unemployed. "They just need to get off their lazy asses and get a job!" So we're supposed to assume that the American worker in 2012 has by some mysterious process become twice as lazy as he/she had been at any other time since WWII. Right.)

Finally, I don't agree with the structuralist policy prescription. Structuralists assume there will be greater wealth in the future, as long as the government gets out of the way (except, of course, for taking actions to stroke the egos of the "job creators"). I assume nothing of the kind. I don't think that the main thing standing between us and a brighter tomorrow is high taxes, or regulation, or uncertainty about what crazy thing government might do next.

I think that what stands between us and a brighter tomorrow is an energy constraint, a constraint that, for various reasons, private markets aren't well suited to solve on their own. So government action, far from being useless, is crucial--the 2009 stimulus bill included funds for home energy retrofits, which some people mocked as obviously a crazy idea, but which look to me like some of the most sensible spending imaginable.

But precisely because of the energy constraint, paying for that government action is tricky. If structuralists assume there will be a brighter tomorrow as long as the government gets out of the way, standard Keynesians assume there will be a brighter tomorrow as long as the government "steps up to the plate." And Karl is unsure of brighter tomorrows in general.

Right now, the U.S. government can borrow at a negative real interest rate: people are so eager to lend Uncle Sam money, that the interest the government has to pay on its new loans is less than the rate of inflation. In normal times, any business would be thrilled to be able to borrow on those terms--if it couldn't turn a profit on a loan like that, it really has no business being in business. And in normal times, it would be a no-brainer for any government to seize the opportunity. Econometric studies routinely show that various infrastructure projects have large, positive effects on economic growth. So if you can borrow at negative interest rates and spend the money in ways that make the economy grow faster than otherwise, you can easily raise the money to pay back the loan without raising the tax rate--your richer economy will bring in the higher tax revenue all by itself.

But what happens if you borrow the money, spend it, and your economy doesn't grow, but merely shrinks more slowly than it otherwise would have?

Maybe borrowing the money is the right thing to do even in these abnormal times, as long as it's spent intelligently. As I said above, I haven't figured this out yet. But this post is already longer than I intended, and I haven't even gotten to Jason's point about Iraqi oil. I'll try to follow up on that soon.