David Perell on the Microwave economy

I hope to write a longform essay myself about this topic, but this was fascinating on multiple levels:

Please read the entire thread, and the threads in almost all of the tweets that make up the first thread (if you see what I mean). Anything that can tie together a microwave dinner, urbanization, and Robert Pirsig is, as they say, self-recommending.

On Interning

It is hunt-for-an-internship season at our Institute, as I suppose is the case all over the country.

The process is trickier than usual, because of the pandemic, and for that reason I wanted to put up a small outline of my thoughts about internships.

  1. At the start of your career, optimize for learning, rather than branding. This means that in your internship, and your first job, you should optimize for firms where you are likely to learn a lot, rather than firms that are prestigious. Prestigious firms are likely to be more bureaucratic, and more about status. This means that the junior employees aren’t likely to get a lot of crucial, really important work. The pay will be better, the Friday parties will definitely be better, but the opportunity cost will be high as well.
  2. Learning how to document the work you’ve done is a very, very underrated skill, especially in internships. One way to be really and truly remembered at the end of your internship is by handing your mentor a docket of what you did, what you wish you had done, and a documentation of all the processes you learnt about.
  3. Best of all, include a section for the next intern in this team. Include stuff like who to meet in payroll, where is the best chai to be had, who in IT is especially helpful etc, along with the obvious stuff. Not only is paying it forward a good idea in and of itself, but that next intern is automatically a friend for life.
  4. Go for all the chai and sutta beaks that you are invited to, even if you don’t smoke or drink chai. Relaxed conversations with your mentors or seniors is invaluable, and soak in all the info you possibly can.
  5. Learn Excel. Here’s a laundry list to get you started: HLOOKUP, VLOOKUP, INDEX, MATCH, OFFSET, SUM, SUMPRODUCT, COUNTIF (and all the variants). Pivots, filters, data analysis add-in, solver add-in, charts, trace precedents, what-if analysis, data tables, goal seek, data validation. You must know all of this in and out, and be able to know what you can use when. YouTube videos, websites will help, but the best way to learn is to sit with a colleague and ask her to help you out. I cannot emphasize this enough – you need to know Excel. It doesn’t matter which role, which team, which department. You. Must. Know. Excel.
  6. Whatever productivity suite your organization is using, soak yourself in it. GSuite, MS Office or anything else. Know the ins and outs of the email system, the calendar tool and the internal messaging tool. Invest the time to make yourself a ninja in it. Trust me, it is worth the effort.
  7. Seek out a mentor in the organization if one isn’t allotted to you. Set up weekly lunch/tea meetings with the mentor, and have her tell you stories about stressful times in the office.
  8. Continue to learn whatever tools you got access to at the workplace. It could be Tableau, Crystal Ball, R, Jupyter notebooks or anything else. Again, soak yourself in the tool, and start on the path of becoming a ninja in it. This will take time, but it is worth your while.
  9. Learn the big picture. Ask your mentor how whatever project you are working on fits into the larger objectives of the workplace. My very first manager told me something I have never forgotten: every single thing you do in the workplace is either raising revenues for the firm, or is cutting costs for the firm, or is improving speed-to-market. If what you’re doing is achieving neither of these three, then it is a waste of time. Ask, until you are clear about the answer, how your project fits into this simple model.
  10. Lastly, about landing an internship. Do not send out blanket emails to contacts on LinkedIn, or elsewhere. Shortlist not less than ten, but not more than twenty people, and write them a personalized note. These folks should have skillsets you want to possess – it doesn’t matter where they work. The note should include a specific question about this skillset. If they answer – and to such specific notes they usually will – take their advice to heart. Incorporate it into a project you are working on. Send them this project, and ask for feedback. Then ask if they can help you land a gig. All the notes I get on LinkedIn just ask for a gig. That’s a waste of a potential networking opportunity.

Reproducibility and Replicability

I and a colleague conducted a small behavioral economics and experimental economics workshop for our students at the Gokhale Institute. It was a very small, very basic workshop, but one of the things that came up was the reproducibility problem, or as Wikipedia puts it, the replication crisis.

