Friends,
Back in March, I wrote a piece about looking for deep value opportunities, driven by changes in market structure. At the time, I noted that pod shops were creating unusual mispricings in equities, as they were operating with a valuation agnostic framework, only chasing rate of change:
“In my realm of value investing, I’m genuinely amazed at how these pods will short high-quality, rapidly growing businesses at under five times cash flow—just because the next quarter will be weak. I don’t understand how that strategy makes money, except during highly truncated bear-raids, yet the pods keep playing at it as they fixate on short-term rates of change. Then right after the negative print, they often accelerate their short selling, pressuring the stock in the pre-market and further spooking the longs. They want to take a bad quarter and stampede things, so that they can cover. Even then, sometimes they don’t cover until the data stops inflecting negatively. Then they cover en masse.”
At the time, I attributed this incredible opportunity set partly to malice, with the belief that that pod shops were trying to spook long-only players and create selling, so that they could cover. What if there were an even simpler explanation to all of this?? What if there were also a quantitative answer??
Fortunately, my good friend, Kevin “The Macro Tourist” Muir, has given me permission to publish a recent paywalled piece of his, where he looks at why stocks tend to move randomly—totally divorced from valuation. (For the record, I think you guys should all subscribe as he publishes one of my favorite blogs).
As noted in my original posting, I believe that pods are one of the greatest sources of Alpha currently in the markets. I want to understand them better, so that I can abuse them more effectively. I want to reverse engineer their formulas, as I intend to use them as my pinatas. Kevin’s piece helps to distill why things seem to overshoot much further than in the past. Understanding this, should help me to improve my timing on position entry (I tend to be too early buying pullbacks), but also help me to dodge landmines along the way.
I think this piece was the missing link on my journey to understand why the market structure has changed so completely. Please read it twice. Enjoy…
Kuppy
The other day I was chatting with a good friend and our conversation went something like this:
“I just don’t understand why my stock is getting hammered,” he admitted. “Sure, the near-term outlook isn’t that good, but by 2026, this stock will print scads of money.”
My friend then goes on this five-minute rant about how this industry is set to explode. He explains the fundamentals in ways that I can only dream about being half as articulate. He knows the story cold. He has thought through every angle. The short-term is challenging, but he is completely convinced about the stock’s long-term prospects.
“I watch all the stock trade, and the short interest is actually increasing! This makes no sense. What do these short sellers know that I don’t?”, he asks.
“You’re thinking about this all wrong,” I tell him. “You’re assuming the market is efficient. It’s anything but. I’ll bet I can explain the movement in your stock over the past year.”
“Really? How?”
“Stocks are overwhelmingly moving on forward EPS revisions,” I explain.
“But why would they short something that’s so cheap?” he asks.
“Cause these managers aren’t thinking about it as an individual stock. They are thinking about it as a portfolio strategy. Let’s pull up your stock with the forward EPS estimates and see if it correlates to the price action.”
“You see how the periods of rising EPS forecasts (positive revisions) saw your stock price gain? And then look at what happened when revisions went negative; the stock price fell. You’re wondering why your stock is getting hammered over the past couple of months, but it’s obvious that EPS revisions are comping negative.”
In reality, there isn’t just one factor like EPS revisions that can explain all the movement, but it was an easy prediction for me to make because this factor has become so important.
Here is Morgan Stanley’s Mike Wilson explaining this phenomenon:
Mike Wilson: Idiosyncratic opportunities for institutional clients — that’s where they’ve been focused, and where they should be focused — and it’s a great stock picking market right now.
Jonathan Ferro: Can you share a single name? Or can you describe it?
Mike Wilson: The factors that are really continuing to work are quality and 3-month earnings revisions. And that’s how we are positioned. That’s the quality cyclical trade. That’s where the earnings revisions are in the better mood.
Some of the big cap growth stocks’ revision factors are starting to fade because they got too extended.
All right, let’s take a step back and explain how this all works.
Analysts make forecasts about what a company will earn in the future. At any moment in time, the company’s price should reflect an approximation of known fundamentals. However, stock prices move on expectations of changes in future earnings. If the company’s prospects improve, then analysts will increase their forward EPS estimates, and the stock should rise in sympathy.
Before quant trading became popular, the vast majority of portfolio managers did bottoms-up fundamental research on individual companies and then decided on a fair price to pay. When the company got cheap versus their expectations, they bought it. When it got dear, they sold it. These legions of fundamental portfolio managers set the price for individual securities.
However, somewhere along the way, quants figured out that they could create “factors” that could generate positive alpha. For example, one factor they realized affected stock prices was the change in forward EPS estimates. To measure its effectiveness, they created a way to monitor the factor’s performance. There are different methods, but here’s how Bloomberg does it; they take a universe of securities, rank them by the factor (in this case, we’ll use the 3-month change in the current fiscal quarter earnings estimate), and then create a portfolio where they are long the top 20% and short the bottom 20%. This portfolio is then rebalanced on a weekly or monthly basis. Since there is no net position, if the market was perfectly efficient, we should be left with a random return path.
