[Joseph E. Stiglitz] What’s wrong with negative rates?

I wrote at the beginning of January that economic conditions this year were set to be as weak as in 2015, which was the worst year since the global financial crisis erupted in 2008. And, as has happened repeatedly over the last decade, a few months into the year, others’ more optimistic forecasts are being revised downward.

The underlying problem — which has plagued the global economy since the crisis, but has worsened slightly — is lack of global aggregate demand. Now, in response, the European Central Bank has stepped up its stimulus, joining the Bank of Japan and a couple of other central banks in showing that the “zero lower bound” — the inability of interest rates to become negative — is a boundary only in the imagination of conventional economists.

And yet, in none of the economies attempting the unorthodox experiment of negative interest rates has there been a return to growth and full employment. In some cases, the outcome has been unexpected: Some lending rates have actually increased.

It should have been apparent that most central banks’ precrisis models — both the formal models and the mental models that guide policymakers’ thinking — were badly wrong. None predicted the crisis; and in very few of these economies has a semblance of full employment been restored. The ECB famously raised interest rates twice in 2011, just as the euro crisis was worsening and unemployment was increasing to double-digit levels, bringing deflation ever closer.

They continued to use the old discredited models, perhaps slightly modified. In these models, the interest rate is the key policy tool, to be dialed up and down to ensure good economic performance. If a positive interest rate doesn’t suffice, then a negative interest rate should do the trick.

It hasn’t. In many economies — including Europe and the United States — real (inflation-adjusted) interest rates have been negative, sometimes as much as minus 2 percent. And yet, as real interest rates have fallen, business investment has stagnated. According to the Organization for Economic Cooperation and Development, the percentage of gross domestic product invested in a category that is mostly plant and equipment has fallen in both Europe and the U.S. in recent years. (In the U.S., it fell from 8.4 percent in 2000 to 6.8 percent in 2014; in the EU, it fell from 7.5 percent to 5.7 percent over the same period.) Other data provide a similar picture.

Clearly, the idea that large corporations precisely calculate the interest rate at which they are willing to undertake investment — and that they would be willing to undertake a large number of projects if only interest rates were lowered by another 25 basis points — is absurd. More realistically, large corporations are sitting on hundreds of billions of dollars — indeed, trillions if aggregated across the advanced economies — because they already have too much capacity. Why build more simply because the interest rate has moved down a little? The small and medium-size enterprises that are willing to borrow couldn’t get access to credit before the ECB went negative, and they can’t now.

Simply put, most firms — and especially SMEs — can’t borrow easily at the T-bill rate. They don’t borrow on capital markets. They borrow from banks. And there is a large difference (spread) between the interest rates the banks set and the T-bill rate. Moreover, banks ration. They may refuse to lend to some firms. In other cases, they demand collateral (often real estate).

It may come as a shock to non-economists, but banks play no role in the standard economic model that monetary policymakers have used for the last couple of decades. Of course, if there were no banks, there would be no central banks, either; but cognitive dissonance has seldom shaken central bankers’ confidence in their models.

The fact is that the eurozone’s structure and the ECB’s policies have ensured that banks in the underperforming countries, and especially in the crisis countries, are very weak. Deposits have left, and the austerity policies demanded by Germany are prolonging the aggregate-demand shortfall and sustaining high unemployment. In these circumstances, lending is risky, and banks have neither the appetite nor ability to lend, particularly to SMEs (which typically generate the highest number of jobs).

A decrease in the real interest rate — that on government bonds — to minus 3 percent or even minus 4 percent will make little or no difference. Negative interest rates hurt banks’ balance sheets, with the “wealth effect” on banks overwhelming the small increase in incentives to lend. Unless policymakers are careful, lending rates could increase and credit availability decline.

There are three further problems. First, low interest rates encourage firms to invest in more capital-intensive technologies, resulting in demand for labor falling in the longer term, even as unemployment declines in the short term. Second, older people who depend on interest income, hurt further, cut their consumption more deeply than those who benefit — rich owners of equity — increase theirs, undermining aggregate demand today. Third, the perhaps irrational but widely documented search for yield implies that many investors will shift their portfolios toward riskier assets, exposing the economy to greater financial instability.

What central banks should be doing is focusing on the flow of credit, which means restoring and maintaining local banks’ ability and willingness to lend to SMEs. Instead, throughout the world, central banks have focused on the systemically significant banks, the financial institutions whose excessive risk taking and abusive practices caused the 2008 crisis. But a large number of small banks in the aggregate are systemically significant — especially if one is concerned about restoring investment, employment, and growth.

The big lesson from all of this is captured by the familiar adage, “garbage in, garbage out.” If central banks continue to use the wrong models, they will continue to do the wrong thing.

Of course, even in the best of circumstances, monetary policy’s ability to restore a slumping economy to full employment may be limited. But relying on the wrong model prevents central bankers from contributing what they can — and may even make a bad situation worse.

By Joseph E. Stiglitz

Joseph E. Stiglitz, a Nobel laureate in economics, is a university professor at Columbia University and chief economist of the Roosevelt Institute. — Ed.

(Project Syndicate)