The replication crisis (also called the replicability crisis and the reproducibility crisis) is an ongoing methodological crisis in which it has been found that many scientific studies are difficult or impossible to replicate or reproduce. The replication crisis most severely affects the social sciences and medicine. The phrase was coined in the early 2010s as part of a growing awareness of the problem. The replication crisis represents an important body of research in the field of metascience.

https://en.wikipedia.org/wiki/Replication_crisis

And further on in that same article:

A 2016 poll of 1,500 scientists reported that 70% of them had failed to reproduce at least one other scientist’s experiment (50% had failed to reproduce one of their own experiments).[9] In 2009, 2% of scientists admitted to falsifying studies at least once and 14% admitted to personally knowing someone who did. Misconducts were reported more frequently by medical researchers than others.

https://en.wikipedia.org/wiki/Replication_crisis

The basic idea behind replicability is very simple: you should be able to take the data and the code from the paper you are reading/reviewing, and replicate the results obtained. You don’t have to agree with the choice of method, or with the results or with anything – you should be able to replicate the results, that’s all.

One basic standard of economic research is surely that someone else should be able to reproduce what you have done. They don’t have to agree with what you’ve done. They may think your data is terrible and your methodology is worse. But as a minimal standard, they should be able to reproduce your result, so that the follow-up research can then be in a position to think about what might have been done differently or better. This standard may seem obvious, but during the last 30 years or so, the methods for reproducibility have been transformed.

https://conversableeconomist.blogspot.com/2021/01/the-reproducibility-challenge-with.html

Now (to me, at any rate) this is interesting enough in and of itself, but at the risk of becoming a little meta, reading the rest of Tim Taylor’s post is worth it because it raises so many interesting issues.

The first is a link to a lovely overview of the problem by Lars Vilhuber, published in the Harvard Data Science Review. It is relatively simple to read, and is recommended reading. For example, Vilhuber draws a careful distinction between replicability and reproducibility, and is full of interesting nuggets of information. I’ll list out the major ones (major to me) here. Note that I have simply copy-pasted from the link:

  1. Publication of research articles specifically in economics can be traced back at least to the 1844 publication of the Zeitschrift für die Gesamte Staatswissenschaft (Stigler et al., 1995).
  2. As the first editor of Econometrica, Ragnar Frisch noted, “the original data will, as a rule, be published, unless their volume is excessive […] to stimulate criticism, control, and further studies” (Frisch, 1933)
  3. …only 17.4% of articles in Econometrica in 1989–1990 had empirical content (Stigler et al., 1995)
  4. As Dewald et al. (1986) note: “Many authors cited only general sources such as Survey of Current Business, Federal Reserve Bulletin, or International Financial Statistics, but did not identify the specific issues, tables, and pages from which the data had been extracted.”
  5. Among reproducibility supplements posted alongside articles in the AEA’s journals between 2010 and 2019, Stata is the most popular (72.96% of all supplements), followed by Matlab (22.45%; Vilhuber et al., 2020) (Note: Do check figure 2 at the link. Fascinating stuff.)
  6. It was concluded that “there is no tradition of replication in economics” (McCullough et al., 2006).
  7. The extent of the use of replication exercises in economics classes is anecdotally high, but I am not aware of any study or survey demonstrating this.
  8. The most famous example in economics is, of course, the exchange between Reinhart and Rogoff, and graduate student Thomas Herndon, together with professors Pollin and Ash (Herndon et al., 2014; Reinhart & Rogoff, 2010). (Note to students: this is a fascinating tale. Read up about it!)

There is much more at the link of course, but Tim Taylor’s post does a good job of extracting the key points. I’m noting them here in bullet point fashion, but you really should read the entire thing.

