What’s more intense than leading the IMF during a financial crisis? For Simon Johnson, it was co-authoring a book with fellow economist (and past guest) Daron Acemoglu. Written in six months, their book Power and Progress: Our Thousand-Year Struggle Over Technology and Prosperity, argues that widespread prosperity is not the natural consequence of technological progress, but instead only happens when there is a conscious effort to bend the direction and gains from technological advances away from the elite.
Tyler and Simon discuss the ideas in the book and on Simon’s earlier work on finance and banking, including at what size a US bank is small enough to fail, the future of deposit insurance, when we’ll see a central bank digital currency, his top proposal for reforming the IMF, how quickly the Industrial Revolution led to widespread prosperity, whether AI will boost wages, how he changed his mind on the Middle Ages, the key difference in outlook between him and Daron, how he thinks institutions affect growth, how to fix northern England’s economic climate, whether the UK should join NAFTA, improving science policy, the Simon Johnson production function, whether MBAs are overrated, the importance of communication, and more.
Watch the full conversation
Subscribe on Apple Podcasts, Spotify, or your favorite podcast app to be notified when a new episode releases.
Recorded March 21st, 2023
Read the full transcript
TYLER COWEN: Hello, everyone, and welcome back to Conversations with Tyler. Today, I am chatting with Simon Johnson. He’s a British American economist and a professor at the MIT Sloan School of Management. He served as chief economist of the IMF. He’s worked on Russian reforms, and he’s the author of numerous books, but most prominently, his new book, co-authored with Daron Acemoglu, called Power and Progress: Our Thousand-Year Struggle over Technology and Prosperity.
We’ll be getting to that book, but since we’re taping on March 21, I’d like to start with a few questions about banking and deposit insurance for Simon, who has written in a book in that area as well. Simon, hello.
SIMON JOHNSON: Hi, Tyler. Great to see you, and great to see you today. These are good Tyler topics, I would say.
COWEN: Yes, so a very simple question: At what size or level is a US bank small enough to fail?
JOHNSON: [laughs] I think we’re in the process of exploring that, Tyler, because previously . . . I actually testified to the Senate in 2015. There was a big argument over many years about this. I said, “If you let banks go over $50 billion, you’re going to have to pay attention to them because there could be systemic knock-on effects.” I talked about Long-Term Capital Management, which wasn’t a bank, which was about $100 billion.
We talked about other episodes in 2008, but I guess the consensus shifted. The view coming out of the reforms or changes of 2018 was that $250 billion total assets was where systemic attention should start to be paid. Silicon Valley Bank was smaller than that; Signature Bank was significantly smaller than that. So, it’s a very good question, Tyler, whether there’s a minimum size. We can talk about that knock-on effects. What does contagion mean in this context? But it’s definitely worrying when you think about what has been done and for whom over the past two weeks.
COWEN: Many of us used to have this notion that if we only split up the banks, made them smaller, stopped them from becoming large, they would be in a position where they’re allowed to fail. It now seems there’s no size where we can let that happen. So, should we give up on the idea of splitting up the banks?
JOHNSON: No, actually, I think Sheila Bair nailed this. She nailed it in the Silicon Valley Bank weekend, if you like. And she wrote about it last week in the Financial Times, which is, what you could have done was have the Silicon Valley Bank uninsured deposits take a haircut. The assets of that bank were mostly high-quality, long-term government bonds that were underwater, but only because of the interest rate increase.
They probably would’ve got about 90 cents on the dollar by reasonable estimates, Tyler, and the FDIC also pays out about half of what they owe in that recovery process in cash after just a few days. So, they would’ve had cash. They would’ve had a haircut.
But then, Tyler, I think the key point that Sheila makes is, you should have insured all the other banks so that, in other words, uninsured deposits at Silicon Valley Bank would’ve had losses. That would’ve addressed some of the moral hazard issues that we’re now very worried about, and all the other banks would’ve been insured, so you wouldn’t get the contagion.
COWEN: Insured above $250,000, you mean?
JOHNSON: Exactly. Sorry. Actually, the Sheila Bair proposal was to insure all uninsured deposits. We could talk about that. I think a very sensible version of that would insure business transaction accounts, or operational deposits, as FDIC calls them. Basically, small-business working capital, payroll, the account from which they make payroll, and so on. I think those are a real issue at the moment. I think insuring all deposits is something you only do when you’re absolutely desperate. Hopefully, we won’t get to that.
COWEN: It seems to be that’s the world moving forward. If we insure all deposits but SVB, then they’re all insured. Then those banks will take a lot more risk, and the problem gets bigger, and you raise deposit insurance premia. More funds move outside the banking system to, say, money market funds, which are less regulated. Doesn’t it just postpone the problem and make it bigger?
JOHNSON: It may do. I don’t think we’re going to get a consensus to insure all deposits, Tyler, for exactly the reasons you’re flagging. I do think that we might leave individual deposits at $250,000, where it is now. And we might have a sensible design of an insurance scheme for small-business transaction accounts, because I think if you have a nine-person startup, telling them to spend time on financial management when they’re trying to build a new hot sauce company — whatever it is — it’s a distraction and rather unfair.
The FDIC already has this category called transaction deposits, operational deposits. They could use that, and we can carve that out. And then, yes, you want to avoid a situation where hedge funds put money on deposit with crazy banks that go off and take crypto risk or other risks that are not well managed.
COWEN: Let me tell you my worry and see what your response is: that once you start playing the credibility game, which ultimately you can’t avoid, that even letting a truly small bank — much smaller than SVB — fail, it sends a signal. And then, since about half of the deposits in US banks are not FDIC-insured, it doesn’t matter how small the bank is; the signal is, in a sense, infinitely large once it happens once. Is that a problem? If so, how do we get around it?
JOHNSON: It’s a very big problem, Tyler, honestly. Uninsured deposits were about $8 trillion at the end of last year; insured was about $10 trillion. Like you say, 50–50, and it’s all about the signal because if you believe that uninsured deposits are in jeopardy, which was the case during that Silicon Valley Bank weekend, then you may be looking for other places to put your money.
I think that that shift in deposits away from — if there is a shift in deposits away from, let’s say, regional banks or mid-size banks towards very large banks, for example, they’re not going to be particularly good at replacing the regional banks in what they do in terms of lending to the nonfinancial sector all across the country, for example. If the shift, by the way, is down to community banks, that’ll be interesting. They might actually be able to step up and fill some of the gaps, but I still think that’s going to be pretty disruptive in a lot of markets.
