Reinvestment Moat Follow Up: Capital Light Compounders

Posted on Posted in Education, Investment Philosophy

This post is the second guest post by my friend Connor Leonard, in what I hope to be a somewhat regular “column” here at BHI (by regular, I mean as often as Connor decides to put the proverbial pen to paper and share his insights with us). Based on the quality of his work, he’s welcome back anytime. Connor and I live in Raleigh, NC, and get together regularly to share investment ideas. I encourage you to reach out to us if you’d like to meet up sometime for coffee. My contact info is on this site, and his contact info can be found at the bottom of this post.

What follows is what I’ll call “Part Two” to his excellent post on Reinvestment Moats vs. “Legacy Moats”.

Here is Connor’s post:

Reinvestment Moat Follow Up

A couple of months ago John invited me to contribute a guest post to Base Hit investing (link) where I discussed the difference between Legacy Moats and Reinvestment Moats. While I encourage you to read the post for the full explanation, below is a quick summary:

Low/No Moat: The typical business you encounter during the day likely falls into this bucket, such as your average convenience store or insurance agency. These are perfectly fine businesses and likely provide employment within the community and a solid product or service to customers. However without a sustainable competitive edge it will be difficult to earn exceptional returns as an investor owning a Low/No Moat business unless you time the entry and exit well. Specifically the game plan has to be to buy at a discount (say $.50 on the dollar) and exit at around fair value ($.95 – $1.00) in a relatively short amount of time[i].

Legacy Moat – Returning Capital: These businesses have an entrenched position within current markets that enable strong and consistent profitability relative to the prior invested capital. However there are few opportunities to deploy incremental capital at similarly high rates, so the management team decides to distribute the majority of the earnings back to owners at the end of each year. This is a prudent move by the management team, and essentially turns the company into a high yield bond. Many “wide moat” companies such as Procter & Gamble and Hershey’s successfully follow this strategy, distributing 80%+ of annual earnings out as dividends. While this investment profile is adequate for many, if you are aiming to compound capital at 15% – 20% rates it likely will not come from owning this kind of business over a long stretch.

Legacy Moat – “Outsider” Management: Here you have a business with all of the characteristics of a Legacy Moat, but the management team decides to retain all of the capital and deploy it into new businesses through a focused M&A program. The home office effectively serves as an internal private equity fund, using the permanent capital supplied by the operating companies to fund a disciplined acquisition effort. When the right businesses are paired with an exceptional capital allocator, the result can be remarkable compounding of shareholder capital such as Berkshire Hathaway (Buffett), Tele-Communications Inc. (Malone), and Constellation Software (Leonard).

Reinvestment Moat: This is the rare company that has all of the benefits of a Legacy Moat along with ample opportunities to deploy incremental capital at high rates within the current business. In my opinion this business is superior to the “Legacy Moat – Outsider Management” because it removes the variable of capital allocation: at the end of each year the profits are simply plowed right back into growing the existing business. This is the purest form of a “compounding machine” and when combined with a long reinvestment runway the result can be a career defining investment. Examples listed in my last post include GEICO, Walmart, and Amazon.

Following my initial write-up I noticed some questions and discussion around a fourth type of business: companies that can grow revenue and earnings without requiring additional capital. In this follow up post I thought it would be useful to discuss the characteristics of these “Capital-Light Compounders”, the playbook for how they should be run, and some current examples. Consider it an addendum to the original write up:

Capital-Light Compounders

As an investor I’m constantly looking for businesses that I believe can increase intrinsic value per share at a high rate over a long period of time. As John outlined in a recent post (link), a simple formula for estimating the rate of increase in intrinsic value is:

Connor Clip 1 Revised

This makes sense, if a company keeps 50% of earnings and reinvests that capital at a 20% rate, over time that should add about 10% to annual earnings power, thereby increasing intrinsic value by 10%. However there are a handful of companies that defy this logic. These “Capital-Light Compounders” are able to increase earnings power with zero or even negative capital employed. How do these companies accomplish this feat?

