Deep Value: Finding Gems in the Cheapest 20% of the Market
Mr. Wellington:
Lyrical Asset Management is a New York City-based equity investing boutique. We are focused on one thing, and that’s value investing in equities, both in the U.S. and in developed markets abroad, so we have U.S., we have international, and global.
We also have a subset of our global portfolio that’s a relatively unique product in the space of impact investing. We believe we’re the only value style impact fund. That product is called GIVEs, which stands for Global Impact Value Equities. That comes from our global portfolio; we have about 60 stocks that we own between our U.S. and international portfolios, and a subset of those, about 25 or so, also are making a quantifiable positive impact on one of the United Nations sustainable development goals.
We were founded back in 2008. We launched with our first client at the start of 2009. Firm assets today are right around $8 billion. The firm was co-founded by myself and Jeff Keswin. He and I have a friendship that goes back over 35 years to when we were undergrads at the University of Pennsylvania, enrolled in the Management and Technology dual degree program. After college, we were roommates, and then about 18 years ago, we started this business together.
Mr. Wellington:
Value investing hasn’t been the most popular investing style for a while, mostly because of the poor performance of the value indices. But if you look at the data, the cheapest stocks historically have had the highest performance. And the opposite is true, too. The most expensive stocks historically have had the worst performance.
When we founded Lyrical, our goal was to generate the highest returns we could over the long run. We didn’t have to be value investors. We could invest anywhere we thought would generate the highest returns. But looking at the data, the cheapest stocks are where you find the highest returns, and so we are value investors. We look for our investments in the cheapest 20% of the market, the cheapest quintile.
This is now my 30th year as a value investor, so I’ve been doing this a very long time. Over the first dozen years or so of my career, I was sifting through this cheapest quintile of the market, and I began to observe a few things. I noticed that I was a lot more successful as an analyst when I analyzed a good business than when I analyzed a bad business.
When I analyzed a good business, things tended to work out a lot more often. The real world is unpredictable, and all kinds of unexpected things would happen. But good businesses found a way to adjust, adapt, be resilient, and still end up making about the same earnings I projected them to.
Bad businesses were the opposite. A lot of things could go right, but if only one or two things went wrong, I ended up with earnings much worse than I projected, and that made them much harder to get right.
And so that was one observation — that good businesses tend to work out a lot better than bad businesses.
The other observation was that the cheapest 200 is a giant pile of junk. As a group, they have much lower quality characteristics, and much lower growth characteristics than the rest of the market.
However, if you carefully, meticulously screen and sift through those cheapest stocks, you will find that there are some really good businesses that slip through the cracks, and are ignored or misunderstood by the market. And while they are rare exceptions, there are just about enough of them that you can build a concentrated portfolio out of just these exceptions.
We refer to these exceptions as the “gems amid the junk.” They’re amid the junk because they’re in the cheapest 20% of the market, where most companies are junk. But they are gems. They have growth rates and returns on invested capital as good or better than the overall market, despite their heavily discounted price. We purely focus on finding these gems.
Probably the hardest thing about investing in them is finding them, because they are so rare. When you find them, you should end up with better investment results. You should get more stocks right and fewer stocks wrong, and your portfolio should generate a higher return.
TWST: In terms of your research and investment process, what other characteristics or metrics do you look for?
Mr. Wellington:
We start with a value screen that we’ve built in-house. That takes the top 1,000 stocks in the U.S., and outside the U.S., the top 1,500 stocks in the developed markets. The screen enables us to sort those large universes by valuation.
While the screen isn’t perfect, one huge way it helps is it tells us which 80% of the market we can ignore, because even if they do grow as much as expected, they are fairly valued or, even worse, they’re expensive. It helps us focus on those stocks that appear to be trading at big discounts.
From there it really becomes a manual process. We’ve honed our skills in finding these gems amid the junk. We know exactly what we’re looking for, or perhaps more importantly, we know exactly what we’re not looking for.
A lot of the names at the very top of the screen, the first five, 10 names, will be cyclical businesses. Basically, companies where the screen’s picking up numbers, but nobody in their right mind would trust those numbers given the history and nature of those businesses.
One of the most important things is to not waste your time on those bad companies, but instead skip over them and keep looking. Sometimes we have to skip over a dozen, and sometimes we have to skip over 30 or 40, but we keep working our way down the screen from the first cheapest stock to the 200th. If you keep doing that, and know what you are looking for, that’s how you will come across these gems.
We look at the returns on invested capital of the business, the volatility of historical earnings, the historical growth rates. Those are all great markers to identify a prospect, and then begin our deep research process.
