Major Window Dressing Exercises and a Dead Federal Funds Market
Many of our readers are probably aware of the quarterly spike in reverse repos, which has previously been amply documented and discussed elsewhere. The Fed has introduced these overnight reverse repos two years ago, and has made them accessible to a wide range of counterparties (altogether 163 at last count), including banks, primary dealers, mutual funds, brokers and GSEs. In these transactions the counterparties are essentially depositing cash with the Fed overnight in exchange for treasury securities.
Photo via izismile.com
The Fed’s counterparties receive interest rather than having to pay interest (currently 25 basis points) when borrowing treasuries in these transactions. By setting the rate it pays at a higher level than the rate on short term t-bills, the Fed encourages participation. The reason for introducing the facility was that the Fed wanted to test various “exit” procedures from its extraordinary monetary accommodation.
The flow of money and securities in repo markets, from a 2013 IMF working paper by Manmohan Singh
Reverse repos will temporarily withdraw liquidity from the financial system, which will ceteris paribus tend to put upward pressure on short term market interest rates. Here is what has happened since the facility was introduced:
On December 31, overnight reverse repo transactions with the Fed spiked to a record $475 billion. The green line is the general collateral repo rate (more on this further below) – click to enlarge.
At the same time, the repos are supposed to relieve shortages of high quality collateral, which have reportedly become a problem as the Fed’s QE programs have lowered the amount of treasury bonds available for trading and swapping. Originally capped at a maximum of $300 billion, the RR facility has been expanded to a maximum of $2 trillion after the rate hike of December 16, which seemingly underscores its primary function as a tool to remove excess liquidity.
We have so far not commented on the occasional spikes in repo transactions. It seemed obvious that they were connected to some sort of window-dressing exercise, given their regular month-end and quarter-end timing, but we never gave the topic much thought (except for concluding that no effect on the money supply was to be expected).
Something slightly curious happened at the end of the fourth quarter though. Apart from overnight reverse repo transactions reaching a record high, the federal funds rate apparently nosedived from 35 basis points to just 12 basis points on December 31, way below the Fed’s target range, while the general collateral repo rate concurrently spiked to a multi-year high:
Effective FF rate and GCF repo rate – click to enlarge.
We have taken the above charts from an article by Jared Blikre, who reported on these events shortly after they happened last Thursday. According to Blikre, the especially large divergence between the two interest rates shown above on December 31 was unexpected and could be a sign of underlying stresses in the credit markets. However, we are actually not so sure about that inference.
When looking at the FF rate data published by the NY Fed, we could find no indication of an effective FF rate for December 31. Assuming that he found the FF rate indication somewhere else and that it is correct, it would still not necessarily mean much. As Lee Adler has recently pointed out, trading volumes in the federal funds market have collapsed, as all the large banks are sitting on enormous amounts of excess reserves.
Why would anyone need to borrow reserves? In fact, since 2008, federal funds trades outstanding have shrunk by nearly 90% (considering that the banks haven’t been particularly constrained by reserve requirements prior to the 2008 crisis either, this is a remarkable decline). We would guess that this dead market will simply tend to be even deader on December 31.
Underlying Problems in Credit Land?
Similarly, the surge in the tri-party repo rate may merely be indicative of a short term lack of liquidity due to year-end, but this is definitely connected to the surge in reverse repos. The big surge in reverse repos to a new record high tells us that the removal of the previous $300 billion cap has intensified the above-mentioned window-dressing activity. We don’t think the movement in FF and repo rates reported at year end is as such saying that there is a problem, but Blikre – after briefly discussing the problems of junk bond fund Third Avenue – adds a comment which we by and large agree with (and which incidentally explains the larger than usual upward pressure on the general collateral repo rate):
“It is possible that mutual funds loaded with distressed debt could use another repo market, the tri-party repo market, to offload their underperforming assets in exchange for cash. This cash could then be used to temporarily buy high quality collateral from the Fed through a reverse repo. At year-end, their balance sheets would appear to be composed of higher quality assets with a higher quality counterparty than would otherwise be the case.
Data is scant with regard to the tri-party repo market, but the size of the year-end Fed auction tells the story. A healthy financial system should not need $475 billion in high quality collateral from a zero-risk entity, such as the Fed—unless there are problems with portfolio composition and counterparty risk bubbling beneath the surface.”
(emphasis added)
This is actually a very good point. One has to wonder why such huge window dressing is held to be needed on reporting days if everything is just fine and dandy. This is especially so in light of the well-known lingering liquidity issues in the corporate bond market that have resulted from post GFC regulations such as the Dodd-Frank Act (as an aside: it is amazing that the regulations supposed to hold future banking crises at bay are named after two Congressmen who for years paved the way for ever looser mortgage lending rules governing the GSEs and did their utmost to stop all attempts to rein these institutions in. This has almost Orwellian qualities. A hat tip to the Daily Bell for pointing this out).
New capital regulations for banks may also be playing a role in this context. In other words, perhaps it isn’t just mutual funds that are responsible for the recent surge in these window dressing efforts, but commercial banks. After all, according to Basel III rules and the various national capital adequacy rules modeled after them, sovereign bonds are regarded as completely risk-free, and therefore require a capital buffer of zero. Banks thus have a very strong regulatory incentive to engage in such temporary asset swaps on reporting dates as well.
Either way though one has to conclude that this cannot be indicative of a healthy financial system. Moreover, it appears as though the central bank is actually assisting large financial institutions to engage in what is ultimately a game of deception by offering these reverse repos – which are ostensibly just a monetary policy tool. If mutual funds and banks are indeed off-loading risky assets in the tri-party repo market to unregulated shadow banking entities on reporting dates and are using the cash they receive to replace them with pristine assets borrowed from the Fed precisely on those days, regulators, shareholders, mutual fund investors or anyone else for that matter, will never be in a position to find out what is actually in their portfolios.
In short, it will be impossible to determine or estimate what risks are actually extant. Similar to Jared Blikre, all one can do is engage in guessing games. It could be that many of the assets that are swapped out are actually perfectly fine, except for being subject to inane regulatory restrictions. However, it seems at least as likely, if not more so, that they are actually quite risky – or as Blikre suggests, even distressed and therefore very risky.
A map of the US tri-party repo market
Conclusion
We wouldn’t be overly concerned over sharp moves in short-term interest rates on the final business day of the year, but the fact that reverse repo transactions on reporting days are extremely large and keep growing quite rapidly (the amount reported at year-end was 40% above the previous record high) is actually raising a few questions.
Either the financial regulations introduced since 2008 are so overly restrictive that they impede the functioning of financial markets unless they can be sidestepped to some extent, or there really is a very large amount of risk out there which is deliberately hidden from the prying eyes of other interested parties (our guess would be that it is most likely a combination of both).
The reason why this topic should be of concern is that there clearly are stresses in several corners of the financial universe (most prominently in junk bonds, commodities and various currencies) – and those are just the ones that are currently blindingly obvious. It is almost certain that quite a few as of yet undiscovered time bombs are lingering in the system as well after seven years of ZIRP.
For instance, during the 2008 crisis, people suddenly became familiar with a great many obscure financial instruments that had previously not attracted much attention. Suddenly they became focal points of the crisis as liquidity dried up (such as e.g. CDOs squared, auction rate securities, et al.). Different obscure instruments are likely to receive attention in the next crisis, but they will have one thing in common with those that rose to prominence in 2008: they will all be the kind of instruments that are great to hold during a bubble and are impossible to sell once it bursts.
Charts and tables by: Yahoo Finance, IMF, Zerohedge
