« Department of Oh-Oh, a continuing series | Main | The Federalization Fraud »
Scream this from the rooftops
We can't just bargain down the prices of pharmaceutical drugs without adverse consequences. It is hard to measure the effects here, but yesterday I came across this piece of serious empirical work:
EU countries closely regulate pharmaceutical prices whereas the U.S. does not. This paper shows how price constraints affect the profitability, stock returns, and R&D spending of EU and U.S. firms. Compared to EU firms, U.S. firms are more profitable, earn higher stock returns, and spend more on research and development (R&D). Some differences have increased over time. In 1986, EU pharmaceutical R&D exceeded U.S. R&D by about 24 percent, but by 2004, EU R&D trailed U.S. R&D by about 15 percent. During these 19 years, U.S. R&D spending grew at a real annual compound rate of 8.8 percent, while EU R&D spending grew at a real 5.4 percent rate. Results show that EU consumers enjoyed much lower pharmaceutical price inflation, however, at a cost of 46 fewer new medicines introduced by EU firms and 1680 fewer EU research jobs.
Here is the paper. Here is a non-gated copy. Here is my column on medical R&D. Here is a previous installment in the series "Scream this from the rooftops."
Posted by Tyler Cowen on November 15, 2006 at 06:40 AM in Medicine | Permalink
Comments
If I am reading this correctly, prices controls led to about a 30% drop in new medicines produced, although part of that is presumably a substitution effect of shifting new medicine creation to the US. Is that a fair reading?
I suspect that whether most people care is partially dependent on what those 46 fewer medicines might have been. While different versions of existing medicines are important to people who experience differing side effects, presumably the medicines foregone were the marginal ones. Would the typical new medicine be a statin inhibitor or a different version of Viagra? Still, it is hard to believe that you can drop 30% of new medicines without eliminating something valuable.
People may generally not care, because it is hard to feel the loss of something you never had. It is not as though drugs that people use were taken off the market; non-existent drugs have no constituency. We could say that one of them might have been a cure for cancer, but that sounds more than a little hyperbolic.
30% is the number I would want to scream from the rooftops, since 46 may or may not be a big thing without context. "You can have cheap drugs now, but you will lose 30% of future treatments every year, now and forever." I wonder how many people would call that a good trade, since the EU had about 45% cheaper drugs in the last period shown. Hey, 45 is bigger than 30!
Posted by: Zubon at Nov 15, 2006 8:13:05 AM
Wait, I said that wrong. The US had about 45% more expensive drugs. That means the EU had about 30% cheaper drugs.
Posted by: Zubon at Nov 15, 2006 8:16:22 AM
Isn't the real story that we won't lose them all, but will pay royalties to Indian and Chinese pharma companies (where the "high-paying jobs" have gone as well)?
Are the American people subsidizing the drugs, the jobs, or both? Why not, it's the American way?
Posted by: Chi at Nov 15, 2006 8:23:04 AM
I am going to cut and paste a bit here...I see no one answering the questions posed by Dean Baker regarding the efficiency of the patent system:
Go here: http://www.prospect.org/print/V12/2/baker-d.html (my apologies for the extended cut and paste, please read the rest at the link)
"Absurdly high prices have put lifesaving prescription drugs out of reach for millions of Americans and for hundreds of millions of people in developing countries. In large part, patent protection is to blame. The patent system is a trade-off: Consumers pay a monopoly price on a drug for 17 years to provide incentives for firms to undertake research that yields large profits. But the patent system is not the only way to support drug research. Alternatives that have a proven track record of success already exist--specifically, research supported by foundations, universities, and the government. Shortening patent terms and putting most pharmaceutical research in the public domain would cut costs for consumers as well as for government. And contrary to industry propaganda, doing so would not reduce innovation.
