In the movie Terminator, various faceless machines, and one especially murderous one almost caused the end of the world. In an ironic twist of life imitating art, the very core premise of our capital markets - the effective allocation of capital to worthy assets on the basis of solid fundamental analysis (and yes, "information arbitrage") is on the verge of being eliminated by the same combination of forces: millions of faceless, anonymous algos, and one destructive endoskeletal machine. Remember: Ben Bernanke is out there. It can't be bargained with. It can't be reasoned with. It doesn't feel pity, or remorse, or fear. And it absolutely will not stop, ever, until both the middle class, and the dollar, are dead.
As ever more carbon-based investors withdraw from stocks, the feedback loop of leaving more and more stock churning and, thus noise creation, in the hands of assorted computer algorithms becomes progressively more acute. And as the only market participants will soon be High Frequency Traders, very soon the whole concept of fundamental analysis will be unrooted at its core, in essence making capital markets, in their traditional sense of allocating scarce capital to innovative, efficient, and high ROI ideas, obsolete and redundant. Yet instead of giving up, perhaps it is our duty, in a very peculiar case of art imitating art, or in this case the movie Terminator, to redeem the freedom of capital markets from the machines.
This is precisely the point of Nicholas Colas' latest essay, in which he observes that "HFT exists among all market capitalizations, from micro cap to the largest of names in the S&P 500. The smaller end of the cap spectrum, however, has far more potential growth associated with it. Somewhere among these companies sits the “next big thing” that is currently undervalued. HFT strategies are unlikely to find these opportunities; human analysts have a much better chance if they can separate the proverbial wheat from the chaff." Colas' fundamental argument relies on the premise that "one of the outcomes of a heavily HFT focused capital market seems to be lower-than-normal P/E ratios" - we disagree, and we have demonstrated repeatedly that algorithms have an upward trending momentum bias in which self-sustaining fractal feedback loops create a micro pyramid scheme, in which incremental losses are compensated by millions of daily liquidity rebates, rendering the risk return profile of a top ticking event moot. What is troubling is that Colas himself acknowledges this possibility: "if we don’t see more private equity transactions or small cap outperformance, it may well be because the dynamics of HFT trading domination is actually overstating real values."
This is correct: courtesy of HFTs stocks are now far overvalued, as seen by the scarcity of MBO, LBO, IPOs, and various other M&A types of events. But worst of all is the realization that this is not just HFT that is to blame for the death of fundamental analysis: another culprit is Ben Bernanke himself: "the Federal Reserve’s monetary stimulus and essentially zero interest rate policy clearly has a role in asset allocation. When bonds pay very little, money flows into stocks and given the popularity of index-based investment products this also has the effect of allocating capital on other than fundamental terms." In other words, computers and the Fed have now destroyed the stock market.
Colas is of the opinion, that Private Equity, long a staple of true information arbitrage will lead the way to a return to true fundamental value. Perhaps - but is that what is truly needed? And is that an example of information arbitrage at its truly best? While it is admirable to believe that PE investors sniff out value, especially in the form of upstream dividending cash flows, they do so at the expense of the business model, the employees, and sucking a firm's assets dry. We are confident that should Nicholas analyze the incidence of default within a 5 year horizon of companies being taken private, he will change his view.
In the meantime, here is his always insightful and stimulating opinion.
From DLJ to HFT, and Beyond
Successful equity investing has always centered on “information arbitrage,” a fancy way of saying that in order to make reliable returns you need to know something that the market doesn’t. For much of the last 50 years that information has been largely fundamental in nature, unearthed by investors speaking directly to company managements, experts in the field, analysts who cover the stock, and other resources that provide actionable information. The last decade has seen a dramatic shift away from stock trading based on such efforts to High Frequency Trading, with HFT now dominating daily trading volumes in U.S. stocks. We don’t think that is necessarily sustainable, if only because this dynamic does not directly address the most basic function of a capital market – to allocate a scarce resource (capital) to its best possible use (companies that merit it). The way back to a fundamentally driven market will have to come from a renaissance of active investing, and we think P/E company takeouts of undervalued companies and small cap growth investing will likely represent the beachheads for this change.
It is tough to nail down the exact start of any important trend, but in my book modern sell-side research started in the mid 1950s when Bill Donaldson, Dan Lufkin and Dick Jenrette met at Harvard Business School and decided to form a company. DLJ was founded on the idea that excellent stock research opened doors across the investment banking landscape, from institutional customer order flow to corporate underwriting to high net worth brokerage. At the time investment analysts at sell-side firms were not especially well-paid or respected, but DLJ brought them front and center. As a strategy, it worked well enough that eventually most of their larger competition followed along, and a “Star system” of analysts with supporting teams dominates sell side research to this day.
