Saturday 11 June 2022

No, Poor Countries Shouldn't Try To Make Their Own Covid Vaccines

Source: UNDP
One of the impressive features of humanity's response to Covid was the development of successful vaccines within only 10 months, and the production and distribution of 12 billion doses around the world. One of the worst features was the inequality in the distribution of those vaccine doses.

As of June 2022, only 18% of people in low income countries have received at least one dose compared to a global average of 66% and an average of 72% in high income countries (Our World in Data/UNDP). 

Obviously this contrast looks very unjust. Many people in rich countries have now received a full course of vaccination and multiple booster shots, even if they aren't particularly vulnerable. They also benefit from access to large well-resourced medical systems and thus high survival rates even if they are unlucky enough to be infected. Covid vaccines are clearly not being distributed to where they would do the most good.

One response to this injustice has been to argue that Low Income countries should be enabled to manufacture their own vaccines. Unfortunately, this is one of those ideas that sound nice but don't stand up well to systematic scrutiny (see previously: ideas vs arguments).

1. Patents are Irrelevant

The patent system confines innovations to the affluent & makes the others wait.
Millions die as a result.
The pandemic makes this injustice visible to all.

— Thomas Pogge (@ThomasPogge) May 1, 2021

One common complaint concerns intellectual property rights, which give the owner of a patent the right to control who gets to produce a medicine. This gives the owner a monopoly position in the market since no other companies are allowed to sell the same product, and this in turn gives them the power to decide what price to sell at (since no competitors can undercut them). Naturally they will choose a price that maximises the amount of profit they make (by selling fewer units at a very high price), not the quantity of vaccines produced (by selling to everyone who can afford to pay the marginal price of production). By contrast, in a competitive market companies attempting to maximise their profits will have to compete with each other on price and so will automatically maximise the quantity of vaccines produced.

The critique is that international trade treaties (e.g. TRIPS) prevent pharmaceutical companies in poor countries from copying Covid vaccines (and many other life-saving medicines) that have been invented elsewhere. They must either import those vaccines from the original manufacturer or negotiate a fee with them to be allowed to use the recipe (patent). This perpetuates monopoly pricing and hence reduces the total amount of vaccine produced far below the number of people willing to pay what it costs to make. 

According to this critique, our patent system has created a humanitarian clusterfuck: we have a terrible disease, and we have an effective treatment that can be produced at an affordable price, but our capitalist institutions are preventing billions of people who need that treatment from getting it! The situation will be worst for poorer countries since the few doses made will go to the highest bidder, i.e. the people who can afford (or whose governments can afford) the exorbitant profit-maximising price. 

As a generic critique of patents in medicine, this makes some sense. However, it is quite wrong to suppose that the solution to Covid vaccine inequality is to suspend or break their patents.

Firstly, it is not the case that poor countries are legally constrained by TRIPS from breaking international companies' patents ('compulsory licensing'). TRIPS does not apply yet to the poorest countries, and in any case has a clear exception for public health cases (like Covid). This exception also allows manufacturers in other countries (like India and Brazil) to make and export such medicines regardless of their patent status. 

Secondly, legal authority is necessary but insufficient for manufacturing medicines. Patents by themselves are useless without mastery of the technical process of production, for which you need specialised, expensive and scarce machinery, personnel, and components. (This is why pharmaceutical firms generally don't even bother to patent their medicines and technologies in the poorest countries). 

The reason the poorest countries of the world don't make their own medicines is not international patent law. It is for the same reasons they don't make other sophisticated products, like cars or computer chips: they are too dysfunctional for exporters to set up in, and too small to support domestic producers. It is not surprising that countries too disorganised to keep electricity flowing 24 hours per day or to distribute even half the vaccines doses donated to them under COVAX also lack a biotechnology industry and so haven't any companies capable of manufacturing Covid vaccines under license.

Thirdly, patents do not provide monopoly pricing power in the specific case of Covid. As of June 2022 there are 9 different vaccines accredited by the WHO; 30 more accredited by national governments (e.g. China); and hundreds more still in development (Wikipedia/WHO). Thus the generic challenge against patents - that they generate a perverse incentive to produce fewer doses of medicine than are needed - does not hold. Every vaccine manufacturer already has a clear financial incentive to maximise production because the only way to make more money is to sell more units than their competitors. 

This matches what we have seen in the real world: capitalist pharmacorps reaching out across borders to any manufacturer capable of being brought up to code to contract for vaccine production. This is a complicated and slow process as it requires technology transfer (sending scarce personnel and equipment) to set up and train new production lines. Nevertheless, it achieved a tripling of the world's total vaccine production from one year to the next. This suggests that capitalist pharmacorps have already had the right incentives to maximise Covid vaccine production. It also puts the burden of proof on critics to demonstrate the source of their confidence that a 'more just' alternative could have done even better (and definitely not worse).