This idea may sound radical, but look at the numbers. The drug industry currently spends around $18 billion a year on socially useful research. If research spending grows at a real rate of 3 percent annually, expenditures would total about $240 billion over the next decade. By comparison, the prescription drug plan proposed by Al Gore would cost $250 billion. If the government just picked up the full tab for drug research and did away with patent protection, most seniors would end up paying less for drugs than if the Gore plan were put in place. The federal government would save $10 billion right off the bat, in addition to savings on drug costs incurred in Medicare or Medicaid. And the rest of the population would pay 75 percent less for prescriptions.
When the patent system for prescription drugs is challenged, or when price controls or liberalized importation rules are proposed, we're warned about the perils of stifling innovation and tampering with the market. But patents are themselves a market distortion--an explicit form of government intervention: The government grants a prize (a 17-year monopoly) to people who innovate. This monopoly in effect transfers income from consumers to pharmaceutical companies. But this is neither the only nor the most efficient way to promote innovation.
Pricing patterns in nations without effective patent protection, and on drugs after their patent has expired, suggest that the free market price for drugs averages about one-quarter of the monopoly price. In the United States, which spent approximately $106 billion last year on prescription drugs, consumers would save $79 billion if this patent protection were dropped.
What consumers get for this $79 billion in higher drug prices is $22.5 billion in domestic research from the pharmaceutical industry (another $4 billion is conducted abroad). Consumers pay more than three and a half dollars to the drug industry for every dollar of research induced by patent protection. Another two and a half dollars goes to industry profits and marketing--and to the legal costs, campaign contributions, and political lobbying needed to protect and extend the industry's patent monopolies.
Worse, much of the research conducted by the drug companies is directed not toward breakthroughs to better our lives but toward finding ways around the lucrative patents of competitors. When a drug company scores a big hit with a drug like Viagra or Claritin, competitors patent comparable but slightly different drugs so that they can enjoy a slice of the market. In a world with patent protection, this kind of limited competition can be beneficial to consumers. After all, some competition is better than none; also, a copycat drug may incidentally benefit people who react poorly to the original drug. In a competitive market, however, pure copycat research would be dropped in favor of research funded by public sources and likely to produce real improvements. [See Merrill Goozner, "The Price Isn't Right," TAP, September 11, 2000.] "
...continued at http://www.prospect.org/print/V12/2/baker-d.html
Posted by: theCoach at Nov 15, 2006 8:30:39 AM
I'm a bit confused by Tyler's argument. If there are high potential profits in the US, why aren't European pharmaceuticals doing R&D for drugs they can sell in the US?
Posted by: EclectEcon at Nov 15, 2006 8:42:24 AM
I'm a bit confused by Tyler's argument. If there are high potential profits in the US, why aren't European pharmaceuticals doing R&D for drugs they can sell in the US?
Posted by: EclectEcon at Nov 15, 2006 8:45:03 AM
Since the market for pharmaceuticals is an international market, with negligible shipping costs, I don't see how price controls in European markets can explain the relative shift in R&D between European and American firms.
Even if prices are controlled in Europe, drug companies, both European and American, have been able to sell their drugs in the US.
I would expect European price controls to have the same effect on drug companies profitability no matter where they were located.
Posted by: Brock at Nov 15, 2006 8:47:52 AM
I'm also a bit confused by Tyler's argument. The government is the largest buyer (in the US at least) of a number of products. Does Tyler expect that the number of models of cars will increase and the price of cars will decrease if the government ceased negotiating fleet sale prices?
Posted by: cactus at Nov 15, 2006 8:59:04 AM
The obvious point being that of course innovation will follow the relative money*, but did the extra money result in the most efficient incentives for innovation?
I have not read the paper in question, but from the excerpt this key question seems unaddressed.
For anyone interested in alternative ways of funding innovation consider the establishment of research prize for drugs that acheive some objective.
*Assume for illustration that innovation of drugs is constant. Assume that country A spends $100/year to companies developing drugs, and country B spends the same. Assuming also that there is some ability to work internationally, if country B were to up its payments to $200/year we would expect that more of the static amount of new drugs would come from country B.
