[Note: If you're short on time, just click through and read the broken window fallacy Wikipedia page. It could change the way you think about economics and how you evaluate government policy. Also, I've put a couple disclaimers at the bottom of this post. I had to move them down there because it was just too much non-content leading the post.]
The latest episode of the Planet Money podcast - Why Japan Will Bounce Back - may have been misleading. A few weeks ago, I wrote the following tweet, but did not publish it (there are actually two drafts here - yes, I write drafts of tweets):
Well, apparently the economic stimulus bug has hit Japan! They'll need to create hundreds of jobs to replace all that infrastructure!
What we call a disaster in Japan, Keynes would call phase one of a stimulus program.
I didn't publish it because I was afraid it would be misread, and people would think I was making light of the devastation in Japan. I was really making light of the tunnel vision that many have regarding government's ability to directly improve the economy through spending. Then, I heard the economist Planet Money booked to discuss the Japanese recovery flirting with the idea that Japan's economy could see a net benefit from the disaster.
At first, they were just discussing economic resilience of industrialized countries. But they took it a little further from "Can Japan bounce back?" to:
Question from interviewer: Is it possible to come out with a net economic positive for Japan?
Economist: In terms of GDP and production flows you might be able to say that, except for the loss of wealth. You could've been using that same money - those same resources - for doing other things other than rebuilding.
In writing, this doesn't seem quite so outlandish. But pretty much everything starting with "except for the loss of wealth" was said almost as a throwaway. It sounded like the economist thought the loss of wealth and redirection of resources would be insignificant relative to the overall "stimulus" provided to the rest of the economy. (At one point, he did compare Japan's recovery to the economic recovery resulting from the stimulus in the States, so I know "stimulus" was on his mind when he gave the previous answer.)
I think what may have happened was the Planet Money team got caught up in the hook for their story: countries like Haiti and other impoverished countries do not bounce back catastrophe in the way that developed superpowers like Japan can. That's true, but they got a little carried away and started veering towards the idea that the disaster could ultimately be a benefit to the Japanese economy.
Another quote (my emphasis added):
You could say, possibly over the next year or two, we'll see some strong positive benefits including the construction of new highways, new railroads, new cell phone towers and telecommunications networks, new housing for people to replace the old stock. So we could end up with more modern, more productive, more efficient amount of capital out there than we have at the moment.
The problem is they are flirting with the broken window fallacy. Of course the result is "strong positive benefits" and newer infrastructure to "replace old stock", but that doesn't mean the Japanese economy will see a net improvement. In order to think that, we have to ignore all of the "old stock" that was lost. That infrastructure had value, and the new infrastructure will cost money to build. This is a net loss to Japan unless all of the infrastructure lost was essentially value-less before it was destroyed. All the old stuff was still there because it still had economic value, not because they just couldn't get around to building more modern stuff. I'm sure there were some condemned buildings or whatever, but most of the stuff they lost was economically useful and so represents substantial economic loss.
I will note that they led with a discussion about how GDP doesn't capture lost assets that were previously created or acquired. The example they gave was a bridge built 10 years ago: if the bridge disappears, GDP doesn't take a hit even though there has been economic loss. So it's possible that the quotes above were said in light of this idea that GDP could look better once Japan starts the recovery (it will). But it didn't sound that way to me. It seemed like they acknowledged the shortcoming of the GDP calculation and then the rest of the conversation - including the quotes above - were about the economy in general (not in terms of GDP).
Ironically, most people would read this and think, "Duh. Of course there was a loss. All that stuff got wiped out." In this case, common sense rules the day. If there is a building, and then the building is gone, that building has been lost. But economists sometimes like to get fancy and out think themselves. Hence the broken window fallacy mentioned earlier.
The disaster will result in economic benefits, just not to Japan as a whole. My guess is that geographically proximate countries will benefit as they will likely need to supply Japan with materials and goods while Japan bounces back. Competing economies will be able to sell more of their own goods (and possibly at higher prices) while Japan is offline. Even within Japan, smaller parts of the economy will benefit at a cost to the overall economy. The construction industry will surely benefit, as will other industries whose services will be required to rebuild. But the bottom line is that Japan lost a lot of its still-useful infrastructure, and that represents an economic loss. Estimates are that it will cost $100 billion to hundreds of billions of dollars for Japan to rebuild, and that's a good estimate for how much wealth Japan has lost as a result of the disaster.
What do you think? Am I just seeing monsters under my bed?
- I like NPR's Planet Money podcast. A lot. I listen to every episode, read the blog and follow them on Twitter. They do a great job of covering difficult material in an objective, approachable and understandable way. I've learned a lot from their podcast and I have recommended it to several people. This is the only episode I remember having serious qualms with, so don't let this discourage you from subscribing.
- I could be totally out of line here. I'm using some quotes from the podcast to make my point, but I encourage you to go listen to it for yourself. It's only 20 minutes long, and you can skip to about the 8:30 mark for the portion of the show that I'm discussing. I'm not quote mining, but I also can't transcribe the entire podcast, so I've tried to find a middle ground without taking things out of context.