Non-Recourse National Strategy

Behaviors you learn in your career are hard to unlearn

There are behavioral ticks you develop in any industry and real estate development and venture capital - while very different in many respects - share an important one: limited recourse. The VC writes a check into a fund vehicle. The fund writes a check into a startup. The startup burns through it. Nobody goes to prison. Nobody loses their house. The GP’s carry is clawed back, if the documents are well-drafted but often another fund has already been raised on marks that can be charitably described as optimistic. The real estate developer draws down a construction loan against the asset, and if the asset goes sideways, hands back the keys. In neither case does the decision-maker internalize the full cost of being catastrophically wrong. This is not a bug, it is a feature of these industries and in the case of real estate is why bank lending is and needs to be heavily regulated.

What it also does, and this is less celebrated in the Sand Hill Road hagiographies, is systematically select for a cognitive style. People who thrive in limited recourse environments develop a very particular relationship with risk. They learn, through repetition and reward, that bold bets on low-probability outcomes are structurally advantaged because the payoff on the upside is uncapped while the downside is bounded. They learn that hesitation is expensive, that deliberation is for people who don’t have conviction, and that the main failure mode is not swinging. They are, in the language of behavioral finance, calibrated to be systematically overconfident in their own ability to identify signal in noise, and structurally indifferent to path dependence because in their world, paths don’t particularly matter — if this startup dies, you do the next one. The option resets.

Ghent and Kudlyak showed this empirically in the mortgage context: non-recourse borrowers walk away from underwater properties at dramatically higher rates, not because they can’t pay but because the calculus is different. The liability is capped. The rational response to a capped liability is to think like an option holder, not a property owner. What happens when you staff an entire executive branch with people whose entire professional formation has been as option holders?

You get the current administration’s approach to tariffs as Exhibit A.

The tariff policy makes complete sense if you model it as a VC making a bold, asymmetric bet as many Trump adjacent people argued it was. The upside scenario, in the administration’s telling, is a wholesale reindustrialization of America, the return of semiconductor fabs and auto plants and steel mills, a renegotiation of the entire postwar trading order on terms favorable to the US. That’s the moonshot. The downside, in the cognitive model of someone whose career was forged in Sand Hill or in Manhattan real estate, is that you hand back the keys, dust yourself off, and do the next deal. The problem is that countries are not startups and trade relationships are not venture portfolios, and the people on the other side of the table are not simply going to dissolve and return their remaining capital to LPs.

Exhibit B is the war in Iran. Heads you win and its regime change, tails…. uh did anyone think about tails? Apparently not - at least certainly not Kushner, Trump and Witkoff the real estate guys.

This is where path dependence enters, and the administration appears constitutionally unable to see it.

Path dependence in diplomacy is not a complex concept. It is simply the observation, formalized in the repeated games literature going back to Axelrod’s tournaments in the early 1980s, that the shadow of the future changes behavior in the present. Countries that interact repeatedly — on trade, on security, on currency arrangements — are playing an iterated game, not a one-shot game. In iterated games, reputation is not a “nice to have” it drives how people play and outcomes. Your counterparty’s willingness to make concessions today is a direct function of what they believe you will do tomorrow, and that belief is formed by what you have done before. Defect once and your partner adjusts their priors. Defect repeatedly and the game changes entirely — they stop cooperating not because they are irrational but because cooperation is no longer incentive-compatible given their updated model of you. Assassinate their leadership repeatedly and you are outside the literature in a very, very bad way.

A venture capitalist does not play iterated games in any meaningful sense. Each company is a new relationship, a new cap table, a new set of counterparties. The GP’s reputation matters at the margin — a truly terrible actor will find it harder to get into the best deals — but the feedback loop is long, loose, and largely mediated by PR. A real estate developer in a large enough market is similarly insulated. Trump’s own career is a case study in this: multiple bankruptcies, creditors left holding losses, contractors unpaid, and yet the deal flow continued because in a market as large and liquid as Manhattan or Florida real estate, there are always new banks, new marks, new counterparties who haven’t played with you before.

That is not how sovereign relationships work. The European Union, Canada, Japan, South Korea, Mexico — these are not fresh counterparties who can be impressed with a pitch deck and a handshake. They are repeat players with long institutional memories, staffed by civil servants whose careers span administrations and who brief incoming ministers on exactly what the Americans did the last time. When the administration announces tariffs, waives them, reinstates them, carves out exceptions, threatens anew, and then claims victory for a deal that looks remarkably like the status quo ante — they are not being seen as tough negotiators. They are being modelled as unreliable, which is categorically different and considerably more damaging.

The VC intuition says: move fast, create urgency, manufacture FOMO, keep counterparties off balance. In a Series B negotiation this sometimes works. In a trade war with a country that has a forty-year planning horizon and a bureaucracy that outlasts any individual administration, it is roughly as effective as the aforementioned frat boy taking a swing at the meth head — the asymmetry of pain tolerance is not in your favor and you have misread the situation badly.

The deeper problem is that the limited recourse cognitive model also produces a specific blindness to what engineers call “path integrals” — the reality that the route you take to an outcome matters as much as the outcome itself, because the route leaves state changes in the world that cannot be undone. The factory that was not built because of tariff uncertainty doesn’t get built retroactively when the uncertainty is resolved. The supply chain that was diversified away from US inputs doesn’t snap back. The Taiwanese semiconductor manufacturer who started building Arizona fab capacity under CHIPS and is now confused about whether the US is an ally or an adversary adjusts their investment case in ways that compound quietly for years. The Iranian negotiator that knows his predecessor was killed is likely to engage in tit-for-tat negotiation in bad faith by, for example, talking about a Hormuz opening but then saying it will only be 15 vessels so that oil products spike hard into the midterms.

None of this shows up in the carry calculation. None of this is visible to someone whose mental model of risk is shaped by a portfolio of bets that either hit or don’t, where the failures are written down and the winners define the narrative. You are, to put it plainly, being governed by people whose professional formation has given them an excellent intuition for one very specific kind of risk-taking, and a near-total inability to perceive the category of risk that now matters most: the slow, structural, irreversible erosion of the trust and predictability that underpins everything else.

The option is not going to reset until the players change and even then, it will take time.