‘Mystical Assertions’ And ‘Many Pretty Pictures’: Is The QT Fascination Misguided?

Market participants’ obsession with Fed balance sheet runoff kicked into high gear late last year and was, by many accounts, one the most pernicious aggravating factors when it comes to explaining the Q4 swoon across risk assets.

That’s amusing for at least two reasons.

First, for all the derisive commentary about how runoff ended up not being akin to “watching paint dry”, it actually was “like watching paint dry” right up until everybody decided to use runoff as an excuse to sell everything. So, was Q4 2018 really the “tipping point” beyond which whatever the mechanical effect of QT on risk assets actually is ended up being “too much”? Or is it more likely that people got spooked for a variety of reasons, dumped risk assets and then mistook the market implications of their own panic selling as “evidence” that runoff finally “caused” a problem?

Second, it is unquestionably the case that the balance sheet issue was amplified by Donald Trump’s “50 Bs” tweet. That, in turn, means that to the extent anybody did sell in Q4 based on concerns about the purportedly pernicious effects of runoff, market participants were implicitly trusting Trump’s assessment of the mechanical impact of QT on stocks, despite the fact that no analyst has been able to come up with a model of that impact robust enough to serve as the gold standard. Hopefully, I don’t need to explain why that is ridiculous, but just in case, it means that folks were making asset allocation decisions based on an assessment of the effects of balance sheet runoff delivered by a man who thinks you spell the word “hamburgers” with two e‘s and a d where the g is.

None of that is to say there’s no mechanical impact of balance sheet runoff on risk assets (there is – see here and here) and it’s not necessarily to say that whatever that impact is didn’t contribute to tipping things over late last year. It’s just to say that the cessation of QE and then the eventual commencement of runoff had been exerting a tightening impulse for years (you can see it in the shadow rate – see here and here) and I have yet to hear a wholly convincing rationale for why Q4 2018 should have been the quarter when runoff finally caused a rout.

All of that said, it ultimately didn’t matter because once enough people believed this was a problem, it became a problem (that’s the self-fulling nature of things as falling stocks tighten financial conditions and wider credit spreads start to manifest themselves in real economic outcomes). And so, we got the January relent. Now, the Fed is widely expected to announce – either this week or at the next meeting – when the official end date for runoff will be.

The above provides the context for a March 13 note penned by Credit Suisse’s James Sweeney and the incomparable Zoltan Pozsar who mercifully decided on a light touch and simply referred anybody wanting to spend 48 hours mired in the plumbing of funding markets to his other work.

As noted here on Sunday, Credit Suisse has a slightly more -hawkish -than -the -market call on the Fed’s rate path which they attribute to their take on inflation, the labor market, and global growth. But the March 13 note focuses on the contention that the Fed’s plans for the balance sheet have more to do with technical considerations tied to funding markets and much less to do with macroeconomics.

It’s not feasible to do the whole piece justice here, so rather than try, we’ll just capture the gist of it by way of what we think are the best soundbites, some of which betray a palpably derisive tone towards the way the balance sheet debate is often framed, where “framed” can be taken both figuratively (i.e., the discussion) and literally (i.e., with charts).

“Investors’ views on balance sheet policy are all over the place, including some almost mystical assertions that QE apparently works but ‘nobody really knows how’ [and] simple charts of balance sheet size versus the equity market or exchange rates are offered as evidence of QE’s power”, Sweeney writes, adding that “investors are widely haunted by the possibility that asset prices are ‘only here because of the Fed’s large balance sheet.'”

He goes on to note that despite everyone’s insistence on the narrative, actually quantifying or otherwise analyzing the relationship between balance sheet expansion and, for instance, credit growth, monetary aggregates, asset prices, and inflation, is almost exclusively the purview of academic papers. As anyone who has ever spent any time in academia knows, the only thing that gets fewer reads than obscure sellside notes are peer-reviewed, academic journal articles, which in many cases are only read by the author(s) and the “peers” who “reviewed” them.

Long story short, Sweeney and Pozsar say that what’s important are the effects on duration supply and the level of reserve balances held by banks, not the overall size of the central bank’s asset pile.

