Turing Test: Larry Summers

Larry Summers Turing Test Photo

Larry’s Career Strategies

  1. Revolving door. Going back and forth between academic positions and policy seems to have paid off.
  2. Invest in relationships. “I’ve always tried to cultivate a culture of surrounding myself with the most able people that I can and trying to practice maximum reciprocal loyalty. People who’ve worked with me, I’m willing to listen to what they think for as long as they want to express opinions and I try for the rest of their lives to look out for their careers and in return, I expect candid advice, hard work and a willingness, once decisions have been made, to throw themselves into execution. And I think I’ve been very, very fortunate in the people I’ve been able to surround myself with. Whether it was young economists like Andrei Schleifer and Greg Mankiw as research assistants very early in their careers. Or Sheryl Sandberg working with me at the Treasury Department with a young Tim Geithner working with me at the Treasury Department. I think a lot of any success that I’ve had has had to do with my ability to attract and select very, very able people and then to collaborate with them.
  3. Consider the counterfactual. I’ve always asked myself in every position I’ve held what’s going to be different because I was here?
  4. Practice the Turing Test.I’ve always believed that if you say something meaningful then it could possibly be wrong and therefore you should always ask what data would disconfirm the view that you have.

People mentioned

Bill Gates

Dustin Moskovitz

Sheryl Sandberg

Andrei Schleifer

Greg Mankiw

Tim Geithner

Related Reading

Wealthier is Healthier by Summers and Prichett

Transcript

Ales Flidr:                        Professor Summers thank you so much for joining us today. Before we delve into the more theoretical questions, many of our listeners might not be familiar with your background story and your career path. So some people are still working on their post-docs in [00:01:30] their late 30s, whereas you became the World Bank’s Chief Economist at the age of 37, if Wikipedia is right. So how did you achieve that?

Larry Summers:               Oh I don’t know, I just kept going. I was a graduate student, it was a different era of people didn’t usually take time off between college and graduate school in those days. And so I was a graduate student at Harvard for four years, and then I started teaching at M.I.T. and then I was lucky enough [00:02:00] in my late 20s to be invited back to be a Professor at Harvard. And I did a lot of research over that period and I think the World Bank was looking for a fresh voice from outside the traditional development field when they were choosing a Chief Economist, and I was among the candidates. And somehow I got to the end and they selected [00:02:30] me, and I thought it was a terrific opportunity to learn about what in many ways was most fundamental in the world, which is what’s happening in developing countries where billions of people live. Thought it was also the first opportunity I really had to have a position of responsibility where I was leading an organization within the World Bank that had several hundred people. And in the end [00:03:00] it proved to be a remarkably satisfying thing to do.

Larry Summers:               During that time I was able to write a significant research study that, at that moment, was novel, though it would be less novel today, demonstrating that the highest rate of return investment, or so I claimed, in the developing world at that time was in educating girls. My research assistant [00:03:30] on that study was a young woman named Sheryl Sandberg, and had just graduated from Harvard. I also had a chance to commission a report on global health that was very much in the spirit of effective altruism because it focused on cost benefit analysis and making the highest return investments available. [00:04:00] And that report had 120,000 copies sold, but one reader was by far the most important. His name was Bill Gates, and he has said that it was the report that led him to devote a large part of his philanthropy to global health because he saw it’s immense human significance. He also saw that it depended upon development [00:04:30] that involved fairly high fixed costs, but then distribution of something that had very low marginal costs. And so in many ways it was like software where he had already been very successful, but this time he didn’t make profits, he saved lives. And so I’ve always been proud that our report helped to draw Bill’s philanthropy towards global health.

Holly Elmore:                   Well that’s …

Ales Flidr:                        Thank you for that.

Holly Elmore:                   You’ve alluded to so many of the topics we wanted to talk about. [00:05:00] Just to stick on the career question for a little bit longer. If you were to pick the most important skills that helped you get to the position of being an influential advisor and policy maker, what would you say they are?

