Patents allow pharmaceutical companies to sell drugs at several times the price they would be able to get in a competitive market. Only a small proportion of the extra money spent by the public goes into research and development (R & D), despite the latter being the supposed benefit of the higher prices.

I looked at a sample of fourteen of the largest pharmaceutical companies.

Between them they made an average gross profit margin of 77%. Having heard the pharmaceutical companies and other advocates of patents, you might imagine that most of this 77% would go to fund research.

In fact only 16% of revenues goes into research. 33% goes into sales, general and administrative costs. A look at the few companies (Novartis and Roche) that separate sales and marketing from administrative and general costs shows that the bulk of this does go into sales and marketing.

The high expenditure on sales is defended as a means of providing doctors with information. It is hard to see how it is better than cheaper ways of providing doctors with impartial and fuller information in the form of journals, training and access to research.

Heavy expenditure on sale also encourages the use of the patented drugs that are heavily marketed rather than cheaper out of patent drugs. I have previously written about a number of other unintended consequences of relying on patents to fund drug research.

The system is so bad that it would be better simply to abolish patents and for governments to increase their funding of researchers. While state sector mechanism are not always efficient, it is hard to imagine that it would do worse than the current system that:

• costs a lot more than direct funding would,
• misdirects money, and
• creates incentives for behaviour that is actually undesirable.

The biggest problem is a zeitgeist that leads people to even consider an alternative to everything possible being done by the for-profit private sector.

Note on methodology: The companies in the sample were chosen by taking the 20 largest companies from this list and removing those whose non-pharmaceutical revenue were more than 20% of the total, unless they disclosed the relevant numbers separately for their pharmaceutical busines. All relevant reported numbers in the most recent available annual report were converted to dollars at current rates and then summed. These simplifications would have introduced small errors, but no consistent bias. The companies in the sample were: Pfizer, Glaxosmithkline, Sanofi-Aventis, Novartis, Roche, Merck, Astrazeneca, Bristol Myers Squibb, Wyeth, Eli Lilly, Amgen, Takeda, Schering Plough and Astellas.