#57 – Tom Kalil on how to have a big impact in government & huge organisations, based on 16 years’ experience in the White House
You’re 29 years old, and you’ve just been given a job in the White House. How do you quickly figure out how the US Executive Branch behemoth actually works, so that you can have as much impact as possible – before you quit or get kicked out?
That was the challenge put in front of Tom Kalil in 1993.
He had enough success to last a full 16 years inside the Clinton and Obama administrations, working to foster the development of the internet, then nanotechnology, and then cutting-edge brain modelling, among other things.
But not everyone figures out how to move the needle. In today’s interview, Tom shares his experience with how to increase your chances of getting an influential role in government, and how to make the most of the opportunity if you get in.
He believes that Congressional gridlock leads people to greatly underestimate how much the Executive Branch can and does do on its own every day. Decisions by individuals change how billions of dollars are spent; regulations are enforced, and then suddenly they aren’t; and a single sentence in the State of the Union can get civil servants to pay attention to a topic that would otherwise go ignored.
Over years at the White House Office of Science and Technology Policy, ‘Team Kalil’ built up a white board of principles. For example, ‘the schedule is your friend’: setting a meeting date with the President can force people to finish something, where they otherwise might procrastinate.
Or ‘talk to who owns the paper’. People would wonder how Tom could get so many lines into the President’s speeches. The answer was “figure out who’s writing the speech, find them with the document, and tell them to add the line.” Obvious, but not something most were doing.
Not everything is a precise operation though. Tom also tells us the story of NetDay, a project that was put together at the last minute because the President incorrectly believed it was already organised – and decided he was going to announce it in person.
American interested in working on AI policy?
We’ve helped dozens of people transition into policy careers. We can offer introductions to people and funding opportunities, and we can help answer specific questions you might have.
If you are a US citizen interested in building expertise to work on US AI policy, apply for our free coaching service.
In today’s episode we get down to nuts & bolts, and discuss:
- How did Tom spin work on a primary campaign into a job in the next White House?
- Why does Tom think hiring is the most important work he did, and how did he decide who to bring onto the team?
- How do you get people to do things when you don’t have formal power over them?
- What roles in the US government are most likely to help with the long-term future, or reducing existential risks?
- Is it possible, or even desirable, to get the general public interested in abstract, long-term policy ideas?
- What are ‘policy entrepreneurs’ and why do they matter?
- What is the role for prizes in promoting science and technology? What are other promising policy ideas?
- Why you can get more done by not taking credit.
- What can the White House do if an agency isn’t doing what it wants?
- How can the effective altruism community improve the maturity of our policy recommendations?
- How much can talented individuals accomplish during a short-term stay in government?
Get this episode by subscribing to our podcast on the world’s most pressing problems and how to solve them: type 80,000 Hours into your podcasting app. Or read the transcript below.
The 80,000 Hours Podcast is produced by Keiran Harris.
Highlights
If the President invites you to a meeting, or even if someone who just works for the President like me, many people will show up. And, there are several different types of meetings that you could have. So, one would just be, hey, we think that this issue X is important. And so, we want to have a conversation about what are the best possible ideas to make progress on the issue X. So, that would be a way of trying to understand the issue, understanding potential courses of action. But, we might also have a meeting where the purpose is to serve as an artificial deadline. And, we would use the time between now and when the event was going to occur to ask people what it is that they might be prepared to do.
So, for example, the President said, in the same way that if you win the Superbowl or the NCAA, you get to come to the White House, the same thing should be true if you win a science fair, or robotics competition. So, he wanted to sort of increase the prestige and status of STEM education, and get more young boys and girls excited about STEM. And so, we would use that event, not just for the President to interact with these kids who are doing these amazing robotics competitions and science fair projects, but to mobilize the entire country to make specific commitments that would advance the ball.
So, the question that I would ask people is, if you were the President and you could call anyone and ask them to do something, who would you call and what would you ask them to do? And, sometimes there are organizations that are not part of the government, but are, particularly if they work together, are really in a position to move the ball forward on a particular national issue.
People have many more things that they’re supposed to do than they have time to do, so if what you do is you show up and say, “I would like to give you another thing to do”, generally people are not gonna be super receptive to that. If there was something that, for example, I wanted my boss to do, my view was I should not show up and say, “I have a monkey that is on my back and I would like to transfer this monkey from my back to your back.” What I would figure out is, how could I make it as easy as possible for him to help me. For example, if I needed help getting a particular member of the Cabinet on board to support an idea that I was enthusiastic about, then I would say, “If I draft an email for you will you look at it, and if you’re comfortable with the substance and the content, will you send it to them?” He would generally say yes to that.
If you want someone to help you, make it as easy as possible. That also requires an understanding for an individual in the context of the particular organization, what’s easy and what’s hard? You have to acquire a lot of fine grained institutional information about how different organizations work, about how decisions are made within that organization, and what’s relatively straightforward for them to do and what’s really difficult to do and what constraints they’re operating under.
Sometimes there will be windows of opportunity, we call those policy windows. Let me give you a concrete example. When the Ebola crisis was going on I had been briefed on a program that was going on at DARPA, which was, could we dramatically reduce the time to go from bug to drug? We have this new emerging infectious disease, we don’t have a vaccine for it, telling people at that point, “Well, give us ten years and we’ll have something for you” is not terribly satisfying. The approach that DARPA was using was as follows, you’ve got someone, they’ve been exposed to a pathogen, they survived, their body produced a set of antibodies that will provide immune protection, let’s identify those antibodies and then create a synthetic oligonucleotide construct that will directly encode for those antibodies.
The process of doing that would be a lot shorter than the traditional process of vaccine development. I knew that we were going to ask for what is called an emergency supplemental, which is, we didn’t budget for Ebola because we didn’t know it was gonna happen therefore we need extra money to be able to contain it. I was able to get some additional funding for this new approach added to that and my primary motivation for it was, maybe it’ll help in Ebola, but almost certainly if it works it will improve our ability to respond to future emerging infectious diseases, or maybe even a world of engineered pathogens.
Let’s say that you decided that a good next step would be that there should be more R and D that helps reduce existential risk and maybe it’s on the pandemic side or maybe it’s on the AI safety side. Well, one thing that you’d want to know is which agencies would have the capability to do that? And how do those agencies make decisions? So for example, DARPA has a budget of $3.5 billion but one of the interesting things to know about DARPA is the P in DARPA stands for projects, and what that means is that every four years they stop working on something. So that means a quarter of their budget is available for new projects every year. So unlike an agency like the National Science Foundation where if you want to get them to do something new, you kind of have to get them some more money because they’re not gonna say, “This year we’re no longer gonna fund condensed matter physics. Instead we’re gonna fund this other thing.”
They’re gonna fund physics in perpetuity, and so the way you influence DARPA is you get someone to go there who wants to pitch that program. So DARPA is a very program manager-centric organization and the director doesn’t necessarily choose what to work on next. The DARPA program manager candidate arrives and their job talk is based on an idea that they have and they have to answer something called the Heilmeier Catechism, which are a set of questions that this DARPA director, previous DARPA director, came up with for evaluating whether or not you have a good idea for an R and D program.
So if you want to influence DARPA, and let’s say you want them to do more in the area of improving our response to engineered pathogens, then the way to do that is to say, “Well, who could we find who would be world class technically and has a great idea and could have convincing answers to the Heilmeier Catechism who would be willing to go to Washington for four years,” right? So if you didn’t know anything about that agency or how it operated, then that wouldn’t occur to you as a path to influence, but if you knew exactly how it worked and its culture and its procedures for decision making and how its budget works, then it becomes a little clearer about how you’d influence that as an organization.
Articles, books, and other media discussed in the show
Tom’s work:
- Policy Entrepreneurship at the White House – MIT Press Journals by Tom Kalil
- Case study of exploring the role of Tom as Deputy Director for Technology & Innovation at the White House Office of Science and Technology Policy by Linda A. Hill and Allison J. Wigen
- The Promise of Incentive Prizes by Jeff Ubois & Thomas Kalil
- Tom’s talks at EA Global: Opportunities in Science & Technology Policy and Policy Entrepreneurship for EAs
Everything else:
- Emerging Tech Policy Careers — expert advice and resources for people interested in US emerging tech policy (AI, bio, etc.)
- Center for Security and Emerging Technologies vacancies
- Our job board with other vacancies
- Strong Medicine: Creating Incentives for Pharmaceutical Research on Neglected Diseases by Michael Kremer and Rachel Glennerster
- Launching The Innovation Renaissance: A New Path to Bring Smart Ideas to Market Fast by Alex Tabarrok
- Ansari X Prize
- The Heilmeier Catechism
- American with a science PhD? Get a fast-track into AI and STEM policy by applying for the acclaimed AAAS Science & Technology Fellowship by Niel Bowerman
- Presidential Management Fellows Program
- Yes Minister
Transcript
Robert Wiblin: Hi listeners, this is the 80,000 Hours Podcast, where each week we have an unusually in-depth conversation about one of the world’s most pressing problems and how you can use your career to solve it. I’m Rob Wiblin, Director of Research at 80,000 Hours.
We’re about the get some advice from someone who has 16 years experience in the White House, both about how to land a job in the US executive branch, and how to get more a lot done once you’re actually in there.
If you’re a US citizen and you are interested in moving to DC to work on AI policy issues, then we would potentially love to chat with you. In the show notes we’ll stick a link where you can apply for free careers advising from 80,000 Hours. And we’re especially keen to meet people with backgrounds in law, policy or computer science, so if that sounds like you do let us know about it!
Alright, here’s Tom Kalil.
Robert Wiblin: Today, I’m speaking with Tom Kalil. Tom Studied political science and international economics at the University of Wisconsin, Madison and worked on the Democratic presidential campaigns in 1988 and 1992, before serving in the White House under Obama and Clinton helping to design and launch national science and technology initiatives, from 2009 to 2017. That involved working at the White House Office of Science and Technology Policy First as deputy director for policy and then as deputy director for technology and innovation.
Robert Wiblin: He’s now a chief innovation officer at Schmidt Futures, a foundation, which among other things works to improve US science policy and identify and pursue 21st century moonshots. Thanks for coming on the podcast, Tom.
Tom Kalil: Happy to be here.
Robert Wiblin: So I hope over this interview to hear your thoughts of long term ism in government and draw out a whole bunch of your advice on how people can best pursue policy careers in the US and potentially in other countries as well.
Robert Wiblin: But first, what are you doing now and why do you think it’s really important work?
Tom Kalil: Well as you noted, I’m serving as the chief innovation officer for Schmidt Futures, which is a philanthropic organization founded by Eric and Wendy Schmidt. I believe in general, philanthropy has big opportunity to make investments in things that both the public and private sector are under investing in.
Robert Wiblin: Now, what kind of investments are those?
Tom Kalil: One of the things I’m interested in is that the federal government is serious about using science and technology to solve certain types of problems in areas like national security, energy, space, health, and basic science. But there are a lot of other areas where the relevant mission agency has little or no capacity to invest in science and technology. Those create systemic gaps in the country’s research and innovation portfolio. That’s an example of an area where a philanthropist or a foundation has more flexibility to address some of those gaps.
Robert Wiblin: Okay. So we’re not gonna focus too much on your current foundation work in this interview, because I’m especially interested to hear about your many years of work in government. I guess to help, to give people a sense of how they could potentially build their own career in policy and government, let’s just briskly walk through through your own career.
Tom Kalil: Sure.
Robert Wiblin: Yeah. What was the path you’ve taken to progressively having more impact or influence in the policy world from, I guess, say your student days to to now?
Tom Kalil: Yeah. My career has been a first or second order consequence of my decision to volunteer on the 1988 Dukakis for president campaign. I did that in the summer of 1987. I volunteered for a part of the campaign called the issues department. That is like bootcamp for would-be policy wonks, because you have to learn how to get up to speed quickly on a new issue, how to create a network of outside advisors. You learn this work ethic of staying up until it’s done, because you can’t ask a presidential candidate for an extension. You learn how to look for new policy ideas that your candidate can get behind, how to prepare them for presidential debates, how to understand what issues they’re likely to get asked about as they travel around the country.
Tom Kalil: You certainly don’t become an expert in one particular area, but you learn how to become a generalist, read lots of things, talk to lots of smart people and then try to figure out what does the candidate really need to know about this issue to be informed and knowledgeable.
Robert Wiblin: Presumably, that couldn’t have been the complete beginning, ’cause you can’t just walk off the street and get a job developing policy for presidential candidates. How do you actually get your foot in the door to take that kind of position?
Tom Kalil: It was really volunteering early. I was volunteering in the summer of 1987 when there were lots of different presidential candidates. There wasn’t the same level of competition that there might be. Then the other thing that I was lucky, is that the issue that I knew something about, namely international trade, was an area of disagreement between the candidates. The fact that I knew something about international trade made them more interested in bringing me on as a volunteer.
Robert Wiblin: How much do you think it ended up boosting your career that Dukakis actually became the nominee?
Tom Kalil: That was important. What was even more important is that a number of the people that I worked for also wound up working for Bill Clinton. In ’92, very close to the general election, in the fall of ’92 some of the people I’d worked with in ’88 contacted me and said, “Hey, do you want to come down to Little Rock and write some of Bill Clinton’s position papers?”
Tom Kalil: As you may remember, Bill Clinton won. The first thing that happened was that Bill Clinton had said during the ’92 campaign, “It’s the economy stupid.’ And he said, “In the same way that we have a national security council that will ensure that the president is focused on foreign policy and national security and defense issues every single day, we need to focus like a laser beam on the economy. So we should have a national economic council.”
Tom Kalil: So Bill Clinton asked Bob Rubin to be the first head of his National Economic Council. Bob Rubin had never worked in government before. He asked some of the campaign people, “Who should I hire?”
Tom Kalil: And they said, “Oh, you have to hire Tom Kalil.”
Tom Kalil: Bob was initially highly skeptical … fortunately I did not know that at the time. I thought that it was more or less a pro forma interview so I was very relaxed. I visited him in New York and he asked me why I wanted to work for the national economic council, what types of things would I work on? Fortunately I knew a lot more about technology policy than he did, since he was an investment banker. I came off as appearing very knowledgeable. He said, “You are as advertised,” and offered me the job. That’s how I wound up working in the White House in my late twenties.
