Wednesday, November 11, 2015

The Me-Too Drugs Fallacy

At that event on drug pricing that I attended recently, someone asked a question about me-too drugs: he referred to drug companies looking to see what their competitors were making a killing off and bringing out their own version of it a couple of years later.  Someone else referred to one company bringing our a pink version of the first company's yellow pill.

The really sad thing about the whole episode was not the nonsense that was being spouted, it was that the nonsense was being spouted by someone who held a position as an economist in the provincial ministry of health.  If you've got that degree of lack of understanding within the ministry, it's no wonder you don't get get good policy being made.

The simple fact is that the only true me-too drugs are generics.  If the original drug has gone off patent, any generic manufacturer can enter the market with a drug containing the same active ingredient as the original, so long as they can prove to the  drug regulatory agency that their copy is bio-equivalent to the original drug.  Bio-equivalent means, basically, that the the generic copy releases the active ingredient into the body at the same rate as does the original drug, so that there is no difference, so far as effect on the patient is concerned, between taking the original drug and taking the generic. (Actually there can be, since the generic might not hold the active ingredient together using the same carrying substance as does the original, and some patients who did just fine on the original drug might react badly to the stuff which is holding the generic drug together, but fortunately that doesn't seem to be too widespread a problem.)

Bio-equivalence isn't quite as easy to achieve as it might sound - not long ago the FDA pulled a couple of slow-release generics off the US market because the active ingredient wasn't, in fact, being released into the body at the same rate as in the original drug.  Still, most generic copies achieve it.

That's what happens when the original drug goes off patent.  If the so-called me-too drug comes onto the market while the original drug is still on patent - e.g. when Company B brings its pink pill out two years after Company A's yellow pill hit the market, while Company A's pill is still under patent - then the second pill must be sufficiently different from the first in terms of the way it works that it does not violate the first pill's patent.  That means that it has to be different enough from the first drug to get its own patent, and to ensure that it doesn't work exactly the same way as the first drug does. And that means that it must have gone through exactly the same process of  clinical trials as the first drug did.  If it took the first drug, say, seven years to go through all three phases of clinical trials and to get clearance to go to market, it will have taken the second drug seven years to go through its own trials, unless it's an absolute miracle drug whose Phase III trial results were so strong that the process was stopped early, in which case there's no way it could be a simple copy of the first drug.  If the pink pill comes out two years after the yellow one, it will be because it entered clinical trials two years after the first yellow one did, while the yellow one was still going through its own series of clinical trials.  And again, it must be sufficiently different from the first drug that it does not violate the first drug's patent.

At least that's the case if you have product patents, as we, and most developed countries do.  The story's different if you have what are sometimes known as process patents.  Oversimplifying, product patents protect the final product itself while process patents protect the process by which that final product is made (I'm not going to get into the current state of TRIPS on this).  So under a simple process-type patent, what you need to be able to do is find a different way of producing the same drug, and you're golden.  Or pink.

As it happens, we know how that story plays out.  Back in the 1960s, when India decided that it wanted to stimulate a domestic pharmaceutical industry, one of the things it did was remove product patents and just keep process patents.  The strategy worked, up to a point.  India developed a very economically significant pharmaceutical industry.  The problem was that it was entirely a generics industry.  Because you couldn't obtain patent protection on the fruits of your research, nobody did any significant research.  Why spend money developing your yellow pill when someone else could come in with an identical pill, just assembled differently, a year or two later.  So the Indian industry focussed on making generic copies, and not just generic copies but generic copies of drugs for rich country diseases, because that was where the money was.  Big Pharma is often criticized for neglecting diseases endemic to poor countries, but here you had India, where diseases of poverty were not just on the doorstep but actually inside the door, and whose stock of human capital was quite remarkable (just think of all of the notable scientists who have Indian names) and its drug industry was ignoring the problem in its own back yard.

And it was ignoring it because in those pre-TRIPS days, India was a place where you could find me-too drugs in the sense the term was used by that guy from the provincial ministry of health.  But when you've got a system of effective product patents, two on-patent drugs aimed at treating the same condition must have clearly different mechanisms of action.  And in an age of personalized medicine, when different people who have the same disease will respond to different pills because of differences in the patients' genetic make-up, having multiple different on-patent drugs aimed at treating the same condition is probably a really good thing.

