Investment Philosophy

Importance of ROIC Part 1: Compounders and Cheap Stocks

Update: For those interested, I wrote a 5-part series about the concept of return on incremental invested capital and also a few other writings that discuss ROIC

A while back, I posted a couple articles on return on invested capital (ROIC) along with some comments on Joel Greenblatt’s Magic Formula. These posts attracted a lot of comments and email questions, and so I wanted to post some more thoughts on the topic of compounding generally, and maybe ROIC more specifically. Here are some links to posts that are somewhat related to this topic:

In this post I want to address a comment that has been made by numerous people regarding Compounders vs. “Cheap Stocks” and how some people have a tendency to completely fall into one camp at the expense of the other. The next post I’ll list some examples of why ROIC is crucial for long term business owners and shareholders.  I’ll also have a post or two discussing the simple math involved with returns on capital and how that relates to compounding.

Valuation vs. Return on Capital–Depends on Time Horizon

First off, one commenter pointed out that various studies have shown that ROIC is a metric that doesn’t actually add much value to returns. In fact, these studies say that better results could be had by simply just buying cheap stocks (low P/E, low EV/EBIT, low P/B, etc…). Forget the quality (return on capital), just focus on valuation (low price to earnings).

I believe this to be true–BUT, with a very important caveat. One very crucial point is often left out of these studies…. Holding period. Most of these studies pick a group of stocks based on some value measure (low P/E, etc…) and then after 1 year (or sometimes 2 years), sell those stocks and replace them with a new set of stocks that match that valuation criteria. Most of the studies turn their portfolios over once a year.

And I think many value investors see these studies and get excited about valuation, leaving quality far behind. Who needs quality when valuation is all that matters?

Here’s my take: I completely agree with these studies and with the practitioners who favor valuation over quality if their holding period is only 1-2 years. On balance, paying a high price for even a great business will not always work out well if you have to sell that business in 1-2 years, or even 3-5 years. But if you plan to hold your stocks for longer periods of time… 5 years, 10 years, or longer, then quality becomes much, much more important than valuation.

Why Does Wells Fargo Beat Everyone Else?

I did a post on Wells Fargo a few months ago where I discussed how WFC’s long term business results trounced all of the small community banks’ long term results, and the WFC stock price had the same outperformance. WFC’s high return on equity (15%+ over long periods of time) was the most important factor in creating wealth for shareholders over time. A bank that produces 15% ROE will always result in better shareholder returns over a bank that produces 7% ROE, given enough time.

Of course, buying a low earning community bank at 50% of book value can work out very well if/when the market corrects itself and the stock gets revalued at 100% of book value. But over 10-15 year periods or longer, paying 2 times book for a business like WFC will nearly always work out better than paying 0.5 times book for a low earning community bank.

The same goes for these studies that look at P/E ratios (valuation) and ROIC (quality). Over short periods of time, paying low P/E ratios or low EV/EBIT ratios will work very well, as the market typically corrects itself over 1-3 years or so. But over time, if you intend to participate in the long term results of your business and own the stock for 5-10 years or longer, you should be much more concerned with the quality of that business.

This doesn’t mean that one category should be focused on with complete abandon for the other. It just means that both play a role, especially in long term results.

I think of the entire debate as follows: my ideal investment is the Fastenal type business that can produce high returns on capital and reinvest large amounts of retained earnings at similar high rates of return (when I discuss ROIC, what I’m really looking for is what return a business can generate on its incremental investments in the future–more on this in the next few posts). The result for shareholders of FAST has been north of 20% annually for the past 25 years or so.

FAST Long Term Chart

If one could only know in advance the next FAST, life would be simple, right?

We can look for clues for the next FAST, and that includes looking for businesses with simple products and simple models that can invest large amounts of capital at high rates of return.

The problem is that capitalism is tough, and unpredictable things can occur that can harm a business’ economics. So I rely on valuation as my safety net. I prefer to find really cheap stocks, but I want them to be businesses that I think can grow intrinsic value. This provides me with a margin of safety not just in the valuation, but also because the gap between price and intrinsic value will widen over time as the business continues to grow its value.