Robert Wiblin: So is this a path that you think listeners could potentially take to get on board with a presidential campaign, early in it’s run, when it’s a bit less competitive to get involved? And someone who is in their mid twenties can potentially get a policy position?
Tom Kalil: Absolutely. Yeah. I think that it’s a career move that has a lot of variance. I don’t think you should do it if you say, “Well, I want to be guaranteed that my candidate will win my party’s nomination, win the general, and then I will get a great job.”
Tom Kalil: I think you need to be prepared for an outcome in which none of those things happen and still be okay with doing it. I think that there are a number of instances during my career where I’ve just been lucky. Things could have easily worked out another way and I would not have had the opportunity to work for two presidents for 16 years.
Robert Wiblin: Did you feel kind of out of your depth a bit when you first got involved in the campaign?
Tom Kalil: No, I don’t think so. I mean, I certainly felt a little bit more out of my depth. Maybe I was too young and brash to recognize it when I was actually doing it. But having the first time I worked in the government be working for the White House is a little unusual. So there was a fair amount of learning on the job that I wound up doing.
Robert Wiblin: How did you choose Dukakis? And I guess if you don’t want to choose a specific candidate, what guidelines could we use to figure out what campaign do I want to get involved with?
Tom Kalil: I chose Dukakis primarily because I was in the Boston area at the time. That’s where his campaign was located. It wasn’t based on a review of all of the candidates and then saying, “Which particular candidate is closest to my policy preferences?”
Robert Wiblin: Should people investigate the candidates a lot and think, “Oh, this is one that I agree with ideologically or should they maybe look at how likely they think they are to win? Or how likely they are to win the general?
Tom Kalil: Oh. So it depends, right? So I think that in a lot of cases, there are actually not huge differences between candidates of a given political party, and a lot of it is about how likely they are to win or what’s their personal story, what are their strengths and weaknesses as a candidate? For example, Mike Dukakis was very thoughtful. He had a number of weaknesses as a political candidate.
Robert Wiblin: Okay. So moving on, what were you working on in the Clinton administration?
Tom Kalil: Yeah, I worked for part of the White House called the National Economic Council, which was this new organization whose responsibility was to coordinate domestic and international economic policy. I handled the technology portfolio. The entire staff might fluctuate between 20 and 30 people. Everyone would have a fairly broad portfolio.
Tom Kalil: When I started, I spent a lot of time working on issues related to information and communications technologies. That was because Vice President Gore was particularly interested in what was then known as the information superhighway. But today we just call the Internet because back in 1993 the Vice President had this vision of this global network of networks that was going to connect computers and all these other devices. Exactly how it was going to evolve was not entirely clear. I did a lot of work on those issues. Then also towards the end of the administration, I worked on something called the National Nanotechnology Initiative.
Robert Wiblin: Yeah. So I imagine once you’re actually in the White House, things are so busy, there’s so much going on that it’s maybe hard to find time to think and to figure out what you actually believe and what the priorities should be. How do you choose areas to focus on? Do you just have advisors? Does everyone have advisors going down and down and down? People who actually have time to read original research and figure out what the most promising technologies are?
Tom Kalil: You’ve always got a situation in which you can have both formal and informal advisors. For example, within the White House, there’s an organization called the President’s Council of Advisors on Science and Technology. There’s an organization called the National Academy of Sciences. It has a different name now, but they’re continually producing reports. What you have is, first of all, some things that you have to deal with. For example, if a crisis like Ebola occurs, you don’t have the option of saying, “Well, this year I’d prefer not to work on Ebola.” When a crisis like that occurs, you have to work on it.
Tom Kalil: That’s one source of the agenda. Second is that presidents campaigned on things. They know that they’re going to be held accountable by the press and potentially the voters , “You said you were going to do X, Y, and Z. What sort of progress have you made?” There’s always this promises made, promises kept exercise.
Tom Kalil: There might be people who are bringing issues to your attention. I’ll give you a canonical example of this. Congress passed a law that said that if the executive branch wanted to change the definition of what constituted a supercomputer for purposes of export control laws, then the president had to make a determination that this was a sound policy. When Congress passed this law, they set it at a particular level in terms of the speed and power and performance of the computer. The problem was is that what is a super computer one day-
Robert Wiblin: Is a laptop the next.
Tom Kalil: Is a laptop. Apple would come to us and say, “Hey, this is going to be classified as a supercomputer. Do you think that makes sense?”
Tom Kalil: We would say, “No.”
Tom Kalil: There would be things like that that people would bring to our attention of saying, “Hey, public policies about to have this deleterious impact on US exports without any corresponding national security benefit. Don’t you think that you should change that policy?”
Tom Kalil: You’re right in the sense that a lot of times you have the intellectual capital that you walk in the door with and the amount of time that you have for reflection and, as the EA community would put it, cause prioritization is limited, because you are constantly dealing with fires.
Tom Kalil: Now I was a little fortunate in the sense that science, technology and innovation are by their very nature longer term issues. I worked on President Clinton’s Caltech speech in which he announced the National Nanotechnology Initiative. He talked about some of the goals of the effort. He said explicitly in that speech, “We may not reach some of these goals for 20 years. But that’s why there’s an important role for the federal government, because it’s very difficult for companies to justify making investments that may not pay off for 20 years.” Relative to someone else in another policy council, like the National Security Council that might be dealing with what’s happening in Bosnia that day, I had the luxury of occasionally taking some time to say, “What should I be working on next?”
Robert Wiblin: So after Clinton left, you went to UC Berkeley during the Bush administration, right?
Tom Kalil: Yes, that’s right.
Robert Wiblin: So again, this is a common cycle, I guess, the other side of politics is in is in power, so you’re in DC. Then when the other side is in power, people go on and do other things and build their careers outside of government. And then come back.
Tom Kalil: Yeah. I think that even if the party that you are affiliated with controls the White House, it’s not like people are going to stay there from the first day to the last day. What I did was somewhat unusual. So people burned out or decide they want to do something else. It’s not unusual for people to move in and out of government.
Robert Wiblin: What were you doing at UC Berkeley?
Tom Kalil: My job was to help the campus develop new multidisciplinary research initiatives that cut across different departments and colleges. So integrating biology, the physical sciences, engineering, computer science and social and behavioral sciences. One of the things that I worked on that might be of most interest to the EA community was a program called Big Ideas at Berkeley. The premise of the program was a lot of times students have ideas of their own and they just need small amounts of funding, a deadline, permission and tips about how to have influence without authority in order to help make their ideas happen. What I found to be very effective was issuing a very broad call for ideas and seeing what students would come up with, what they were intrinsically motivated to work on. Then, as I said, to give them funding connections to people on and off campus and tips about how to be effective in the world.
Robert Wiblin: Yeah. So influence without authority is one of these ideas that you are constantly returning to. Do you want to explain that one?
Tom Kalil: Sure. Yeah. Part of this stemmed from the roles that I would have in the White House, which is that I would have a desk and a phone and a computer and a business card that said the White House on it. But, particularly when I was working for President Clinton, it wasn’t like there were a bunch of people reporting to me. In order to be effective, not only would I have to come up with an idea, but I would need to figure out how to build a coalition around that idea, even though the people who I ultimately had to convince to say, “Yes,” in no way reported to me. The ability to do that, to build coalitions around ideas that you’re excited about, is very important in a policy role when it’s not like you have dedicated funding that you have direct responsibility for and a large staff that is reporting to you, that you have this command-and-control relationship with.
Robert Wiblin: Yeah. Is having influence but not authority mostly just a disadvantage because you can’t tell people what to do? Or does it have advantages as well, because maybe people are more open with you and less defensive?
Tom Kalil: I think the advantage is that the number of things I was able to work on was much higher. For example, if I had been someone who had a responsibility for a particular issue and had a budget and a staff, then people would look at me strangely if I said, “Well actually, I have an opinion about what you’re working on as well.” I think what I found intellectually engaging about the roles that I had is that I had a fairly broad remit. The name of my division was technology and innovation and we used to joke, “If it’s not technology, it’s innovation.”
Robert Wiblin: I guess another possible benefit is if you have lots of people reporting to you, then your whole time is taken up managing people and dealing with programs that are already running. Whereas you can think about new things and calling people up and just explore.
Tom Kalil: Absolutely. Yeah.
Robert Wiblin: Okay. So yeah. Ultimately you ended up at the White House Office of Science and Technology Policy for the full eight years from I guess 2009 to 2017. What were your key responsibilities there?
Tom Kalil: When I started, I thought that my primary role would be advising the person that President Obama had designated as his Science Advisor. This was Doctor John Holdren. I had known John because he was one of Clinton’s external science advisors, but he’d never worked in the government before. I was playing a role in the transition period. This is the period after the election has been held and before the inauguration has occurred. The Obama campaign called me up and said, “Would you run the transition team for the office of Science and Technology Policy?”
Tom Kalil: My job was to help whoever Obama was going to appoint to be a science advisor, get ready to do that job. So what are the issues that they’re going to have to deal with in the first hundred days? What are the commitments that Obama made during the campaign that they’re going to have a role in implementing, that type of thing.
Tom Kalil: So when Obama appointed Holdren to be a science advisor, Dr. Holdren asked me would I be willing to stay on as his deputy? I thought that my primary role would be, because this was John’s first time working in government as opposed to nearly advising government, that I would be able to give him some institutional knowledge about how to get things done. That’s how I sort of initially conceived of my job.
Tom Kalil: Then I asked John if I could start recruiting people to work on various projects that I thought were interesting. For example, during the Bush administration, I’d served as the chair for the Global Health Working group for the Clinton global initiative and gotten really interested in global health issues because of the high benefit-to-cost ratio of a number of global health interventions. I started recruiting people to work on things that I thought were important and then discovered that I really liked that, because I had the opportunity to recruit people in their twenties and thirties and just be able to work on a number of things in parallel. Then also, I enjoyed mentoring and coaching the next generation. I had discovered that as well. Then I really started building a team both to identify issues that I thought were important and then to try to find out someone who could work on those issues full time. Or occasionally just finding someone who I thought was really talented and giving them the opportunity to engage in public service.
Robert Wiblin: Okay. So maybe let’s dwell on this on this time for a bit. I mentioned that after someone gets elected they’ve got like 75 days or something until they actually take office. There’s going to be this very manic period when they’re trying to staff up quite a large executive operation.
Tom Kalil: Yes.
Robert Wiblin: I guess you were brought in soon after the election I suppose to figure out what are we going to do with the OSTP in this new administration?
Tom Kalil: Yes.
Robert Wiblin: I imagine that’s a time when if you can talk to the right people, there’s potentially a good chance of getting a job or a very unusually good chance of actually finding a role, so this might be a scramble, I guess, for people who are qualified to take positions that you’re thinking, “Oh well we want to talk about now in technology, want to have someone to do that.” Then you’ve got to very quickly find someone to do it.
Tom Kalil: Yes.
Robert Wiblin: Yeah. Do you have any comments on, I imagine people are probably trying to figure out how they’re going to staff the next administration in 2021 already, because you have to prepare so much.
Tom Kalil: Yeah, there’s actually not a very structured way of doing that. It tends to happen in earnest when the two parties have their nominees and then they can start working on their transition. Transition is usually sequencing and prioritization. What’s the order in which you want to try to get things done? Then staffing the administration. What you want to do is not only begin to identify candidates, but then you have to start vetting them, because particularly if they’re going to go through the process of Senate confirmation, there are all these reasons why people can fail to make it through that. So you’ve got to have these very difficult conversations with people to say, “If we put you through that process, is there anything that’s going to come out about you that’s going to embarrass you and the president elect.”
Robert Wiblin: So is there a bit of a phenomenon where if you’ve already served in a previous administration, then you’re viewed as safe? ‘Cause you’ve already-
Tom Kalil: If you’ve gone through that, yeah. If you’ve gone through the FBI background check or you’ve gone through … Senate confirmation is not necessarily for all positions. It’s necessary for a subset of those positions, particularly those that are presidentially appointed and Senate confirmed. But yeah, that’s a difficult process. Then what happens is that the administration will usually start with the most senior positions. For example, you need to have a Secretary of Defense right away. You can’t say, “Oh well, six months from now.”
Tom Kalil: It’s even to the point where in some cases, they might ask someone from a previous administration to stay. For example, the Trump administration asked our Deputy Secretary of Defense who was seen as not particularly political, he was a former colonel, a Colonel Bob Work, to stay on until they’d managed to get to identify and get confirmed their Deputy Secretary of Defense.
Robert Wiblin: Yeah. Do you have any advice for listeners who might be at a point in their career where they could imagine working in the next White House? What kind of things that they can do now to bring themselves to the attention of the right people? So they’re in a good position come November, 2020.
Tom Kalil: Obviously, it depends on what point in their career they are. If they are at the beginning of their career, then the question is, is there someone that can serve as an advocate for them? For example, if they’ve served as a research assistant for a professor who is advising people in government or has moved in and out of government before, they might say, “Oh, once we know who is filling this particular position, they’re going to be looking for a young, energetic, special assistant. That’s a role that you could play.”
Tom Kalil: Politics, all careers are this mix of know-how and know-who. I think that a lot of the reasons that you tend to pick people that you’ve worked with in the past, or people that you know, vouch for, is that there are a bunch of things about these roles that may be difficult to observe from the outside just by looking at your resume.
Tom Kalil: So is this the type of person who has a very strong work ethic? We used to call this the “Stay up until it’s done.” You can’t ask the President of the United States for an extension. The ability to produce a lot of work in a short period of time, because you might be in this situation where something has to be done by the end of the day. The ability to function without a lot of babysitting. Your boss tells you, generally, the problem that needs to be solved and you go figure out how to solve it. You’re not constantly coming back and saying, “Well what about this?” Right? Having a lot of common sense. Being someone that people like to work with. If you have this aversive personality and no one likes to work with you, that’s going to be a disaster in a political environment, whereas that might be less problematic if you’re an individual contributor as a software engineer.
Tom Kalil: These are characteristics and qualities of someone that you’re more likely to know if you’ve worked with them as opposed to looking at their resume. I think that there is this high level of reliance on recruiting people that you’ve worked with or people whose opinion you really trust, have worked closely with. Because it’s sort of like being in the foxhole with someone.
