Ken Grant - Hark Off

Ken discusses the method of 'Hypothesizing After Results are Known', and its application from Kavanaugh to Musk.

Hark Off

September 30, 2018

“Hark the whimper of a seagull, He weeps because he’s not an ea-gull,

Suppose you were, you silly seagull, could you explain it to your she-gull?”

-- Ogden Nash

A lot to cover after a busy week – particularly with a likely-busier week following on its heels. I begin by bidding a bittersweet farewell to Martyn Jerel Buchwald, better known as Marty Balin, founding member of the Jefferson Airplane, who took flight from his mortal coil this past weekend. Marty probably never wanted to be either a seagull or an ea-gull; clearly he preferred to take to the skies in an Airplane. Or a Starship. One way or another, this High Flying Bird took off, and I wonder if looks on me as he flies, so high. But darling look at me, yeah, I’m rooted like a tree, yes I am... …and so are most of y’all.

Separately, and as the years go by, I find myself increasingly marking their passage by my survival of U.N. Week in NYC. I managed to get through another one, but I wish the whole enterprise would just hark off. Strip clubs and illegal parking by night; lecturing about America’s moral failings by day. Can’t they find another spot for this? Say, for instance, the peace/freedom-loving port city of Havana? The Castros, I’m sure, would be only too glad to welcome them with open arms and show them a good time.

But what really caught my eye this week was an editorial in the WSJ describing the downfall of a former high-flying Cornell University scientist somewhat ironically named Brian Wansink (wanna sink?), who was revealed by his colleagues to have blatantly and repeatedly sinned against science. Know that his transgressions, were gall and wormwood to us unfortunate wretches who wallow in data analysis. Specifically, he has shown to have been a serial curve fitter: one who massages data and reruns results until they match up to the point he’s trying to make. More to our purposes here: he is accused of HARKing: an acronym derived from a method known as Hypothesis After Results are Known. For shame, Sink, for shame.

On the other hand, who among us can truly attest to have never curve fitted, never having HARKed? Because all of us statistical seagulls have pined, at one time or another, to be scientific ea-gulls. So part sympathizes with Professor W, and that portion of my psyche might even go so far as to say that not only is an attack on the practice mean and unnecessarily hurtful, it’s positively un-American.

And if you doubt this, I suggest you just take a look around you. If you do, you’ll see HARKing seagulls everywhere you point your peepers. Of course, Exhibit A is Judge Kavanaugh and the spectacle surrounding his prospective elevation to the eagle’s nest of the United States Supreme Court. Whatever one’s predispositions concerning this episode, and however it ultimately plays out, it is beyond dispute that from the outset, his political enemies wanted to crush his ambitions at any cost, and that knowing the desired result, they went on a frantic search for an appropriate hypothesis. At this point, it’s about a flip of the coin as to whether they can pull off the HARK, but one has to give them mad props for a flawless execution of this timeless and time-tested approach to jurisprudence.

I also feel, though with some reluctance, that we should give a HARKing shout-out to everyone’s favorite bong-toting entrepreneur: one Elon Musk. Under full attack from virtually every quarter, he socialized the notion that if investors didn’t like the way he was rolling with Tesla, he might just take his battery-powered ball home and go private. More likely than not, he knew that he couldn’t do so at anything other than a discount, but went ahead and hypothesized that investors were lining up to buy out the public at a premium. In an assist to my thematic struggles, all of this transpired in the seagull/eagull-rich, avian realms of Twitter. But not only didn’t his HARK achieve the desired outcomes, the Feds and the SEC are now coming after him, and, like the HARKs that they are, the stripped him of his chairmanship.

But the whole HARK thing doesn’t end there, not by a long shot. Recent data shows multi-year records in both stock buybacks and insider sales. Now, this, my friends, is indeed the American Way. Use investor capital to buy back shares, as helpfully supplied by the corporate chieftains who are authorizing the purchases. Ladies and gentlemen, this is the HARK as it might’ve been ordained by the Market Gods. Valuations are at or near an all-time record and, having accumulated a galaxy of shares in their company, a CEO might justifiable believe that now is as good a time as any to cull his or her herd of shares. But why risk doing so on a tape suffering gravitational pull? Well, then, why not sell your stock, in contemporaneous time, to your own treasury, at prices you yourself specify? Who is the wiser, who, in fact is harmed? HARK, the herald angels sing.

