The Dollar Funding Shortage: It Never Went Away And It's Starting To Get Worse Again

Very quietly, in the last few weeks, cross currency basis swaps (CCBS) related to the dollar have reversed their rise and started moving deeper into negative territory… again. This might not be of much interest to buyers of global equity markets at this point, but it is signalling ominous signs of growing funding stress in the financial “plumbing”.

As Bloomberg notes  “cross-currency basis swaps, which money managers and corporate treasurers outside the U.S. can use to borrow in dollars, remain close to the widest levels since January even after quarter-end, when such financing strains typically dissipate. The market was a key indicator of stress during the financial crisis, and while it’s nowhere near the alarming levels of that era, it’s still garnering the attention of analysts.”  

In simple terms, the CCBS is the cost in basis points (typically for three months) of swapping these currencies into dollars over and above prevailing interest rate differentials. In a benign environment the CCBS should trade at zero, not in negative territory. The latter implies a shortage of US dollar balance sheet (credit) offered by the global banking system. As the chart above shows, dollar liquidity became extremely tight in December 2016, especially for Yen borrowers, although it not nearly as bad as what happened in May of 2015 when we first brought attention to this little followed corner of the financial system. Despite the weakness in the dollar during much of the current year, the dollar liquidity issue never completely disappeared.

There are several reasons why, like in recent years, financing in dollars is becoming more expensive.

Among the reasons cited by strategists, are the political tensions in Spain related to Catalonia’s independence push and the slow pace of Brexit talks, which may be heightening the perception of credit risk for the region’s banks. Combine that with the prospect that a U.S. tax overhaul could trigger dollar repatriation, and the outlook for monetary-policy divergence with the Federal Reserve starting to unwind its balance sheet, and analysts see the trend only worsening.”

"This is keeping a lot of people feeling uneasy,’ said Gennadiy Goldberg, an interest-rate strategist at TD Securities in New York. ‘This now seems more of a political story, with Catalonia, U.K. Brexit negotiations and potential U.S. tax reform and repatriation. Spreads could keep widening. While Republican efforts to get a tax plan through the Senate may be off to a rocky start, any framework that spurs U.S. companies to repatriate cash could compound the scramble for dollar financing. Although that’s probably a story that will play out in the second quarter, it may already be factoring into expectations."

While we couldn’t disagree, there’s one important factor they’re missing, the US Treasury’s account (Treasury General Account) with the Federal Reserve.

By running down this account by $400bn in the first quarter of 2017 (mainly due to the debt ceiling issue), the Treasury effectively increased dollar liquidity (bank reserves) by the same amount. This not only helped to ease the dollar funding problem, but was a factor in the dollar’s weakness.

Since last month, the Treasury has rebuilt the balance in its account at the Fed from $38bn on 6 September 2017 to $170bn on 11 October 2017, for a net increase of $132bn…not insignificant. Obviously, if and when the Treasury rebuilds its account at the Fed to the previous level, dollar liquidity could become extremely tight again, especially if the Fed is tapering its balance sheet at the same time.

We have been wondering whether the Fed governors fully understand this, although some of the boys at 33 Liberty no doubt do. Credit guys also understand it “there’s another reason the strain is set to grow. The Fed is set to boost the pace of its balance-sheet roll-off each quarter, potentially putting upward pressure on U.S. rates relative to Europe and making it tougher for global investors to get dollar funding," according to Mark Cabana, head of U.S. short rates strategy at Bank of America Corp.”

Clearly the issue is attracting the attention of investors as BoA analyst, Cabana writes in a recent report, and explains that “we have received a number of client questions recently about the outlook for banking reserves both in the near and medium term due to the Fed's balance sheet unwind and potential swings in Treasury's cash balance.

In summary, Cabana expects a large reserve drain in Q2 2018 with banking reserves dropping by more than $1 trillion by the end of 2019, which “highlights the potential for funding strains to emerge around Q2 next year and uncertainties around the Fed's longer-run policy framework… This reserve drain and the Fed’s portfolio unwind should pressure funding conditions tighter through wider FRA-OIS and more negative XCCY (cross currency basis swaps) levels.”

Here are his views in more detail for the rest of this year, next year and 2019-20.