2. Poor Countries Could Not Afford To Make Their Own Vaccines

The patent law challenge fails so obviously that some proponents have actually accepted the point and revised their argument (eg). Instead they propose that poor countries be helped to develop their own pharmaceutical manufacturing sectors so that they are in a position to make use of patents in future and so won't be at the end of the line for vaccine doses in any future pandemic. Unfortunately the logistics of this are implausible.

The point of trade is that it allows specialisation and economies of scale which translates into much greater production of valuable things for the same amount of resources. The world is richer thanks to trade because we can consume more than we would otherwise be able to. It follows that smaller poorer countries benefit the most from trade since their domestic markets are too small to allow very much economy of scale. With access to global markets their producers can lower their costs by economies of scale, and their consumers can access cheaper products from efficient foreign producers. On the face of it, it is therefore rather strange to see small poor countries being encouraged to develop their own pharmaceutical manufacturing industries as a kind of insurance policy against being overcharged or left at the end of the line in public health emergencies. 

To start with, building and maintaining a modern pharmaceutical manufacturing sector capable of rapidly scaling up to produce vaccinations for the whole population would be very expensive. How would poor countries pay for this insurance policy? 

On the one hand, it could be funded by government subsidies, either by direct capital transfusions or guaranteed contracts to make medicines for local demand. In either case there is no path to financial sustainability, which is why previous efforts to establish a domestic pharmaceutical industry in small poor countries have failed repeatedly in the past. The result is that we would see the governments of very poor countries spending a large amount of their limited tax revenue or foreign aid on a much more expensive way of obtaining medicines than just importing them. 

On the other hand, it could be funded by exports. If those exports were to rich countries then this would quickly run into legal and diplomatic opposition. Violating foreign companies' patents in order to undercut that company's prices in rich country markets would be very much against TRIPS and such exports would almost certainly be blocked. But suppose TRIPS and other relevant laws were somehow changed to allow this. Then we would have the situation where the economics of drug development would be under threat. If competitors are able to sell at just above the marginal cost of production, then that is the only price the drug developer will be able to charge too. But then it won't be possible to fund the cost of drug development (including the cost of all the research that doesn't result in a successful medicine). So spending on drug development would fall to zero and there would be no super fast development of a vaccine to the next pandemic for the poor country manufacturers to copy. 

Alternatively it could be funded by exports to other poorish countries. This wouldn't take many sales from the big international pharmacorps and is allowed under TRIPS. There are two problems with this though. 

Firstly, there are already huge pharma exporters in the developing world with economies of scale (in India and Brazil) which supply generic versions of patented medicines to poor countries. It seems unlikely that a plant in Malawi would be competitive with those. 

Secondly, it is logically impossible for every poor country to create a domestic pharma industry by becoming an exporter of essential medicines to other poor countries. No country would accept those imports because it would undermine their own export financing model, and also create the kind of international dependency (and hence vulnerability) that the whole domestic production scheme is supposed to prevent. 

Constructing their own pharmaceutical industry as an insurance policy will definitely make poor countries significantly poorer. Unfortunately it it still wouldn't be much help during a pandemic. The problem we encountered in the case of Covid is that even though the patent owners increased production as fast as they could, it still took a long time to scale up. One result was scarcity despite the existence of multiple competing vaccines (i.e. the absence of the artificial scarcity of intellectual monopoly). Rich world governments could afford to bid higher to be at the front of the line to receive what was produced. (It helped that they had prefunded a lot of the production expansion in exchange for preferential treatment.) But governments also acted nationalistically whatever their wealth level: for example India as well as the US and EU all restricted exports of vaccines and key components produced by companies in their territories at key moments. 

If there were to be another epidemic like Covid and every poor country had prepared by building up its own domestic pharma manufacturing capacity (at great expense) they would still be vulnerable to these other kinds of scarcity. There would be a surge in competition for the scarce personnel and equipment needed to retool production lines to make the new vaccine. There would also be competition for the key components needed for making the vaccines (some 280 from 86 suppliers across 19 countries in the case of Pfizer). And the winners of that competition would again be the larger and richer countries. Malawi's pharma factories would be last in line for what they need to produce vaccines for the same reasons that Malawi's people were last in line to get imported Covid vaccines. Countries like Malawi are just never going to win a game of vaccine nationalism. Instead we should be looking for a way to change that game.