But most of us are not Cornell researchers, CEOs of publicly traded automotive concerns, or, for that matter, CEOs of any kind. So what are we to make of all this HARKING jazz?

Well, somehow, like those solemn U.N. rituals described above, perhaps the best that can be said about recent market history is that both September and Q3 are over. My anecdotal evidence suggests that few market types will miss them. I know I certainly won’t.

As we enter the home stretch of this extremely quirky year, it strikes me that we do so with market factors in odd configuration. The Gallant 500 enters the proceedings 17 skinny handles below its all-time highs; the NDX charts are in approximately the same relative zenith proximity. Last week was painted in red for both benchmarks, but the action – with beaten down names like NXPI, FB and TWTR subject to incremental pastings -- suggested to me that a large number of institutional capital pools were selling down losers so as not to be compelled to list them on their 13Fs and other quarterly position disclosures.

Meanwhile, mid-week and with little fanfare amid the Washingtonian circus, the Fed not only nudged its overnight rates to above 2%, but did so in unapologetic, almost aggressive matter. Longer-term yields, while wavering a titch, remain visibly above the 3% threshold. All of this, albeit in delayed fashion, gave a rational boost to the USD:

I suspect, however, that at least some of this recent USD love/EUR hate is due and owing to economically unfortunate decisions by the Italians to spend tax dollars that they don’t have like Sicilian Sailors, in the process thumbing their collective noses at their paymasters in Brussels and Berlin.

Italian stocks and bonds sold off dramatically and in Pavlovian fashion, and I’d share those charts with you if it weren’t for their buzz-killing nature (to say nothing of space constraints).

But in the U.S., all appears to be good in the hood. The VIX has drifted down, seagull-like, to just above 12, and who’s to say it won’t waft benignly lower -- into the 10-handle, last breached in the early days of August?

There are, however, some signs that investors are antsy and showing some desperation for respectable yields. And who can blame them? After all, they’ve got to eat too, as do their capital providers.

A somewhat alarming manifestation of this is evident in recent credit spreads, which show pretty unambiguously a net flow of funds out of Investment Grade paper and into the more pulse-quickening but historically capricious embrace of High Yield.

I must state that over my long career (which carbon-dating now traces back to the days of Edward the Confessor), I’ve never seen this sort of thing play out without ending badly, but I’ll stop short of trying to pull a HARK here, because, well, I just don’t have the chops.

The one thing about which I’m most certain is that October should be wild and wooly, with virtually every pertinent, market-impacting factor in play. Perhaps the informed action begins in earnest with Friday’s Jobs Report, but before we even get to this, our senses are likely to be assaulted by breathless revelations about the Kavanaugh Hark, the Rosenstein Roast, or maybe even some news out of the trade wars with Canada and China. Everyone, of course, will be watching for impacts on the Mid-Term polls, and rightfully so, because different electoral outcomes imply widely divergent market paradigms. The pollsters, as we all are aware (and it pains me to write this), have shown themselves to be less than infallible, but given the stakes involved, it’s pretty hard to ignore them in terms of deciding how much risk one wishes to sustain, and how best to distribute it.

I also am compelled to remind my crew that after the Jobs Report, the monthly/quarterly macro data and earnings cycle will accelerate, and that the results may impact the qualitative data flows referenced above, setting up for a potential for some crazy action, just around the bend.

It is thus my further grim but unshakeable duty to remind everyone that given: a) the relative absence of factor volatility over recent weeks; and b) the likelihood that this respite is about to end, the backward looking risk analytics (including those provided by yours truly) are probably underestimating the price dispersion their portfolios are likely to experience over the next 6 weeks. Moreover, this pending action is arriving against the backdrop of what I can only acknowledge to be a difficult performance environment for most funds through the first three quarters, and the associated need for these tables to be turned in the 12 short weeks we have left to HARKy ’18.

And, reverting to our main theme, I must inform you that in a simpler, less tear-veiled world, I might be tempted to counsel everyone as follows: if all else fails, just go pull a HARK. But I don’t think this option is open to many. We’re not politicians, we’re not CEOs, we’re not research scientists; we are, for the most part, seagulls. But consider this, even for the eagulls among us, as I have hopefully demonstrated in these pages, the strategy is by no means guaranteed to work.

So, in conclusion, whatever else happens, I’d say the HARK is off.

TIMSHEL

Comments