COWEN: Let’s say you had been in charge with Credit Suisse. What would you have done in advance to have eased the path of the mess we’re now in? Because they’ve been poorly managed for a long time, right?
JOHNSON: [laughs] Yes, absolutely, and they’ve had many cautions from the regulators. I think that the losses that have been imposed on their convertible bonds, the so-called CoCos or AT1 — additional tier 1 — bonds . . . we’re going to see what happens. But I think if that works, and that sends a signal that these are actually going to be treated like equity — or more than equity, Tyler, because the equity holders got some compensation in the purchase. They got $3 billion for a bank that was worth $8 billion on the Friday before this happened, but the AT1 bonds were taken to zero. That’s the current position.
COWEN: Sure, but isn’t that a huge problem? We’re signaling the CoCo bonds don’t make any sense, you ought to just buy the equity. Haven’t we abolished CoCo bonds as an institution moving forward, at least in that context?
JOHNSON: I think there’s going to be a lot of repricing in that market for that reason. But I think the concern, Tyler, was that the convertibility would not happen, that those bonds would be protected precisely because of the kind of contagion fears that you mentioned a few moments ago.
At least the authorities felt confident enough to do that, and that presumably means, hopefully, they know who owned those bonds, and they were not owned by highly leveraged entities that could themselves be in trouble. So far — it’s less than 48 hours after that happened — so far, so good, but let’s see exactly what the linkages are between various financial-sector firms in Europe.
COWEN: What’s the likelihood and time frame for a dollar central bank digitally managed currency? Is it ever going to happen?
JOHNSON: I think it’s pretty far off, Tyler. Look, you mentioned money market funds a moment ago. One thing that could happen, of course, is people might decide to take their uninsured deposits, which are a claim, a liability of a bank, and you may have different views about the sustainability or vulnerability of that bank, and you could buy short-term government debt through a money market fund, and then you could just sit there and wait to see what happens.
If you had a central bank digital currency, I think a lot of people would’ve migrated, if they weren’t already holding that — they would’ve migrated from the bank account into that central bank digital currency, and that would be a big shift in the ability of the banks to fund their loan book. That could have serious and difficult ramifications in a period of stress like this. I think the complications of a central bank digital currency are more apparent this week than they were a couple weeks ago.
COWEN: So, we’re just going to live with privately issued stablecoins, and that will be what we use? Or do we somehow abolish all attempts to make money programmable?
JOHNSON: That’s the right question, Tyler. I think these privately issued stablecoins are pretty unstable, personally, and there’ve been a number of pressures there and a number of disappointments, as you know. I think they’re not well regulated, and the Systemic Risk Council, of which I’m co-chair now with Erkki Liikanen from Finland — we’ve sent some letters really cautioning about that. That’s the main part of the crypto space that we think has systemic implications.
But you’re right. Could we get programmable money? Should we want programmable money? What could that look like? That’s the right question to be asking. Perhaps there are other, better protocols that could be developed, given what we’re currently experiencing.
COWEN: Regulating who or what is a bank — should that be expanded?
JOHNSON: It’s already pretty strenuous, the regulation of that, and I think supervision —
COWEN: But who counts as a bank? Money market funds don’t count as a bank in the same way. Crypto exchanges don’t count as clearinghouses in the same way. Financial regulation — should we extend it to more entities?
JOHNSON: I think Gary Gensler — when it comes to crypto exchanges, I think he’s got the right view, which is, if it walks like an exchange and talks like an exchange, it’s an exchange. Doesn’t matter what you call it. I think there’s a pretty clear definition also of what’s a security, which Gary is following up on.
I think the bank-nonbank distinction is a really interesting one, Tyler, because of course, it all goes back to allowing money market funds to appear in the 1970s, which, in itself, was a reaction to high inflation and controls and interest rates.
Once that genie has been let out of the bottle, it’s very hard to put it back in. I would not run around handing out a lot more banking licenses to things that are not banks, but I’d also be very careful and cautious of anything that provides bank-like services without a banking license, and that’s exactly what stablecoins are.
COWEN: The American separation of banking and commerce — does it still make sense? Why don’t we just abolish it? Plenty of countries don’t have it. Maybe their banks have been more stable than ours. Why have that form of the Glass-Steagall Act?
JOHNSON: That’s a great question. I think, in America, one of the things we’re really good at is allowing people to get really big and powerful fast. If Amazon owned a bank, or if Microsoft owned a bank, or if Apple owned a bank, would the system be more stable or less stable? Would it be more fair or less fair? It’s a very good question. I think we’re not going to do it politically. I think that separation is something that’s sufficiently ingrained, and people are accustomed to it, but you’re right to push them, Tyler. Absolutely.
COWEN: Universal banking, as we’ve had in Japan, Germany — I remember the 1980s, parts of the ’90s. People said how great it is. Were they wrong in overrating universal banking? Those nations have not done perfectly since then, right?
JOHNSON: Yes, exactly. The Japanese banks ran into big trouble end of the ’80s and the ’90s, and subsequently have been pretty rough, and the European banks are also regarded as an imperfect model. Deutsche Bank, for example, is often held out as being very problematic, although it’s backed by the German state, so they can get away with a lot of things. These Swiss banks that were struggling this week had universal bank features at some points in their history. I think banking is a difficult business and rife with problems for the rest of us, Tyler, however you organize it.
On
COWEN: Should very small countries be allowed to have such large banks? If you look at Credit Suisse liabilities relative to Swiss GDP, I’m not sure what the number is now, but those two things are not so far apart, right?
JOHNSON: Yes.
COWEN: So, it’s an issue: Can Switzerland even bail out its own very large banks credibly?
JOHNSON: Credit Suisse, when it was in serious trouble, had a balance sheet of around half a trillion dollars. The Swiss National Bank has foreign exchange assets of over or around about a trillion dollars. Yes, size matters, Tyler, but size relative to the ability of the government to stand behind its banks, including in foreign currency when they’re operating not in the domestic currency. I think, on that score, the Swiss did have that capacity.
But that’s not always the case. Iceland, for example — when Iceland had big banks that failed in 2008, Iceland could not stand behind their banks, and that probably should not be allowed.
COWEN: Well, Switzerland now has one really big bank, bigger than it had been a week ago. Should the Swiss be more worried? Should they split it up? What’s the equilibrium here? It’s moved against the direction you want.