  1. Economic data – our understanding of the phrase needs to change, because a lot of it is in fact not publicly available today.
  2. “Vilhuber writes: “In 1960, 76% of empirical AER [American Economic Review- articles used public-use data. By 2010, 60% used administrative data, presumably none of which is public use …””
  3. Restricted Access Data Environments is a new thing that I discovered while writing this blogpost. “…where accredited researchers can get access to detailed data, but in ways that protect individual privacy. For example, there are now 30 Federal Statistical Data Research Centers around the country, mostly located close to big universities.” We could do with something like this in India. Actually, we would be a lot happier with just dbie working the way it was supposed to, but that’s for another day.
  4. Data that is given by creating a sub-sample, data that is ephemeral (try researching Instagram stories, for example) and data that you need to pay for are all challenging, and relatively recent, developments.
  5. I worked for four years in the analytics industry, so believe me when I say this. Data cleaning is a huge issue.
  6. Tim Taylor writes five paragraphs after this one, but this is a glorious para, worth quoting in full:
    “As a final thought, I’ll point out that academic researchers have mixed incentives when it comes to data. They always want access to new data, because new data is often a reliable pathway to published papers that can build a reputation and a paycheck. They often want access to the data used by rival researchers, to understand and to critique their results. But making access available to details of their own data doesn’t necessarily help them much.”

If there are those amongst you who are considering getting into academia, and are wondering what field to specialize in, reproducibility and replicability are fields worth investigating, precisely because they are relatively underrated today, and are only going to get more important tomorrow.

That’s a good investment to make, no?

A Pro-Classroom Argument

I am, if anything, against how learning is delivered today. Much lesser classroom teaching, much more discussions, much more of arguments, much more of thinking and writing (this ought to turn into a separate post!) is how I would prefer learning takes place.

But, if I had to force myself to think about what about traditional classroom teaching is good…

  1. Traditional classroom teaching, where the teacher talks and the students listen for the most part, allows for a much more systematic completion of the syllabus, and reduces the burden on the teacher. Teaching ought to become easier, and therefore (assumption alert) better.
  2. The teacher is able to focus on one particular aspect for the duration of that one class, and therefore is able to prepare accordingly. Random questions and answers, taking the class off on a tangent is all well and good, but you suffer, inevitably, a loss in depth when you go wide.
  3. A one-to-many mode of teaching ensures that all students have the same notes, and are in agreement about what was taught. Group based discussions (breakout rooms is what we call these things these days) for example, leaves students unaware of what was said in the other groups. Debriefing helps, but never completely.
  4. Do we underrate “sit still and listen” these days? Yes, long classes and having to focus on the voice that drones on is easy to make fun of, but have we collectively lost the art of sitting still and listening? Might we be inculcating the value of sustained concentration by having traditional classes, and might this in fact be a good thing?
  5. If a class is going to be about listening on a one-to-many basis, does this reduce the cognitive load on the student? Freed from all other requirements, classroom teaching might free up the student to learn more, by reducing the amount of effort demanded from her?
  6. Two points about discussions and debates. Doesn’t limiting the scope for discussions and debates in class make it better, by having only genuine doubts and disagreements being raised? Forcing students to take part in a discussion or a debate, when most of them seem to not want to, can end up making them uncomfortable. It can also be a time-consuming affair, and all for points that perhaps were not worth it. On the other hand, leaving only ten minutes or so for discussion at the end will “bubble up” only the most willing, most eager and most well-thought out responses. That is a good thing, right?
  7. Is a classroom really the best place to debate and discuss? Is not the opportunity cost of having to listen to your peers, rather than the person with the most amount of knowledge about the subject (the professor), very high? Students can (and should!) debate issues raised in class – but outside.

I find myself unable to come up with more, but I’m sure there are other arguments to be made for classroom-based, one-teacher-talks-many-students-listen-based model. What am I missing?

Back to College

I am very interested in the future of higher education.

I have learnt much more outside of the classroom than inside, and this was truest when I was a student. I want to stick around in higher education because I want to try and change this for everybody in college today.

Change it through two ways:

  1. Make classes more interesting than they were back in my day. Also make them more interesting than the typical run-of-the-mill classroom experience today. (This is a hard problem, it requires hard work and it does not scale. But learning how to teach better is an invaluable experience.)
  2. Help change college into something more than drab old sit-in-class for six hours a day, six days a week. What a horrible way to learn!

