Looks like steak won't be on the menu tonight. Photographer: Daniel Acker/Bloomberg
Looks like steak won't be on the menu tonight. Photographer: Daniel Acker/Bloomberg

Ben Domenech of the Federalist and James Pethokoukis of the American Enterprise Institute are arguing about inflation and what it means. Domenech points to rising food prices as evidence of bad government policy; Pethokoukis blames structural factors. This is a long-running and lively meme that’s current not just among Republicans, but also among moderates. Last weekend, I heard a two-time Barack Obama voter express the view that the government was deliberately excluding food and oil from its calculations in order to hide inflation and cheaping out on cost-of-living adjustments for federal workers and retirees.

Because this is not true, I thought I should weigh in on what’s going on, how and why it matters, and what the government should do about it.

First note: The cost of food really has risen quite a lot over the last 10 years. When you compare commodity prices to the summer of 2004, you’ll see that the price of pork has risen nearly 50 percent, chicken has risen by more than a third, and beef has roughly doubled, as has wheat. All this is well above the general inflation rate. Oil . . . well, I don’t have to tell you what has happened to gas prices, do I? They’re more than twice as high as they were in 2004, and they have even flirted with a triple-bagger.

That cuts into household budgets considerably: The McSuderman household, for one, has not seen a steak grace its grill for quite some time, yet our grocery budget has increased by more than 30 percent. So it’s no wonder that people are incredulous and angry when the government tells them that inflation is low. Unless you’ve got a good Crock-Pot recipe for LCD televisions, it’s no good pointing out how marvelously cheap they’ve gotten when John Q. Public asks you how he’s supposed to feed a family of five at these prices. Or even to point out all the amazing deals now available on Oreos and Fruit Roll-Ups; Americans have become accustomed to setting their tables with meat, eggs, milk and fresh produce, and they get touchy and anxious when they’re told they can’t have as much of those things as they used to.

And yet, as Pethokoukis notes, the Federal Reserve is fundamentally right to ignore food prices when it's trying to figure out what monetary policy should be. Food prices aren’t increasing because of bad monetary policy; they’re increasing for a number of reasons, which are as varied as pig viruses and a billion Chinese consumers who have started eating higher on the food chain. Fed policy doesn’t cause those things, and it can’t fix them, either.

What we’re seeing in this debate is a sort of short circuit in the policy process that crops up from time to time: a disconnect between two different, and equally valid, understandings of what statistics are for.

To see what I mean, let me wind the clock back for a moment to an old debate that happened a long time ago, on a blog far, far away. OK, actually in New York, which is certainly convenient by Acela but still pretty far, all things considered.

In 2008, the Wall Street Journal ran an op-ed from Andrew Wilson that said, in part, that unemployment was still high at the end of the Great Depression, with nearly 1 in 5 workers unemployed as late as 1938. Historian Eric Rauchway said Wilson was “lying”; using an alternative method that included public work-relief programs such as the Works Progress Administration, Rauchway said unemployment was more like 12 percent. Alex Tabarrok of Marginal Revolution called Rauchway’s accusation nonsense. I got involved, somewhat intemperately.

Six years later, I still agree that the accusation of lying was nonsense. But on the broader question -- how bad was unemployment during the Great Depression? -- the interesting thing is that both measurements were completely valid. The debate really hinged on the question of what unemployment statistics are for.

If you primarily think of unemployment as a measure of human misery, then Rauchway is right: Whatever else you think of programs such as the Civilian Conservation Corps and the WPA, they reduced the number of people who couldn’t find work. Because being unemployed is one of the worst things that can happen to you, that’s really important. These programs may even have kept people employable who otherwise would have dropped out of the labor force permanently, though I’m more skeptical on that point because the labor shortages of World War II meant that anyone who could breathe and carry a tool bag ultimately ended up back at work.

But while this is an important reason to measure unemployment, it is not the only reason to measure unemployment. We also look at unemployment to gauge the underlying health of the economy. Wilson and Tabarrok’s point, which is also totally fair, is that in 1938, the economy was still generating disastrously low demand for labor.

When I wrote about this six years ago, a lot of commenters got sidetracked into discussions of whether government jobs were “real” jobs, and if not, why not. This is not the point. The WPA jobs were certainly real enough to the people who collected the paychecks, and I am sure most of those people performed their work honorably. The fact remains that those jobs were created mostly for the purpose of creating jobs; whatever actual work was performed was quite secondary. It was the mass equivalent of the industrialist who creates a cushy job for his useless nephew -- the nephew is employed because the uncle wants him to be employed, not because his labor is worth more to the company than his salary.

That matters because a healthy economy is one that generates things for people to do by itself. I’m not arguing against job creation programs in time of crisis -- indeed, I’ve suggested that we might look into them as a possible way to solve our own current employment problems. Nor am I denigrating the people who took those jobs -- unlike the proverbial useless nephew, those people were victims of macroeconomics, not their own incompetence. I’m simply pointing out that as long as the government needs to create millions of jobs for the primary purpose of employing people rather than for the primary purpose of doing things that we would like done, then the economy is still in a very bad way. In 1938, the normal processes of generating demand for labor and goods and services were still very much broken; the make-work programs were alleviating some of the misery, but they did not, as President Franklin Roosevelt had hoped, actually jump-start the economic engine.

Now back to inflation. As with unemployment, there are two different, and equally valid, questions we could be asking when we look at inflation statistics:

  • Are people finding it harder to maintain their standard of living?
  • Is monetary policy too tight or too loose?

These are actually very different questions, which doesn’t mean that one is more important than the other. The Fed is interested in whether there is too much money in the system, which would show up as a broad increase in all prices. That’s why it is probably more useful for them to look at “core” inflation, which excludes food and energy prices. Food and energy are very volatile because they’re vulnerable to supply shocks -- a drought, a pig virus, unrest in the Middle East that might disrupt oil shipments. The Fed doesn’t cause those things, and it can’t fix them, either. If they use broader measures that include all this volatility, then a big oil shock would signal that it should raise interest rates -- even though an oil shock is a recessionary force and probably calls for looser money to offset the economic contraction. The end result would be higher unemployment every time Iran gets frisky with its neighbors.

But that doesn’t mean that the Fed is denying what’s happening in the food aisle. It just means that it's denying it has the power to do anything about it. Which is perfectly true.

Domenech argues that there are other policy areas where food prices should matter, such as agriculture policy. And, certainly, I’d be thrilled if rising food prices caused us to revisit our unnecessary, counterproductive agricultural protections. But we should also be modest about how much difference those policies would make. Ultimately, what’s driving food and energy prices higher is rising demand in Asia, combined with various local production problems. There’s no policy cure for those things; mostly it’s going to be solved by consumers reaching for the chuck roast or the tofu instead of the prime cuts.

To contact the writer of this article: Megan McArdle at mmcardle3@bloomberg.net.

To contact the editor responsible for this article: Brooke Sample at bsample1@bloomberg.net.