Reserves through Year End: The aggregate amount of reserves outstanding will decline only modestly between now and the end of the year, minimizing any near-term funding pressures. The $30 bn reduction from the Fed's portfolio this quarter along with near-term fluctuations in Treasury's cash balance due to the debt limit will result in only modest swings in overall reserve levels. Treasury's cash balance will need to decline ~$100 bn between now and December 8, but should rebound to ~$200 bn by year end via corporate tax receipts and ~$70 bn in bill supply during the last 3 weeks of the year.


Reserves in 2018: A more material drain of over $600 bn bank reserves during 2018 should occur due to the increased pace of Fed portfolio unwind and build in the Treasury cash balance post debt limit resolution. The Fed is projected to have $381 bn in Treasury and agency MBS roll off of their portfolio next year with the pace of reduction accelerating to $90 bn in Q2 and around $115 bn in each of Q3 & Q4 (the Fed is expected to have monthly redemptions below the cap in these quarters).

The sharpest reserve drain is likely to come from a boost in Treasury's cash balance after the debt limit resolution in March. Treasury will likely increase the cash balance to $350 - $400 bn early in Q2 through tax receipts and higher bill supply. This reserve drain and the Fed's portfolio unwind should pressure funding conditions tighter through wider FRA-OIS and more negative XCCY basis levels.

Reserves in 2019 & 2020: Large reserve drains of $400 - $500 bn are expected in each of these years. This will primarily be driven by the expected $425 and $337 bn Fed portfolio reduction as well as growth in currency in circulation, which we project to average around 5% per year ($80-90 bn / yr). We also expect slightly lower usage of the Fed's reverse repo facilities as the Fed's balance sheet shrinks due to more attractive short-term investment opportunities amidst higher Treasury supply.


We expect to see signs of reserve scarcity emerge at some point over the course of 2020 or in early 2021. While the total amount of required reserves is currently unknown we have previously estimated that required level of reserves in the system is likely between $600 bn - 1 tn. This is consistent with the NY Fed's surveys of primary dealers and market participants where the median respondent believes reserve balances will total $613bn, while the 75th and 25th percentile of responses were $1tn and $406 bn. As reserve scarcity is reached, we expect to see continued upward movement in LIBOR as well as higher rates and volumes in the fed funds / OBFR markets.”

Cabana finishes with a discussion about how bank reserves will fit into monetary policy and the potential appointment of a new Fed Chairman.

Framework question: A key question in thinking about the longer-term outlook for reserves is the monetary policy operating regime that the Fed will employ, which will be heavily influenced by the next Fed Chair. As our economists recently noted, if Chair Yellen or Governor Powell leads the Fed they would likely be in favor of maintaining a "floor" regime (relying on IOER & ON RRP). In contrast, Fed Chair contenders Warsh and Taylor would likely favor a "corridor" system that relies on a scarcity of reserves that would reduce IOER usage, increase fed funds trading activity, and require frequent open market operations to adjust reserves in order to hit the Fed's target.


It seems current staff at the Fed have a strong preference to maintain a "floor" regime. The November 2016 FOMC meeting minutes noted a number of advantages of such a system and recent Fed research has highlighted that abundant reserves can smooth interbank payments, reduce daylight overdrafts at the Fed, and lead to less discount window usage. A "floor" system would also aid in satisfying bank LCR HQLA requirements while reducing the need for frequent open market operations to smooth volatility in Treasury and financial market utility deposits at the Fed. This regime would keep fed funds below IOER and likely ensure that the ON RRP remains a key fixture of the Fed's monetary policy well into the future.

Negative cross currency basis swaps indicate that the structural tightness in dollar liquidity never disappeared despite the weaker dollar. If dollar funding markets get a lot tighter again, this won’t be good news for EM markets with offshore (Euro) dollar debt in the region of $10 trillion. Rolling over dollar debt periodically will be uncomfortable, to say the least, for some of the region’s banks.


Paul Kersey Wed, 10/18/2017 - 20:11 Permalink

The article references the billions of dollars that the Fed will shed off of its balance sheet, but, when the Fed, over a relatively short period of time, had gifted the international banking cartel (the member banks from which the FRBNY's Primary Dealers come) over $29 trillion (yes, that's trillion with a 't'), and there are over $700 trillion in derivative deals outstanding (most involving U.S. banks) out there, what is the real significance of a paltry few hundred billion dollars less in Fed reserves? Damn that was one long sentence.