Robert Wiblin: Okay, we’ll come back to the career advice and what would make you a good fit later on, but what’s a big win that you’ve had at some point over your career? What kinds of things can people accomplish?
Tom Kalil: Yeah, I’ll give you a couple of examples. One was when I worked for President Obama, I was able to recruit at any one time, 20 people. Each of those people were working on a handful of things that were consequential. That wound up being pretty significant.
Tom Kalil: Let me give you one example. A young woman emailed me and the subject line of your email was, “Cass Sunstein says I should work for you.”
Robert Wiblin: That’s a strong subject line.
Tom Kalil: Good subject line. So I did a little research on her. It turned out that she had been a child violin prodigy with Itzhak Perlman, had won the major Yale undergraduate awards, was a Rhodes scholar, and was wrapping up a post-doc at Stanford in Decision Neuroscience. I went out on a limb and I decided to take a chance on her. Her name was Maya Shankar. I asked Maya, “What do you want to do?”
Tom Kalil: She said, “The UK has created this organization called the Behavioral Insights Team, which is taking these insights from people like Kahneman and Tversky and Sunstein and Thaler and using them to inform policies and programs. These are all US researchers. Why don’t we have something like this?” She said, “I would like to create that.”
Tom Kalil: Sure enough, in her late twenties, she arrived with no money, created this new organization called the Social and Behavioral Sciences Team, recruited 20 behavioral scientists to the federal government, got them to launch 60 collaborations with federal departments and agencies and got President Obama to sign an executive order institutionalizing this new entity.
Tom Kalil: I think that’s pretty consequential for someone in their late twenties to be able to accomplish. That’s one thing I did, was to recruit people of that caliber and teach them how to get things done in the federal government because the government doesn’t come with an operating manual.
Tom Kalil: I also launched a several dozen research initiatives. One you mentioned was the National Nanotechnology Initiative. That’s resulted in a $23 billion investment in nanoscale science and engineering. That has survived the transition from Clinton to Bush to Obama, to the current administration. In the Obama Administration, I worked on something called the Brain Initiative where the goal was to do for a neuroscience with the genome project had done for genetics. I was able to get DARPA prize authority, which they used for the self driving car competition. The second time they ran that a team won. Larry Page was at the finish line, he promptly acquired the winning team. That’s where the Google X division came from, was from that work on the self driving car. It’s also where Waymo came from. Then when I came back into government, I was able to get all agencies the authority to do incentive prizes for up to $50 million. If you’ve go to challenge.gov, you’ll see over 800 instances in which agencies have used this authority. Those are a couple of examples of the things that I’ve done.
Tom Kalil: But the thing that I’m most proud of, is the team that I built. Not only because of the impact that they had while they were there, but because I think many of them are going to return and engage in public service in the future. They learned some things about how to go from an idea to things happening in the world.
Robert Wiblin: So one of the biggest impacts was fostering young people to go off and have a great impact for their careers. It’s a little bit like 80,000 hours.
Tom Kalil: Yeah, absolutely.
Robert Wiblin: You get a lot of leverage potentially from that. Are there any products or businesses that you can point to that came out of, say, the Brain Initiative of an nano technology work? Like think of products that have reached fruition and had an impact?
Tom Kalil: Yeah. So the advances in Moore’s Law, the ability to double the number of transistors on an integrated circuit, increasingly rely on advances in nanoscale science and engineering. It’s having the biggest impact in areas like energy, particularly around energy storage, which is one of the big challenges that we need to address.
Tom Kalil: I have a story about the National Nanotechnology Initiative. In the late nineties I started interacting with scientists and engineers and said, “Well, if we decided to make this an area of focus, what are some of the things that might occur if we did that?”
Tom Kalil: They said things that were totally incomprehensible. They would say things like, “Well, we might be able to develop functionalized nano engineered MRI contrast agents. We might be able to develop a material where the Young’s modulus is of this many gigapascals. We might be-
Robert Wiblin: You were nodding along, pretending to understand.
Tom Kalil: Able to develop molecular electronics with a storage density of 10 to the 15th bits per cubic centimeter.”
Tom Kalil: What I did was I turned that into English, which was: we will be able to detect cancerous tumors before they’re visible to the human eye; store the equivalent of the Library of Congress in a device the size of a sugar cube; and make materials that are stronger than steel on a fraction of the weight. Armed with those examples, I was able to convince everyone, including ultimately the President, that we should make a bet in this area. For those people who are interested in science and technology, but don’t have a deeply technical background, one role that I’ve been able to play is just serving as an impedance match between the policy world and the scientific and technical world, because they’re generally not trained about how to communicate the importance of what they’re doing to a broader audience.
Robert Wiblin: Yeah. Okay. I’ll bring the other two questions I was gonna ask you. One is, is it just generally the case that science and tech people struggle to communicate with policymakers, ’cause they just thinking in a different way? And I guess, did you ever regret not having a more technical or scientific background yourself, having studied economics and political economy?
Tom Kalil: I have found the things, that there is a genuine value added that I can provide. Now the reason I’m able to do that is because I have a network of people who are deeply technical. Anytime I have a question like, “Is this a good idea or is this a crazy idea?” I have a network of people that I can talk to. But that there is a role that I can play, which is if I can understand it, then I’m generally capable of communicating it to someone who hasn’t spent … I’ve spent a good chunk of my career interacting with scientists and engineers. Most policymakers are not doing that. A value add that I can provide is the ability to talk to scientists and engineers, get the gist of what it is that they’re talking about, and then communicate that to a much broader policy audience.
Tom Kalil: Then the other thing is that- that there is a virtuous circle that kicks in with doing this. I’ll give you an example of this. In 1993, I was one of maybe a handful of people in the White House that was interacting with the computer networking research community. I was interacting with DARPA. DARPA had a very fast connection to the Internet that was say, a 100 to 200 times faster than a dial-up modem. I saw what Mosaic, which was the predecessor to Netscape, looked like, when you had a 45 megabit per second connection. It was immediately obvious to me that the ability to create this global information space that anyone could contribute to with all these hyperlinks and that anyone could use, was going to be a big deal.
Tom Kalil: The reason is that we were allowing the research community to live in the future. It was like going in a time machine and seeing what the world was going to be like 10 years later. Right?
Tom Kalil: And then, I could do the information arbitrage of going around and showing everyone in the White House what this was going to be like and why it was important. And so, for most people, the first time they heard about the Internet was from me. And so, because in 1993 it was like-
Robert Wiblin: It was not going to be on people’s radars as much.
Tom Kalil: It’s not on people’s radar screen. Unless you are a computer scientist or something like that, it’s probably not something that you were familiar with. So after that, when I said nanoscale science and engineering is going to be important, people are like, well, he was right about that Internet thing so maybe he’s right about this as well.
Robert Wiblin: I think it feels to me like grant makers and I guess anyone including me, whenever you hear about a new thing that you could fund in science and technology or something you can go and work on, there’s always a question is this bullshit or is this something that can happen. And I guess, how do you know who to trust? Because kind of even if someone’s trying to be very honest scientist, then it’s they’re so invested in their own thing that it’s very hard to know whether they’re hyping it beyond what’s technically practical.
Tom Kalil: I think that the important thing to do is to be taught to talk to more than one person. Right? So, for example, in late 2011 and 2012 when people brought me the idea that ultimately became the Brain Initiative, my initial reaction wasn’t: “great! Let’s announce this tomorrow”. It was to do due diligence on the idea by asking lots of other people, is it the right time to work on this? Right? Have various technologies from new materials to new types of imaging technologies advanced to the point where this is actually doable, or is this radically premature? And, if we were successful, how big a deal would this be? Is this a single or a home run. And then also, a lot of times, smart people disagree and you have to try to figure out what’s driving that disagreement. And then, if you’re the government, sometimes you might take a portfolio approach, right? So, if there’s a goal that you think is really important. As an individual researcher, you may care intensely about which path is the right way to do that because you have to stake your career on that. But, as a policy maker, you can be like-
Robert Wiblin: Give everyone a little bit of money.
Tom Kalil: Yeah, let’s support both and see which one works out.
Robert Wiblin: So, you’re like moderately technical, I guess. And, you’ve spoken to a lot of scientists and you’re trying to figure out whether something’s worth investing in. You speak to different experts and they maybe disagree. There’s a bit of controversy here. At that point you’re just like, well, maybe we’ll fund a bunch of different things, we’ll take a portfolio approach. Have you found any rules of thumb for figuring out who to trust more?
Tom Kalil: I think part of it is looking at someone’s track record. So, there are some people who they’ve just consistently delivered, number one. And number two, they do not have a track record of overhyping their work. So, if they say, I think that over the next five to ten years it will be possible to do X, you’ve just attach more of a weight to a statement like that than someone who has a track record of over hyping and under delivering. So, definitely someone’s track record. And then, the other is just ensuring that you’re talking to lots of people and not just one charismatic individual who may have an incentive to overhype their future research directions.
Tom Kalil: The other thing is also what’s the level at which you’re making a bet. So, for example, the idea of the National Nanotechnology Initiative, the bet that we were making was that at the nanoscale, new and potentially useful properties emerge. So, it’s like adding another dimension to the periodic table of elements. We did not say that carbon nanotubes is going to be the answer or buckyballs. We should put all our money in buckyballs. It was just there’s this new class of materials. They have new and potentially useful properties. If we can make nanoscale materials, devices, and structures, that’s going to have a broad range of applications.
Tom Kalil: So, similarly with the Brain Initiative, we didn’t say this is the way in which we’re going to figure out how the brain encodes and processes information. We were making the following bet informed by the research community, which is that if you look at the scientific revolution, what triggered it? It was advances in things like telescopes and microscopes that allowed us to see new things that we didn’t see before and allowed people to say, hey, Ptolemy and Aristotle, they got some things wrong. So really, if you look at say, what’s the driver of the scientific revolution? It’s the ability to see things that we couldn’t see before.
Tom Kalil: Similarly, the view in the research community was that neuroscience was tool limited. We can measure a very small number of neurons with high levels of temporal and spatial resolution, or we can take a fuzzy picture of your entire brain using something like fMRI, but we can’t do anything in the middle. And so, the hypothesis was: we could develop a set of tools that would allow us to ask and answer new questions about how the brain and encodes and processes information. So, part of it is also what’s something that’s big enough to matter, but it’s not so focused on a particular pathway that it’s the equivalent of going to Vegas and saying, I’m going to bet all my money on 12 and then if 12 doesn’t come up then I’ve lost all the money.
Robert Wiblin: Are there any things that you’ve helped to fund where in retrospect you think, oh damn, I should’ve realized that was not going to work or is that not how things function?
Tom Kalil: There were definitely things like that. So, for example, one of the things I learned is that if the United States is behind in a technology, it’s very difficult to try to re-establish a leadership position.
Robert Wiblin: Interesting.
Tom Kalil: Yeah.
Robert Wiblin: So, it’s better just to look at something else.
Tom Kalil: We tried to do that in the area of technologies like flat panel displays and we invested some money, but I don’t think a whole lot came out of it.
Robert Wiblin: So, is it like once there’s one city or one region that has really big agglomeration effects where everyone’s going to-
Tom Kalil: Once Korea and Japan dominated the market for things like active matrix, liquid crystal displays, then trying to get the United States back into that market is really, really hard, and might require more money [crosstalk 00:41:43] than the US is willing to put into it. Because obviously we believe that the primary role of government is to create the right environment for the private sector. It’s not to engage in this sort of heavy handed-top down industrial policy that you see a China engaging in, for example.
Robert Wiblin: So, I guess then you’d think, well, instead we should look at what’s the next thing going to be where we can potentially jump ahead?
Tom Kalil: Yeah. So, for example, we invested in this idea of flexible electronics where the idea is – maybe you have a display that’s a piece of paper that you can roll up and put into your pocket. And, if that’s an area where no one has established a clear leadership position, that’s more likely to be effective than saying, okay, we’re going to duke it out in some market that we’ve kind of already lost.
Robert Wiblin: You’ve already alluded to the fact that the administrations change every four or six years, well I guess occasionally it’s 12, and so you have to be thinking, well what can we do to potentially keep this policy around, to sustain it over a long term? What things do you do to make sure the kind of projects you make are built to last and how often do you succeed at that?
Tom Kalil: First question is you have to be sure that this is something that you really think needs to be sustained over a long term. So, for example, there might be some policies where you’re like, hey, there’s a really important role for the government for the first five to 10 years. But, that doesn’t mean that the government needs to be involved indefinitely. The technology might get to the point where it’s really time for the private sector to pick it up and run with it.
Tom Kalil: In the same way that the government began investing in the Internet in 1969 and eventually at some point the private sector, [crosstalk 00:43:14] beginning in the 1990’s, there was an explicit policy by the government of, hey, it’s time for the private sector to commercialize this and offer it as a commercial product or service. And now, we’re going to think about what’s after the Internet. So, we started at that point investing in something called the Next Generation Internet, which was an internet that was a thousand times faster than what was available through commercial products and services.
Tom Kalil: But, to your question, the thought experiment that you engage in is all of the people who are affiliated with the current administration walk out the door, and then what happens the day after? So, that’s the thought experiment that you have to do. And so, one of the things that you can do is you can ensure that there is a group of civil servants who genuinely thinks this is a good idea and will continue to work on this unless they’re explicitly told by the [crosstalk 00:44:10] incoming administration to stop.
Tom Kalil: So, for example, there have been a number of areas in environment and climate policy where the current administration has instructed the departments and agencies like EPA, I want you to reverse the Obama Administration’s policy on this, right? But, there are other areas like nanoscale science and engineering where in the absence of a direct instruction that all the agencies that are involved in this will continue [crosstalk 00:44:47] to work on it because they think it’s the right thing to do.
Tom Kalil: So, civil service buy-in, I think is really important, and it’s also very useful if Congress passes a law. So, there were some policies that we worked on, which they were done through regulation as opposed to through law. And, one of the things that makes those more fragile is if they can be done through regulation, they can also be reversed through regulation.
Robert Wiblin: Yeah. Legislation is much harder to change.
