Recently there's been an interesting back and forth between David Beckworth of Macro and other Market Musings and Cardiff Garcia on the interaction of monetary policy, interest on excess reserves (IOER), quantitative easing, and money market funds, as well as the implications for policy. Garcia puts forth the argument that paying interest on excess reserves is a key mechanism keeping yields on treasuries above zero, which maintains the solvency of money market funds. As a result, cutting interest on excess reserves would be dangerous because it would effectively disable a $2.7 trillion industry that plays a critical role in fodern financial intermediation. This would lead to zero rates on deposits, withdrawal of cash from banks, and massive money hoarding. On the other side, David Beckworth argues that the regime change embodied by a reduction in IOER along with a committment to NGDPLT would spur safe asset creation and allow money market funds to stay afloat in spite of zero IOER. So, yes, while an incremental decrease in IOER would have a negative effect, if only it is combined with a large enough policy change it will be expansionary.
I have previously written on the relationship between money market funds and conventional monetary policy tools such as quantitative easing. The core of the argument is that, in this world of nonlinear expectations, policy results become non-linear and non-monotonic as well.
A major problem with this scenario is that the market response function becomes highly nonlinear. With no policy action the market contracts. With some policy action the market contracts further. Only with outsize policy action are private agents convinced of a stable future NGDP target, and do collateral chains continue to expand. In this situation, it would be hard for a central banker to tell how many asset purchases "would be enough". Consequently, a credible NGDP target becomes even more important. First, it facilitates private safe asset creation. Second, it assures the market that the Fed won't end up in the middle where collateral chains contract and monetary expansion fails to raise output or prices. This is a critical argument in favor of NGDP targeting in an increasingly complex world. In spite of all of the complications in modern finance and banking, stable nominal expectations can help smooth those problems over and maintain the processes of safe asset creation.
So can lowering IOER improve MMF solvency? Only if, as I have written above, it can translate into higher trend nominal GDP growth. In other words, reducing IOER translates into concrete committments about the future money supply, ala Woodford. Beckworth does point out such a channel:
Here is my take. The FT Alphaville story fails because it ignores the broader effect of the Fed lowering the IOER. Such an announcement, if credible, would send a loud signal to markets of more monetary stimulus. And if done right, this signal would have a huge impact because lowering the IOER is tantamount to saying the Fed is going to permanently increase the monetary base. A permanent increase in the monetary base implies a permanently higher price level and permanently higher NGDP level down the road. In other words, lowering the IOER would permanently raise expectations of future nominal spending and income. As a result, demand for financial intermediation services would increase today as firms, households, and governments planned for the higher level of NGDP. The increased demand for credit would raise financial firms' net interest margins and more privately-produced safe assets would appear. No doomsday for MMF and a recovery ensues.
But does it? What I think is missing from the discussion is the importance of zero. When IOER approaches zero while deposit rates are low, there is a non-zero probability for the FT Alphaville scenario to play out, and people start taking out the money as currency and hoarding it. While this adds to the monetary base, it cripples the money supply. Beckworth argues that "demand for financial intermediation services would increase today as firms, households, and governments planned for the higher level of NGDP" but I find that doubtful. Financial markets move quite fast. So if many money market funds suddenly broke the buck and there was another run on the banks, the effect would be severe and immediate. In the words of Garcia.
David assumes that the the signaling effect would be enough to raise companies' demand for loans, and that this would raise banks' net interest margins (meaning that banks wouldn't have to play the MMF funding/excess reserves arbitrage). Even if he is right, this would surely take time, whereas an MMF breaking the buck or a run on MMFs by nervous investors can happen extremely quickly. See: 2008, September.
On the other hand, the real economy inherently contains lags and would respond second. For as much as money market funds are risky ventures, they do help a lot of businesses with payroll and inventory financing. As a result, such a hit to the financial sector would undoubtedly reduce the expectation that nominal GDP was on path to being restored. At this point, if the expectation is uncertain, IOER would not function as an effective regime shift. This is what differentiates lowering IOER from FDR's bold move off of the gold standard. While lowering IOER can have an ambiguous effect, there was no such ambiguity for the gold standard. Inflation expectations were going to go up, real interest rates went down, and investment increased. This, of course, ignores the fact that there was nowhere near a well developed system of shadow banks and money market funds in that era.
Even if IOER was an effective regime shift, the credibility of the new regime is still an issue. If IOER is going to release so much currency, and if that currency really was going to be spent instead of just hoarded, then there could be a massive burst in nominal GDP. If the public does not see such a massive boost in nominal GDP as something the central bank could credibly maintain, then how would the nominal GDP target be credible?
So are there any alternatives? I think the role in IOER in maintaining solvent balance sheets should not be underestimated. It is keeping money market funds afloat, and as a consequence much of the modern financial intermediation sector. Without these banks, much of the buying and selling of treasuries would be hurt, and monetary transmission would be reduced. However, we do need more expansionary monetary policy and must chart a path forward. One proposal is to engage in quantitative easing until a certain dual threshold, such as 7% unemployment or 3% inflation, is reached. Alternatively, the Fed could committ to forward guidance on interest rates while purchasing MBS or more long term treasuries to show its committment to low interest rates in spite of better economic conditions. This would accomplish what Beckworth calls for in committing to a larger future monetary base without jeopardizing the financial stability of money market funds. These would all increase expectations of nominal GDP as a result. But no matter the final option, we cannot neglect the role of IOER in maintaining the financial sector, the key mechanism of monetary transmission, and the possibility of effective monetary policy.