Buried in the very long thread at Crooked Timber wastefully eviscerating Megan McArdle’s data-free and slanted musings on medical research is a very interesting nugget of actual fact from Daniel Davies (comment 132 (sic)). This deserves more fame:
For Glaxo [GSK], the USA is 39% of revenues by location of customer and 30% by location of subsidiary, but only 27% of operating profit (source: p.116 of the 2008 annual report, which is page 118 of the pdf file.)
Well, well. DD couldn’t find similar breakdowns for other companies. GSK is an intelligent and aggressive company, but no more than its competitors, so the working Bayesian assumption is that since Pfizer (US) and Roche (Swiss) face the same pressures in the same environments, the results are unlikely to be very different. National health insurance isn’t bad for Big Pharma: after all, six of last year’s top ten companies by sales are based in countries that have it. The list is unstable, as the blockbuster drugs go through their life-cycles, but here they are:
What possible explanations are there for DD’s finding? National health care does squeeze prices; American patients and insurers pay twice as much per pill as the European average (here, table 5, page 14). We can I think rule out differences in manufacturing, distribution, regulatory and administrative costs: the industry’s costs are mainly R&D and marketing. (WHO: “After marketing costs, R&D is typically the second biggest item in the spending profile of large pharmaceutical companies.”)
I see two hypotheses:
1. The GSK accounts are no basis for anything because the company (like presumably its competitors) is gaming the numbers to avoid tax, labelling R&D costs as US-based because they are tax-deductible, shifting profits through transfer pricing to low-tax jurisdictions, etc. Plausible, but can you really get away with this for billions rather than millions? Tax domicile would matter a lot here, and GSK is a British company, so the direction of the effects may not generalise. Big Pharma relies heavily on government goodwill, and has very strong incentives not to risk killing the goose that lay such golden eggs by behaving like hedge funds over tax. Count me very sceptical.
2. Evil healthcare socialism constrains marketing expenditures for all participants. These are larger than R&D so the impact would be considerable. Only the USA allows marketing of prescription drugs to consumers: that’s an asymmetric $4bn a year extra for the industry alone. I suspect American rules on junkets (travel to conferences, continuing education seminars in the Caribbean) are weaker, though I may be wrong. The big difference arises from a far looser and more diverse structure. Take the British NHS. What’s the point of flying the board of the Rochdale Primary Health Care Trust to Cancun? They aren’t allowed to take such perks (as I read the rules), and anyway take their pharmaceutical decisions from NICE.
The companies will of course try to game any system. Bribing NICE would on paper be cost-effective but it’s horribly dangerous: there’s a very high risk of failure, discovery, disgrace and even jail. The Orwellian-named NICE is of course staffed by very thick-skinned and doubtfully nice people, and it works under intense public and professional scrutiny. The pricing negotiations are a lot less transparent. There was recently a scandal in Italy. When I started to live in France, the Sécu used to reimburse a lot of harmless me-too formulations with aspirin, vitamin C and the like, presumably to prop up a weak domestic drugs industry: less when I left. Anyway, even if such practices do restore rents to the industry, they do so at much lower overall cost than the marketing fiesta in the USA.
To the extent that drugs marketing is a zero-sum arms race, all the corporations benefit if it’s forcibly cut back.
Here’s Dr. Jeffrey F. Caren, a cardiologist in Los Angeles, with a display of the hundreds of pens given to him by the drug industry:
.. Relentless pressure on doctors, exerted by pharmaceutical marketing masquerading as education, is common.
How much worse can it be in the USA?
Footnote on Megan McArdle
I’m less inclined than Mark to cut McArdle [Update: cheap shot play on her name removed, sorry, we don’t choose them] any slack, she’s not an obnoxiously clever undergraduate (memory lane …) but a paid journalist under a professional obligation to check her facts, which she doesn’t. Her baseless assertion that the USA is responsible for “80-90%” of medical innovation is a dinner-party-circuit factoid. The USA is responsible for just under half (WHO: 42% in 1998, here table 2.2; British government: 49%, year not given) of world pharmaceutical R&D. To read her, you would think that only American teams could have come up with the discovery of the HIV virus, a gene therapy path to an AIDS cure, monoclonal antibodies, the bacterial origin of ulcers, computer tomography for scans, heart, hand, and face transplants, collagen scaffold trachea transplants, and the big daddy, the cracking of the genetic code. Would any of them have been possible without American science? Surely not: but interdependence works both ways.
The big difference between the USA and the rest of the developed world in medical research isn’t anything to do with Big Pharma, but the asymmetrical high public funding through the NIH: about half. Germany (here, table 2.2) and Britain (idem, figure 2.2) have much lower ratios of public to private health research expenditure – I guess the same would hold for Japan. So I agree with Mark that some burden-sharing would be fair.
This is all the more necessary in that there’s growing evidence that the Big Pharma model of innovation is broken. Quite apart from the shoddy marketing of diseases to fit the drugs available, and the natural bias towards the maladies of the well-heeled, the pace of real innovation has slowed. But should we be surprised? Having rival research teams work competitively in secret was the plan the Third Reich chose to develop a German atomic bomb, with the success we know. The academic, DARPA, and open-source model of competition plus full information-sharing seems to work a lot better. If national health care leads to a further socialisation and internationalisation of medical research, we should see that as an opportunity not a problem.