Tom Kalil: Legislation is much harder to change. So, for example, as I said, I was able to get a DARPA prize authority working for President Clinton and then every agency prize authority. And, agencies have continued to use this authority. Whereas if it had just been done through executive order, an incoming administration said, well we don’t like prizes, then they could just repeal the executive order. And, that’s a lot easier than changing the law.
Robert Wiblin: Yeah. Looking at the big picture of kind of US science and technology or innovation policy, what things could be changed that would allow the US to get significantly more R&D done, or to increase GDP growth or increase scientific advancement?
Tom Kalil: Yeah. So, as I said earlier, one of the things I’m really interested in is what problems do we have where science and technology could make a difference. So, right now our defacto answer is that we can use science and technology to advance the frontiers of human knowledge about ourselves and the world around us, to help with national security and intelligence, energy, space, and health. And, we have lots of other problems. And, for those problems, for one reason or another, we have not made a decision as a country, oh, let’s use science and technology to help solve that problem.
Tom Kalil: So, I think a really interesting thought experiment would be to take an agency that works on a really important problem, say K through 12 education. And say, what if that agency had the same capacity to mobilize the research community that DARPA did? Number one, what goals would it set? And number two, what are examples of projects that it would support in order to achieve those goals? So, right now in K through 12 education, productivity is negative. What I mean by that is we’ve doubled real per pupil expenditures without any corresponding significant improvement in student learning outcomes.
Tom Kalil: So, I believe that if we had a research agency that had the sort of same capability to mobilize the scientific and technical community that say a DARPA has, that there are things we could be doing to help address that problem. So, an example is DARPA supported this project and the goal was to reduce the time for a new Navy recruit to earn a technical skill and be able to do that in months rather than years. And, that’s developed some really dramatic results in terms of training. And so, you could imagine applying that approach to the training problem and to the education problem.
Tom Kalil: Similarly, a big problem that we have in the United States is the intergenerational transmission of poverty. So, we like to say that we have equality of opportunity, but we know that by the time kids show up at kindergarten, there are already large differences in school readiness, brain development, and vocabulary size. And, our schools do not narrow those divides. So, if you have low income parents, the odds of you struggling in school and struggling to make the transition from learning to read, to reading to learn, to graduating or going on to college. So, even though we talk about equality of opportunity, there are large differences in people’s life chances based on their parents, or what’s going on at home.
Tom Kalil: And, my view is not science and technology will solve all those problems. My view is if you are going to come up with a list of things to try, science and technology would be one of the things on that list. So, the thing that I’m really interested in is when we think about science and technology, by which I don’t just mean the natural sciences and engineering, I’m also talking about the social and behavioral sciences. What are the goals that we would like to use science and technology to help us achieve in the same way that we believe that we need science and technology to avoid technological surprise for the United States and to create it for our adversaries.
Robert Wiblin: You mentioned prizes before. I guess there’s different ways of trying to fund science. One is grants where you try to guess what’s going to work and you’ve got patents, I guess, as a market driven mechanism, and you’ve got prizes, which I think you’ve kind of been championing as perhaps something that the government should put more money into. Do you want to talk about that?
Tom Kalil: Yeah, absolutely. So, one of the things that I think is really perverse is that right now the federal government routinely will make a financial commitment that is contingent on failure. So, what do I mean by that? That is if you were to look at the finances of the United States government, you would find roughly $2.6 trillion in loan guarantees. So, that is the government saying if this individual or organization goes bankrupt, Uncle Sam as good for it. So, that is a financial commitment that is contingent on failure.
Tom Kalil: We are just barely scratching the surface on our ability to make financial commitments that are contingent on success. And, let me give you some examples of how when we do do that, it can have a large impact. So, there is a market failure associated with vaccines for diseases of the poor. Left to their own devices, drug companies won’t work on vaccines for poor people because poor people have no money. So Michael Kremer, I think you interviewed Michael’s wife-
Robert Wiblin: Oh yeah.
Tom Kalil: … on 80,000 Hours. So, Michael and Rachel wrote a book about this, but they came up with this idea for addressing this market failure, which is called an advanced market commitment. So, as a result of their work, five countries and the Gates foundation went to GSK and Pfizer and said, if you develop a vaccine for this disease, which is safe and effective, then we commit that we will buy X million doses at seven dollars a dose. And, that one intervention was enough to eliminate the 10 to 15 year gap that usually exists between when a vaccine is available in wealthy countries and when it’s available in low income countries. So, it’s a financial commitment that is contingent on success. So, it’s the Gates Foundation and five countries saying if you develop a vaccine which is safe and effective, then we’ll buy it.
Tom Kalil: So, the global health community has developed the capacity to do three things. Number one is identify a specific unmet need. The second is the ability to develop a performance based specification for what a successful innovation would look like. In this case, a vaccine that works and is safe. And the third is, if there’s a market failure, if there’s a huge gap between the social return and the private return to create some incentive that motivates the private sector to solve this problem.
Tom Kalil: So, my view is we should be doing more of this, right? Why is it the case that we have $2.6 trillion in financial commitments that are contingent on failure when we barely have any financial commitments that are contingent on success? So, that’s my view. And, these come in a variety of different flavors. There’s incentive prizes. So, that’s like the Ansari X prize saying if you develop a rocket that can go up a hundred kilometers, carry the equivalent of three people, repeat that within a two week period and do that without any backing from the government, then we’ll give you $10 million. So, that’s an incentive prize.
Tom Kalil: The advance market commitment or an advanced purchase commitment is if you develop X, then we will buy it. And then, there are other approaches like milestone payments. So, this is how the NASA collaboration with SpaceX worked. So, when we retired the Space Shuttle, the United States Government had to start spending a lot of money with the Russians in order to send astronauts to the International Space Station. So, they developed a collaboration with private sector companies like SpaceX, and they said, we want a rocket that will go up to the Space Station, deliver and retrieved cargo, and ultimately astronauts. And, we’re going to define a set of milestone payments that’s associated with getting there. And, every time you meet one of them, we will send you a check. And so, NASA got a capability on the order of $400 million that would have cost them two to four billion dollars using a business as usual approach. And, that’s what allowed Elon to finance the development of the Falcon 9 rocket. So, that was a win-win between the government and the private sector.
Robert Wiblin: Yeah, I think Alex Tabarrok, the economist has written a book, I think Creating the Innovation Renaissance, something like that. We’ll stick up a link to it where he talks about some of the pros and cons of prizes versus other funding mechanisms. Why hasn’t it taken off more? I guess one concern that I might have is that the government wouldn’t know how much it should offer or what’s an appropriate priced offer so it could end up accidentally giving away much more for something that’s too easy. And, maybe the political economy of it is bad because it gets taken over by businesses that are going to try to milk this system of prizes to get money for things they were going to do anyway. What kind of concerns do you think we will have that mean that prizes are just currently not a big funding mechanism for research?
Tom Kalil: Yeah. Well, certainly they’re not the appropriate way to support science in all instances. So, it works if you have a clear goal. So, if you know kind of the general direction that you want to go in, but you can’t articulate a clear finish line, then it makes a lot less sense to use a prize mechanism. So, in many cases, a researcher will know the question that they’re trying to answer, but they won’t really know what success looks like. So, that’s one thing is that is that it’s only appropriate for certain classes of problems where there’s a much clearer finish line. The ideal situation in the prize is, you almost don’t need judges, right?
Tom Kalil: So, if you think about the Ansari X prize, it was like, did the rocket go up a hundred kilometers? Was it carrying the equivalent of three people? Was it able to repeat that with an a two week period? So, you don’t need a panel of experts to decide, well did they actually do all those things? So, that’s sort of like the ideal case as opposed to-
Robert Wiblin: It’s more discretionary.
Tom Kalil: … something that’s more-
Robert Wiblin: Fuzzy.
Tom Kalil: … subjective. Right? It’s like, oh, is Rob asking an interesting question? Right? That we rely on peer review in order to be able to answer. So, it’s a lot more work upfront in order to do this as opposed to saying, well, I’m generally interested in ideas in this broad area as opposed to I can define what success looks like.
Robert Wiblin: I guess, is it possible that some research groups or businesses might find it hard to fund the money, if the money only comes later rather than beforehand.
Tom Kalil: Yeah. So, that’s why I’m interested in milestone payments because if someone had gone and said, okay, we want a rocket, and once you’ve developed it, then we’ll buy rides on the rocket. Particularly if it’s a new entrant, the fact that the government is saying, well I’ll buy it once you make it, might not allow them to be able to finance that.
Robert Wiblin: Are there any common themes of the ideas or the policies you’ve pursued that have succeeded versus the ones that haven’t panned out, or that failed later on?
Tom Kalil: That’s a good question. I don’t know that there is a single theme that connects all of them. I think that in terms of the role that I played, it is to bridge different worlds, right? So, there’s a policy world and there’s a science, technology, innovation, entrepreneurship world. And, by participating in both of those, I can help move ideas that are in one realm or community to another. So, that’s something that I’ve done over and over again. Or, when I found out about how the global health community was thinking about solving its market failures, I thought, this has broader relevance. It’s not just relevant to global health, it’s relevant to lots of other areas where there’s a market failure, and particularly when the innovation in question has high fixed costs and low marginal costs. So, part of it is just you encounter particular situations, and you discover that there’s sort of this underlying pattern language of approaches to solving a particular problem that can be applied in some other policy domain?
Robert Wiblin: It seems like a common thread of the other work that you’ve done and your writing is to focus on kind of these moonshots, these grand challenges, what are the things that are going to be really transform the world? Why focus on that rather than more incremental improvements. And, do you think, it’s important to have people work on incremental improvements, you just want to focus on the big picture?
Tom Kalil: Absolutely. Yeah. No, no, no. So, I mean, if you said, why has the world gotten better since 1820, it’s things getting better two percent per year, right? That has a huge impact over a long period of time. I think that part of it is that I think that setting these ambitious goals is important. And, I also think that it helps get people excited about investing in science and technology. Now you can take that too far, right? So, you can overpromise and underdeliver. So, I do think that there are some risks if you do that in an irresponsible way.
Tom Kalil: But, I have found it a way to get people excited about investing in science and technology more broadly or just focusing on particular problems. So, the UK for example, has done this report on the longterm economic consequences of living in a post penicillin world and it’s $100 trillion, right? So, I think people should be jumping up and down saying, hey, we should be doing more in this area. So, I do think that there are some areas where we’re underinvesting relative to the importance of the problem.
Tom Kalil: And so, I think that what you need to do is to go from … the experience of policy makers is they don’t have a shortage of problems. So, just showing up and saying, oh, and here are five more things for you to worry about. That’s not very effective. Right? And so, what you need to do is to say, here’s the problem and here’s what you could do about it. Right? So, what I find to be more effective is if I’m able to articulate both the problem and the solution because their experience is they have a list of problems as long as your arm, and just dumping another one on them, they’re like, gee, thanks. Well, what the hell do you want me to do about that?
Robert Wiblin: You said that what potentially the most valuable thing you were doing at the OSTP was recruiting the right people, was hiring. I suppose it’s feverish competition, obviously, over the best people. So, what could you do to get an edge in either figuring out who the best people were, where other people hadn’t realized how talented they were, or in convincing them to work at the White House where otherwise they wanted to do something else?
Tom Kalil: A lot of it was about me being able to be very specific and granular about what I had accomplished and what other similarly situated people had accomplished. So, obviously, I couldn’t compete on the basis of salary or the quality of our cappuccino machine. It was, hey, you’re going to be in the White House overseeing this federal government that allocates $4 trillion a year, and there’s no shortage of opportunities to make improvements relative to the status quo. And, here are all the tools that you have.
Tom Kalil: You can convince the President that this is an area that we should invest more money in. You can work with Congress to pass laws. You can identify things that federal departments and agencies can do through executive action, and you can leverage the President’s convening power to build coalitions, not only of government agencies, but of companies, research universities, philanthropists and foundations, nonprofit organizations, in order to achieve a particular goal. So, if there is something that you’re excited about doing, you will have access to tools in order to advance that agenda that you wouldn’t have anywhere else. So, it’s giving them, what some people call definite optimism, not just like general optimism of, oh, I think things will get better, of oh, there’s something concrete-
Robert Wiblin: Specific.
Tom Kalil: … that I can do to make things better.
Robert Wiblin: When people were skeptical, or kind of reluctant to come work for you, what were their reasons?
Tom Kalil: Well, I think that if your perception of the government is based on reading the press, what do you read about? You read about scandal, you read about gridlock, right? And, it’s not like those things are incorrect, but it would be like if your perception of New York were based on only reading the crime pages. I mean, say yes, those crimes do occur, but New York also has-
Robert Wiblin: Restaurants.
Tom Kalil: Restaurants.
Robert Wiblin: Parks.
Tom Kalil: Symphonies and amazing art museums. And so similarly, it is true that, particularly when you have divided government, it’s very difficult to pass legislation. And so, there’s a lot of things that we need to do that really require passing legislation. And, the fact that we’re not able to do that is a bad thing. So, I do not discount gridlock, partisan gridlock, as a problem. But, I have a different view that is based on, even subject to those constraints, things that I’ve been able to accomplish that I think are important. And so, I try to give that perspective as well.
Robert Wiblin: Yeah. So, do you think people underappreciate how much the executive branch can just do autonomously?
Tom Kalil: Yes. Yeah. Not only what it can do, but the President’s ability to convene.
Robert Wiblin: Explain that.
Tom Kalil: Sure. So, if the President invites you to a meeting, or even if someone who just works for the President like me, many people will show up. And, there are several different types of meetings that you could have. So, one would just be, hey, we think that this issue X is important. And so, we want to have a conversation about what are the best possible ideas to make progress on the issue X. So, that would be a way of trying to understand the issue, understanding potential courses of action. But, we might also have a meeting where the purpose is to serve as an artificial deadline. And, we would use the time between now and when the event was going to occur to ask people what it is that they might be prepared to do.
Tom Kalil: So, for example, the President said, in the same way that if you win the Superbowl or the NCAA, you get to come to the White House, the same thing should be true if you win a science fair, or robotics competition. So, he wanted to sort of increase the prestige and status of STEM education, and get more young boys and girls excited about STEM. And so, we would use that event, not just for the President to interact with these kids who are doing these amazing robotics competitions and science fair projects, but to mobilize the entire country to make specific commitments that would advance the ball.
