Skip to content

That $9 trillion bill on the sidewalk: Why don’t third-world countries borrow to get vaccines?

Gaurav Sood quotes this from June:

“To get roughly 70% of the planet’s population inoculated by April, the IMF calculates, would cost just $50bn. The cumulative economic benefit by 2025, in terms of increased global output, would be $9trn, to say nothing of the many lives that would be saved.”

Gaurav continues:

The Economist frames this as an opportunity for G7. And it is. But it is also an opportunity for third-world countries, which plausibly can borrow $50bn given the return on investment. The fact that money hasn’t already been allocated poses a puzzle. Is it because governments think about borrowing decisions based on whether or not a policy is tax revenue positive (which a 180x return ought to be even with low tax collection and assessment rates)? Or is it because we don’t have a marketplace where we can transact on this information? If so, it seems like an important hole.

Here’s another way to look at this point. Among countries where the profits mostly go to a few, why do the people at the top not come to invest together so that they can harvest profits later? Brunei is probably an ok example.

Economists talk about the $20 bill on the sidewalk. In statistics we talk about posterior predictive checks. Bill James uses implausible estimates to reveal problems with a model.

On this blog we’ve written about the two modes of thinking of microeconomics: (1) People are rational and respond to incentives. Behavior that looks irrational is actually completely rational once you think like an economist. and (2) People are irrational and they need economists, with their open minds, to show them how to be rational and efficient.

These two modes of thinking roughly correspond to what we do in statistical workflow, or in science in general. A model is like a horse. You ride it as far as it can take you, and then when it collapses you get off and walk. Or, better still, a model is like a car: you drive it as far is it can take you, you refill it with data where possible, you give it repairs as needed, sometimes prophylactically, and then when it finally dies on you, you take a look and figure out what went wrong with the goal of designing a new car that will handle this difficult terrain.

In the case of the two modes of economic thinking, the default model or starting point is to assume that people are making rational decisions, but at some point the economist might break down and say that, no, these decisions could be improved through outside intervention. It’s tricky. On one hand, people often do have good reasons for their behavior; on the other hand, we seek out advice all the time, which suggests that we will not necessarily make rational decisions by default.

Anyway, to return to Gaurav’s point: I don’t know enough about economics, vaccines, or the politics of international investment to have a sense of the merits of the argument, but it seems like, again, this could go in two directions:

Possibility 1: It seems obvious that these countries could raise the $50 billion and get this amazing return on investment. The fact that they aren’t should lead us to understanding the good reasons whereby they are rationally deciding not to do this.

Possibility 2: They should do it already! They’re not doing so because of short-term thinking, the sort of thing that economists can help shake them out of.

As I said, I’m not sure how to think about this one; just wanted to share with you.


  1. Dale Lehman says:

    Or a third possibility – mundane as it is – there are “transactions costs” that get in the way of mutually beneficial trading. No irrationality is required (although there may be some), only the difficulties of arranging the loans, repayment terms, etc. When it comes to international politics, I am no expert, but existing institutions seem far from ideal from designing and executing trades such as this, particularly in a short time frame.

    • Andrew says:


      I guess I’d count transaction costs as part of possibility 1 above, but maybe it deserves its own bullet point. Another way to put it would be to add the following:

      Possibility 3: Problems with the international financial/political system that get in the way of this efficient transaction being performed.

  2. Rahul says:

    ..or maybe countries do not trust IMF estimates?

    The covid era hasn’t really boosted confidence in most estimates.

  3. Wonks Anonymous says:

    Economists don’t necessarily assume that governments behave in a socially optimal way. Or at least they haven’t since the development of public choice theory.

    • Andrew says:


      The logic of assuming governments behave rationally is the same as the logic that people or firms behave rationally. From one direction, governments have goals, incentives, etc., just as people or firms do. From the other direction, people and firms are composed of different, contradictory elements, just as governments are. Whether we are discussing governments, people, or firms, economists are free to use argument #1 or argument #2 as they see fit.

      • jim says:

        “The logic of assuming governments behave rationally is the same as the logic that people or firms behave rationally.”

