An exercise in connecting the dots: the Libor-OIS spread

The Libor-OIS spread is such a technical topic that I will leave the details to professionals with a much more profound understanding of it (plus Shawn already wrote a very good post about it). I am just trying to investigate the subject from a different angle.

I have been thinking about a few things some very smart investors shared with me and I am finding connections between the following:
·     A tactical long view on USD and the fact that investors will, during the next downturn, sell the most liquid asset they own first (although I personally believe a proper USD bounce should happen only after a credit event of some sort).
·     The recent Libor-OIS spread widening, which got a lot of attention. I am not convinced by either bearish and bullish views.

·     The bullish case concentrates on the fact that most of the widening is coming from T-bill increase in issuance, that should decrease in the near term. There is no clear correlation between the recent increase in T-bill issuance and an increase in CP yields (I understand the RV argument, but that cannot explain the full move alone).
·     The bearish case concentrates on the fact that the wider spread would signify a tightening of funding conditions and, potentially, a Dollar shortage. If this was the case, I guess the xccy basis for Eur and Jpy should widen too (this has happened in the last few days, but almost as a reaction to the Libor-OIS move).

I think the pressure is mostly coming from tension in the commercial paper/short duration credit space (as T-bills widened earlier in 2017 and they recently outperformed vs Eurodollars and Libor, not showing much of a sign of overwhelming issuance). This is potentially confirmed by the fact that low beta, short dated paper, have been under-performing the credit market in beta-adjusted terms. What is interesting is that the same part of the credit curve is under pressure also in the European market.

CP yields rising relative to Fed Funds (%). Source: Citigroup

1-5y US HG corporate bond spreads vs CP rates. Source: JP Morgan

A recent JP Morgan research piece pointed out that the tension might not be coming out of repatriation flows only, as that would have called for a flattening of the credit curve, which, during the last leg of the move in the Libor-OIS spread, did not happen.

Change in spreads on the JPM JULI Domestic index of US high grade corporate bonds by maturity bucket in bps. Source: JP Morgan

I think these developments should be followed very closely. In fact, at least for the moment, I see banks` excess reserves remaining ample, which should support the FX swap market. This could change if an increase in T-bills issuance starts weighing indirectly on banks` excess reserves.

I wonder if we might be closer to a downturn than the market expects, with market participants starting to sell liquid assets, but in a slightly different fashion, as we are still in a very low yield environment. They might have started liquidating some very expensive government bonds between end of 2017 and beginning of 2018, but, after a certain level of yields was reached, moved into selling short-dated corporate paper. If this trend in low beta corporate paper selling should continue (which is a big if), together with a continuation of the increase in T-bills yields (driven also by an increase in issuance and higher Fed rates), it could create pressure at funding level. This would potentially translate (at a certain point) to a shortage in Dollar funding, which could theoretically be supportive for USD itself (like in 2008, with the USD bouncing only after Bear Stearns at the beginning of the year), which would be in line with the view of a potential tactical bounce.
I think one of the scenarios the market is not pricing is that the Fed might have already tightened (or it is very close to) financial conditions too much, but other elements may have hidden this, especially in 2017 (i.e. Chinese liquidity injections). Something like what I tried to explain above might be one of the canaries in the coal mine. I understand it is a bit of a far-fetched hypothesis, but I would appreciate any thoughts or feedback, especially those that go contrary to my view.

(Disclaimer: I took some liberty in using some of the analysis done by Citigroup (Matt King). You can find most of its piece on Zerohedge in case you have no access to Citigroup research (https://www.zerohedge.com/news/2018-03-20/why-citi-suddenly-freaking-out-about-exploding-libor-ois))