Tom Kalil: So, the question that I would ask people is, if you were the President and you could call anyone and ask them to do something, who would you call and what would you ask them to do? And, sometimes there are organizations that are not part of the government, but are, particularly if they work together, are really in a position to move the ball forward on a particular national issue.
Robert Wiblin: Yeah. So, I guess I would have been skeptical as well that this power, at least with businesses, was very effective because you can imagine, you call in the CEO’s of various companies, and they come and play nice and they’re like, oh yes, we’re very interested in your priorities. Maybe because they’re scared of the government apart from anything else. But then, they go back and they do what they were going to do anyway because their responsibility is to the shareholders, and so they don’t have so much discretion to decide things based on what the President cares about.
Tom Kalil: Yeah. So, I think that if you asked them to do something that is contrary to the interests of their shareholders, you’re not going to have a lot of luck with that. But, for example, one of the issues that we were dealing with was the imbalance between demand and supply for technical workers, particularly jobs that you could get that didn’t require a four year degree, but you could get through some sort of accelerated training program. And, the status quo was within a given region, the companies were just stealing their workers from someone else. And so, we got them together and said, hey, what if as opposed to doing that, you grew the pie, and tried to figure out what sort of relationships you would need to have with various training providers so that there would be more workers with these skills as opposed to there’s this limited pool and what you’re doing is poaching someone else’s workers. Yeah, that makes sense. Right?
Tom Kalil: So, I think you’re right in the sense that if we said, we want you to take some action that is going to lower your stock price by 20 percent, they would have said no. Right?
Robert Wiblin: Go jump off a bridge, yeah.
Tom Kalil: Right, Or, hey, if you want to do that then you’re going to have to make that a law. Right? So, we don’t make it optional about whether or not you get to drive on the right hand side of the road. That’s just a law. And so, there are some instances where if you want people to do things, you have to pass a law. There are other instances where you can get people doing things by asking them, number one. And number two, putting it in the context of a broader social movement around a particular issue.
Robert Wiblin: And, do you think people do that because you’ve brought to their attention something that might’ve been sensible for them to do anyway, you just made it more salient? Or, perhaps is it they just had to buy into this vision and they’re like, actually this would be really cool, and this is something that this isn’t too costly, so I’ll happy for my business to be a part of it.
Tom Kalil: I think a part of it though is that you’re putting it in a broader context. So, a lot of times when we would talk to individuals, they would say, Tom, I feel like I am trying to put out a forest fire with an eyedropper. Right? So, I feel like my individual contribution relative to the scope of the problem doesn’t seem all that significant. But, if you were getting together all of the stakeholders, and we’re defining a set of mutually reinforcing steps that different sectors can take, then the piece that I’m able to contribute feels a lot more meaningful. So, that’s the thing that we can do. Because a lot of problems can’t be solved by an individual or a sector.
Tom Kalil: So, if we can say, here’s what government is going to do at the federal level, here’s what state and local government is going to do, here’s what foundations and nonprofits can do, here’s what the private sector can do, here’s what skilled volunteers can do, then everyone’s role feels more meaningful. Think about a campaign. The fact that you can only knock on a hundred doors, you’re like, well, what impact can that have? But, the reason it feels more meaningful is that-
Robert Wiblin: You know other people-
Tom Kalil: … you know other people are doing their part as well.
Robert Wiblin: You motivate people by coordinating them, by getting them all together and saying, well, if everyone else would do this, then it’s useful for me to do my part.
Tom Kalil: Yes.
Robert Wiblin: Yeah. Interesting. So, taking a step back to the hiring thing, were there any particular criteria you look for in hires when you’re interviewing people or looking at their [crosstalk 00:33:57]?
Tom Kalil: I was always more interested in people who had at least an initial theory of what they wanted to accomplish, that they were intrinsically motivated by. So, I put a lot of weight on intrinsic motivation because generally I find if someone really wants to achieve something, they’ll figure out how to do it. Whereas if they’re just responding to extrinsic motivation, they’re less likely to run through walls in order to achieve the goal.
Robert Wiblin: Interesting. So, typically people came in with a project that they were already interested in rather than arriving on day one and your being like, oh, what do I do?
Tom Kalil: Yeah, exactly.
Robert Wiblin: Interesting. So, I guess that means there’s pretty regular changeover on projects as people come in and out.
Tom Kalil: Yes, that’s right. Yeah. But, there were instances where I said I want to achieve X and therefore I want to find out the best person who can do that. So, for example, I knew I wanted to do more in incentive prizes. And so, after we passed legislation that gave every agency prize authority, I specifically looked for who can help lead a community in practice within the federal government so that more departments and agencies will actually start using this new authority. So, there were instances where I saw specific opportunity and then I mapped backwards from that, like, who could work out that.
Robert Wiblin: Is that a common situation, that if you want to get hired into government it’s important to have an area of passion, a policy that you really care about?
Tom Kalil: I think that’s helpful, yeah. I think that there are other instances where someone is just looking for a utility infielder. A lot of times someone will say, “I want someone who is really smart, hard working, has common sense, good people skills, the ability to pick up a new area quickly, the ability to write effectively”, so a set of skills that are useful in lots of different contexts. Then, they’ll just be able to catch whatever I throw at them and they don’t have to be an expert in a particular area.
Robert Wiblin: Are there other things you learned about hiring over 16 years in the White House so far?
Tom Kalil: I didn’t bat a thousand.
Robert Wiblin: Are there any indicators of people who didn’t work out?
Tom Kalil: Well, I guess some of it is learning to trust your instincts. If you have a small voice in your head saying, “I’m not sure about this”, it may be your intuition is trying to tell you something. That’s one thing, I think the other thing is, unfortunately sometimes you can really never tell until someone is in this environment the extent to which they’re gonna succeed or fail, so one of the things I benefited from is that a lot of these appointments were short term in nature and then if I really like the person I would work on getting it renewed, but the default was, they would time out after a year, or something like that. So that limited some of the downside associated with some of the decisions I made that I wish I hadn’t in retrospect.
Robert Wiblin: You often use this term, policy entrepreneurs and policy entrepreneurship.
Tom Kalil: Yes.
Robert Wiblin: Do you just want to explain why you think that’s an important concept and what that is?
Tom Kalil: Yes, I think a lot of times when people think about the government, maybe they have something in mind, like “Yes, Minister”, where the people in the government are trying to prevent anything from happening. Did you ever watch that show?
Robert Wiblin: I’ve watched all of it, yeah. I just say, looking at the UK at the moment, I think Yes, Minister would be a blessing if the government worked that well.
Tom Kalil: Right, right, right. No, but like, Sir Humphry is always trying to figure out how to maintain the status quo. Right?
Robert Wiblin: Exactly.
Tom Kalil: I do think that there are people whose … my main motivation for being in the government, if you had said, “Oh, your job is just to preside over the status quo”, I would’ve left after three months.
Robert Wiblin: Right, yeah.
Tom Kalil: That’s not … I wouldn’t find that intellectually engaging or worthwhile, so what I was always interested in was, what were the new ideas that were bubbling up from the research community or entrepreneurs that I could get behind and champion. The policy entrepreneur, in the same way that entrepreneur is identifying some unmet need in the marketplace and then trying to produce some product or service that addresses that unmet need through this process of opportunity recognition, the policy entrepreneur is identifying an opportunity for a new or improved policy that he or she thinks will advance the ball in some policy domain, and then is trying to figure out, what is the coalition that I would need to build in order to make that happen, and then actually going off and doing that.
Tom Kalil: So, you might have a couple different types of policy entrepreneurs. One is, someone who’s passionate about a particular issue. There was a woman in my staff, Jennifer Ericson, who was really passionate about reducing the waiting list for an organ transplant, kidneys in particular. She was super passionate about that issue and it’s be like midnight, or something like that, I’d say, “Jennifer, you gotta go home and get some sleep”. She was super fired up about working on that, so you have people like that, and then you have people like myself who are open to a pretty broad range of ideas. I view my value as, not that I’ve specialized in any one particular issue, but I’m curious about lots of different things, interested in lots of different things, therefore, like being in this situation where I’m getting a continuous flow of high quality ideas and then picking some of them and trying to figure out how many of them I can get to happen, as opposed to saying, “By the time I retire I want to have accomplished X”.
Robert Wiblin: Do you have any comments on the opposite of policy entrepreneurs? The Humphreys of the world who just want to kind of … there’s a role for those as well, people who try to stop things from being destroyed that are already working.
Tom Kalil: Well, yeah, so I think that it’s not always the case that change is a good thing. Right? I think that it’s important for there to be a productive relationship between political appointees like myself and civil servants because the civil servants: A- they’re the ones who implement things; B- they have a lot of institutional memory, so they can say, “Oh, we tried that before and here is what happened”. They’re also the ones who will be there when you leave, in many instances. Again, if you want a policy that persists, you want them to think that it’s a genuinely good idea, as opposed to something that we’re doing as long as you’re watching them.
Robert Wiblin: Yeah, that makes sense. What do you think people most misunderstand about policy entrepreneurship? One thing is that they just might not think of the policy world as being as dynamic as that.
Tom Kalil: Yeah.
Robert Wiblin: But what do they misunderstand about the process?
Tom Kalil: Well, I think a number of things. One is thinking that when the policy is announced the work is done, so one of the things I discovered is that that was maybe a third of the way through and that just because the president had signed an executive order or presidential memorandum, or Congress had passed a law, that you had to devote equal, if not more attention to implementation. I think people’s view was, “Well, if the President tells an agency to do something, of course they’re gonna do it.”
Tom Kalil: It’s like, –
Robert Wiblin: Maybe not.
Tom Kalil: Particularly as you are in the middle or towards the end of an administration an agency can rightly say-
Robert Wiblin: Maybe we’ll work this out.
Tom Kalil: Yeah, or just “Hey, the President has signed so many executive orders at this point that I can’t possibly do all of these and so therefore in the absence of some guidance I’m gonna figure out which are the ones that I think are the most doable.”
Robert Wiblin: What actually happens there? If the White House sends executive orders to agencies and they’re like, “We can’t do all these things, we just don’t have the resources to do everything we’re being asked”, do they prioritize themselves? Is there any recourse when it’s like, in a sense they’re not doing what they’re told to do?
Tom Kalil: Then what happens is that if the White House really cares, if the President really cares then there’ll be really frank discussions about some sort of down selection of which of these … it’s appropriate for an agency to say, “You’ve given me all these things to do, I can’t possibly do all of them.” I think you see that particularly towards the end of an administration. We were told, “Look, this particular part of the federal government, they’re gonna be focused on these three things and if you want them to do something else, “too bad”, or, “you’re gonna have to take it up with the Chief of Staff” because there’s been an agreement about what they’re gonna focus on and that means the only way for you to have a priority is to say you’re not gonna do certain other things. Like, if everything’s a priority, nothing’s a priority.
Robert Wiblin: How common would it be for there’d be a policy announced, it would go out to agencies and they would just kind of … it doesn’t really go anywhere ’cause there’s no one dedicated to it and you have to keep following up, calling every day?
Tom Kalil: Basically what you would see is a range of agency responses, some agencies taking really seriously and doing something meaningful, and other agencies sort of checking the box and engaging in the minimum level of compliance that they thought they could get away with.
Robert Wiblin: Is that ’cause they disagree about the policy sometimes, or just that they’re overwhelmed-
Tom Kalil: It could be that they’re overwhelmed, could be that they were never all that excited about the idea to begin with. Also, they might not have the capacity. For example, the President was more excited about STEM than his department of Education was so there was a large imbalance between the level of personal attention that he gave this issue and the extent to which the department was responsive to that.
Robert Wiblin: That means you kind of have to keep following up to-
Tom Kalil: Yeah.
Robert Wiblin: Keep reminding them.
Tom Kalil: Yes.
Robert Wiblin: Are there any other things that people misunderstand about policy entrepreneurship?
Tom Kalil: I think they believe that if you have divided government between the Congress and the Executive Branch then nothing gets done.
Robert Wiblin: Yeah, but actually lots of things going on.
Tom Kalil: Lots of things going on.
Robert Wiblin: Just without legislation being passed.
Tom Kalil: Yeah. The government is large and it has no shortage of legal authorities to do things and there are areas where it has significant amounts of discretion and it has the ability to convene coalitions of willing and able in order to achieve a particular goal.
Robert Wiblin: You’ve written this great article called, Policy Entrepreneurship at the White House, which is kind of lessons from your time in government. We’ll stick up a link to that so people can explore it in full if they’re interested, but briefly, what are one or two rules of thumb that you can give with working with big organizations?
Tom Kalil: One thing that I used to ask people is to imagine that you have a 15 minute meeting with the President in the Oval Office and he says, “Rob, if you give me a good idea for”, pick your cause, “reducing existential risk, then I will call anyone on the planet. It can be a conference call so there can be more than one person on the line. If it’s someone from inside the government, that I can direct them to something because I’m their boss, and if it’s someone outside the government then I can challenge them to do something. So, you not only have to tell me, what is your idea, but in order to make your idea happen, who would I call and what would I ask them to do?”
Tom Kalil: There are several reasons for this thought experiment. One is that if you work for the President you have the ability to send the President a decision memo and have him check the box that says yes. Over time that give you a sense of what psychologists call agency, a sense that many things that you see in the world around you are the result of human action or inaction, as opposed to the laws of physics. That’s one thing, a more expansive view of what do you think is potentially changeable. The second is, it’s sort of a version of the Hamming question, presumably if you really did have a meeting with the President you’d use it to describe an issue that you thought was really important as opposed to a secondary or third tier issue. The third is that many complex problems cannot be solved by a single individual organization, they require coalitions.
Tom Kalil: You can’t build a coalition if you can’t articulate, number one, who are the members of the coalition, and number two, what are the mutually reinforcing steps that you would want them to take. That’s one thing that I talk about in the Policy Entrepreneurship, then I also talk about something that people don’t ever really appreciate, which is that policy makers do things with words. What do I men by that? Well, think about when the priest says, “I now pronounce you man and wife”, he has changed the state of affairs by virtue of, A, him being a priest, and B, him saying, “I now pronounce you man and wife”. Similarly, the way that a policy maker both frames and makes a decision and implements that decision is through documents. When the President does it we call it an executive order or presidential memorandum, when a regulatory agency does it we call it a rule, when the Congress does it we call it legislation. But in all instances it is a document that you are creating or editing, so part of the policy process is that you are able to figure out what’s the document or documents that you need to create or edit and who is allowed to take that something from being a Word document that is on your screen to something that has some force in the world?
