Someone asked me why EV/EBITDA matters. Why shouldn’t we just look at a stock’s free cash flow yield? Isn’t that closest to a discounted cash flow approach? Isn’t that closest to Warren Buffett’s owner earnings approach?
The answer is yes. Free cash flow divided by market cap is the best measure of your hold return. But most investors will not hold a stock from now till Judgment Day. They are buying at a certain price. And they are selling at a certain price. If the multiple – P/E, EV/EBITDA, etc. – they are buying at is far enough from a normal multiple, the change in the multiple (up or down) over even a holding period as long as 15 years may add or detract 3% to 4% annually from their returns. Over shorter holding periods – 10 years, 5 years, 3 years, etc. – the headwind or tailwind caused by buying at an especially high or low multiple can completely erase the hold return you earn by having good free cash flow while you own the stock.
An EV/EBITDA approach captures two ideas: price and leverage. It is like a leverage-adjusted P/E ratio. This is important because two things that will have a big influence on your return in a stock are whether the stock’s P/E rises or falls from the time you buy it to the time you sell it and whether the company has more debt or less debt when you sell it. Ideally, you want to always buy a stock you will sell at a higher P/E than you bought it at. And – although fewer investors think of this – you want to buy the stock at a lower debt level than you sell it at.
The reasons for this are obvious. Both P/E expansion and added leverage increase your annual returns, while both P/E contraction and subtracted leverage decrease your returns.
Other things equal, you want to buy good merchandise that is low P/E and low debt. And then you want to sell that same good merchandise when it is high P/E and high debt. You also want to collect a good return in the meantime.
The problem Warren Buffett-type investors and Ben Graham-type investors often have in communicating is that one is talking about “hold” returns while the other is talking about “sell” returns. The Warren Buffett investor wants to keep a cow and milk it. The Ben Graham investor wants to sell a cow for more than he bought it for. Both approaches work. And they work very well together.
The EV/EBITDA ratio can help keep an eager Warren Buffett type grounded. While you can make a lot of money in a good company over time – the key phrase is “over time.” There is a reason Warren Buffett made his best annual returns – in the 1950s – when he invested in the cheapest stocks he ever bought. This Ben Graham-type approach can generate returns faster. Of course, some of Buffett’s best early investments – like GEICO – bordered on a combination of high quality and low price. His later investments – like Heinz – will not generate as good annual returns. That’s because the buy price does not allow for a good “sell” return and Buffett is no longer interested in selling.
Great stock returns (20% to 30% a year) almost always involve at least some sell return. They are either a Ben Graham approach or a combination of Buffett and Graham – like Buffett’s early days. Buy and hold investing works. But it is most likely to generate consistently good returns (10% to 15% a year) rather than great returns.
I hope the lesson you learn from reading this article will not be that the later Buffett approach is best or the Graham approach is best. I hope you will learn that if you pay attention to business quality at the same time you pay attention to EV/EBITDA – if you think both about what would happen if this stock were sold in a couple years and what you would earn if you never sold it – you’ll find ways to make sure your investments are good enough to qualify in some sense as both value and quality investments.
And even if you choose to pick stocks purely on value or purely on quality – I hope you will not be blind to the influence each has on your annual returns over different holding periods.
Let’s look at a good, overleveraged stock. I think this is an attractive buy and hold investment right now. I also think the EV/EBITDA ratio – recently around 9 – is not an obvious bargain.
A great example right now is Weight Watchers (WTW). I think it has good 10-year or so return potential. However, EV/EBITDA is not high. In fact, in thinking about the company one of the 3 risks I identified that could cause future returns not to be so great is a buyout offer. In other words, the control owner (a private equity company) could offer to pay something like a 25% premium to the stock price and they could get a fairness opinion, shareholder support, etc. Now, in all actuality, I think the 10 to 15 year return potential of WTW would actually be quite good even if the stock was 25% more expensive. And yet I don't think anyone would say such a buyout was being done at a high price.
So, EV/EBITDA is not important to me as a buy and hold approach. It is important in negotiated transactions. That is how I use it. The same could be said at John Wiley (JW.A), IBM (IBM), Dun & Bradstreet (DNB), Birner Dental (BDMS), etc. In those cases, I expect 10 to 15-year returns in the stock to be better than their EV/EBITDA suggests. But this is caused by capital allocation. I think they can generate free cash flow and get good returns on free cash flow. In some cases, they may be able to grow 5% a year and buyback 5% of shares and pay a 2% dividend all at the same time. Now, really, that gets you to good returns even without any multiple expansion if you hold the stock.
