Featuring
Jason and Friedberg called in reinforcements: Gavin Baker brought the market tape, Ben Shapiro brought the political theory, and Phil Deutch brought the energy spreadsheets. The sharpest clashes were over whether clean-energy subsidies kneecapped nuclear power, whether socialism self-corrects after failure, and whether tariffs can raise useful revenue without making the economy pay twice. Phil had the best episode by refusing Friedberg's solar-versus-nuclear trap and grounding the argument in how energy capital actually moved.
Spice rack
Would failed socialist policies quickly discredit the movement, or cause government to ratchet up its intervention?
Original point: Mamdani was likely to win, but if he implemented policies such as municipal groceries and police cuts, their failure would be obvious and would quickly end socialism's appeal.
What everyone argued
David Friedberg
Friedberg says intervention is a one-way ratchet: when food assistance or rent control disappoints, political advocates ask government to spend and control more rather than reconsider the premise. He expects affordability anxiety to make socialism spread well beyond New York.
Ben Shapiro
Ben directly rejects Gavin's quick-correction theory. He argues that failure takes years to become visible, can be blamed on wealthy people or businesses leaving, and may intensify redistributionist politics if a wider economic downturn lets voters say the rich gained while everyone else lost.
Gavin Baker
Gavin says Mamdani's charisma could win the election, but actual implementation would expose bad policies—especially municipal groceries—and bring a quick end to socialism's momentum.
Winner circle
Ben wins provisionally. He identifies the decisive flaw in Gavin's forecast: policy failure cannot quickly discipline an ideology when implementation itself takes years and political actors can contest the cause. Friedberg adds the stronger ratchet mechanism but overstates it into a national inevitability. The final outcome remains unresolved until the stores operate and voters react.
Commentary
David Friedberg
Assumptions and fact checks
Failed government programs generally produce demands for larger versions of the same programs.
Why it mattersPolicy feedback and organized constituencies can create a ratchet, but repeal, redesign, fiscal limits, and electoral backlash also occur. The direction depends on whether voters attribute failure to the program's premise, implementation, or insufficient scale.
Affordability pressure will cause socialism to sweep the United States.
Why it mattersAffordability is potent, but one mayoral victory does not establish a national realignment. Electoral institutions, local conditions, policy performance, and the opposition's response all constrain the forecast.
Nearly half of Americans are employed directly or indirectly by government.
CheckThe direct-employment portion is far smaller: BLS counted about 23.6 million government payroll jobs out of roughly 159.5 million total nonfarm jobs in July 2025, about 15%. 'Indirectly' is undefined and cannot turn that statement into a reproducible employment measure.
Ben Shapiro
Ben wins the timing argument without proving the ideological forecast. His best point is prosaic: a program scheduled to open years after the election cannot quickly teach voters anything about its operating performance.
Assumptions and fact checks
A downturn would make redistributionist politics stronger than evidence of a municipal program's failure would make it weaker.
Why it mattersEconomic distress can boost anti-establishment and redistributionist appeals, but the direction of backlash depends on who voters blame and what alternatives are credible. The causal forecast remains underdetermined.
Gavin Baker
Gavin deserves credit for making the most testable prediction and for correctly calling the election. He loses the debate by treating policy feedback like a quarterly earnings miss: municipal programs move slowly, use contested metrics, and come wrapped in political narratives.
Assumptions and fact checks
Poor results from a municipal grocery program would be visible quickly and decisively discredit socialism.
Why it mattersThe stores will take years to open, their subsidies and service goals create multiple success metrics, and voters can assign blame in competing ways. Even clear operating losses would not mechanically settle a broader ideological debate.
If elected, Mamdani was likely to proceed with municipal grocery stores.
CheckMayor Mamdani's administration allocated capital, identified sites, and issued an operator RFP for five municipal grocery stores, with the first planned for late 2027.
Did wind and solar subsidies divert capital from advanced nuclear power, or solve a different market problem?
Original point: Stable, rational subsidies helped solar, wind, and electric vehicles scale, and reversing them across administrations damages U.S. innovation and energy capacity.
What everyone argued
David Friedberg
Friedberg argues that subsidized returns pulled private capital into solar and wind and away from nuclear R&D. He says this prevented natural cost discipline, left the United States behind China's advanced reactors, and illustrates how government incentives always create an unseen loser.
