Why the Fed dislikes negative rates
We are in remarkable and unprecedented times.
Fears about the coronavirus pandemic and its effect on the world economy have led to a dramatic collapse in the returns on US Treasury bonds, with both the 10-year and 30-year bonds now yielding less than 1 per cent.
Even more unusual is that investors are now expecting short-term interest rates to be close to zero for the next couple of years, and stay well under 1 per cent for the next five years.
What markets are telling us is that the US Federal Reserve’s recent emergency 50 basis point rate cut and its decision to pump trillions of dollars into the financial system on Thursday have failed to do the trick. Further cuts in the policy rate, right down to almost zero, may also not be enough to stabilise the economy and return inflation to the 2 per cent target.
Markets are beginning to think that “lower for longer” will become “even lower for even longer”, with policy rates uncomfortably close to zero for much of the 2020s. This is not a good outcome for the yield curve or the global economy.
That raises the question of whether the Fed should go negative. The European Central Bank and the Bank of Japan have already shifted policy rates into negative territory, essentially charging institutions for parking cash with them. If the Fed joined them, it might be a sufficiently dramatic move to convince investors that the current economic shocks are temporary, prompting long-term rates to rise again. But the Fed leadership seems reluctant to do this, even under the extreme liquidity strains seen recently.
When the problem of the zero lower bound on rates first reared its head a decade ago, the major central banks initially thought rates could never go into negative territory, because this would induce a stampede out of bank deposits into notes and coin, which of course yield zero.
In the event, this shift did not occur in the expected size, and some central banks — such as Denmark’s — have been able to move policy interest rates to minus 0.75 per cent. This has now become the “effective” lower bound, though the ECB on Thursday refused to go below minus 0.5 per cent.
Former Fed chairman Ben Bernanke has suggested that negative policy rates in the US could be useful in some circumstances, but this has not seemed to persuade Mr Powell or current Federal Open Market Committee members.
Here are the three main reasons why.
First, it is not clear that the central bank is permitted under legislation to take this action, as Michael Feroli of JPMorgan and others have pointed out. The 2006 law that allows the Fed to pay interest directly to banks only says that depositors “may receive earnings” and does not contemplate the charging of fees. Some lawyers have interpreted this as ruling out negative rates. Congress could, of course, change the law, but the Fed is reluctant to ask for this, in case the ensuing political debate leads to demands for additional congressional oversight of rate decisions.
Second, the specific institutional features of the US financial system make it difficult to go negative. In particular, the existence of money market funds, which hold about $4tn in assets and are sometimes treated by depositors like bank accounts, could be a problem. Back in 2008, it caused enormous turmoil when one such fund “broke the buck” and was no longer able to protect investors’ principal. Although regulations for such funds have changed, the Fed still seems concerned that negative returns in these funds could cause panic in stressed market circumstances.
Third, and most important, the experience with negative policy rates in Japan and the eurozone does not provide conclusive evidence that such a dramatic change actually restores confidence and economic activity. Instead, there appears to be a “reversal” rate of interest. Below a certain point, really low negative rates not only do not stimulate the economy, they weigh it down.
This occurs because negative rates act like a tax on the banking system. Banks may respond by restricting credit rather than making the additional loans needed to get money out into the real economy. Although the evidence on these effects is mixed, the Fed certainly does not see a conclusive case for action.
It is possible that if markets continue to sink, the world economy stumbles badly and governments do not step up with fiscal stimulus, the Fed may have to think again about the fundamental reforms that would be necessary to make deeply negative rates possible in the US.
For now that still appears to be off the agenda. Nevertheless, global nominal interest rates have collapsed to zero, with the US seemingly close to joining the “permanently zero” club. For the markets, this will be a new world order (see box below).
Zero global rates across the yield curve
If policy rates stay close to zero for long periods, but never go negative, there would be important consequences for the likely future shape of the yield curve, the optimal mix of bonds and equities in investors’ portfolios, and the dollar. Several of these consequences have already become apparent in Japan.
While it is theoretically possible for bond yields to turn negative, even if policy rates are expected to remain in positive territory indefinitely, it seems unlikely that this could last for long. As John Maynard Keynes implied in his Treatise on Money in 1930, bond yields carry duration risk which normally should require a positive risk premium compared to cash. He thought there would be a “limiting point” on the minimum possible level for bond yields.
This remains true today, so bond yields would remain very low but positive if short rates stay close to zero. The implication is that long duration bond yields could fall no further and would no longer play any useful role in hedging equity risk in a balanced portfolio.
Another implication is that quantitative easing or other similar measures would no longer tend to reduce policy rates or bond yields, so may not send the dollar lower. In fact, in a risk-off situation where there is a flight to quality in global markets, the dollar would be likely to rise, even if the Fed is trying to ease monetary policy. This may already have started in recent days.