Anonymous analyst Crémieux: MFN drug pricing will cause 10% US price drop but 90% increase in Canada, threatens pharma R&D

Summary

Trump's Most Favored Nation executive order on drug pricing is being read by markets as a likely failure — but the anonymous analyst Crémieux, appearing with a voice changer, argues the opposite: it will succeed, and the consequences will be bad.

The policy bans geographic price discrimination, requiring pharmaceutical companies to sell drugs in the US at no more than the lowest price paid by any other developed country. The naive read is that Americans save 50–80% on drug costs. Crémieux's models put the actual US price reduction at around 10%. The redistribution, though, is severe: Canada, a small market with limited pricing power, should expect a roughly 90% increase in drug prices as pharma companies reprice globally to comply.

The mechanism is straightforward. Drug companies currently maximize profit by charging what each market will bear. The US pays more not because it's being exploited but because it consumes more — American household incomes are high, and US drug launch lags average 6 months versus roughly 6 years in the UK. Strip away that pricing flexibility and total pharma revenues fall, because the regulation forces companies to price above the profit-maximizing level in most other markets.

R&D is the deeper concern. Pharma IRRs are already below the average cost of capital. The GLP-1 boom provided a temporary lift, but generic semaglutide is expected around 2033, and Crémieux argues that once cheap generic Ozempic arrives, most patients will opt for it over tirzepatide or retatrutide, ending the boom. With IRRs already marginal and the MFN rule compressing them further, pharmaceutical R&D in the US is, in his words, "teetering on the brink." The likely result is R&D migration — and given China's current clinical trial environment, the destination is predictable.

China overhauled its trial regulations in 2016, allowing nationwide patient recruitment rather than constraining trials to a single center. The US still requires single-site recruitment for many trials, which raises costs, constrains sample sizes, and inflates the rate of false positives and negatives. As of this year, China runs more clinical trials than the US, with larger per-trial enrollment. The FDA could implement equivalent reforms via direct final rule without legislation, and could also adopt WHO Good Manufacturing Practice guidelines to cut production costs — neither has happened.

The 1991 precedent is instructive. A similar most-favored-customer rule was tried for firms, and patent-protected drugs with no generic competition saw no price change — other countries simply weren't buying them in volume anyway. Generic drugs were also unaffected. The only price movement was a roughly 4% increase for drugs facing generic competition, driven by compliance costs and the pricing distortions of uniform pricing. The net result was lower IRRs and no meaningful consumer benefit.

The Louisiana model is Crémieux's white pill. Several years ago, Louisiana negotiated a subscription deal with a manufacturer of Epclusa, a hepatitis C cure then priced at around $27,000 for a 28-day course. The state paid a flat $30 million for unlimited supply, granting the manufacturer exclusivity within the state. With zero marginal cost to the state, Louisiana screened prisons, Medicaid patients, and the broader population aggressively. Hepatitis C death rates in the state fell by roughly one-sixth within a year. Organ transplant demand dropped. The state saved money. Crémieux argues many pharma companies would accept this model — predictable, high-volume contracts at lower margins still improve IRRs — and that scaling it nationally could effectively solve the US high-price-drug problem for any drug with a generic equivalent, without touching R&D incentives.