Launch order has quietly become a pricing decision

For most of the last two decades, the sequence in which a company entered global markets was an operational question. You went where you had a partner, where the regulatory path was shortest, where the commercial team had bandwidth. Price was negotiated market by market, and each negotiation stood more or less on its own.

That is no longer true. Under most-favored-nation and reference pricing, the order of entry has become a pricing decision, and most teams have not caught up to it.

What changed

A growing web of pricing rules now ties the price a product can command in one market to the prices it carries elsewhere. In the United States, a series of measures now benchmark US prices against prices in a defined set of comparable countries: the May 2025 executive order,¹ the voluntary manufacturer agreements announced from September 2025 onward,² the proposed GLOBE and GUARD payment models,^{3 4} and the pharmaceutical tariff proclamation of April 2026.⁵

The specifics matter more than the label, because they are now written down.

The proposed GUARD model would set a Medicare Part D benchmark by reference to 19 OECD countries, selected on two published criteria: a GDP per capita of at least 60 percent of the United States on a purchasing power parity basis, and an aggregate real economy of at least 400 billion dollars, also PPP adjusted.⁴ The resulting set is Australia, Austria, Belgium, Canada, Czech Republic, Denmark, France, Germany, Ireland, Israel, Italy, Japan, the Netherlands, Norway, South Korea, Spain, Sweden, Switzerland and the United Kingdom.

Read that list against your own launch plan. Most mid-size launch sequences touch four or five of those countries in the first eighteen months, usually the ones with the fastest regulatory routes rather than the ones with the best prices.

The companion GLOBE model would do the equivalent for Medicare Part B, on an international benchmark drawn from the same kind of economically comparable country set.³ Both are proposed as mandatory for manufacturers of the drugs they cover, not voluntary.^{3 4} GLOBE proposes an October 2026 start; GUARD proposes January 2027, running five years and applying to randomly selected geographic areas covering a quarter of Part D enrollees.^{3 4} Comment periods on both closed on 23 February 2026 and final rules had not published as at this writing.

The implication is easy to state and hard to absorb: a low-price launch in a small market can move the reference point for the largest market in the world. A concession made to enter one country, made for entirely sensible local reasons, can follow the product into every market whose price is benchmarked against it.

Reference pricing is not new. Roughly 23 of 31 European countries already used it as their primary systematic pricing criterion more than a decade ago.⁶ What is new is the density of the web, and the fact that the largest market is now inside it.

Why teams miss it

Most commercial and market access teams still run market by market. Country teams optimize for their own launch, their own access negotiation, their own timeline. That is how the function is organized, resourced, and rewarded.

The interdependencies exist, but they live in spreadsheets and in the heads of a few people, rather than in the decision itself. A pricing concession in one market and its downstream effect on a reference market are usually owned by different people, evaluated at different times, against different targets. By the time the connection is visible, the sequence is already set.

The result is that companies optimize each market in isolation and inherit the interactions by accident.

This is not a theoretical failure. When Germany introduced international referencing as a supporting criterion in 2011, prices fell in the countries that reference Germany by between 3.5 and 7.5 percent, and rose in the countries inside Germany's reference basket by 5.5 to 8.4 percent, because manufacturers negotiated harder in those markets in anticipation of Germany looking them up later.⁷ Nobody announced that strategy. The prices moved anyway.

The companies that adjusted did so because someone modelled the basket. The ones that did not simply paid the difference.

What a company should do instead

Three shifts follow.

First, model markets as an interdependent system, not a list. The relevant question is not "what price can we achieve in each market" but "what does a price here do to prices there." The reference relationships between markets are the structure of the decision, not a footnote to it.

They are also, for the most part, public. The Dutch maximum is the arithmetic mean of four named countries, revised at least twice a year.⁸ Brazil caps the factory price at the lowest price charged across 14 countries plus the country of origin.⁹ Japan adjusts a domestic price downward if it exceeds 1.25 times the average of the US, UK, German and French prices.¹⁰ Canada benchmarks new medicines against the median of a regulated eleven-country basket from which the United States and Switzerland were removed in 2019.¹¹ None of this is confidential. It is simply not assembled.

Second, optimize the sequence, not just the markets. Two companies entering the same set of countries in a different order can end up with materially different global outcomes. The sequence is a lever in its own right, and it can be modeled before a single launch. The European Commission's own simulation work found that launching in high-GDP countries first, versus low-GDP countries first, produced significantly divergent average prices over a ten-year horizon.⁶

Third, treat every small-market access negotiation as consequential beyond its borders. A price agreed to unlock a modest market may be the price that resets a much larger one. Small markets are no longer only small opportunities; they are inputs to the price of every market that references them.

The honest caveat is that this cuts the other way too, and the evidence is not uniform. The Netherlands evaluated its own 2020 basket change and could establish no difference in availability before and after.¹² A Canadian cohort study of 349 new patented medicines found launch rates fell after the 2017 policy shift, but fell similarly in comparator countries, and found no overall attributable effect.¹³ Sequencing is a real lever. It is not the only thing moving.

The shift underneath

None of this makes market entry harder in principle. It makes it a different kind of problem. Entry has moved from a sequence of local decisions to a single interdependent one, and the tools most companies use, built to answer one market at a time, were not designed for that.

The companies that adapt will treat launch sequencing as a pricing strategy, modeled deliberately and early. The ones that do not will keep making locally rational decisions that turn out, market by market, to be globally expensive.