This current semester, I want to try and get as many projects off the ground as possible. This has meant getting some BSc students started on projects of their own, it has meant involving some of them in work I am currently engaged in, and it has meant trying to get some workshops going.

Some of these things will stick, and grow into something much larger than just my involvement. Others will fail. That’s ok. This semester is about trying out new things.

One of these things is a podcast.

I had tried this out in 2019 (link here), completely as a solo effort, but I got only five episodes in. 2020 is a mess I’d rather forget. And now, in 2021, we’re back with another season of Back to College.

What is Back to College?

The idea is simple: speak to people about how they would approach college differently, if they got the chance to do it all over again.

  • What would you do more of, what would you do less of?
  • What technologies that are available today would have been a blessing, and how could they also have been a curse?
  • Is bunking a science or an art? How should you choose which classes to bunk, and which to not – and why?
  • How would you have built out networks better?
  • Would you give exams the same importance with the benefit of hindsight? Why or why not?
  • Which books helped you?
  • How overrated are textbooks, or are they not? Why?
  • What in your current job are you able to do well because of what you learnt in college?
  • What in your current job makes you wish you had been taught differently in college?
  • … and the list goes on and on and on.

We’re beginning with Gokhale alumni, and we’ll add more folks in as we go along. But the idea is to build a repository of interviews for folks to listen to, any time, to get an idea about the careers they want to get into.

And this time around, it ain’t a solo effort. I have the energy of youth on my side! Praneet, Rahul, Vaishnavi, Simran, Shashank, Jay, Anshi, Nivida and Amogh are helping me out on this project, and the hope is that eventually, this will become a completely student run thing.

New episodes will be up every Friday, and we have two out already. Neha Sinha spoke with me about public policy, and Binoy Mascarenhas and I chatted about urbanization. In each case, of course, I touched upon some of the questions above. This Friday will be a conversation I had with Rohith Jyothish on understanding the ‘P’ in GIPE.

Please do give it a listen, and to all the GIPE alumni reading this, please – pretty please! – don’t hesitate to reach out if you think you would like to be on the podcast. We’ll set up a time at your convenience. (Non-GIPE folks, same offer applies to you in about a couple of months. I’ll do another post then).

Thank you, as always, for reading – and now for listening too!

Understanding fiscal deficits

Fiscal deficit is a phrase that is bandied about every year, but not very well understood – both in terms of how to arrive at it, but also in terms of what it means.

In the first part of today’s post, I’ll explain how to arrive at it. In the second part, I’ll rely on a couple of lines from an excellent article written by Rathin Roy a while ago.

I work at the Gokhale Institute of Politics and Economics, Pune. This means that while I continue to be employed at the Institute, a salary will be credited into my bank account every month. I can also choose to augment my income by, say, breaking a fixed deposit, or by taking a loan. The first part is my “recurring” income, while the latter is a one-time income.

I stay in a rented accommodation. This means I have to pay rent every month. I also have to buy groceries, pay for utilities, and pay the salaries of everybody who works at my household. But also, every now and then, I can, say, buy a car. Or a laptop. Or a house. These are not monthly expenses – at least not in my household they aren’t! The first set of expenses are “recurring” expenses, while the latter are one-time expenses.

Taken together, what matters in my household is that I must be able to arrange for ways to meet my monthly expenses. Let’s write down some very simple numbers:

  1. Assume that my recurring expenses are one lakh rupees – one hundred thousand INR. (Groceries, rent, salaries, petrol, eating out etc etc)
  2. Assume that for the month of March, my capital expenses are also one lakh rupees. (Maybe I’ve chosen to buy the latest M1 Macbook. One can dream.)
  3. So, my expenses, all told, are two lakh rupees for the month of March 2021.
  4. Assume that Gokhale pays me seventy thousand rupees as my salary. Assume that I augment this income by teaching courses in a couple of other colleges. Let’s assume that I earn one lakh rupees through this recurring income (salary plus visiting faculty income is one lakh per month)
  5. I have no other sources of income. So: 1+2 are my total expenses, against which my total income is 1 lakh rupees (4).
  6. Let’s say I am unwilling to break into any of my savings to purchase this laptop, and choose to borrow the amount instead. That is, no capital income, only borrowing.

