Negative Interest Rates: A Useful But Limited Tool

September 16, 2016 11:30 AM

Negative interest rates have attracted much attention this year, most of it critical. Many investors are outraged at the idea that the value of their deposits or bonds might decline steadily over time instead of grow. Yet one of the options the Bank of Japan is considering for the forthcoming review of its policy framework is to lower the interest rates it pays banks further below zero.1 The evidence shows that negative interest rates can be a useful form of monetary stimulus, although the net benefits differ across countries. In all economies, the scope for pushing rates into negative territory is limited.

This year’s Geneva Report on the World Economy, “What Else Can Central Banks Do?” describes a range of tools central bankers can use to stimulate growth and achieve moderate inflation when the traditional policy interest rate has fallen to zero. These tools include forward guidance about future policy actions, negative interest rates, quantitative easing (QE), and raising inflation expectations through various channels including changes in the level or specification of the central bank’s operating target.

Interest rate cuts in negative territory operate through several channels in ways comparable to rate cuts when rates are positive. These channels include money and capital markets, foreign exchange markets, and wholesale banking markets. The notable exception has been retail banking, where deposit and lending rates have not followed policy rates down.

The unwillingness of banks to charge retail customers negative interest rates on their deposits has raised concerns that negative rates will reduce bank profitability and could reduce bank lending. A recent IMF paper suggests that negative interest rates have been useful in spurring growth in the euro area without damage to bank profits, but it warns that further cuts may have diminishing net benefits and could hurt bank profitability.

Former Federal Reserve chair Ben Bernanke recently argued that negative rates should not be ruled out in the event of a future US economic slowdown. As noted by my colleague, Adam Posen, negative rates are likely to be more economically effective and politically acceptable in countries with open and diversified financial systems, such as Switzerland and the United Kingdom, whereas negative rates are likely to be less effective and more controversial in countries with bank-dominated and less open financial systems, such as Germany and Japan.

Until the past few years, many economists believed that interest rates below zero were not even possible. The argument was that investors would prefer to hold paper currency (which has a constant value) than bonds or deposits with a negative interest rate and thus a declining value over time. We have now learned that investors are willing to give up a modest amount of income to hold deposits and bonds because transactions with deposits and bonds are cheaper, faster, and safer than dealing in paper currency.

However, the disadvantages of paper currency are not so large as to allow for unlimited negative interest rates. The Geneva Report shows that for negative rates down to -0.75 percent, there is little evidence of any large-scale shift into paper currency. But, especially if rates are expected to stay negative for a long time, there may not be a lot of room to cut further before triggering such a shift, which would mark the true lower bound on interest rates. Overall, it appears that other policies, notably large-scale asset purchases or QE, offer greater opportunities for monetary stimulus if and when it is needed.

Note

1. I have argued elsewhere that the Bank of Japan should expand its asset purchases, particularly in equities. My colleagues Olivier Blanchard and Adam Posen have urged structural and fiscal measures, including raising government wages and minimum wages. Lower negative rates would help but are unlikely to be sufficient by themselves to get Japan to its goal of sustainable inflation of 2 percent per year.