Does the US Really Account for So Much Pharma Profit?

There's quite a lot of question about whether Obama can turn health insurance reform around when he comes back from vacation next week.  My guess is that he has to do it pretty quickly--within a few weeks--or any idea of really substantial reform is fairly hopeless.  But I certainly don't think it's impossible.  The man can talk.

But I gather that liberals are getting more and more worried.  Why?  Because the gratuitous nastiness from across the aisle.  Take this post from the American Prospect:

I'm not saying The Atlantic should fire her. But perhaps her bosses ought to sit her down and have a discussion about the rigor with which one should approach writing, even blog writing. For instance, there are claims of opinion ("Coffee ice cream is tasty"), which require no particular support or justification, and then there are claims of fact ("Pharmaceutical companies make most of their profits in the United States") which do require that one be accurate. McArdle doesn't seem to know the difference. Fortunately, today we have this thing called "the Internet," which on a whole range of topics allows one to quickly and easily verify whether the impression one has, or something one vaguely remembers hearing somewhere, is actually true. If you can't be bothered to look it up, you might try inserting some qualifiers - "I seem to remember that..." or "I believe that..." or "I'll have to check this, but I think that..." - before making emphatic empirical claims. That way, if it turns out that you're wrong, you can easily correct the record, without looking like an idiot or a jerk (or maybe both).

If they had that conversation with her, then maybe she'd be less likely to find herself saying things like, "It wasn't a statistic - it was a hypothetical."

The reference is to my off-the-cuff remark about slashing pharmaceutical profits by 80%.  I should note, to be fair, that there were two portions of the comment:  one in which I repeated an estimate I had heard from several people, that the US accounted for something in the range of 85% of pharma net profits after you accounted for various issues, which I then turned into 80-90% when typing--a fairly common way to give a range on an uncertain verbal statistic.  And then I said, "So if you slashed pharma profits 80% . . . "  When asked about it on the Washington Post live chat, I forgot the first, and thought the commenter was referring to the postulated hypothetical destruction of all US profits.  It's not clear which part of the comment they are referring to. 

But this "error" that I didn't check was not, contra Waldman, in a blog post, but in a comment, followed by a live chat on the Washington Post's site.  Waldman doesn't seem to know that, which implies that he didn't look.  I mean, I'm not saying that Waldmann should be fired.  But maybe his bosses should sit down and have a talk with him about primary sources.

I may be in error on that--I've heard 80-90% from people in healthcare consulting, and I've seen that sales and profits in the US are usually larger when they're broken out on financial statements, which they aren't all that often.  But they were not speaking on the record, and  financial statements are not necessarily a very good guide to allocating the net profitability of a drug, because of various tedious pricing strategies involving market timing that you can read about in an exhaustive volume from the OECD that I have on my desk, if you want to come to my office, or spend $100 to buy it yourself.  There are also issues of the way that companies allocate profits across international borders, which vary for all sorts of reasons, including the location of the company.

For example, GlaxoSmithKline, which Waldman mentions as a counterargument to my estimate, just had a catastrophic collapse in its US revenues due to the expiration of important patents like Wellbutrin XL, Paxil CR, Imitrex, and Lamictal.  Waldman would have known this had he, like, Googled it.

But there are other ways that we can back into the question of which markets are the most important.  Sadly, there are no definitive numbers on the topic that I am aware of.  And I've asked a lot of researchers, left and right.

We can, first of all, look at financial statements longitudinally.  We can examine what happens to profits of pharmas when sales in the US decline, while sales in Europe (and overall revenues) rise.  The answer, for GlaxoSmithKline, was that profits fell 13%.  The US is punching massively out of its weight class on their balance sheet.

Next, we can look at where the revenues come from.  According to the OECD's invaluable Pharmaceutical Pricing Policies in a Global Market, which really is a bargain at the price, 45% of global sales come from the United States, 30% from Europe, and 9% from Japan, meaning that the US accounts for the lion's share of profit--sales in other countries are too low margin to be currently important markets, though they're undoubtedly nice gravy, and serve an investment function.  The OECD also says that the United States accounts for more than half the growth in pharmaceutical revenue.  And it singles out the United States as the "important exception" to the otherwise iron rule that no country's prices much affect the level of R&D spending.

We can look at per-capita pharma spending.  The United States spends about twice the OECD average, and as aforementioned, does more than half of the OECD spending on pharmaceuticals.

We can back into it from prices.  We know that prices are higher in the United States than anywhere else for on-patent drugs (our generics can actually be cheaper, which further ups the implied profit on first-line treatments).  How do we know this?  Because the American Prospect, among many other media outlets, spends a great deal of time complaining about the fact that we don't use Medicare's "purchasing power", etc. etc.  If the prices are higher here, so are the profits.