Idea generation is incredibly important, because research is a garbage in, garbage out process. Or maybe it’s better put as quality in, quality out process. By that I mean that if you give a great research team a mediocre stock, at the end of that process you’ll have a well-researched mediocre investment. So, the better you can generate the best ideas, the better your portfolio is going to end up being.
Our research process is a thorough due diligence of the company, a deep understanding of the opportunities and threats, and then we ultimately develop our own projections of future earnings and cash flow. Our portfolio ends up with the very best stocks that come out of that process.
Our turnover is very low. We typically own 33 stocks in the U.S., and a similar number in international, and we’ve averaged about five replacements a year. That’s about 15% annualized turnover.
So, when we go through the screen, we don’t need to find 33 new names every month. We’re just looking for one or two or three names. Sometimes the market gives us four or five. Sometimes it gives us zero, and then you just wait, and if you give it a few more weeks or a couple more months, great opportunities will arise.
You don’t want to force it. But with our low turnover, we don’t need to find a ton of new ideas; over the course of 12 months, on average, we only need to find less than half a dozen.
Mr. Wellington:
Exits are driven by two things. We got it right, or we got it wrong.
We aim to buy great companies at cheap prices. When we get it right, over time they become great companies at fair prices. Our discipline is to buy cheap and sell fair. And so, the highest percentage of our sales, about two-thirds of them, occur when a stock has risen in valuation and has approached our fair value and it’s time to move on.
Additionally, about once per year, on average, we sell a stock because it gets acquired.
And then about a quarter of the stocks we sell are the ones where we got it wrong. This is a humbling business. It’s not a business where you’re going to bat a thousand. You’re trying to get as many right as possible, and at least get several more right than you get wrong, but you’re never going to get them all right. We do get stocks wrong, and when we get to the point where we recognize that we no longer have conviction in the investment, we sell.
TWST: Can you tell us about a few holdings?
Mr. Wellington:
A good example of a holding is Ameriprise (NYSE:AMP). This company is a classic example of a gem amid the junk. But before I get into Ameriprise I should touch on the concept of analyzability.
We have three pillars to our investment approach — Value, Quality and Analyzability. I’ve talked about value and quality already. Analyzability is about owning simpler companies that are easier to understand, and thus easier to get their future earnings right.
Over my 30 years as a value investor, I’ve owned hundreds of cheap stocks. Every stock I’ve ever owned has been a cheap stock, and a lot of them have worked out, which is why I’m still here 30 years later. But a lot of them haven’t worked out. And the only thing that separated the successes from the failures was whether the future earnings came out close to what we expected them to be.
We don’t need to get the earnings exactly right, because cheap stocks have a big margin of safety. When we expected the future earnings to be $10 a share and we got $10, that was great. If we got $12, that was amazing. If we got $8 or $9, that was also really good. But if we got $2 or $1, well, we’ve never owned a stock cheap enough to work with that kind of terrible outcome.
So, you’ve got to get the future earnings right. Everybody does deep research and analysis to try to get the future earnings right. What I think we do differently is, we’re very good at figuring out which companies we have a better chance of getting the earnings right and which ones we have a worse chance of getting the earnings right.
Some companies are much easier to analyze and get right than others; it’s in the nature of their business structure. A very cyclical business whose earnings fluctuate wildly, that’s a lot harder to get right than a business that has very steady and stable earnings.
Now, it’s a lot more common to find a cyclical business than a stable one among the cheap stocks, but the stabler ones are there, they’re the gems, those rare exceptions. And so, we think very long and hard about what kinds of businesses are analyzable or not.
One of the things that really makes Lyrical stand out from most of its deep value counterparts, besides the quality and growth in our portfolio, is that we don’t own any banks. We haven’t owned a bank in over 15 years, and that is because we don’t believe banks are analyzable. The credit risk, interest rate risk and liquidity risk inherent in the business of banking create significant tail risk that is extremely difficult to assess.
Despite not owning any banks, financial services is currently the largest sector exposure in our portfolio. How can we have such a big exposure to financials without owning any banks? Well, we have found many great businesses that provide services that are financial, but don’t have the huge tail risk that banks have, and are thus much more analyzable. And that brings us back to Ameriprise.
Ameriprise is in the wealth management business with about 10,000 financial advisers. They are also in the asset management business, primarily consisting of Columbia Threadneedle. Rather than lending money and hoping to get it back like a bank does, Ameriprise takes in money from clients and charges them fees on it, so it doesn’t have the tail risk that a bank does.
Ameriprise has been a very well-managed company. If you go back to the dawn of the financial crisis, the end of 2007, before the crisis started, they’ve compounded their earnings at over 15% a year since then. And yet, over that same period of time, the S&P 500 has only grown its earnings at about 5.5%.