However, that’s not what we get.
Look at this factor’s performance. Sure, there are some decent sized drawdowns, but on the whole, if you managed those dips, this is a fabulous strategy. Especially since COVID.
I am not suggesting that there is a bunch of money trading a strategy this simple. But this is a good indication of the amount of alpha that is out there for quant portfolio construction. Sure, it takes some work. All factors are not instant money makers. For example, here is the 3-month EPS change if we use the entire fiscal year instead of the first quarter.
Not nearly as good. But strangely enough, if we use 3-month change on following fiscal year, it’s better.
And the 3-month timeframe is an arbitrary number. Here is the 1-month EPS change of the current fiscal quarter.
And the 1-month change in following fiscal quarter works well too.
Although the current fiscal year doesn’t work well with 3-month change, the 1-month change is a different story.
The point that I am making is that you can create a portfolio that focuses solely on factors that generate alpha, add in some money management, and then a healthy dose of leverage, and it’s a terrific strategy.
Here’s another way of looking at it. One of my buddies is an ardent supporter of a piece of software called Portfolio 123. It allows him to create quant systems that are amazingly easy to program and yet shockingly robust. He was kind enough to take our simple 3-month EPS revision factor and run it through the P123 system. They have a different way of showing the results, which I thought were illuminating.
These are the annualized forward returns for the different deciles of the Russell 3000 3-month change in EPS revision (rebalanced weekly) for the period from the end of 2003 to last week.
The bottom 10% of the stocks returned only 1.01% on an annualized basis during this period. Meanwhile, the top 10% returned 14.61%. I would like to make something clear. I am not suggesting that this factor works all the time. Look closely at those charts. During the period from 2015 to 2020, EPS revisions were not a top performing factor.
However, when a factor is generating alpha, it has a tendency to trend. Its success attracts more proponents, which only makes it work even better. As Mike Wilson highlighted, right now, it’s one of the best performing strategies out there.
_________________________________________________________________________________________________
What does this mean for your trading?
If you are short-term oriented, you should be aware that this is a big determinant in what’s driving stocks. The other day, a bunch of my trading pals were grousing about how a stock that a particular sell-side firm upgraded was declining in price. It was as if the firm had issued a sell recommendation instead of a buy. I checked to see if my theory held. The stock was Aritzia, and the upgrade was BMO’s October 15th, 2024 raising of the target price.
On the morning of the upgrade, the stock opened strong, but then proceeded to sell off for the whole next week.
Why the bad performance after the analyst’s upgrade?
Because even though the analyst raised his target, the forward EPS came down.
Unexpected EPS revisions are often at the root of unexplained stock movements.
_________________________________________________________________________________________________
It works for macro too!
Did you catch Mike Wilson’s slip where he said “the factors that are continuing to work are earnings revision breadth?” He quickly changed that to 3-month EPS revisions because he realized he was talking about individual stocks, but the reason he mentioned breadth is that this analysis can work on a macro country level as well.
One of Mike’s competitors, Bank of America, creates a monthly report where they measure the “breadth of earnings revisions.” They count how many companies’ EPS forecasts are being revised higher and then compare it to the amount being revised lower to create a “earnings revision breadth measure.”
For example, over the last three months, Japan is the only country with positive earnings revision breadth.
Over time, if you buy the higher ERR (earning revision ratio) countries and short the low ERR countries, you have another winning strategy.
Earnings revisions offer alpha in a variety of different strategies.
_________________________________________________________________________________________________
Are we sure this is progress?
If you talk to professional investors, many will lament that the markets have become “stupider” over time. I don’t disagree. And for all of us that feel this way, we can take comfort in the fact that we are in good company. Recently, AQR’s Cliff Asness, wrote a paper about this very subject: THE LESS EFFICIENT MARKET HYPOTHESIS.
Although I will gladly accept Cliff’s theories as playing a part in the “dumbing down” of markets, I think Cliff overlooked a rather large contributor. And his blindness might come from the fact that his own strategies are partly responsible.
Over the past two decades, quant shops have attracted a tremendous amount of assets — especially the pod shops that use many of these factors as the basis of their trading. There can be no denying that today there are way less fundamental-type managers. Heck, some traditional long-only funds even have quant screening applied to their list of securities they are able to purchase. These factors are increasingly driving portfolio construction.
Bringing this back to my buddy with the stock that’s falling in price — he’s a rare breed thinking about long-term fundamentals. But he won’t necessarily be wrong. If he is correct about the stock’s earnings in 2026, then eventually analysts will increase their forward EPS and his stock will go from screening negatively to positively. The quants shorting it today because EPS estimates are declining will cover, and then when earnings start increasing, the quants will rush in to buy.
