Research Notes: ARK Innovation, Finding A Bottom, And Part I Of A Second-Quarter Update
Looking around the market, and at a few of our deep dives so far
Welcome to Overlooked Alpha. Subscribe below to get our best investing ideas in your inbox every Sunday morning:
Last month, I argued that a single Tweet showed why Ark Innovation ETF (ARKK) was uninvestable. Here is that Tweet:


As with so much that has happened in the market (and in life!) over the past few years, it takes some distance to truly appreciate how ridiculous the claim was.
Projecting 30-50% annual GDP growth is akin to a basketball analyst arguing that the Golden State Warriors will win a few hundred games in a row
. 30% (let alone 50%) GDP growth simply isn’t how the economic world works any more than a 150-game winning streak is how professional sports work. Whatever the accuracy of the underlying analysis (that AI might transform the economy, or that the Warriors will be an excellent team next year), the sheer insanity of the claims instantly destroys any credibility of the claimant.The market hasn’t really agreed with that analysis. Fund flows into ARKK, using the excellent fund flow database from ETF.com, were about flat over the next month. But, of late, they’ve picked up again; on Monday, Bloomberg noted the fund’s longest streak of inflows in almost a year. Net inflows totaled $639 million over the eight trading sessions, dented not by that Tweet or ARK’s insane “model” forecasting 66% annualized growth in Zoom Video Communications (ZM) over the next five years.
Right now, most investors are trying to figure out where the ‘bottom’ is. In our first post, in late March, in analyzing that same question, I pointed to the resilience of inexperienced investors, writing that “a bottom can’t be reached until the novices are run off, properly chastened.”
Quite clearly, that hasn’t happened. There’s no way to spin the recent rebound of ARKK flows as anything but those novice investors still hanging around, hoping for/believing in a rebound. There is no good investment process that ends with an investor owning ARK.
An investor can perhaps own some (or, heck, all; it’s possible sentiment really is too negative) of the stocks in the ARK Innovation portfolio; an investor can bet that the sell-off in growth has gone too far. But paying a fund manager (who a month ago forecast 30-50% annual GDP growth) 75 basis points for the privilege of owning her oversold picks is not something good investors do. It’s something that novice investors do when they’re hoping the bottom is in, when they don’t really know any better.
The argument for owning stocks here isn’t crazy. One underdiscussed aspect of the 2020 market rally is how quickly investors jumped back in. Part of it was what Matt Levine called the “boredom market hypothesis", in which retail investors traded stocks (and thus bought stocks) because there was almost literally nothing else to do. Part of it, perhaps, was low interest rates and/or fiscal stimulus.
Even considering those factors, however, the market took the long view — even for the less retail-heavy stocks that make up the S&P 500:
2020 chart. source: YCharts
The index was positive for the year by July, and at new highs by August. Yet for most of the steepest part of the rally, the outlook remained exceptionally uncertain.
At points during the bull market, I wondered whether investors — particularly professional investors — were far better than their predecessors. (On occasion, I would meet some of those investors and be disabused of that notion, but I digress.)
In theory, investors should be much better. Humans generally get better at everything over time. The amount of data available to retail investors now is substantially better than what was on offer 30 years ago to their institutional counterparts.
That aside, in the post-financial crisis bull market, there clearly was a willingness to take the long view (as quality investors should). That willingness was best reflected in the collective decision to pay multiples for growth stocks that on their face seemed potentially too steep. The long view definitely paid off:


The key reason for hope now is that investors, as they have for most of the past 13 years, will again take that long view. If that’s the case, the debate over the likelihood of a recession wouldn’t be nearly as important. In theory, recessions now should have a less significant impact on equity values, as the market has a far greater proportion of value relying on less cyclical companies in software and tech.
The same improved investors who drove the market higher during the bull run might be able to keep this market afloat. But that requires that this time be different for the same novice investors who, starting in late September 2020, pushed the market too far. As discussed in our first post, seemingly every bubble ends with the dumb money running off with its tail between its legs.
The inflows into ARKK are a concerning sign that hasn’t happened yet. So are the moves in Revlon (REV) and Redbox (RDBX). The crypto world, hammered by scams and imploding stablecoins, still doesn’t look like it’s had its reckoning. The same is true in venture capital and private equity, both of which benefited from their own dumb money investments. (Dumb money absolutely is not a description that applies exclusively to retail investors.)
Back in May, on this site we looked at still-strong inflows into ARKK as a reason for concern. The market took a steep slide south soon afterward.
History is rhyming. There’s still more reckoning to come, which in turn likely means there’s more selling to come. Indeed, as I write this section on Tuesday afternoon, equity markets are turning south. We continue to believe there’s more downside ahead, and it’s not crazy to think that investors need to look at only one metric — flows into and out of ARKK — to know when the bottom truly is in.