So, in essence, the amount that I need to borrow after all possible sources of income have been thought of, in order to meet my total expenses…

That borrowing is my “fiscal deficit” for the month of March 2021.

Homework: to check if you have understood this, try reading the budget at a glance document, and see if you get how Nirmala Sitharaman and team arrived at the fiscal deficit for the government. Page 3 in the PDF.*

OK, so now we know what the fiscal deficit is, and how to go about arriving at it. But is a high fiscal deficit a good thing or a bad thing, and how does one decide?

Well, it depends on what you are borrowing for! For example, as I often say when I am talking to students, they are and should be running a fiscal deficit in their own, personal lives. They’re spending money (rent, food, movies, college fees) but not earning anything at the moment. The idea is that this money is being spent in order to acquire skills that enable them to earn much more in the future. Much more, in fact, than they spend on acquiring that education – or that, at any rate, is the plan.

But what if they instead spend an equivalent amount of money, but not to acquire an education. They spend this money, instead, on buying a Honda Gold Wing. (Yes, I know education isn’t quite that expensive just yet.)

That would be problematic, because you are taking on debt, but for acquiring a depreciating asset (a bike that gets worse over time) and not an appreciating one (your education and your years of experience get more valuable over time).

Or as Rathin Roy put it in a recent Business Standard column:

If the government is merely borrowing to fund consumption expenditure then this is difficult to justify.

https://www.business-standard.com/article/opinion/political-economy-of-fiscal-responsibility-121010701581_1.html

and a little while later, in the same piece…

For example, the “golden rule,” which states that governments must finance consumption expenditure out of revenue receipts and borrow only for investment.

https://www.business-standard.com/article/opinion/political-economy-of-fiscal-responsibility-121010701581_1.html

There is much, much more to take away from Rathin Roy’s piece, of course (and I’ll write a follow-up piece later this week) – but as a first step towards understanding fiscal deficits, this is more than enough.

*If, for whatever reason, the budget at a glance document is not clear, let me know in the comments below. If more than ten people are interested, happy to arrange a quick video call about it (because, you know, there have been so few of ’em this past year!)

Scrambled Eggs on a Sunday

And which is why I scheduled this for 7.30 am, rather than the usual 10 am slot.

I discovered this channel via (who else) Krish Ashok

Cesar Hidalgo’s Viral Thread

You might first want to learn about who Cesar Hidalgo is…

… and then go through this thread.

Update: I scheduled this post a couple of days ago, but then came across this excellent Twitter thread by Rathin Roy, which is also worth reading:

Three Excellent Games to Play (Econ related)

We’re running a small workshop on experimental and behavioral economics at the Gokhale Institute, and we had great fun playing these games yesterday.

  1. Bad News: a game that teaches you how (worryingly) easy it is to get addicted to generating, sharing and amplifying bad news. Yes, I know this isn’t news, exactly – but playing this game allows us to actively participate in the process. And like I said, what is worrying is the ease, and the addiction. Give it a shot, doesn’t take more than 10 minutes.
  2. Go Viral!: In similar vein, but more topical.

Both of these I found out about via Behavioral Scientist.

And the excellent, excellent, kiviq.us. I’ll be using this again and again in the years to come – a very simple, very hands-on way to help students understand double oral auctions.

Have fun – and please reach out if you need help running any of these games. I’d love to help out, if I can in any way.

Airbnb and the Asymmetry of Information

Devon Zuegel (@devonzuegel on Twitter, and definitely worth following) was less than happy with Airbnb recently:

And so of course I thought about Akerlof (1970)

This paper relates quality and uncertainty. The existence of goods of many grades poses interesting and important problems for
the theory of markets.

Akerlof, G. (1970). The Market for “Lemons”: Quality Uncertainty and the Market Mechanism. The Quarterly Journal of Economics, 84(3), 488-500

It’s a paper that every undergraduate student ought to read. Not just economics undergraduate student, mind you, but every undergraduate student. Because it helps you get an understanding of many modern businesses today.