Larry Summers:               I’m not sure, I’d probably have to ask the people who hired me into various positions that question, but I’ll tell you some things I’ve always tried to do. I’ve always tried to see both sides [00:05:30] of problems, and never been sure that I was right, and always been eager to see both perspectives or multiple perspectives on any given problem. I’ve always believed that if you say something meaningful then it could possibly be wrong and therefore you should always ask what data would disconfirm the view [00:06:00] that you have. I’ve always asked myself in every position I’ve held what’s going to be different because I was here? Many people aspire to hold positions for the sake of having held them, and I always tried, sure I wanted to have leadership positions, but I always tried to ask what was going to be different if somebody looked back and looked at the consequences of my leadership. [00:06:30] What did I want to accomplish? And to have a clear answer to that question, and if I didn’t have a clear answer to that question, then I was probably the wrong person for that leadership role.

Larry Summers:               I’ve always tried to cultivate a culture of surrounding myself with the most able people that I can and trying to practice [00:07:00] maximum reciprocal loyalty. People who’ve worked with me, I’m willing to listen to what they think for as long as they want to express opinions and I try for the rest of their lives to look out for their careers and in return, I expect candid advice, hard work and a willingness, once decisions have been [00:07:30] made, to throw themselves into execution. And I think I’ve been very, very fortunate in the people I’ve been able to surround myself with. Whether it was young economists like Andrei Schleifer and Greg Mankiw as research assistants very early in their careers. Or Sheryl Sandberg working with me at the Treasury Department with a young Tim Geithner working [00:08:00] with me at the Treasury Department. I think a lot of any success that I’ve had has had to do with my ability to attract and select very, very able people and then to collaborate with them.

Ales Flidr:                        So maybe it would be easier to just go over the list of most influential people in the world and cross out those that weren’t somehow influenced by you.

Larry Summers:               I don’t know about that.

Ales Flidr:                        Yeah. So outside [00:08:30] of career impact, many effective altruists are also interested in how to best spend their money in terms of charitable donations and there are some large charitable funds. So if you were in the Bill Gates or Dustin Moskovitz position of having multi-billion dollar fund, what would your approach be and, I mean, both in terms of the objective and the cause areas that you would look into, but maybe more interestingly, how would you try to influence the non-profit sector with this amount [00:09:00] of money?

Larry Summers:               It’s not a position I’m in so it’s not something I have thought hugely carefully about. I think you have to decide what you care about. And I don’t think you have an obligation to choose the most important problem in the world. Some people believe they want [00:09:30] to help the whole world, some people believe they want to help America, some people believe they want to help the city in which they live. I don’t think one of those preferences is more legitimate than another. I think if you are supporting a new hockey rink at the prep school you attended, [00:10:00] that’s a reasonable thing to do, but I don’t think of that as being altruistic in quite the same way as supporting the education of young children in disadvantaged areas. But in general, I think there are many areas and many different values that one can bring to being effectively altruistic, and I think that people [00:10:30] who are engaged in philanthropy, there’s nothing wrong with some part of that philanthropy being a display of loyalty to institutions or individuals who made a big difference in their lives. But I would hope that some part of the philanthropy would be directed at being altruistic towards our broader society.

Larry Summers:               If I look at philanthropists, I think the [00:11:00] very difficult issues in philanthropy involve the concepts of fungibility and here there are different ways of looking at it. If you give more money for financial aid to Harvard, what’s [00:11:30] probably going to happen is Harvard is going to be in a position to contribute less of it’s budget to financial aid and therefore Harvard is going to have more money, but probably your money is not going to contribute to incremental financial aid. And so I think philanthropists would do well, as a general matter, to ask themselves what will the ultimate impact, rather than the stated impact, [00:12:00] of their money be?

Larry Summers:               On the other hand, I think some philanthropists appreciate that insufficiently. I think there are other philanthropists who over appreciate that and who think that everything they need to do needs to be catalytic or needs to leverage other people’s money. Well you can’t have everybody leveraging and no-one being levered. [00:12:30] So I think there’s an important role for the lever. And so I suspect if I had a large amount of money I would decide that the difference between the philanthropy sector and capitalism is this, in capitalism there are 100 start ups. And if you look after five years, 85 of them have failed, [00:13:00] 10 of them are puttering along, three of them have been absorbed by other organizations in a substantial way, and two of them are their own large institution. If you look at the non-profit sector, if 100 start ups were formed, 70 of them are still kinda puttering along, 25 have [00:13:30] failed, the rest have doubled, and nothing has become massive, and nothing has become absorbed by another organization.

