Are pharmaceutical drug prices too high?

Hillary Clinton has proposed a new plan to bring down prescription drug prices, but so far the reception is cool.  Here is one comment:

But when I ran it by some health economists and other health policy experts, several strongly disliked the idea because it misunderstands the diversity of companies in the pharmaceutical industry. They say it would create perverse incentives that could raise instead of lower the costs of developing new drugs.

“This is an astonishingly naïve approach,” said Amitabh Chandra, a professor of public policy at Harvard University, in an email. He argues that the plan could encourage wasteful research spending without necessarily doing much about the prices charged for medications.

That is from Margot Sanger-Katz at the NYT, not the Heritage Foundation.

I would stress a different point, and this concerns pharmaceutical prices more generally, not just the Clinton plan.  Higher prices induce more innovation, and those innovations benefit patients in many countries.  Note that connection is true even if you think most innovations come from universities or the NIH rather than being hatched Big Pharma.  There is still a pot at the end of the rainbow for the significant innovators in this process.

OK, so how much does innovation go down if prices go down?

Here is an earlier post by Alex on Frank Lichtenberg’s estimation: “Thus, price controls or other restrictions that reduce prices are almost certainly a bad idea.”

Here is another earlier post by Alex, citing Megan, noting that Apple spends three cents of every dollar on R&D, and other inconvenient facts and that was from 2009.

If the advocate of lower drug prices does not have clear quantitative evidence for a conclusion of “lowering drug prices will not harm innovation very much,” commit the analysis to the flames, for it harbors nothing but sophistry and illusion.  And while agnosticism about elasticities might weaken the argument for keeping prices high, that’s not an argument for lowering prices, that is an argument for agnosticism.

This same point applies to most commentaries on TPP I might add, and intellectual property analysis.  Write it on the bathroom wall: “Without an elasticity, there is no answer.”  And scream it from the rooftops while you are at it.


The fact there is so much variation between countries with regard to pricing for what is effectively a global market indicates that the market is highly rigged already. So it doesn't automatically that reducing pricing would reduce innovation following classical market economics. btw, You could significantly reduce prices without reducing innovation by reducing the cost of innovation itself (FDA etc, which has been a previous subject on this blog) but that might have consequences for patient safety - ones that would not easily have a market based mechanism as a replacement.

Highly rigged, as in price controls in countries other than the US.

yes, and also other kinds of market distortion in USA (micro-regulation for purposes other than safety, regulatory capture, etc)

Also, monopoly pricing in the U.S. even on off-patent, generic drugs. The news story about Turing is just the latest example. Allergan, Impax and Lannett have been subpoenaed by the DoJ as part of a criminal investigation into possible anti-trust violations. The fact that Medicare now covers prescription drugs but is not allowed to bargain over their prices gives monopoly generics manufacturers a huge incentive to name their own prices, knowing that taxpayers will pick up the tab.

"Rigged" is such a lefty silly nonspecific accusatorial empty word.

Non-USA have price controls.

These prices cover marginal but not fixed costs.

So USA payers subsidize free riding world.

So if by "rigged" you mean socialists stealing from (relative) capitalists, yeah.

An alternative explanation is that Americans are simply getting fleeced.

I don't know anything about Clinton's plan but I don't think the current system is functional.

When you have an effective monopoly on a life saving drug I'm not sure supply and demand makes sense as a way to determine the price. The very fact a company would raise the price of a drug by 5000% and it's not completely unprecedented suggests to me that there aren't good price signals.

Does anyone even know the process that is used when a company is literally the only manufacturer of a life saving treatment?

Your correct the price signals are horrible in medicine.

I suspect that you mean that short term demand is highly inelastic in regard to price, but that tells us little about long term elasticity. I would bet that there are very few medicines that don't have close substitutes that could be developed fairly quickly.

It is a common problem actually in economics, what is efficient long term may not be considered fair in the short term. Disaster pricing (gouging) is another example, allowing gouging would improve storage of relief materials and speed up business response to provide additional supplies, but at the short term cost of some people.

I guess us slightly less emotional types just have to kill our annoyance at this woolly head sob story thinking and just accept that this is the way a lot of people think and that as a result society just will never be optimized.

How short is the short term? How long is the long term?

See my later post about the $1500 drug I was taking. There was a close substitute. The price signals were horrible because I had no idea how or that I even should look for substitutes (I just trusted the doctor), and the doctor didn't know or didn't care about cost.

OK I see your point now.

Gouging is actually great in the short term as well usually. Attracts more sellers and increases supply.

Someone has to pay the higher prices though, and that's who will be on CNN.

Except in this case, the gouging is being done by monopoly holders. Who often resort to any means imaginable that can be enacted without getting caught to maintain those monopolies. Good example is the updating of "the little purple pill" to renew monopoly status.

What I saw during a couple of years working in big pharma was typically far more devious than that. But still of questionable ethical consideration, at least in my book.

However, one thing is guaranteed: as soon as other sellers can enter the market and increase supply, thereby lowering cost, the monopoly holders frequently move on to newer products, where they are again monopoly holders. The company I worked with openly stated their policy was to unload any non-monopoly product. I am pretty sure they are not alone.

I would bet that there are very few medicines that don’t have close substitutes that could be developed fairly quickly.

Maybe, but you couldn't get them through FDA approval fast enough to prevent shocks.
The FDA is the long pole for many drugs.

What impresses me in this context (and many others) is how little the people debating it, actually know, about the subject at hand. As in, almost nothing.

The pharmaceutical industry is a case study in diminishing returns. After WWII, the pharmaceutical industry invented numerous innovative drugs (words like "blockbuster", "breakthrough", "astounding", etc were quite justified). Prices were quite low and the drugs were marvels. For a generation or so, that has no longer been true. Each new generation of drugs is substantially less impressive than its predecessor... But costs vastly more. A reasonable analysis shows that drug development productivity has fallen by a factor of at least 100 (perhaps 1000). Note that increased regulation accounts for only a small fraction of the productivity decline.

