Feud Breaks Out Between BofA And Credit Suisse Over Cause Of Libor-OIS Blow Out

One of the most vexing aspects of the blow out in the Libor-OIS spread, which from 20bps at the start of the year has blown out to 60bps and refuses to come down (despite the end of the T-Bill issuance deluge), is that nobody could agree on what exactly caused it in the first place. As we explained one month ago, among the 6 most often cited reasons was everything from the mundane - a simple supply/demand technical imbalance pulling demand due to a spike in T-Bill issuance, to Fed policy (as Citi's Matt King suggested), all the way to a funding crisis and blowing out bank CDS confirming bank liquidity is suddenly in peril:

  • an increase in short-term bond (T-bill) issuance
  • rising outflow pressures on dollar deposits in the US owing to rising short-term rates
  • repatriation to cope with US Tax Cuts and Jobs Act (TCJA) and new trade policies, and concerns on dollar liquidity outside the US
  • risk premium for uncertainty of US monetary policy
  • recently elevated credit spreads (CDS) of banks
  • demand for funds in preparation for market stress

Then, two weeks ago, a novel theory "explaining" the blow out in Libor-OIS emerged, when money-markets maven and former NY Fed analyst, Zoltan Pozsar, currently at Credit Suisse suggested an entirely new reason for the spike wider in L-OIS. 

In one of his delightfully lengthy notes, Pozsar suggested that the the cause for the move in short-term funding markets comes back to the U.S. tax overhaul, starting with its incentives for U.S. companies to repatriate money from abroad. But there’s also a provision known as the Base-Erosion and Anti-abuse Tax, or BEAT, which is spurring changes in companies’ investment behavior.

The BEAT tax makes interest paid by U.S. affiliates on money received from their foreign-domiciled headquarters no longer tax deductible in America. The result is those flows are shrinking, and being made up locally in the U.S. through unsecured borrowing such as commercial paper, Pozsar suggested.

“BEAT is forcing foreign banks to substitute FX swaps with unsecured funding and also leads to temporary overfunding on the margin,” Pozsar wrote. “BEAT explains why cross-currency bases are tighter while Libor-OIS is wider, and also introduces upside risks to the forward Libor-OIS spread."

Well, maybe not, because shortly after Pozsar suggested the theory, Citi's credit guru, Matt King summarily dismissed it, proposing instead that the changes in short-term funding rates were merely a function of shrinking central bank liquidity and reserves, which are having an outsized effect on short-term funding.

But the most direct takedown of Pozsar's theory emerged this morning, when Bank of America's rates strategist Mark Cabana, who correctly predicted the blow out in short-term funding as a result of Trump's tax reform all the way back in October, when LOIS was in the teens, released a note titled "Missing the BEAT" in which he said that while BEAT has been attributed to diverging demand for USD evidenced through short-term unsecured and cross currency basis markets, "we are skeptical BEAT has meaningfully contributed to these moves using evidence from CP, Fed data, and bank reporting."

Instead of BEAT, however, Cabana and team continue to claim that the drives for the move are ongoing uncertainty over repatriation, an elevated money market supply as well as reserve draining. At the same time the recent narrowing in cross-currency bases reflects reduced demand for USD hedging via FX forwards, and that "opportunistic" issuance from those with access to USD funding markets has contributed to narrowing in cross-currency basis swaps.

And so the first great feud in explaining the blow out in Libor-OIS was born.

While the above covers the issues in a nutshell, below we lay out a more detailed explanation from Cabana why he believes the Pozsar, and the BEAT approach, are wrong.

According to the BofA strategist, the "BEAT impact on money markets likely overstated" for the following reasons:

The base erosion anti-abuse tax (BEAT) has recently received a large amount of market focus as potentially contributing to diverging demand for USD evidenced through short-term unsecured and cross currency basis markets (Chart of the Day). While BEAT may be contributing to this divergence on the margin, we are skeptical that this is a primary driver of the recent move. Instead we continue to believe the increase in 3m LIBOR-OIS (L-OIS) is primarily related to repatriation uncertainty and elevated money market supply while movements in the cross currency basis reflect reduced demand for USD hedging via FX forwards and opportunistic increase in USD funding.

We think the 3m L-OIS spread should partially retrace, but likely slower than the market anticipates. We also expect short-dated tenors in major USD cross currency basis pairs will move to price in greater demand for dollars, but this demand will not be as acute as in recent years.

Cabana then claims that BEAT poses risk for foreign-parented companies

The BEAT provision in the tax law enacted at the end of last year was intended to reduce erosion of the US tax base. BEAT is applied to multinational companies that generate more than $500mn in US revenue where the US entity makes large payments on royalties, reinsurance, or interest on loans that are payable to related companies overseas. BEAT is akin to an "alternative minimum tax" where US entities need to pay the greater of (1) their normal tax liability inclusive of deductions and (2) BEAT liability, which adds back deductions due to inter-affiliate interest costs, a portion of net operating losses, and reinsurance payments to the parent. If the BEAT liability is larger than standard taxes due, the difference is subject to an additional tax rate of 5% in 2018, 10% in 2019, and 12.5% in years beyond 2026 (Exhibit 1; rates are 1ppt higher for banks and brokers). Lawmakers have indicated BEAT was meant to discourage companies from inappropriately channeling profits generated in the US to lower-tax regimes.

In this context, BofA believes that BEAT may be problematic for inverted and foreign-parented companies that rely heavily on intercompany loans.