Posted by: theCoach at Nov 15, 2006 9:01:42 AM
I'm with Brock. Tyler's argument here doesn't seem to make much sense. European companies can (and do) sell their drugs in the U.S. There are no formulary rules that bar drugs from abroad or give assistance to local drugs. And American companies, for that matter, sell their drugs in Europe (since even with price controls, drug prices in Europe are far above the marginal cost to the pharma companies). So where is the evidence -- or even the theoretical argument -- that demonstrates the connection between European price controls and European investment in R&D?
Posted by: William Goodwin at Nov 15, 2006 9:02:45 AM
TheCoach: "Alternatives that have a proven track record of success already exist--specifically, research supported by foundations, universities, and the government."
These institutions don't have a proven track record of bringing drugs to market. I'm also wondering whether any honest financial analysis has been done on the costs of academic "pure" research. I also genuinely doubt that gov't funding can effectively and efficiently replace private-sector research. Baker's solution is full of holes.
Posted by: jult52 at Nov 15, 2006 9:08:29 AM
Posts in this Blog frequently point out that the EU economy as a whole did not perform well over the time period studied in this paper. How much of the decline in the stock performance, profits, and R&D spending in the EU was due to general economic conditions?
Posted by: joan at Nov 15, 2006 9:10:50 AM
Isn't the real story that we won't lose them all, but will pay royalties to Indian and Chinese pharma companies (where the "high-paying jobs" have gone as well)?
"Have gone?" The jobs that have gone to Indian and Chinese pharma companies already are not the high-paying or particularly research-oriented jobs discovering new compounds. The big market that's already gone to India and China are things like "replicate lots of known compound X."
If there are high potential profits in the US, why aren't European pharmaceuticals doing R&D for drugs they can sell in the US? and
Even if prices are controlled in Europe, drug companies, both European and American, have been able to sell their drugs in the US.
From the first page of the paper itself, if anyone had bothered to read it:
"Although many pharmaceuticals are sold worldwide, EU (U.S.) firms typically sell proportionately more in the EU (U.S.) (see Vernon, 2005)." They also demonstrate a correlation between the proportion of drugs sold in the US and the rate of growth in R&D research among firms. The more a company sells to the European market compared to the US market, the slower the R&D growth.
Another story is how much of the R&D done by European pharmaceutical companies has switched to being located inside the United States, to be located closer to the target market. This paper measures by firms, so it doesn't account for how much of, say, Glaxo's research is being done in the US now.
Another two and a half dollars goes to industry profits and marketing--and to the legal costs, campaign contributions, and political lobbying needed to protect and extend the industry's patent monopolies.
Ridiculous to imply that companies wouldn't market or lobby if a different plan were implemented. In addition, marketing is designed to increase sales and profitability; good marketing pays for itself. Lots of drug marketing is designed at getting people who otherwise wouldn't know that their condition is treatable to go see their doctor and get a drug. (For "lifestyle" diseases like allergies or impotence that people might choose to live with rather than seek treatment.) In cases like that, marketing simply increases overall sales of the drugs, spurring investment into research.
[A]lso, a copycat drug may incidentally benefit people who react poorly to the original drug. In a competitive market, however, pure copycat research would be dropped in favor of research funded by public sources and likely to produce real improvements.
The author sounds completely ignorant when he pretends that we can a priori know that whether a "copycat drug" will have fewer side effects or benefit people who react poorly to the original drug. In reality, that tends to be discovered at the end of the long, expensive process of research. There is absolutely NO way that public sources could determine at the onset whether a drug would be helpful in that way. Thus, there is no way to guarantee that one would fund only "real improvments," and the inclusion of the comment convinces me that the author of the piece has no idea what he's talking about.
A suggested compromise: People who believe that federally supported research is the answer should concentrate on federally supporting research into drugs aimed at diseases with small or poor (or both) patient populations, diseases that it is often unprofitable to develop a drug for. Unfortunately, like any political process, federally supported research money goes towards popular diseases suffered by the middle class and up, so it's not really better than for-profit research as far as that goes.