There are a couple of common characteristics in almost all Capital-Light Compounders, specifically negative working capital, low fixed assets, and real pricing power.

Negative Working Capital:

To determine the working capital structure of a company examine the balance sheet over the last few years and do some quick math to calculate the typical levels:

Connor Clip 2

Note: For this calculation I advocate using round numbers and rough estimates for excess cash. Working capital is dynamic and it is not necessary to calculate a precise number down to the last dollar to arrive at a general conclusion[ii].

A typical business will have a positive number for this calculation, however certain companies will be consistently negative – these are the ones we are looking for. Negative working capital often means the customers are paying the company cash up front for goods or services that will be delivered at a later time. This is a powerful concept for a growing company, as the customers are essentially financing the growth through pre-payments. Best of all the interest rate on this financing is 0%, pretty tough to beat. It is common to see negative working capital in subscription-based business models where customers pay up for recurring service or access. Some examples include SiriusXM, Verisk Analytics, and Atlassian. Because revenue is recognized when the service is performed, which is after the cash comes in, these businesses typically have operating cash flow that exceeds net income.

Low Fixed Assets:

The second characteristic of a Capital-Light Compounder is low fixed asset intensity, which can be analyzed by comparing net PP&E and/or capital expenditures to annual sales. If a typical manufacturing business wants to grow it will require significant capital investments in new factories, machinery, and trucks. Instead we are looking for companies that make money based off intangible assets such as brand name, intellectual property, or developed technology. A classic example is a franchisor, such as Dairy Queen, Burger King, or Winmark. In this business model the franchisor collects a royalty from franchisees in exchange for the use of the brand name, business plan, recipes, and other proprietary assets. The overall system grows as franchisees supply the capital to build new locations, enabling the franchisor to increase revenue and earnings without deploying additional capital. The key factor to focus on when analyzing a franchisor is the cash-on-cash returns the franchisees earn from building new locations. If this metric remains strong, the brand should have a long runway of unit growth ahead.

Real Pricing Power:

Finally if the business provides a product or service that is differentiated, has high switching costs, and is critical to customers it may be able to consistently raise prices at levels exceeding inflation. This is the simplest way to grow earnings without additional capital because the flow-through margins on price increases should be extraordinarily high. Companies such as CapitalIQ and See’s Candy have long histories of raising prices at or above inflationary rates, and Buffett considers this one of the most important variables when analyzing a business:

“The single most important decision in evaluating a business is pricing power. If you’ve got the power to raise prices without losing business to a competitor, you’ve got a very good business. And if you have to have a prayer session before raising the price by 10 percent, then you’ve got a terrible business.”

So if you run a Capital Light Compounder that is fortunate to have all of these characteristics, what is the playbook for maximizing intrinsic value per share? One option would be to allocate the excess capital into mergers and acquisitions in an effort to grow earnings power. The issue here is that if you start with an exceptional business like Visa or Moody’s, it’s almost certain that the acquired business is inferior and will dilute the overall quality. Additionally, acquisitions can end up becoming a distraction that take management’s time and focus away from the core “crown jewel” business. The classic example of this pitfall is Coca-Cola in the 1980’s, which allocated proceeds from the core business into acquiring Columbia Studios before refocusing a few years later.

Instead my preference would be for management to undertake a systematic share buyback program, what Charlie Munger would affectionately label as “cannibalizing” their own share count over time. Instead of acquiring a new business and the risks associated with that strategy, the management should instead direct M&A funds towards acquiring more of the exceptional business that the shareholders already own. Aggressive share shrinkers such as NVR, Inc., AutoZone, and DirecTV successfully reduced share count by over 50% within ten year stretches.

