“The Fed is resuming QE (‘QE-Lite’ with MBS reinvestments into USTs) later this year, back in the bond buying business”, Nomura’s Charlie McElligott writes on Monday morning.
As we detailed on Sunday courtesy of Goldman’s quick and easy calculations, the Fed’s decision to reinvest principal repayments from agency and MBS securities starting in October “subject to a maximum of $20 billion a month” translates into the following “flow” effect:
With an estimated $17bn a month in MBS runoff and the average duration of USTs outstanding at ~54 months (approximately that of a 5y note), this roughly equates to the Fed taking ~8-10bn in 10y equivalents out of the market; that’s compared to $30, 60, and 40bn 10s of monthly purchases under QE1, QE2 and QE3 respectively.
Although that flow is relatively modest, it is what it is – ‘QE-Lite’, as McElligott calls it.
That said, Charlie again reminds you that “QT OF MBS REMAINS A VERY-REAL ‘THING,’ as the Fed will continue to whittle-down their mortgage holdings eventually to zero, with lumpier runoffs having-shown to support cross-asset volatility.” That’s a reference to several analysts’ work which has shown that MBS runoff does appear to mechanically push up volatility and dent risk assets.
Read more
Ok, Sure! I’ll Take A Look At That Fed MBS Runoff Vs. S&P Returns Note…
As WSJ Tips ‘Earlier-Than-Expected’ End To QT, One Bank Assesses Impact On VIX, Rates Vol.
Here’s a quote in that regard from Nomura’s George Goncalves:
As we’ve said before, QT tightening on FCI is part science, part art; however we can see that MBS dispersion from the Fed B/S has a good correlation with vol. This is important because we think the Fed will continue to let their MBS book shrink and return MBS back into the private sector (which should hedge their MBS). In theory the Fed was a seller of vol (by holding MBS post-GFC on an unhedged basis). Since the MBS rolloffs will continue until further notice, the Fed in practice is going to let vol (from private hedgers) leak back in.
That’s the nuance behind the “Fed is short vol.” meme, which is still misunderstood by the layperson (read our explainer here).
For McElligott, Friday was more than a simple “growth scare” risk-off move. In his Monday note, Charlie describes the extent to which it “set-off widespread portfolio / strategy rebalancing.”
“This ‘acceleration towards the end-game’ drove a powerful mean-reversion in Cross-Asset Risk Premia, as many crowded ‘YTD winners’ saw large blasts of monetization / ‘size-down,’ while ‘losers’ were much more nuanced”, he writes, before laying things out via the following set of handy bullet points which together show the “thematic- / sector- / factor- performance-dispersion.”
- US Equities “Beta†factor -4.4%, the largest one-day drawdown since 6/27/16 (post-Brexit vote)
- US Eq “Volatility†factor -4.1%, the largest one-day drawdown since 10/24/18 and second-largest since Brexit
- US Eq “PEG†factor (Growth) -2.1%, the third-largest one-day drawdown since 5/3/12, ‘bottom 5’ worst day since 2010
- US Eq “Default Risk†factor -2.0%, the largest one-day drawdown since 1/22/19 and fourth-largest since 5/24/17
- US Eq “Commodities†factor -2.1%, worst day of 2019
- US Eq “Predicted LTG†factor (Growth) -1.7%, worst day of 2019
- US Eq “Analyst Coverage†factor -2.0%, 2nd worst day of 2019
- US Eq “Cash / Assets†factor (de facto “Growth vs Valueâ€)  -1.8%, the third-worst day of 2019
- US Eq “Volatility Adjusted Momentum†factor +1.6%, best day of 2019
- US Eq “Dividend Payout†factor +1.7%, 2nd best day of 2019
- US Eq “ROIC†factor (Quality) +0.9%, 2nd best day of 2019
- US Eq “ROA†factor (Quality) +1.0%, 2nd best day of 2019
- US Eq “Debt / Equity†factor four-day return (+2.1%) the largest since the four-session period ending 3/7/16
After documenting the “dumpster fire” in financials, Charlie warns (again) that the S&P is still dancing around levels that could ultimately see the dreaded short gamma “flip”, a critical point considering the extent to which long gamma positioning has helped keep volatility suppressed. A move to the “dark side” (so to speak) would mean directional moves could be exacerbated/accelerated/reinforced.
(Nomura)
As far as the CTA trigger levels getting pulled mechanically higher (and this is part of the “March surprise” thesis that we’ve documented exhaustively here), McElligott expects the model to flip short for the Russell on Monday. Here’s an update on the trigger levels for other indices:
- S&P 500, currently 100% long, selling under 2777.31 to get to 50%, more selling under 2636.65 to get to -100% , flip to short under 2636.93, max short under 2636.65
- NASDAQ 100, currently 100% long, selling under 7048.5 to get to 50%, more selling under 6677.31 to get to -100% , flip to short under 6678.05, max short under 6677.31
- Euro Stoxx 50, currently 100% long, selling under 3175.37 to get to 50%, more selling under 3026.85 to get to -100% , flip to short under 3027.17, max short under 3026.85
- Hang Seng CH, currently -50% short, more selling under 11311.46 to get to -75%, max short under 11310.31, buying over 11712.02 to get to 100% , more buying over 11710.87 to get to 25% , flip to long over 11710.87, max long over 11710.87
Â
