February 25th, 2016 | By Nick Sargen
Over the past two years an increasing number of central banks including those in Sweden, Switzerland, Denmark, the eurozone, and most recently Japan have introduced some form of negative interest rate policy on bank reserves to counter deflationary forces and stimulate economic growth. Yet, while markets previously applauded unorthodox policies such as quantitative easing that were intended to bolster economic growth via asset price appreciation, there has been less receptivity to the adoption of negative interest rates. According to the Financial Times, “Some investors and analysts fear the moves are an alarming reflection of dwindling central banking firepower, and that the new weapon could even be dangerous.” (See “Negative Thinking”, February 17, 2016.)
One of the initial objections to the NIRP concept is that economists believed there were narrow limits for driving interest rates below zero on bank reserves, because they assumed financial institutions would rather hold physical cash than pay interest on their excess reserves. The initial experiments, however, have shown that some central banks have been able to establish lower limits for interest rates than had been foreseen originally. The Swiss and Danish central banks, for example, currently have set negative rates of 75 basis points and 65 basis point, respectively, for excess reserves with no significant signs of a shift into cash balances.
According to a report by J.P. Morgan’s economists (February 9, 2016), European central banks ability to drive interest rates deeper into negative territory without undermining bank profitability is a result of central banks introducing a tiered system of reserve charges. In these regimes, only a portion of reserves are charged the lowest (most negative) marginal policy rate, with that portion determined by what is needed to keep money market rates down at the lowest marginal rate. The rest of the reserves are exempt from charges, which limits the pressure on bank profitability and the incentive to switch into cash.
The Morgan report notes that central banks are learning by doing as they explore unchartered waters. Its main conclusion is that the lower nominal bound on the policy rate is lower than they had expected initially, and that further rate reductions are likely:
“In our view, the combination of a tiered reserve charging regime, limitations on the ability of banks to hoard physical cash, and other innovations in the money market regime could open up significantly more room, with the scope to plausibly push short-term rates down to below -1%. Signs that central banks were increasingly willing to deploy these ideas could ease some of the markets concerns about a lack of monetary policy space.”
While some economists may take comfort from this experience, the question remains how effective NIRP is in boosting growth and inflation toward central bank goals. One channel for transmitting negative interest rates on bank reserves is via their effect on bank deposit rates and lending rates. In this regard, however, the Morgan report offers less reason for optimism, as it observes that it is more difficult to drive household deposit rates below zero. One reason is that the costs of holding cash balances are lower for households than for banks. Another consideration is that there could be a political backlash to negative deposit rates on grounds that it punishes savers unduly. For these and other reasons the Morgan report concludes: “As a result, it is highly likely that the segmentation of the retail and wholesale markets would dampen the transmission of monetary policy changes as rates move further below zero.”
In that event, there are still two other channels in which the effects of NIRP can be transmitted to the economy – namely, via changes in asset prices and currency values. However, there is little basis to be confident of how such policies will actually play out.
A noteworthy example is the announcement by the Bank of Japan that it would begin charging negative interest rates on a small portion of banks’ excess reserves beginning in mid-February. While the announcement came as a surprise, it only succeeded in weakening the yen for a brief period; subsequently the currency appreciated against the U.S. dollar, as market participants revised their expectations about U.S. monetary policy. Indeed, the more central banks resort to NIRP, the less effective the impact is likely to be on currencies, as actions by several countries may cancel the impact on individual currencies.
Furthermore, one cannot rule out the possibility of adverse market moves, such as those that occurred in Europe earlier this month, when bank share prices plummeted as investors concluded continued low or negative interest rates would hurt bank profitability. A recent Morgan Stanley report (February 17, 2016) called NIRP a “dangerous experiment” with diminishing positive impact. The report contends that policy actions by Switzerland, Denmark, and Sweden, are not a good gauge for future ECB action to be considered at its upcoming meeting in March. The report argued that negative rates would likely erode bank profit margins and incent banks to shrink, rather than grow.
My bottom line is that we are nearing the point where central banks have very limited ability to bolster economic growth via further reductions in interest rates. To try to do so by driving interest rates on reserves deeper into negative territory increases the risks of distorting capital markets further, and would likely outweigh the potential benefits. Also, while negative interest rates may work in theory, it remains to be seen how effective they are in practice.
Finally, Chair Yellen was asked whether the Fed would contemplate NIRP if the U.S. economy were to weaken. She indicated the Fed’s economists were studying the issue, but that it was not under active consideration by the FOMC at this time. Her answer is very appropriate, as the U.S. economy is less at risk of deflation than either Japan or Europe, and there is little reason for the Fed to experiment with untested theories when it doesn’t have to do so.