You might say that’s dumb. Why short something today that you’ll likely be covering six months from now and then chasing it higher when earnings improve? The answer lies in the fact that, however dumb that trade might seem on an individual basis, when you apply this strategy to a portfolio, it works. Quant trading is attracting assets because it is offering superior returns.
_________________________________________________________________________________________________
The problem with quant strategies
I have absolutely no problems with any of these quant strategies. I am not one to rail against a new method of trading. I want to understand all forms of trading and investing so that I can adapt, but if you expect it to always remain the same, you’re in the wrong line of business.
For my buddy whose stock has negative EPS revisions, which is causing the stock to decline, the situation will eventually work itself out. At some point, the fundamentals will become too obvious for analysts to ignore, and earnings forecasts will be jacked higher, and the stock price will soar. If you have a long time horizon, then you can simply hang tough through this period. Or you can try to time the changes in forward EPS and play the game. Once you understand what’s moving the stock, it’s easier to trade around your position.
However, there is another type of quant trading strategy that offers a problem that is not as easily remedied. Mike Wilson mentioned two factors that have been performing well lately; 3-month EPS revisions and quality. We’ve explained 3-month EPS revisions, but what about quality? The problem with quality is that it’s not easily definable.
Here is GMO’s Jeremy Grantham explaining his take on quality:
Interviewer: So you referenced the late 90s period in the basket of internet stocks, many of which were pretty junky companies. I wanted to see if you could contrast that to the A.I. companies which seem to be a higher quality basket of companies. You have been interested in quality for many years. You launched GMO’s quality strategy fund twenty years ago, and GMO recently launched an ETF around quality. Could you contrast some of the A.I. fundamental characteristics with those companies that led the bubble in the late 90s period.
Jeremy Grantham: This seems like a PhD paper question. There were obviously a lot of Pets.com ideas in the internet light-weight companies. There are probably quite a few this time. And history will decide which ones they are. But the very nature of quality is monopoly.
We define quality (and always have) as high, stable returns. The only way, in a competitive society to have high, stable returns, is to find a nook or cranny that is not that competitive — ie: you have a monopoly. You have a defensive moat.
So what we are talking about is the dramatic emergence, in the last ten years or so, of what you might call; great global monopolies. Almost instant in some cases, where winner takes all. First come, first serve. They grab the market. They defend it brilliantly. Ferociously. It’s what capitalists do. It’s what they are paid to do. And it’s what all the brilliant leaders manage to achieve.
Although Jeremy veers towards explaining quality as near-monopolies, it’s the “high, stable returns” that most people agree is attractive. You will often see EPS volatility as one of the screening factors for quality.
[As a quick aside, I recently interviewed Warren Pies from 3-fourteen Research and he had an interesting take on quality. He argued that companies shouldn’t be punished for upside earnings volatility — only downside. So his quality screener only penalized downside volatility. You can check out his quality ETF FCTE here. Not a recommendation to buy, just passing along for your education]
No matter how you measure quality, quant investors have flocked into these sorts of stable, high return companies. The factor has worked.
The problem is that it’s become a self-reinforcing feedback loop. As more money has flowed into quant strategies, they have chased the same high-quality names. In doing so, some of these stocks have reached dizzying valuations.
I am lucky enough to have an extremely shrewd hedge fund manager as a subscriber/friend who likes to say that everyone’s myopic focus on the MAG7 is misplaced. He argues that there is a modern-day Nifty 50 of extremely expensive “quality” stocks.
One of those is Costco. It’s an absolutely wonderful business with a stable, high return profile.
From 2002 until 2014, it traded between a 16x and 24x multiple on forward 1-year EPS. Then, from 2015 to 2020, it rose to a 22x to 30x range. However, in the post-COVID environment, this P/E has exploded all the way up to 50x!!!
Is this a “quality” company? Sure! No doubt. But the problem with factor-type quant trading strategies is that there isn’t necessarily a “human sanity” check to stop absurd valuations.
This is a bubble. No doubt about it. It’s not a bubble in the traditional sense where people are talking about it at cocktail parties. No one is discussing their exciting Costco position. Yet, the money flowing into these strategies has created a group of investors that will invest in “factors” with little regards to overall valuations. After all, the strategy is working. Pundits that worried about Costco being expensive at 25x look foolish as we trade at 50x, so who is to say we’re not going to 100x? And yes, that sort of thinking works as long as investors keep buying these “quality” companies regardless of valuations, but over the long-term, there is little chance that these purchases will prove wise investments.
Factor investing is a victim of its own success. Like all things on Wall Street, it’s a good idea taken too far. I don’t know when “quality” stocks will correct, but I would be careful with assuming that just because a factor has been working, that it will continue forever. In the meantime, get to know these factors. Figure out what’s working and try to understand how they are affecting markets. Markets are evolving and even old fogies like me need to adapt to the new reality.
Thanks for reading,
Kevin Muir, the MacroTourist
The Fed Is Fuct (Part 4)…