Larry Summers:               And so I would say that the niche that is unfilled in philanthropy is not starting new things, it’s taking small things that have succeeded [00:14:00] to far larger scale. And I think philanthropy of that kind would fill an important gap and, because it generated competition to be the one that was taken to large scale, would also promote evaluation and metrics which would facilitate the other parts of an improvement. Because if you had more metrics in order to try to go big, [00:14:30] you would have more discovery the things a lot of shrink and go small. So if were to enter, if I were in a position to enter, the philanthropic space, it would be what in capitalism would be called growth stage investing.

Holly Elmore:                   We want to move on to our audience question. It’s known that you disagree with Bill Gates on the need for a robot tax, you suggested retaining workers instead. And first, I’d just like to get your views on that and then we’d like to [00:15:00] give you the Ideological Turing Test and ask you to give the best possible restatement of Gates’ position or possibly something that he’ll wish he’d thought of himself.

Larry Summers:               Bill’s a friend and he’s very, very thoughtful. I’m not sure he was right this time out. I think technology does have substantial dislocative effects. It always has had substantial dislocative effects. [00:15:30] Time magazine had an article in 1960 about how this generation of technology was not so much going to make existing workers more productive as it was going to remove a lot of jobs. So this is a common theme. I do think there’s cause for concern, but it seems to me that rather than taxing the robots, it’s much better to make our pie as large as possible by accelerating [00:16:00] innovation, and then making sure that once we have that pie, we tax the whole pie and use it to cushion the disruptions. And so I think an approach of redistribution that lets the market operate to maximize the size of the pie is much better than an approach of pre-distribution that seeks to limit the size of the pie [00:16:30] in the name of equity because I don’t think approaches of that kind have historically been hugely effective.

Larry Summers:               I also think the idea of a robot tax is very funny. Why tax robots and not dishwashers? Why tax robots and …

Holly Elmore:                   Maybe it should be retroactive?

Larry Summers:               Why tax robots and not tax word processing programs? Why tax robots and not [00:17:00] tax driers? Why tax robots and not tax electricity? And so I don’t think the idea of saying that we should slow technology down is a very fruitful one with all the want that still exists in the world, and all the poverty that exists in the world, and all the problems that need to be solved.

Larry Summers:               What’s the best argument [00:17:30] for Bill’s position? Robots are a metaphor. I mean, choose the cutting … He really means, choose the cutting edge technologies that are most disruptive and directly substituting for labor and put an extra tax on them, not with the objective of driving them out of business, but with the objective of slowing things down. And where the productivity benefit [00:18:00] is only minimal and the displacement benefit is large, if we slow things down, we’ll have to get some benefit. Yes, the tax is distortionary, but every tax is distortionary, and if we don’t have that tax then we’ll have to have higher taxes on people who work, higher taxes on people who save, and those will discourage good things too. Right now we have all kinds of tax breaks for investment. [00:18:30] We have accelerated depreciation, we have the special deduction that you’re able to take if you borrow money in order to make an investment. And so a small robot tax would be a leveling of the playing field in terms of worker’s and capital because currently we have subsidies and that would be an improvement. That would be the best argument that I can think of for his position.

Ales Flidr:                        What is your impression of effective altruism as a movement [00:19:00] and what advice would you give?

Larry Summers:               I’m here because I believe in it and I admire the idea that you can both count and care. And I think too often people who care think that that removes from them the obligation to count and be rigorous and be evaluative and be tough minded. [00:19:30] And I’ve always thought that the morally important something is, the more important it is to be brutally analytical and clinically calculating as you go about making decisions. And I think that’s part of a philosophy of effective altruism, and that’s something I admire.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s