Rational Discussion of Controversial Topics
Economics Education Ethics Foreign Policy Government History Politics Religion Science

Perverse incentives in medical research

melian          14 May 2015 05:00 PM

Imagine the following scenario. A pharmacological company wants to choose its next big research project. The options are:

a) To search for a new drug that would marginally improve the condition of patients with some popular chronic disease (such as diabetes). The drug would have to be taken daily and would be very expensive to produce.
b) To search for a drug that would totally cure all diabetes patients. The drug would have to be taken only once and would be very cheap to produce.

Obviously, the company’s chances to succeed with the second goal are far lower and this option would be seriously considered only if the potential profits are correspondingly large. In reality, however, the potential profits would be smaller.

The reason is simple. In case A, the pharmacological company might easily charge $100 for a pill that costs $50 to produce. If the pills have to be taken regularly, within several years the company would earn thousands of dollars from each patient, allowing it to recoup the money it spent on research. In case B, however, to compensate for research costs (not to mention the risk premiums), the company might have to charge $10000 for a pill that costs only $1 to produce. Realistically, the company would never get away with it, which is why it makes far more financial sense to invest only in type A projects.

Any ideas how to fix this?

Would you like to read similar articles in the future?
-2 -1 0 +1 +2

                  Post Comment                   

Recommended for You Optimates Populares Centrists

Show comments            Sort by        

FrameBenignly 18 May 2015 02:32 PM

If anything, option B is much more attractive because option A wont have a high enough margin. A marginal improvement only has marginal value to the consumer. They would right from the start have to price option A competitively with generics or they won't sell very many. There are issues with insurance which distort this basic mechanism, but it generally remains in place. A cure on the other hand has a very high market value, and they can charge a lot more for it.

And this all depends on the company having both options available around the same time ready for human trials. How often does that happen?


melian 20 May 2015 05:44 AM

A marginal improvement only has marginal value to the consumer.

As you wrote yourself, "there are issues with insurance which distort this basic mechanism". As long as consumers do not have to pay for it (which is usually the case in developed countries), they have no reason to reject even marginal benefits.

A cure on the other hand has a very high market value, and they can charge a lot more for it.

The price one can charge for an intellectual property is limited not only by its market value, but also by ability to enforce the set price. The market value of a full cure for diabetes would be measured in trillions of dollars. Do you think governments would allow any pharmacological company to earn that much?


ChristianKl 15 May 2015 03:18 AM

Costs to produce drugs aren't essential to how drugs are priced.
Gilead does manage to charge a high price for Sovaldi even through it doesn't has to be taken a lifetime.

Realistically a company would be able to charge 10,000$ for a pill that cures diabetes no matter what it costs them to produce the pill.

The effect size of a new diabetes drug also isn't clear when producing the drug. In many cases the drug company doesn't know whether a drug has to be taken a year or ten beforehand.


melian 15 May 2015 06:57 AM

a company would be able to charge 10,000$ for a pill that cures diabetes no matter what it costs them to produce the pill

Theoretically, it has the legal right to do so, but in practice... Consider countries like Argentina, China or Russia which have their own pharmacological industries, but where most patients would not be able to afford a $10,000 pill. It is hard to expect their governments to favor patent rights of a western company over the interests of their own population. Most likely, local pharmacological companies would not only produce the drug for domestic market, but also export it to other countries.

Even in developed countries the property rights are not set in stone (especially, for companies which are perceived to engage in “excessive profiteering” at the public expense). If a pharmacological company tried to earn hundreds of billions in profit, after spending just several billions on development, the governments and courts are bound to interfere.


ChristianKl 17 May 2015 04:22 AM

To me your analysis looks like you didn't spend much time actually looking at how our present system works and instead try to rely on guesswork.

Drugs aren't priced the same way in the US and in Argentina anyway.


Fwiffo 26 May 2015 08:41 AM

Anyways it means that the profit limit is more on what they dare to set it with a straight face without getting into too much confrontation with to goverment. That is the price isn't strongly responsive to market mechanics.


Fwiffo 26 May 2015 08:54 AM

There is a core issue that the patient pays for the pill when they really ought to pay for the costly part of research. However as additional pills don't require additional research our curretnly way of handling economic transactions doesn't handle it well.

I have been trying to think of an alternative system and have tried to gather them under the name fo "Contributionism". Under this system the reserach costs would be split evenly between the pills (here the pills ingredient costs are small enough to be neglible). When new patients take the drug the price per pill would actually go down. This would be sensitive whether it's a oneshot or a chronic intake type of pill. And the only way to make new money would be to make new research. It should be noted that under current system a medicine company can make new money while not furthering its medications. It's way heredox and I am not confident I can adequately or workingly argue the case for it. The most counterintuitive thing is that it doesn't feature prices that stay at a fixed level.

The thing here is copyability. A new pill beyond the first doesn't represent any genuine new work being done. Normally in economic theory you can assume that the effort per product is substantial. But for moderately many things this doesn't happen. When you design a new chair the fact that you don't have to redesign the next chair can be rounded off and only the focus on the price of wood and work be focused on. However with pills it's exactly the reverse. The design of it is the essential thing and the materials can be rounded off. This makes the prices of intellectual properties so volatile. THere is no good one to one correlation between work done and customers served.


ChristianKl 26 May 2015 11:21 AM

Do you know what you mean when you say the "research costs"?
Which costs are you talking about?


Fwiffo 26 May 2015 01:54 PM

The salaries of researches. The cost of computers time and lab equipment to search throught various compounds. The salaries of people running animal and human clinical trials to demonstrate feasibility to consumption. Biolologist and other reseraches involved with biomimicry for nature reverse engineering. Salaries of chemists of making up a chain of reactions that can effectively produce the end molecyle (same for coming up with the candidate compounds)(you make up the molecyle from parts but which parts do you combine first and from which start compounds do you aquire them?). These are all things that need to be done once per pill type but do not need to be additionally done when making more pills.

View Replies (32)