Cancer drugs are a tragic example. The single most important drug for treating many types of cancer is 5-FU (Fluorouracil). It was invented back in the 1950s and costs just a few dollars (literally). The newer anti-cancer drugs (which are used with 5-FU and don't replace it) add a few months of (miserable) life expectancy, but cost 1,000-10,000 times as much (literally).

Here is an easy way to understand the modern pharmaceutical industry. Imagine if every 5 years, new PCs cost twice as much, but were only half as fast. But of course, they were very popular because the government made them "free" and trial lawyers (plus Sarah Palin) got together to demonize anyone who dared to challenge the status quo.

Hillary's solution is wrong. However, as I wrote in a very different context

“The bottom line is that focusing on …. …….. is yet another example of “free market” absolutism and libertarian idolatry”

The original context was “the Jones Act”. However, substitute “drug innovation” and the point is the same.

What impresses me in this context (and many others) is how little the people debating it, actually know, about the subject at hand. As in, almost nothing.

Welcome to the Internet!

Given how so many non-generic drugs are either prohibitively expensive, or purchased through intermediaries like insurers and the state, isn't the Schumpeterian excess profit somewhat of a blunted signal? What is the goal of pharma innovation anyway? I thought it was new treatments at lower costs. If rolling over IP keep prices on effective treatments perpetually high that kind of defeats the purpose.

That doesn't make price controls a good idea. But why not start by stop doing stuff like this:

Because India doesn't recognize US patents and basically steals pharma IP?

Becasuse government will do whatever it takes to protect its cronnies' profits?

A lovely feature of a political system where political donations are capped at ... no limits with little/poor disclosure.

Would you trust a company that gives large donations to BOTH Republicans and Democrats?

No other country recognizes US patents. That isn't how patents work.

The complaint of US pharmaceutical companies is that they don't like how the Indian Patent system works.

That article is about generics that are not under patent anyway.

I think the FDA has a fairly decent rationale for banning these bathtub hooch pharma companies in India from pawning unapproved drugs in the states.

Anyone who says "but Indian generics!" with a positive undertone has clearly never been party to a Hatch-Waxman litigation about how those drugs are made.


"Anyone who says “but Indian generics!” with a positive undertone has clearly never been party to a Hatch-Waxman litigation about how those drugs are made."

Agreed. However, reputable generic producers exist in other countries (Europe, Israel).

Intent and outcome are two very different things. Yes, the FDA should be regulating it, but there is very clear evidence that they are way too bureaucratic about it. See Alex's posts during the "drug shortage" a couple of years ago. The FDA regulates not just how the drugs are made, but exact quantities as well, restricting the ability of companies to quick react to supply & demand fluctuation, etc, etc.

Is there any authoritative website that tracks when the price of particular drugs will come down due to finally reaching open competition among generic makers? I've tried looking in the past and found most informative to be websites for pharmaceutical reps for high priced brands swapping predictions about which month they'll all be fired because the drug is going generic.

It seems like there are a lot of lawsuits among drug companies that mostly serve for reaching agreement to keep the price up another year or two.

It isn't quite that simple. A generic manufacturer must still go through a very expensive process to get approved for bio-equivalency. You can certainly look up the expiration date for the patents, and for many of the most popular drugs generic manufacturers will begin the process before the patent even expires. (A few years back, the makers of the birth-control drug Yasmin even started to manufacture their own generic called "Ocella" a little while before patent expiration. The strategy involved getting it on approved formularies for insurance companies.)

However, the most recent issue at hand has nothing to do with patents. The rare drugs that Turing/Retrophin (and many others) buy and then price hike have patents that are all expired years ago. So that's why Tyler's last line making a dig at TPP critics doesn't make any sense at all.


"Is there any authoritative website..."

Type "drugs going off patent" into Google for some data.

"It seems like there are a lot of lawsuits among drug companies that mostly serve for reaching agreement to keep the price up another year or two"

Quite a bit of truth to that. Sometimes criminal.

Higher prices induce more innovation, and those innovations benefit patients in many countries.

It is obvious to me that higher prices should lead to more innovation. New drug production has slowed down. We need more of it. Therefore drug prices are too low.

How anyone can argue with this, I do not know. We all benefit from drug innovation. The solution may not involve only prices though. We may need less regulation.

The correct answer is that the low-hanging fruit have been picked. Drug companies are now seeking to develop niche drugs and me too drugs that they can charge very high prices for. What we need to do is incentivize companies to make effective products for conditions most in need of treatments, not just pay whatever they want for the crap they bring to market regardless of how well it works, which is what we do now.

"Drug companies are now seeking to develop niche drugs and me too drugs that they can charge very high prices for."

But they *can't* charge very high prices for 'me too' drugs when there are close off-patent alternatives. And what are 'niche drugs' except treatments for relatively uncommon diseases? And how would you provide incentives for developing drugs for relatively small market niches other than prices? And are people suffering from these uncommon conditions better off with ineffective treatments rather than expensive ones? (We know insurance companies and state care systems would be better off if those new, expensive treatments didn't exist, but not the patients).

Do you know how much they charged for Nexium when it came out? The perceived value was purely marketing driven.

I am not saying they shouldn't be developing those niche drugs. Of course people with those conditions need treatments, but the cumulative health gains across society are not going to be even close those from the blockbuster drugs of the 90s, for example. This is the new reality.

The question for you: is there a limit for aggregate spending on drugs? Even if they are very effective, should drug spending supplant spending on other types of health care, on education, on transportation infrastructure? What if it approached 10% of GDP? Is that ok? It seems everyone wants to restrain overall health care spending--it is almost a no brainer. But with drugs, the question makes people get very uncomfortable.

"The question for you: is there a limit for aggregate spending on drugs? Even if they are very effective, should drug spending supplant spending on other types of health care, on education, on transportation infrastructure?"

Why do you assume drug R&D crowds out spending on education or infrastructure but not, say, smart phones or Hollywood blockbusters?

Because so much of drug spending is public money.

Also, just for frame of reference, Americans spent about $10 billion seeing movies last year, while Medicare Part D spent more than that just five drugs. This does not even include the more expensive injectible drugs for cancer and other serious conditions that are covered under Part B.

Nexium is just purified and overpriced Prilosec.

And what better way to determine what conditions are most in need of treatment than the price signal?

