The broken window fallacy fallacy
The broken window fallacy fallacy
That which is chosen, and that which is not chosen.
Last week, the NZ Association of Economists’ Twitter account shared a link to an opinion piece in the Financial Times titled “Taylor Swift and the fallacy plaguing modern economics”. I’ll give them the benefit of the doubt and note that retweets are not endorsements; indeed it seems they were just trying to get in on some of the recent Swift-related economic commentary, rather making a particular point. The ‘Eras’ tour was reportedly big enough to have an impact on measured inflation and economic activity in some instances.
The FT article is more of a general lament about activities that add to measured GDP but are otherwise of dubious value to society. But for a piece about bad economic reasoning, it itself starts off with some bad economic reasoning:
Most recently, commentators have asserted that Taylor Swift’s concert tours have added hundreds of millions to the US and UK economies. What they fail to consider is the counterfactual: how Swifties would have spent their ticket money otherwise.
Saying that money spent on one thing could have been spent on something else is true by construction, so this tells us nothing. What the author is insinuating here is that they could have spent it on something better. And better according to who? The author, of course. Never mind that the fans might be the best judges of what value they place on the Tay-Tay experience.
On its own, defending the honour of Swifties probably wouldn’t warrant more than a tweet from me. But it reminded me of a topic that I’d had on my list to write about at some point. (Certainly I came up with the title a long time ago.)
The article brings up the ‘broken window fallacy’, an idea that comes from French economist Frédéric Bastiat in an 1850 essay called “That Which Is Seen, and That Which Is Not Seen”. He uses the parable of a shopkeeper whose son accidentally breaks a window:
Suppose it cost six francs to repair the damage, and you say that the accident brings six francs to the glazier’s trade – that it encourages that trade to the amount of six francs – I grant it; I have not a word to say against it; you reason justly. The glazier comes, performs his task, receives his six francs, rubs his hands, and, in his heart, blesses the careless child. All this is that which is seen.
But if, on the other hand, you come to the conclusion, as is too often the case, that it is a good thing to break windows, that it causes money to circulate, and that the encouragement of industry in general will be the result of it, you will oblige me to call out, “Stop there! Your theory is confined to that which is seen; it takes no account of that which is not seen.”
It is not seen that as our shopkeeper has spent six francs upon one thing, he cannot spend them upon another. It is not seen that if he had not had a window to replace, he would, perhaps, have replaced his old shoes, or added another book to his library. In short, he would have employed his six francs in some way, which this accident has prevented.
In other words, repairing a broken window creates work, provides someone with an income, and adds to measured GDP (although that concept didn’t exist in Bastiat’s day). But it would be ridiculous to conclude from this that paying people to break windows, then paying other people to fix them, could be the basis for a functioning economy.
The point here is that the strength of our economy is best represented not by ‘flow’ measures like output or income, but in terms of wealth. If an asset is destroyed, then replaced, we are in no way better off than we were before - and probably worse off, because the time and effort spent on replacing that asset means that we miss out on other opportunities to expand our wealth. (So why do we obsess over GDP? Short answer is that it’s easier to measure.)
Of course, this has nothing to do with Taylor Swift concerts - unless you think that they are themselves a destructive force, descending on each town in the manner of a tornado. Regardless of your personal tastes in music, you’d have to be a real snob to say that with a straight face.
Where Bastiat’s argument might be relevant, though, is with natural disasters. Indeed, my first encounter with it was after the Christchurch earthquake in 2011, but there have been plenty of occasions to think about it since then - the Kaikoura earthquake, Cyclone Gabrielle, and of course the prospect that climate change will make severe weather events more frequent.
Estimates of the cost of the Christchurch rebuild varied a lot as time went on, but not too long after the event they converged at around $30 billion. That was about 10% of annual GDP at the time - spread out over several years of course, but at its peak, it was quite likely that it would have a visible impact on GDP at the national level.
At some point, a handful of economists began to promote a different line. No, no, you’ve got it all wrong, the rebuild isn’t good for the economy at all! Don’t you know the broken window fallacy? Replacing things that were broken doesn’t make us richer! At the time we didn’t feel there was any serious need to address this - our work was simply about tracing through the effects that the rebuild would have on the economy, with no connotations of ‘good’ or ‘bad’.
But it did get me thinking about what would be the best way to counter this argument. And where I came to is that there’s an overlooked assumption in here about the role of choice.
Choosing to smash things and then paying someone to fix them doesn’t make the country better off. But we don’t choose to have earthquakes. They just happen, and when they do, the only choices we have are to rebuild or to not rebuild. And of those two, rebuilding is clearly the better option for the economy. Why? Because those those things that we lost had value to us. And how do we know that they had value? Because we built them in the first place.
So: natural disasters bad, rebuilding after natural disasters good. And only one of these things is within our control.
Or maybe not? As I said, we don’t choose to have earthquakes, or floods or cyclones - but we can, and do, choose to build on fault lines and flood plains. Why do people take on that risk?
One reason is they might judge that the lifetime benefits of that location outweigh the costs. There are plenty of real-life examples of this. In developing countries, people will live on flood plains because that’s where the most fertile soils are. And closer to home, it’s the strongest reason to not shed too many tears for those who own beachfront property and will (eventually) have to contend with rising sea levels.
Another reason is that they may be banking on someone else covering the cost when disaster strikes. This seems to raise the concept of moral hazard, where people may take on greater risks once they know that someone else, such as an insurer, will bear the downside of that risk.
I don’t think that moral hazard is relevant in this example though. The only risky behaviour that’s possible here is to put yourself in the path of a natural hazard. That’s not a post-contract change of behaviour; an insurer can observe this in advance, and decide whether or not they’re willing to provide cover. If they are, and if the property owner is willing to pay a premium that reflects the risk, then there isn’t a problem.
Things get a bit greyer when there’s the possibility that the government, rather than an insurer, might underwrite the risk. To be clear, a government-led bailout can be the best solution in some situations - for instance, Christchurch hadn’t previously been identified as a high-risk area for earthquakes, so neither property owners nor insurers could be held responsible for taking on that risk. The trouble is that elected governments find it difficult to avoid granting repeated bailouts, which in turn gives people less incentive to remove themselves from the hazard.
On the whole, though, I don’t think we’re making the case that repairing things is a loss to the economy. The key to whether the ‘broken window fallacy’ applies is determining the point at which choice comes into play. If a window is broken by accident and we choose to fix it, that’s better for the economy than the alternative. If we choose to break windows, that’s a different matter.
In Bastiat’s essay, he imagines people offering this consolation: “what would become of the glaziers if panes of glass were never broken?” Let’s try replacing his glaziers with another profession: “what would become of the doctors if people never got sick?” It’s a redundant question; the fact is that people do get sick, and not by choice. And in those circumstances, it’s not hard to imagine that the money we spend on doctors might actually be the highest-value use of that money.