This could be especially true for foreign bank US branches and subsidiaries that are partially funded in USD out of their parent. BEAT could force these US entities to re-evaluate their inter-affiliate funding. To the extent they can adjust their funding, they may wish to raise USD out of their local branches where they can still claim interest deductibility (subject to interest below 30% EBITDA). This risk is particularly pronounced for German, Japanese, and UK banks whose US branches and agencies rely on relatively large amounts of parent-related funding (Exhibit 2).

And here comes Cabana's direct slam of Pozsar, or as he calls the former NY Fed analyst, "some market participants":

Some market participants have concluded BEAT has been a meaningful contributor to the sharp increase in 3m L-OIS and major short-dated USD cross currency (XCCY) basis pairs since the start of this year. We are more skeptical. BEAT could have led to these market moves if the foreign parent reduced reliance on funding via the cross currency basis and had the US entity instead raise funds through the domestic CP/CD market. While this narrative helps make sense of relatively counterintuitive USD LIBOR and XCCY basis movements, we do not find much evidence to back it up when drawing on the following evidence:

  • Commercial paper market: Elevated levels of foreign financial commercial paper outstanding have been cited as potential evidence of a shift in funding behavior due to BEAT (Chart 1). Our understanding is that the foreign financial issuance contained in this series is from the parent and not from the local US subsidiary. A list of the largest foreign financial CP program sizes furthers this view by indicating most USD CP issuance is linked back to the offshore parent (Table 1). Further, the Fed's CP data also shows that domestically issued CP from entities with a foreign bank parent has actually declined YTD (Chart 1).

  • Fed foreign banking data: Fed H.8 data also provides limited evidence of a shift in foreign banking behavior due to BEAT. Foreign-related banking institutions in the US have reduced their usage of large time deposits (our proxy for Yankee CD issuance) and also increased the net amount of their inter-affiliate funding (Chart 2). The YTD trend in both of these series runs counter to the expected impact from BEAT.

  • Bank commentary: Our conversation with foreign bank branches in the US suggests no shift in behavior as a result of BEAT. However, these conversations also reveal elevated uncertainty around the BEAT law change and a desire for further clarification from the IRS and tax teams about the potential future applicability. As a result, BEAT appears not to have impacted behavior yet, but could do so in the future. Public commentary from bank 1Q earnings also reveal a limited expected BEAT impact:
    • Credit Suisse: "our current assessment is that it is more likely than not that we will not be subject to this tax. I'd caution, though, that this assessment is subject to the further guidance that we would expect to achieve in due course from the U.S. Treasury and from the IRS." (Earnings call)
    • Natixis: "Neither of these changes [US tax cuts and BEAT payment] were deemed likely to have a significant impact on the income gained from the reduction in the federal tax rate." (Annual report)
    • Citi: "Citi does not expect to be subject to the Base Erosion Anti-Abuse Tax (BEAT) added by Tax Reform. However, U.S. Treasury guidance regarding BEAT could affect Citi's decisions as to how to structure its non-U.S. operations, possibly in a less cost efficient manner." (Annual report)
    • JP Morgan: "Moving on to the BEAT tax. This is an area where there do remain open questions, however, at this point we do not expect to have a BEAT liability. But if we were wrong, we would not expect it to be material." (Earnings call)

Graciously, instead of slamming Pozsar, Cabana leaves the door open... just barely:

To be clear, we are not saying that BEAT has not impacted USD funding markets. It could very well be influencing USD funding conditions on the margin. However, our analysis suggests BEAT has likely not been a primary driver for the recent widening in USD 3m L-OIS. It certainly could end up shaping foreign bank US branch or subsidiary activity, but we are hard pressed to find clear evidence of this now.

So if it's not BEAT, then what is it? Here not surprisingly, Cabana goes back to the original theory he first proposed back in October.

If not BEAT, then what? The narrative surrounding BEAT provided a simple intuitive explanation to an otherwise challenging-to-explain rise in 3m USD LIBOR and decreased USD demand via short-dated XCCY basis. We continue to believe the same set of contributors we wrote about previously here have been pressuring 3m L-OIS wider, though we would likely place greater emphasis on repatriation uncertainty and reserve draining (Chart 3, Chart 4).

Incidentally, this is also the thesis proposed by Citi's Matt King (more here). The biggest surprise, however, is that even today, Cabana admits he is still not sure what is causing the move wider:

We acknowledge these factors leave much to be desired and suggest that an unaccounted-for factor has been contributing to the recent market movements. However, the data available to us suggest BEAT is not the main culprit and that the aforementioned contributors and higher bill supply are likely the main drivers.

This also means that any time you hear a macrotourist tell you that the L-OIS blow out is benign and not to worry about it, you can not only mute them, but  ignore them going forward as they have no idea what is going on, and certainly have no explanation why L-OIS refuses to tighten now that net T-Bill issuance is back to a very nominal amount.

So with the provision that BofA - unlike most of its peers - admits it is not certain what is reallycausing the blow out in LOIS, here is what it thinks happens next:

Given our list of primary divers of the widening in 3m L-OIS, we expect to see a modest retracement in funding levels over coming months, but believe the extent of this retracement will not be as rapid as market prices suggest. We believe the bill supply outlook will provide a near-term temporary reprieve to USD 3m L-OIS pressure and expect that repatriation uncertainty may soon be reduced as large offshore cash holders report earnings in the next few weeks. That said, we believe the Fed's ongoing portfolio reduction and higher Treasury supply outlook should continue to contribute to relatively tight USD funding conditions.

And now that Wall Street is reduced to engaging in virtual deathmatches over the reasons behind such cryptic concepts as blowing out funding spreads - which interest roughly a few dozen people in the entire world - at a time when everything else is centrally planned by the central banks, we now look forward to Pozsar's response...