‘India’s vaccine ecosystem is messed up’

The public narrative reflects loss of confidence in the Centre’s handling of the vaccination drive. The Centre’s decision to decentralise procurement of vaccines to let states take this up on their own was taken in April in response to the criticism. The Supreme Court has questioned if decentralisation is the best approach and raised questions about equitable access to the vaccines. Was the earlier policy of centralised negotiation and procurement and decentralised distribution better? What ails India’s vaccine policy?

Unfortunately, the whole vaccine ecosystem is messed up at the moment. For starters, the government has not ordered enough vaccines. While India needs around 2 billion doses to vaccinate the eligible population with two doses each, barely 12% have received one dose. In fact, till very recently the government had only put in orders for 100 million doses of Covishield, the vaccine that was supposed to be the workhorse of India’s vaccination programme. Since then it has added another order of 110 million doses of Covishield. The other vaccine, Covaxin, is similarly under-provisioned with an initial order of 8 million followed by a recent order of another 50 million.

Perhaps even more troubling is that the annual production capacity of Serum Institute of India (SII), which produces 90% of vaccines in India, is currently around 700 million doses which they are hoping to ramp up to 1.2 billion doses. Bharat Biotech’s annual production capacity is even lower at around 150 million doses. Clearly, these two entities do not have the capacity to vaccinate India on their own. Bizarrely, the government neither ordered any other vaccine nor did it invest in increasing the production capacity of these two producers until very recently. Procurement, thus, has been shambolic with no planning whatsoever.

The earlier distribution policy was just hamstrung by limited supply. If you don’t have it, you cannot distribute it. The pricing policy has two aspects: the price paid to acquire the doses and the price charged to consumers. The central government had negotiated a fabulous price of $2/dose with SII. That is one of the lowest prices in the world. They were also inoculating people for free. It is hard to argue with either of these strategies.

The new three-tiered policy that has been announced is more problematic. Tiered pricing by itself is not new. It is a standard strategy that enables producers to cover their average cost while meeting public health goals. Externalities in public health imply that charging an average price to all consumers can induce under-provisioning. Hence, charging differential prices in different segments of the market potentially allows low (zero) prices for poorer segments and higher prices for richer segments through cross-subsidization (though there are significant issues with keeping markets with different prices separate in order to prevent resale).

The policy announced is to sell 50% at the $2 price to the central government, and the balance to state governments and the private sector at significantly higher prices. This has two problems. First, it is not clear why state governments should pay a different price from the central government. Having state governments procure allows the central government to shift the burden of the consumer subsidy to the states. But why should their per dose subsidy cost be greater than that of the central government? This is particularly galling since the states have barely any independent revenue generating powers after the introduction of GST.

Second, the state governments are branches of the consolidated governance structure of the country who are also inoculating citizens of India using taxpayer money. Third, in this moment of acute vaccine supply scarcity, letting the states procure independently at a fixed price of ₹400 a dose only implies that there will be quantity rationing.

Moreover, the supplier will now decide how to allocate the remaining 50% of scarce doses. Since the vaccine supplier is neither consolidating public health information, nor accountable to the public, this has the recipe of a complete disaster in vaccine allocation.

What corrections would you recommend?

First, I would get rid of this decentralized procurement policy for now, along with the tiered pricing.

As long as supplies are so severely limited, we will have heavy rationing of produced doses. In this environment, it is best to have one procurer (the central government) who procures at a common price for the whole nation and distributes them based on a transparent, clearly announced allocation rule. The central government must strive hard to signal that it is an honest broker in distributing doses without political considerations coming into play.

Second, the vaccine should be distributed free of cost to the eligible population. The cost of vaccinating the entire adult population of the country would be between 0.5-0.7% of GDP. In the overall scheme of things, this is nothing. A 3-week shutdown or loss of economic activity will cost around $165 billion or 5.9% of GDP. This cost is 8-fold greater than the cost of free vaccinations for adults.

Third, India must authorize and procure vaccines beyond Covishield, Covaxin and now Sputnik. While it is probably late in the day to get quick access to doses from some of the larger global players, one should nevertheless begin negotiations with Pfizer and Johnson & Johnson (despite their safety concerns) about setting up domestic manufacturing and distribution networks that could also be used to ease the global supply crunch of vaccines. Pfizer has done this in China with Fosun. There is no reason for India not to be able to do the same.

Fourth, our goal has to be to have enough supplies so as to sustain a vaccination rate of 5 million a day. After having achieved a rate of around 4 million a day, we are now down to 2.5 million per day due to vaccine shortages. The 5 million a day target is the lower bound of what we have to aim for since even at that rate, it will take a year for us to get everyone two doses. The situation unfortunately is very grim.

