The Greeks have voted no in today’s referendum, which makes it more likely (though not certain) that Greece will leave the euro. Some economists believe that this action would help the Greek economy. Devaluation would be an expansionary monetary shock, raising both nominal and real GDP. I have some sympathy for this view, although in the end I favored a yes vote, out of fear of where Syriza was taking Greece.

When economic shocks are very complex, I tend to look to the markets for guidance. In my research on the 1930s, I noticed that markets seemed to understand the implications of policy shocks much earlier and more accurately than pundits and politicians. Ironically, the closest parallel to the current crisis was the German economic crisis of 1931, where Germany was the beleaguered debtor nation, struggling to stay on the gold standard (which was the eurosystem of the interwar years.)

There’s a news story that US stock futures dropped about 1.5% on news of the no vote, and tomorrow morning most asset prices are expected to fall sharply all over the world. But why is that? Greece is a very small economy and is already deeply depressed. Its GDP probably won’t decline much further, and in any case the US exports only a trivial amount of goods to Greece. What’s the mechanism for the stock decline?

One popular answer is “uncertainty”. Sorry, but that’s a cop out. Uncertainty has to be about something important. There’s uncertainty over who will win the World Series in baseball, but it doesn’t affect stock prices. It turns out that the interwar period might give us a clue.

In 1931, even Germany was much too small to have a major impact on US stock prices. And German troubles did not affect US stocks until mid-1931, when the Germany financial crisis began to threaten the international gold standard. And then when the US left gold in early 1933, German difficulties no longer correlated with US stock prices. This sort of “time varying correlation” is a powerful tool for understanding causality.

In the 1930s, actual devaluation was expansionary, and helped countries recover. However expectations of devaluation were often contractionary. This was because fear of devaluation led people to hoard gold, which increased the value of gold and reduced prices in all gold standard countries. Greece is important not because of Greece itself, but because of fears that it could lead to turmoil in other much larger southern European countries, and lead to a run on their bonds. The public might choose to hoard safer assets like US dollars, putting deflationary pressure on the US.

The one asset price that seems to respond positively to Greek turmoil is safe bonds. The price of these bonds rises, as their yields fall on eurozone turmoil. This is a good indication that Greece’s no vote was a significant deflationary shock to the world economy. I choose the term ‘significant’ for a reason. It’s not a major shock of the sort that is likely to lead to another global recession. But it’s also not trivial. Recall that stocks were already somewhat depressed by fears of a no vote, so the total impact on US stock prices is certainly more than 1.5%. Of course European stocks are affected much more significantly.

Let’s suppose that economists like Paul Krugman are completely correct. The no vote was the right move. Let’s also assume it leads Greece out of the eurozone. Let’s also assume that next year Greece begins to recover strongly with a devalued currency. None of this is certain, but let’s assume it’s all true. It still remains true that today’s vote was a deflationary shock to the global economy. It still remains true that several years from today the global unemployment problem will probably be worse than if there had been a yes vote. The no vote will probably have a net negative effect on the global economy.

Now of course Greek citizens were fed up with the rest of the world, and I can’t blame them. Average people don’t have a sophisticated understanding of economics (indeed I’m not sure I understand all this stuff). All they know is that they keep doing things the EU tells them to do, and Greece keeps getting worse off. Perhaps Greek voters could be blamed for electing corrupt politicians in previous decades, but that’s not how they would view the current crisis. So I’m not surprised that Greeks would vote in the way that they thought was best for Greece.

Indeed in my forthcoming book on the Great Depression I praised FDR for devaluing the dollar in 1933, even though his actions may have worsened conditions in Europe. I suppose I could reconcile that view with my “yes” view on this referendum, by arguing that it was far easier to leave the gold standard than to leave the euro. The gold standard was clearly on the way out, and the sooner it collapsed the better. If the eurosystem eventually collapses, those pundits who suggested a no vote will be proved right and I will be proved wrong this time around. If it survives, as I think it will, then the Greek vote will turn out to harm the global economy, and was regrettable on utilitarian grounds.