Jan's personal regulatory board?

Jan September 24, 2015 at 7:14 am

The correct answer is that the low-hanging fruit have been picked.

I don't think that is true. All innovation is done by a small number of young men working on their own. We have driven them out of the medical field. Now you have to be a big multinational to produce drugs or maybe, at best, part of a team at a university. Too much paper work. Too many regulatory hurdles to over comes. Hence no really good new drugs.

Drug companies are now seeking to develop niche drugs and me too drugs that they can charge very high prices for.

True. And niche drugs for people who are either going to die anyway or are not really sick. You don't want to develop a drug for someone who might have health problems because they will sue. Fen-fen was probably the turning point. Because of some utterly insignificant number of people taking it developing heart problems (it was for obese people. Of course they had heart problems) a major drug company was sued out of existence. No one else is going to risk it.

What we need to do is incentivize companies to make effective products for conditions most in need of treatments, not just pay whatever they want for the crap they bring to market regardless of how well it works, which is what we do now.

Well sure. But how? The system is broken. We aren't going to fix it any time soon. The drug companies are our best bet. We need to let them make money. By preference on real drugs. That means tort reform.

Ultimately we all benefit a lot more from new drugs than we do from stupid regulation and paying extortionate prices. You can't put a price on another six months of life.

The cost of developing a new molecular entity (new drug) has risen pretty dramatically over the past few decades, while the regulatory burden has remained the same, or even declined as FDA scrambles to "facilitate innovation." Many people's answer will always be that regulation is the problem, but those standards haven't changed and productivity has gone down. It is telling us that the easy stuff has been done.

We do let the drug companies make money, and lots of it. It is not really helping in recent years. It is possible that if other counties paid anything close to what we do for drugs the massive influx of money would allow large scale research and innovation that produced some great discoveries. But I don't see that happening anytime soon. Also, I am skeptical that more dumb money is the answer, as companies already have relatively high profits and continue to sink almost as much into marketing as they do into R&D.

the low-hanging fruit have been picked

Not even close to being true. We've barely started reaping the rewards from sequencing the human genome. Prior to that, drug development was fishing in the dark. But there's a vast new field opening up in which scientists can actually design drugs tailored to particular genes - to turn on and off specific genes based on an actual understanding of what those genes do.
And the genome is vast, vast beyond measure.

The genome revolution in drug development has been on the cusp for, what, 20 years now? See my response above on evidence that low-hanging fruit have been picked.

Hazel Meade, Jan,

"the low-hanging fruit have been picked"

Rather than debate this, take a look at the data. See "NME Output versus R&D Expense – Perhaps there is an explanation" (

Drug development productivity has fallen 98-99% (perhaps more).

I think in this case it's kind of like the lag in waiting for investments in personal computers to pay off in productivity.

Having sequenced a genome of some particular human gives us ONE jigsaw puzzle, but we're not even quite sure how it's cut yet. It could be decades before major economic returns start to flow from medical advances which rely on the genome, but I think it's time will come.

I would tend to consider that more of that underlying research still needs to be done, which is more likely to happen in university laboratories than corporate R&D.


"The correct answer is that the low-hanging fruit have been picked"


Very true with tons of statistical data to support it. A classic case of "The Great Stagnation".

That's not an argument for higher drug prices, it's an argument for more research funding.

I took a drug that cost $1500 a month. Course my deductible was $4000. So by April of every year I was paying $90 instead of $1500. No idea if the insurance company paid the full $1500. There are substitutes (with slightly worse side effects) that cost as little as $4 a month (yes $4). I only discovered this when I temporarily lost my job and insurance. I went to one of those cheap doctors that won't take insurance just cash and he told me about it. I don't think my original doctor even new what various drugs cost.

Same here. My doctor knew about it though.

She asked about my insurance and I said it was so-so. She prescribed the $8 medicine instead of the $180. I had to take 3x a day instead of once.

Drug prices and reimbursement are just about the furthest thing from a free market in this world.

What the US should do is tie drug reimbursement to evidence. Unfortunately, in Medicare Part B, which pays for the really expensive infused drugs, there is no link at all between what they pay and the effectiveness of the drug. Medicare basically covers anything and everything and companies charge what they want. This is also true for six categories of drugs in the privately run Part D plans. The regulations state that the Part D plan must cover all drugs in these classes (e.g. cancer, anti-depressants), no matter what the price and no matter what their effectiveness. It is idiotic and no other country on earth does it like we do.

If one thinks a stupid US drug policy is required for innovation, they are deeply confused.


I'd look at the price difference between medicaid and medicare. It isn't that much.

Also prices in Europe vs here. Again some outliers but Harvoni is only about 20% of a discount in the EU and Japan.

I'd guess medicare has to cover 65% of rx in the US? And everyone else just follows the medicare price.

As a stalking horse, I suggest that Medicare put out a price list of "here is how much we will pay for a cure for hepatitis-C", and the prize should be in the tens of billions.

It will be the same price, but the fact that it was announced ahead of time will defuse some of the dunderheads.

It's silly to think that the only way to innovate drugs is by profit-oriented companies. Government-funded research can do a pretty good job.

It already does a very good job. Government grants allow for a lot of experimentation and failure. Almost all of the innovation we see being brought to market can be quickly traced back to science grants at the academic level. (I am aware this is probably an unpopular notion with the readership of this blog)

But the innovation is only the beginning of the process. It is massively expensive and time-consuming to run all the trials and get FDA approval to bring a new drug to market. Initially promising treatments can fail only after many years and enormous investments. Like this:

$800 million and 16 years down the drain. Pfizer ended up not just canceling the drug but also exiting the cholesterol-drug business entirely and shuttering the massive R&D site.

Sure, that's why patents are granted for drugs which pass. I don't think anyone is arguing against awarding patents.

Pfizer still makes Lipitor... so they are very much in the cholesterol business.

Yes, I'm sure they'll continue to crank out Lipitor as long as they can sell it at a profit in competition with generic alternatives, but they've given up developing any new cholesterol drugs.

