How to Spot a Compounder

And why you should be willing to pay up once you identify one!

Following my post on TerraVest earlier this week, I wanted to share my thoughts on compounders, how to spot them, and why they deserve to trade at lofty multiples.

I think the term “compounder” gets thrown around loosely, so let me start by explaining my mental model for compounders. To me, a compounder is a company with a long runway for reinvesting its profits back into its business at high rates of return (+15%). Classic examples of business models that fit this mold are companies that grow by acquisition (e.g. Constellation Software, TransDigm) and retailers/restaurants with long runways for expanding their store bases (e.g. Tractor Supply, Starbucks). In essence, existing operations generate a baseline level of sales growth and profitability, while profits from those existing operations are used to acquire other businesses/open new locations at attractive economics. Let me go into further detail with a few examples:

Constellation Software

Constellation Software is a consolidator of vertical market software businesses. Their growth formula is broken up into two components:

• Organic growth of underlying businesses. Each acquired software business grows organically through the acquisition of new customers/users and annual price increases.

• Acquisitions: Using the profits from its owned businesses, Constellation acquires other vertical market software businesses.

Historically, Constellation’s organic growth has been modest, generally ranging from 0-5% annually. The reason why Constellation’s businesses grow at unexciting rates compared to other software businesses you might be familiar with (probably growing in the double digit range) comes down to incentives. Constellation’s managers are incentivized to drive (i) capital deployed, (ii) ROIC, and (iii) organic growth (to the least extent). Naturally, this has pushed Constellation’s managers to become relentless M&A machines while ensuring pricing discipline. Mark Leonard, Constellation’s chairman and founder, has made many remarks in his president’s letters about the challenges of sustaining organic growth over the long-term. One example (Q1 2009 President’s Letter): “The toughest challenge in the software business is intelligently trading off profitability and organic growth. Many entrepreneurs have a huge bias towards growth at the expense of profits. Most private equity owned software firms have the opposite bias. At Constellation we try to find an optimum position where incremental investment still generates good incremental long term returns.” Building off that quote, Constellation’s managers are forced to make the following trade off: invest in their businesses to generate organic growth, which comes at the expense of profit (minimizing the R in ROIC), or hustling for new acquisition opportunities at attractive prices to maximize capital deployed and maintain/juice ROIC through discipline.

From the beginning of 2007 to September 30, 2023, Constellation has generated ~$9.0B of CFO and has deployed ~$7.3B on acquisitions, implying a reinvestment rate on acquisitions of +80% (the balance of CFO has been used primarily to fund modest dividends and capex). This was done all while maintaining an ROIC that has hovered in the ~20-40% range. It’s worth noting that prior to 2021, Constellation maintained a +30% ROIC based on internal hurdle rates on acquisitions. Thereafter, Mark Leonard’s directive was to lower the company’s ROIC hurdle on acquisitions to something in the ~20% range (actual hurdle rate not disclosed, but ~20% range supported by recent financial performance) to allow the company to deploy closer to 100% of its CFO at an increasing scale. This change was welcomed by shareholders, many of whom (including myself) recognize that they’re better off letting Constellation allocate 100% of their capital at a lower, but still attractive, ROIC. Mark Leonard addressed this point in his last president’s letter written in 2021.

Since Constellation’s IPO in 2007, the company’s share price has risen by more than 130x (+150x including dividends), implying a CAGR of 35.4% (36.8% including dividends). The components of the company’s monstrous performance can be broken down into the appreciation of the company’s (i) CFO/share and (ii) Price/CFO per share multiple. Since its IPO, Constellation’s CFO/share has risen ~61x, while its Price/CFO per share multiple has risen by ~2.1x, with the multiple increasing from ~15x in 2007 to ~30x today*. For clarity, when taking those two components together (61 * 2.1), you get ~130x. Obviously, what has driven the performance of Constellation’s share price since its IPO has been the compounding of its CFO/share. By reinvesting the lion’s share of earnings at a +20-30% ROIC, coupled with getting a few percentage points of growth organically (which requires zero capital), a powerful combination of compounding is created.