JOHNSON: Right, absolutely. I think they should be very careful and supervise that bank extremely carefully. Of course, part of why UBS is so big — it has very large wealth management, and if wealth management is run properly, it’s basically a custodian with fees for rich people. When you strip that out, I think their total balance sheet is a bit more than the reserves that the Swiss National Bank has, but they still have strong enough national balance sheet and a credible enough central bank.
Let’s not forget the swap lines with the Fed, Tyler — really important in this context to be able to support banking systems. Not that many people have open unlimited swap lines so they can access double liquidity whenever they want, but the Swiss do.
COWEN: Sure, but if you believe in splitting up the big banks, isn’t that the first place you should look, and the Swiss, say, two months from now, should turn around and split up UBS into three, four, five different parts? Or do we give up on the idea? What do you consider? What’s your view?
JOHNSON: Well, I’ve always said, Tyler, that a strength of the US system from its financial history is that we’ve had a multitude of banks, and we’re always very suspicious of big banks. The Europeans went for the big banking model, universal banking, a long time ago — late 19th century, actually. So, there was a divergence there. The US economy, US innovation, US productivity was not built on big banks. It was not associated with big banks.
Large banks or banks that are large relative to our GDP are something that’s developed much more recently, since the 1980s. And even JPMorgan Chase, which is about a $3.5 trillion bank, is not big relative to our GDP like Deutsche or UBS’s in Europe. I remain very concerned about the concentration of economic and political power in the hands of those very large banks, but I’ve always said and argued that that’s not the only problem to worry about. You also have to worry about weaknesses and contagion in other parts of the system.
COWEN: Given what a comeback housing prices have made in many countries around the world, should we not think that 2006–2007 wasn’t really much of a housing bubble? Prices were a little bit ahead of their time, but the actual crash was a kind of negative bubble in the shadow banking market,, and home prices then were close to rational — right or wrong?
JOHNSON: I think there’s a lot of validity to that. It was obviously also a derivative — I don’t know if bubble is quite the right word — but there was a view — and I was at the IMF, as you know, Tyler, in 2007 — and we had a view that was completely wrong, that derivatives had spread the risk around the world. In fact, they concentrated the risk, and it was that concentration of risk in these very large financial institutions in the US and in Europe that caused a lot of the panic, a lot of the problems, and that generated downward pressure on house prices, the negative to which you’re referring.
Certainly, in some markets, maybe many markets, house prices did bounce back, but unfortunately, not before millions of people lost their homes. That was a really tragic set of events that we need to try to avoid in the future.
COWEN: Speaking of the IMF, today, how would you reform the IMF?
JOHNSON: I would remove the Russians from the board of directors. I think that having the Russians there in that role is inappropriate, problematic. I think the way that Russia has acted over the past year, and arguably longer, is outside what’s acceptable for nations. I’m very sad that this has not happened.
I think when autocrats behave badly and aggressively and invade other countries and otherwise mess with the world and the world economy, you should not have them at the big table. If they want to behave better, sure, let them back in, but I don’t think the Russians belong at the IMF right now.
COWEN: I’m fine with that, but how would you improve the positive program for what the fund does?
JOHNSON: I think that the fund has long been primarily engaged with emerging markets and developing countries. I understand there’s a rhetoric of evenhandedness. There’s a rhetoric of the IMF should speak truth to authority in all countries. I think they struggle with that, just given the power structure of the world. So, their ability to lend, when necessary, to help people who are genuinely in trouble, but not to get sucked into situations where deeper reforms are needed, and reforms perhaps the IMF can’t help deliver on.
I don’t have easy answers, Tyler. Luckily, it’s not my job anymore. They work hard on it. They grapple with it, but I don’t know that there are imminent solutions.
COWEN: We have Pakistan entering, what, its 23rd or 24th IMF program. It surely seems something about the system is broken, right? Should there be a way we can just kick countries out and say, “No more”?
JOHNSON: Well, you can. Countries can be —
COWEN: But we don’t very much.
JOHNSON: Well, certain countries don’t get excluded, countries that have, let’s say, some geopolitical significance and a good enough relationship with the G7. Sure, maybe there should be a special recovery arm of something, but it’s very hard to create these new institutions. We haven’t really done it or done it well since 1944. So, there isn’t that much other than the IMF. I agree, Tyler, it’s extremely awkward and not clear that it’s good for the people of Pakistan to continue on this sort of trajectory.
COWEN: Here’s a reader question, and I’m quoting: “Ask him if Indonesia’s money-financed fiscal program in which their central bank monetized national debt has worked. The nation’s currency has modestly appreciated. The program seemed to get them through the pandemic without ill effects.” Opinion?
JOHNSON: Well, I’m not an expert on Indonesia, but I do think they’ve come a long way in the past 20 years in terms of responsible macro management. Of course, if you have a good reputation for keeping your budget under control . . . The joke of the IMF and elsewhere is that IMF stands for It’s Mostly Fiscal. If you’re credible on the fiscal side, I think you do create space, Tyler, for extraordinary operations at a time of crisis. COVID was a time of crisis, a pretty severe crisis in some places.
Let’s hope that the Indonesians can keep that going. I think you’ve got to be very careful when it comes to money-financed deficits anywhere, and use those kinds of measures only when there’s no alternative.
COWEN: Okay. Now turning to your new book with Daron Acemoglu, Power and Progress. Could you please summarize that book for us as a high-quality GPT-4 program might? That’s my prompt.
JOHNSON: [laughs] Okay, I haven’t had a chance to ask GPT-4 that, but it’s a great question. We are challenging, I think, some modern techno optimism view, which happens to be quite prevalent, Tyler, which is that technology just kind of happens. It raises productivity. It improves how people live, including their health, and it provides them with more opportunity, and everyone benefits eventually. Well, when we look back over the past 1,000 — and in some ways, we can look back 10,000 years — we find it to be more complicated.
Sometimes, technology is massively beneficial, absolutely. The Industrial Revolution did work out well for many people eventually, but eventually took more than a hundred years. I think getting better outcomes for more people sooner — including from, for example, artificial intelligence right now, Tyler — is something we need to put our minds to.
We have a number of recommendations about making technology more people-centric and more about generating new tasks, helping people become more productive and creative, rather than displacing workers and pushing them out, just using automation to displace workers. That is not the way we recommend going forward.
