Saving is not a problem
By Scott Sumner
I recently came across a WSJ article that is a gold mine of teachable moments for EC101 students:
KORSCHENBROICH, Germany–Two years ago, the European Central Bank cut interest rates below zero to encourage people such as Heike Hofmann, who sells fruits and vegetables in this small city, to spend more.
Policy makers in Europe and Japan have turned to negative rates for the same reason–to stimulate their lackluster economies. Yet the results have left some economists scratching their heads. Instead of opening their wallets, many consumers and businesses are squirreling away more money.
When Ms. Hofmann heard the ECB was knocking rates below zero in June 2014, she considered it “madness” and promptly cut her spending, set aside more money and bought gold. “I now need to save more than before to have enough to retire,” says Ms. Hofmann, 54 years old.
Recent economic data show consumers are saving more in Germany and Japan, and in Denmark, Switzerland and Sweden, three non-eurozone countries with negative rates, savings are at their highest since 1995, the year the Organization for Economic Cooperation and Development started collecting data on those countries. Companies in Europe, the Middle East, Africa and Japan also are holding on to more cash.
Economists point to a variety of other possible factors confounding central-bank policy: Low inflation has left consumers with more money to sock away; aging populations are naturally more inclined to save; central banks themselves may have failed to properly explain their actions.
But there is a growing suspicion that part of problem may be negative rates themselves. Some economists and bankers contend that negative rates communicate fear over the growth outlook and the central bank’s ability to manage it.
Here we see a failure to understand the “paradox of thrift”, and also a case of reasoning from a price change. Because the article focuses on demand-side effects, let’s look at the equation of equation:
MV = PY
If we hold M fixed, then the impact of lower rates on NGDP will depend on their impact on velocity. But what is that impact? It depends:
1. Assume base money pays no interest and the market interest rate on other assets falls. Then velocity will decline, and so will NGDP. This is why bearish growth expectations often created recessions back in the gold standard days—they led to lower nominal (and real) interest rates and this led to more demand for gold. The supply of gold was fairly stable in the short run. As gold velocity fell, NGDP also fell. Larry Summers and Robert Barsky wrote a paper explaining this effect.
2. If, however, the central bank reduces the interest it pays on base money, then this will reduce the demand for base money, and increase velocity. That’s why European asset markets responded positively to a cut in the interest rate on reserves, even into negative territory.
It’s not enough to talk about the impact of low interest rates, you need to first ask why interest rates are low.
The second problem is that the article seems to assume that more saving is bad for the economy. Actually, saving tends to fall during recessions and rise during booms, for reasons related to the permanent income theory—people tend to save income flows that are temporarily above normal.
The authors may be confused about the old Keynesian “paradox of thrift”. Many people wrongly believe that this paradox predicts what will happen if people save more. Not so, in the paradox of thrift, actual saving does not increase at all. Rather, an attempt to save more leads to a fall in national income, and people don’t end up saving any more at all. Indeed in more sophisticated Keynesian models with investment accelerators, the ex post quantity of saving (and investment) will actually rise during a boom, even as a share of GDP.
The problem with the article is that they cite lots of evidence that actual, ex post, saving has been rising in Europe. If true, that’s usually a bullish sign for the economy. In contrast, the paradox of thrift is all about increases in ex ante, or intended saving. And these shifts do not result in increases in actual saving, or if they do then there is no contractionary impact.
The more fundamental problem here is that people confuse saving and hoarding. Actual saving has no first order effect on GDP. It doesn’t cause anything. Nor do interest rates cause anything. Rather saving and interest rates are caused by more fundamental factors, such as monetary policy, business confidence and productivity shocks. To evaluate the effect of these changes on the economy, you always need to start with the more fundamental shocks. In contrast, an increased demand for money—i.e. hoarding—does have a contractionary impact on NGDP. It is a fundamental factor.
And finally, the entire premise of the article seems flawed. Yes, Europe is not doing particularly well in absolute terms, but growth has picked up over the past few years. So there’s no need to explain why negative rates failed to boost growth, because growth actually did accelerate.
A better argument is that negative IOR is weak medicine, and the ECB needs to change their target, preferably to NGDPLT, but at a minimum to price level targeting.