The Dual Asset

TAM-CV.06 · The Capital View · The Approximate Mind

Marcus has moved the trawler.

It is on the windowsill now, behind him, where he would have to turn around to see it. He does not seem to notice this. He is talking about governance structures, about board representation and preferred terms and the specific mechanics of a dual-asset exit, and the trawler is behind him catching the afternoon light while he works through the logic of something he has been building toward for six months.

He sounds different. Not more certain. More careful, in a way that reads as different from caution. Something has shifted in the way he describes what he is building, in the vocabulary he reaches for, and the shift is worth attending to before the mechanics.

The Conventional Mistake

#

The standard private equity approach to technology in a service rollup is to build it in-house, keep it proprietary, and value it at exit as an embedded asset. The technology is a tool, inseparable from the business it serves, and at exit it is valued accordingly: at services multiples, not technology multiples. Twelve times EBITDA for a healthcare services company that happens to have built a good internal platform is still twelve times EBITDA. The technology is in the number. It is not a separate number.

This is a significant valuation discount on the technology asset, and most deal teams accept it because the alternative requires a structural complexity they find uncomfortable. The alternative is to treat the technology platform as a separate entity, with its own capitalization, its own customer base, its own revenue, and its own exit.

Two entities. Two tracks. Two exits.

The rollup portfolio acquires agencies, consolidates back-office functions, installs the AI orchestration layer, accumulates outcome data, and differentiates into three service tiers. At exit, it is valued as a healthcare services business with a technology moat: fifteen to twenty times EBITDA, bought at six to eight. The multiple expansion is the thesis.

The technology platform serves the rollup portfolio. It also serves independent agencies, health systems, payers, other PE-backed rollups in adjacent markets. It has recurring revenue from multiple customers. It has a proof record that does not depend on any single client. At exit, it is valued as a technology platform: eight to fifteen times revenue, in a different universe from services multiples.

The same AI layer, valued two different ways, because the structure separated what the conventional approach kept bundled.

The structural insight Marcus has been working toward is this: the technology platform becomes more valuable if it is not captive to the rollup. Every external customer the platform serves increases its valuation without diluting the rollup’s operational advantage. The rollup has the outcome data from its own clients, the demand density that no independent agency can replicate, the horizontal bundle that is the product category. Competitors can license the platform. They cannot replicate what was built on top of it.

The moat is layered. The platform is available to anyone. The implementation at scale is not.

The Tension

#

The governance question is real and Marcus does not paper over it. If the rollup’s operations run on a platform that also serves the rollup’s competitors, the platform’s interests and the rollup’s interests will diverge at specific moments. A feature that serves the rollup’s competitors might be the feature that makes the platform more valuable to the acquiring company at exit. A pricing decision that is good for the platform’s independent customers might not be the pricing decision the rollup would choose if it controlled the platform entirely.

Board representation, preferred terms, information barriers, right of first refusal on competitive features: these are manageable. Deal lawyers have built the contracts for structures like this before, in other industries, and the contracts work well enough.

What cannot be contracted away is the structural tension itself. The platform is more valuable as an independent entity than as a captive tool, and the independence that creates the value also creates the possibility of interests that diverge. The rollup accepts this tension in exchange for the valuation gap between services multiples and technology multiples. The calculation is usually correct. It is not always comfortable.

Marcus describes this tension the way someone describes a known risk they have decided to take rather than a problem they have failed to solve. The language is specific and unpanicked. He has thought about it. He has structured around it. He has accepted what cannot be structured away.

The Half-Life Question

#

The harder question, the one he circles back to twice without being prompted, is about the outcome data.

The thesis depends on accumulated outcome data being a compounding asset. More data, better proof of care quality, higher premium from payers and families who are choosing between providers. The early mover who accumulates outcome data at scale has a moat that late entrants cannot close simply by deploying the same platform, because the platform is available to everyone but the data is not.

This is correct, as far as it goes.

The question Marcus is now asking is how far it goes.

If the platform standardizes care delivery across all its users, if the protocols it recommends and the outcomes it produces converge across the rollup and the independent agencies and the health systems using the same orchestration layer, then the outcome data across providers begins to converge. The rollup’s clients have good outcomes. The independent agencies’ clients have good outcomes. The data that was a differentiator becomes, over time, a baseline. The moat was never the data. It was the lead time in accumulating it, and lead time is a depreciating asset.

The rollup is racing to accumulate something that may be depreciating even as it acquires it.

Whether the depreciation is fast enough to matter within the hold period, typically five to seven years, is the bet. Marcus believes it is not, or more precisely, he believes the lead time advantage is long enough that the exit happens before the convergence erodes the premium. This may be correct. The demographic wave is large enough that the market keeps expanding even as margins compress, and expanding markets absorb a lot of convergence before the pressure becomes acute.

What he is less certain about, and says so, is whether the platform business holds its valuation if the outcome data story weakens. The technology platform at exit is worth technology multiples if it can show recurring revenue, proven orchestration, and a proof record of measurably better outcomes. Two of those three are durable. The third depends on the half-life of the data advantage, which depends on how quickly the market converges, which depends on how widely the platform is adopted, which is partly within Marcus’s control and partly not.