COWEN: Say we take the Industrial Revolution. There’s a new paper by Joel Mokyr in the Journal of Political Economy, and he argues in the areas where you had an Industrial Revolution, wages went up a fair amount pretty rapidly. Performance in England, as a whole, was mixed, but it was the areas that didn’t have the IR where wage performance was poor. Doesn’t it typically happen pretty rapidly that if you have sustained advances in technology, most people are much better off?
JOHNSON: We did talk to Joel a lot about that in writing the book, and he was immensely helpful. We used his work a lot. I think the distinct difference we would make, just in terms of those facts, is that if you look at how people lived in cities — for example, in Manchester in the 1830s, which Engels wrote about — whether or not you like Marx and Engels, that was a really good, powerful description of working-class conditions.
It was bad, Tyler, and if you look at the conditions of workers and children working in coal mines in the 1840s, which was subject to a big investigation in the UK, it was also absolutely terrible and much worse than the conditions for children before the Industrial Revolution.
Sure, Joel is right that some people definitely had some gains in some areas. But I don’t think that the living standards, taken in any modern sense, of people in and around the textile factories of Manchester or the coal mines of Northumbria or the town I’m from, Sheffield, which was steel that was just starting to emerge at that time — I don’t think people really saw much by way of gains until after the 1850s.
Our view is, it took a hundred years for this really to pay off. Then it does pay off. Joel’s right about the importance of entrepreneurs tinkering with technology — love that take on the driving force, but that wasn’t enough to generate shared prosperity. It took a bit more than that.
COWEN: As we know, until fairly recently, there have been significant increases in wealth inequality in many Western nations. But don’t they coincide with a period of relatively low TFP, not relatively high? Productivity growth is pretty slow since 1973. Income inequality goes up. But the story you’re trying to tell in the book is, “Oh, you have a lot of tech advances, and then income inequality goes up.” But we’ve been seeing almost the opposite of that. The problem is not enough tech advances, so wages are somewhat stagnant until lately.
JOHNSON: We would like more tech advances and more productivity growth, to be clear. We’re not anti-tech at all. And you’re quite right about the coincidence of what’s happened since the 1970s. What we say is that a lot of the digital technology, for example, Tyler, was quite disappointing in terms of productivity effects, but it was nevertheless deployed because management thought it would be helpful to displace workers.
Daron Acemoglu, my co-author, and Pascual Restrepo coined the term so-so technology, where you automate even though it doesn’t boost productivity — marginal worker productivity and, therefore, wages — but you do displace workers. And their favorite example is self-checkout kiosks at supermarkets, for example, where you shift the work onto the consumers. You’re not making the workers more productive. You don’t see increases in the wages in supermarkets where they adopt self-checkout kiosks.
That, more broadly, lines up with what you just described in terms of the macro phenomenon, which is technology change doesn’t become productivity growth, and it’s also consistent with widening inequality.
COWEN: What do you think of the claims of your colleague, David Autor, that now we’re entering an era of wage compression, say, the last four or five years?
JOHNSON: I hope he’s right. David, Daron, and I run a little center at MIT for the study of the future of work, so we talk about this a lot. David is always really sharp, on the money, and into the latest data. I think he would say — he does say — that this, after coming out of the pandemic — it’s not clear if this is a trend or a blip, but I think some wage compression would be very helpful at this point, Tyler.
COWEN: If we look at wealth, if we look at the very wealthiest people in this country — Elon Musk, Mark Zuckerberg — it seems they’re much less wealthy than they were two or three years ago. More than cut in half. Maybe they’ve lost three-quarters of their wealth. If market returns are a random walk, that’s more or less our best forecast for the future. So, hasn’t this problem largely fixed itself in the last few years? Top wealth has really been cut down a lot.
JOHNSON: I’ve never been bothered by the top wealth, Tyler. It’s never been a major concern of mine. What bothers me more is lack of good wages, rising wages for people who haven’t completed college, let’s say, or didn’t go to college. I think it’s the stagnation of that since the 1970s, which was quite different, of course, from what happened for the 30 or 40 years before that. I think that’s unnecessary and we can do better, but it’s a very hard process to reverse.
It’s not a new process, and it’s a process very rooted in the fact that in early Industrial Revolution, people like Henry Ford — when they brought in automation, they also created a lot of new tasks for workers that didn’t previously exist. Workers even without a lot of skill became much more productive. That productivity was shared with them through higher wages, through unions and other things.
That whole mechanism has broken down more recently. I think with AI, there’s a lot of obviously huge discussion about all kinds of dimensions, but the piece that we really focus on and are concerned about is that more workers may be displaced from previously well-paying jobs, and without creating new tasks. Therefore, worker productivity doesn’t go up. You may pay some people — maybe the designers of AI algorithms — more money, but most people may well be paid less in real terms.
COWEN: You’ve probably seen estimates that ChatGPT is the most rapidly adopted computer application in all of human history compared to any technology. Now, if that’s the case, why is it that Open AI has only a market valuation of about $30 billion? Which is high, but it’s not close to Apple. In a way, it’s remarkably low. Doesn’t that suggest most of the gains from AI will go to other parties and not to the people who own and design the algorithms?
JOHNSON: Well, that’s a great question, Tyler. I suppose we should go back and look at when exactly people started to think IBM was so valuable, Microsoft was so valuable, Google was so valuable. There is certainly a lot of discussion about the value chain of AI and whether the place to be is some more fundamental generative algorithm, or whether it’s going to be in the applications, the downstream part of AI.
I think it’s all up for grabs, personally. I think it’s also correct that we many times — it’s a cliché, but a correct cliché — that we overestimate the immediate impact of technology and underestimate the longer-term impact of technology. Presumably the stock price would reflect people’s views of the longer-term impact.
COWEN: Shares of Nvidia — they’re definitely up, but they’re not up by so much. It’s not become one of the world’s largest companies. There seems to be more and more evidence that AI can be commodified. You have a lot of competitors. There’s talk of training systems on a single GPU [graphics processing unit]. So, doesn’t AI mean we’ve entered this new age again, where the gains mostly don’t go to corporations?
JOHNSON: Well, possibly. That, I think, could be quite positive. Of course, we thought that about crypto, as well, at the beginning. And there were many discussions I recall — I’m sure you recall — around the beginning of the internet, where there were thoughts that, oh, this could become a more distributive model. Maybe power seeps away from large corporations. I think we tend to recombine things in this country in particular, Tyler. Again, we get big quickly, and we let people get powerful quickly.
