Tag: tinstaafl

Recursively. That’s The Only Change I’d Make.

The ability to exercise good judgment is the binding constraint in development is the title of Gulzar Natarajan’s blogpost on Oliver Kim’s essay, which we’ve covered earlier here.

Almost all of doing development is about making non-technical decisions (the technical ones are easier, have limited degrees of freedom, and mostly slot themselves into place). Such decisions are invariably an exercise of judgment by taking into consideration several factors, one of which is the technical aspect (or expert opinion). The most important requirement for the exercise of good judgment is experience or practical knowledge. In the language of quantitative science, it’s about having a rich repository of data points that one can draw on to process a decision.

Source: https://urbanomics.substack.com/p/ability-to-exercise-good-judgment

This applies, Gulzar Natarajan says later on in the essay, to “industrial policy and promotion of industrial growth, macroeconomic policy and inflation targeting, and programs to improve student learning outcomes or skills, increase nutrition levels and health care outcomes, improve agricultural productivity, and so on”.

Let’s take one of these and think about it in slightly greater detail: health care outcomes. Let’s do this in the context of India. Answer these questions, for yourself:

  1. What is the best possible health care outcome you would wish for, in India’s context? Define it however you like – everybody should have excellent healthcare so long as they can pay for it is one option. Everybody should have excellent healthcare regardless of whether or not they can pay for it is another option. Everybody should have free healthcare until we reach a per capita income of x dollars (adjusted for inflation and purchasing power parity, if you prefer) is a third option. You can whistle up a million more, and feel free to let your imagination run wild. You get to define the best possible health care outcome for India – setting the standard is up to you.
  2. Microeconomists will call this the indifference curve, and ask you about the budget line. Mathematicians will call this the objective function, and ask you about the constraints. Humans will say “Haan woh sab to theek hai, magar bhaiya, kaise?”. This is the part where we encounter the bad news – what are you willing to give up in order to achieve your best possible outcome? Best possible health care outcome subject to we spending not more than 20% of our GDP on health might be a constraint you choose to apply, for example. Other people may start to froth at the mouth at the thought we spending 20% of our GDP on health, but you do you (for now). Reduce spending on defense, and pensions, and highways, and on education, you get to say (for now). In my little ivory tower, you get to say, I want India to focus on healthcare outcomes, and healthcare outcomes alone – and I’m ok spending x rupees on it. We get to not spend those x rupees elsewhere, of course – remember, opportunity costs are everywhere, including in imaginary ivory towers.
  3. Here’s another way to think about the same problem. You could also say, I’m optimizing not for healthcare outcomes in the short run, but for free markets. I’m not doing this because of my love for free markets per se, but because of my conviction that this path, and this path alone is the only way to deliver the best possible healthcare outcomes eventually. Sure there will be mistakes along the way, and sure some healthcare services will be denied to people who need it the most for now. But eventually, the market will correct all of these errors, and that in ways we simply cannot know right now. Why can we not know them right now? Well, because we are not omniscient. We don’t know what errors will crop up, and we don’t know what solutions will work best for whatever errors will crop up. If we did know this, we could have avoided those problems in the first place, no?