Three Months In
In context, our efforts three months in have been reasonably solid (figures as of Wednesday’s close):
Our average idea has posted positive returns. In a sharply down market, we take some pride in our results and a long idea performance of -3.9%.
Here’s an update on a few our picks so far:
AppLovin (APP)
Our first idea so far has been our worst idea in terms of performance, dropping 33%. APP has been hit by the sell-off in growth stocks, worries about recessionary impacts on online ad spending, and the uncertainty surrounding privacy changes from Apple (AAPL) and Alphabet (GOOG) (GOOGL).
Long-term, though, AppLovin still seems like it’s in decent shape. Valuation is particularly attractive here, and count me among those who believe that either a) Apple and Google won’t actually “blow up the digital ads business” and/or b) the industry will adapt somehow, because ad targeting is simply too important to all but the world’s largest companies.
There’s going to be a lot of volatility here, certainly. I added at the lows following a strong Q1 earnings report, though I may look to trim if we get another bounce to $40 (from a current $35). This is the kind of position to trade around in this kind of market.
Long-term, though, I still believe AppLovin is a pretty significant part of the digital ad ecosystem, and I believe that ecosystem eventually will find a new post-IDFA equilibrium.
de-SPACs
Our first Research Notes covered what was then a pretty dazzling fall in de-SPACs, companies that had gone public via mergers with SPACs (special purpose acquisition companies). At the April 6 close:
The average de-SPAC was down 32% (including one acquisition);
Barely one-sixth were above $10 (the entry price into the pre-merger SPAC);
Almost as many (48 of 291) had declined more than 75%.
Six weeks later, the figures were substantially worse:
Average returns of -53% (including two acquisitions);
Median de-SPAC traded at $3.77, versus $5.52 at Apr. 6;
7.9% of de-SPACs (23 in total) were above $10;
The total of 75%-plus decliners had exactly doubled to 96, or 33% of the total.
In the context of the market over the past six weeks, the updated numbers
perhaps don’t seem as bad as one might expect:Average returns of -56% (including two acquisitions and the first de-SPAC bankruptcy, Electric Last Mile Solutions);
Median return of -67%;
23 still above water, same as in mid-May;
39% are down 75% or more.
Overall, however, it’s been truly a stunning run. 23 de-SPACs are positive; 114 — one shy of five times as many — are down at least 75%. Yes, there’s been a bear market recently, but even in that context the performance of de-SPACs is awesome (in the original sense of the word).
Incredibly, it’s still not that easy to see bargains in the wreckage. There are more stocks that look like clear zeroes than even worthy of due diligence from the long side.
NeoGames SA (NGMS)
Our thesis for NeoGames remains largely unchanged. News since publication in mid-April has been fairly minimal. Q1 earnings were solid The acquisition of former parent Aspire Global closed, though that’s little surprise (NeoGames had agreements with major shareholders before the deal was announced). An agreement to launch in Brazil this week seems more bullish than the market is giving the stock credit for, but it’s not necessarily a transformative move.
For now, the original thesis seems good enough. This is a defensive, growing business with a more-than-reasonable valuation. NGMS has defied the market downturn so far, gaining 3% since our recommendation (even with a sharp sell-off over the past two sessions). It’s the kind of stock that can continue to do so going forward.
1-800-Flowers.com (FLWS)
We got FLWS wrong. Full stop. We looked at a big decline and a ‘cheap’ valuation, and ignored the cyclical exposure (particularly for the high-dollar gifting business, including fruit seller Harry & David) and thin EBITDA margins. 1-800-Flowers.com tanked its fiscal Q3 report in late April, and there seems more risk to the outlook going forward.
We do eat our own cooking. In this case, my portfolio got food poisoning. To stretch the metaphor, at least the toxin has been expelled.
CPaaS Stocks
On April 22, we took a look at CPaaS (Communications-Platform-as-a-Service) stocks. The members of the group — Twilio (TWLO), Bandwidth (BAND), Kaleyra (KLR), and Zenvia (ZENV) — had declined between 63% (KLR) and 87% (BAND) from their highs. Valuation looked much more reasonable; the long-term case for the industry still seemed intact.
Yet…looking with fresh eyes, valuation didn’t seem that compelling for a sector with relatively low gross margins (at least by tech standards).
To some degree, that analysis was correct: the group wasn’t compelling. What it was, however, was a pretty attractive short opportunity that we hinted at but didn’t execute:
source: YCharts. performance since April 22
CPaaS stocks represent yet another data point showing how insane valuations were at the 2021 peak for growth stocks. Gross margins weren’t high to begin with, and in a relatively commoditized industry, had (and have!) room to compress further. Yet all four stocks, to varying degrees, received ‘tech’-type price-to-revenue multiples (where ‘tech’-type often meant ‘SaaS-type’).