3. One Partial Solution: Incentivise Pharmaceutical Companies To Maximise Health

Generally I have little time for Thomas Pogge's ideas about global justice (he thinks poverty is a conspiracy by rich country governments against the poor, and so is economics whenever he doesn't like what it says). Nevertheless there is one idea of his that does stand up to scrutiny and is very applicable to this case, the Health Impact Fund. This creates an alternative opt-in patent system in which companies are paid according to how many Quality Adjusted Life Years (QALYs) their treatments save. 

The Health Impact Fund's key innovation is to explicitly link the good that companies can do for people's health to their profits. This avoids the monopoly pricing problem of traditional patents in which companies make more money by charging so much that only a few can afford to pay (and focusing their research on rich people's problems). To some extent it also evades the vaccine nationalism that occurred with Covid: the gross misallocation of scarce vaccines from where they would have done most good to where the people with most money lived. (It is less successful at countering direct government interventions such as export bans.) 

If pharmacorps were paid according to the QALYs their treatments saved, then they would direct vaccines to the places and to the particular people in those places whose health would most benefit from receiving them. If this had played out in the Covid pandemic, this would have meant far more vaccines being sent far earlier to poorer countries and directed to vulnerable populations such as the elderly and those with significant risk factors.

Interestingly, this approach would also encourage pharmacorps to take responsibility for system level consequences - something we nowadays expect from governments but don't always get (especially in poor countries). For example, a pharma corp might realise that prioritising vaccines to health workers would generate more QALYs, because they would save more people from Covid and other illnesses if they didn't get sick themselves. 

But pharmacorps might get more involved than that. The global map of vaccinations shows a lot of variation in poor countries' performance in distributing the vaccine doses they have received. At one end of the spectrum Mozambique has managed to distribute almost 100 doses per 100 people (counting each dose individually) by distributing almost all they received (WHO). At the other end, Congo has managed to distribute only 1/6 of the scanty supply it has received, leading to a total vaccination rate of less than 3%. From the perspective of most donors, Congo's dysfunction makes it a poor choice to give scarce vaccines to (which is presumably why it has received so few doses in the first place). However, from the perspective of a company with a Health Impact Fund patent, Congo looks like an attractive place to mine QALYs because so many people's health could be improved so cheaply, and this would more than compensate for the expense of building its own distribution system. In other words, this kind of patent aligns pharma companies' incentives with maximising health far more systematically than regular patents do, and also far more systematically than national governments demonstrated. 

There are two major practical challenges to the Health Impact Fund idea: funding and measurement. Pogge suggests the fund would need to pay out $6 billion per year to be an attractive alternative to the standard patent route and to make a real impact on public health. On the one hand this is a very small amount of money to find in a $120 trillion world economy and in the context of the economic cost of a pandemic (estimated at $350 billion per month at the peak). On the other hand, the world has not previously shown much enthusiasm for funding public goods: before Covid the entire World Health Organisation had to scrabble by on a budget of $4.4 billion (a substantial chunk of which was a private donation from the Gates foundation). Measuring the health impact of medicines in order to calculate the rewards due to different pharmacorps will necessarily be complicated and imperfect. Nonetheless, the attempt in itself would yield valuable data about how well medicines actually work on normal populations in non-controlled circumstances that gold standard RCT drug trials only hint at.


Inequality of access to Covid vaccines (and many other life-saving medicines) is a terrible and shameful injustice. We should think hard about how to prevent it in future. One idea that certainly will not work is to lean into vaccine nationalism and try to build a pharma-manufacturing industry in every poor country. More promising is to accept international dependence and find ways to revise the capitalist incentives of pharmacorps so that they oppose vaccine nationalism rather than amplifying it.

Several commenters have mentioned the case of Cuba as a counter-example. Cuba got a lot of positive media attention for inventing 2 Covid vaccines of its own, one of which they have manufactured in sufficient quantities to vaccinate their own population (and perhaps some people in allied countries like Venezuela and Nicaragua - the details are unclear or uncorroborated). However, there are several reasons why Cuba's success does not undermine my argument.
  1. Cuba may look poor compared to the US, but it is classified as an upper middle-income country. As the graph at the top of my essay shows, countries in this category got all the vaccines they needed only very slightly slower than rich countries. Cuba is in an entirely different category of resources and organisation to really poor countries like Malawi.

  2. Cuba has a medical research and export industry, yes. But that doesn't mean that all poor countries could have one at the same time. (The essay's argument)

  3. Cuba has not succeeded in exporting the vaccine at scale because it hasn't achieved accreditation by any internationally respectable regulatory agencies. These would vouch for the credibility of Cuba's self-reported safety and effectiveness data, and also the quality and safety of its manufacturing process. People in poor countries should have access to medicines produced to the same standards as in rich countries. In the case of vaccines especially, independent credible quality control is essential for people to have faith in them - and without faith, take-up will be low and so the vaccine's real effectiveness falls.