*Given that Constellation completes acquisitions throughout the year, the earnings impact of acquisitions completed part-way through a given year are only partially reflected in Constellation’s financials. For simplicity, no pro forma adjustments have been applied to Constellation’s earnings. Given that Constellation reports in USD, but the company’s stock price is CAD denominated, Annual CFO has been translated from USD to CAD at average annual FX rate.

Tractor Supply

Tractor Supply is the the largest rural lifestyle retailer in the U.S. Stores are located on the outskirts of large cities and in rural communities. Stores typically carry ~20k SKUs and store assortment primarily includes farming products, hardware, and a diverse array of home goods. Their growth formula is broken up into two components:

• Same store sales growth (SSSG). Base of existing stores increase sales through price increases, improved merchandising, higher average ticket per transaction, etc.

• Same store sales growth: Using the profits from its existing store base, Tractor Supply opens new stores.

From the beginning of 2000 to September 30, 2023, Tractor Supply has generated ~$10.7B of CFO and has deployed ~$5.3B on capex designated for new store openings/remodels (~$4.7B) and acquisitions (~$550M), implying a reinvestment rate of ~50% over nearly 24 years. The balance of cash generated was primarily used to fund share repurchases (~$4.7B) and dividends (~$2.1B). This was done all while maintaining an ROIC generally in the ~20-30% range.

Since 2000, Tractor Supply has experienced exceptional SSSG, averaging ~6%, with SSSG dipping into negative territory just once (negative 1.1% in 2009 - all things considered, I don’t think that’s so bad for 2009). Coupled with that, Tractor Supply increased its store count from 305 in 2000 to nearly 2,400 through September 30, 2023 (nearly ~8x in ~24 years). The trifecta of powerful SSSG, new store openings, and improved net margins from economies of scale (EBIT margin more than doubled from ~5% in 2000 to ~10% today) has enabled Tractor Supply to grow its net income ~70x (~20% CAGR) since 2000. Overall, Tractor Supply’s share price has increased ~201x (~25% CAGR), or ~238x including dividends (~27% CAGR), broken down as ~70x from net income appreciation, multiplied by ~2.3x from an expanded P/E multiple (~9x to ~20x), and finally multiplied by ~1.3x from a reduced share count from share buybacks. I know that’s a bit of math, but here’s my mental model: new store openings generally have paybacks of ~2-4 years. For argument’s sake, after-tax, let’s call that a 25% cash on cash return on a new store opening. Each year, half of the company’s profits are reinvested into the business for new store openings. From that, you can expect 12.5% earnings growth from new store openings (50% reinvestment rate * 25% return). On top of that, you get ~5 percentage points of organic growth that, let’s assume, requires no incremental capital. Combined, you get 17.5% revenue growth annually. Layer on margin improvement from economies of scale and share buybacks and you get to your low 20s EPS CAGR.

But how long can compounders compound for?

This is naturally the hardest question to answer. If you look at Constellation, you have to ask yourself “how many software businesses can they possibly buy? Will they reach a point of scale whereby it eventually becomes impractical to continuously deploy +80% of FCF (now 100% of FCF) at a +20% ROIC?” This is the question that has plagued Constellation’s wannabe investors on the sidelines forever. The company has continued to prove their doubters wrong, and in the last couple of years alone, the company has managed to deploy +$2.5B of capital on acquisitions, exceeding their CFO over the same time period (they drew debt to fund the shortfall), all while maintaining a +20% ROIC. It’s also worth mentioning that Constellation claims to maintain a growing database with over 100k software companies.

As mentioned previously, given Constellation’s scale, Mark Leonard decided to lower the company’s hurdle rate on acquisitions from +30% to +20% to be able to continue to deploy the company’s free cash flow at an ever increasing scale. Historically, Constellation has focused on smaller acquisitions (<$10M in sales) that generally fall beneath the radar of Software PE funds. In addition, Constellation opportunistically pursues larger acquisitions.