He does not resolve this. He describes it as the honest risk in the structure, the thing he discloses to LPs who ask carefully enough.

What the Language Reveals

#

Six months ago Marcus described what he was building as a care coordination business with a technology moat. The vocabulary was operational. The questions were about labor markets, regulatory environments, acquisition multiples, integration timelines.

He uses different words now.

He talks about the platform as infrastructure. Not infrastructure in the loose sense that technology companies use when they want to sound important, but in the specific sense: the thing that other things depend on, the layer beneath the service layer, the part that has to work for anything else to work. He talks about what the platform is for in a way he did not six months ago. Not what it produces, which is outcomes data and care coordination and operational efficiency. What it is for, which is something he takes a moment to find the right word for.

He says: dignity at scale.

He says it the way someone says a phrase they have not said out loud before, testing whether it survives contact with air. It does not embarrass him. He does not qualify it immediately. He lets it sit for a moment and then continues talking about governance structures.

I do not think he arrived at this phrase through sentiment. I think he arrived at it through the investment logic, through the half-life question and the floor-becoming-ceiling risk and the agent-to-agent scenario that dissolves toll booths that cannot justify their margin. I think he understands that a platform built to optimize metrics rather than the thing the metrics are supposed to measure is a platform that will be competed away by someone who builds it right. I think dignity at scale is, for Marcus, a structural argument before it is a moral one.

This does not make it less true.

The infrastructure that forgets what it is for tends to be replaced by infrastructure that remembers.

Whether he knows that this is what he is building, or whether he is building it and noticing the description afterward, is a question I am not sure how to answer. Possibly both. People discover what they are doing partly by watching themselves do it, and Marcus has been watching himself build something for six months that turned out to be larger than the thesis he started with.

The trawler is on the windowsill. He has not mentioned it. He is talking about preferred terms and board representation and the specific mechanics of a dual-asset exit, and behind him the afternoon light is moving across a wooden boat he bought twenty years ago in a country he has not been back to, for reasons he has never explained to anyone.

He does not fish. He never has.

This is the sixth essay in The Capital View, a nine-essay arc examining the AI transition from the position of capital. It returns to Marcus, the PE partner introduced in TAM-CV.01, six months into building the structure the first essay described. The essays that follow complete the arc: TAM-CV.07 names the general pattern of capital enclosure across industries, TAM-CV.08 traces the asymmetric deployment of AI across populations and the feedback loop this creates in what gets built, and TAM-CV.09 makes the practitioner case directly to the PE audience. This essay connects to the choreographed market argument in TAM-051; to the dissolved middle in TAM-059; and to the autonomous pipeline and utility layer questions in the Ungoverned Frontier series. The half-life question it raises without resolving is taken up in TAM-CV.09.

References

#

Platform Economics and Dual-Asset Structures

Evans, David S., and Richard Schmalensee. Matchmakers: The New Economics of Multisided Platforms. Harvard Business Review Press, 2016.

Parker, Geoffrey G., et al. Platform Revolution: How Networked Markets Are Transforming the Economy and How to Make Them Work for You. W. W. Norton, 2016.

Parker, Geoffrey G., and Marshall W. Van Alstyne. “Two-Sided Network Effects: A Theory of Information Product Design.” Management Science, vol. 51, no. 10, 2005, pp. 1494-1504.

Private Equity Exit Structures and Valuation

Appelbaum, Eileen, and Rosemary Batt. Private Equity at Work: When Wall Street Manages Main Street. Russell Sage Foundation, 2014.

Kaplan, Steven N., and Per Strömberg. “Leveraged Buyouts and Private Equity.” Journal of Economic Perspectives, vol. 23, no. 1, 2009, pp. 121-146.

Data as Asset and the Half-Life Problem

Mayer-Schönberger, Viktor, and Kenneth Cukier. Big Data: A Revolution That Will Transform How We Live, Work, and Think. Houghton Mifflin Harcourt, 2013.

Varian, Hal R. “Artificial Intelligence, Economics, and Industrial Organization.” The Economics of Artificial Intelligence: An Agenda, edited by Ajay Agrawal et al., University of Chicago Press, 2019, pp. 399-419.

Infrastructure, Dependency, and Control

Bowker, Geoffrey C., and Susan Leigh Star. Sorting Things Out: Classification and Its Consequences. MIT Press, 1999.

Plantin, Jean-Christophe, et al. “Infrastructure Studies Meet Platform Studies in the Age of Google and Facebook.” New Media and Society, vol. 20, no. 1, 2018, pp. 293-310.

Value Migration in Technology Markets

Christensen, Clayton M. The Innovator’s Dilemma: When New Technologies Cause Great Firms to Fail. Harvard Business School Press, 1997.

Mazzucato, Mariana. The Value of Everything: Making and Taking in the Global Economy. PublicAffairs, 2018.