Monday, November 2, 2015

Coase and Pharma

A question came up at an event I was at the other day about the structure of the pharmaceutical industry - in particular the issue of companies buying new drugs from other, smaller companies, with the purchasing companies bringing the drugs to market.  The argument was made, and I've heard it before, that the bigger company hadn't actually done any of the research and development underlying the drug  (although it's not unusual for this kind of buyout to happen while the drug in question is still in clinical trials), so the company that brings the drug out can't make the usual claim that it needs the monopoly revenues from the remaining period of patent protection  to recover its research costs.

There's been an ongoing debate in the pharma Industrial Organization field about what form of organization would be most productive of new drugs.  One argument is that the best structure is the large, integrated corporation because it's got more fire-power to devote to research and because it has a portfolio of drugs in development, so that the failure of some can be compensated for by the success of others.  The opposing argument is that big pharma are a bunch of slow, ponderous dinosaurs and that it's the small, agile, very focussed research outfits which are most likely to come up with breakthrough drugs, but that because those outfits are small they don't have the resources necessary to take new drugs all the way through clinical trials, so they should sell promising new drugs to companies which can pay for large scale clinical trials and use the proceeds of the sale to start work on developing still more new drugs.

From the economist's perspective, this is an example of Ronald Coase's (1937) question about the nature of the firm (pdf here).  As we usually express it, Coase was asking why large, vertically integrated firms exist at all.  After all, economic theory argues that the price system is the most efficient mechanism for allocating resources (see Adam Smith's description, in Wealth of Nations, Book I, Chapter 1, pp. 4,  of the market-linked stages in the production of a labourer's woollen coat), so why do we have what are essentially command-and-control structures with resource allocation decisions handed down from the top, rather than have firms at the final stage of the production chain buy all of the components from competitive suppliers?

Back in the latter part of the 19th century and the early years of the 20th century, the American auto industry  didn't consist of a small handful of very large, vertically integrated corporations.  In those days if you bought a Buick, while Buick would have assembled the bits and put its name on the car, it would have bought most of the components which went into the final product from other small, specialized manufacturers.  It was Henry Ford who invented the modern automotive firm, bringing all of the stages of production under one roof and under one controlling force - his.  One suggestion as to why he did this was that it was inspired by a strike at one of his parts suppliers, which put a serious hitch in his production plans.  If he bought all of his parts suppliers and brought them all under his own roof, it would take a firm-wide strike to stop his production - no single part of the chain could go out on strike on its own.  So he created a new form of automotive producer, and the rest of the industry followed him.

Presumably it followed him because he had it right - vertical integration and command and control had advantages over the market-coordinated system which had preceded it. The explanation usually given for what those advantages are tends to involve transaction costs - there must be extra costs associated with market coordination that can be eliminated through imposing at least some degree of command and control.

Most industries seem to have more or less settled down into what are presumably transaction-cost determined structural equilibria.  Pharma, though, seems to be an exception.  In Pharma the wheel seems to turn without ceasing.

To come back to the starting point for this post, if there are no transaction-cost advantages to vertical integration, it wouldn't matter which organizational structure the industry adopted.  If the market could co-ordinate everything, there would be no difference in costs, profitability or the ultimate price of the product between the case where everything was done within big Pharma and the case where everything was done by small, specialized firms which sold their products and services (services in the case of CROs -  Contract Research Organizations - which specialize in organizing and running clinical trials of drugs under contract to whatever firm owns the drugs) which means that the price of drugs would be unaffected and the research would ultimately be paid for by the marketing company which bought the final product with the aim of putting it on the market.  The money would simply pass from firm to firm along a chain of markets instead of passing from division to division within a single firm.

One thing which might create a transactions cost equilibrium is the cost of financing the research in the first place.  The research will eventually be paid for out of market revenues, but especially in the case of a small, new firm, the research will have to be funded from other sources before the revenue from marketing the drug comes in (assuming the drug doesn't wash out before getting to market, as about 83% of drugs which enter the clinical trials process these days do).  In pharma, as F. M. Scherer  has shown, most research is funded out of retained earnings.

In fact, Gerben Bakker has suggested most research intensive industries fund their R&D out of cash, rather than, for example, by borrowing.  One possible explanation is informational asymmetry: the borrower is likely to have a better idea than the lender which R&D projects are likely to be more successful and which are riskier.  Since the borrower has an incentive to downplay the riskiness of any individual R&D project for which it is trying to obtain funding, lenders will tend to add an extra risk premium element to the interest rate charged for any loans raised for R&D purposes.  As a result, it might be cheaper for the firm to finance its R&D out of retained cash.