Notably, this year, in what may be the fleece of a lifetime, Constellation acquired Optimal Blue, a SaaS platform in the mortgage origination space. Optimal Blue was sold by Black Knight as mandated by the FTC to facilitate Intercontinental Exchange’s acquisition of Black Knight. Constellation acquired Optimal Blue for ~6.3x EBITDA of ~$110M. The $700M acquisition was financed with $200M of cash from Constellation and a $500M promissory note issued by Constellation to Black Knight payable over 40 years at a 7% interest rate. Meanwhile, Black Knight acquired Optimal Blue in tranches over 2019 and 2022 at EBITDA multiples in the ~24x and ~29x range. Sure, mortgage originations have dropped off in the last couple of years given interest rate hikes, but the combination of the ~6.3x EBITDA multiple and incredibly attractive financing structure makes this deal seem like an incredible bargain purchase for Constellation. One point I want to make from this is that because the sale was mandated by the FTC, Intercontinental Exchange needed a buyer to move swiftly so they could close their acquisition of Black Knight. In my opinion, a transaction like this requires a buyer with deep software expertise and a proven M&A track record to pull this off. Constellation fits the bill perfectly. All to say, Constellation has a proven track record of taking down smaller deals in volume and larger deals opportunistically while maintaining pricing discipline.

Let me add one more thing on Constellation: they eat their own cooking. Although Mark Leonard is a material shareholder (owns ~2% of shares outstanding, worth ~C$1.5B), one of my favorite elements of Constellation’s culture is their employee compensation scheme. I mentioned the employee incentive structure above, but what’s also interesting is that employees must use their year-end bonuses to purchase Constellation stock on the open market, and the purchase itself vests over 4 years. This culture has kept share dilution to zero since the company’s IPO. This is of course in stark contrast to most other tech companies that issue stock-based comp that dilutes shares by +10% annually.

Given Constellation’s exceptional track record and the incredible culture Mark Leonard has built over multiple decades, I have faith in the company’s ability to maintain a +80% reinvestment rate at a +20% ROIC for years to come.

On the other hand, if you look at Tractor Supply, you have to ask yourself “how many stores can they possibly add to an already lofty base?” Over the last ~10 years, Tractor Supply has opened ~100 new stores per year. Of course, that amounts to tremendous growth when your existing store base is 500 or even 1,000. But, as you scale, the incremental ~100 new stores move the needle less and less. Now, at ~2,400 stores, an incremental ~100 stores amounts to ~4% growth. Sure, Tractor Supply can expand into adjacent retail models, but is that a bet I want them to take so they could sustain high levels of capital deployment? Probably not. At this point, given the company’s scale, the days of Tractor Supply’s sales growth in the teens % range are likely over, and their growth algorithm is realistically in the mid to high single digit range (~3-5% from SSSG plus ~3-5% from store openings).

Given the above dynamics, my view is that it’s much harder to maintain continuous compounding as a retailer than it is to maintain continuous compounding as a growth by acquisition business. You brush up on the natural limits of store expansion before you brush up on the natural limits of acquisition opportunities if you play in a large enough industry. Obviously, there aren’t an infinite number of businesses in the world. But, with a strong culture that incentivizes capital deployment and ROIC, steady, double digit compounding can be maintained for many years even at significant scale (if you don’t believe me, see Berkshire Hathaway).

So why do compounders trade at such rich multiples?

I think this is most easily communicated by way of example. The image below has a high level model (with dummy numbers) running the following assumptions:

• Base FCF grows by 2% annually

• 80% of FCF in a given year is redeployed on the first day of the following year at a 30% ROIC (equivalent to a 3.3x multiple of earnings)

• 20% of FCF used for share buybacks

• Entry and exit multiple of 20x

As shown at the bottom of the image below, FCF/share increases by 11.3x over 10 years, equating to a 27% CAGR. Given that the entry and exit multiples are both 20x, share price also grows at a 27% CAGR.

Let’s consider this scenario: what if I had so much confidence in this company, and I knew they were going to redeploy 80% of FCF each year at a 30% ROIC? How much would I be willing to pay today? As outlined in the sensitivity table below, even paying 35x earnings (and assuming an exit multiple of 20x) would yield a 16% IRR and 4.5x MoC, a still very strong outcome over a 10-year period.