Tom Kalil: I would see this all the time, something would go from being a Word document on my computer to being a presidential executive order, it always seemed like this slightly magical transformation from a Word Doc to something that is instructing relevant members of the Cabinet to take some action.
Robert Wiblin: I guess this makes you more ambitious, then you’re like, “What is the best thing, what is the best memo-
Tom Kalil: Yeah, exactly.
Robert Wiblin: … that I could write.
Tom Kalil: Yeah, exactly, yeah. But also you have to be able to articulate some coherent relationship between ends and means. I would … a lot of times someone would come visit me and they would say, “my issue is important”.
Tom Kalil: I’d say, “great, let’s say that I’m prepared to stipulate that, what is that you want me to do?”, then they would look at me and they would say, “You should make this a priority”.
Tom Kalil: I’d say, “What would that look like?” People were not able … they were able to tell you that their issue was important and that they thought the President should devote more time and energy to it, but then when you said — “Alright, what is it, literally … let’s say we got the President super interested in this issue, what would they do?” They weren’t able to articulate that part.
Robert Wiblin: Yeah, I’m sympathetic to that ’cause there’s a lot of things that I think are very important, but I’m also not sure what should be done. I suppose maybe it makes sense for people to think more about that once they’ve gotten people to care about it, but at the same time, maybe it’s hard to get people to care about something if you have no actual concrete steps that they can take, they’re like, “Well, I don’t know what to do”.
Tom Kalil: Yes. Yeah, because that’s assuming that … because what you’re saying is, I’ve really thought about this issue a lot and I think it’s really important, but I don’t know what to do.”
Robert Wiblin: That’s a bad sign. So you should think about it.
Tom Kalil: Right?
Robert Wiblin: What were some of the aphorisms you had on the white board in the White House?
Tom Kalil: One was, the schedule is your friend. What that means is that things can take a while for things to happen in the government, to get done, but if you were able to land an event in which the President was gonna participate, then everyone had to meet your deadline. For example, if the President said, two weeks from now I’m gonna do an event on the brain initiative, then you’re allowed to go around and say, this needs to get done because it’s for the President. That’s what, the schedule is your friend means.
Tom Kalil: There’s an aphorism, talk to who owns the paper. What that means is that generally, as I said, if a executive order or a speech is being prepared it’s on someone’s computer so you want to find out who that person is. If you want to be-
Robert Wiblin: Where’s the file?
Tom Kalil: Right, if you want to be able to edit that or change that you need to talk to the person who ultimately has that document on their computer while it’s still … it’s still this sort of internal discussion about what it should say. People would always say things to me like, Kalil, how did you get the President to say X in the State of the Union?”
Tom Kalil: And I said, “Well, I talked to the speech writer.”
Robert Wiblin: I wrote the sentence, copy and paste.
Tom Kalil: It was probably like when Sherlock Holmes would explain how he’d figured something out and Watson was then disappointed, because once you say that it seems really prosaic. It’s like, “well, there’s this person, they write the speech, I talked to the person who wrote it, who was responsible for writing it and I told them to insert this line and he did it.” But unless you’ve been in the policy process you don’t really understand.
Robert Wiblin: I suppose a lot of people aren’t in a senior position to have a lot of those conversations. .
Tom Kalil: Yeah.
Robert Wiblin: Yeah, organizations. Another one that stood out to me is making it easy for people to help you is almost always a good idea. Do you want to expand on that one?
Tom Kalil: Yeah, people have many more things that they’re supposed to do than they have time to do, so if what you do is you show up and say, “I would like to give you another thing to do”, generally people are not gonna be super receptive to that. If there was something that, for example, I wanted my boss to do, my view was I should not show up and say, “I have a monkey that is on my back and I would like to transfer this monkey from my back to your back.” What I would figure out is, how could I make it as easy as possible for him to help me. For example, if I needed help getting a particular member of the Cabinet on board to support an idea that I was enthusiastic about, then I would say, “If I draft an email for you will you look at it, and if you’re comfortable with the substance and the content, will you send it to them?” He would generally say yes to that.
Tom Kalil: That’s what I mean, if you want someone to help you, make it as easy as possible. That also requires an understanding for an individual in the context of the particular organization, what’s easy and what’s hard? Right? You have to acquire a lot of fine grained institutional information about how different organizations work, about how decisions are made within that organization, and what’s relatively straightforward for them to do and what’s really difficult to do and what constraints they’re operating under.
Robert Wiblin: Yeah, the OSTP team has been famous for having something of a no credit ethos, something that you’ve talked about. Do you want to explain that? Was that something that just appeared ’cause of the kind of people that were there, or was it kind of an active effort?
Tom Kalil: I think it was an active effort. It wasn’t like, “Oh we’re allergic to ever getting credit for things”, it was, “if there’s a trade off between claiming credit and getting something done, always choose getting something done”. In other words, if someone else really cared about who got the credit, then we’d say, have at it.
Robert Wiblin: Donate it to them.
Tom Kalil: Right? If it’s like … there were situations in which a particular agency really wanted to be the person who got to speak immediately before the President and introduce the President. Right? So, if something symbolic like that is important to someone and it increases the extent to which they are excited about the idea, then let them do that.
Robert Wiblin: Did that extend to being inside the team, as well, that people didn’t try to jockey too much for who was responsible for doing things?
Tom Kalil: Yeah, absolutely. One of the members of my team, Doug Rand said it was a team of dolphins, not sharks.
Robert Wiblin: Is being able to trade away the credit for doing something unusually important in politics compared to other areas.
Tom Kalil: I think it’s probably generally useful, but I think in politics, given that there are some people who care, then I think it’s particularly useful to care less than they do.
Robert Wiblin: Did you ever worry that if OSTP isn’t getting seen as taking credit for things, then that would be bad for the agency ’cause then people wouldn’t care about-
Tom Kalil: No, part of it depends on who your audience is. I think that that by being a White House staffer as opposed to someone who’s going around and giving speeches, I had a lower public profile and so fewer people knew what it was that I had accomplished, but there was definitely an audience of people who are like, “oh, that’s the guy who actually did it.”
Robert Wiblin: Some people are known.
Tom Kalil: Right? It’s like your famous for 15 people, but as long as they’re the right 15 people then-
Robert Wiblin: So it’s kind of an apportionment of the credit, where it’s like to the general public the Secretary is responsible, but the people in the office know that it was you and so … yeah, okay, that’s interesting. What’s this bit it’s very important to be able to get excited about other people’s ideas? Do you want to explain that one?
Tom Kalil: Yeah. There are definitely ideas that I have come up with personally, but I think I have a higher than average ability to get excited about other people’s ideas. What that means is that the range of things that I can work on goes up considerably. I was not sitting at my desk and saying, “gee, I think we should launch a bit neuroscience initiative”, but the fact that a group of scientists and engineers contacted me and told me about this idea and I said, “Wow, this seems like a big idea” and was able to get excited about pushing it. For me, I didn’t need to be the person to originate the idea, I just had to be someone who was in a position to help it in order for me to get intrinsically motivated to work on it.
Robert Wiblin: From your experience in government, do too many want to originate ideas rather than follow them up?
Tom Kalil: Yeah, that’s a good question. Certainly within academia if you took the position that I did of going around and being excited about other people’s ideas then that wouldn’t be great for your career. People want to know what intellectual contribution have you made to the field, so someone being in my position of saying, “well, I didn’t come up with the idea for the brain initiative, but I played this important role in moving from an idea to things happening in the world”, that’s not something that you could really get away with in academia as a rule.
Robert Wiblin: Can you explain your magical laptop thought experiment?
Tom Kalil: It’s similar to the one that I talked about before which is, you have the meeting with the President. Instead, the thought experiment is that you have a magic laptop and the power of the laptop is any press release that you write will come true and what you have to do is to write a headline, which is a goal statement, several paragraphs that provide context, and paragraph level descriptions of who is agreeing to do what in the form, A does B, so C.
Tom Kalil: Again, the idea is, here we are in 2019, there’s some more desirable future that we’re working towards and we have to articulate who would need to do what in order to increase the chances of that desirable future coming true.
Robert Wiblin: I guess the benefit of this is it forces you to be very concrete about exactly what you want to happen?
Tom Kalil: Yeah.
Robert Wiblin: Where previously things are too abstract to … this is like, “oh, it should be a priority but it’s too vague to deal with”.
Tom Kalil: Yeah, it also gets you to think about what Bucky Fuller called the Trim Tab. What’s the Trim Tab? The Trim Tab is the thing that moves the rudder that moves the ocean liner. You can occasionally find instances in which there is an organization or an individual that is in a unique position to do a particular thing. I mentioned that I served as one of the chairs of the Clinton Global Initiative, one year Walmart announces that they were going to green their supply chain. Walmart has 180 thousand suppliers, or something like that, they’re all highly responsive when Walmart asks them to do something. Walmart greening their supply chain had far more of an impact than had they written a check to an environmental NGO. This A does B, so C really gets you to focus on the, what do you think in a particular problem or policy domain are the key leverage points and who would you need to act in order to take advantage of those leverage points.
Robert Wiblin: Yeah, when thinking about what actions you want people to take, how often did you get to think about actors outside of the United States?
Tom Kalil: It’s a good question, I did a little bit of that but not much. I was really much more domestically focused.
Robert Wiblin: Yeah, in effective altruism we’re often very interested in international coordination things. It seems like often existential risks and long term issues require us to look even beyond a national level. Do you have any thoughts on how you can get more coordination with the international level, or is that kind of outside of your bailiwick?
Tom Kalil: Well, one thing that I would observe is that individual personal relationships were still very important. Dr Holdren played a really important role in climate policy and one of the reasons that he was able to do that is there were a number of very senior people that he had worked with in the Chinese government that he had worked with when he was a Harvard professor with appointments at the Kennedy school and other research institutions. Those were relationships that have gone on for a long time and there was the reservoir of trust and mutual understanding that he could draw on when he was talking with his counterparts. Whereas, if these people are interacting with each other and they’ve never met before, that’s gonna be far more difficult.
Robert Wiblin: Do you think the one on one personal relationships are more important at the international level perhaps, ’cause we don’t know … their institutions and organizations and things going on in other countries are less scrutable to us so it’s all we can know is just this one person who we’ve met, whereas-
Tom Kalil: Yeah. No, I think so. Again, if the countries have diametrically opposing interests, then the fact that two people know each other … You know, you think about the negotiation with Iran, the two people who were involved with those negotiations at the technical level were both MIT PhDs, on both the US side, that was Ernie Moniz and his counterpart on the Iranian side. That doesn’t solve the game theoretic problems associated with the coordination-
Robert Wiblin: But at least they don’t hate one another, potentially.
Tom Kalil: It’s very … I don’t think you can overestimate the importance of people and having relationships that are based on trust, and mutual understanding and reciprocity. One of the reasons that I’ve been able to be effective is that there are a lot of people in the science and technology community that I’ve worked with on some specific project and if that experience went well for the other person when I call them up and say, “hey, let’s work on X”, I’ve got this sort of reservoir of goodwill that I’m drawing upon as opposed to, “who the hell is this guy?”.
Tom Kalil: There are these increasing returns that can kind of kick in at a certain point of your career where people say, “oh, Kalil is making this a priority. In the past when he’s done that, things have happened, therefore I should take it seriously.”
Robert Wiblin: It seems like this is a big debate the people have, how much that happens in governments or institutions is about the individual people and their personalities and their decisions and their relationships, versus it’s the institutional structure or it’s the incentives, it’s the gain theory, that kind of thing. I guess you want to say no, the people do really matter-
Tom Kalil: Yeah, people matter. I think unfortunately the answer to that is, both. There are definitely times when I have felt under constrained, so I felt like, I decide to work on this, I could have done otherwise. There were no institutional reasons that were saying, “oh, you should work on nanotechnology” I felt that I had autonomy, and I chose to work on it, not because … there was no political economy, there was no interest group story of saying, “Tom, you should work on nanotechnology”. It was just, I thought it was interesting and I decided to do it. As I said, because I’d been the person who showed everyone the internet in 1993 they were prepared to give me the benefit of the doubt.
Robert Wiblin: As you know, longtermism is kind of one of the distinctive things you find pretty frequently in the effective altruism community. How common is that long term style thinking or values in the policy world at the federal level?
Tom Kalil: I would say that there are pockets of it, but I think a lot of policy makers are focused on dealing with more immediate and shorter term issues. For example, imagine that you are President Obama and you’re coming in and the economy is in a free fall, you’re gonna focus on stopping the bleeding. If you’re involved in foreign policy and defense, a lot of your life is about dealing with immediate crisis. You might try to carve out some time to think about things that are important but not urgent, but it’s very difficult and requires a lot of discipline.
Robert Wiblin: Are there any departments or agencies or people who you mentioned who are more sympathetic to this world view, we can get things done?
Tom Kalil: Yeah. I think that in the science and technology area there is this notion that we’re investing now and it’s not gonna pay off for 20 or 30 years, so I think science and technology by its nature tends to be more open to making long term investments. When I say long term I’m talking about 20 years not, “Let’s think about the far future and maybe they’ll be a trillion people who haven’t been born yet and we should think about their welfare”.
Robert Wiblin: In practice it seems like long term is often thinking on the multi decade level ’cause it does … when you’re thinking centuries it gets a bit hard to figure out what to do.
Tom Kalil: Yeah. I think policy makers also think, “well, I have a difficult time figuring out whether my actions are going to influence things in the right direction over, say, several years, let alone something that may not happen for a hundred years.” Their level of epistemological modesty is gonna go way up at that point.
Robert Wiblin: Yeah, with that kind of background, is there much that individuals can do to get people to think about the long term consequences of policy, or about technological change in government, or is it-
Tom Kalil: I think one of the things … there’s a couple things. One is that sometimes there will be windows of opportunity, we call those policy windows. Let me give you a concrete example. When the Ebola crisis was going on I had been briefed on a program that was going on at DARPA, which was, could we dramatically reduce the time to go from bug to drug. We have this new emerging infectious disease, we don’t have a vaccine for it, telling people at that point, “Well, give us ten years and we’ll have something for you” is not terribly satisfying. The approach that DARPA was using was as follows, you’ve got someone, they’ve been exposed to a pathogen, they survived, their body produced a set of antibodies that will provide immune protection, let’s identify those antibodies and then create a synthetic oligonucleotide construct that will directly encode for those antibodies.