If that's the case then over the long run the integrated firm is likely to dominate in pharma, at least. On the other hand presumably the transactions cost advantage to being large and integrated can't be overwhelming, or the Industrial Organization wheel wouldn't keep on turning the way it does.

Sunday, October 25, 2015

Keynes on Saving and Investment

There's an interesting discussion going on over at David Glasner's excellent Uneasy Money blog, dealing with what Keynes had to say in the General Theory about saving and investment and how that fed into his theory of the rate of interest.  This is an area which we tend to take very much for granted, and it's one of several points of theory with regards to which you could argue that we have discarded the economics of the General Theory and simply slapped the label "Keynesian" onto what is really AC Pigou's macro model.

Keynes is quite clear about S and I in the General Theory: each is the gap between income and consumption, and since each equals Y-C, they must be equal.  They are, however, behaviourally different: Investment is a choice variable in an intertemporal optimization problem, but saving is not - it is consumption which is the choice variable in the household's intertemporal utility maximization problem.  Keynes had not forgotten Frank Ramsey's work on saving, he did see the individual household as solving an intertemporal optimization problem - see the chapters in the General Theory on the propensity to consume - , it was just that he took the view that, especially in the short run, income was the predominant determinant of saving.  Further, whatever decisions the individual household might be making, in the aggregate saving had to be equal to investment by definition, and the equality was not driven by saving.

The important thing, to Keynes, was what this meant for the determination of interest rate.  Since S was always identically equal to I by definition, the interest rate could not be a price which equated S and I, so it must be determined somewhere else in the economy.  This tied into his criticism of the way standard texts of his time treated interest rate policy - if the rate of interest, r, was determined by the S(r) and I(r) schedules, how was it that the central bank could influence r?  What did monetary policy affect, S or I, and note that in the classical model this effect would have to involve a shift of whichever of the curves it was operating on, not a movement along it (or them).

Keynes' answer to this was that the rate of interest was determined by portfolio choices; not  the level of saving but the way individuals allocated their savings across financial assets.  The default choice of holding savings was in the form of money (in the General Theory, money pretty broadly defined) and here Keynes agreed with Irving Fisher that the fact that money is, by definition, something which confers complete liquidity on its holder means that there is a liquidity return for holding cash (including certain types of bank money) and this liquidity return creates a certain preference for liquidity.  The rate of interest, then, depends on how strong this preference for liquidity is and what kind of return it takes to get people to put part of their savings (the level of which in the aggregate is determined by income) into non-monetary, interest bearing assets.

Keynes knew, by the way, that there was more than one interest rate - in the General Theory he simply assumed for simplicity that the spectrum of rates across duration and risk of assets was stable, so that a change in the short term rate would serve as an index of the shift in the whole spectrum of rates.  He also assumed that what mattered for economic purposes was the long term rate of interest - that was what drove Investment spending - but he held that the bank rate could influence the long term rate in a stable manner.  Hawtrey argued that it was the short term rate itself which mattered and further that, if you looked at data over a Century of Bank Rate, you'd find that the relation between long and short rates wasn't so stable after all.

When we consider what Keynes had to say about S=I it's important to remember that he was trying to counter three theoretical arguments. There was the classical one which we're familiar with, to the effect that S equalled I because the interest rate made it so.  But in addition to that, there was Hawtrey's argument that  I equalled S plus any new bank credit which the banks created to finance investment that period.  And, ironically, there was Keynes' own model from the Treatise on Money, according to which (using what he later admitted was a non-intuitive definition of income) investment could differ from saving.  In the Treatise, the central bank could influence the market rate of interest (exactly how was rather taken for granted) and Keynes defined a natural rate of interest as the rate which would make S = I (using to the definitions of the Treatise).  So he had to convince people that his own earlier writings had been wrong.  Thus what he needed to do if he wanted to convince people of what was really important, i.e. his monetary theory of the rate of interest, was to convince them that S had to equal I by definition.

There's much more in the Uneasy Money post, in particular about Keynes' comments on Irving Fisher's concept of the real rate of interest, but commenting on that'll wait for another post.

Valeant and Health Economics (1)

The guys at Valeant must really hate the guys at Turing.  It was Turing's decision to jack up the price of Daraprim which brought the whole pricing strategy thing out into the open, prompting blog posts like this one from Derek Lowe, and news articles like this one from the Financial Post and this one from the Globe and Mail, along with claims from a short seller that Valeant had a somewhat unhealthy (for the payer) relation with a specialty pharmacy.  Valeant is apparently denying that there's anything improper in their business relations with Philidor, and they may well be able to make their case, although the whole thing's going to leave a bad odour for a long time.

