Recession risk and the circularity problem
By Scott Sumner
[Trigger warning for acrophobics like me.]
I have fond memories of growing up in the Midwest, and seeing big white mushroom-shaped water towers on the outskirts of small towns:
Back in 1970, a friend of mine told me about some guys he knew that played a game of dare. They’d go to the top of one of these, and then gradually crawl down as far as they dare, before climbing back up. But if you make a mistake and go to far, over the horizon, then it’s all over. I’m a bit acrophobic, so it makes me queasy every time I think about this story, which is several times a year.
The reason I frequently think about this story is because it reminds me of the “circularity problem” in economics. Consider the following parable:
A great power elects a leader who is not well informed on economic theory. He is inclined to adopt reckless economic policies, especially trade policies. But this leader is also a former businessman, and pays a lot of attention to the performance of the stock market. Indeed he views it as a sort of indicator of his effectiveness.
Stock investors see a healthy economy, with the main risk being reckless trade policies imposed by the leader. Whenever investors see the risk of these policies increasing, stock prices fall. But they don’t fall very much, as investors know that the leader pays attention to the stock market, and in the end will likely pull back from reckless policies. The leader knows that the stock market doesn’t like some of his polices, and uses it as a sort of gauge to determine if he’s gone too far.
The problem here is that the stock market will only give “warnings” to the extent that they think things have already gone to the point where they won’t be corrected, and there’s increased risk that the economy will slide into recession. Not a 100% risk of recession, but slightly higher than before. Meanwhile the leader knows that stocks bounce around for lots of reasons, and that only a big drop is a clear sign that he’s gone too far and needs to pull back. But investors know that the leader knows that investors are evaluating the risk of a policy accident. Think of an infinity of mirrors in a barbershop, another memory from my childhood.
In 1997, Ben Bernanke and Michael Woodford wrote a paper on the “circularity problem”, where Fed policymakers look for market guidance that they are off course, and then correct policy on that basis. Do you see the problem here? If the markets know the Fed will correct, they’ll never give a warning in the first place. As a result, I’ve always designed my NGDP futures targeting proposals with their paper in mind. The key is not to have markets predict the outcome, but rather predict the instrument setting that leads to an on target outcome (which is also their view, BTW.)
In the parable with the economically ill-informed leader who watches the stock market for guidance, no such mechanism is in place. It’s still true that watching the market is probably better than ignoring the market, but it’s not true that the market will definitely give the leader sufficient guidance to avoid a big mistake, for the simple reason that if that were true then there would be no guidance in the first place—stocks would rise along a path reflecting the healthy state of the economy, complacently assuming the leader would in the end avoid major mistakes due to market guidance.
In the water tower case, no one actually died (as far as I know). But there might have been a point where it was too late to turn back, and the reckless young man (why is it always men?) slid at increasing speed to his death. In macroeconomics, there may come a point where it’s too late for policymakers to get back on course in time to prevent a recession. In that case, stocks may plunge as fast as that unfortunate daredevil on the water tower. October 1929 and October 2008 were two such examples.
PS. I’d love to see a film like Free Solo, as I love mountains. But when that camera looks straight down . . .