Phil Deutch
Phil says renewable incentives successfully drove cost reduction and deployment, while nuclear's binding constraints were bureaucracy and the absence of firm buyers. He argues capital returned to advanced reactors when hyperscalers committed to power purchases and government began easing regulation, so solar and nuclear need not be treated as rivals.
Winner circle
Phil wins. He offers the better causal account and the more realistic portfolio: build fast renewables, preserve firm nuclear options, and fix each technology's actual bottleneck. Friedberg correctly warns that subsidies create opportunity costs, but his claim that they largely caused America's advanced-nuclear delay outruns the evidence and exaggerates China's deployment scale.
Commentary
David Friedberg
Friedberg finds a real tradeoff and then promotes it to the whole explanation. His case needed comparative evidence showing renewable credits materially reduced nuclear funding after controlling for licensing, project overruns, reactor economics, and the absence of committed buyers.
Assumptions and fact checks
Capital invested in subsidized renewables would otherwise have funded advanced nuclear R&D and deployment.
Why it mattersSome crowd-out is plausible, but the technologies have different risk, duration, licensing, and customer profiles. Phil's evidence that capital arrived after hyperscaler commitments offers a stronger proximate explanation for nuclear's recent financing.
Unsubsidized markets would direct capital toward the best long-term energy technology.
Why it mattersMarkets discipline cost but can underinvest in long-duration R&D, public goods, and technologies exposed to policy risk. 'Best' also depends on reliability, emissions, construction risk, and time horizon, not just expected private return.
Government pays half the cost of solar and wind investments.
CheckThe standard federal clean-electricity investment credit is 6%, rising to 30% when labor requirements are met. Domestic-content, energy-community, and low-income bonuses can raise some projects further, but government does not generally pay half of every project's cost.
China is deploying Generation-IV nuclear systems at scale while the United States has none.
CheckThe IAEA reports that China's HTR-PM entered commercial operation in December 2023. That is an important demonstration, but a single 210 MW-class project does not support the broader phrase 'deployed at scale.'
Phil Deutch
Phil wins by refusing the false choice. Solar incentives can distort marginal investment and still have accelerated a useful technology; nuclear can deserve support without rewriting every missing reactor as a solar-credit casualty.
Assumptions and fact checks
Renewable subsidies were an important cause of solar and wind cost reductions and deployment growth.
Why it mattersCredits improved project economics and deployment, which supported learning and scale. Global manufacturing, technology improvement, financing, and supply-chain competition were also major causes, so subsidies should not receive all the credit.
Regulation and weak demand, not competition from subsidized renewables, were nuclear's dominant constraints.
Why it mattersNuclear's long licensing and construction timelines, execution risk, and need for bankable offtake are unusually consequential. Renewable competition matters at the margin, but Phil identifies constraints more specific to the technology.
The cost of solar power fell roughly 80% to 90%.
CheckDOE reports that photovoltaic levelized costs declined by roughly 90% since 2011. The exact percentage depends on system type and start year, but Phil's range is well supported.
Did the 2025 tariff experiment justify its revenue and leverage, or impose too much inflation and instability?
Original point: Early tariff triumphalism was premature: weaker demand could hide inflation temporarily, the goals of revenue and reshoring conflict, and constantly changing country deals leave investors unable to plan.
What everyone argued
David Friedberg
Friedberg says the negotiation contains private gamesmanship and that Trump's unpredictability may be strategically optimal. He likes running the experiment but worries that a durable customs-revenue stream would become another federal spending crutch.
Ben Shapiro
Ben says the administration pursued conflicting goals—tariff revenue, reshoring, and containing China—through ad hoc country rates based heavily on bilateral deficits. He credits tax cuts and deregulation but argues tariff volatility dampened growth and made businesses unsure which rules would last.
Gavin Baker
Gavin argues the early damage was smaller than critics expected, retaliation was limited, and trying tariffs would reveal whether U.S. market leverage could move partners toward reciprocity. He trusts Trump to back away if markets clearly reject the policy, while conceding that capitalism needs stable rules.
Phil Deutch
Phil worries that tariffs on Chinese solar equipment raise U.S. energy costs and that supply-chain security is being purchased by reversing decades of beneficial globalization. He ultimately calls the macro outcome 50/50 because the unintended consequences are too large to handicap confidently.