This post is not about the legalities or the accounting rules involved - plenty of other people are posting about those.  I'm interested in how I could use the Valeant example to teach health economics (and my view on that has changed over recent days - I see a wider gap between Valeant and Turing than others do).

It comes down to two well known pieces of economic analysis of the US (and to lesser but increasing extent Canadian) market for pharmaceuticals.

The one I'll deal with in this post is the well known fact that most people don't pay the full cost of their prescriptions.  Their insurance pays most of it (unless they're uninsured, which is a matter for yet a different post).  The cost of the prescription to the insurer depends on the price set by the pharmaceutical company.  When a drug is under patent, with no on-patent competitors (and if you don't think that on-patent competition matters, see this paper by Lichtenberg and Philipson) the insurer basically just pays the price set by the supplier.  When the drug goes off patent, generic competitors can come on the market.

We used to speak as if a generic was automatically a lot cheaper than an on-patent drug: Turing has punctured that illusion.  the price set by a generic company will depend on the degree of competition it faces.  No competition means monopoly, which means monopoly price.  But for the moment let's assume that the old assumption holds and that generics are cheaper than their brand-name counterparts.  Insurers want to shift as many of their insured across to the generic as possible, for obvious reasons.

Sometimes they do it through mandatory generic substitution.  When one of my meds went off patent, my pharmacist shifted me across to the generic with my next refill (I was relieved, by the way, to see that he gets his generics from Teva and not from one of those generic manufacturers whose quality control has been shown to be, shall we say, less than sterling recently - seems brand name matters even in the case of generic drugs).  Alternatively, if the insurance plan operates using co-payment (a flat patient charge per refill, as distinct from co-insurance under which the patient pays a percentage of the price of the drug) the insurer will put the brand name drug in a high co-pay tier and the generics in a low co-pay tier, and count on the patient's elasticity of demand with regards to out-of-pocket payment to induce him to shift from the brand name to the generic drug.

What's been happening recently (and Valeant is by no means the only firm doing this, so too are a number of brand name companies - take a look at what happened when Lipitor went off patent) is that manufacturers have been offering to cover the co-pay for any patient who has his or her doctor specify no substitution - i.e. has their doctor insist that the patient get the brand name drug.  The patient pays next to nothing out of pocket, the brand name company makes the sale and the insurer has to pay the price which the brand name company set (unless the insurer chooses to refuse to cover the brand name drug at all, which some insurers decided to do in the case of Lipitor).

So it looks as if part of what Valeant might be doing is using a specialty pharmacy to cut the price to the person in the prescription chain who has a big say in the buying decision and who is likely to be pretty sensitive to price - i.e. the patient.   If so, they're doing it differently, if they are indeed using a specialty pharmacy to do it rather than sending out co-pay credit cards, but it may well be exactly the same pricing strategy as a number of brand name drug companies have engaged in in recent times. Basically, they are making it very convenient for patients to order their prescriptions through one distributor, who can pay the co-pay on behalf of the patient, and so keep those patients loyal to Valeant's products.

As I say, the legalities and accounting rules re Valeant's relation with its specialty pharmacy are way outside my competence, but from the perspective of teaching health economics, it's not impossible that Valeant is actually engaging in a variant of a now pretty well established pricing policy.

There's more health economics theory to be invoked here, but that'll wait for another post.

The Pharmaceutical Sector is a Godsend if you Want to Teach Old Fashioned IO (1)

The recent kerfuffle about Turing Pharmaceuticals having jacked up (and then maybe jacked down) the price of the anti-parasitic drug Daraprim should provide an opportunity to do some serious old-fashioned industrial organization  (i.e. pre-game theory) teaching, if only because of the way it highlights just how complex the IO of the pharmaceutical sector is.

First, you've got the division between on-patent (or brand name) drugs and off-patent generics.  The usual notion is that generics should be cheaper than brand name drugs because the generic companies don't have to put their products through clinical trials, they just have to demonstrate bioequivalence with a brand name drug whose patent has expired.  The Daraprim incident shows that the only thing which will keep prices down is competition.  Alex Tabarrok had a post on Marginal Revolution a little while ago dealing with the competition issue in this case - the FDA's regulations set the bar for entry into the US market by foreign (or for that matter any new) producers so high that once a number of existing producers drop out, it's not uncommon for the remaining producers to find themselves in a monopoly/duopoly position.  Daraprim isn't the only recent example, it's just the most dramatic.