This strategy creates a “double dip” for shareholders that can greatly enhance the compounding of intrinsic value per share. Imagine you own shares in a Capital-Light Compounder that is about to begin a decade long run of share cannibalization. Over that stretch, the business may increase earnings power at 10% per year, which would typically result in a ~10% return to an owner if the valuation multiples were held constant. However in this case additional capital was not required to grow, so instead 100% of earnings power was available for ongoing share repurchases, raising the IRR on the investment to 17.9% (refer to calculations below). This formula is how certain companies can turn solid growth into exceptional shareholder returns over long stretches.

Connor Clip 3 Revised

Note: even the best Capital-Light Compounders require some annual capital expenditures, so the amount of earnings allocated to share repurchases will probably be less than 100%. This example is more for illustrative purposes.

Are “Capital-Light Compounders” superior businesses to “Reinvestment Moats”? My current thought is that in an inflationary environment the Capital-Light Compounder is the preference because the lack of physical assets enables revenues to increase without the corresponding need for heavy capital expenditures at inflated rates. It is an interesting topic to debate, one which certain investors have weighed in on overtime:

“The best business is a royalty on the growth of others, requiring little capital itself.” – Warren Buffett[iii]

[i] Many great investment careers have been built on this method, I am not knocking it at all, it’s just a different approach from the one I focus on. I think the key to identify what approach works best for your personality and then be disciplined within that framework.
[ii] Calculating excess cash is more art than science. Some investors would advise you to remove all cash from this calculation, personally I believe you need a reasonable amount of “cash in the drawer” for a business to run.
[iii] From John Train’s The Money Masters – which has an excellent chapter on Buffett

Connor Leonard is the Public Securities Manager at Investors Management Corporation (IMC) where he runs a concentrated portfolio utilizing a value investing philosophy. IMC is a privately-held holding company based in Raleigh, NC and modeled after Berkshire Hathaway. IMC looks to partner with exceptional management teams and is focused on being a long-term owner of a family of companies. For more information visit or reach Connor at, or on Twitter at @CataumetCap.

17 thoughts on “Reinvestment Moat Follow Up: Capital Light Compounders

  1. Interesting article. It’s a finger in the air question but what returns do you think would be possible from a portfolio of ‘legacy moat’ businesses? Assuming you waited for a good valuation and sold at expensive valuations rather than buy and hold forever.

  2. Excellent article.

    Footnote [iii] ought to read “Train” and not “Thain”. Train wrote Money Masters, while Thain was the Chairman and CEO of Merrill Lynch.

  3. Hi John, Connor, I am a big fan of both your writings. I had a question- companies that use earnings to reduce debt- in which bucket would you classify them?

  4. Thanks again! Nice article with clear explanations. A bit off topic, but do you have a framework of evaluating network type companies and their resulting moats?

  5. Liked the article, thanks for it: agree that buybacks sometimes are profitable use of capital but that depends on intrinsic value vs market price doesn’t it? it’s not a strategy you can always implement … there are a couple of Buffett letters on this topic

  6. Great post, and perfectly articulates what I look for when investing in companies. While I use “net operating assets” rather than “net working capital”, the concept is rather similar. When this number is negative, it can often indicate that the company’s operations are being financed by float (from customers and suppliers). While rare, this is phenomenal trait to have. I would strongly recommend Stephen Penman’s Financial Statement Analysis and Security Valuation textbook which goes into further detail on these concepts, and really helped clarify my thinking around them.

    Also agree regarding asset-light companies. One my highest conviction stocks is a little snow plow manufacturer named Douglas Dynamics (PLOW) which was trading between $17-$18/share earlier in the year (~35% discount to my estimate of intrinsic value). I loved how asset-light they were, management’s track record of returning money to shareholders, reinvesting in the business through prudent and sensible acquisitions when possible, and multi-moat business. One doesn’t often come across these opportunities, but it’s good to load up when we do.

    Thanks again for a great post!

  7. Another way of thinking about these capital light compounders is that they can increase the growth rate by focusing on ROIC, not reinvestment, and they do so by focusing on the drivers of ROIC, which you get from a DuPont analysis.