  4. There are, in other words, unseen consequences to Bastiatian solutions as well. That’s just a fancy way of saying there are opportunity costs everywhere, but let me make the point more explicitly: the opportunity cost of an immediate application of a completely free market solution to healthcare is poor health outcomes for at least some folks today. I might be wrong about this, so please, don’t hesitate to tell me the how and why of it.
    For example, let’s say that government stops spending even a single rupee on healthcare (no CGHS, no PMJAY, no ESIC, no Jan Aushadhi, no government run hospitals, no PHC’s, no government run vaccination programmes, nothing) at 12 pm today. Will health markets be Utopian at 12.01 pm, or will they transition to Utopia eventually? How long is eventually? What problematic outcomes will occur along the way, and do we correct for them? How?
    For example, we may learn that poor families in rural Jharkhand now do not have access to healthcare because all government intervention has stopped. Do we do something about this? If yes, what? If not, why?
  5. This cuts both ways, of course.
    For example, let’s say that government doubles its current expenditures on healthcare, at 12 pm. Will health markets be Utopian at 12.01 pm, or will they transition to Utopia eventually? How long is eventually? What problematic outcomes will occur along the way, and do we correct for them? How?
    For example, we learn that corrupt practices when it comes to invoicing in procurement departments have gone up because government expenditure has gone up. Do we do something about this? If yes, why? If not, why?
  6. Given your ideological bent of mind (and we all have one, learn to live with it), you have an urge to say “Ah hah, exactly!” and “Oh, c’mon!” to pts 4 and 5 – in that order. Or to pts 5 and 4 – in that order – it depends on what your ideology is. But if both of those things was what you ended up saying, that’s just you being bad at elementary economics, because there is no such thing as a free lunch, regardless of what your preferred solution is. You can have inequitable outcomes today and therefore a relatively more efficient outcome tomorrow, or you can have equitable outcomes today and therefore a relatively more inefficient outcome tomorrow.
    Equity today and efficiency tomorrow is like those real estate ads offering you high returns and low risk – it doesn’t happen.
  7. Which brings us back, in a very roundabout fashion, to the point that Gulzar Natarajan was making in his post. When he says that “it is not one decision, but a series of continuing, even interminable, decisions”, I interpret it as two different but very related things.
    One, if you’ve chosen to optimize for equity today, you have to be explicit about the fact that you’ve sacrificed optimized efficiency (today and tomorrow). The worst manifestations of these sacrifices must be adjusted for at the margin. And ditto if you’ve chosen to optimize for efficiency today! You have to be explicit about the fact that you’ve sacrificed optimizing for equity (today and tomorrow). The worst manifestations of these sacrifices must be adjusted for at the margin.
    Two, no matter what your favored path is (and I envy you your conviction if you know that your path is The Best One For All, I really do), there will be errors along the way. That’s just life, there will be unexpected surprises along the way. Call it risk, or uncertainty, or whatever you like (and yes, I know, comparing the two is like comparing Knight and day) – but account for the fact that your battle plan will meet the enemy, and it will not survive.
    You must adapt, and said adaptation will involve a series of continuing, even interminable decisions.
  8. Those adaptations will land you somewhere in the middle of efficiency and equity. At which point, you can adapt your will to your circumstances and say you’ve found the truth, or you can continue with your decision-making. It is, after all, interminable.
  9. “Wait, so there’s no end to this?!”, I hear you ask in righteous indignation. “What is the eventual outcome? Or are we doomed to keep making these interminable decisions forever?”. Kids these days, I tell you. They’re just like kids in those days.

