Pharma bad behaviour is not new. Credit: Erik McGregor/LightRocket via Getty


December 4, 2021   8 mins

The pandemic has not been bad for everyone. Vaccine manufacturers – notably Moderna and Pfizer – have gained billions of dollars in value from making and selling the jabs. The rise of the Omicron variant means they get to do it again; the CEO of Pfizer thinks we’ll be needing boosters every year for years, and it seems unlikely that he’ll be too disappointed about that. The vaccines have ended up in rich arms rather than poor ones, leading to accusations of vaccine “apartheid”.

But pharma bad behaviour is not new. For example: there’s a thing in patent law called “evergreening”. It’s most famously used by big pharma companies who don’t want their expensive drugs to reach the end of their 20-year patent and become available as a generic, so they develop a very slightly different version of the same drug and get a new patent on that.

Venlafaxine is an antidepressant, marketed as Effexor. As it neared the end of its patent, the manufacturer developed a new version – desvenlafaxine, marketed as Pristiq. Desvenlafaxine is what the body naturally breaks venlafaxine down into; your liver takes the venlafaxine and metabolises it into desvenlafaxine. It is also either less effective or no more effective than the original.

The patent for Effexor expired in December 2008; Pristiq entered the market in early 2009. By 2014, Pristiq was the second most prescribed antidepressant in the US, despite being “a slightly worse version of an older antidepressant with no proven advantages that also costs fifteen times as much”. (A month’s supply of Effexor at the time cost $20; a month’s supply of Pristiq cost $300.)

And it’s not the only bad thing they do. When a drug reaches the end of its patent, other companies can make generic versions. To encourage that, the US FDA says that the first company to do so gets 180 days of exclusivity, so it can establish itself in the market. “Until recently,” says Dr Vishal Gulati, a venture capitalist specialising in healthcare, “it was legal for the original company to pay off the company to not launch the product. Companies would get a license to make the generic, and then be paid to not launch it.” They were literally paying to prevent patients from getting cheaper healthcare.

What’s going wrong here? There’s a very tempting explanation, which is that big pharma is evil. But I never find that satisfying as an explanation. Instead, I’d rather think about incentive structures.

There’s an inherent tension at the heart of any knowledge-based business — anything which develops new technology or information and sells it. You want it to do two things: to create that knowledge, to advance our understanding; and to spread it around, so that the world can benefit from it. Information is a “non-rival good”: if I benefit from it, that doesn’t stop you from benefiting from it too. An example is a lighthouse. If I see the beam, and avoid the rocks, it doesn’t stop you from doing the same. By contrast, a hamburger is a “rival good”: if I eat it, you can’t.

People can make money selling hamburgers. But it’s much harder to do so by operating a lighthouse.

The trouble with knowledge-based businesses is that they are more like lighthouses than like burgers. If you create a new piece of software, or a new technology — or a piece of investigative journalism — then there’s nothing stopping me from simply copying it. And then you won’t make any money off it. That makes it less likely that you’ll spend the time and effort to create any more new things.

So we created intellectual property laws in order to stop that. Creators of a thing can get a patent or a copyright, and are granted exclusive rights to produce that thing for some number of years — 20 years, in the case of pharmaceuticals.

Problem solved, right? Well, obviously not. Because we don’t just want a world in which new things are created. We also want a world in which everyone can gain access to those things. Intellectual property rights prevent us from making extremely cheap copies of things that we already know how to make. If a foreign correspondent reports human rights abuses in Yemen, it’s best for society if as many people as possible to read that report, so that we can act on it. But intellectual property rights (and their downstream effect, paywalls in journalism) stop that from happening.

This is a direct trade-off. “My mental image of all this is we’re basically squeezing a balloon,” says Owen Barder, a development economist. You can squeeze the top, and move the problem to the bottom. Or you can squeeze the bottom, and move the problem to the top. But the balloon is still there.

And this isn’t the whole problem. We want pharma companies to develop drugs and vaccines for the developing world. But people in the developing world can’t pay as much money as people in the rich world, obviously.

Pharma R&D is expensive: sometimes billions of dollars to research one drug. And for every drug that is successful, there might be 20 that aren’t, and the research into those needs to be paid for as well. That money needs to come from somewhere. At the moment, it comes from patients in rich countries paying sometimes hundreds of dollars for pills that might, individually, cost a few cents to make. The “marginal cost” of each dose is tiny — Jacobin and Oxfam fume that the Covid vaccines, for instance, are priced at many times the manufacturing cost — but that cost needs to cover the “fixed cost” of all the R&D (and marketing, staff costs, etc) you’ve put in.

The ideal solution to this would be selling the drug to everyone at the maximum cost they’re willing to pay. Charge hundreds of dollars in the US, a bit less in the UK and EU, much less in Bulgaria, and almost nothing in Malawi. “If you could perfectly price-discriminate, you’d be charging above the marginal cost in lots of countries, and at the marginal cost for the ones who couldn’t afford more than that,” says Barder.

But that’s not possible. For one thing, if they tried it, people would buy the pills in Malawi for a few cents and fly to the US and sell them at a profit. But more important, Congress or Parliament (and the press) would kick off – why are our citizens paying hundreds of times over the odds? It would be politically impossible to do. So, instead, they make the drug at a single price, which only rich countries can pay, until the sales in those countries have paid off the R&D costs. “Drug companies are just responding to the incentives we’ve set,” says Barder. “They’re doing the thing we asked them to do.”

