.png)
SummaryExternal reference pricing is the most widely used price setting instrument in Europe and the least visible in the resulting data. A country picks a basket of other countries, reads their prices for the same product, applies a rule, and derives its own. Of 31 countries examined in our work on European price data, 29 apply it in some form. That makes a single national price decision a function of decisions taken in several other markets, at a defined level, on a defined date.
External reference pricing, also called international reference pricing, sets or caps a national price by reference to prices for the same product in other countries.
It is not the same as internal reference pricing, which compares a product with therapeutically comparable products inside the same country. Germany's fixed reimbursement amount is the best known internal system, and we describe it separately in our glossary. The two are often conflated in summaries, and they behave completely differently: one imports price levels from abroad, the other groups products at home.
Three parameters define any external system.
The basket. Which countries are read.
The level. Which price in those countries is read: ex-factory, wholesale or retail.
The rule. What is done with the values: lowest, average, average of the lowest three, and what happens when a reference country has no price.
Change any one of the three and the result changes, which is why a comparison of national systems without those three parameters says almost nothing.
The Netherlands calculates maximum prices from Belgium, France, Norway and the United Kingdom.
Switzerland compares against Belgium, Denmark, Germany, Finland, France, the United Kingdom, the Netherlands, Austria and Sweden, converting foreign ex-factory prices into Swiss francs at rates the authority sets on 1 January and 1 July.
Romania references manufacturer prices from a European basket and converts them into lei at the average exchange rate of the national bank for the last completed quarter.
Austria's procedure works from an EU average price, and the same regulation that defines it also supplies the margin percentages that several Nordic back calculations rely on.
Two properties stand out when you read those four together.
The levels differ, and the countries in a basket do not publish at the same level. The Dutch basket contains Belgium, which publishes ex-factory, France, which publishes a gross consumer price, and Norway, which publishes maximum pharmacy purchase prices. A basket calculation therefore has to normalise before it compares, and normalisation is the part most descriptions skip.
The exchange rate is a policy parameter, not a market fact. Switzerland fixes it twice a year. Romania uses a quarterly average from its central bank. That means a currency movement changes a national price at a scheduled moment rather than continuously, and it can move a price without any commercial decision behind it.
If your price in one market is read by several others, the order in which you launch is a pricing decision in itself.
A low price agreed early in a small market can travel into larger markets through their baskets. A delayed launch avoids that, at the cost of revenue and access. This is not a theoretical concern: it is the mechanism behind launch sequencing as a discipline, and the reason confidential rebates are so widely used, because a confidential net price does not enter a foreign basket while the visible list price does.
That last point closes a loop that runs through all our pricing work. Reference systems read published prices. Published prices are list prices. The gap between list and net is protected in part precisely because reference systems exist.
If most countries read most other countries, what anchors the system?
In practice three things do. Some countries set a price first, without a usable reference, usually the markets a product launches in earliest. Some systems mix external referencing with an internal comparison against therapeutically similar products, which introduces a domestic anchor. And some apply the reference only as a ceiling, leaving room for a negotiation underneath it.
The circularity is nevertheless real, and it has a measurable consequence: a price reduction in one market can propagate through several baskets over the following review cycles, arriving in countries that had no part in the original decision. Because review cycles differ, the propagation is staggered rather than simultaneous, which makes it hard to see in a snapshot and visible in a time series.
That is the main argument for keeping a dated history of national prices rather than only a current state. A snapshot shows where prices are. Only a series shows how a decision travelled.
Modelling a reference system means reproducing, per product and per date, what the authority itself would read. Four problems recur.
Level mismatch. Basket members publish at different points in the chain, so each has to be normalised with its own statutory rules before averaging. Our work on trade margins and on VAT exists for exactly this reason.
Coverage gaps. A reference country may have no price for the product. France carries a price for 13,648 of 20,900 packs, Spain covers roughly 20,500 of about 67,000 presentations. A rule has to define what happens when a basket member is silent, and the definition changes the result.
Pack comparability. A 30 unit pack in one country and a 100 unit pack in another are not the same object. Per unit normalisation depends on structured pack data, which several countries do not publish.
Timing. Prices move at different rhythms, from daily in Belgium to quarterly in Poland. A basket read on the wrong date reproduces a state that never existed simultaneously.
| Country | Published level | Comment for a basket calculation |
|---|---|---|
| Belgium | Ex-factory | Directly usable, daily |
| France | Gross consumer price | Needs VAT and two margins removed |
| Norway | Maximum purchase and retail price | A ceiling rather than a price |
| United Kingdom | Reimbursement to pharmacy | A different concept, outside the EEA |
| Germany | Manufacturer level, with negotiated amounts visible | One of the few markets showing a negotiated figure |
| Poland | Manufacturer price plus three further levels | Published, but quarterly |
Read that table as a warning rather than a shortcut. Four of the six rows need either a derivation, a label or a caveat before the numbers can sit in the same column.
It does not reveal what anyone paid. It does not capture confidential rebates, tender outcomes or managed entry agreements, all of which sit below the visible level. And it does not make national prices comparable in an economic sense, because it compares regulated outcomes rather than market outcomes.
What it does do is make national price levels interdependent in a way that is calculable. For a market access team that is the practical value: the effect of a price change in one market on the others is not a matter of intuition, it is a function with published parameters.
Applied in 29 of 31 examined countries, external reference pricing is the closest thing Europe has to a shared pricing mechanism, and it runs on published list prices at levels that differ from country to country. The concept is simple and the implementation is not: baskets mix price levels, coverage is partial, pack sizes differ and exchange rates are policy parameters. Anyone modelling it should state the basket, the level, the rule, the date and the currency in the same breath as the result.
Related reading: European Drug Pricing Database: how cross-border prices work and Pharma pricing Germany: how the system works.
pharmazie.com is the consolidated pharmaceutical data platform by DACON Datenbank Consulting GmbH that bundles 25+ specialist databases into a single search, exclusively for healthcare professionals. Price coverage focuses on the DACH region and a number of further EU countries, with more countries following in the coming months.