And that’s why I say what I did at the start of today’s post. The only change I’d make is the addition of one word:

The ability to exercise good judgment recursively is the binding constraint in development.

A Tweet, A Reply, And So A Blogpost

It goes without saying that I do not know enough about the details, but I certainly treat this tweet as good news. And in case you missed reading about it, there’s also this from last month.

It is remarkable how much progress we’re making in the medical field, and based on what little I understand of the developments over the last two to three years, we’re only getting started.
But it was a reply to this tweet that had my EFE antennae really and truly perk up:

There’s so much to analyse in that short little tweet!

  • Autonomous cars are coming – they’ve “been coming” for a long time, it is true. But whenever they do, y’know, actually come, will they make the world a better place or not?
  • We can (and do) worry about what impact this will have on employment, car ownership patterns, parking lots within cities and lots of other things. But what about fatalities?
  • Do I mean fatalities caused by having autonomous cars, or fatalities avoided because we have autonomous cars? Well, the net effect, of course.
  • This tweet makes the claim that fatalities will, on net, go down because of autonomous cars. Maybe you agree, maybe you don’t. But especially if you do not, I would argue that you should focus on not just newspaper reports about deaths caused by autonomous cars, but also check to see if fatality statistics drop as autonomous cars become more prevalent. This is where a carefully designed econometric analysis can be truly useful. Counterfactuals really and truly matter!
  • But let’s assume, for the moment, that fatality statistics will actually come down. If they do, surely that’s a good and wonderful thing?
  • But ah, TANSTAAFL! What this tweet is really getting at is the opportunity cost of a reduction in fatalities as a consequence of greater deployment of autonomous cars. That is, the author of the tweet assumes that fatalities will come down with autonomous cars… but then asks about some of the second order effects.
  • And one second order effect, he says, is that we simply will not have as many organs up for donation as we used to earlier. Fewer fatalities by definition means fewer deaths (which is awesome), but it also means lesser organs up for donation (which is not so awesome)
  • And so we need to get a move on in biomedical sciences, and make sure we figure out how to grow organs suitable for human transplants.
  • Have fun going further out on this limb if you are a student of economics. Imagine, for example, what a world with abundant organs for transplants might look like. Will people end up being less careful about their health? Is that a good thing or a bad thing?
  • You might be tempted to say it is a bad thing. But consider this: will not this cavalier attitude towards health lead to greater demand for better quality of transplants and at lower prices?
  • Note that I have no clue what the “correct” answer is! I’m simply trying to point out that simple applications of simple economic concepts can help you frame better and more thought-provoking questions.

Not Even Sunlight is Free

Teaching the Principles of Economics course to the first semester students at the Gokhale Institute is my absolute favorite thing to do. Both of these things matter – teaching this specific course matters, because I get to help young people realize how awesome economics can be. But teaching at the Gokhale Institute also matters, because it is the place where I myself learn economics.

We just finished the first week of teaching this past week, and in the last class I just about got enough time to introduce to students the concept of opportunity costs.

There is no such thing as a free lunch.

Read that line carefully, and tattoo it onto your brain. That’s the easy bit – tattooing it onto your brain. Applying it ceaselessly, day in and day out, to everything that you do – that’s the really, really hard bit. I wish I could say I practice what I preach, but let’s quickly move on.

So one of the keys to thinking like an economist is always remembering that everything has a cost. This may be one reason economists have fewer friends than they otherwise would. Sometimes people are very happy holding on to the naïve view that something is free. We like the idea of a bargain. We don’t want to hear about the hidden or non-obvious costs. Thinking about foregone opportunities, the choices we didn’t make, can lead to regret. Choosing this college means you can’t go to that one. Marrying this person means not marrying that one. Choosing this desert (usually) means missing out on that one. Sometimes, people just want to eat their cake and have it, too, without being reminded that they missed out on a spectacular piece of pie.
All true. But if you want to get the most out of life, you have to take account of the opportunity cost, the foregone alternatives. Better to make good choices and learn how to live with them than make bad choices in blissful ignorance that lead to ruin.

https://www.econlib.org/library/Columns/y2007/Robertsopportunitycost.html

Again, there is no such thing as a free lunch.

Part of the reason teaching this is so much fun – besides it’s innate importance – is challenging students to come up with a “free lunch”. And I enjoy extending this challenge to my students: give me an example of a free lunch!

Here’s Raghav Agarwal:

In today’s session you said nothing in life is free and there is always an opportunity cost. That got me thinking about and identifying all the costs I incur in my daily life but there is one thing I failed to identify a cost to and that is sunlight. So the purpose of me writing this mail is to understand the price we pay for sunlight (if there is any) and that’s about it.

Whatay enjoyable question! Before you read my answer, try answering it yourself. Is not sunlight free? Once you’re done, read my take:

Whatay lovely question!
1. How do we use sunlight? If it is something as simple as sitting in the sun, and basking in it’s warmth, the opportunity cost is doing something else with that time (sitting in an air-conditioned room, for example)
2. If sunlight is used to generate electricity, there will be opportunity costs in terms of the capital (the machinery) used to generate this electricity. That machinery (and the raw material used to make it) could have been put to different use.
3. If sunlight is used to make plants grow, something else could have been done with that land (construct buildings, for example).So if you ask me, there’s opportunity costs at play here too!

So if you ask me, nope, not even sunlight is free.

I hope some of you disagree with my answer. I hope some of you have other contenders for “But this is free, surely?”. Let me know, and I and my students shall ponder over your responses!

Here are other posts on EFE with the tag “opportunity costs“.