        This is incorrect.

        Economists believe people behave rationally *in aggregate*. There are billions of people. There are about 100 governments in this subset. Also selecting this subset of governments is akin to selecting the low-performing half of a small population.

        But that point is probably moot, because chances are good there’s no puzzle here at all.

        Like most social “investment” forecasts, the assumptions that went into the original calculations are surely wildly optimistic – or more likely outright false. Even this subset of low performers would likely catch on to such a purportedly immense opportunity. The link doesn’t give the time frame for the purported return of $9T – a whopping 17900%!!! – but presuming even a twenty-year time frame that’s an annual return of 30%, which pretty much any moron can figure out.

        • Andrew says:


          I wouldn’t say that economists “believe” people behave rationally, whether as individuals or in aggregate. Rationality is a model, and economists understand that the model imperfect but can often be descriptive as well as giving insight when it fails. To the extent that economists use this model, they do use it to characterize individuals, not just groups. An extreme example is the ludicrous model of “rational addiction,” which is indeed intended to apply to individual addicts. But this comes up all the time in microeconomics. There’s a separate strain of thought in economics, the “as if” argument by which aggregate outcomes can look the way you’d expect if individuals were behaving rationally, even without requiring rationality of the individuals, and there’s yet another argument claiming efficiency in the stock market based on the potential for arbitrage, which again does not require that all players behave rationally, just that there are some players in the system who can enforce rationality. So, lots of theories out there. But the theory of individual rationality comes up a lot too.

        • Jo says:

          I strongly disagree with this statement.

          The economic theory of rationality has pretty concrete implications for individual behavior (mainly in the form of consistency of behavior), but only puts very loose restrictions on aggregate behavior. So even if people are individually rational, that does not mean that aggregates behave “as if” they are rational. The most famous of these results is Arrow’s theorem (there is no fair way to aggregate individual preferences into a single utility function), but in macroeconomics you similarly have the Sonnenschein–Mantel–Debreu theorem, which shows that individual rationality places only very few restrictions on _aggregate_ demand. E.g. even if individual demand is downward sloping (so if prices go up, demand goes down), there is no guarantee that this is true for aggregate demand (if prices go up, aggregate demand might go up). In other words, economic theory does not imply that a billion people together act more rationally than any individual. In fact, if you only assume individual rationality, almost “anything goes” for the aggregate.

          You can of course get around this by making stronger assumptions (indeed, a lot of modern macro simply assumes there to be an aggregate household that acts as if it is a rational representative of the whole population and political economists make extra assumptions to make the problems of Arrow’s theorem go away), and you’re free to believe that individuals are not rational but that aggregates are. Even if grouping a few people together makes errors cancel out, it is not so obvious that this is true if you group together the whole world population.

          • somebody says:

            Seconded—that individuals are irrational and groups are rational in aggregate is essentially the exact opposite of the modern economics position. Perhaps the more precise statement you’re looking for is that “most but not all people are individually rational”, which is closer to the practical economist’s theory.

      • Wonks Anonymous says:

        The incentives of government are different. They don’t go out of business like a firm. The question instead is whether they can get re-elected or not (or overthrown, if it’s not a democratic government).

  4. Koen says:

    Speaking as an economist, I don’t think that the problem is the demand or supply of money, perhaps not even the supply of vaccines. In most cases, the governments don’t even really need to borrow to purchase the vaccines; given the importance of the vaccines, they can free up even limited budgets. The problem is the complexity of the licensing, storage and administration. So you bought hundreds of millions of vaccines and now what? Many governments do not have a well managed field force and planning capacity to prepare for this on the short term. Health systems are answering a bunch of difficult questions – who will manage the vaccine distribution and with what safeguards? What is the communication plan, how will you ensure that vaccines won’t spoil, how will you handle citizen complaints, how will you ensure that the population will develop trust in the vaccines and their administration, how will you handle personal data of vaccine recipients. Some countries e.g. in South Asia are handling this quite well, but they have an enormous and relatively professional civil service. The same is not true everywhere. Recall that distribution of vaccines that have been developed 50 years ago is extremely difficult for many low income countries. Note that this can mask as a money problem. An international lender will not provide funds unless these basic issues are answered. Not acting will be bad, but making all the fundamental errors in distribution is likely to hurt more on the medium and long term. The reputational impact for international lenders to make these basic mistakes is also large. Instead of asking why don’t these lenders give money (which of course is a fair question), the question will become why were these lenders so unprepared that people died?

