Transcript Pricing Issues in Improving Access to Essential

Pricing Issues in Improving
Access to Essential Medicines
Keith E. Maskus
UC-Berkeley Law School
Course on Neglected Diseases
February 17, 2009
Scope of Problem
• Halting progress in treating 3 major diseases:
– HIV/AIDS declining?
• Estimated global deaths 2.7 million in 1999; 2.0 million in 2005
• Falling mortality, somewhat reduced morbidity
• Reasons: better prevention, more access to ARV treatments developed for rich
countries.
• Still, less than 10% of patients in poor countries have ARV access.
– Malaria growing?
• Estimated global deaths 1.1 million in 1999; 1.9 million in 2005
• Rising mortality and morbidity
• Continued race between disease resistance and drug development
– Tuberculosis stabilized?
• Estimated global deaths 1.7 million in 1999; 1.6 million in 2005
• Better prevention and treatment
• Health problems likely reduce GDP growth by 1-2 percentage points
per year (could worsen with current economic crisis).
Rising Global R&D Investments in
Neglected Diseases
• Total R&D $2.7 billion in 2007
• HIV/AIDS (Type II disease) $1.083 billion (not much
aimed at specific DC conditions)
• Malaria $468 million ($84 million in 1993, US military)
• Drug development 46%
• Basic R&D 24%
• Vaccines 19%
• TB $410 million
• Drug development 35%
• Basic R&D 32%
• Vaccines 20%
• Others $739 million
Sources of 2007 R&D
• OECD+ governments and multilateral
organizations: 69% (of which USG = 70%)
• Philanthropic and not-for-profit groups: 21% (of
which Gates Foundation = 84%)
• Private companies: 9% of which
– Research-intensive MNEs = 80%
– Small companies, biotech = 20%
Data source: Moran, et al, Neglected Disease R&D: How
Much Are We Really Spending? PLoS Medicine Policy
Report, Feb 3 2009
General Market and Policy Failures
in Inducing R&D
• Small market demand fails to cover fixed investment costs.
• Standard appropriability problem for innovators:
– High R&D costs;
– Low marginal costs and ease of copying or inventing
around.
• Consumption externality in vaccines: spillover gains reduce
individual willingness to pay.
• “Time consistency” problem that governments want low price
ex-post. Can achieve this through price controls, purchase
agreements, compulsory license threats, etc.
• Lack of exclusive rights reduces incentives to launch
medicines in specific markets.
Access to Medicines: Problems in Poor
Countries
• Low household incomes in relation to other needs;
• Very low investments in public health and low shares devoted to
medicines;
• Inadequate health care infrastructure and delivery systems;
• No public and/or private insurance mechanisms for pooling risk
(except for richest groups);
• Tariffs, taxes and monopolized distribution in imported or licensed
drugs;
• Weak generic competition from domestic or regional producers
(even without patents);
• Patent protection an insufficient incentive for investments in new
drugs and other forms of treatment;
• Results: limited availability and affordability of medicines.
Need for Global Action
• Patent reforms (WTO/TRIPS) are unlikely to offset
income-based demand deficiencies for investment
incentives.
• R&D externalities (collective action problem): policy by
any one country to encourage R&D (raise price, adopt
patents, subsidize R&D) would shift benefits to other
countries.
• Pricing externalities: policies to reduce prices close to
marginal cost in poor countries may inefficiently affect
prices in richer countries.
• Importance of divorcing global research incentives
from pricing/access policies in poor countries.
Global Pricing Issues: Definitions
• Differential pricing: same drug is priced at
different levels across countries for any reason
(eg., price controls).
• Discriminatory pricing: prices differ due to profitmaximizing strategies.
• Equity pricing: price in each country is related to
income or other measure of economic capacity
and patient need or disease burden. Set by
government or firm policy to increase
affordability for poor patients. (Also called tiered,
preferential or access pricing.)
Price Discrimination
• The practice of charging different prices to consumers with
different marginal values for a good.
• In diagram below, region M are profits for monopoly if just
one price can be charged. If (expected) M exceeds R&D costs,
investment is made. Net consumer gains are area A.
Consumers in region B go unserved.
• First-degree PD (“perfect” PD): charge each consumer her
marginal valuation. Monopoly extracts all net consumer value
as profits.
– All consumers are served.
– More R&D is induced ex ante.
– Not sustainable if consumers are in integrated market.
p
Monopoly vs. perfect price discrimination
A
pm
M
B
c
mr
prescriptions
Price Discrimination
• Second-degree PD: different prices for
different quantities sold.
• Third-degree PD: different prices for groups
segmented by location (or other feature).
– Requires market power protected by exclusive
rights, brand power, etc.
– Demand elasticities vary by market group.
– Markets are segmented and insulated from
arbitrage by trade costs, restraints on parallel
trade, etc.