So as I mentioned in the previous related posts, I want to have my cake and eat it too. I like quality businesses, I like great capital allocation, I like high returns on capital, but I demand value (as Buffett once said).

Compounders Come in All Shapes and Sizes

One of my current holdings is a small bank that I purchased at around 60% of tangible book value and a P/E of around 7. There are many opportunities to own small banks at these metrics. The difference with this particular bank is that it has grown its intrinsic value by somewhere between 8-10% per year by my estimation. Book value has averaged over 9.3% annual growth since 2000 without a single down year, and the bank pays out a steadily rising dividend as well (it hasn’t cut, lowered, or missed a dividend payment since the 1920’s when it was founded, even through the Great Depression!). The bank is also extremely risk averse, and capitalized at around 17% equity to total assets.

These types of situations are rare (both cheap and high quality), but they exist, and they are the type of ideas I’m looking to find for our portfolios. These types of businesses are compounders.

By the way, many seem to associate “compounders” with these mega cap quality businesses that Buffett buys (Coke, Walmart, Exxon, JNJ, P&G, etc…).  Numerous people have commented lately that you can’t find compounders cheap except during market crashes. But there are many small, relatively unknown, quality businesses that are quietly growing their intrinsic values per share at 10%-12% per year that you can occasionally buy quite cheaply.

I know of one such business that has a market cap of under $250 million that has grown its intrinsic value per share (by my estimate) at a rate of around 15-20% per year for the past 25 years. In fact, the stock is up more than 50 fold in that time, a compounded annual rate of return of about 17% per year before factoring dividends.

You might ask: how could it have compounded at that rate and still be so small? Well the short answer is it was very small 25 years ago, but in their annual report they discuss the incredible amount of capital that they’ve returned to shareholders via buybacks and dividends over the years. Basically, the company sold shares initially for a grand total of $1.7 million in proceeds, and that is the only the time the company has ever issued stock in its history. Since that time, the company has returned around $130 million to those happy original shareholders via buybacks and dividends, and the stock itself is 50 times more valuable than it was then as well. The company is small and doesn’t necessarily have huge reinvestment opportunities (thus the large buybacks and dividends), but it’s a great business with a strong niche and great management, and it has created enormous wealth over time for its long term owners.

Businesses like this are out there, and occasionally they can be purchased well below the conservatively estimated present value of their future cash flows. Occasionally it’s a no-brainer. Those are the situations we’re looking for. Like the bank I mentioned with long term management, very stable business model, steadily growing book value, all for a price that represented 60% of tangible book and a fraction of estimated intrinsic value.

It doesn’t mean these ideas work out each time, but given enough of a sample space and enough time, they are the situations that collectively aggregate to create a low risk portfolio with a high probability of achieving an above average rate of return.

So there you have it. Ideally, I want high quality… but since I’m unsure of the accuracy of my crystal ball (I don’t like predicting the future), I don’t want to pay much for what has heretofore been a quality business.

This sets up what academics call “asymmetric upside”. I prefer a more simple explanation—one that I stole from Mohnish Pabrai: “Heads, I win… Tails, I don’t lose much”.

I’ll have a few more posts discussing this topic along with the math behind it so we can better understand its importance, and more practically, be able to identify these types of qualities in our own prospective investments going forward.


John Huber is the portfolio manager of Saber Capital Management, LLC, an investment firm that manages separate accounts for clients. Saber employs a value investing strategy with a primary goal of patiently compounding capital for the long-term.

John also writes about investing at the blog Base Hit Investing, and can be reached at

20 thoughts on “Importance of ROIC Part 1: Compounders and Cheap Stocks

  1. Any hints on the two compounder examples? Outside the U.S.? And do you exclude stock options from the “the only the time the company has ever issued stock in its history” statement?

  2. John, great post. Completely agree with you that it doesn’t make sense to be dogmatic in choosing compounders vs. cheap companies. If you can find one or the other at different points in the cycle, that’s great…if you can find a company that combines both, that’s fantastic.

    Question for you…any chance you’d disclose the names of the two companies you mention in the last section?