Tom Kalil: The process of doing that would be a lot shorter than the traditional process of vaccine development. I knew that we were going to ask for what is called an emergency supplemental, which is, we didn’t budget for Ebola because we didn’t know it was gonna happen therefore we need extra money to be able to contain it. I was able to get some additional funding for this new approach added to that and my primary motivation for it was, maybe it’ll help in Ebola, but almost certainly if it works it will improve our ability to respond to future emerging infectious diseases, or maybe even a world of engineered pathogens. So that’s one thing you can do-
Robert Wiblin: Try to find a present thing that hooks on to some problem-
Tom Kalil: Yes, that’s one thing. The second thing is that the long term thing is actually aligned with the mission of an agency. I’ll give you an example of this. I started reading about the idea of potentially hazardous asteroids and I found out that although you might think that NASA was all over this, they were actually making a pretty small investment. I was able to work with the leadership of NASA to get them to do more of this, so in that case there was this long term issue, very low probability, but high consequence event and there happened to be an agency that you could say, “Hey, you should be thinking about this as part of your mission”. I was lucky that Lori Garver, who is the Deputy Administrator of NASA shared my interest in this. She thought it would be really cool if we had actually made investments so that if that ever did happen we were prepared, as she put it, “being smarter than the dinosaurs” that was her tag for it.
Tom Kalil: A third thing that you can do is, you may have a really long term goal, but you’re able to identify ways of making intermediate progress towards that goal. One of the things I blogged about was a researcher who said, what we should do is having this long term goal bootstrapping a solar system civilization. What did he mean by that, expanding the frontier of human activity beyond Earth and out into the solar system. One of the things that is preventing us from doing that is that all of the energy and matter that we use in space comes from Earth so it’s very expensive. What you’d like to be in a situation is to be able to do something called in-situ resource utilization, where more and more of the energy and matter that we’re using in space is coming from space.
Tom Kalil: If you think about being able to recapitulate the entire terrestrial supply chain, but in space, that’s gonna take a long time, that’s not gonna happen over night. But, what you can say is, “well, we should at least be able to get fuel in space”. You have this long term view of, would it be desirable to expand the frontiers of human activity out into space and ultimately that will require that more and more of the things that we do leverage energy and matter from space, and that is a goal that will take a long time, but there’s something that we can get started on. For example, we can say, “Where is all the ice on the moon”, we can characterize that and we can say, “how would we take that ice and turn it into fuel and how would we use that to create fuel depots”? That’s a long way from saying all the things that we’re going to make in space ultimately rely on materials in space so that we have self replicating robotic factories, but we can start to make incremental progress towards that goal.
Tom Kalil: One metaphor that I like about technology is that it’s like you’re walking in a building and you enter a room and in each of those rooms are three doors. So you can’t skip four rooms ahead, you can only advance one room at a time. So that is, from a pragmatic point of view, even if you had this really long term goal which is humanity being this space faring civilization, you have to say, “What is the next incremental step that we can take towards that goal.”
Robert Wiblin: So what door out of these three do we want to open?
Tom Kalil: Yeah, exactly.
Robert Wiblin: How did you pitch the idea of prioritizing asteroid defense? What’s the motivation for other actors to offer money to this, given that I don’t imagine the public is calling their senator, right?
Tom Kalil: They’re not clamoring for it. Well, there was research that had been done that said, “Hey, even though this is low probability, it would be really bad if it did happen,” number one, and number two, we had a sense that we’d only observed a small percentage of the asteroids that were out there that were potentially hazardous and that also, even if we were able to locate them, we would have no idea about what to do about it if we did encounter one.
Robert Wiblin: We’ve made pretty big progress on that now, right?
Tom Kalil: Yeah, we’ve made some progress on it but there’s a lot more that needs to be done, both on the characterization of the risk and then knowing what to do about it as well.
Robert Wiblin: So if there’s other people in the executive branch now or just somewhere in the US government who are interested in promoting policies that reduce existential risk like that one, where you’re thinking about asteroids, do you have any advice for them on how to pitch this to other people in government? How to frame it?
Tom Kalil: Yeah. Part of it was what we were talking about earlier, which is that just showing up and saying, “Hey, this is a problem. Do something about that,” is not terribly effective. So let’s talk about engineered pathogens. I think you could get most people to say, “Yeah, engineered pathogens is a problem and ultimately we’re gonna have to have the capability to prepare it,” but a more effective approach might be to say, “Look, we know that mother nature throws all these emerging infections diseases that we’re not prepared for and so we should figure out what are all the things that we would need to do to get ready for that? Because we see that occurring with some level of regularity.” You know, Ebola, Zika, MERS, different strains of Avian flu. Some of them could have increased transmissibility. So we know we have to get ready for that, and is there a way that we can get ready for that that would also have added benefits with respect to a scenario around engineered pathogens as well?
Robert Wiblin: It sounds like you think that we might be pretty bottlenecked by our ability to actually come up with policy ideas that we think are useful, that if you can do that, then you can potentially find the right people to talk to and there’s enough sympathy.
Tom Kalil: Yeah. I mean, I think that that’s not the only thing that you need to do, right? So for example if you come up with an idea and it’s incredibly expensive, and it also requires some major trade off with some other value that we care about, it’s not like it’s a walk in the park but I guess what I’m saying is that’s the first thing to do, right? If you can’t articulate some concrete thing that you want the US government to do that is different from what it’s currently doing, then you are gonna have a difficult time getting people’s attention.
Robert Wiblin: Yeah, do you have any technologies in mind that you thought of over the last 20 years that you think might help to reduce existential risk?
Tom Kalil: Well, as I said, I think that there’s a lot more that we could be doing on the biology side, and in a couple different domains. One is dramatically reducing the time to go from an emerging infectious disease or an engineered pathogen and some sort of response. In policy circles, that’s called reducing the time from bug to drug, and the other is moving away from the idea that we have to have the unique response for every pathogen. So that is called moving away from the one bug, one drug paradigm.
Robert Wiblin: Broad spectrum vaccine.
Tom Kalil: Yeah, or boosting your innate immune response to anything, or dramatically improved capability on the surveillance side so that you know if something is emerging.
Robert Wiblin: I guess that’s not the kind of thing that NIH, possibly DARPA, like other groups fund already that they considered as one of their responsibilities, so it’s possible to go and get funding for this.
Tom Kalil: Yes. So I think that … So for example I talked about this idea of gene encoded antibodies providing temporary immunoprophylaxis. So in those cases where we don’t have a vaccine and we have an engineered pathogen or something like that, just because you’ve done the research doesn’t mean that you actually have a capability that’s ready to go if that occurs. So I think there’s a lot more work that needs to be done, not only in developing the technology but knowing on day one what you would do once the next Ebola occurs or if there is an accidental or malicious use of an engineered pathogen or a pathogen that is released on purpose.
Robert Wiblin: As much as you’re worried about extinction risks or global catastrophes, do you get any benefit from the fact that defense and security and emergency management is such a big part of the federal government? There’s other people to collaborate with?
Tom Kalil: Yeah, no, I mean I think you get some of that, but it’s not always the case that just because something is really important, that someone is responsible for it. So I think sometimes that there are situations where I think that governments in general and probably human beings, for that matter, have a difficult time of dealing with things that they perceive as being low probability, high consequence. So an example is solar weather leading to EMP and knocking out the grid. There were some people in the administration that were working on this but you know, it was certainly not the case that the government was leaving no stone unturned with respect to its capacity to recover after something like that, in part because some of the things that you would do to prepare are just really expensive.
Robert Wiblin: I guess they face competition from other things.
Tom Kalil: They face competition with other resources, right? So for example if an EMP knocks out the electric grid and you have to have these really expensive and bulky transformers and right now the private sector has no incentive to do that, then are you gonna force them to do that? Are you gonna subsidize them to do that? So, sometimes if someone is focused on an issue, they can come up with a very long list of things that would be prudent to spend money on, but in an environment where that’s perceived as zero sum, those may compete less well against other things.
Robert Wiblin: Yeah, so one way that you could help with that competition is to get the general public to care and pay more attention to these issues, which you can try to do. You could try to make them worried about these risks, freak people out even potentially, as sort of that they’ll then call their senator or call their members of Congress, hassle people about it. Do you think it’s possible to mobilize people about these low probability, high impact problems? And would it actually help, at the end of the day, someone who’s in the White House trying to move funding?
Tom Kalil: We are struggling to deal with the high probability, high consequence threats, right?
Robert Wiblin: Thinking of climate change, perhaps.
Tom Kalil: We’re massively underperforming on climate change, so I think until we have that one under control, I’m not sure I would run around getting people concerned about EMP or something like that.
Robert Wiblin: One thing is that that one became very partisan in a way that asteroids is not. There’s no asteroid deniers, so that kind of helps. It seems like in general, things, sort of more straight forward engineering problems, it’s easier to get people on board for some reason or they feel very concrete. We think we’re gonna find the asteroids, or build a missile to destroy it. The more abstract something becomes, it’s like an AI is more … It’s further away from just something like a straightforward engineering challenge that makes it harder to get people to really believe in it and really feel it and do something about it. Do you have any idea where there are pressure points here where the right people can be spoken to? Because ideally I guess we’d have this … Like lots of people would be interested in the general public and would understand it, and they would elect people based on this and that would put pressure on all the stuff. You’d have this whole system here, but it’s very hard to build that potentially. Is there any way, any shortcuts you can get?
Tom Kalil: I mean, I think that if you have a group of opinion makers who get organized and say, “We think it’s really important for the government to do X and here’s how it could do X.” So one thing that you see in policymaking circles is the equivalent of a blue ribbon commission. So for example if a critical mass of AI researchers were all to say, “The government needs to double its level of investment in AI safety. Here are a number of the research directions. Here’s how much it would cost,” you wouldn’t need a consensus. It’s not like you’d have to get every AI researcher to sign it, but if you had critical mass and you said, “This is really important and here’s what the government can do,” then certainly on a technical issue that, again, doesn’t involve this sort of massive conflict with some other set of values and is not gonna cost tens of billions of dollars, then that’s something where the opinions of experts and opinion makers coupled with an internal champion can make a significant difference.
Robert Wiblin: Sounds like you might think that another angle is to convince people that it’s affecting them now. If you say, “Climate change is hurting people now,” people are trying now with some success, but maybe you could try to say artificial intelligence … Well I guess people do make this case. We’re already seeing problems with artificial intelligence doing things that we don’t really intend and we don’t understand how it works.
Tom Kalil: No, we’re seeing algorithms. We’re seeing algorithmic driven decision making, right? I mean, it’s sort of generally the case that sometimes when you use metrics and you optimize for that metric to the exclusion of other things, that leads to bad outcomes, right? So we’ve known about that for a long time, so for example, the National Health Service said, “We want to reduce wait times and so we’re gonna evaluate people on whether they’re doing that.” That didn’t lead to shorter wait times, it led to patients being driven around in ambulances longer.
Robert Wiblin: So they arrived at the right moment.
Tom Kalil: Right?
Robert Wiblin: Yeah.
Tom Kalil: So that ought to tell us that maximizing any objective function and substituting that objective function for human judgment can be a bad thing.
Robert Wiblin: Do you agree that potentially trying to connect it with concrete things [crosstalk 01:55:49] today definitely makes it easier to get people…
Tom Kalil: Yes, as opposed to saying one day you could have recursively self improving systems, that it’s gonna seem more abstract to people.
Robert Wiblin: What do you think of the idea of building a broad based, longtermist movement among the general public, among voters, at least among people who read the New York Times?
Tom Kalil: Well, what I don’t know is how many people you could mobilize around the abstract principle of long termism, as opposed to a concrete problem that we should pay more attention to from a long term perspective. And I guess the other question is how long term is long term? So as I said, we’re having a difficult time acting on things that are deleterious now and are going to get worse over time. We’re having a difficult time … We’re not responding adequately to something like combating antibiotic resistant bacteria. We’re probably not investing sufficiently in Alzheimer’s. These are very high probability and high consequence.
Robert Wiblin: Right, we’ve already failed it.
Tom Kalil: It’s not like we’re doing nothing. There’s a large gap between what we are doing and what we should be doing. So I guess I would say if you were gonna organize a movement around it, I would focus on the high probability, high consequences and build up our civilizational muscle for solving those problems and then after we’ve got those done, then we can build up a muscle around the low probability, high consequence things.
Robert Wiblin: Yeah, what’s a piece of advice that you might give to people who are part of this kind of new subfield of AI policy?
Tom Kalil: Well, I think we’ve talked about one of them which is that, again, if you ultimately want to impact policy, I think that you can have some influence by just saying that this is important and that more attention should be paid, but you’re gonna have more influence if, number one, you’re able to articulate something that you want the US government or other governments to do that they’re not currently doing, that your suggestions are based on a detailed understanding of the status quo, the capabilities of different institutions, and that there are people at different levels. The people who are making the decisions, the people who are informing the decisions and the people who implement the decisions, who share at least some part of your normative and empirical beliefs.
Robert Wiblin: Yeah, let’s talk quickly about the effective altruism community, you’ve had some exposure to it for a couple years. I guess haven’t been super involved so I don’t want to put too much pressure to give opinions that you don’t have. What are some of the mistakes that you think that we might be making today and how might we steer away from them?
Tom Kalil: Yeah, so let me tell you what I find attractive about it first. One is that I think this idea of long termism is really interesting. The idea that we should weigh for far future more heavily than we do, number one. Number two, the altruism part that is the sort of worrying about the welfare of other people, maybe even other species. The rigor of trying to say not only how can we do good in the world but what’s the most … Is it possible to think about the particularly impactful ways to do good in the world? So those are all things that I find attractive.