I think in the case of social media, we regretted that. Look, I think it’s very hard to stand in the way of these powerful social processes like the adoption of technology like GPT. We can maybe find some policy adjustments along the way. I think we can certainly push people to find ways to use the technology, the ways to use the technology that would enhance creativity, enhance productivity in a distributive way so all the gains don’t go to corporations, but no guarantee at all we’re going to get that outcome.
COWEN: The people who said all along that social media are more competitive than they look — haven’t they been proven right? TikTok and Facebook, Google competing with Bing, Amazon laying people off repeatedly. Those companies have lost their luster. They don’t look like unassailable giants anymore, to say the least. Meta shares were down, what, three-quarters in value not too long ago?
JOHNSON: Well, sure. There’s definitely competition in that sector, and there’s definitely been overexpansion and overexuberance. I think the regret actually there, Tyler, is much more about the impact of social media on political, economic, and social discourse, the way that people talk to each other, the way — what has happened to civility. Of course, there’s lots of concern, which is still concern — I’m not saying this has been completely established — about the impact of social media on young people, young people’s minds, young people’s mental wellness, and so on.
I think some of those social media companies could have been, and should still be, much more responsible and much more careful. Perhaps it’s just a natural learning process. Perhaps we have to go through this and figure out how to be more responsible. I think there are some encouraging uses of technology out there. That’s one thing we emphasize in our book, right, Tyler? It’s all about choices.
Choices that we make, choices that we make socially, and choices that you and I can argue about or discuss on this kind of show that can impact what people develop and how they use it, in addition to whatever levers the government may find available that, perhaps, could be pulled or not pulled.
COWEN: From my point of view, it seems that you and Daron overestimate how much market concentration matters. There’s one passage you have where you’re criticizing the soft drink market for being highly concentrated. Coca-Cola dominates and so on. But I used Google, I used GPT, and everything I could find tells me the price of soft drinks, in real terms, has been falling.
There’s also a lot more choice. You go to an airport, you see this long array of hideously colored things that I don’t even want to touch, but they’re not, in general, put out by Coca-Cola. Variety is going up, and if real price is falling compared to inflation, why be upset about concentration in the soft-drink market?
JOHNSON: Oh, I think that was just an attempt to illustrate what market concentration means, Tyler. Perhaps I’ll go back and look and see if we need to insert some errata on that.
In general, we are just trying to get across the point that if concentration prevents innovation, if concentration prevents development of varieties, if concentration keeps prices high — and of course, there’s a long history of concentration in various sectors where some aspects of that is true — then it’s problematic. If it’s fully satisfying the consumers, and if you have enough progress innovation in a particular sector, then yes, sure, there’s less to object to.
COWEN: What do you think is the main thing you changed your mind on over the course of writing this book?
JOHNSON: The Middle Ages.
COWEN: How did you change?
JOHNSON: Well, I bought into this idea that you get from school books and so on about there was a dark age of very little innovation between the fall of Rome and the emergence of the Renaissance. Turns out, when you look more closely, there was a huge amount of innovation in agriculture, in commerce, even in early industry, but what didn’t happen — it didn’t filter down to regular people.
There was an increase in wealth or money that came to the church and to the state, for example, in medieval England and other parts of Western Europe, but a lot of that was funneled into these massive cathedrals that you see.
I actually went in and dug up some of the economics of cathedrals and cathedral building, which is fascinating. I never paid attention to that. And you could see — because these were simple economies — how the money was squeezed out of peasants. This was not free labor, obviously. They were coerced or pressed and restricted in their choices of employer. They became more productive, but they were also squeezed harder. The Normans were very good at squeezing the English, for example.
Those resources came to that higher, let’s say, 5 percent or 3 percent of society, and they spent it building cathedrals. Next time you see a cathedral, next time anyone sees a cathedral, think about them as symbols of medieval despair and increasing productivity, but the failure to create shared prosperity in a very particular labor system that was with us not too long ago.
COWEN: But wasn’t your original view correct for at least 400 years, say, centuries six through nine? And the revival you’re talking about — maybe that’s 10th century, and then for England, the Normans in 1066. But if the view you held earlier is correct for 400 to 500 years, it’s not that bad a view, right?
JOHNSON: Good point. I think that it was certainly right for several hundred years. Whether it started in 800 or 900 — I think you can debate that. The record’s not very good. The general idea that was put around by Renaissance scholars like Petrarch was that nothing good had happened from the fall of Rome for 900 years, and that turns out not to be true.
COWEN: If you think about how you read your own book, and how Daron reads the book — I’m not saying the two of you disagree — but if you had to explain the difference in emphasis, how each of you views the book, how would you present that to your potential readers?
JOHNSON: Well, I think Daron spends a lot of time thinking about the future of AI. He’s very deeply in discussions with that community. He’s always looking down the road, and Daron is very good at seeing what’s on the horizon, analyzing it, and then putting a mental structure around it.
I think I tend to look back at history a lot more, Tyler. I’ve read a lot for the book, but I’m reading more about Henry Ford. I’m reading more about the use of technology in World War II. I like to go back and read some of those medieval-type sources again and think about how that played out. I found the book has helped me understand a lot of things, but I tend to look at the historical experiences and try to drill down into those. Daron’s a very forward-looking person.
COWEN: Working with Daron — how would you describe what that’s like? Obviously, he’s super productive, as are you, but he might be renowned as the very most productive person in the entire profession. How does that manifest itself through daily interactions?
JOHNSON: It’s hard to keep up with Daron, to be totally honest. We talked about writing this book, and we kicked ideas around for about five years. And then all of a sudden — it was a pre-planned moment, but all of a sudden, pretty much January last year, Daron said, “Right, let’s start writing.” We’d got some pieces already, but we worked flat out for six months, and that was super intense.
Then we spent a lot of time doing fact checking and cleaning things up and tightening text. It was probably the most intense six months of my working life, I would say. That’s more intense than working at the IMF during the financial crisis, so that’s quite a benchmark.
COWEN: Have you and he been surprised at how rapidly AI has evolved? Large language models in particular.
JOHNSON: I have been surprised. Daron has not, again, because he spent a lot of time with that community and with computer scientists in general. I think he called this in our internal discussions pretty much right. We were lucky that some people gave us a heads-up on not exactly what was coming with GPT, but we did have some, let’s say, orientation to that. So we were not taken by surprise by some of the things that surprised many people, let’s say, particularly in December of last year. We took that in stride. I think it would’ve been quite disorienting if we had not had that fairly clear in our minds already.
