Thread by @surajbrf

This article exhaustively collects order data and current delivery data for the US, India, EU/UK plus some other countries.

It specifically addresses the assertion that India failed by not making large bulk orders during mid 2020 like the west did by spending billions .

1/

https://twitter.com/prasannavishy/status/1393382997307510788

Let us look at US data since they were at the forefront with Operation Warp Speed; all numbers here are backed by reference data.

This shows how much of total US order volume has so far been fulfilled:

2/

Observations: three of six OWS contracts have yielded 0 doses to date. The J&J contract has yielded 15m doses, all imported because entire US production has been discarded for QC reasons, and they have been written off as a source:

politico.com/news/2021/04/2…

3/

All US order volume to date is Pfizer/BioNTech + Moderna. The latter is only able to supply by deferring its obligations to EU/UK. The US embargoed all vaccine exports until May 2021, but allowed Pfizer production in Belgium to be imported, plus J&J from Netherlands.

4/

Now lets look at the EU. The EU’s approach was to make several large (half a billion $) non refundable deposit bets with major manufacturers, with very high order sizes totaling 4.2 billion doses, almost 10x their population:

5/

However, due to inability to coordinate between states, EU had trouble ramping up production. UK-EU tensions affected Astra Zeneca output, leading to a lawsuit. Meanwhile they order 1.8 billion doses from Pfizer:

6/

politico.eu/article/eu-pre…

The EU’s approach to making large orders and then sitting back and waiting, has not worked. Both India and US showed - data to follow - how production scaling needs constant handholding, coordinating multiple lines and incremental support.

7/

Europe also supported multiple vaccine development efforts that have not yielded any result as yet - GSK/Sanofi (dropped, Sanofi now makes Moderna), Curevac, Novavax. It’s order yield ratio is even lower than the US, due to poorer production scaling relative to order size.

8/

Now let us look at a relative success, the UK:

Total order volume 520 million doses, shipped to date ~55m doses. Mostly AZ and Pfizer, with a small amount of J&J. UK also got emergency Covishield supplies from India. UK only has 66m people so the target is smaller.

9/

So far we’ve looked at major countries that have local production bases. What about those that don’t ? Here’s a few:

Canada: dependent on EU-made Pfizer supplies due to US embargo until May 2021.
Australia: gave up 2021 full vaccination plans using AZ due to low supply.

10/

Russia: having trouble scaling Sputnik, whose doses are different and have different production yields.
Brazil: Dependent on AZ from India and now Sinovac.
Japan: Has so far vaccinated 2% of population and has received a fraction of its 120m order made mid 2020.

11/

Now finally India:

The orders are being made every month in close coordination with the producers. Like US, India orders 100m doses at a time. Turn around time is 1-2 months from order to the volume being consumed. All Indian supply to date is local production.

12/

Now let us look at the share of each region to total vaccine production to date, and correspondingly the number of orders they have made in number of doses:

13/

India has made far fewer orders than US or EU. And much later than them. And yet India has produced the same volume as them - one home grown and one license made (both of which had import content) . So have huge order volumes enabled production ? Nope.

14/

On the contrary the huge volumes reflect lack of coordination and risk transmission. Yes, vaccine development is tricky and has finite failure probability. But western development costs are extremely high too.

15/

Western order sizes reflect investing very high upfront amounts despite low probability of immediate success. Western supplies are mostly just Pfizer + some Moderna, both of whom depended on publicly funded development to make massive profits:

16/

nytimes.com/2021/05/04/bus….

Their total order size to production so far is wildly out of sync, despite many orders being 6-12 months old:

17/

In effect, production scaling is a problem requiring independent attention and order volumes are not what ensures it. Depending on kind of vaccine, scaling takes 3-9 months. For example Covaxin scaling effort began Nov 2020:

pib.gov.in/PressReleasePa…
pib.gov.in/PressReleasePa…

18/

The latest scaling estimates asserting 2 billion doses by end 2021 indicate the following monthly volume increase. There are several unknowns here, but Covishield + Covaxin output will likely maintain estimates.

19/

Covaxin may also add the Pune Biotec facility and Panacea Biotec, in addition to Haffkine, which means more than the current estiamte of 110m doses/month production from August:

Haffkine: pib.gov.in/PressReleseDet…

indiatoday.in/india/story/bo…

20/

In addition, the current scaling estimate does not count imported vaccines under the accelerated approval pathway. J&J has just applied for EUA under this path:

reuters.com/world/india/jj…

21/

pib.gov.in/PressReleasePa…

Developing, producing and scaling up vaccine prodn is a complex exercise. It requires continuous management. The US, India, EU/UK, China have all made 300-450m doses each. Indian production rate benchmarks favorably against major peers, though we are by far the poorest.

22/

This month India hits the 100m doses/month production mark due to Covaxin scaling up. In June, it hits a likely 150m doses (i.e. 5m doses/day).

Next few months should see substantial acceleration of current capability; each new available vaccine is just an added bonus.

23/23