You may have picked up on this by now, but the above scenario is effectively what Constellation has been since its IPO. A collection of vertical market software businesses that generate low single digit organic growth coupled with an acquisition machine that deploys +80% of FCF generated at a +30% ROIC.

As mentioned above, it seems that Constellation’s hurdle rate on acquisitions changed in 2021 to +20% ROIC to allow the company to deploy closer to 100% of its FCF, so let’s run the following scenario through the model:

• Base FCF grows by 2% annually

• 100% of FCF in a given year is redeployed on the first day of the following year at a 20% ROIC (equivalent to a 5x multiple of earnings)

• Entry multiple of 30x (Constellation’s current multiple) and exit multiple of 25x (implicitly assumes regression to mean (S&P500 average) at the end of 10 years)

As shown at the bottom of the image below, FCF/share still increases 7.3x over 10 years, equating to a 22% CAGR. Given the assumed multiple compression, share price increases 6.1x over 10 years, equating to a 20% CAGR. Still a nice outcome!

Now let’s compare the Constellation-type scenario to a Tractor Supply-type scenario:

• Base FCF grows by 5% annually (assumed SSSG)

• 50% of FCF in a given year is redeployed on the first day of the following year at a 25% ROIC

• 50% of FCF used for share buybacks

• Entry and exit multiple of 20x

As shown at the bottom of the image below, FCF/share increases by 6.6x over 10 years, and both FCF/share and share price grow at a 21% CAGR. By all means, this is still an excellent outcome, but not quite as good as Constellation generating 2% organic growth, but deploying 80% of FCF at a 30% ROIC.

So how do you identify a compounder in the making?

If this were an easy question to answer, I’d be much richer. That said, in studying some of the greatest entrepreneurs and capital allocators, I have the following to offer:

• You get what you incentivize (thanks Charlie Munger for instilling this in me). If you want your company to compound earnings at high rates, you should be incentivizing capital deployment and ROIC to combine scale and discipline. Ultimately, employees need to think like owners and work towards building long-term value.

• The larger a business gets, the more difficult it becomes to centrally manage operations. Some of the world’s best CEOs of all time (Warren Buffett, Nick Howley, Mark Leonard) run extremely decentralized operations where all decision making, except major capital allocation decisions, is delegated to subsidiaries.

• Two of the best business types:

  • Businesses that require little to no capital to grow

  • Businesses that can employ large amounts of incremental capital at very high rates of return

If you’re interested in getting smarter on capital allocation, I’d highly recommend reading The Outsiders by Will Thorndike and Mark Leonard’s President’s Letters. There’s infinite wisdom in their content, and Mark Leonard’s letters are a masterclass in how to run a high performance conglomerate.

I know this post turned into a bit of a Constellation celebration, but I’m really trying to drive home the point that when you identify a great capital allocator in a business with a long reinvestment runway, invest and pay up!

Of course, investing in Tractor Supply and/or Constellation in their early days would have worked out wonderfully. What I hope you can take away from what’s included above is a framework for how to evaluate the makings of a future compounder.

Bringing it all back to TerraVest

I see elements of compounder qualities in TerraVest today. Solid incentives and insider alignment, decentralized organization, and acquisition machine. If I compare TerraVest to Constellation, I prefer Constellation’s end markets, but I think with discipline, there are probably better deals to be had in TerraVest’s end markets given their un-sexiness relative to software. I like what I’ve seen from the TerraVest team over the last 5 years, and if they can continue to redeploy close to 100% of their FCF at a +15% ROIC (and prudently use debt to juice capital deployed and ROE), naturally, the company will be a top performer.

Given that TerraVest is still relatively small (<C$700M), I think they’ve got tons of runway to continue to deploy FCF at high rates of return. If their next 5 years match the underlying performance of their past 5 years, I think their multiple should rerate to at least +15x (and in light of the modelling I included above, I wouldn’t be surprised if their multiple eclipses +20x).