  8. Hi Connor, thanks for the excellent article. What are your thoughts on the impact of market prices on extent of share repurchases? I.e. in your example, at what prices should our company cease buying back shares?

  9. A few responses, in order:

    Tom – the answer depends on a number of factors, mostly how wide was the initial discount and how long did it take to close? The longer you own the business, the more your investment returns will match the underlying returns of the business. But if you are constantly churning out your ideas (avg. turnover of 1 year or less lets say), then its more about your entry valuation and identifying a catalyst. I think numerous investors such as Ben Graham, Walter Schloss, and Seth Klarman have produced ~20% returns over stretches using this approach. Those are the “Hall of Famers”, but it certainly can be done.

    Vikas – I would probably classify those as Legacy Moats – Return of Capital. Private equity investors tend to focus on Legacy Moat businesses that generate steady cash flows to pay off debt. If you mess around with an LBO model you will see that a company acquired for 8x EBITDA with 4x of debt doesn’t need much EBITDA growth to generate a solid 15% – 20% return. That’s the formula PE investors go for, and over time its produced very strong results.

    Harry – A lot of VC investors have great frameworks for network effects, I basically borrow their concepts. Some of my favorites are Andreessen Horowitz, Peter Thiel, and Bill Gurley.

    Eli/RV: I think buybacks make sense when the share price is trading for less than intrinsic value per share. This is where it’s really important to understand management’s framework for capital allocation. If they state that “the company bought back 1mm shares to offset dilution” I would run the other way. Buybacks should have the same consideration as any other use of capital, its an investment. The ideal scenario would be a company that is perpetually undervalued while the “cannibalization” is occurring. In my 10 year example I assigned a constant multiple of 15x to the business, and buybacks at this level enhanced the returns from 10% earnings growth to an 18% all-in IRR. That shows me that the multiple of 15x was undervaluing the company throughout the 10 year run. Most investors assign multiples by looking at EPS growth rates and comparable companies. However in this case the Capital-Light Compounder’s 10% growth rate should be “worth more” because it also comes with tons of distributable FCF. I think most analysts/investors miss this point and its a potential source of edge for a longer-term investor.

    1. “I assigned a constant multiple of 15x to the business, and buybacks at this level enhanced the returns from 10% earnings growth to an 18% all-in IRR. That shows me that the multiple of 15x was undervaluing the company throughout the 10 year run.”

      I was wondering if you could elaborate on your conclusion of the 15x undervaluing the company. I changed 15x to 100x. The IRR is still above the 10% earnings growth rate. Actually, the IRR would be be marginally above 10% for a multiple of 99999999x. Basically, this view suggests that in an inflationary environment in which a stock is trading at its intrinsic value and 100% of earnings will be used to repurchase shares, there is no multiple too high. Thoughts?

  10. John & Connor,

    The posts on reinvestment opportunities and capital allocation are among the best I’ve seen on value investing – many thanks to both for the insight and perspective. It may be worth highlighting a crucial component to the argument made in the article – cannibalizing shares ought to happen only when shares are trading at a discount to intrinsic value. In the examples cited (AZO, NVR, etc…), the cannibalization created value because shares were bought in at discounts to intrinsic value over long periods of time. I think that the reason for the persistent undervaluation was due to the sustainable growth in unlevered free cash flow almost regardless of economic backdrop.

    Even in the theoretical example posted above, unlevered free cash grows 10% per year for a decade and the going in price was about 10x EBIT (I’m assuming no debt and a 35% tax rate). The growth in unlevered free cash drives the degree of undervaluation, which makes the cannibalization work so well for ongoing shareholders. Almost any reasonable valuation framework would reveal that ~10x EBIT for such an opportunity is too cheap.

    Another company that is smartly repurchasing shares is Home Depot. Management maintains a view of its intrinsic value per share based on its internal estimates of unlevered free cash flow growth and its perceived cost of capital. Shares are repurchased only when priced at a discount to management’s view of intrinsic value.