Almost all of the innovation we see being brought to market can be quickly traced back to science grants at the academic level

Not really, unless you get REALLY stupid about "well they wouldn't know to research drug X for hypertension unless someone 20 years earlier had researched hypertension at a school."

When you pose a question that can actually be answered -- "where are the drugs that make it to market initially discovered?" -- you get less than a quarter coming from academia (which includes academia funded by pharma and biotech).

"R&D" isn't really homogenous. If you need a simple heuristic, business and capital markets do well pumping a lot of cash into decently developed ideas and scaling them to national and international markets. This is a unique skill-set, that academics do not have. This research is complementary, not antagonistic.

The incentives in the current system are wrong. we produce drugs for some conditions, like toe nail fungus, of the insured but not a vaccine for Ebola which is a greater danger

Toenail fungus is probably a much bigger health problem in the US than Ebola.

Correct, few Americans contracted ebola. If drug companies dropped hundreds of billions of dollars on every stupid CNN-scare, they'd have no money, and I would have uncontrolled high blood pressure, and my brother would have untreated schizophrenia, and my Wife would've been dead from a tumor in her head. All REAL problems REAL Americans REALLY face, and not tail-risk CNN-ratings-boosting pandemic coverage.

BBC reports about 11,000 people died worldwide from Ebola. But only a handful were Americans and it wasn't terrorism so I guess you shouldn't give a shit.

"BBC reports about 11,000 people died worldwide from Ebola. But only a handful were Americans and it wasn’t terrorism so I guess you shouldn’t give a shit. "

The US DOD alone spent $140Million on Ebola treatment research in 2014. Exactly how much money did your country spend?

So why should a private pharmaceutical company put hundreds of millions into developing an Ebola vaccine? Do you think they're going to recoup those costs in West Africa?

Urstoff has done an excellent job of making Joan's point.

Those 11,000 people in Africa earned an average of $500 per year.

It's reasonable that human civilization doesn't allocate scarce pharmaceutical research resources to help a small group of extremely low income people when it could meet the needs of a large group of high income toenail fungus sufferers.

Ebola wasn't a danger to anyone who could afford to pay for the drugs.

Do you have any idea how many people suffer from hypertension or diabetes?

There is a bias toward drugs for chronic diseases as opposed to acute conditions? As in, a drug for type II diabetes may be taken for the rest of the patient's life, but an antibiotic to treat an acute infection might be taken for only a week or two.

And in addition, if you do develop a new antibiotic, its use will be mostly restricted to infections that are resistant to existing antibiotics, so usage may remain low for many years (perhaps for the life of your patent).

"Apple spends three cents of every dollar on R&D, and other inconvenient facts"

It's not so much an inconvenient fact as a frankly irrelevant one. Apple is a company that is brilliant at designing and marketing high-end consumer electronics. It is not, however, a company operating on the frontier of science and technology, as a quick reading on the history of smart phone or tablet technology (and their largely failed efforts to stop Samsung from selling comparable products) will make clear. It used existing technologies, put them into a pretty package with a nice user interface and then outsourced the production to other firms. Intel with its R&D at 15% of revenue is a bit more relevant.

"OK, so how much does innovation go down if prices go down?"

Hillary's proposal is to link R&D spending to revenue, not to cap prices. That approach may be misguided and result in wasteful R&D spending but it is tough to see how it would cause less innovation.

This is leaving aside the elephant in the room, which is the over-pricing of generic, off-patent pharmaceuticals. This seems to be enabled by a combination of barriers to entry (yes, over-regulation by the FDA may play a role), restrictions on importing generic drugs from other countries which do not seem to have this problem at all, and Medicare's inability to bargain over drug prices. Whatever the arguments for restricting Medicare's bargaining power for patented medicine, it is hard to see any viable argument for why Medicare shouldn't bargain over old, off-patent drugs.

The recent drug price controversy has nothing to do with innovation. In these cases, a company sees an old, cheap drug with only one manufacturer. Unsurprisingly, some people really depend on that drug as part of their treatment regimen. The company takes advantage of the lack of competitors making the same old, generic drug by buying it and jacking up the price 5000%.

Paying this company more--or less--for the product they bought is not going to have any impact on innovation whatsoever.

"Higher prices induce more innovation, and those innovations benefit patients in many countries." Notice how Cowen combines two concepts in one sentence, one (higher prices induce more innovation) debatable, the other fact (innovations benefit patients in many countries). He's good, very good. My question: where is the "clear quantitative evidence" that higher prices induce more innovation, innovation that justifies the higher price? As for Ms. Clinton's proposal to cap monthly drug costs at $250, such cost shifting (higher premiums paid by everyone) is the nature (and rationale) of "insurance", and it would have appeal to those who believe they may need a similar drug at some time. Most people believe it won't happen to them and, therefore, it is just one of the many risks in life that the individual must bear. In matters of risk, I always recall Professor Shiller's dictum that in a perfect world all risks would be shared, the per capita cost would be nominal, output much, much higher, and the standard of living for humanity vastly improved. In a world of gamblers, however, Shiller's dictum has little appeal.

If there was a betting market for TC's activities, we would all be rich. Of course he immediately jumps up to defend high prices on drugs, as he will always defend a private entity's right to charge whatever they want for a product. Of course, what he fails to mention in his shallow and misleading post:

* The drug in question is monopolistic, which is why a 5000% increase in price could be conceived of in the first place; were there a competitive marketplace, obviously that would never have happened.

* Higher drug prices *in the US* benefit the world all over, because their nations have price controls. We are subsidizing the entire world's medical research; while that may be good for the rest of the world, it's certainly not good for Americans.

* And finally, why so many people seem to eel Economics is such a junk science in the first place; it's theories so often seem disconnected from the real world and regular people. TC's basically saying "It's OK if some poor people who's lives used to be saved easily and cheaply die today, because in the long run, innovation!" Until economics stops viewing the world as an aggregate and a timeline, and can instead speak to individuals and what's important to them today, it will always have a more limited impact on our world than it should, and that's a shame.

Let's see a result that says limiting Medicare's ability to negotiate with drug companies incentivizes those companies to conduct research. Because that's what you're arguing here.

The economist's coverage of the situation is *so* much better than your naive application of a flawed economic model.

