Revisiting What ‘Matters’: Signal Versus Noise – Seeking Alpha
On Friday in “Au Contraire, Heisenberg,” I took a look at the counterargument to my contention that European investment grade credit is asleep at the proverbial wheel when it comes to pricing political risk.
That post, while seemingly esoteric (I imagine the decent viewership metrics had more to do with the title than the subject of the post), was in fact quite important.
You’ll recall what I always say about what I choose to highlight for readers: I endeavor to never waste your time. And it’s funny, because there’s a kind of paradox involved with my selection of discussion topics. To a certain extent, I have to exercise a bit of what we might call “benign paternalism.”
There’s a whole universe of content out there and there’s only so much time in a day. I want to make sure that if you choose to spend some of your valuable time reading what I write, that I make it worth your while. But the problem is, sometimes that entails writing about things that you may think are a waste of time – “useless noise”, as one reader recently put it.
But I know that what I’m giving you is anything but “noise.” In fact, all you get with Heisenberg is signal. There’s zero noise. I don’t write to maximize page views. I write to maximize the extent to which I can add to readers’ understanding of markets and what actually drives asset prices.
You may not realize it’s important – this is where the benign paternalism comes in – but you’ll be a better investor for reading it even if you don’t immediately understand why. Do you know what won’t make you a better investor? Stuff like “What Would Warren Buffett Buy Now?” or “Removing The iPhone Home Button: What It Means For The Stock” (note: I made those titles up, so as the old movie disclaimer goes, “any resemblance to actual persons, living or dead, or actual [articles] is purely coincidental”).
The piece linked above (“Au Contraire, Heisenberg”) is no exception. The thrust of that post was to take a closer, more granular look at the extent to which European high grade credit is pricing in French political risk. Why should you care? Well, it’s simple. First, French corporate debt comprises nearly a quarter of the entire universe of â¬ investment grade debt (according to BofAML). Thus, if French voters were to elect Marine Le Pen in May and Le Pen were to follow through on her promise to take her country out of the EMU, a good portion of the European corporate bond market would be thrown into turmoil.
Let me simplify further. Let’s say you own a US IG ETF like iShares iBoxx $ Investment Grade Corporate Bond ETF (NYSEARCA:LQD). What do you imagine would happen to that ETF in the event a large portion of the European equivalent was suddenly converted from euros to a “new” French franc? Don’t know? Me neither. But that’s the point. You can bet that if suddenly, the fate of 23% of the entire European investment grade bond market is subject to considerable uncertainty, there will be knock-on effects for you as an investor in US credit. To be sure, the knock-on effect might be positive. That is, suddenly, everyone could flee â¬ IG for US IG. Your shares would rally sharply due to the inflows. Isn’t that something you think you might want to know about ahead of time? Of course it is.
So in the interest of continuing that very discussion, consider that one of the things I noted on Friday is that â¬ IG has underperformed US IG by around 7bps this year. If you’re thinking “7bps, that seems trivial,” you’re exactly right. And once again I get to flick on the light bulb for you and say “see, that’s the point”! Given the political risk associated with the French elections, and given that such a large percentage of the â¬ IG market comprises French corporate debt, doesn’t it seem like US IG should have tightened much more than 7bps relative to â¬ IG? If your answer is “probably,” then you have just uncovered mispriced risk.
Let me give you another example of the same dynamic. Have a look at this chart:
Obviously, foreign-law bonds are less susceptible to redenomination risk than domestic-law bonds. Now look at the (too lightly) shaded spread in the background. Why in the world has the spread between those two types of French IG bonds not spiked? That is, which do you want to own in the event Marine Le Pen wins and introduces a new franc – the bonds that would be at risk, or the ones that wouldn’t? Easy answer, right? Ok, well then explain to me why you’re only getting compensated to the tune of ~25bps for taking on redenomination risk. Again, that’s a trade. It’s clearly mispriced risk. Objectively, there’s no way you can tell me that “what would Warren Buffett buy?” is more “actionable” than that.