Even in Europe, the track record of publically supported pharmaceutical research is very thin. Yes, there are plenty of stories of excellent basic research that does lead to drugs. But people constantly underestimate the amount of effort to take research demonstrated on a plate to a real drug.
Posted by: John Thacker at Nov 15, 2006 9:12:21 AM
How many of the new drugs in the US were for controversial illnesses for which pharmaceutical intervention is a new development? I'm not just thinking of the obvious case of Viagra, but also drugs to counteract behaviours that have recently been medicalised - think Ritalin, for example.
Posted by: h2o at Nov 15, 2006 9:14:02 AM
So where is the evidence -- or even the theoretical argument -- that demonstrates the connection between European price controls and European investment in R&D?
From the first page of the paper itself:
"Although many pharmaceuticals are sold worldwide, EU (U.S.) firms typically sell proportionately more in the EU (U.S.) (see Vernon, 2005)."
They demonstrate further a correlation between the rate of growth in R&D and the proportion of sales in the US as opposed to the EU.
There's your theoretical argument. I'm sorry that it wasn't contained in Professor Cowen's post, but required you to click on the free link.
Posted by: John Thacker at Nov 15, 2006 9:14:56 AM
From the paper: "Given the EU’s restrictive price regulations, one would expect EU firms to be more negatively affected than U.S. firms because proportionately more of their revenue is likely to come from the EU. Unfortunately, few firms report comprehensive data on the
geographic distribution of firm sales. Compustat Segments database includes sales by geographic area for firms that report it in their financial statements after 1997. Between 1998 and 2004, 43 (13) U.S. (EU) firms report separate U.S. and EU sales. These data cover only about 40 percent of the firms in our sample. Nevertheless, they show that on average U.S. (EU) firms generated 76 (53) percent of combined U.S. plus EU sales from the U.S. This proportion is stable across years for the U.S. firms, with 76 percent in both 1998 and 2004, and only slight variations in other years. But the proportion for EU firms
increases consistently from 43 percent in 1998, to 57 percent in 2004. This illustrates how EU firms’ sales have recently shifted toward the U.S., perhaps because the U.S. offered better pricing than the EU during the period."
Posted by: Tyler Cowen at Nov 15, 2006 9:17:22 AM
but also drugs to counteract behaviours that have recently been medicalised - think Ritalin, for example.
Ritalin's a bad example if you're talking about recent research or recently medicalized behaviors.
1) Not on patent, available generically.
2) Invented in 1954.
3) Invented by a Swiss company (Ciba, a precursor to Novartis.)
4) Prescribed for hyperactivity starting in the 1960s.
However, one will concede that prescription of Ritalin did greatly increase in the 1990s in the USA, but it hardly has to do with patents or research or anything.
A great deal of the advertising for drugs has to do with "lifestyle" diseases or illnesses for which pharmaceutical treatment is a new development, though, yes. That's because potential patients may not know that help is available. (If you have cancer, of course you're going to get whatevertreatment; advertising is unnecessary.) This effect probably biases perception; since most of the advertising is for this kind of drug, people may heuristically assume that most new drugs are this kind.
Posted by: John Thacker at Nov 15, 2006 9:21:06 AM
Also from the paper (regarding shift to US-based R&D):
"Consider that in 1990, European firms spent 73 percent of their R&D in Europe and 26 percent in the U.S. By 2002, they spent 58 percent in Europe and 34 percent in the U.S. Major European firms have moved their research or operational headquarters to the U.S., including Pharmacia in 1995, Aventis in 1999, GlaxoSmithKline in 2000, and Novartis in 2002... Indeed, Thiers, Sinskey, and Berndt (2006) show that the number of multinational clinical trials performed in England, Germany, and Italy between 2002 and 2005 has declined by over 15 percent, while the number in the U.S. has been stable."Spain and France set a price for Bayer’s second best-selling drug (Adalat) 40 percent below its price in England. When Bayer restricted supply to French and Spanish wholesalers who were reselling their supplies in England, the European Commission fined Bayer 3 million Euros in 1996, and ordered them to stop restricting supply. World-wide sales of Adalat, which had been growing strongly, subsequently fell by four (three) percent in 1998 (1999), mostly due to French and Spanish wholesalers reselling Adalat at relatively low prices. As a consequence, Bayer cut its European R&D spending by one percent in 1998 and by 14 percent in 1999. Conversely, Bayer increased its U.S. R&D spending by eight percent in 1998 and by 31 percent in 1999."