I don't see many people showing their elasticities, but I do see a variety of rhetorical moves designed to skirt the point. It's to the Humean flames for many of your comments!

Supernormal profits get chucked into the Humean flames because you get to set the parameters of the argument, huh?

You suggested that a 10% decrease in drug prices would result in a 5% drop in innovation. Since 2008 (coincidentally around the time the Lichtenberg paper was written), the price of branded drugs has increased 110%. Where's all the innovation?

Ask Martin Shkreli he knows the answer!

Ignorance of alternatives is not an argument for the status quo.

Without elasticities of your own, the only point you yourself can make is rhetorical.

Your question was how much lower drug prices would harm innovation.

But even assuming all those excess profits went directly into innovation if price signals aren't operating properly we can't even assume that we're spending an appropriate amount on innovation in the first place.

There's also a question of objectives, the societal goal of drugs isn't to maximize innovation, it's to maximize welfare. One highly affordable and moderately effective drug could produce a far greater benefit than an extremely expensive and extremely effective drug.

There is a deeper point here that seems to have been missed. Where is the evidence that pharmaceutical innovation is a good thing? Or stated differently, at what price does pharmaceutical innovation become a net burden on society?

Say a drug company invents a new anti-cancer treatment that adds 1 month of life expectancy for 100,000 people at a cost of $10 billion dollars. That $100K per month of (miserable) life. My numbers are hypothetical. However, some drugs are actually worse than my example.

That's a (large) net loss to society as a whole.

Inventing new ways to make society worse off is not an activity we should encourage.

That's only a problem because society is stupid enough to spend $100K to extend the life by a miserable month. If it wasn't medication it would instead be some other procedure.

The fact that government JUST CAN'T STOP ITSELF from spending that $100K isn't the fault of someone who presents products to the government to buy.


"That’s only a problem..."

I generally agree. However, we have Libertarians arguing in favor of the status quo (at MR for example). Real Libertarians oppose both government price controls and health care paid for by government. Pop (trendy, fashionable, hip) Libertarians aren't willing to take the (popularity) hit that goes with opposing government handouts in health care. So we a toxic combination of passionate opposition to price controls (the "free market" as a sacred cow) and either tacit approval of handouts or explicit advocacy of freebies (remember "forced savings"?).

Of course, this doesn't make for a coherent philosophy of government (or governing), but it does make for nice happy talk. It's Bread and Circuses thinking. Drug companies get unrestricted prices and profits. Consumers get "free" medicine. Everyone "benefits" from "innovation". What's there not to like?

Indeed, Bush put all of this into action. The Medicare Plan D contained explicit prohibitions on drug price negotiations, combined with "free" drugs for seniors. Old people got "free" meds. Drug companies get higher prices and volumes. What's there not to like?

Pity that the Bush administration ended with an implosion of stellar proportions. Bigger pity that the Bush crash was the inevitable consequence of Bush's policies.

If society paid me more money I would give society more innovation.

If society paid you more money your bank account would get bigger...the second part is mostly bullshit...companies are interested in maximizing their own profit, not society overall. Just ask Marin Shkreli.

Yes, Moreno, we know it. But for some reason many others are fooled into believing the drug companies are withholding their discoveries from us until we pay them enough money.

There are a fair number of promising orphan drugs. They just don't get pursued to see if they really work because the market is too small to justify the expense of the approval process. Think congress passed a law to try and help this issue, no idea of the results.

At least in the UK, the thresholds for approval and reimbursment are different for orphan drugs than for common drugs

"the market is too small to justify the expense"

This is called economics. It explains why a lot of great and wonderful ideas do not exist as actual products.

I agree its economics. But, for small volume potentially life saving drugs. Why not cut a few corners on fda approvals to make the economics better.

Tyler strangely thinks that the onus is on the purchaser of the drug to show how much innovation will go down quantitatively if we start paying less for drugs. What if we just set up a metric to measure the value of a drug and pay based on that? What a novel concept.

If we think that innovation really is tied to price, then we ought to see the most innovative, effective drugs enter the market at higher prices. But guess what, there is basically no correlation between a drug's health impact and its price. The single best predictor of a new cancer drug's price is not by how many months it extends life, or quality of life it brings the patient. No, the price of the most similar drug is the best predictor. So a new cancer drug, even if it is worse that its closest competitor will almost always enter the market a few percentage points higher than the other drug. Why? Because that is what we will pay for a new drug. Not because we reward innovation or because we think this new drug is any better.

The fundamental question is whether the extremely high price for drugs (in the US only, as other countries are not as stupid as we are) is correlated to innovation. I would argue that it is clearly not.

"Higher prices induce more innovation, and those innovations benefit patients in many countries."

Higher prices for *actual* innovation could have this effect, yes. What we often have, though, are higher prices for NON-innovation -- higher prices for monopolizing a generic market, higher prices for me-too drugs that are no better than an existing drug, higher prices based on the pharma company's asserted costs rather than the marginal value of the drug, higher prices for no reason other than the buyers' refusal to bargain.

So the post is premature in calling for "clear quantitative evidence" on the effect of "lowering" prices. Let's first have clear quantitative evidence that paying high prices for NON-innovation and refusing to let the buyers negotiate or pay for marginal value (as they do in any actual market) is somehow producing more innovation.

Y'all really need to have some faith in markets. There are many extremely strong players who have strong incentive to push back on these practices, or start alternative businesses.
Meanwhile Hillary Clinton's crusade against "gouging" has annihilated market caps in the bio-tech sector. I am sure Hillary Clinton will just have typical Dem answer of shoveling money at the problem, which will turn out awesome, just like Cash For Clunkers and Solyndra.

Have you read her plan, or was that a partisan cheap shot?

The part where she's limiting business expenses or the part where she's capping monthly costs? We're talking about typical naïve garbage, and none of her supporters can offer any useful defense of her plan here in these comments.

If a tweet from a not yet nominated candidate can "annihilate market caps in the bio-tech sector", the sector is way over priced.

She is not nominated, not elected and cannot determine policy for another year and a half in any event. Not to mention that a GOP House [and possibly Senate] would hardly pass legislation putting price controls in place.