The temptation, when faced with problems like these, is to argue that big pharma should be destroyed, and all drug research run by publicly funded university laboratories. And maybe that would work – but it’s a huge gamble. Pharma companies do some bad things, but they objectively do make drugs that are hugely valuable to society. And it’s not that publicly funded bodies are free of bad incentives. Academica has huge problems of its own — academics are rewarded for publishing lots of papers, rather than for necessarily finding out true things. Government’s incentives are to remain popular, rather than to fund the most effective things: it would be easy to imagine lots of money going to fund treatments for picturesque children with cancer, rather than for, say, diabetes, even if it were a much less effective way of saving lives.

Still, it might work. At the moment, as Barder says, there’s a tendency to “socialise the losses and privatise the gains”: private companies get rich off research that is often begun in university departments. A starting point might be to pump lots of money into university research, let them bring drugs to market, and see if they can outperform big pharma. What would be crazy, though, is to destroy big pharma first, and then hope that our new nationalised version can keep new drugs coming through.

A more low-key version, says Gulati, might be for academic institutions to become better at demanding equity in pharmaceutical products that are based on their early research. He also suggests that countries such as the UK could negotiate cheaper drugs by offering the NHS as a source of clinical trial subjects — as has happened with Novartis’s new cholesterol drug inclisiran, aka Leqvio. That’s hugely valuable to pharma companies, and it’s something the NHS can do easily and safely, with its huge, centralised, well-protected data systems.

Those ideas might help make drugs cheaper in the UK and other rich countries: getting them to poorer countries is a different problem, with different solutions. Governments could buy out patents — if a firm thinks it can make $10 billion over the next 10 years for its product, we could say that we’ll give them the $10 billion now (or a bit less) in exchange for the rights to make the drug available at cost.

Barder likes one idea, put forward by the late economist Jean Olson Lanjouw. “Her suggestion was,” says Barder, “that if I’m AstraZeneca and I show up at the UK patent office asking for IP protection for a new drug, the patent office should say ‘Well done. Can you tell me what you expect the market value for this drug to be in all 200 countries in the world?’”

Then AstraZeneca or whoever would say “I expect most of my revenue to come from the US, UK, EU and Japan, and relatively little from sub-Saharan Africa and Bangladesh.” And the patent office would grant them a patent, on the condition that they license it for free use in those countries that make up the bottom 2% of their revenue. “It’s like a tax of 2%,” says Barder, “but those countries might well make up 80% of the disease burden.” It hasn’t been tried, as far as I know, but it’s worth thinking about, and it would avoid the problems of Congress or Parliament demanding that the drug be made cheap here — although it would only work for global diseases that affect the rich world as well as the poor — diseases like cancer, or heart disease, or hypertension. It wouldn’t help incentivise research into diseases like malaria or dengue, which have no impact on the rich world. The “advanced market commitment”, which I discussed here, and involves Western governments promising to give pharma companies a bonus for every dose bought by developing nations, might be more effective for that.

Andrey Zarur, the CEO of the biotech firm GreenLight[1.  I have done several pieces of paid writing for GreenLight over the last two years], who are producing their own mRNA vaccine for Covid at the moment, comes at the whole thing from a different angle“Pfizer was not designed to make low-cost therapeutics available to every corner of the world,” he says: it’s a 150-year-old company with settled investors and a particular way of working.

He compares it to Apple. “You have a $1,000 iPhone 13,” he says. “Who’s that designed for? My children. Idiot teenagers with rich parents.” Poorer countries need smartphones too, but the solution is not to force Apple to sell smartphones to Ethiopia at a discount. “What you need is an innovative company with different processes.” Instead of demanding changes from 150-year-old companies that are very good at the specific things they do, create smaller, newer companies which do the thing you want. “There’s six billion people in the developing world,” says Zarur. “You should be able to figure out a way to turn a reasonable profit with reasonably priced drugs.”

It’s obviously true that pharma companies have done bad things. It’s also obviously true that they’ve done marvellous things — I have relatives who are alive today who wouldn’t be without the pharmaceutical industry’s products.

Perhaps there are other systems which could have produced those drugs, other than the undeniably cutthroat capitalist system we have today. But any system would have bad incentives and obvious, easy-to-publicise disasters.

The job of government in this situation should be to find the bad incentives, the market failures, and to patch them; to make it work, to decide which end of the balloon to squeeze. The Indian government, for example, decided a few years ago that it would not accept evergreening any more — it passed a law saying that it would only offer new patents on drugs that were sufficiently different from existing ones.

In 2013, the government won a court case against Novartis, which had tried to get a new patent on a cancer drug, imatinib mesylate — a crystalline version of an existing cancer drug, imatinib. The government said that the new version was no better than the existing one and was simply an attempt to squeeze more money out of the healthcare system.

It worked. You could argue (as Novartis did) that it will reduce innovation; but the point is you can change the system, change the incentives, encourage the behaviour you want, without smashing the system entirely.


Tom Chivers is a science writer. His second book, How to Read Numbers, is out now.

TomChivers