    • BrianM says:

      I would add another option – vaccine availability. While most (all?) countries want to deliver the vaccine, first world countries paid up to get the vaccine first.

    • Mike says:

      This strikes me as being on the money (ha!) — even assuming an entirely elastic supply of vaccines and money, you run up against there not being an infinite well of human hours that can be diverted for vaccination. There’s even fewer useful human hours for things like supply chain management and distribution, considering how hard a problem this is.

      The US military does an absolutely fantastic job of managing logistics, because that’s more or less the hardest part of war. Hence the massive military involvement in the US vaccine rollout, why I got my shot at a National Guard-run station. Most developing countries don’t have the same government-controlled easily-diverted logistics expertise at hand.

    • morris39 says:

      Good points but there maybe other factors in play. One such is the tremendous and mysterious inflation in select endeavours e.g. education.
      There is a story probably apocryphal, but maybe with some truth about smallpox vaccination in NYC in 1911. According to this tale most of the people needing vaccination (city pop.~5 million) got it on the sidewalk, in about a weeks time.
      As an aside I told this story as a joke to the nurse giving me my shot
      and she asked me if I was there with a straight face.

    • Fred says:

      South Korea is a good example of a country where a) people are very willing to get vaccinated, b) the government is rich enough, and c) storage, administration, etc. is set up well but the vaccination has been pretty slow.
      Supply is still a binding constraint for many countries before the problem of logistics (which is real and likely underrated) kick in.

  5. oncodoc says:

    A widespread benefit may be hard to recover. The $9 teradollars won’t pop up as a convenient pile of money in a single location. Instead it will be spread over the entire population. This means it will take time to manifest itself. Governments actually seem to be very shortsighted are not willing to do things that are beneficial in the long run. Each of us, whether right, left, or in the middle, can come up with examples of this. Almost every war is an example of governmental short term thinking.

  6. kj says:

    Rationality depends a lot on those estimates. I’m trying to track down where the IMF got their benefit estimates from. These two seem to be the sources, but I’m not sure:

  7. MJM-WA says:

    Could this arise in part from the fact that C19’s impacts have been largely a “ first world” problem? The enormous disparities between the population fatality rates in Europe and North America versus Africa and Asia-PAC are quite striking. ( ). There are exceptions to this as we see remarkably low PFRs in “first world” countries such as Japan, Korea, and Singapore…. And relatively higher pop fatality rates in certain less developed LATAM countries.
    The variations across countries —and across states in the USA— are quite striking and puzzling !

  8. jim says:

    Possibility 3: a 17900% return on investment BS.

  9. Neil says:

    Or perhaps third world countries don’t want to take on even more debt because their economic hit is comparatively small and the slice of the economic benefit will be minor compared to the countries whose companies are selling them the vaccines.

    As Koen writes there are also organisational issues related to distributing and administiring the vaccines beyond just purchasing them.

  10. Michael Nelson says:

    Based on comments, the Economist quote has been misread by many. First, it doesn’t say it costs 50B to buy vaccines, it says it costs that much to get people vaccinated. That figure may seem wildly implausible, but the quote doesn’t say that that’s the cost for inoculating 70% of people in 3rd world countries. That figure is referring to the “easiest” 70% to get vaccinated. The easiest subgroup is people who are already vaccinated! I don’t know what percentage of the world that is, but it costs zero dollars to vaccinate them. The second-easiest group to inoculate is people in countries that already have the vaccine, but who have chosen so far not to take it. Unless the 50B is accounting for the cost of persuading the holdouts, their contribution to the cost is trivial. The rest of the people in that 70% are probably spread across many countries, in large cities or other easily accessible areas. That’s the answer to Sood’s question: the benefit goes to the world, not any one country, so no one country is incentivized to pay up. That’s consistent with the article’s point: rich countries have to pay because the benefit is to the total economy, which means it will go mostly to rich countries.