One way to think about this is that different sources of value have different lifespans. A dividend yield is fairly constant. That's why people like - and maybe overvalue - it. If you buy a stock with a 5% yield it is likely to pay you the same - or slightly rising - cash return as a percentage of your purchase price each year. So, if we broke down returns over possible investment time horizon we'd see:
1 Year; 5%
3 Years: 5%
5 Years: 5%
10 Years: 5%
15 Years: 5%
Or we might even see some rise in the dividend over time. So, annual returns might even get a little better over long holding periods. However, the tax on dividends and the relation of the dividend growth rate to our discount rate is likely to make the after-tax present value of such increases pretty flat. So, a dividend is mostly a stable return across the holding time spectrum result.
Of course, this depends on how fast the dividend is rising. But the dividend growth rate will often be something you can analyze through your expectations for the earnings growth rate. In other words, dividends may add a lot of value over time. But that doesn’t mean you need to analyze the current dividend very carefully. It may make just as much sense to analyze the ability to pay future dividends rather than the tendency to pay dividends now.
What about multiple expansion or contraction? It might be worth 30% if the multiple expands fully (from say a P/E of 12 to 16) this year. I'm looking at a stock right now - Weight Watchers (WTW) - that I think should have a P/E multiple (remember, it's leveraged up at the moment) about two-thirds higher than it is right now. So, over a one-year holding period this multiple expansion (from a P/E of under 10 to a P/E of 16) would add about a 65% annual return. However, over a 15-year holding period it would add only 3.5% to my return each year.
This is where the P/E ratio, EV/EBITDA, etc., come in. If you are buying something to sell it later, the two numbers that matter are how much more you will sell it for (future multiple/current multiple) and how many years you hold the stock for. The goal in this kind of undervaluation exploitation is to buy something that will one day trade at a high multiple of what you are paying today and will trade there soon. As I explained, returns from this source of value can vary from 65% in one year to 3.5% over 15 years. So holding period matters.
This is a "trade" approach. Even when we are talking as long as 15 years, it's still a pure "trade" approach. It's just a very long-term trade. But you aren't expecting to make money holding the stock - the way you do with a dividend - you are expecting to make money selling the stock.
You are clearly an investor who thinks in terms of your "hold" return. That is why EV/EBITDA sounds so strange to you. It's not a critical variable in calculating your return in holding a stock. In fact, if you hold a stock forever the importance of the EV/EBITDA you originally paid is very slight. I say that as a value investor. But it's true. The longer you hold a stock the less important the price paid becomes and the more important the quality of the business becomes. What really becomes hugely important is capital allocation. You focus on good (or at least durable) businesses with high free cash flow that must be allocated somewhere outside the business. That makes you a hold-style investor.
There are people who have made 10% to 15% a year as true hold investors. If you are very, very selective you can make 15% a year as a true buy and hold investor - with a focus on the hold. It is hard to ever make more than about 20% a year using this method because companies themselves very, very rarely have the opportunity to earn returns on equity of 20% while growing retained earnings. The one exception - and this seems to be an area you pay attention to - is companies who can grow free cash flow without adding much (if any) shareholder equity to the business. This creates a dual - or even tri - stream of returns. You get growth. You get buybacks which create more per-share growth. And you get dividends. Sometimes you get multiple contraction or expansion.
The best scenario for - even pretty long-term - great returns is free cash flow growth that requires almost no equity growth coupled with a buyback and bought at a low multiple. Imagine a company trading at 8 times earnings. Now, imagine it can grow 5% a year without increasing equity used in the business at all. Now imagine it buys back 5% of its shares every year. Let's look at what will likely happen to this company as a 15-year investment.
Income Growth: 5%
EPS Growth: 10.3% (caused by combination of income growth and 5% buyback which becomes 5% plus 5.3% due to arithmetic of declining denominator)
P/E: Goes from 8 to 16
The result is a return of 15.5% a year over 15 years. In this case, the P/E was 8 which suggests a return of only 12.5% on an earnings yield basis. Overall growth was just 5% which suggest only a 5% return. Even counting overall EPS growth, we only get to a 10.3% return which is far below the 15.5% actual expected return in the stock.
And I’m lying about 15.5%. That's a bit under what the return is likely to be. Can you tell why?
Notice that I said the company would buy back 5% of its shares a year and grow 5% while requiring no added equity. Well, that means free cash flow is left over every year. In fact, in the first year there will be 7.5% of your purchase price just lying around if the company only buys back 5% of its shares.
This is critical because it means the company can pay a 7.5% dividend. So, your actual return in the stock over 15 years would be closer to 23% a year if you paid 8 times earnings for a 5% organic grower that needed no added equity to grow and that paid a 7.5% dividend and bought back 5% of its shares each year.
Although 23% earned over 15 years sounds shockingly high - my math isn't wrong. It's just that the market would have to be very wrong to offer a business like that at a price like that. It rarely does.
