Nearly three months on, the question of what the hell happened in US dollar repo markets in September — when interbank repo rates spiked to uncharacteristically high levels sparking emergency funding moves from the Fed — is no closer to being properly answered.
But there have been a few developments that are worth summarising.
First off, the BIS published this over the weekend. Diehard repo watchers will no doubt have already read finance guru Frances Coppola’s neat evaluation of the paper’s key observations, one of which comes by way of this chart:
What this shows, as Coppola notes, is the shocking degree to which the funding market has come to depend on four key banking names.
The blurb from the BIS puts it as follows:
The big four banks appear to have turned into the marginal lender, possibly as other banks do not have the scale and non-bank cash suppliers such as money market funds (MMFs) hit exposure limits (see below).
In that context, Coppola says Wednesday’s diagnosis by Randal Quarles, the vice-chair of the Fed, that banks’ own internal stress testing may have led them to hoard cash rather than lending it in the overnight repurchase — or repo — market, is probably on the right lines.
With respect to potentially self-sabotaging post-crisis regulation, he also noted this:
We understand the independence of monetary policy, but where regulatory policy is impairing or impacting monetary policy, I think we need to make sure that we’re doing the right thing with these regulations.
But Coppola also notes that the BIS believes leveraged money market arbitrage trades (another way of describing cash and carry curve trades) by hedge funds, indirectly funded by MMFs, may also have had a role to play. These MMFs, the BIS says, are currently only able to lend to high-risk counterparties because their trades need to go through banking central counterparties, who take the credit risk. It’s a view that implies the possible presence of an LTCM-style player in the market picking up pennies in front of a proverbial steamroller — a notion that’s hard to swallow for many repo-watchers. But it does beg the question if it’s a not a large hedge fund or a series of hedge funds taking risk on cash and carry trades, who else could it be?
In short, the notion that there’s an immense negative rate arbitrage out there to be exploited (f you’re brave enough isn’t entirely mad. The madness comes if players think they can continuously roll it over with no risk. For the market, the concern comes in the repercussions if they suddenly can’t.
All in all, the bigger picture is that Coppola isn’t convinced the Fed knows what it’s doing here. Its policy action seems confused and contradictory. On one hand it’s raising rates trying to wean the markets off its crisis-era stimulus, on the other hand it is intervening in the repo market to dampen rate rises. Not it, or anyone in general, seems to properly understand what’s happening in repo markets.
As she notes:
But to my mind simply replenishing the lost reserves is not good enough. We need to start asking hard questions about what the role of the repo market, and the big banks upon which it has become dependent, should really be. As the interbank market slowly dies, and reserves return to being scarce rather than abundant, the repo market is becoming the principal market through which monetary policy is transmitted. But it is poorly understood and extremely concentrated. Do we really want the transmission of monetary policy to depend entirely upon four large banks?
This is a big deal considering the imminent global move away from Libor to SOFR, a collateralised rate, which at this point looks like it will be largely determined by an even smaller cadre of mutually interested too big to fail institutions.
In that context comes Zoltan Pozsar’s latest report, which anticipates the Fed will be forced to restart quantitative easing to plug the liquidity hole at the heart of the system ahead of the crucial year-end funding period.
As well as looking at all the usual issues affecting year-end balances and funding needs, Pozsar’s report zooms in on another aspect of this whole mess that’s very close to Alphaville’s hearts. The role changes to payments systems have played in increasing liquidity pressures on the system.
From his report (our emphasis):
The payments system used to be a credit system… where banks routinely incurred daylight overdrafts – that is, negative balances – in their reserve accounts at the Fed. The Fed’s intraday credit provision to the payments system ensured that payments between banks never bounced.
Large money center banks like J.P. Morgan Chase Bank were the biggest and most active users of daylight overdrafts, and a once deep overnight (o/n) fed funds market was where large banks with negative reserve balances borrowed from small banks with positive reserve balances to get the reserves necessary to pay down their daylight overdrafts at the Fed by sunset. The payments system morphed into a “token” system under Basel III…as liquidity rules require large money center banks – which, under Basel III, we call globally systemically important banks or G-SIBs – to pre-fund their 30-day outflows, intraday liquidity needs and resolution liquidity needs.
As he goes on to note, in the post Basel III framework, banks would never voluntarily incur daylight overdrafts at the Fed, because this would breach their intraday liquidity needs or resolution liquidity needs. That means overdrafts are out the window and the system must clear with permanent reserves in circulation, as there’s no incentive to borrow temporary reserves from the Fed at all.
That puts all the pressure on the bank with the most reserves (JP Morgan), which must act as a lender of next-to-last resort for the system to balance. That indirectly turns it into a sort of Saudi Arabia-style swing producer of funding, on the basis it holds the largest spare reserves and is best positioned to balance the system. Or as Pozsar explains it:
J.P. Morgan’s dominance in repo and FX clearing meant it had to hold more excess than other banks – its portfolio of excess reserves was the “Bakken Shale” of global dollar funding markets
Anything that impacts the ability of JP Morgan to transform those excess reserves fluidly into funding, obviously risks choking the system. To keep the Saudi Arabia analogy going, imagine the impact on oil prices and oil price volatility if environmental regulation was to impose a huge penalty on Saudi Arabia every time it tapped over and above its permitted production quota or if indeed its reserves were rendered entirely unusable via international sanctions?
A final point on Fed daylight overdrafts comes by way of recently released Fed data for September, available here. The data confirms that peak overdraft use grew substantially during the period of repo distress, from about $12bn to $19bn in the month. And yet — perhaps surprisingly — it also didn’t explode. 2013, for example, saw periods with much greater overdraft usage. The pre-QE peaks of 2008, meanwhile, consistently hit above $200bn levels.
But there is one data point that is notable. It’s the percentage of peak overdrafts that are collateralised, which has been significantly decreasing of late:
The data only goes back to 2011, because that’s when the Fed overhauled its daylight overdraft framework system encouraging collateralised daily overdraft borrowing by offering a zero fee on collateralised overdrafts, while continuing to charge for uncollateralised arrangements.
It’s significant, we’d argue, that during the peak stress period some 22 per cent of overdraft borrowing came on an uncollateralised basis. These are levels that were last seen when the new framework was originated, after which point the amount of aggregate fees raised by the Fed from its daily overdraft facilities fell sharply.
How RTGS inadvertently killed system liquidity – FT Alphaville
How RTGS killed liquidity: US tri-party repo edition – FT Alphaville
A story about a liquidity regime shift – FT Alphaville
There’s a black hole in the dollar funding market – FT Alphaville
The blind Federal Reserve – Coppola Comment