We can look at prices in Canada and Mexico.  They're higher than they should be, relative to the general price level in those countries and other national health systems.  This implies that companies are protecting a very profitable US market.

Or we could go to the academic literature.  Not the literature from advocacy groups which too often fills the pages of political magazines on the left and right, but something from someplace like Rand.  And fortuitously, Rand happens to have published a paper on this very topic!  Their analysis of the effect of a 20% price decrease--about what they estimate we could get:

Exhibit 1 illustrates the impact of introducing U.S. price controls on the longevity of cohorts ages 55-59, using our baseline parameter values. It shows that the introduction of price controls would reduce life expectancy by two-tenths of a year for Americans ages 55-59 alive in 2010 and by one-tenth for Europeans ages 55-59 alive in the same year. In percentage terms, these correspond to 0.8 percent and 0.7 percent declines from the status quo.

The longevity effects are larger for the older cohorts, because the effects of price controls take time to set in. The early cohorts are not exposed to innovation reductions for a number of years. This dampens the impact on their life expectancy. By 2060, Americans and Europeans in this age group lose almost 0.7 years of life expectancy as a result of U.S. price-control implementation.

On the benefit side, U.S. price controls reduce spending on dr ugs and medical care. Exhibit 2 quantifies this effect. Price controls adopted in 2005 would reduce lifetime per capita health spending by $9,000 in the United States and $400 in Europe, for those ages 55-59 alive in 2010. Reductions in Europe come about as a result of reductions in life expectancy. The U.S. effects combine life expectancy reductions with direct reductions in cost. For those ages 55-59 alive in 2060, Americans can expect $14,400 less in lifetime spending; Europeans, $2,100 less.

Exhibit 3 shows that U.S. price controls have very modest benefits in the present but substantial costs in the long r un. For the 2010 cohort, price controls pro-duce $1,100 of net per capita benefit in the United States but $8,000 of net per capita cost to Europeans in that cohort. By 2060, the cohorts ages 55-59 lose $51,000 and $54,000 in the United States and Europe, respectively.

If you're wondering how much levels of spending matter, you could go to Acemoglu and Linn, who estimate that a 1% increase in market size (aka revenue) for pharmaceuticals results in a 3-4% increase in the number of drugs being approved.

I don't want my off-the-cuff comment, based on conversations with people who were not speaking on the record, to become the source of a fake statistic for the right.  80% may not be right, and I can't back it up with any hard numbers, because there are no hard numbers available.  But multiple corresponding sources suggest that the number is well over 50%.  60% is probably the floor of likely.  

The thing is, the US is less than 25% of the population of the OECD.  And we're the ones driving the majority of the revenue growth.

The really irritating thing is that this is not remotely controversial to anyone who has ever spoken with anyone who is in the pharmaceutical business.  It is not controversial that the center of gravity of pharmaceutical research and biotech is slowly but steadily tipping towards United States, though the reasons for this are still debated.  It is not controversial that when the US starts looking like it might move towards price controls, R&D departments hunker down and wait.  It is not controversial that the United States is by far the most important market for pharmaceuticals as a whole.  This is only a surprising finding to people who do not want to sully themselves by getting any information from the pharmaceutical industry, who, after all, are interested parties.  Rather than risk contaminating themselves with the dreaded taint of business, they rely on reports written by "experts" who have never run, or even studied, businesses.

Perhaps I shouldn't use scare quotes, because of course, those experts often are real experts in important things.  But their work usually betrays the belief that running a business does not take any very interesting skill beyond a ruthless willingness to kill babies for fun and profit.

So we can quibble about whether the percentages are in the high double digits, or the very high double digits, but what does the exact percentage matter?  Even if we had a beautiful, firm number, it would change two years hence--probably, if history is any guide, in the direction of making the US a higher percentage of profits.  The upshot is that the overwhelming weight of the available evidence indicates that the effect of price controls in the US would be real, significant, and bad.

This is particularly rich since in his actual written columns, not his off-the-cuff remarks, Waldman displays a certain penchant for inflating already inflated statistics:

Fifty million uninsured, the highest per-capita costs in the world, millions of people pushed into bankruptcy by medical bills, worse health outcomes than most of the industrialized world? Are you kidding me?

We don't even have "millions" of people pushed into bankruptcy by all causes. He's inflated the number of uninsured by almost 10%, and that number is itself inflated by the inclusion of people who lack insurance for as little as a day.  Worse health outcomes is, to say the least, highly contested outside of Waldman's circle of advocates.  Etc.  

The left used to be the side that said "You're entitled to your own opinion, but not your own facts."  The idea that the United States market is by far the most important pharmaceutical market in the world, and that any significant change in the profit margins on drugs sold here will have enormous impact on the future of pharmaceuticals, is as close to a fact as we can get in this vale of uncertainty.  This is not the hill on which liberals should try to stake their claim to the moral high ground.