Now getting back to the mispricing I discussed in the “au contraire,” post, let’s take another look at US IG versus â¬ IG. Here’s Goldman, from a note out Friday afternoon (my highlights):
With the next few weeks and months likely to bring more clarity on the new administration’s policy agenda, we take a step back and look at the relative performance across the USD and EUR credit markets. To do so, we estimate the beta adjusted monthly excess returns of various pairs of US and European cash and synthetic credit indices. These beta-adjusted excess returns are then ranked on percentile basis relative to the history of the past three years. Exhibit 1 provides these percentiles in a heat map that flashes red when the index on the left side has outperformed its beta to the one at the top, and blue in the opposite case. Exhibit 1 shows two clear themes since President Trump’s inauguration. The first is the outperformance of the US vs. Europe, in both the cash and synthetic markets. For example, the “beta adjusted” outperformance of CDX IG vs. iTraxx Main ranks in the 92nd percentile vs. the history of the past three years. The second visible theme is the outperformance of HY vs. IG, in both cash and synthetics and across both sides of the Atlantic.
So this is really, really interesting if you understand everything said above.
Think back to what I said about the relative outperformance of US IG versus â¬ IG and how 7bps didn’t seem to cut it considering the political risk in Europe. Well, what Goldman is suggesting is that a more nuanced look at things indicates that in fact, synthetic US IG’s beta adjusted outperformance relative to synthetic â¬ IG falls in the 92 percentile versus history. In the cash market (i.e. iBoxx US IG versus iBoxx â¬ IG), the percentile ranking is 72nd.
In fact then, we may be wrong to assume that European credit is entirely asleep at the wheel in the face of French election risk, and in turn, that would mean that we may be overestimating the reward we can expect to reap from preferring US IG versus â¬ IG.
Once again, someone will say this is all “noise” or otherwise too esoteric to be of any use. But remember that one of my main goals in writing for public consumption is to reduce the information asymmetry between the “pros” (whatever that actually means) and average investors. Part of that pursuit involves giving you a window into the kind of things that professional investors take into account when making decisions on how to allocate large sums of money. If you don’t consider that valuable, then for you, the information asymmetry to which I referred above will remain very, very pronounced.
Finally, consider the second trend Goldman mentions in the excerpted passages (i.e. high yield outperformance versus investment grade). I’ve consistently argued that while both IG and HY (NYSEARCA:HYG) are grossly rich after 12 months of virtually uninterrupted spread compression, HY is even more overvalued than its high grade counterpart. This trend is mirrored within HY. Consider the following, again from Goldman (my highlights):
As we discussed last week, this “down in quality” view has performed well over the past months as tighter HY spreads and a solid macro backdrop have provided strong incentives for investors to move down in the quality spectrum. With CCC-rated bond spreads standing at their 23rd percentile vs. the postcrisis period, compared with the 6th and 2nd percentiles for B and BB-rated bond spreads, respectively, we think the tactical case for more compression within the HY rating spectrum remains compelling. But over a longer investment horizon, we think it is important to keep in mind that the rally over the past few months has significantly eroded the premium net of losses embedded in CCC-rated bonds, both on an absolute basis and relative to the BB and B buckets. Exhibit 2 illustrates this and shows a decomposition of current spread levels into our estimates of the annualized expected loss rate and the credit risk premium, the five-year excess IRR on a buy and hold strategy of diversified credit portfolios.
That chart is nothing short of incredible and it speaks volumes about the dynamic I’ve discussed before regarding how the central bank-inspired hunt for yield is driving investors down the quality ladder.
What you’re looking at there (essentially) is how your cushion – as a HY investor – has eroded with the rally. The first thing to observe is that that cushion has shrunk meaningfully across buckets, but in the lowest quality bucket, it’s vanished almost entirely. So that chart simultaneously says something about the trip down the quality ladder and about the extent to which that has served to pretty much eliminate the risk premium investors can expect in the lowest quality credits.
As usual, I’ve tried to keep a running tab of the takeaways. That is, if you go back read this again, you’ll see that for each subject broached, the takeaways for investors are explained on the fly.
The larger point, is that what you’ve just read represents another attempt on my part to help readers get their minds around the type of analysis you need to make truly informed decisions about investing at the asset class level.
Everyone is willing to glance at someone’s DCF model on a particular stock that while impressive on the surface, was really just auto-generated after someone plugged in some assumptions. But retail money exhibits a generalized aversion to understanding how asset classes are priced and how macro factors influence that pricing. The credit market (described above) is a prime example.
That aversion is unfortunate because as the Heisenberg raison d’Ãªtre clearly states, going forward geopolitics are going to influence macro, macro is going to drive asset prices, and the collapse of stock and sector correlations notwithstanding, I contend that overall trends in asset prices are going to be more important than idiosyncratic factors in determining the trajectory of individual securities.
Have a good weekend.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.