    1. PSD and Anon,

      I’m reluctant to provide the specific companies because I’d prefer the prices going down. I mention them to provide a few general examples of small companies that are building value, but these two aren’t unique. There are others, and they often are off the beaten path. I may get more specific on a few investment ideas or current holdings, but the ones I talk about are usually quite liquid so there is no impact on my partners. As for the small ideas, we’re always trying to buy more, and so we need the prices to be going down–and not up!

      Thanks for reading!

  3. Great post. have you done any work testing the high ROIC theory? It makes perfect sense on paper, but if you were to buy high ROIC stocks 20 years ago and hold them until now, would you have compounded capital faster than the market?

    1. Hi Admirer,

      I haven’t done any tests like that. Basically, I don’t really invest using backtests or formulas, although some investors have done that successfully. To me, I just try to stick to the common sense principles. I consider buying good companies at cheap prices to be logical, and each investment is different. Also, ROIC can be calculated differently depending on the specific company and the specific industry the company is in. So to me, there wouldn’t really be an effective blanket rule to follow that says: “ROIC above X% is good”, etc… I’m really analyzing the specific dynamics of each business model and trying to determine the cash on cash return that the company can achieve on the capital that it retains and reinvests.

      Thanks for the comment.

  4. Love the column and agree BUT how much then is too much to pay for these great compounders? I fear if one gets too focused on quality you start to forget about the price — wasn’t that the fallacy of the Nifty 50?

    Peter Lynch’s PEG ratio comes to mind here. Is that ratio one you observe?

    Many thanks!

    1. Hi A in P,

      Yes, everything boils down to value. Price is what you pay, value is what you get. So one can’t (or shouldn’t) forget about price as it directly impacts the amount of value you are receiving per dollar invested. So quality is a part of value–it’s joined at the hip as Buffett said in one of his letters. Regardless of how good a company is, there is a certain price that if paid will only produce mediocre or worse results. And the key thing to remember is that quality is one aspect of value.

      Businesses that are growing intrinsic value over time (compounders) in general are attractive companies to analyze because their intrinsic value is growing, and that means the gap between price and value grows over time, thus giving you room for error if your estimate of value is off.

      But certainly, great companies can get overvalued, just as mediocre companies can get undervalued.

      The main concept I want to convey is that over time, ROIC is the most significant factor in long term business returns. In extreme circumstances (1972 Nifty 50, Coke 1998, etc…) these compounders can get to levels that are destined to yield subpar returns over long periods. In most situations, the higher quality businesses will yield far better results than the lower quality businesses given enough time.

      But as I say in my piece, I’m not interested in just paying any price for great businesses. I’m interested in finding huge gaps between price and value.

      More thoughts to come… thanks for reading!

  5. Hi John,

    Thanks for the post. Would like to know how you determine intrinsic value and how can we screen for these kind of stocks? Thanks.

  6. Hi John,

    Great post! So how can we determine the intrinsic value? And how can we screen for these stocks that grow intrinsic value?

    1. Noob and Only,

      Thanks for the questions. I’ve had a lot of questions on the computation of intrinsic value. Basically, I think Buffett explains it best in his Owner’s Manual (I linked to this and reference his definition in my previous post). Basically, it’s the present value of all the future cash that can be extracted from the business.

      1. The problem with present value computation is we would have to forecast future cash which is subjected to a lot of assumptions. So how can we bypass that?

        1. Well I wouldn’t say you can bypass it, but you’re right, it is easy to fall prey to the “DCF model”. I don’t use spreadsheets or models when I’m doing valuation work (I use spreadsheets to compile past years data for viewing, but I never use them to predict or assume future data).

          Some use these types of models and can do well, but I’ve never had success with them.

          However, it’s important to keep in mind (as Graham, Fisher, and Buffett would all agree) that the value of a business is the present value of the future cash that you can extract from it (either from a sale, liquidation, or most likely the cash flow it generates).

          So that is an accurate way to think about value, but the problem is that using precise models make it very easy to justify values that you are looking for.

          The best thing to do is think very simply. Try to find large, obvious gaps between price and value. Don’t worry about being so precise. If it’s not obvious, then it’s not undervalued enough. Look for something that’s trading at 10 and is worth 20. Maybe you’re wrong and it’s worth 18 or 16, or 22, but find something you are fairly certain it’s worth much more than 10.