So, this is a company that’s growing two to three times faster than the S&P 500. And yet the S&P 500 has a multiple of over 20 times earnings, and Ameriprise has a multiple that is less than 14 times earnings. We like getting two to three times the growth of the market for a huge discount.
On the other hand, while it’s had this great growth rate for over 15 years, and it’s had a low multiple for over 15 years. The market has not recognized this great earnings stream yet. So, it’s taken a lot of patience, but just because we haven’t gotten the multiple expansion doesn’t mean that it hasn’t been a worthwhile investment. Because the earnings growth has been so good, the stock has still been able to do very well without that multiple expansion, and we can continue to patiently hold on to it.
Our thinking is that if they continue to grow their earnings faster than the market and their multiple doesn’t go up, well, that’s a good outcome. But the market should be greedy and selfish and recognize that it can get a bargain and get this great earnings growth at a much cheaper price, and that should drive the multiple up over time. And if that happens, then it’s an even better outcome.
Just because re-valuation takes a long time doesn’t mean that’s a bad outcome. It’s suboptimal. You always have a better outcome when re-valuation happens quickly. But this is the importance of owning good businesses that can compound their earnings.
The only way you really lose on any investment is you get the earnings wrong or you get the price wrong. The way you get the price wrong is you pay too high a multiple for a stock, and even if over time you get good earnings, the multiple compresses and hurts your return.
By sticking to cheap stocks we take away the risk of overpaying, and it just comes down to the underwriting risk of getting the earnings wrong.
Mr. Wellington:
Expedia (NASDAQ:EXPE), the online travel agency, is such an example. I think Expedia is another great example of a gem amid the junk. It, too, has grown its earnings at about 15% a year going all the way back to 2007, before the financial crisis.
While Expedia’s earnings grew through the financial crisis, its business was hit very hard by COVID, when you had a complete shutdown of all travel. Expedia makes most of their money through hotel bookings; you also can book car rentals and airfare, but most of the profit stream of the business comes from the commissions they earn booking hotels. And instead of booking hotels during 2020, they had to refund all the bookings. The business went in reverse, and they had some pretty substantial losses.
They were making $8 a share before COVID broke out, and by the start of 2022, two years later, they were back to making $8 a share. So, it was a very resilient business. Currently they’re projected to earn $15 a share. Once they recovered, they’ve had really attractive growth since then.
The stock got cheap starting in late 2022 and into 2023. Today you have a company with really strong earnings growth and a resilient business model. It’s capital light; they just get commissions on booking hotels. They also own vrbo.com, so they have good growth in alternative lodging.
And really, in our opinion, the crown jewel of the business is their B2B business. If you have a travel rewards credit card, and you go to the credit card’s website to book a trip, there’s a good chance that technology is actually being run on Expedia’s platform. That business is growing about twice as fast as the online travel agency and is very profitable.
Expedia today is trading just above 11 times earnings, again for a company that’s growing two to three times faster than the S&P 500 and has almost half the multiple.
Amongst the cheap stocks, there are not a lot of Expedias and Ameriprises. They’re very rare. You really have to put a lot of time and effort into sifting and searching for them. But if you do, the reward is you get these great businesses at very cheap prices. On average, cheap stocks work, despite their low quality. But we believe faster growing, higher quality cheap stocks work a lot better.
Mr. Wellington:
We want to build portfolios without a theme. We just want a collection of the best opportunities. The only theme in the portfolio should be that all the stocks are cheap, and they are all quality companies that are analyzable. Beyond that, we want our stocks to have little else in common.
We never want to own too much of any one thing, because we know we get things wrong, and we never want to get something big in the portfolio wrong. The only way we can guarantee that happens is, we never make anything big.
When it comes to themes in the market, as we go through the screen, we might see that there are a lot of oil stocks that are cheap, or there’s a lot of banks, or there’s a lot of managed care companies. We notice those themes, but just because there’s a lot of something that’s cheap doesn’t mean any of it is any good, and so that shouldn’t be a guide of where it is best to invest. For example, there were tons of banks that looked cheap in 2007, and that was a terrible industry to invest in at that time.
And so, while we observe these themes, we don’t let them influence where we invest. We just try to focus on the next best stock that’s out there, based only on the merits of each stock, regardless of any broader theme or trend.
There are also broad market themes. Recently, the market has fallen in love with AI stocks. But those market themes are typically bad investments, unless you’re in very, very early. They can work well in the short run as the mania builds, but they typically go terribly wrong in the long run.
Investing in what’s hot today does not have a good history of long-term success, because even when you get the earnings right, you’re usually paying such a high price that the stock produces poor long-term returns.