Growth and multiple expansion can be seen as seller returns. They are returns you get when you sell the stock. So, we can model the result of paying 10 times earnings for a 5% grower and selling at 15 times earnings after 15 years as $1 * 1.05^15 = $2.08; $2.08 * 15 = $31.20. You will sell the stock at $31.20. I said you bought it at 10 times earnings, so you paid $10 for the stock. Over time, a rise of $10 to $31.20 (all in capital gains) over 15 years is an annual return of 8%.
So, that's all the gain you'd get out of a 5% grower with a P/E of 10 that eventually sells for a P/E of 15.
That is a seller return. It's a capital gain. But once you buy a stock you aren't just a seller. You're also a holder. Investors make money by holding stocks and selling stocks. Often, the hold return of a good long-term investment is high. Often, the very best long-term investments combine hold return with sell return.
In the 23% return example almost 13% came from the hold value. The other 10% came from the sell value. This is where EV/EBITDA comes in. Let's imagine you know a company should - at the end of 15 years - trade at a P/E of 16 while having no leverage.
What if the company is levered up right now? Or what if the company has surplus cash?
That's simple. Your hold return changes. An overleveraged company has to use up some of your hold return - not buyback stock and pay dividends - to instead pay down debt. Meanwhile, an "anti-levered" company can do the reverse. If a company has $300 million of debt and its cash flow can support $300 million of debt - it's possible to return $600 million to shareholders. We can pay a special dividend, pay regular dividends, buy back stock, etc.
Good (contrasting) examples of this are Q-Logic (QLGC) and Weight Watchers. Q-Logic is anti-levered. It has cash it is trying to find a way to get rid of. Weight Watchers is leveraged. It has debt it is trying to pay down. Q-Logic's return will be higher than its free cash flow to the extent it also pays out surplus cash. Weight Watchers' return will be lower than free cash flow to the extent it pays down debt.
By the way, I think both stocks are better investments than the S&P 500. And I think both companies are worth your research time. One is more than double the price of the other on an EV/EBITDA basis. However, I'm not at all sure - over a long-term holding period - that the cheap stock (Q-Logic) will outperform the expensive stock (Weight Watchers). Actually, over 15 years, my money would be on Weight Watchers outperforming both the S&P 500 and Q-Logic - and probably most other investments you can find. This is despite WTW having a non-value EV/EBITDA ratio.
Why do I think this? Because I think WTW will grow 5% a year over the next 15 years. I don't think it will need added equity to do that. I think it will never completely deleverage. It's controlled by a private equity firm. And I think they'll pay down half their debt and then releverage again through a stock buyback. I think these stock buybacks can cause share count to drop 5% a year. And I think the stock can end up trading at a P/E of 16. If that all happens, you're looking at a 14% return over 15 years (ignoring dividends). In other words, I think Weight Watchers could be a $280 stock in 2028. It's a $41 stock today.
Weight Watchers is not a value stock. It is not a seller stock if you expect to be around for a year or two. It is a holder stock. A lot of free cash flow-focused investors are really hold investors.
I tend to prefer situations where I make money holding and money selling. I want to pay 75 cents for a dollar that compounds. I want the dollar to compound as good as any other dollar out there. But I don't want to spend a dollar for that dollar. I want to spend 75 cents for that dollar.
The truth is that - even leveraged up - Weight Watchers is the kind of business that could still compound as well as the market will compound your money even at 16 times earnings. Even with debt and a P/E of 16 it could do fine. One reason for this is that it can handle the debt. The other reason is that it can grow while returning its cash flow. This combination makes it an adequate compounder even at 16 times earnings.
However, I would rather pay less than 10 times earnings for something that is valued relative to other stocks probably worth 16 times earnings and valued on a pure DCF basis probably worth 20 times earnings. This creates a seller return in the stock. It avoids just depending on a holder return.
How important is EV/EBITDA if you are a buy and hold investor buying a controlled company?
Not very. A buy and hold investor doesn't want to sell. And a control owner can block takeover attempts and can use unorthodox capital allocation. I like situations where you have a good, controlling capital allocator. They are the best stocks to hold. In those cases price-to-free cash flow - fully levered - is what matters.
Most stocks don't match those circumstances. Biglari Holdings does, Berkshire Hathaway does, maybe Weight Watchers does, BDMS seems to, maybe ARKR does, and so on. Some corporations are not controlled but do have an obsession with total shareholder return-type approaches - see IBM, DNB, etc. Plenty of companies do this. They just aren't public.
What you're describing in focusing on price to free cash flow and capital allocation is what a control owner - like a private equity holder - would focus on while they held the stock. However, in buying and selling something (taking and giving control) even these kinds of owners will look at EV/EBITDA. So it makes sense to consider EV/EBITDA as an entrance and exit measure even if it has little to do with the return you get while holding the stock.
The return you get from holding a stock and the return you get from selling a stock are both important. Don’t ignore either of them. Always think about both.