Determinants of Price Elasticity
• Per-capita income (presumption: demand becomes less priceresponsive as incomes rise);
• Existence of product substitutes;
• Marketing and prescribing practices;
• Concentrated buyer(s) or individual purchasers;
• Income distribution;
• Insurance markets;
• Taxes, trade barriers, other factors.
• Basic markup equation: pi = c/(1-1/ei) for market i. p = 2c for e
= 2; p = 1.25c for e = 5; p = c for e very large. The lower the
elasticity the higher the price.
Third-degree PD
• Permit patent-holder to sell in 2 segmented countries.
• In diagram below, price is higher in less-elastic demand
market H and lower in more-elastic demand market L.
• If firm is forced to sell at uniform price it may choose not to
sell in market L in order to protect profits in H.
• Is PD superior in welfare terms to single price? Yes if the lower
price serves more markets without reducing supply in highpriced markets.
• Note that PD generates net consumer gains in L of area A and
contributes profits of area M, raising R&D incentives.
• It is possible that in serving market L the firm would restrict
supply in H, causing higher price there.
p
Prices with different demand elasticities
pH
A
pL
M
c
prescriptions
Do Prices Rise with Income?
• In general, yes but correlation is weak (around
0.35 for sample of 15 drugs in 1998).
• Next 2 charts show that identical drug baskets
can have higher wholesale prices in lowerincome economies.
Average Drug Prices and Incomes
(1998)
Source: Maskus, WIPO Report, 2001
Relative Wholesale Prices for 15 ARV drugs, 1995-99
Source: Scherer and Watal, JIEL 2002
Why the limited price variations?
• Price controls and single-buyer price negotiations may be
more rigorous in OECD countries.
• Generic competition is often stronger in OECD (but note low
prices in India).
• Firms sell at high markups to high-income consumers with
private insurance in DCs.
• Firms set or negotiate higher prices in DCs to avoid:
– Smuggling;
– Parallel trade back to high-price markets;
– Reference-pricing pressures on price controls;
– Political pressures to reduce prices in home countries (for
Type I and II disease drugs)
Comments on Reference Pricing (RP)
• Forms of pharmaceutical price regulations:
– Cost-based (estimate production costs, perhaps some allowance for
R&D and fixed-cost recover, and a negotiated profit margin);
– Price negotiations based on costs and “therapeutic value” or “patient
needs”;
– RP: Authorities set price based on prices of similar products on home
market and/or similar and identical products on foreign markets.
• Examples of RP:
– Portugal, Ireland and Italy use baskets of lowest prices in EU and
several DCs.
– Spain and France use such baskets as benchmark for negotiations.
• Global “web” of RP is extensive (diagram) but most rich countries do not
include prices of poor countries.
• The bigger “information spillover” is political pressure on home markets to
match prices in foreign markets.
Source: Dr. E. Schoonveld, Report on the Workshop for Differential Pricing and Financing of
Essential Drugs, WTO & WHO
Comments on Parallel Imports (PI)
• PI is unauthorized cross-border trade in legitimate goods
protected in import market by some IPR.
• Legality depends on scope of rights exhaustion (varies by
country or region and by type of IPR).
• PI takes place overwhelmingly at wholesale or distributor
level, not consumer arbitrage.
• Main benefit of PI should be to reduce prices in importer
nations.
– One detailed study: Ganslandt-Maskus, JHE 2004. Relative wholesale
prices of medicines in Sweden subject to PI falls as much as 12-18%.
Significant entry of PI firms in blockbuster drugs. Not much impact on
retail prices as hospitals and pharmacies absorbed higher margins.
Comments on Parallel Imports (PI)
• Main static costs of PI:
– Transport costs and tariffs in act of trading identical goods.
– Higher prices in exporting nations if significant volumes leave. GM
study found little evidence of this (Sweden is small compared to EU).
– Strategic market segmentation behavior.
– Strategic pricing: IPR-holding firms may set/negotiate higher prices in
export markets to limit arbitrage.
– Risk of “importing price controls”.
• Main dynamic costs of PI:
– Delayed launches of new products in lower-priced countries (evidence
supports this claim).
– Diminished profits of originator firms may reduce R&D incentives
(probably, though R&D indicators in EU are approaching those in US).
RP and PI Policy Issues for Equity Pricing
• The objectives:
– A sustainable global policy of greatly tiered equity pricing;
– Encouraging pharma and vaccine companies to be willing to launch
new products in poor countries.
• This outcome presumably requires:
– An agreement by OECD and middle-income countries not to set price
controls with reference to poor countries.
– A political agreement not to try to import prices from poor countries.
– Sufficient exclusive rights in poor nations to encourage product entry
(imports, licensing) at a price covering marginal cost.
– Permission of PI into poor countries.
– A ban on re-exportation of goods from poor countries.