    As an area of follow up, I’m interested in your thoughts regarding the optimal capital structure for capital light compounders. I realize that it varies from industry to industry, but I am struck by how low leverage levels tend to run for some of these companies (NVR and ORLY as prime examples). HD has said that they won’t take up lease-adjusted leverage above 2.0x debt/ebitdar; AZO doesn’t go above 2.5x. How would you characterize the right levels of financial leverage for these cannibals?

    1. That is a great question. Seems like with rates at the lowest levels in history, it would make a lot of sense for companies to take on more leverage, especially for companies with sustainable, predictable growth. In the above example, the returns would have been magnified that much more by funding a portion of the buybacks with debt. It doesn’t make sense to over do it, but seems like for a lot of companies, prudent use of debt could add tremendous value to the equity over time.

    2. Thanks for the comment Danny. That’s a good question regarding optimal leverage. I’m not sure what the correct answer would be, but I think it depends on the company and the nature of the industry it operates in. My very general take is that I’d almost always prefer lower levels of leverage (and thus lower ROE’s) in exchange for much greater safety in the event of a severe economic downturn that could impact volumes, revenues, and earnings. Some companies like Fastenal has historically produced excellent returns on equity through its business operations with very little use of financial leverage. They have recently taken on some debt, but throughout their history they’ve had very little use of leverage. Despite the stock being one of the all-time great compounders, they probably could have juiced returns by taking leverage up a notch or two. Fastenal obviously is a very rare bird (ability to produce high returns on incremental capital over a long period of time). But I think Buffett said something to the effect of (loosely paraphrasing): if you have to use leverage to produce attractive returns, then you probably don’t have a business that is all that great.

      That said, I don’t think I have any problem with firms like HD using leverage to increase returns, but I’d lean on the side of conservatism if I were in charge of the capital allocation. I think piling up cash isn’t a bad idea, as eventually market and/or economic conditions give you a chance to use it opportunistically (through internal investments, buybacks at better prices, acquisitions, etc…). As for AZO, they’ve been slightly more levered than some of their peers (at least relative to ORLY), and I think that could be a slight problem for them going forward as ORLY has built out their distribution network to a level where they now have significant ability to grow their store count without as much capital investment. AZO has focused on buybacks (financed in part by holding a slightly higher debt load), but now has to play catch-up in an attempt to build out their distribution capabilities.

      So I don’t think there is a right or wrong answer, but I’d probably favor a more opportunistic approach to capital allocation that gives the flexibility to take on leverage at times, and at other times let cash build up (or let leverage decline) in anticipation of utilizing dry powder elsewhere… I think most stable companies seem to lean more toward a consistent leverage ratio though…

  11. Nice post.

    I personally believe it is difficult to get an analytical edge by identifying negative/no working cap businesses as it seems to be common knowledge and no secret. I think the difficulty becomes how much do you pay for these businesses?

    How do you value a business that needs no capital? What multiple of FCF are you happy to pay? How much should we pay for a similar business that needs working cap?

    This is the question I struggle with.

    1. Yeah that’s always the tricky part, and a huge piece of the puzzle. As Charlie Munger has said, it’s easy to see that IBM would beat some crappy steel company (he said this during a time when IBM dominated). But IBM traded at 6x book value and the steel company traded at 0.3 times book value. It wouldn’t be as clear cut which one offered more value.

      Generally speaking though, I think that the better businesses almost always tend to perform better than poor businesses, at least over time. There might be a chance to buy a steel company cheap and sell it a few months or a year later for a nice profit, but in 3-5 years it will likely be mired in the same state it is now with meager returns on capital and miserable earning potential, whereas the business that can produce above average ROIC will likely be worth much more.

      As for how much to pay, it varies from business to business. It’s really hard to put a number on a general situation, so I’d have to just say that each business has their own unique position in the market that warrants an individual analysis. Even among capital light FCF generators that don’t require any capital, some will be much better positioned than others, and thus will demand a higher valuation.

Leave a Reply

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