I think investors were just looking for a reason/excuse to dump their stocks.

With barriers to market entry like the hundreds of millions or billions price tags to bring a drug to market, I think that normal "faith in markets" sort of logic starts to lose credibility for very simple theoretical reasons.

When there is a billion dollar barrier to entry, one can expect that encumbents will try to milk any monopoly position that the hold to the last drop, regardless of the effect that this may have on aggregate welfare. Why? Because we count on profit-maximizers to maximize profits, not social welfare, particularly in non-"perfect" markets (greater than zero barrier to entry, for example).

Where does the price data come from? The stated price of a drug is often much, much higher than the price paid by the consumer. This is probably most true for drugs of chronic conditions, where the drug company would rather pay the patient's deductible and then have insurance pay for the drug for the rest of the year than have a patient not be able to afford it at all and not buy any of the drug. R&D is a sunk cost, so some sales are better than no sales.

Wouldn't that be reflected in higher insurance premiums?

Insofar as insurance companies paying anything raises premiums, yes, but in these cases the drug company is simply absorbing the patient's deductible.

Big pharma and big kickbacks! Hide the true cost from the consumer so there are no valid price signals!

Perhaps but it's also the case in the USA the idea that someone should have to make the choice of spending X or dying in the very near future, or dealing with a high level of pain is too often treated as a choice they should never have to make. They will but we have a number of industries largely operating on a denial of that reality.

Why not just treat single-source patent-expired drugs as natural monopolies? The companies can do whatever they want while the patent is on (as they do today), and then when the patent has been expired for some amount of time, a reasonable profit margin is imposed given the market's natural monopoly. This would prevent both bad incentives: TC's fear that lower drug prices will discourage innovation (surely people are not inventing new drugs with the notion that in 50 years they will have a monopoly) and the practice of wildly marking up drugs with long-expired patents (these are a natural monopoly due to the cost of FDA approval versus size of market).

Or just let anyone who can show they're making the same chemical that had been under patent can sell a duplicate. It's not hard to make pyrimethamine (people in India do it every day for a few cents a pill):

But it's expensive enough to prove to the FDA that you're got a bioequivalent drug that no one is going to do make a competitor, so Turing can crank prices for at least a short period of time. The rules for making a generic drug should be similar to making a vitamin pill (does the the pill contain 99-101% of the labeled chemical on the side? and did you include only food grade additives?).

I could get behind this. People already pay a premium for non-generics. At least this way, there'd be some conceivable reason for doing so.

Obviously other countries should pass laws setting price floors for certain drugs. Innovation would shoot through the roof.

OK that was smartass but let's work with it. Tell companies, if you produce a drug that prevents X disease with Y efficacy, we agree to pay for it at $__ a dose, inflation adjusted, for at least ten years. I guess the problem there would be adverse selection - the companies would work on the "easy" drugs rather than the ones that would do the most good.

I think it was clearly stated and not at all smartass.

Do high prices still encourage innovation when pharmaceutical companies are being started by hedge-funders solely looking to buy up the rights to sell drugs that are priced lower than they could be?

This is not just about drug prices, but on the crappy patents that are protecting the high drug prices. Many of the patent protecting these supernormal profits are on very obvious applications of well-known chemical compounds.

Why do strong patent rights over crappy inventions in the pharma industry increase innovation, but strong patent rights over crappy inventions in the tech industry decrease innovation?

My proposal is that the rhetoric of such is due to the identities and relative political power of the entities that are holding the crappy patents.

A lot of truly ignorant comments above (and got about halfway through before giving up)- and you can identify those comments because they all claim most drugs are really the result of government grants, or they complain about marketing costs and me-too drugs.

It isn't that the low hanging fruit has been all picked- what has happened is that the hurdle for getting drugs to the market is higher than it was in the past- even the recent past. Many of our most famous treatments- the miracle drugs so to speak- wouldn't even be pushed into clinical trials today simply because they would never be approved by the FDA today, or that the legal risks would be too great, given the profiles of the drugs themselves. I worked in drug discovery for nearly 20 years, and the effort to improve the profile of clinical candidates was an ever escalating process.

Given the regulatory structure that applies (and that includes the tort system), I am not sure there is a good substitute for the present patent system to give the proper incentive to develop new drug treatments- and especially if one wants to continuously improve the profile of what does make it to the market. Drugs with a large enough market will eventually be generics with multiple potential manufacturers. We could have the government buy out patents or give large, one-time prizes for new drugs, but then we would be arguing over what drugs deserve to be bought out, or what developments deserve the cash reward, and how big that reward needs to be. For example, would a large cash reward be appropriate for a new drug that 100,000 people in the entire world might need to survive, or even just live a more normal life.

Now, as for the generics of today that are niche markets and limited number of manufacturers (the ones that are in the news today for gouging), there are two possible approaches- make it easier for competition to enter/re-enter the manufacture of the drug, or simply treat such single-producer drugs as monopolies- an appropriate use of anti-trust law, in my opinion. I think it would be better to do the former- there is always the danger than niche generics like that end up with no supplier at all at some given reimbursement rate.

Is there literally ANY evidence to back up your assertions? Everything I read says the FDA's basically approved every turd that gets dropped on their doorstep in the past 10 years. Are there reams of new regulation making it tougher to get drugs to that door step? Are insurance companies/medicare/medicaid refusing to pay for new drugs? You said that something has changed. What's changed?


There is plenty of evidence that regulatory approval is harder to get now vs. 30-50 years ago. The Thalidomide tragedy led to a notable tightening of U.S. drug laws (even though Thalidomide was never approved in the U.S. under the old laws). It's also true, that the trial lawyers are a much bigger burden than they were historically.

However, the counterargument is also true. The low-hanging fruit is long gone. Even without the FDA (the current laws) and the trial lawyers, drug development productivity was doomed to crash. 99% to 99.9% is a reasonable guess (for the decline in development productivity).

Let me offer two extreme cases. Back in the 1920s, two Canadian researchers isolated Insulin in an tiny lab without almost no money. The did get the Nobel prize in 1923. No clinical trials were needed, because the effect was so dramatic. These days pharmaceutical companies run clinical trials with 10,000+ participants. Why? Because with a smaller sample size it would not be possible to show a statistically significant benefit.