          1. Thanks for the reply. It goes back again to how can we determine that it’s worth much more than 10. Other than DCF, is there other ways we can use to determine it’s intrinsic value? peer comparison of valuation ratios?

          2. Well you can look at the value a private buyer would pay. But even in that type of a valuation, you’re still going to be looking at the cash in-cash out method. That’s really how you can determine what a private buyer would likely pay. But intrinsic value could be defined very simply as the price that a rational private buyer would pay for the business.

            I sometimes look to see what comps are trading at, but I’ve never been a big fan of relative valuation. What if the comps are overpriced? What if the general market is expensive? I don’t think you can judge value very well by what other things are trading at. I’m not saying that you shouldn’t notice what the comps are valued at, but I never make decisions based solely on that information. Occasionally you might find assets that a business owns that are either hidden or undervalued in the market. Sometimes those create interesting special situations where the company can sell or spinoff parts of the business. So each situation is different, and needs to be analyzed differently.

            In the end, I try to think about it simply. Reread Buffett’s latest annual shareholder letter about his real estate investments. He was looking for 10% pretax returns on those investments, and he wanted that to grow over time. So his 10% yield on cost grew each year. That’s a nice way to think about a business. Look at the current earnings (or normal owner earnings), and ask yourself if you can see those earnings being higher than they are now in five years. You don’t need any models or spreadsheets. Just figure out what the normal earning power of the business is, and decide how much its worth to you. You don’t need to know if the growth rate should be 8% or 12%. Just determine the current earnings yield and make sure you have an understanding about the future prospects of the business to determine if you think those earnings will be higher or lower over time. Try to find situations where you can easily see the gap between the current price and your estimate of fair value. Try to think like a business person and look past the current quarter and the current year, and think in terms of downside first. When a business person thinks about risk, they think about the competitive position of the business, the competition, the future prospects, and the management of the business.

            It’s somewhat of an art form, and each investor will come up with different intrinsic values when given the same set of facts, but occasionally, you can spot drastically mispriced merchandise. Like Buffett said regarding his BAC investment, he bought it because he “saw a large gap between price and value, not because he calculated some specific P/E ratio”. When you see a stock trading at $8 with somewhere between $1 and $2 of earning power that will likely grow over time, that’s likely a good bet. You don’t always get such extreme situations, but the idea is to combine current earnings yield (current valuation) with common sense logic about the business to determine if the earnings will grow over time. Just like Buffett’s apartment building where he knew he was getting a 10% yield up front, but he also figured that over time, rent would likely increase as prices rise and occupancy increases. Same with the farm: 10% current yield, but he figured that corn prices and productivity–although volatile–would increase over the long term.

            Hope that helps…

  7. Great article as always.

    Something I’d add is that when you have intrinsic value growth that also compounds with multiple expansions. Although, this works both ways. As way of example, if your investment’s intrinsic value grows 50% and the multiple expands 50% the market value of your investment is now up 125% not 100%. On the flip side if value contracts but your multiples expands (p/e, p/b, ev/ebit, etc…) you still might be down. For example if value is down 50% but the multiple is up 50% your still down 25%. So when you buy a good company at a cheap price you get a positive compounding effect (hopefully!).

    Also as you allude to, when you buy a cheap and good company you get two margins of safety. One with the business and the other on the valuation. Unfortunately, cheap and good stocks are harder to find than cheap or good.

    1. That’s right Chris. Yes, multiple expansion or contraction matters, especially in the near term. In many cases, I try to assume the same multiple (assuming it’s a fair multiple), and I prefer to get my results through the internal results of the business. I try to buy cheap so that multiple expansion provides the added kicker that you speak of.

      But over time, the results of the business are the most important (assuming you’re not paying extreme prices).

  8. Thanks for posting these thoughts, they are very helpful. Echo much of what The Brooklyn Investor has been writing about, especially with regards to how Buffett thinks about normalized earnings and wants to get a 10% pre-tax yield from day one.

  9. By chance is the bank NASB? Been following it a bit myself and seems to match the stats you gave so I was curious if it might be the same.

  10. Pingback: Valuing Growth

Leave a Reply

Your email address will not be published. Required fields are marked *