The basic IO story here's pretty simple.  The actual physical manufacture of drugs is pretty close to being a constant marginal cost industry.  For brand name drugs the research costs are fixed costs; generics don't have those.  For a while, the generic drug sector was very profitable, but nobody noticed because of the halo effect of the notion of generics.  New producers entered and the price was continually driven down, at least in the US.  In Canada, provincial drug plans regulated generic prices by tying them to the prices of brand name drugs and, as so often happens, what was supposed to be a price ceiling turned into a price floor and we would up paying considerably more than the Americans did for generics.  More on that in a later post.

Anyway, in the US the profitability of the generics sector attracted entry, and prices and profits were competed down, causing some producers to leave the market.  Getting back in's a lot harder than getting out (the reverse lobster trap effect) so the ones which stayed found themselves with market power, which they started to take advantage of by raising their prices significantly, but not catastrophically from the payer's point of view.  (Is that like raising the temperature in a pot of water containing a frog?)

Bottom line on generics: standard economic theory works.  Price regulation will hold prices down for a little while but only competition will keep them down long run.  For a while it looked as if Indian generic manufacturers were going to fill the competitive niche, but quality control problems knocked some of the biggest of them out of the American market.  If the Trans Pacific Trade Partnership negotiators want to do something useful, they should agree on a new reciprocal production quality regime and encourage free trade in generics among TPP members.  Canada's got an exporting generic industry - bets that that could be significantly expanded under the right trade rules?

Pharmaceutical IO (2)

So what about the IO of the brand name sector?  There's a New York Times piece on it here which suggests that you could use it to teach Coase's question on the nature of the firm.  To paraphrase, since economic theory teaches that production and exchange can be coordinated through the price mechanism, why do firms, especially big, vertically-integrated firms exist at all?

The usual answer is transactions costs.  There is something intrinsic to a sector which prevents the price mechanism from working its magic (I'd say something about preventing the Invisible Hand from working, but I don't want to wind up on one of Gavin Kennedy's Looney Tunes lists) and which requires direct, command-and-control assignment of resources to tasks.

So the structure of a particular sector, whether it be made up of a few large vertically integrated firms (as the auto industry is now) or of a lot of smaller firms which sell parts to each other and to final assemblers (as the auto industry was before Henry Ford created the modern auto firm) is going to vary across sectors according to the nature of the transactions cost in each sector.

But the research-based pharmaceutical sector has both extremes.  You've got the traditional Big Pharma integrated firms which do everything from basic research through production and marketing, and you've got small specialist companies, some of them doing basic research, others buying the products of the first firms and bringing them to market, and you've got CROs, Contract Research Organizations, which specialize in running clinical trials for firms which are hoping to bring drugs to market.  And the structure isn't static - it's a wheel which is always turning; sometimes being small and nimble is the way of the future ans sometimes being large and having a portfolio of drugs and drug candidates which can see you through bad times is the only way to survive (according to that year's favoured theory).

So the research based pharmaceutical sector is one in which Coase's firm structures are in a constant state of churn.  Which has got to mean something really cool as far as the industry's Coasean transactions costs are concerned.

Friday, October 2, 2015

Prizes as Incentives

There's a strand of the literature on the economics of intellectual property that argues, harking back to the Longitude Prize, which holds that prizes are better than patents as devices for stimulating invention and innovation.  Some argue, for example, that they have a lot of promise in drug development.  I am sceptical on that front - among other things I'd argue that the prize proposals underestimate the true cost of the development of a successful drug, since that cost includes the cost of all the blind alleys which research-based drug companies go down.  I'm also inclined to suspect that supporters of prizes in pharmaceutical development haven't considered how that approach would affect the development of competitors for the winning drug, or, more likely, that they don't believe that having a number of drugs targeting the same problem is a good thing.

There are, however, areas where I'd argue that prizes make a lot of sense, and that this is one of those areas.  I'd argue that it would have made a lot more sense had, say, the government of Ontario offered a prize for the development of a climate-friendly energy technology rather than pushing existing technologies with assorted subsidies.  Just because a technology exists doesn't mean that it's the best long run solution, but the more governments support existing technologies, the harder it is for new ones to gain a foothold, even if they're superior.  So if you think we need to reduce carbon emissions, impose a carbon tax and create a green-energy prize.