But first, a relatively simple explanation of the core idea of the paper:

Suppose buyers cannot distinguish between a high-quality car (a “peach”) and a “lemon”. Then they are only willing to pay a fixed price for a car that averages the value of a “peach” and “lemon” together (pavg). But sellers know whether they hold a peach or a lemon. Given the fixed price at which buyers will buy, sellers will sell only when they hold “lemons” (since plemon < pavg) and they will leave the market when they hold “peaches” (since ppeach > pavg). Eventually, as enough sellers of “peaches” leave the market, the average willingness-to-pay of buyers will decrease (since the average quality of cars on the market decreased), leading to even more sellers of high-quality cars to leave the market through a positive feedback loop.

Thus the uninformed buyer’s price creates an adverse selection problem that drives the high-quality cars from the market. Adverse selection is a market mechanism that can lead to a market collapse.

Akerlof’s paper shows how prices can determine the quality of goods traded on the market. Low prices drive away sellers of high-quality goods, leaving only lemons behind. In 2001, Akerlof, along with Michael Spence, and Joseph Stiglitz, jointly received the Nobel Memorial Prize in Economic Sciences, for their research on issues related to asymmetric information.

https://en.wikipedia.org/wiki/The_Market_for_Lemons#

Now, one way to understand the value of many businesses today is to realize that they’re solving asymmetry of information problems. Or at least, that’s how I think of it when I end up looking up the rating for a restaurant on Zomato in a unfamiliar part of town. I don’t know enough about this part of town, and I certainly don’t know this restaurant. Should I walk in for a meal or not?

I could always check if the people already inside are smiling or not, of course, but let’s face it, most of us will simply Zomato our way through this problem. Zomato is reducing the asymmetry of information problem. Successfully or not is a matter of opinion and perhaps controversy. But my argument here is that this is a potentially useful way of thinking about the problem: how to decide where to eat?

How to decide whom to recruit? Linkedin.

How to decide whom to trust? Look ’em up on Facebook, or Twitter, or Instagram, or wherever it is that people look up people these days.

How to decide which product to buy on Amazon? Check out the user ratings. In fact, sort by average user ratings! Yes, Amazon does provide this option.

How to decide which book to read? Goodreads.

How to… you get the drift, right. Part of the reason these firms are so highly valued by the public is because they solve the asymmetry of information problem.

And so does Airbnb. Or does it?

And that brings us back to Devon Zuegel’s tweet.

Every review left on Airbnb informs potential users about the quality of a stay at a particular host’s place. The more information they are able to glean from reviews left by previous users, the more they are likely to definitively transact…or not. That is, potential users will either stay at a particular place, or will definitely not.

Since Airbnb gets a cut from each transaction, but not from each no-stay, they have an incentive to put up only positive reviews. And that is the problem that we have to think about when we read Devon Zuegel’s tweets. Is Airbnb incentivized to leave only positive reviews up? Short answer: yes. Therefore, will they leave only positive reviews up? I’d say it’s a question of horizons, but it is also a question of the calculus.

Airbnb will not last for very long if they pull down every single negative review, because that will destroy trust.

But:

  • every now and then…
  • particularly for really highly rated hosts…
  • especially during a pandemic…
  • will the odd negative review…
  • have a higher chance of being pulled down?

Nothing in life is ever black and white, and the truth lies somewhere in the middle. So no, Airbnb will not pull down every single negative review, but we also shouldn’t assume that it will leave every single negative review up.

More information in the hands of the consumer is a wonderful thing, and it does reduce the asymmetry of information. But who is providing the information to the consumers, and what are their incentives? What if the providers of the good/service are the ones that are making information available to the eventual consumers? Will that need to be regulated, and if so, how?

Zomato, LinkedIn, Uber, Airbnb – it’s a great time to be alive, because these firms, and many others like them, have provided for many services that would simply have not been possible otherwise. They have successfully reduced the asymmetry of information problem. But it’s not the end of the asymmetry of information problem, not just yet.

If anything, it just got more interesting.