There isn’t a whole lot known about the Federal Reserve’s Foreign Reverse Repo accommodation, and I believe that is intentional. The rate which the Fed pays to “borrow cash” from foreign central banks and governments is unknown. What is known is just how much in total the Fed is “accommodating” foreign dollar business. This RRP, in sharp contrast to the OMO RRP for domestic use, had been surging through February in almost perfect tandem with the liquidations.

The reason for that spike is supposedly, however, liquidity. On the one hand, Wall Street banks, we are told, have been “stuck” with an unusually large inventory of treasuries as overseas central banks have been allegedly selling them. These foreign central banks have then placed those balances from the treasury sales into the RRP as a liquidity buffer.

On Monday [Feb 22], the New York Federal Reserve’s executive vice president Simon Potter said the Fed’s repo program for foreign central banks has increased because “the constraints imposed on customers’ ability to vary the size of their investments have been removed, the supply of balance sheet offered by the private sector to foreign central banks appears to have declined, and some central banks desire to maintain robust dollar liquidity buffers.”

There is a lot here that doesn’t make sense, including no reference whatsoever to collateral or collateral strain. The mainstream usually refers to the RRP, either this or the other one, as a place to “park cash” but in reality it is a last resort of collateral supply. In the first place, why would foreign central banks suddenly decide to swap holding UST’s for RRP balances? Did they suddenly fear liquidity in the UST market that was, at that time, actually surging? It’s as absurd as suggesting Wall Street was suddenly “stuck” with UST’s that they couldn’t offload. As I wrote in March, it only grates against logic and sense the further you think it about it:

These banks are supposed to have had trouble selling UST’s during a period when everyone was buying them? Even in the aftermath of the second liquidation phase, since February 11, all the benchmark yields remain subdued; meaning that there has been a continuous bid for UST paper both through the liquidation and then the partial retracement. To suggest these banks have had so much trouble offloading their “duty” of emergency support, to the point of being “forced” into worse and worse funding and hedging prices, does not just strain credibility it is patently absurd.

It forces you to circle back to the question of what central banks were doing in the first place, leaving collateral as the likely difference. There is a fundamental distinction between wholesale mechanics and traditional understanding, and nowhere is that more a barrier than in UST’s as collateral (for more than just repo, as derivative positions are not just collateralized but also marked, most often, daily). In other words, were central banks actually “selling UST’s” to Wall Street banks, or were the UST’s being recalled by Wall Street in collateral calls?

This would make far more sense bringing in the radical spike in the RRP; central banks “lost” UST collateral to primary dealers who were hoarding them (rehypothecation chains decaying) for their own difficulties or had funded past wholesale positions for these same central bank’s activities during past illiquidity (the ticking clock striking zero) now requiring more posting. Parsing Simon Potter, the “supply of private balance sheet offered by the private sector to foreign central banks” was collapsing at that moment, a fact that is established by the very global and intense liquidations concurrent to all of this. It would make sense, then, that foreign central banks were not using the RRP as a “liquidity buffer” but rather as an alternate means to collateral availability that either they could not get from Wall Street (hoarding) or needed to get to pay into Wall Street (collateral calls).

If we use the price of especially emerging market currencies as a rough estimate for this trend, the “rising dollar” against them means a more expensive funding environment for these central banks (again, unnamed) to try to work around local “dollar” shortages; thus, the likely collateral flow was against them and very likely intensely so.

ABOOK May 2016 Collateral Foreign RRP ABOOK May 2016 Collateral Foreign RRP Recent

It is unsurprising, then, that the RRP balance peaked the week of February 17 – the very week after the February 11 end of the “dollar” run. The coincidence in timing and intensity again directs our attention to collateral and thus “dollar” shortage both here and overseas. It is a nod to obvious systemic illiquidity that was at that time most obvious. The relevance now is the nod to less obvious illiquidity since sentiment has shifted so far in favor of a much brighter future. However, as we see in so many other places, while the RRP balance has not pushed higher it also hasn’t dropped, either.

ABOOK May 2016 Collateral Repo Fails ABOOK May 2016 Collateral Repo Fails Recent ABOOK May 2016 Collateral Dealer Hoarding

From that one indication alone, it would be entirely reasonable to suspect that collateral shortage and lack of free flow continues to be the dominant condition or setting. Sure enough, we find repo fails still elevated and lumpy even after the shocking surge in mid-March (which was actually held over for a second straight week to show just how much collateral has been misaligned even though sentiment and risk have gone in a very different direction). Not long after that huge repo warning, actually during the second week of the dreadful surge in fails, we found a great deal of trouble in Japan, which also isn’t surprising given that Japan seems to have fallen into the primary cohort of foreign sources “selling UST’s.”

ABOOK Mar 2016 TIC Japan

Further, given the continued elevation in the foreign RRP it is predictable to also find that US primary dealers continue to hoard UST collateral. In what is also not coincidence, the peak net positive balance for dealer holdings of coupons was the week of February 10. Those coupon holdings have been pared back only somewhat since then, rising again a bit recently, which we are supposed to believe is due to dealers having continued trouble selling out that inventory into a still robust appetite for UST paper? Not a chance; hoarding continues because systemic illiquidity and collateral problems do. All of these related data points agree on that count.

As with the huge spike in repo fails back in March and the continued irregularity since, it is only consistent with the idea that the “dollar” run and liquidations may have ended but systemic illiquidity did not (subscription required) and is once more only lurking beneath the surface.

ABOOK May 2016 Eurodollar Curve