COWEN: How do you feel that the progress of AI will influence your future career as an economist?
JOHNSON: I’m not sure. I’m planning to have quite a long future career, by the way, Tyler. I am —
COWEN: But it should change what you do, right? I’m not saying it puts you out of work, but you ought to want to work with it in some manner. If it has big impacts, as the book says, it should have an impact on you.
JOHNSON: Maybe, or maybe it should have an impact on a lot of people, and there should be niches for some of us that are less impacted by AI. I’m not sure, Tyler, it’s a very good question.
I was working this morning on a plane, editing a document that was shared with a colleague over the internet. It just occurred to me how far we’d come since the 1990s that I was actually using the internet on a plane to edit a document that was shared with a colleague on the other side of the world. I didn’t expect that 10 years ago, let alone 25 years ago, so I think it remains to be seen. I don’t see the future that clearly, Tyler. Perhaps you need to get Daron on the show as well.
COWEN: We’ve had him on.
Cowen: A general question: what institutional arrangements protect a country from the middle-income trap?
JOHNSON: Well, if you think about countries that have really done well but not entirely broken free of being middle income — South Korea is one of my favorites because my wife’s parents were born in South Korea, and I spent a lot of time talking with them and thinking about the country and talking with Korean economists.
I think that you need to spread an education system that really encourages creativity and really helps you break from some of the strictures of the past. I think you need more entrepreneurship. You need a lot of integration with the world economy, a lot of exchange of ideas.
Ireland, Tyler, is one of my favorite countries, and a country that I would say has definitely broken free of the middle-income trap. They did it with a lot of direct foreign investment. You don’t have to have that, but they did it with an amazingly globalized people and obviously a lot of integration into American society and other European societies.
It’s probably at the human capital and probably about language, and probably about being able to be productive in multiple different kinds of organizational settings and cultural settings. You find fantastic Irish leadership all over the place. In fact, I believe the chairman of UBS — we were talking about UBS few moments ago — is Irish origin.
COWEN: If institutions are the key to economic growth, as many people have argued — Daron and yourself to varying degrees — why, then, is prospective economic growth so hard to predict?
In 1960, few people thought South Korea would be the big winner. It looked like their institutions were not that good. It was a common view: oh, Philippines, Sri Lanka — then Ceylon — would do quite well. They had English language to some extent. They seemed to have okay education. And those two nations have more or less flopped. South Korea took off. It’s now, per capita income roughly equal to France or Japan. Doesn’t that mean it’s not about institutions? Because institutions are pretty sticky.
JOHNSON: Yes, I think of institutions as being part of the hysteresis effect, if you can get it in a positive way, that if you grow and you strengthen institutions, which South Korea has done, it makes it much harder to relapse. There are plenty of countries that had spurts of growth without strong institutions and found it hard to sustain that.
You make a very good point about the early 1960s, Tyler. There wasn’t that much discussion that I’ve seen about institutions per se, but education — yes, absolutely. Culture — people made the same comparisons. They said, “Confucian culture is no good or won’t lead to growth. That’s a problem, for example, for South Korea.” That turned out to be wrong.
I think institutions are sticky. I think history matters a lot for them. They’re not predestination, though. You could absolutely carve your own way, but the carve-your-own way is harder when you start with institutions that are more problematic, less democratic, more autocratic control, less protection of property rights.
All of these things can go massively wrong, but building better institutions and making them sustainable, like Eastern Europe — the parts of the former Soviet Empire that managed to escape the Soviet influence after 1989, 1991 — I think those countries have worked long and hard, with very mixed results in some places, to build better institutions. And the EU has helped them in that regard, unquestionably.
COWEN: I’d say three years ago, I thought, incorrectly, that Ethiopia and Ghana had reasonably promising futures because they were on good tracks. They had racked up a number of years of high growth rates in a row. Now, it seems both of my views were wrong. What exactly is the predictive content of institution-based theories of economic growth? What’s the prediction we would make now?
JOHNSON: It’s a good question, Tyler. I think that if institutions have some predictive value, it’s fairly long term. This is something I worked on at the IMF. Other people were interested in this as well. I don’t think it helps you that much. Think about one year, two years, even five years.
I do think it helps to take a view on whether this is a more robust system, whether this is a system that can handle crises, whether the system will suffer some sort of relapse and major economic collapse. A collapse would be hard to reverse, of course.
I think the US has strong institutions. They don’t prevent us from having a very bumpy ride, and the ride is getting bumpier, it seems, about the last couple of decades.
COWEN: Here’s a simple and easy question about growth: How do we fix northern England?
JOHNSON: I think that’s a fantastic topic. I love the idea of more investment and infrastructure, of connecting the people of northern England. I’m from a town that’s on the border between the Midlands and North of England. Connections to other, more prosperous places, reducing travel time, encouraging remote work, moving the high-value jobs of headquarters and so on to lower-cost real estate places — these are all great ideas.
Of course, they have been discussed many times for decades, and it hasn’t happened. London is still an amazing magnet for talent, and that swamps everything else. But Manchester has two good football teams, and that might turn things around eventually.
COWEN: We’ve had a lot of autonomous advances in work from a distance that have nothing to do with northern England per se. Working over Zoom is one of many. AI may prove to be another one. Are you predicting a resurgence in northern England? Or is it just too much the case that the high-capital individuals want to leave for London in the South?
JOHNSON: I would love to predict a boom for northern England. It’s certainly doing much better now than it than it did in the 1980s, when I left the North of England. To me, it seemed easier to break in and have success in the United States than in London, actually, at that time. It is a struggle for talent, you’re absolutely right, Tyler. Where do smart people want to live? Where do they want to settle down? Where do they want to have families?
People are willing to have long, arduous commutes into London. I think that it would be great if either the market, or maybe the government, could induce and persuade people to move to other parts of the country, including further north, but it has been a long-term struggle.
COWEN: If you compare, say, Britain to France or Germany, the number two city in France would be Lyon. Lyon is a very nice place. It’s not as rich as Paris, but you wouldn’t say it has structural problems different from those of France as a whole. Germany has plenty of cities that are doing fine. Why is England, in particular, so unbalanced in terms of its urban geography?
JOHNSON: Well, I probably blame the Norman Conquest for that one, Tyler. Just kidding a little bit in terms of going back a thousand years.
