Intended for healthcare professionals


High cost of new drugs

BMJ 2016; 354 doi: (Published 27 July 2016) Cite this as: BMJ 2016;354:i4136
  1. Mariana Mazzucato, R M Phillips professor in the economics of innovation
  1. Science Policy Research Unit, University of Sussex, UK
  1. M.Mazzucato{at}

Why government must negotiate a better deal for publicly funded research

The investigation by The BMJ and Cambridge and Bath universities into the availability of breakthrough hepatitis C drugs raises important questions for NHS England about access to lifesaving drugs.1 However, the main question is why medicines are so expensive in the first place. The pricing strategy of Gilead for sofosbuvir (Sovaldi) and ledipasvir-sofosbuvir (Harvoni) raises questions that go well beyond the UK.

The BMJ’s investigation is just an example of a more general problem. What is the right price to pay for a particular drug, and how should this be determined?

Pharmaceutical innovation should be structured so that it focuses on unmet health needs globally and delivers therapeutic advances that are affordable and accessible to all,2 not just profitable for manufacturers. This requires an approach that directs effort towards therapeutic innovations rather than “me too” drugs, and a financing and pricing structure which is transparent, focused on access, and reflects the collective investment and risk taking involved.

Drug companies have often ignored the collective element of innovation and argued that their research and development investment justifies the extraordinarily high prices for some medicines, even though they refuse to provide transparency on their costs. The Drugs for Neglected Diseases initiative has documented much lower drug development costs,3 4 and several authors have shown the extent to which taxpayer funded investments subsidises those costs.5 In the US alone, tax payers fund $32bn a year of research and development expenditure channelled through the National Institutes of Health (NIH).6

Sofosbuvir and ledipasvir, the two drugs on which The BMJ investigation is based, relied on early stage funding from the NIH and the Veterans Administration.7 Sales of the two drugs were around $12bn in 2014,8 far in excess of the $880.3m which Gilead reported for sofosbuvir related trials from 2012 to 20147 showing a complete disconnection between price and development costs.

As high prices are hard to justify based on research and development costs, drug companies have been turning to a different line of defence: they argue that their prices are proportionate to the intrinsic value of the drugs—that is, the costs to society if a disease was not treated, or if treated with the second best therapy available. “Price is the wrong discussion,” declared Gilead’s executive vice president, Gregg Alton, responding to criticism over the price of sofosbuvir, “value should be the subject”.9

But there is no consistent link between a drug’s price and the associated medical benefit.10 A study published in 2015 in the Journal of Economic Perspectives examining a sample of 58 cancer drugs approved in the US between 1995 and 2013, shows that the increasing trend in the price of these medicines is not explained by the survival benefits they provide to patients. Over two thirds of new medicines reaching the market do not represent any therapeutic advance for patients, with many patents based on a reshuffling of old combinations or on additional uses for existing ones.11

An effective pricing system should ensure accessibility but also reflect the public contribution so taxpayers don’t pay twice, through publicly subsidised research and high priced medicines. In such a system, drug prices do not need to be so much higher than manufacturing costs. We could, for example, limit patents on new medicines (the current source of company profits) and instead establish a competitive prize system that rewards well targeted pharmaceutical innovation. This would allow widespread access to drugs at competitive prices through generics, while pushing drug companies to focus their energy on delivering innovations that fulfil real medical need. In any case, patents should not be so upstream to affect scientific research, and should remain relatively narrow so as not to close off future discoveries around a therapeutic area.12 In other words, they should foster innovation, not stifle it.

Importantly, drug pricing must be completely transparent, so that governments can negotiate for better value on behalf of their populations. Public funders could, for example, retain the lion’s share of intellectual property rights (patents) produced by public research so that spillovers through licensing can be better managed to foster diffusion, and governments could ensure that the prices of new drugs reflect the burden of financial risk borne by the tax payer. In the US, the 1980 Bayh-Dole Act that allowed publicly funded research to be patented includes a clause enabling the government to cap the prices of drugs that are largely publicly funded. The US government has never exercised this right, despite repeated petitioning.13

The international debate about unsustainable drug prices, including those for hepatitis C drugs, presents us with an opportunity to rethink the therapeutic innovation ecosystem—the direction and the accessibility of the drugs that result. Realising that government has power to actively shape and create markets, and not just remain on the sidelines fixing broken ones, especially in the area of health that is heavily subsidised by the public, is the first important step to reaching a better deal.14 15


  • Analysis, doi: 10.1136/bmj.i3718
  • @MazzucatoM

  • Competing interests: I have read and understood BMJ policy on declaration of interests and have no relevant interests to declare.

  • Provenance and peer review: Commissioned; not externally peer reviewed.


View Abstract