[Ip-health] HIF and economies of scale and scope

[James Love]

This note again focuses on the issue of economies of scale and scope,
and why the elimination of open licensing from a prize fund is harmful
to patients.

I'll start with some data.

If you look at the June 20, 2008 UNITAID  cost benefit report, available
here:
http://www.keionline.org/misc-docs/1/cost_benefit_UNITAID_patent_pool.pdf

On page 13, table 2, is a comparison of 2004 generic API prices for HIV
drugs, based upon two characteristics:

1.  Purchased as Generics in Brazil

2.  Purchased from Brand Name Manufacturers in Brazil

Here are the results:

Table 2: Difference in Raw Global API prices (2004) and Patent Status in
Brazil
        Price per kilo   low/high

1.  Purchased as Generics in Brazil

Average: 382 582

2.  Purchased from Brand Name Manufacturers in Brazil

Average: 1,540 2,760
------
Note that all prices are for generic products.

Here are a few other facts.

in 2004, generic API prices for Lopinavir (LPV) were higher than $4000
per kilo.  By September 2007, the fully formulated cost of LPV/r had
fallen to less than $1,400 per kilo.

A long run competitive generic price for Lopinavir APIs would be lower.
How much lower?  Certainly less than $1k.  Probably less than $500k.
Maybe lower yet.... if bought, in quantities, from generic suppliers,
with some time for learning by doing.  So under the HIF, what exactly is
the data used to set prices?  Would it have been the the entry level
price of $4k+ for APIs?  Or some guesses about the longer run
equilibrium price?   And what assumptions would be made about the
counter-factual of a price with larger generic demand?

Note that in the short run, sometimes the brand name company is more
efficient, because it also sells in the US and Europe, and has the
advantages of scale.  But often, over time, the generic companies beat
the brand name supplier -- as they have for AZT, 3TC, d4t, NVP, and many
other drugs that have the high level of utilization.

An important fact here is that generic competition has a dynamic impact
on prices, over time.  It may take a couple of years for prices to reach
lower levels, for lots of well known reasons.  In the beginning, maybe
there is only one API manufacture, and competition is not intense.
Maybe the generic manufacture has to recoup fixed start-up costs, or has
not figured out the most efficient production technique.  Maybe the
demand is too small to be efficient, due to registration barriers, IPR
barriers or other factors.

So these are points one and two.  Generic competition is highly
sensitive to economies of scale.   Generic competition also is dynamic,
and the best prices may be years away.

Third, there are major economies of scope.  A firm that can make AZT and
3TC APIs can probably make d4T, TDF or FTC.  A firm that can make LPV
can probably make atazanavir, if it expects buyers.

Fourth, if patent owners have a monopoly on new products, they will
outsouce API manufacturing to firms, ON THE CONDITION THEY Don't supply
APIS to generic competitors.  This happens a lot now.

Why do you think Ranbaxy and Aspen were so coy during the compulsory
licensing battles?  They had complex relationships with big pharma
companies, as suppliers of APIs and finished products.  They didn't want
to risk losing those contracts.  Fortunately CIPLA stepped up, and
fortunately TAC took on GSK and BI in the Hazel Tau case.

As I mentioned in my last post, the more you buy from brand name
companies and the less you buy from generic suppliers, the less
possibility there is to get products from the generic suppliers.  They
get co-oped, or they they don't have enough market to work with.

Economies of scale was 100 percent of the issue in the Doha paragraph 6
debate.  The smaller the generic competition, the less generic
competition there will be.

The decision to reject open licensing is one that would have a an
important (not a zero) impact on the future of generic competition in
developing countries.  It will strengthen the monopoly supply chain.
That's why big pharma cares about this.  That's why we care about this.
That's why donors should care about this.  If you want to provide
sustainable access to medicines in developing countries , I don't see
cannot be done without a strong un-cooped generic sector.  I don't see
that the HIF is helpful or strategic in that respect.

I hope you can respond to this as a substantive criticism of the HIF.

As you know and you might acknowledge, every single comment that you or
Aidan have sent to ip-health has been published on ip-health, and your
contributions are welcome.  You are opposing a core element of the
access to medicines NGO campaign and lately WHO policy, which is to
protect the space for generic competition.  As I have been telling Aidan
for a year, we consider this a blunder by the HIF, and we don't like it.
If I thought it was easy to get reasonably price generic products for
new drugs coming out under the new patent regimes, I would not care.  It
won't be easy, and the HIF policy on licensing would make things worse,
not better.

  Jamie

--
James Love, Director, Knowledge Ecology International
http://www.keionline.org | mailto:james.love at keionline.org
Wk: +1.202.332.2671 | US Mobile +1.202.361.3040 | Geneva Mobile +41.76.413.6584