    Of course, the original article is paywalled, so the quote may not represent the article, but that’s what the quote literally says.

    • The actual Economist article is quite good, and I certainly disagree that its point is that “rich countries have to pay because the benefit is to the total economy, which means it will go mostly to rich countries.” Rather, as economics isn’t a zero-sum game, the rich countries have an opportunity to spend $50bn to do something that vastly benefits everyone, including themselves.

      “After four years when America often seemed more interested in kicking sand in the world’s face than helping it to its feet, vaccines provide it with the ideal means to rehabilitate its image. The United States and its allies displayed great technological mettle by developing several highly effective vaccines and scaling up their production in record time. They therefore have something tangible and urgently needed to offer the rest of the world. Mr Biden is unlikely to get a better chance to demonstrate the benefits of American leadership, and the power of democracy and free markets to boot. Making sure that the rest of the world is inoculated as quickly as possible would be a shot in the arm for America and its allies, too.”

      • Michael Nelson says:

        Thanks for clarifying. Since I couldn’t read the article, the inference that the majority benefit will go to the rich is my own, not from the article. Whether the majority benefit goes to the richest countries depends on how you define “benefit”: it’s indisputable that more of the 9 trillion dollars will go to the G7 than to any one poor country, but the marginal benefit will definitely be much greater for the poor country.

        My main point (after correcting misreadings) was that the incentive for a poor country to pay up, even if they have 50B lying around, is less. Yes, it would be the rational choice and would benefit them more than the alternative. But they’d be creating what would superficially appear to be a massive free rider problem. It would mean paying for vaccines in other countries even as some significant proportion of their own country went unvaccinated. (Remember, the IMF’s goal is to get a total of 70% vaccinated, with no guarantee that any one country will benefit more than the others.)

        Rationally, benefitting others isn’t really a “problem” for the country that pays but more of an ancillary benefit to the world. But it will be immediately obvious to a country’s people that it is paying for vaccines in other countries, whereas it will not be obvious that any improvement in the country’s economy over the next four years is directly caused by doing so. The impact of the investment is both delayed and discreet. Their alternative is to build a coalition of poor countries who pool resources, but the barrier there is self-evident.

        Which leaves rich countries, who already have economic coalitions (like the G7), or a single rich country. Americans would also largely see this as charity rather than a rational economic investment–we aren’t inherently more rational than people in other countries–but it’s such a small amount to us that Biden won’t really have to worry about riling up the “America first!” crowd.

  11. S says:

    >but at some point the economist might break down and say that, no, these decisions could be improved through outside intervention

    The ability to improve people’s decisions with outside intervention isn’t mutually exclusive with those decisions being rational. All rationality means is that people behave rationally subject to the information they have (which really just means people have clear preferences which have certain properties, like transitivity). It’s perfectly possible for me to prefer A to B, and think (incorrectly) that doing C will give me more of A. If you tell me C will actually lead to less A and more B, you’ve just improved my decisions, but I’m no less rational for it.

  12. Byeongmin says:

    I believe the problem is that although raising funds might be rational for the whole G7 in aggregate, it is not for each singleton country. The whatever benefit global vaccination would generate is going to be distributed to the whole world (in an uneven manner), and therefore Germany, for example, might find it not profitable to donate the necessary funds if it expects the returns (be it financial, diplomatic, or else) to the country relatively small.

  13. Stevec says:

    According to CDC estimates for the US from March 2021 (found on The Spectator website), the infection fatality rate:

    18-49 years: 0.05%
    50-64 years: 0.6%
    65+ years: 9%

    I would expect, based on the lower rates of obesity etc in developing countries these numbers would be lower there.

    If you are in the government of a developing country with all kinds of problems, including lack of clean water, perhaps covid might not seem like a top 10 economic priority.

Leave a Reply

Where can you find the best CBD products? CBD gummies made with vegan ingredients and CBD oils that are lab tested and 100% organic? Click here.