That’s no longer the case — and, man, it’s not hard to be a little tempted. KLR and ZENV, in particular, look like high-risk, high-reward bets here. Neither company appears to have substantial odds of near-term bankruptcy; both are still growing revenue; both should have a path to at least breakeven operating cash flow, which is enough to at least maintain some optionality in the equity value.
It may be too early. But as bearish as we are about the market, the sell-off in growth stocks has at least moved to a point where there are some possible fundamental bargains out there. Without using anchoring bias and relative valuation (the dangers of which have been proven so far this year), it was much tougher to make that case only a couple of months ago.
Nathan’s Famous (NATH)
Our biggest winner from the long side, NATH stock has gained 23% since our call on May 1. To be completely honest, it’s not entirely clear why.
To be sure, the stock was too cheap, and as we wrote a better business than unfamiliar investors might think. Fourth quarter earnings earlier this month did look solid, and perhaps ameliorated some of the inflation/demand concerns that had led the stock to a 52-week low.
Packaged food stocks have had a nice multi-session run (though NATH took off a bit earlier than that). More broadly, NATH is pretty thinly-traded, and it is a stock that makes surprisingly large moves on occasion for reasons that, as is the case at the moment, aren’t entirely clear.
In this market, there’s a case for taking some profit. At $47, the argument was that the stock traded at about 12x free cash flow pro forma for some of its potential moves. We’re now at ~16x FCF; attractive for what essentially is a licensing business, but obviously not the zero-growth multiple on offer five weeks ago.
In addition, one of the company’s options was to refinance its debt, which currently bears a 6.625% coupon. But in the new interest-rate environment, those savings may disappear. Indeed, that debt (callable at par starting in November) now is priced at 97. The move below 100 clearly seems driven by the interest rate environment, not any real risk in the business model (Nathan’s remains under 2x net leveraged).
It’s worth letting this run play out; this kind of strength in this kind of market is something to be respected. But positions here probably can at least be trimmed, to either de-risk the portfolio against a further market-wide decline (these stocks historically are the last to go) or to fund new positions in more aggressive bets when a bottom indeed is approaching.
Quanex Building Products (NX)
When we pitched NX stock in early May, we argued it was the best way to play a long-term housing shortage in the U.S. Thanks largely to a blowout earnings report a month later, the stock is 15% higher. As with NATH, in this market, that does raise the question of taking a quick profit.
There are some reasons to do so. Quanex is not just a small-cap building products supplier, but one whose business is heavily reliant on end markets. The window replacement market is somewhat anti-cyclical (there’s really no putting off a broken window), but that’s about 50% of sales. In other words, ~half of the window business (in turn, 90% of profit) still is directly tied to housing starts in the U.S. and Europe.
Over time, those housing starts are going to rise. They simply have to. But “over time” is a pretty big qualifier at the moment with mortgage rates soaring. Yes, earnings were strong, and yes, the stock is cheap. But it’s a stock that’s going to be cheap pretty much forever (as we wrote, it’s not the most attractive business model out there), and it’s a stock that (again, as we wrote), was almost hysterically rangebound before the novel coronavirus pandemic arrived.
The trade before 2020 (for almost a decade) was to buy NX at $16 and sell it at $23. We bought it at $19; in 2022, we have to at least consider selling it at $23.
More updates to follow in next week’s notes.
Disclosure: As of this writing, Vince Martin is long APP, NGMS, NATH, and NX. He may exit, add to, or reduce those positions without further notice.
Disclaimer: The information in this newsletter is not and should not be construed as investment advice. Overlooked Alpha is for information, entertainment purposes only. Contributors are not registered financial advisors and do not purport to tell or recommend which securities customers should buy or sell for themselves. We strive to provide accurate analysis but mistakes and errors do occur. No warranty is made to the accuracy, completeness or correctness of the information provided. The information in the publication may become outdated and there is no obligation to update any such information. Past performance is not a guide to future performance, future returns are not guaranteed, and a loss of original capital may occur. Contributors may hold or acquire securities covered in this publication, and may purchase or sell such securities at any time, including security positions that are inconsistent or contrary to positions mentioned in this publication, all without prior notice to any of the subscribers to this publication. Investors should make their own decisions regarding the prospects of any company discussed herein based on such investors’ own review of publicly available information and should not rely on the information contained herein.
For those not familiar with the NBA, the longest winning streak ever was 33 games.
At some point, the value of targeted advertising needs to be made clear to consumers. I’m not unsympathetic to privacy concerns, but the trade-off isn’t just free search and social media. It’s having a diverse world of unique businesses and products that quite literally will go away if there’s no ability to target and track.
These analyses all cover the same cohort, of those de-SPAC mergers that closed by Apr. 6. Deals that closed post-4/6 thus aren’t included. They wouldn’t change the figures much, and the apples-to-apples comparison here is somewhat useful.