COWEN: No, it’s fine as the answer, but tell us how.
JOHNSON: London has long been this amazing place, this very dynamic place, a very entrepreneurial place. If you think about what started to go right in England before the Industrial Revolution, which obviously, the Industrial Revolution was a British Midlands phenomenon more than a southern phenomenon. But the commerce of London, the power of London as a trading center, the openness of London to new ideas and new products — that was a very important part of what made Britain susceptible to and open to industrialization.
I think it’s the appeal of London and the success of London over more than 500 years, Tyler. The North of England had a very good run, let’s be clear, but it was a run that was based on factories. And factories and concentrated workforces and all of that turned out to be fairly footloose, particularly after the empire declined, and the British found they didn’t have quite enough scale to compete with the Americans and not quite enough government support and corporate support, maybe financial support, to compete with the Germans. The North of England had a rough ride for a while.
COWEN: Should the UK join NAFTA? Which would mean chlorinated chicken, right? Would it be worth it?
JOHNSON: [laughs] Look, the UK should join the European Union, but —
COWEN: Let’s say that’s impossible, or maybe they won’t take you because you can’t pre-commit to staying, but NAFTA would take you.
JOHNSON: Maybe it would. I think that the people in NAFTA should worry about the British turning their backs on that, too, honestly, Tyler.
COWEN: But we wouldn’t mind. Our feelings wouldn’t be hurt. It’d be a much easier separation.
JOHNSON: Oh, I see what you’re saying — it wouldn’t be an emotional breakup. Yes, possibly. That’s a very intriguing notion and probably should be considered. I think NAFTA or USMCA, as it is now, has enough political baggage as it is in the US. The US, Tyler — we haven’t talked about this at all, but I think trade and trade policy and the politics of trade remain immensely complex in the US, and that’s not getting easier. I’m not sure that adding a significant additional piece to USMCA would be helpful or even feasible.
COWEN: My whole life I’ve admired the British system of government, but lately, it just feels to me that it’s not working. You have how many prime ministers in how many years? Is it four in seven years? I’ve lost track. What, if anything, has gone wrong, and how would you fix it? Procedurally, at the so-called constitutional level — I know you don’t have a formal written constitution — but what would you change? And what’s gone wrong?
JOHNSON: There is no constitution, exactly. There’s just heuristics, and there’s just what happens, and then establishing traditions. Look, it’s not unusual in the British parliamentary system, certainly, for one party to gain control, to keep control for a long period of time. That’s what the Labour Party did previously. Now the Conservatives have had this long run, and you run out of steam, I think, Tyler. You’ve run out of ideas.
Alternation of power would be good, but the parliamentary system is one in which a small . . . First, you can get a majority with a minority of the vote, and also, a relatively small shift in the vote can give you a big majority, and those big majorities are unassailable. I think it is the unfortunate downside of what is otherwise a pretty robust and definitely entertaining parliamentary system.
COWEN: I’m a big fan of your book, Jump-Starting America, about science and science policy. It’s with Jonathan Gruber. I have a few questions about science policy. You argue for a place-based policy to jump-start different regions in America. I suppose I’m worried in the same way that you’re worried about northern England, that we could do a lot of good things, which I would be in favor of, but most of the gains will go to the coasts.
Since you wrote the book, we’ve had mRNA vaccines. Well, Moderna — that’s in the Boston-Cambridge area; Open AI — that’s in San Francisco. Those are phenomenal developments, but they’re not in Rochester. They’re not in Iowa. Are you now more despairing of our ability to use place-based policies to jump-start these other regions?
JOHNSON: No, actually, we’ve gone the other way. We’re a bit more encouraged, Tyler. You’re right, of course, about those big breakthroughs, and we are all about the breakthroughs there. But one of the things that strikes me as being a strength of the US is the depth of the bench. All these other places where real estate prices are not so high — there are plenty of talented people who want to live there. Of course, there are massive spillover effects from investing in basic science, which is the basis of our argument.
When we make those public-sector investments — that’s to get the spillovers — why not look more broadly outside of Boston, outside of New York, outside of San Francisco or Seattle, and attempt to make those places more central, tap into the talent that already exists in those places? I think the shrinking of distance, or a different view of distance, Tyler, after the pandemic, has got to be a helpful part of that.
COWEN: Reading the book, much as I like many parts of it, I feel you’re underrating the benefits of sunshine. When you list your most promising places for regeneration, the top three on the list are Rochester, Pittsburgh, and the Syracuse area. Now, Pittsburgh may be doing okay, but Rochester is still losing population. Syracuse area not doing so well.
The entire southeast — not the entire, but most of the southeast — has done well. Nashville — a star performer, not on the list. Most of Texas, doing well. What if we just had a simpler theory that said it’s sunbelt, and a good university helps. Isn’t that a better predictor for the up-and-coming places than like Rochester, Pittsburgh, Syracuse? Cleveland is on your list. Ithaca, New York; Cincinnati — they’re all in your top 10. They don’t seem like winners to me.
JOHNSON: [laughs] Well, what we were flagging there was potential, potential that could be tapped. On our website, jump-startingamerica.com, there’s an interactive map, and you can put in different criteria, so if you like the idea of ranking places, you can change the ranking or change the weight on different elements. We did not put hours of sunshine in there, Tyler. Maybe we should in the next version, and then you could put that in the index. Totally valid point.
I think the bigger argument that we’re making there is, look beyond the East and West Coast. There are some really great places full of talent in the US. Which ones will surge to the front? I don’t know. That depends a little bit on who gets their act together and who gets focused on this. I do think the importance of a good university should not be understated, but that university should also feel pushed to engage with business, engage with finding jobs for graduates, helping start-ups — the kind of things that my university does and your university does very well, Tyler.
COWEN: Circa 2023, with some more years behind us — apart from wanting to spend more money, how would you improve US science policy for any given level of expenditure?
JOHNSON: I think that looking for places where you can invest in the basics, and you can imagine that there will be commercial applications. The Human Genome Project is one of our favorites. That was actually turned down by private venture capital in the 1980s because they said, “Yes, you can do this, map out the human genome, but we don’t know how you can monetize it.”
That turned out to be a brilliant moneymaker and a great job creator because the basic science then could be appropriated in an appropriate manner by downstream firms who could build technology-based solutions. I think we got 300,000 jobs out of that, most of which are good-paying jobs. Now, that requires some imagination. That requires some commercial orientation.