Winner circle
Ben wins. The tariff experiment raised serious money and did not trigger an instant recession, but those facts do not answer his systems critique. Later evidence shows consumer-price pass-through and weaker investment and output, while frequent changes made it harder for firms to plan or for observers to identify success. Gavin earns credit for calibrated uncertainty; Phil correctly spots the energy-cost channel.
Commentary
David Friedberg
Friedberg is disciplined about uncertainty until he labels uncertainty itself optimal. The missing receipt is a negotiating ledger: what concession did volatility buy, and what investment did it defer?
Assumptions and fact checks
Unpredictability is the game-theoretically optimal tariff negotiating strategy.
Why it mattersSurprise can create bargaining leverage, but persistent uncertainty also delays investment and raises contracting costs. Optimality requires comparing concessions gained with those economy-wide costs, which Friedberg does not do.
Tariff revenue would reduce political pressure to shrink federal spending.
Why it mattersA large new revenue source can soften the apparent budget constraint. Whether it actually increases spending depends on appropriations and fiscal politics, but the incentive concern is coherent.
Ben Shapiro
Ben has the cleanest systems argument: a policy cannot indefinitely maximize both taxed imports and their replacement. Later evidence supports his cost warning, although it also shows the revenue was too large to dismiss as mere theater.
Assumptions and fact checks
The economy would have grown faster if the administration had paired tax cuts and deregulation without the broad tariff war.
Why it mattersThe exact counterfactual is unknowable, but CBO's modeled direction is clear: the tariff changes lowered investment and real output relative to otherwise similar policy.
Tariff revenue and reshoring are ultimately mutually exclusive goals.
Why it mattersSuccessful import substitution erodes the tax base on which customs revenue depends. Both can coexist during a transition, but they cannot both be maximized indefinitely.
The country-specific 'reciprocal' tariff schedule was tied to bilateral goods deficits rather than simply matching partners' tariff rates.
CheckThe April 2025 executive order explicitly treated large and persistent bilateral goods deficits as evidence of nonreciprocal practices and assigned higher rates to listed partners; it was not a schedule that merely copied each partner's tariff rate.
Gavin Baker
Gavin is right that the feared day-one catastrophe did not arrive and that U.S. market access creates leverage. He underprices the slow costs: tariffs passed through over months, and moving rules prevented businesses from knowing which experiment they were participating in.
Assumptions and fact checks
Limited immediate retaliation validated the view that the United States could impose tariffs at relatively low cost.
Why it mattersRetaliation is only one cost channel. CBO and Federal Reserve evidence shows domestic price, investment, output, and employment costs even without symmetric foreign tariff retaliation.
Running a broad tariff experiment would produce useful knowledge even if the policy failed.
Why it mattersReal-world experimentation can reveal bargaining leverage, but simultaneous tax, trade, immigration, monetary, and country-policy changes make attribution difficult. The knowledge benefit is real but smaller than Gavin suggests.
The 2025 tariffs produced substantial customs revenue.
CheckCBO reports about $300 billion in total customs duties collected from January 2025 through February 20, 2026, roughly half under IEEPA authority.
Phil Deutch
Phil's caution ages well, especially on energy inputs. He would have contributed more by stating a threshold: how much extra cost per watt or how much supply-chain concentration is worth paying to remove?
Assumptions and fact checks
Tariffs on Chinese solar equipment materially raise the cost of expanding U.S. electricity supply.
Why it mattersA tariff raises the landed cost of covered equipment unless suppliers absorb it; substitution and domestic production can offset some of the increase over time. The magnitude depends on exemptions, sourcing, and domestic capacity.
The long-run security benefit of localization may justify near-term energy and consumer costs.
Why it mattersResilience has option value, but it should be compared sector by sector with diversification, stockpiles, allied sourcing, and targeted subsidies rather than assumed to justify a broad tariff.
Did Apple's enormous buybacks crowd out innovation, or expose an execution problem that money alone could not fix?
Original point: Apple spent roughly $700 billion buying back shares while failing to ship enough category-defining products, and should have redirected far more capital toward R&D, acquisitions, cars, or AI infrastructure.