Peter, Great Stagnation does not suggest IP laws are less relevant or less needed. If we are really are coming up against a hard frontier, any marginal innovation is more costly and requires stricter IP law to ensure profit. It does also suggest that we should be spending less money in this research field, too, and more in alternative health improvement frontiers.


I am not opposed to IP laws. My job depends on IP laws.

"If we are really are coming up against a hard frontier, any marginal innovation is more costly and requires stricter IP law to ensure profit."

At some point, innovation has negative utility to society as a whole. We are probably well past that point in pharmaceutical research. The core problem is that negative utility drugs are "free" to patients and highly profitable to producers.

From Jan

"Also, just for frame of reference, Americans spent about $10 billion seeing movies last year, while Medicare Part D spent more than that just five drugs"

The number one drug is Nexium ($2.53 billion). Anyone, with any knowledge of chemistry, will tell you that the Nexium patent was a fraud. Nexium is the S-enantiomer of omeprazole (Prilosec). The separation of enantiomers was well known chemistry by 1848 (Louis Pasteur)

Disagree with your first two paragraphs. There is no evidence for what you say. Every regulated entity hates the regulation, so I am not surprised you would say what you did.

The innovation argument I think is a variation on the trickle-down theory -- or perhaps once can flip those. I think it applies in quite a number of cases and history certainly seems to support the idea from simple casual observation. That said, we don't seem to see the same working when looking at things like the wealth effect in equity and monetary policy. SO the question might be to what extent and under what market structures that line of thinking works.

With regard to the lack of elasticities one might see that as a setting where market theory that say markets produce good outcomes socially (the private vices, public virtues/invisible hand themes) will fail. In such cases other market regulation mechanism will be sought and will provided superior outcomes I suspect. So I think there are answers but they won't be in the form of price competition among price setting competitors or consistent with the goal of allowing pure profit maximizing behavior by producers.

"Higher prices induce more innovation, and those innovations benefit patients in many countries."

Is this limited to drug prices? Shouldn't we then want higher prices for everything in medicine? Would this be limited to medicine? Should it be assumed that higher prices always lead to more innovation everywhere? Does this happen instantly and if not, what are the costs when it takes a while?


Good point. Seems innovation in consumer electronics (a free market utopia compared to medicine) seems to continue at a rapid pace despite only short periods of high prices.

I think some types of innovation in electronics are far easier to measure. Could that explain the difference? (I have no idea...)

For example, 2Ghz is precisely twice as fast as 2 Ghz. But when comparing two drugs of similar efficacy, you might be weighing very different profiles of potential side effects, for example.

Consumer electronics is just the application of physics that is pretty much understood while biology seems to remain much hazier. Maybe medicine is more trial and error and less straightforward than consumer electronics.

Consider that a gene may produce a given protein, but a) the same parts of that gene may also be spliced differently and used as a part of another gene, and b) the protein that is produced may spend different percentages of time in different bio-molecular positions which can be estimated by computers using our extensive knowledge of biochemistry, but for many purposes remain essentially unknown because the protein may "prefer" different formations in the cellular context which is of greatest medical relevance.

These are among the reasons that a lot of trial and error are involved.

In short, yeah, I think you have a pretty straightforward explanation.

If the advocate of lower drug prices does not have clear quantitative evidence for a conclusion of “lowering drug prices will not harm innovation very much,” commit the analysis to the flames, for it harbors nothing but sophistry and illusion.

If the defender of high drug prices does not have clear quantitative evidence for a conclusion of “lowering drug prices will harm innovation very much,” commit the analysis to the flames, for it harbors nothing but sophistry and illusion.

From a recent Health Affairs article:
"A recent investigation of 54 new cancer drugs approved by the FDA over the past decade found that 74 percent had no proof that they extended life, and few had proof for improving quality of life. “A veneer of innovation” is how Peter F. Thall, of MD Anderson Cancer Center in Houston, described the reliance on surrogate endpoints such as tumor shrinkage, or biomarkers, for drug approvals. With the costs of these drugs running upwards of $100,000 annually, these ineffective drugs are a drain on individual patients, families, and the health care system. About 90 percent of new drugs approved in the United States and Europe do not have any clinical advantage over existing drugs, according to teams of physicians and pharmacists in several countries."

So the empirical evidence suggests that we are currently paying top dollar for very little overall innovation. Maybe, though, we could pay less overall and get more *actual* innovation. For example, what if we started paying for drugs according to their marginal value in extending quality-adjusted lifespan, and not just whatever the manufacturer demanded? Then manufacturers would have a clearer incentive to find innovative ways to extend quality-adjusted lifespan.

In any event, I suspect that the relationship between "drug prices" and "innovation" is quite a bit more complicated than the simplistic view that they have a linear and monotonic relation at all points.

we are currently paying top dollar

There's the problem: the "we."

As long as there is some agency out there filling the role of "make some drug and we'll pay for it," people will make whatever they can. And as long as that agency is the government, "we" will be paying for it.

Private insurance companies have refused to cover drugs as not being worth it. Take a wild guess what the public reaction was.

It seems that drug companies are turning towards marketing SUPPOSED benefits rather than creating ACTUAL benefits in new drugs.

Why aren't we talking about how political correctness is making it hard or even impossible for pharmaceutical companies to take race differences into account in the making and marketing of drugs? Can someone quantify the inefficiencies caused by this?

Was not aware of this. Have you heard of Bi-dil? It's a drug just for African Americans with heart disease. There is no great stagnation.

That would be a pretty stupid situation if it were true.


"Why aren’t we talking about how political correctness is making it hard or even impossible for pharmaceutical companies to take race differences into account in the making and marketing of drugs?"

It's a real issue, but not a major one. Sex differences are actually more important and doctors are highly aware of them.

"Can someone quantify the inefficiencies caused by this?"

Quite small.

Probably a bigger issue is differential tendencies towards disease in different racial and ethnic groups. There is some evidence that Hispanics are genetically prone towards diabetes (the official position of the CDC is yes). There is weaker evidence that blacks have a tendency towards hypertension (probably not true). Sadly, certain groups have genes that increase the probability, and aggressiveness of some serious cancers.

