Via the excellent Damien Boey at Credit Suisse:
Over the past few days, we have seen chaos in US money markets. Repo rates on general collateral (ie secured borrowing rates) have spiked higher to as high as 10%. The New York Fed has been forced to intervene, conducting a $53 billion overnight “system” repo to inject liquidity into the system. However, the intervention attempt encountered “technical” problems, possibly related to inexperience on the New York Fed’s trading desk. And in any case, once the repo operation had been conducted, rates remained uncomfortably high, suggesting that the intervention was not enough.
It has not just been repo rates that have behaved in a disorderly fashion. We have also see the Fed’s interest rates (the Fed funds rate, and interest on excess reserves) drift towards the top end of their target ranges, and even break out to the upside. Reserves in the system have seemingly become quite scarce in a short space of time.
Why does all of this matter? Because if the Fed cuts rates by 25bps, but money market spreads rise by even more, the Fed’s easing will be perceived as ineffective. There is a transmission problem to deal with. And until it is resolved, the natural tendency is for investors to lower their estimate of the neutral rate, as the Fed must work harder to deliver effective stimulus and relief to the economy. All of this supports bonds and curve flattening. Within equities, it is good for quality, good for defensive momentum, and bad for value. All of this said, we think that the markets are experiencing a brief hiccup that will be dealt with in the near-term. We remain of the view that value rotation will continue in earnest.
What have been the catalysts for the drying up of liquidity?
- Large scale issuance from the US Treasury, draining reserves from the interbank system (as banks use reserves to buy bonds in open market operations).
- The September corporate tax payment date, pulling out bank deposits and money market fund cash balances.
The longer term issue is that the definition of “excess reserves” has changed in a Basel III environment. Previously excess reserves were defined by the Fed’s standard. If banks held more in the Federal Reserve accounts than they needed to settle transactions with one another, they had excess reserves. Now, under Basel III, excess reserves are defined by a global standard. Banks not only need enough reserves to settle accounts with one another at the end of each day – they also need to hold enough reserves for liquidity and capital buffer purposes. Indeed, reserves are better “high quality liquid assets” (HQLA) than even US Treasuries. In this context, at the beginning of 2017 (when Basel III really kicked in), banks found themselves with excess reserves by Fed standards, but deficient reserves by Basel III standards. Making matters worse for a little while were the Fed’s balance sheet reduction efforts, draining reserves from the system.
There have been other unintended consequences from the Fed’s re-plumbing of the system. In theory, the Fed needed to completely drain excess reserves from the system, before choosing to raise rates in 2015, because mechanically, excess reserves drive the Fed funds rate down to zero. Instead, the Fed chose to recreate the system by:
- Segmenting the market for excess reserves, using a new instrument of “interest on excess reserves”. Previously, banks got no interest on excess reserves. But once the Fed started to award interest, these “excess reserves” were effectively segmented into a new market.
- Introducing money market funds as active players into the market for reserves. As a result of interest on excess reserves, banks are no longer as significant at the margins in price discovery in the Fed funds market. Instead, government agencies and money market funds have become the dominant, marginal price setters. The Fed enabled this by creating a new reverse repo facility to engage with money funds. Effectively money funds have been enabled to park their proceeds directly in a Fed (rather than commercial bank) account, taking bonds as collateral.
The trouble has been though that these plumbing changes have interacted with the regime shift into Basel III. Banks have been incentivized to hoard reserves as HQLA – but the Fed has been actively trying to take these away through its balance sheet reduction efforts. Banks have responded by offering lower deposit rates. But in turn, this has caused savers to park their money with higher yielding money funds, who earn higher interest via the Fed’s reverse repo facility. The system has experienced a decline in deposits, but an increase in reverse repo. This has been as good as a reduction in money supply, because money parked in the reverse repo facility rarely sees the light of day again.
We mention all of this, because the plumbing of US money markets has become incredibly complicated. Experience matters in dealing with the complexities. But experience is no longer there at the New York Fed, because the overseers of quantitative easing have long gone. Therefore, it is unsurprising to see that a few technical glitches have emerged in recent intervention efforts.
Also, we note that the complexity of the system makes it very difficult for the Fed to quickly change the rules. It has just moved from “Fed funds system 1” to “Fed funds system 2” (interest on excess reserves, reverse repo facilities). To try to re-write the rules of the system in short order to fix current plumbing problems is an extremely difficult ask. It is akin to creating “Fed funds system 3”, on the run.
All of this said, there are short-term, and long-term solutions to the problems at hand:
- More overnight repos, with the New York Fed pre-committing to supplying liquidity on particularly heavy days. This course of action would at least give money market participants and borrowers some confidence in market functioning.
- The standing repo facility. The Fed needs to open up liquidity provision to calm markets in the new regulatory regime. But it will need time to decide whom it will allow access to, and on what terms.
- Quantitative easing. This is the bazooka option, as it involves a near-permanent exchange of reserves for bonds with the system. Resumption of quantitative easing is quite possible longer term. But in the short term, most Fed watchers do not believe that the officials are willing to embrace this course of action.
Note that small tweaks to existing interest rates are unlikely to have much an effect, as effective borrowing rates have already gone wild. Actual use of the Fed’s balance sheet is very much needed. The good news is that the Fed is willing and able to act. The only question is how long the Fed needs to deliver a comprehensive, rather than stop gap fix.
We expect big things from the upcoming Fed meeting:
- The market has been calling for more rate cuts, as the expedient solution to the offshore USD black hole. In other words, the foreign reverse repo facility is sucking in too many USD deposits from abroad, because it is offering terms that are far too attractive in an inverted yield curve environment. The Fed needs to either re-think and re-design the facility – an unlikely option in the near term – or drop the Fed funds rate materially to make it less attractive.
- The Fed also now needs to address the blockages in US funding markets through balance sheet adjustments.
Failure to deliver on both of these fronts might cause more panic, and a broader flight to quality. The saving grace is that if bond yields fall, effective mortgage rates will come down anyway, serving as an automatic stabilizer. And eventually, the plumbing problems will be fixed. Indeed, the solutions are being pitched by the street to the Fed, thick and fast. Therefore, while we appreciate the gravity of recent disruptions, we are not overly worried. We still favour curve steepening and value investing.