What everyone argued
Jason Calacanis
Jason contrasts Apple's buybacks with a shopping list of acquisitions and products it might have funded, arguing that a supply-chain-oriented CEO optimized capital returns while Siri deteriorated and competitors pulled ahead in AI and smart glasses.
David Friedberg
Friedberg says claims of Apple's demise recur whenever it appears late or derivative, as they did before the iPad became a mass-market hit. Apple's system is unusually durable, so critics should not infer permanent incapacity from one weak product cycle.
Ben Shapiro
Ben agrees that Apple has not shipped a consumer product that feels exciting in years and sees Tim Cook's visible focus on supply-chain deals as evidence that operations, rather than product imagination, dominate the company.
Gavin Baker
Gavin calls buybacks-versus-R&D a false dichotomy because Apple could afford both. He still says the company should have spent perhaps $200 billion more on data centers and AI, and judges its user experience and execution harshly while preserving the possibility that Apple can recover in augmented reality.
Winner circle
Gavin wins. He preserves the valid criticism—Apple was executing poorly in generative AI—while rejecting the lazy balance-sheet story. The filings show Apple could and did fund enormous repurchases and growing R&D simultaneously. Friedberg is right not to count the company out, but past resilience does not excuse the present gap.
Commentary
Jason Calacanis
Jason makes the buyback total do too much work. It proves Apple had alternatives, not that his retrospective basket of companies was purchasable or that bigger checks would have repaired Siri.
Assumptions and fact checks
Redirecting a large portion of buybacks to acquisitions or data centers would have made Apple a leading AI company.
Why it mattersMore compute and acquisitions can help, but model strategy, talent, product integration, data, organizational incentives, and execution determine results. Apple's available cash and growing R&D make simple underfunding an incomplete diagnosis.
The historical market capitalizations of Tesla, Netflix, Uber, Disney, Robinhood, and BMW represent realistic acquisition costs Apple could have paid.
Why it mattersControl acquisitions generally require premiums and face financing, board, shareholder, regulatory, and integration constraints. Adding past market caps is a thought experiment, not a feasible purchase ledger.
David Friedberg
Friedberg punctures the funeral mood but does not defend the current execution. 'Apple has surprised critics before' is a useful prior, not a product roadmap.
Assumptions and fact checks
Apple's integrated product system makes another late, category-winning entry likely.
Why it mattersHardware, silicon, distribution, privacy, and ecosystem integration are real advantages. Generative AI also depends on research cadence, compute, data, and service iteration that differ from Apple's classic device playbook.
Ben Shapiro
Ben offers a clean customer receipt—nothing has felt exciting—but does not turn it into evidence that buybacks caused the problem. He is a useful witness, not the winner.
Assumptions and fact checks
A long gap between exciting consumer launches indicates that Apple's leadership is over-optimized for supply chain rather than product innovation.
Why it mattersLeadership priorities may contribute, but product cycles, platform maturity, regulation, and the difficulty of creating new hardware categories also matter. Consumer excitement alone cannot isolate the cause.
Gavin Baker
Gavin wins by changing the diagnosis from cash scarcity to execution. His own giant spending number is speculative, but it does not damage the central point: Apple had money for both shareholder returns and innovation, so product leadership had to be managed, not merely purchased.
Assumptions and fact checks
An additional roughly $200 billion of AI infrastructure spending would have made Apple a serious frontier competitor.
Why it mattersCompute is necessary at the frontier but not sufficient. The claim needs a model, talent, data, distribution, and product plan before a dollar amount can be judged.
Apple could fund large buybacks and substantial R&D at the same time.
CheckApple reported $90.7 billion of common-stock repurchases and $34.55 billion of R&D expense in fiscal 2025, with R&D up from $31.37 billion in 2024. The figures do not prove optimal allocation, but they disprove a simple either-or budget constraint.
Apple continued buying back shares while increasing R&D into fiscal 2026.
CheckApple's first-quarter 2026 filing reports $25 billion of repurchases and year-over-year R&D growth driven by infrastructure, headcount, and engineering program costs.
Friedberg supplies the strongest general mechanism but weakens it with apocalyptic scale. The useful claim is that programs can create constituencies before their results are measurable; the unsupported claim is that this makes a national socialist sweep inevitable.