clock menu more-arrow no yes

Should the Fed binge on M&Ms in September so it can go on a diet next year?

Alex Wong/Getty Images

When the economy enters a recession, the Federal Reserve normally responds by reducing interest rates to try to stimulate demand for business equipment, houses, cars, and durable goods. When I wrote that the Fed shouldn't raise rates this fall given how benign the inflation rate remains, many people emailed me arguing that the Fed should raise interest rates sooner rather than later because right now rates are at zero, so they can't really be cut. A hike in September or October would give the Fed more ammunition for the next recession.

This is a doubly misguided view. First, it's not true that there's nothing the Fed can do to stimulate the economy with interest rates at zero. Second, even if it were true that the Fed has no other tools, the proposed rate hike wouldn't solve the problem.

Returning to the status quo is no better than sticking with it

It is true that if the Fed raises interest rates this fall, then it has the option of lowering them in the spring if the economy looks weaker. But it is also true that if the Fed doesn't raise interest rates this fall, then it still has the option of deploying zero interest rates in the spring.

The hike-now-to-cut-later proposition is a bit like saying that I should gorge myself on M&Ms this afternoon so that I can lose weight next week by cutting down on snacks. This kind of thing might work as a psychological trick — I smoked more than ever in the three months before I quit smoking — but it doesn't accomplish anything biologically or economically.

The Fed can do more stuff

Moreover, it's simply not true that there are no stimulative monetary policy measures the Federal Reserve can take when short-term interest rates are already zero.

Policy options include:

  • Printing money to buy longer-term government bonds (quantitative easing)
  • Printing money to buy foreign government bonds (currency devaluation)
  • Abandoning inflation rate targeting in favor of price level (or nominal income) targeting

Last but by no means least, even though academic economists thought for a long time that it would be impossible to have interest rates that go below zero, this turns out to be empirically false. Academics had thought that because paper money pays a zero percent interest rate (i.e., it's a piece of paper), electronic bank deposits could never pay a lower rate than that, since people would just hold cash instead.

Why the "zero bound" turns out to be a myth

But it turns out that storing gigantic piles of cash in a safe way is expensive and inconvenient, and using giant piles of cash as a payment medium is even more expensive and inconvenient.

You may have heard, for example, that Apple has a $178 billion cash stockpile. This is obviously not an actual pile of physical cash. If he really wanted to, Tim Cook could presumably order the construction of a massive warehouse full of money located in an offshore tax haven. He could hire armed guards to watch over the complex. He could commission a fleet of ships to carry cash around the world to pay Apple's suppliers and collect its revenue. But this would be very expensive. And for all the same reasons that it would be annoying for Apple to conduct business in this manner, Apple's key suppliers and subcontractors would be pretty annoyed if Apple tried to pay them in the form of trucks full of paper money.

Academics are not practically minded people, so this was not really in their minds when they formed the conventional wisdom about the impossibility of negative interest rates. But over the winter of 2014-'15, plenty of European countries experienced negative interest rates, so we know they are possible.

"Higher rates today to cut rates tomorrow" turns out to be a bad solution to a problem that doesn't even exist.