Posted by: John Thacker at Nov 15, 2006 9:31:25 AM
Perhaps this is addressed somewhere else in the paper, but the fact that EU companies sell more of their drugs (proportionally) in the EU is entirely unrelated to the relevant IP question: what *variety* of drugs is available in the EU.
Here's a thought experiment: Imagine the US is the only country with IP protection on drugs, and the protection is such that 100 drugs are profitably invented. Imagine another entity (call them the EU) decides to protect IP; they can either spend half what the US produces and incentivize the production of a further m<50 (diminishing returns of R&D) drugs, or spend the same as the US and incentivize n<100 further drugs. From the EU standpoint, they can get 100 drugs for free, 100<100+m<150 drugs for x amount of money, or 100<100+m<100+n<200 drugs for 2x amount of money. Sure, they miss out on some new drugs by not paying more, but international IP often allows a weak free rider problem.
The same logic applies to, for instance, China's decision not to enforce many IP laws; they're better off because of it!
Posted by: cure at Nov 15, 2006 9:36:42 AM
You do not need a 5th copy-cat version of Prozac.
Posted by: Oskar Shapley at Nov 15, 2006 9:41:58 AM
Sure, they miss out on some new drugs by not paying more, but international IP often allows a weak free rider problem.
I don't think that anyone disputes the free rider problem issue. It's absolutely there, and certainly in many cases it pays an individual country (especially a small one) to free ride on drugs. This merely demonstrates that price controls do have an effect on R&D, a contention often disputed. Also note that the EU has not "picked up the slack" with government-supported research, though again they may simply be free riding.
The free riding problem would only be magnified if all the European drug companies switch R&D production entirely to the USA. There would then be fewer political restraints on price controls there.
You do not need a 5th copy-cat version of Prozac.
Tell that to the person for whom all the other versions don't work, or produce horrible side-effects, such as a tendency to suicide. You do realize that some people have depression but that SSRIs cause really awful adverse side effects in them, right? So some research into better SSRIs is entirely appropriate. In any case, in a private system, you do not need to STOP a drug company from researching a copycat version of Prozac, either. It either pays for itself or it does not. It's not particularly hurting anything now since Prozac is off-patent and available generically.
Posted by: John Thacker at Nov 15, 2006 9:53:33 AM
Anyone know what the pharma companies spend on marketing and advertising in the EU versus the US?
Just curious.
Posted by: Kevin at Nov 15, 2006 9:59:30 AM
The only way to know if we need a 5th copy-cat version of Prozac is to see if people are willing to buy it.
Posted by: Brant at Nov 15, 2006 10:20:15 AM
John Thacker,
The most interesting point that Baker makes, I think is that the amount of R & D that we get for the monopoly-priced drugs.
"What consumers get for this $79 billion in higher drug prices is $22.5 billion in domestic research from the pharmaceutical industry."
That is a big number to overcome. The money allocated to marketing is beside the main point, but i would argue that you are at the very least overstating the benefit of these costs. Some marketing makes people aware of drugs, but there are also reports of patients demanding and getting advertised drugs that are not helpful to them. Regardless of that, the big question is patent-driven research money 3 times more efficient than some other mechanism?
Posted by: theCoach at Nov 15, 2006 10:27:35 AM
And the only reason Europe doesn't have its own Intel and Sun is because the GSA doesn't bargain when it buys servers? Isn't this just libertarians calling for industrial policy for a favored industry?
Posted by: bjak at Nov 15, 2006 10:29:56 AM