The Supreme Court has asked the government why the option of compulsory licensing under the Patents Act is not being considered. Earlier, former prime minister Manmohan Singh recommended the same. What is your view?

It is certainly worth thinking about. It may be worth negotiating with Bharat Biotech’s Covaxin. One solution would be to grant a compulsory license to a number of manufacturers to manufacture Covaxin but pay Bharat Biotech a fee for each dose. It probably would be better to do this through the negotiation route whereby the government and Bharat Biotech can come to an agreement regarding a satisfactory per dose licensing fee. Clearly, Bharat Biotech wants a higher license fee than what manufacturers are willing to pay since otherwise they would have licensed manufacturers already. This is where negotiations and the power of the Indian Patent Act as a tool to nudge may be useful. It is definitely a national public health emergency.

The SC has asked why the model of the national immunisation programme has not been adopted for covid vaccines. Can/should it be?

If you are referring to rolling out a vaccination drive whose cost is paid entirely by the central government, then yes. In that respect, until now the vaccine program has been free of cost to the public. The difference is that vaccines for the coronavirus are in short supply globally. This makes the short-term challenge slightly different.

Should the manufacturers be directed to offer identical procurement prices across buyers for contracts involving different timelines, quantities?

In the current situation of acute scarcity, there should just be one procurer in my view who procures at a negotiated price and distributes at zero cost. The taxpayer funds the subsidy. Differential pricing and multiple procurers in the current supply situation is problematic.

One of the questions the SC has posed is why the price for the states has been set higher than what the centre secured for its initial order. Would a directive fixing the price at the level of the union government’s initial order with SII, or the imposition of price control be a good idea?

If you have only one buyer (as was the case till now), the buyer has some bargaining power to negotiate with the seller. This can get you a better price than otherwise. That said, vaccine development has overhead costs, specifically for R&D. Selling at marginal cost causes losses for the producer since she cannot recover these overhead development costs. It is important therefore to not go overboard with price controls that end up disincentivizing producers.

The price would have to be above marginal cost. I do not know if the initial price of $2 a dose is marginal cost pricing that SII somehow agreed to.

If so, it would be difficult for it to continue supplying at that price without the government investing in the company or the production process somehow. That is the other way to reduce the purchase price of the vaccine. All that the Union government has done is to extend an advance payment against vaccine orders that will be delivered in future. That is not the same as investing in the company or in product development.

There are demands for price controls and nationalization of hospitals and vaccine producers. Will state interventions of this nature improve the situation?

I am totally against both price controls and nationalization of hospitals and vaccine producers. One glance at the state of government hospitals should make it clear that nationalization of hospitals is not the answer. The solution has to involve partnering with the private sector. This is typically done by incentivizing the private sector. Looking for solutions where the state forces private operators to supply public goods at losses or by appropriating their private capital is the wrong approach. The state has to pay the subsidy for the vaccine out of its own pocket.

There is a view that the distribution of the 150 million vaccine doses procured and sent by the Centre to states before 1 May was biased in favour of select states. You have studied the states-wise data on doses administered so far. Have you found evidence for biases?

My examination of the vaccine distribution data till 27 April does not reveal any bias. As you will see in the Chart, the national vaccination rate (at least one dose) on 27 April was around 11.5%. When you look at the vaccination distribution rate till that date (in terms of doses distributed as a share of the state population), the top states that were markedly below the national average were Assam, Bihar, Tamil Nadu and Uttar Pradesh. Among states that received significantly greater doses than the national average were Gujarat, Kerala and Rajasthan. So, there’s no pattern of any systematic bias.

What does the data tell us about the criteria that the Centre may have used to allocate doses it procured across states and the capacities of states to deploy allocated doses?

My reading of the data is that to a rough approximation, the state-wise vaccine distribution has followed the population distribution. There would clearly be some adjustments that would have to be made. Thus, places with smaller population shares of people above 45 would have received lower allocations since only the 45+ age groups were being vaccinated till 30 April. Similarly, I am sure adjustments must have been made for places with more heavy outbreaks of the virus. The cases of Bihar and UP where the doses distributed are just 5-6% of the population, suggest that the distribution possibly also considered the absorptive capacity of the states. That clearly depends on the state of the public health infrastructure in the state.

These vaccines were for the ‘highest priority’ population, as defined by the Centre. What is your view on the segments included, for whom the Centre will continue to procure and distribute free vaccines going forward?

I personally think that a fruitful vaccination strategy would be to prioritize the working age groups of the population. Large-scale lockdowns are neither feasible nor effective in highly dense urban and peri-urban parts of India. So, in these places, the vaccination drive needs to prioritize the workers who cannot afford to stay at home.

Puja Mehra is a Delhi-based journalist