Compounder vs Deep Value: Allocation Frameworks for Different Market Environments

The difference between buying a wonderful company at a fair price and a fair company at a wonderful price can mean thirty percentage points over a decade.

PROFESSOR: Charlie, you spend your days watching how headlines move markets. Let's talk about something that doesn't change with the news cycle but should absolutely change with the environment. When do you pay twenty-five times earnings for a quality compounder versus scraping the bottom for statistical bargains trading at eight times?

CHARLIE: The framework I use starts with the risk-free rate, because that's what changed everything in 2022. When ten-year Treasuries yielded one and a half percent, paying thirty times earnings for a business growing fifteen percent annually with high returns on capital made mathematical sense. You were getting a ten percent earnings yield growth trajectory against a one-point-five percent alternative. But when that ten-year hit five percent, the math flipped. Suddenly that same compounder is competing against a guaranteed five percent return. The margin of safety evaporated.

PROFESSOR: Walk me through how you actually built positions differently in those two regimes. Give me real names, real timing.

CHARLIE: In 2020 and 2021, I owned Visa at twenty-eight times forward earnings. The business had thirty-percent-plus returns on invested capital, was growing high single digits organically, had pricing power embedded in a two-sided network, and required almost zero maintenance capital expenditure. I didn't care that it looked expensive on a P/E basis because I was buying a perpetual cash-generating machine in a zero-rate world. The duration of those cash flows worked in my favor when discount rates were pinned to the floor.

CHARLIE: Fast forward to late 2022, I sold half that Visa position not because the business deteriorated but because the opportunity cost shifted. I redeployed into things like Cheniere Energy trading at seven times earnings with a contractually locked-in revenue stream through 2040. The quality was lower, the growth rate was lower, but I was getting paid an earnings yield north of fourteen percent versus Visa's three-point-five percent. When rates are elevated, current cash flow trumps distant compounding.

PROFESSOR: That's the interest rate axis. But you've also got market valuation levels, which don't always correlate perfectly with rates. How do you layer that in?

CHARLIE: Right, because you can have low rates with high market valuations or high rates with reasonable valuations. The second filter is where the overall market sits. I look at the equal-weighted P/E ratio of the S&P 500, not the cap-weighted version, because the cap-weighted gets distorted by whatever five mega-caps are dominating that year.

CHARLIE: When the equal-weighted P/E is above eighteen, which is roughly the historical average, I tilt heavily toward quality compounders even if they look expensive in isolation. Why? Because in a fully valued market, the statistical bargains are usually cheap for legitimate reasons. The management team is questionable, the balance sheet has hidden leverage, or the industry has structural headwinds. You end up with value traps. In that environment, I'd rather pay twenty-two times for O'Reilly Automotive, which has compounded book value at thirteen percent for two decades, than pay nine times for some troubled retailer that might be worth twelve times if everything goes right.

CHARLIE: But when the equal-weighted P/E drops below fifteen, like it did in March 2020 or October 2022, the calculus inverts. Now you're finding genuinely good businesses trading at statistically cheap multiples because fear is indiscriminate. That's when I'm buying energy infrastructure at six times EBITDA or regional banks at eight times earnings with clean loan books. The quality gap narrows when everything sells off.

PROFESSOR: Let's add the third dimension. Your own time horizon. Because this isn't academic if you're managing against monthly redemptions versus running a personal account you won't touch for fifteen years.

CHARLIE: The time horizon is the permission structure for the whole framework. If you're running institutional money with quarterly reporting and clients who get nervous when you underperform for six months, you have to acknowledge that reality. Deep value requires patience because the catalyst is often unknown and the timing is uncontrollable. You might be right about a stock trading at sixty percent of tangible book value, but it could trade at fifty percent of book for another eighteen months before the market recognizes it.

CHARLIE: Quality compounders give you optionality on timeframe. A business like Costco or Sherwin-Williams will likely be worth more in one year, three years, or ten years because the underlying earnings power keeps expanding. You're not dependent on a specific catalyst or multiple rerating. That matters enormously if you're managing other people's money or your own psychology can't handle extended drawdowns.

CHARLIE: But if you're genuinely a long-term capital allocator with no external constraints, deep value in the right environment offers better mathematical outcomes. Ben Graham's studies showed statistically cheap baskets of stocks outperformed over rolling five-year periods by three to four percentage points annually. That doesn't sound like much until you compound it. Four points of annual alpha over twenty years is the difference between a ten-million-dollar portfolio and a twenty-two-million-dollar portfolio.

PROFESSOR: Give me a decision tree someone could actually use. They wake up tomorrow, they've got capital to deploy. How do they synthesize these three factors into an actual allocation decision?

CHARLIE: Start with the ten-year Treasury yield. Above four and a half percent, you tilt toward value and current cash flow. Below three percent, you can afford to pay up for duration and compounding. That's the first cut.

CHARLIE: Second, check the equal-weighted S&P P/E ratio. Above eighteen, stay quality even if you're tilting value, which means find compounders on temporary setbacks rather than structurally impaired businesses. Below fifteen, you can hunt genuine statistical bargains because the market is probably mispricing risk across the board.

CHARLIE: Third, inventory your genuine time horizon. Not what you tell yourself, but what you've proven in past drawdowns. If you sold anything in March 2020 out of fear, you're not a ten-year holder. Be honest. If your real horizon is three years, weight toward quality. If it's genuinely seven-plus years and you have the disposition for it, you can load up on deep value when the first two conditions align.

CHARLIE: Right now, in early 2025, we're sitting at ten-year yields around four-point-three percent and an equal-weighted P/E around seventeen. That's a mixed environment. I'm running about sixty percent quality compounders that have sold off from highs, think companies like Texas Roadhouse or Floor & Decor, and forty percent statistical value in specific pockets like small-cap industrials or energy services where earnings yields are still north of ten percent. I'm not in one extreme or the other.

PROFESSOR: Last thing. The mistake you see people make most often when they try to run both strategies simultaneously.

CHARLIE: They contaminate the frameworks. They buy a statistically cheap stock but then expect it to compound like a quality business. Or they buy a quality compounder but get impatient when it doesn't work in six months and sell it like a value trade. If you buy something at eight times earnings because it's cheap, you need a catalyst thesis and an exit price. If you buy something at twenty-three times earnings because it compounds, you need to give it years and ignore the quarterly noise. Mixing those timeframes and expectations is how you underperform both strategies.

See you Wednesday. Pay for quality when money is free and the market is efficient; hunt for bargains when cash has a cost and fear is expensive.