After noting the growth in the size of the Fed’s balance sheet post-crisis, Credit Suisse sarcastically writes that “many pretty pictures have been drawn of this balance sheet trajectory versus a range of variables.”

Those “pretty pictures”, the bank reckons, aren’t much use, or at least not compared to other “pictures” which show “the level of reserve balances held by banks and the level of duration risk removed by central banks when they purchase long term assets.”

On the duration point, Sweeney reminds you that it’s not just the Fed’s purchase of long duration bonds that matters. There’s also the demand side of the equation and in addition to shifting demand, the supply side is also influenced by the actions of the Fed’s global counterparts.

The following charts (duration supply in 10-year USD swap equivs) show that, quote, “duration holdings by the Fed peaked in 2014; duration holdings by the BoJ and ECB exploded in 2016 and are larger than the Fed’s; duration supply is influenced more by growing US deficits than by falling Fed holdings.”

(Credit Suisse)

The bank then goes on to neatly summarize the takeaway, which is basically just that the Fed’s duration holdings have suppressed long-term yields mechanically (i.e., distorting the supply-demand picture) and also via making forward guidance more credible (i.e., if you’re willing to deploy the balance sheet to that extent, you must be serious). But, importantly, Sweeney notes that “quantifying the basis point impact of these things precisely is tricky, because the magnitude of changes due to US Treasury behavior (especially deficit size) and foreign central bank purchases of duration risk dwarf the plausible impact of the Fed.”

After delving into the usual discussion of the impact of ECB and BoJ asset purchases on the US term premium, Sweeney ventures off the fairway and into the rough with Zoltan. Without getting too far into those weeds, the point (and this speaks to the idea that the Fed’s decision to end runoff is motivated more by technical concerns than macroeconomic factors) is summed up in the following relatively brief excerpts:

Zoltan Pozsar’s Global Money Notes publications have chronicled the details of the new system and how central banks’ balance sheet policies and Basel III are changing the way global funding markets trade. In the new system, a complex web of regulations affect different financial institutions differently across jurisdictions and through time, thereby creating a demand for reserves that is less linked with credit demand than it used to be. Libor-OIS moves, spikes in repo rates, and spikes in the dollar FX basis – which affect the cost of hedging dollar assets from abroad – have been closely related to changes in the level of reserve balances, and thus the level of reserve balances matters regardless of what is happening to the aggregate size of the balance sheet.

[…]

Dislocations visible in funding markets now are serious enough to force the Fed to consider ending balance sheet decline soon. However, it is important not to exaggerate the importance of these channels to macroeconomic performance. In the grand scheme of things, these are technical problems, not recession or deflation threats.

With that out of the way, Sweeney goes on to assert that going forward, the “ordinary stuff of monetary policy” (i.e., rates and forward guidance) are what matters for macroeconomic outcomes.

Obviously, forward guidance has taken on a much more critical role post-crisis and while Credit Suisse doubts whether a neutral rate estimate based solely on history plus the Fed’s guidance can ultimately prove accurate by anything other than sheer luck due to the unforeseeable impact of “budget, productivity, demographic, geopolitical, and technological developments that are likely to shape [the] investment horizon”, the bank does note that “the evidence is that the Fed’s guidance has indeed anchored the market’s expectations, and this has kept long term rates lower than in the past.”

The bottom line, then, is that when it comes to what actually matters in terms of policy tools/variables, the level of short-term rates combined with forward guidance is what counts.

As far as the balance sheet goes, it’s not the aggregate size that matters, but duration holdings and the level of reserves, with the latter mattering more “on the margins where professional fixed income investors operate” and feel the effects of dislocations.

Ultimately, Sweeney has the following to say about how folks will act with regard to the announcement of an end date for balance sheet runoff and if you sense a bit of disdain in the following excerpt, you won’t be alone:

Market participants will likely respond to the end of balance sheet decline as though this is an easing of policy, and greet it as a “reflationary” event. The standard charts of the S&P versus the balance sheet and others will be widely circulated. We would expect that in the short run some of these relationships will appear to continue to hold, but after a while these relationships will return to ambiguity in the conclusions of serious analysts. (Those are the ones who, for example, look at long term charts of the balance sheet and the dollar or the S&P and find no relationship!).