The analog on the buy side is most likely Fidelity’s research department generally and legendary manager Peter Lynch specifically. Back in the 1970s and 1980s Lynch took the Magellan Fund from $18 million in assets to +$10 billion on the back of consistently excellent performance. The analyst team that stood behind Lynch did deep fundamental work on hundreds of companies and spoke to literally every management team on a quarterly basis if they held the stock or had an interest in making an investment. Fidelity’s research department regularly had – and still does, for that matter – its pick of the litter from any business school in the country.
It feels like a long way from DLJ and the Magellan Fund to now, as the majority of daily trading volume today is focused on opportunities that may not last longer than the blink of an eye. While the phrase “High Frequency Trading” is the popular catch-all term for this type of money management, there are several actual styles of HFT. Some seek to arbitrage Exchange Traded Funds and their underlying stock constituents and lock in risk-free profits from fleeting anomalies in price. Others just keep tabs on the myriad places where stock trades are executed – exchanges, alternative trading venues, broker owned pools of the liquidity and the like – and looks for quick buy/sell opportunities. And yes, some look for the tell tale signs of large blocks of stocks in order to trade ahead of those executions.
While HFT has its share of detractors and the May 6th flash crash seems to have only added fuel to their criticisms, I think it is also safe to say that the current market structure for trading stocks in the U.S. is not going to materially change in the near future. Everyone from the Securities and Exchange Commission to countless market observers have had their bite at this apple, and unless there is another violent intraday swing we doubt there is much appetite for revisiting the model of listed stock trading in the U.S. for at least the next 2-3 years (the next time there may be a change in the Executive Branch).
The story doesn’t end here, however, since a marketplace where the majority of trading is unrelated to long term fundamentals seems to me to be inherently unsustainable over the long term. I am the first to admit that the old model, with market makers/specialists and buy/sell side analysts had its own set of conflicts and problems. Analysts get things wrong (witness the 1990s tech bubble) and specialists/market makers can certainly get too greedy. But the fact that the majority of trading was motivated by fundamental analysis at least fulfilled the basic purpose of capital markets to attempt to allocate capital to its best possible use. If you agree it is unlikely that the SEC or other regulatory agency is going to curtail HFT, then the only way to reestablish the dominance of fundamentally driven stock prices is to have money flow into investment products that feature human decision making at the core of the investment process. Performance is the one thing guaranteed to get investors excited about active investing, so here are two investment approaches that stand a chance of leveraging the current environment and performing well enough to draw new capital.
- Looking for private equity takeout candidates. One of the outcomes of a heavily HFT focused capital market seems to be lower-than-normal P/E ratios. A handful of stocks might buck this trend – mostly super-cap technology companies – but there does appear to be a strong correlation between the increasing share of HFT trading volumes and declining P/Es over the last 5-10 years. Whether this is correlation or causation is less relevant than the question, “Does this make for an opportunity?”
- Private equity takeouts represent a pathway where listed stock markets can travel back to fundamental reality. There is no HFT in private equity land. These firms buy the whole company, operate it, hopefully grow and improve it, and then sell it. So they do their homework very carefully and try to only swing at “fat pitches” of companies with real opportunity and a cheap valuation. Stable and rising cash flow is the hallmark of a good PE story, as well as the possibility of international growth and industry consolidation. HFT strategies don’t have much of a clue about any of these factors, of course, and don’t hold stocks long enough to close valuation gaps to private company valuations. Human investors do.
- Small cap growth investing. Make no mistake – HFT exists among all market capitalizations, from micro cap to the largest of names in the S&P 500. The smaller end of the cap spectrum, however, has far more potential growth associated with it. Somewhere among these companies sits the “next big thing” that is currently undervalued. HFT strategies are unlikely to find these opportunities; human analysts have a much better chance if they can separate the proverbial wheat from the chaff.
I’ll make two other points to close out this note.
- The first is that both our P/E and small cap points work in both directions relative to what will pull the market back towards fundamental versus HFT dominated trading. If we don’t see more private equity transactions or small cap outperformance, it may well be because the dynamics of HFT trading domination is actually overstating real values. That’s a disturbing possibility, if only because U.S. stocks have already seen a lost decade of performance.
- The second point worth a mention is that HFT trading is far from the only factor generating non-fundamental trades for U.S. equities. It is too large a topic fully describe in this note, but the Federal Reserve’s monetary stimulus and essentially zero interest rate policy clearly has a role in asset allocation. When bonds pay very little, money flows into stocks and given the popularity of index-based investment products this also has the effect of allocating capital on other than fundamental terms.