Quality Industrial Compounders

The question

Among US-listed industrial operators with multi-decade records of high returns on invested capital, which are priced today with an imputed probability of continued reinvestment at historical rates that is materially below what the business quality would justify?

Base rates

The reference class is the long-run universe of industrial compounders -- businesses that have sustained ROIC above 15% for at least a decade and have paid out less than half of free cash flow while still growing book value. The historical base rate for this cohort, measured across rolling ten-year windows since 1985, is a 3-4 percentage-point annualized return premium over the broad US equity index. That premium is not random. It is mechanically produced by the reinvestment math: a business that can redeploy a retained dollar at a return above its cost of capital compounds at a rate the market consistently under-prices because the duration of the advantage is longer than the typical sell-side model window.

The imputed probability embedded in current multiples for the quality industrial cohort is that ROIC mean-reverts toward the cost of capital within a five-to-seven-year horizon. That is the default forecast assumption in most DCF templates. The base rate for that specific mean-reversion outcome in this cohort is roughly 35-45%; the base rate for ROIC staying within one standard deviation of its trailing ten-year average for another five years is closer to 55-65%. The spread between those two base rates is the inefficiency.

Why the consensus view is wrong (or incomplete)

The consensus view is not wrong that these businesses are expensive on trailing multiples. It is incomplete because it evaluates quality at the multiple rather than at the reinvestment-adjusted earnings yield. A business trading at 22x forward earnings that reinvests two-thirds of earnings at 20% ROIC has a higher economic return to the owner than a business trading at 14x that reinvests one-third of earnings at 10% ROIC. The multiple is the wrong frame for this cohort. The right frame is the prospective compounding rate of retained capital, and on that measure the names in this book are trading at or below their long-run median.

The causal mechanism that keeps the inefficiency alive is that the investors who most value quality -- long-duration family offices, endowment portfolios, some fundamental public books -- cannot scale their positions without moving the price. The investors who most move the price -- quant-factor books and passive flows -- weight on trailing multiples, not reinvestment economics. That structural mismatch is what makes the quality premium persistent rather than arbitraged away.

Position construction

Four near-core positions anchor the book. TEL at ~19% is the global connectivity franchise -- sensors and connectors into the vehicle and industrial-automation stack, with a multi-decade record of mid-teens ROIC. URI at ~18% is the equipment-rental consolidator whose operational record through the 2015-2016 oil shock and the 2020 pandemic established that its fleet-utilization discipline is a genuine operating edge, not a cycle illusion. CMI at ~16% is the diesel and industrial-powertrain platform with the hydrogen-engine option embedded inside a cash-generative legacy business. PCAR at ~14% is the Class 8 truck franchise with a parts aftermarket that has compounded book value through every cycle since 1980.

WAB at ~9.6% is the rail aftermarket and digital locomotive-controls franchise. DOV at ~6.6% and NDSN at ~3.8% are the specialty-industrial platforms with diversified end markets and the cleanest reinvestment records in the book. RBC (~3.4%), WCC (~3.0%), and AIT (~2.7%) are the bearings, electrical-distribution, and MRO specialists -- smaller weights reflecting thinner competitive moats relative to the top of the book. AEIS (~2.1%) is the precision-power franchise into semiconductor and industrial end markets. AZZ (~0.85%) is the galvanizing business at the lowest weight; it fits the quality screen but carries the thinnest reinvestment runway.

Asymmetric payoff

If the cohort sustains trailing ROIC through 2028 and multiples hold, the weighted book compounds roughly 12-16% per year against an SPY base rate near 8%. If US industrial activity contracts and multiples compress one turn, the book returns roughly -8 to -14%. If reshoring capex accelerates and TEL, URI, CMI, and WAB see incremental operating leverage, the right tail is 22-30%.

At a 60% base, 25% bear, and 15% bull weighting, expected value is roughly +12 to +16% annualized. The margin of safety is the reinvestment math: even with flat organic growth, retained earnings compounded at historical ROIC produces the base-case return without requiring multiple expansion.

Three things that would change our mind

1. Trailing four-quarter ROIC for the weighted book drops below 12%, indicating that the quality characteristic that defined the screen is eroding rather than being cyclically pressured.
2. Two or more core positions materially increase goodwill-heavy M&A at premium multiples, indicating that management teams are running out of organic reinvestment opportunities.
3. Class 8 truck orders, non-residential rental-fleet utilization, and rail carload volumes turn down simultaneously, indicating that the US industrial cycle has actually rolled rather than flattened.

What we are explicitly NOT betting on

We are not buying deep-cyclical industrials priced for trough earnings. The quality screen excludes them on purpose. We are not buying aerospace-primes, defense primes, or pure-play semiconductor-capital-equipment names; those belong in other books with different reference classes. We are not making a short-horizon call on the US industrial cycle. The thesis compounds on a five-to-seven-year horizon and depends on reinvestment rates, not on cycle timing. That is a narrower and more durable claim than a cyclical call, and the portfolio is sized for it.