How well does medicine address these issues...? More work needs to be done. Does PC constrain how well medicine handles these questions? Yes, to a degree.

You cast some of my earlier posts into the flames, so here's an estimate of "innovation elasticity" from the BMJ (see "Foreign free riders and the high price of US medicines"). They suggest that just over 1% of pharma companies' gross sales is reinvested in basic research.

Sort of hard to reconcile with your claimed figure of an elasticity of 0.5, so where's the discrepancy?

Given how much pharmaceuticals spend on marketing (some of which adopts highly questionable marketing practices), compared to other countries where drug prices are lower, perhaps the effect of lower drug prices would be less annoying drug ads on TV and a great share of (remaining) profits into actual R&D?



Using TV ads to promote marginal (in efficacy) drugs with astounding prices paid for by insurance (public and private) is a parody of how capitalism is supposed to work. But wait... It's Libertarian.

I have discussed this with a few doctors (which I am not). In private they are appalled. However, when a patient specifically asks for Nexium, and insurance will cover the bill, they aren't inclined to take the grief of saying no.

The reason that people put up with high pharmaceutical prices is the belief that it will provide for more R&D. Of course, it's at least partially true.

But let's be honest here. There are major principal-agent problems. Theoretically, the goal of any business is to maximize shareholder value, not social welfare.

And "worse" than that, is the obvious fact that the goal of executives, salespersons, etc. is to maximize this year's income (make big bonuses), not the long-term shareholder value of the company.

Of course we can wave around some hidden-hand wands and hope for the best, but it is clear that the current set-up will have problems. That doesn't itself militate for intervention. But could the benefits of intervention (restraining bad sales/marketing tactics via real prison time for rule breakers, forcing executives and sales persons to take half of bonuses priced in shares 20 years out, price controls, etc.) outweigh the costs of the status quo? I think the answer is yes.

This is one of my favorite blogs of all time, but on the topic of pharmaceuticals, the posts generally seem uninformed.

For example, a BMJ article this week ( ) states as follows:

"85-90% of new drugs provide few or no advantages for patients.4 The FDA’s flexible criteria and low threshold for approval do not reward more research for breakthroughs but instead reward more research for minor variations that can clear this low threshold. The growing number and widening application of expedited review programs are accompanied by evidence that many of the clinical trials accepted by an industry compliant FDA have features that contribute to biased results and compromised science (see box).. . . The result is an ever larger number of drugs approved on the basis of weaker evidence and in shorter time periods."

For another example, the FDA's report "FY 2012 Innovative Drug Approvals" points out on page 13 that "The effectiveness of Voraxaze, an orphan drug, was evaluated on the basis of a single clinical study of 22 patients, all of whom received the drug." No control group! Yet the reform proposal (to have the FDA raise the required p-value for possibly life-saving drugs) seemed to be completely ignorant of the fact that the FDA is *already* doing just this -- in fact, the FDA has already loosened the p-value standard so much that it sometimes doesn't require a p-value at all!

It would be far more interesting if Tyler or Alex actually addressed head-on all of the evidence as to how much (or little) innovation is actually occurring, how drug markets actually work, and how the FDA actually approves drugs.


"This is one of my favorite blogs of all time, but on the topic of pharmaceuticals, the posts generally seem uninformed"

This sentence has one clause too many.

"It would be far more interesting if Tyler or Alex actually addressed head-on all of the evidence as to how much (or little) innovation is actually occurring, how drug markets actually work, and how the FDA actually approves drugs"

Facts tend to interfere with ideologies. They need to be carefully avoided.

Isn't it quite possible that most Democrats (maybe most USAers) prefer less effective drugs with lower prices than better drugs with higher prices?

To illustrate the point, an argument could be made that we would all be better off if we never tested for or treated cancer but once a treatment exists we feel we should get it with all the nasty side effects. The same with new drugs maybe Democrats feel so bad about poor people not being able to afford them that they prefer that had not been invented?

Here is bully EZEKIEL J. EMANUEL talking about end of life care: That’s how long I want to live: 75 years.


"Isn’t it quite possible that most Democrats (maybe most USAers) prefer less effective drugs with lower prices than better drugs with higher prices?"

Very hard to say. Before "Death Panels" became part of the Republican attack on the ACA, the Democrats were conspiring with the trial lawyers to sabotage any attempt at cost control in medicine. The case study in these things is the Nelene Fox scandal. From Wikipedia.

"Nelene Hiepler Fox (January 9, 1953 - April 22, 1993) was a California woman diagnosed with advanced breast cancer. She requested her HMO to pay for High-Dose Chemotherapy and Bone Marrow Transplant (HDC/BMT) to treat her cancer. Her health maintenance organization, Health Net, declined her request, stating this therapy was an unproven, experimental therapy. She ultimately received her BMT after raising $212,000, but died eight months later, aged 40. Her estate sued and received $5 million due to the denial. Subsequent research proved that HDC/BMT was a harmful treatment for breast cancer patients, and it is no longer used."


"By September 1994, the Federal Employees Health Benefits Program, which covers employees of the United States' federal government, began requiring all of its health plans to pay for HDC/BMT for advanced breast cancer. The State of California subsequently passed a law requiring health insurance to pay for HDC/BMT. Subsequent research reportedly shows that HDC/BMT for advanced breast cancer does not extend life, worsens quality of life, increases the number of days hospitalized, and costs an additional $55,000 "

However, it was a wonderful political issue for the Democrats (bad HMOs versus suffering patients) and richly profitable for the trail lawyers. It turns out that it was also a goldmine for the doctors and hospitals. After Nelene Fox, no insurer dared to say no or question the bill. Billions for "medicine" that overtly harmed patients.

One title from a book of the period should give you an idea of how the left and Democrats exploited the issue

"Health Against Wealth: HMOs and the Breakdown of Medical Trust"

Wait, shouldn't the assumption be that patents don't increase innovation and it be those promoting greater rewards to patents be on the hook for justifying it?

Priors matter.

NeLytmqthotPYzVCYB 3358

Comments for this post are closed