Mr. Wellington:
We get this question a lot — when do value stocks do well? And the answer is, value stocks have done well in every possible kind of environment. There are decades and decades of data on value stock performance. Value stocks have done well in high inflation and low inflation, high interest rates and low interest rates, rising rates and falling rates, and good economies and bad economies. You can find examples of value stocks doing well in all of those environments.
The more interesting question is, when do value stocks not do well? That’s much easier to narrow down. Value stocks have not outperformed all the time, notably, they did poorly in the tech bubble, they did poorly in the global financial crisis, they did poorly during COVID, they did poorly in the AI mania of the last couple years.
So, it's easier to find when value stocks don't do well, and when value stocks don't do well is when there's a mania or when there's a panic.
When there’s a mania like the tech bubble and stocks rise to crazy high multiples, value investors don’t own them, and while those expensive stocks go higher and higher, value stocks underperform by comparison.
The other time value stocks do poorly is in panics like the global financial crisis or the first weeks of COVID, when everybody thinks the world’s going to end. The cheapest stocks become the ones the market overreacts to most, and they tend to keep going down until the panic ends.
But outside of those extreme environments, outside of those rare circumstances, value stocks have either outperformed or performed in line most of the rest of the time.
Mr. Wellington:
What we do to best position our portfolio is sift through the best bargains, the cheapest stocks in the market, and try to pick out the very best companies. Trying to own the best companies at the cheapest prices is the right formula for every environment. It doesn’t mean it works in every environment in the short run, but I can’t think of an environment where you wouldn’t want to do that.
Most of the narrative written about value investing is based off the performance of the large-cap value benchmarks like the S&P 500 Value and the Russell 1000 Value — and I can’t emphasize enough that they are horrible proxies for value stock performance.
One simple way to understand this is, if you look at a Russell 1000 Value fact sheet today — this is as of the end of April — there’s a thousand stocks in the Russell 1000, and there’s 869 stocks in the Russell 1000 Value. So, the Russell 1000 Value does not own the cheapest 200 stocks of the Russell 1000. Or rather, it does own those 200, but it also owns an additional 669 stocks, such that the performance of those cheapest 200 is diluted away.
With 869 stocks, that means that 87% of the names in the Russell 1000 are in the Russell 1000 Value. It’s really a core index that owns everything except for about 130 of the most expensive stocks in the market. It’s not focused on the cheapest stocks. And as a result, over time, it’s been a really poor proxy for the returns of value stocks.
Value stocks have done really well. Value stocks outperformed coming out of the financial crisis, but the value indices underperformed. Value stocks have been outperforming for the last five years, even with the Magnificent Seven; value indices have been underperforming the last five years.
And yet, every story written in The Wall Street Journal, every guest on CNBC — except when I’m on — talks about value based on how these value indices are performing. And when they’re trying to talk about what’s wrong with value, there’s nothing wrong with value stocks, but there’s something seriously wrong with the value indices.
It is not a well understood story that these value indices are such poor proxies for value stock performance, and especially with the rise of passive.
I don’t have a problem with passive. Passive is a good investing option for investors who can’t do good research and diligence on managers. It’s OK to invest in passive. It’s more than OK to invest in large-cap value stocks. But it has not been OK to passively invest in large-cap value indices. You don’t get your value from an index. Passive and large-cap value has been a terrible choice.
Mr. Wellington:
As terrible as the value indices are, there is really nothing better out there. If we do a good job at being fundamental, bottom-up stock pickers, we should be able to beat any benchmark over time.
The benchmarks all have their pros and cons. The S&P 500 is very mega-cap growth focused, so it’s not really an apples-to-apples comparison. That said, if our approach can’t beat that, then what’s the point of doing our approach? We’ve still got to beat the S&P 500 over time.
Given the skew from the Magnificent Seven, another index to look at is the S&P 500 Equal Weight. It’s not distorted by market cap weighting, but it still is a mix of value and growth stocks and a lot of core stocks. So, it doesn’t capture our value style, but at least in environments where mega caps are having an outsized influence, it gives you a clearer view of how the average stock is performing. So, we look at that index a lot lately.
The large-cap value indices, despite their deep flaws, still serve a purpose. Those indices are what all of our institutional investors compare our returns to. We’re not going to follow those indices, and no one should want us to follow them given how poorly they’ve performed. Our clients know we’re going to have high active share and high tracking error. They know we see tracking error differently. Given the poor performance of the large-cap value indices, we see tracking error as the error of tracking those indices.
I’d also add that if an active product tracks very closely to some passive index, then you don’t need that active manager. You can replicate what they’re doing much cheaper through some passive construction. There’s a fundamental issue: A good active management style captures something that isn’t available in an index or a benchmark. That’s exactly one of the ways active managers add value, by bringing something different to the portfolio that can’t easily be found elsewhere.