• May be useful to permit regional marketing with PI to encourage scale
economies with price discipline.
Direct Equity Pricing Policies in Purchasing
Countries
• Price regulation with exclusive rights (patents, EMRs);
• Pooled procurement policies and bulk purchases;
• Enhanced threats of generic entry through:
– Limitations on patent scope (see India 2005 patent law);
– Short periods of protection for confidential test data;
– Compulsory licenses (CL) issued to domestic competitor(s).
• Note importance of generic competition (graph).
The Importance of Generics
Figure 1: Originator and Generic Drug Prices for a Sample ARV Triple-Combination
$12,000
Originator $10,439
Originator price
$10,000
Generic price
$8,000
$6,000
$4,000
Generic $2,767
Originator $931
$2,000
Generic $295
Originator $727
Generic $209
Generic $350
2000
2001
2002
Notes: Sample of ARV triple-combination: stavudine (d4T) + lamivudine (3TC) + nevirapine (NVP). Low est w orld prices per patient per year.
Source: Médicines sans Frontières, December 2002, "Untangling the Web of Price Reductions," available at w w w .accessmed-msf.org.
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
$0
Impact of Global Trade Rules
• Main framework: TRIPS Agreement at WTO
• Parallel trade: Article 6 is interpreted to mean each country
can set its own policy.
• Legal challenge: South African Medicines Act 1997, US
eventually dropped lawsuit.
• Test data: must be protected but terms and duration are left
unspecified.
• Compulsory licenses:
– extremely complex set of conditions under Article 31;
– Or declare public health emergency;
– Issue government-use license.
Problems with CL Regime
• Article 31.f permits CL only where production is
“predominantly for supply of domestic market” to prevent
industrial policy leading to exports.
• Of no value for countries without domestic production
capacity who need to import.
• The fix: 2003 WTO Waiver (made permanent Dec. 1 2007):
– Firms can export under CL issued by their governments to
meet needs of importers;
– All countries without production capacity are eligible
importers but “understanding” limits use to poorest.
• So far it’s been sparingly used: just one case (Rwanda
imported the ARV TriAvir from Canada 2007-08).
Direct CL Experience
• US has actively used CL as part of antitrust rulings.
• The “cipro” scare in 2001.
• Canada frequently issued CL in medicines prior to
CAFTA/NAFTA.
• US and EU discourage CL use in middle-income
countries and in FTAs.
• Recent Thailand case.
• Conclusion: CL a useful tool in some circumstances.
Other Approaches to Date
•
•
•
•
Voluntary price discounts by pharma firms.
Voluntary licenses of (generally older) technologies.
Drug donations with tax incentives.
These have sometimes been effective, especially with threat of generic
competition.
• Difficulties:
– Perhaps sub-optimal selection of products;
– Sustainability is a question;
– Agreements may be short-term and not encourage investments in
delivery and treatment.
– Conditions may be attached that risk lock-in with particular drugs.
– May not work with newer therapies that must be patented in DCs,
particularly where markets exist in middle-income and rich countries
(eg, HIV/AIDS, heart disease).
Summary Assessment of Global Regime
•
•
•
•
So far largely reliant on market incentives and philanthropy.
No global pricing agreement exists.
Markets have not generated sufficient price differentiation.
Global purchase programs (eg, Global Fund, PEPFAR, World
Bank) have helped but offer no R&D incentives.
• Philanthropic and public-private R&D mechanisms are
improving but not yet sufficient.
• Licensing issues regarding IPR in new therapies remain
difficult but clear allocation of rights is essential.
• Resort to compulsory licenses has been contentious and
generally unworkable in poorest countries.
Broader Approaches
•
Proposals to deal comprehensively with investment and access issues.
– Example: DEFEND proposal
• International publicly funded program to buy patent licenses in target
markets.
• Payments to cover portion of fixed R&D costs and reflect social value of
drugs.
• Distribute drugs at price just above mc.
• Direct subsidies to delivery in poorest countries.
• Ban on re-exports outside of target areas.
– Tough questions:
• How to determine which countries to include? Many underserved
patients exist in middle-income countries.
• Which firms would be eligible to receive licenses? Generic exporters in
middle-income and developed countries?
• Determination of license fees?
• How to purchase licenses on drugs with multiple patents?
Broader Approaches
• Virtually coerced abandonment of patent
rights in poor countries:
– Lanjouw Proposal
• Patent buyouts
• Patent pools:
• UNITAID proposal 2008
• Similar problems of implementation exist for
buyouts and pools.
Concluding Remarks
• Working on price differentiation has been a
long and difficult process.
• Relying on market segmentation and
philanthropy is insufficient.
• Compulsory licenses are not a realistic
solution for poorest nations.
• Broader approaches that reward R&D but
permit targeted price advantages and delivery
investments need to be developed.