We are not saying that the government should do what the private sector is already doing in terms of financing ventures. We’re saying, look for those big potential breakthroughs that could have commercial applications and seek solutions that address anything that the private market is not able to do along the way. If the private market can build it, Tyler, let them get on with it, and then we’ll figure out later if there are consequences we need to deal with. I think the problem is, we don’t do enough of the basic science, and we don’t convert enough of that into commercial products.
COWEN: A few questions about the Simon Johnson production function. What is your most unusual successful habit?
JOHNSON: I don’t know. I play tennis. I find playing tennis to be immensely mentally challenging. It just makes you clear your mind. If you have anything else in your mind at all, you miss the ball and embarrass yourself. I just find that to be completely refreshing.
COWEN: What’s your favorite novel and why?
JOHNSON: Oh, Neal Stephenson’s Snow Crash. First of all, because Paul Krugman said in 1992 or something that it was the best possible predictor of the future, and it turns out to be pretty much right. I think Stephenson predicted the metaverse. Stephenson predicted a bunch of things that happened at the political-social level. It’s an exaggeration; it’s a fun novel. I’m still looking for any science fiction of any kind that’s a better predictor of the future than that one.
COWEN: We have an episode with him. Do you think our actual future will be as dystopian as his novelistic future?
JOHNSON: Well, he wrote the book 30 years ago, and I think we’ve had some dystopian moments along the way, Tyler. I think the key idea that he has is that our technology-powered future will lead to a fragmentation of society and to a collapse of the United States, a sort of political disunion. There’ve been plenty of pressures that were . . . maybe even the pressure of today was unimaginable compared to when he wrote 30 years ago. I don’t think it ends up quite so bad.
I’m an immigrant to this country. I like a lot about this country. I like the American political system, Tyler, which I know is a deeply unfashionable thing to say, but I like the fact that people are always taking swings at each other. We don’t have a closed elite. You didn’t have to go to Oxford and study a certain degree in order to become prime minister. I like the intensity of the back and forth in American politics.
Wasn’t it Winston Churchill who said the Americans — a little condescending, but Churchill’s mother was American, and let’s remember it — the Americans always do the right thing after first exhausting all the other possibilities.
COWEN: What’s your favorite movie, and why?
JOHNSON: I did like Everything Everywhere All at Once. I know that’s just the most recent movie I saw. Also, it did give me a headache. I had to close my eyes at a certain point, but I did like the whole idea that we have multiple empowering personalities that we can draw on in different circumstances. I thought that was supremely wonderful, and I did really like the fact that the Asian actors won so many Oscars. I thought that was outstanding.
COWEN: Now, I’m quite sure the position of MIT is secure in our future, but what’s the future of the MBA more generally, beneath, say, the top 10 or 20 schools? Will we still have MBAs in 20 years? Will anyone go and pay all that money?
JOHNSON: It’s a great question. I think a big part of what went right in the era of Henry Ford was the so-called engineer managers, the people who created the new tasks, the people who designed companies, the people who built these organizations. I think they got a little bit of a bad name later, Tyler. It’s like middle management. Perhaps that was deserved. Perhaps things became too bureaucratic.
But I do think that people who come from engineering or people who like engineering and people who want to be managers — they want to organize things and make things more productive — I do think that’s a very important group of people. That’s the people who I spend most of my time with in the classroom. I push them to think about this issue of what’s the social angle? What’s the task creation? Why does that matter? If you’re just focused on automation and displacing workers, what are the consequences of that?
I have to say, these people are extremely engaged in that conversation, Tyler. Of course, that’s just the classroom conversation. What will they do in their careers 10, 20, 30 years from now remains to be seen. I hope I get to see it.
COWEN: My intuition is that I would rather hire people with technical training and no MBA than hire people with the MBA, maybe even now. Again, not Harvard, not MIT, where a lot of it’s selection and networking. But in terms of what you actually learn, it seems to me the other backgrounds are more valuable.
JOHNSON: I think an engineering or a science undergraduate degree is a very strong preparation for the modern world. I also find that for the right people, where it’s a match, an MBA-type qualification — and they come in many shapes and sizes, as you know — I find that that can be really quite helpful to them. I know some remarkably talented people who fit exactly that description, so I’m sticking with the business model, Tyler.
COWEN: Last two questions. First, when you’re evaluating students — not just who will make a lot of money, but someone you might want to work with or you think will change how the rest of the world thinks about ideas — what are the non-obvious qualities you look for in those students? Sure, they should be smart, they should work hard, but what else?
JOHNSON: Ability to communicate. Daron and I actually have a course based on the book we’re teaching to MIT undergraduates right now. It’s a communications elective, Tyler. We have a fantastic writing coach working with us, and some superb TAs. We’re trying to help the students learn how to write persuasively about technology. Technology policy, sure, but about technology more broadly. It’s a lot of fun, and, I think, a good complement to their other parts of education, which are obviously much more technical, much more mathematical.
I’m interested in, can people communicate ideas? Can they communicate with me? Can they communicate with other people, particularly about technical topics? And do it in a way that isn’t condescending, a way that explains things clearly, a way that invites further debate and conversation? And then, do they listen to the feedback they get and sharpen the messages?
COWEN: Last question: What will you do next, other than promote the book? Again, to repeat, everyone: Power and Progress: Our Thousand-Year Struggle over Technology and Prosperity, Simon Johnson and Daron Acemoglu. But what next?
JOHNSON: Well, talking a bit about the book is a big part of what’s next, Tyler, because it’s immensely fun. You get to go around the country and some other countries and really get feedback and hear the sorts of questions and pushback that you’ve been giving me today. That’s immense.
I think the next thing I’ll do is write another book. I like writing books. I like working with my friends. We’ve got various ideas and things to explore. It takes three or four years to bring one of these books to fruition.
Then, at the end of the day, people might not be interested. Maybe the news is something completely different for a while. It’s a big bet in terms of your personal time, but it’s also fun. It’s also fulfilling, and I feel it makes a contribution that I enjoy. I really enjoy talking to people about the book, Tyler, and I’ve really enjoyed talking to you. I always enjoy talking to you, Tyler. I really appreciate this opportunity. Hopefully when the next book comes around, you’ll have me on your show again.
COWEN: A pleasure to chat with you, Simon Johnson, and very good luck with the book.
JOHNSON: Thanks a lot.
Photo credit: MIT