Tom Kalil: I think where I may have a difference of not “I’m right and the EA community is wrong”, but where I have a difference of perspective is that sometimes I think that the EA community has in mind the following thought experiment, which is you’re sitting behind a desk and someone is showing up at your desk and saying, “What should I do with my life?” And you’re saying, at the margin, what should that person do? And so you would say, “Well maybe you shouldn’t work on climate because other people are working on climate and so maybe if you work on this thing that we think is really neglected relative to its importance, that would be the quote-unquote ‘best thing’ for you to do and let’s evaluate all the issues on a scale “neglectedness and tractability”, and let’s measure all those things on a scale of one through 10 and let’s tell people to work on the things that are 30 first.”
Tom Kalil: And I mean it’s not like that’s a crazy point of view but my sort of emotional and intellectual reaction is more on the total level of effort that needs to go into it. So it’s not that I would say, “Oh, don’t work on climate change. You should work on this dimension of X risk.” My view is we’ve got to do a lot … The total amount that we’re doing on climate is grossly inadequate relative to what we can and should be doing. What we’re doing to combat antibiotic resistant bacteria is grossly inadequate relative to what we should be doing. What we’re doing on Alzheimer’s is inadequate relative to what we should be doing, and to me, that seems the more salient and resonant perspective as opposed to the thought experiment in which one person at a time is showing up and asking for career advice.
Tom Kalil: But I’m not saying that my perspective is right and the EA perspective is wrong. It’s just that I’m sort of far more interested in the getting lots more people interested in solving hard and important problems, and so I’m interested in the what is our current level of effort relative to what I think is necessary to solve the problem, not so much the perspective of I have a high level of confidence that this problem is a 30 and this other problem is a 27. To me, that feels like false exactitude.
Robert Wiblin: So I think the EA community often has this issue that there’s these problems that we’re concerned about, but then we get to the point of making concrete policy recommendations, and maybe we can come up with a bunch of ideas for this but then we find it very hard to kind of predict exactly what effects they would have and often we worry that they would backfire in some way that we haven’t thought of. So there’s a bit of reticence there. How can we become more mature about improving our policy recommendations to the point we actually feel confident enough to say that some of them should happen?
Tom Kalil: You might start off with the policies that you feel more confident with respect to a “no regrets,” right? This is a concept in climate policy. So there’s very few situations in which it seems like a bad idea to reduce subsidies for the fossil fuel industry, right? So you might say that we need more training of people who know both AI and policy, right? So it seems difficult to imagine a scenario in which that would be a bad thing. So you might get your feet wet by maybe identifying those things that seem like more in this “no regrets” policy.
Robert Wiblin: Should we also just be talking to way more people perhaps to get a better understanding of-
Tom Kalil: Well, so I think that there’s a lot of homework that you can do. So for example you can have a more fine grained understanding of what’s already happening, number one. Number two, what are the individuals and organizations that are involved in shaping policy both inside and outside the government? Number three, what are their existing normative beliefs and empirical views? And what are things that have been tried in other areas that might be potentially relevant? So I think that there is a lot of that sort of homework that can be done that would help you generate more ideas and the other thing is that yeah, you can talk to people who have been involved in the policy process and can give you feedback or can explain how particular institutions work.
Tom Kalil: So for example, let’s say that you decided that a good next step would be that there should be more R and D that helps reduce existential risk and maybe it’s on the pandemic side or maybe it’s on the AI safety side. Well, one thing that you’d want to know is which agencies would have the capability to do that? And how do those agencies make decisions? So for example, DARPA has a budget of $3.5 billion but one of the interesting things to know about DARPA is the P in DARPA stands for projects, and what that means is that every four years they stop working on something. So that means a quarter of their budget is available for new projects every year. So unlike an agency like the National Science Foundation where if you want to get them to do something new, you kind of have to get them some more money because they’re not gonna say, “This year we’re no longer gonna fund condensed matter physics. Instead we’re gonna fund this other thing.”
Tom Kalil: They’re gonna fund physics in perpetuity, and so the way you influence DARPA is you get someone to go there who wants to pitch that program. So DARPA is a very program manager-centric organization and the director doesn’t necessarily choose what to work on next. The DARPA program manager candidate arrives and their job talk is based on an idea that they have and they have to answer something called the Heilmeier Catechism, which are a set of questions that this DARPA director, previous DARPA director, came up with for evaluating whether or not you have a good idea for an R and D program.
Tom Kalil: So if you want to influence DARPA, and let’s say you want them to do more in the area of improving our response to engineered pathogens, then the way to do that is to say, “Well, who could we find who would be world class technically and has a great idea and could have convincing answers to the Heilmeier Catechism who would be willing to go to Washington for four years,” right? So if you didn’t know anything about that agency or how it operated, then that wouldn’t occur to you as a path to influence, but if you knew exactly how it worked and its culture and its procedures for decision making and how its budget works, then it becomes a little clearer about how you’d influence that as an organization.
Robert Wiblin: Yeah, you just have to have more direct experience and connections and actual knowledge about how things work rather than just a vague sense of-
Tom Kalil: The government.
Robert Wiblin: The government.
Tom Kalil: Right?
Robert Wiblin: It’s like millions of people.
Tom Kalil: A lot of times people talk about the government as if it’s a unified, rational actor that has a set of preferences that it’s trying to maximize, and, you know, occasionally if the president has something as really a priority, then the White House can manage to coordinate the activities of multiple departments and agencies, but it’s still better understood at the sort of more fine grained institutional level.
Robert Wiblin: So what maybe should the EA community be doing to gain more influence with senior policymakers or major philanthropists? Are there any things that you think that we should stop doing that are maybe hindering us?
Tom Kalil: I would say that I’m not sure how many people in the world will necessarily look at the world through the EA prism, but I think that if you decided that there are more things that we should be doing that will respond to an EA priority like reducing existential risk, then you can say, “Who are the people who already care about that issue?” So for example if you said that you were concerned about pandemics or engineered pathogens, that there are lots of people who are interested in that issue, and then you could say, “Is there some value added that we can add to that conversation?” First of all, a certain amount of humility about a bunch of smart people have thought about this. Are there ideas that they have that are not happening where we can help inform a philanthropist who is taking an EA perspective to say, “Oh, there’s a really important leverage point here.”
Tom Kalil: So for example, I mentioned that I think people are really important and so I think something that we don’t do nearly enough of is to say, “All right, here’s an area that’s really important where policy has a big impact, and where the quality of the people who are in those positions has an impact as well, and so someone should be thinking about what are the key positions? Who are strong candidates for those positions? And what sort of effort would increase the chances of the right person winding up in the right position?” That’s an example of something that I think we’re under-investing in relative to its expected return.
Robert Wiblin: Yeah. Are there any other ideas for how we could improve? I’m thinking perhaps try to recruit different kinds of people who we’re currently lacking or become knowledgeable about other problems that we haven’t focused on so far?
Tom Kalil: Well, I would say that I would think about this inductively, right? So, for example, on the question of existential risk, I would say could you find out whether there are particular types of people that the community of people who is working on pandemics lacks? So let me give you an example in a totally different field not connected to EA, which is that after the Healthcare.gov crisis, the administration made a full court press to recruit more people who had expertise in areas like software engineering and human centered design, and so there was a growing awareness, “Hey, we need more people like this in the government because more and more services are gonna be delivered digitally and if we don’t have people who don’t know about cyber security, software engineering, human centered design, then the government is not gonna be able to function in the 21st century.”
Tom Kalil: So, the hard way, we recognized that we needed more people with those skills in the government. I think you could proactively go out in these different issue areas and ask the people who are currently involved in that, both directly in policymaking roles but in terms of trying to influence policy through think tanks about whether they see particular types of people that we don’t have either involved in developing options, making decisions and implementing those decisions. So I think that’s a difficult question to answer in the abstract but you could constrain it and you could say, “Are there options that we’re not considering or do we have a really limited capacity to execute on policy? Because there are certain types of people whose skills we don’t have around the table involved in shaping, making and implementing policy.”
Robert Wiblin: How good a fit do you think the EA community’s culture is to actually running for elected office? Is that something you think plausibly listeners should consider doing?
Tom Kalil: I think some ideas of the EA community could be expressed in ways that would have broad resonance. So I think that there’s certainly an overlap between the ideas of EA and say the ideas of evidence based policy, which is to say that we should do more things that work and fewer things that don’t work and we should be engaged in continuous experimentation to find things that work even better than things that we know about today. So that’s more on maybe like the give well part of the EA community.
Tom Kalil: I think there are other ideas that the EA community has that are pretty far out of the mainstream of things that people worry about. I don’t know of a lot of voters who are concerned about whether if there are 10 to the 80th hydrogen atoms and they’re all suffering then we should do something about that. So I think it sort of depends on where in the EA, in the sort of spectrum of consequentialist views that you have about whether someone in the EA community would be a good fit for public office.
Robert Wiblin: With the time that we have left, let’s just try to get as concrete as we have with kind of career advice for listeners who might be interested in going into a government or policy career in one form another. Broad open question is kind of try to be as impartial as we can. We’ve got to be making the case here for government careers. How strongly she would be recommending policy careers relative to other promising options like science research itself or business entrepreneurship, nonprofits? I think at 80,000 Hours we have a sense that currently it’s a bit underrated among our audience, but how are you gonna weigh up given that there’s many different ways that one could make a difference?
Tom Kalil: I’m biased because this is what I’ve spent a good part of my career doing.
Robert Wiblin: Yeah, and it’s worked out pretty well for you.
Tom Kalil: And it’s worked out well but I think there was a certain amount of luck there. If you have the right person with the right skills and mindset and the right position, they can wind up having a large impact. So for example it’s difficult for me to imagine unless I were Jeff Bezos and I suddenly decided that nanoscale science and engineering was the most important thing in the world, being in a position where I could launch something that would last for 20 years and result in a well over $20 billion investment in nanoscale science and engineering or to introduce a very different way for the government to solve a problem which is that rather than picking someone to solve the problem, saying, “This is the goal and we’re gonna provide some sort of financial incentive to whichever team is successful in achieving that goal.”
Tom Kalil: So introducing just a whole new way of solving problems into the government and having it be used over 800 times. So, again, I think that this is a way of having an impact on the world that can be pretty significant.
Robert Wiblin: Yeah. What are the best arguments against going into government or for us to prioritize government careers more?
Tom Kalil: Well, I think a couple things. One is that there are areas where you have good luck, where it’s very difficult to get things done, number one. Number two, that particularly if it’s something that is long term, you have changes at administration. So, someone can work really hard on something like energy and climate policy only to see it be completely overturned by the next administration. You can’t take the politics out of politics, so sometimes one party will be against something merely because another party is for it. So, people are not saying, “Oh, what’s your Bayesian prior on this? And what do you think the benefit to cost ratio is?” So the types of things that EA is interested in with respect to applying rigor and trying to figure out what the expected recurrent of different actions is … Believe it or not, it’s not the sole criteria that is used to make decisions within the government.
Robert Wiblin: Yes. Speaking of the changing parties, in practice, does someone kind of basically have to pick a party that they’re going to be aligned with and then accept that, well …
Tom Kalil: It depends what role they want to play, right? So if you want to be a political appointee then it’s very difficult to be a political appointee for different parties. It’s not to say that it never happens. So think about someone like Bob Gates who was secretary of defense in both Democratic and Republican administrations, but I would say that’s the exception rather than the rule. But there are plenty of people who work at a think tank and can say, “I will talk to anyone who I think is acting in good faith and is willing to listen to me,” or as I said, you have people who are civil servants who will serve under multiple administrations, or you could have someone who is doing what we call the tour of duty, who say, “I’m not gonna work in government my whole life. I’m going to contribute a particular skill or expertise that I have and within reason, I’m willing to serve under either party.”
Tom Kalil: So I don’t think it’s necessarily the case, but I think it is the case that if you’re going to be a political appointee, that generally you’re going to be affiliated with one party or another, not both.
Robert Wiblin: How do you feel about the tour of duty model versus the “I’m going to have a government career most of my life” model?
Tom Kalil: I think that one of the reasons I’m excited about it is that I think that it can increase the number of people that would be willing to consider public service. So, the way that DARPA works is that when you show up, you have a date on your badge that says when you’re going to be leaving. There’s a couple benefits that flow from that. One is there’s a lot more people who are technical, who say, “Yes, at some point over the arc of my entire career, I would be willing to spend three to four years in government.” There’s a much smaller universe of people who say, “Yes, sign me up for my entire 80,000 hours working in the federal government.” So it just allows you to draw from a broader talent pool, the sort of tour of duty model. That’s one thing.
Tom Kalil: The second thing is if the position requires being at the very cutting edge of science and technology, you’re more likely to get that if you’re recruiting someone as opposed to they come to government and then they stay there for 40 years. Third thing is if they only know that they’re gonna be there for three or four years or even shorter, they have a greater sense of urgency and they’re more willing to take risks. So one of the things that you hear about civil servants, that I think there’s a certain amount of validity to, is if someone says, “I’m gonna be here until I’m 65,” their appetite for doing something that is gonna get them in trouble may be lower than someone who’s like, “Oh, fire me.”
Tom Kalil: “Send me back to California.” It’s like sort of, “Don’t throw me in that briar patch.”
Robert Wiblin: Yeah. Would you be able to kind of map out the space of different career tracks that people should have in mind? I guess there’s federal government, there’s the executive branch, legislative branch, there’s political appointees, there’s bureaucrats with non-political think tanks. How do you kind of conceptualize that whole space?
Tom Kalil: I think that one of the things that is more difficult about it is that there is not the same sort of uniformity that there is in other careers. For example you say, “I’m an undergraduate, graduate, post doc, assistant professor, associate professor, full professor, named chair, Grand Poobah,” or working your way up an organization from entry level employee to the CEO, right? So I don’t think that there’s this level of uniformity. I think it’s much more the case that people are, again, particularly on the political side, are moving in and out of government. So again they’re not serving in government their entire career. They’re coming into government. They’re serving for a while and then they’re leaving and they’re doing something else and then they’re coming back maybe at some later point in their career.
Robert Wiblin: How do you feel about the non-politically appointed roles in the US federal civil service? Are they a good place to start or to pass through?
Tom Kalil: Yeah. Particularly in the area of science and technology policy, one way in which people get started is doing this AAAS fellowship in the executive branch and then that gives you an opportunity to try it out and then if it’s something that you’re really enjoying, to keep on doing it, and then if you decide it’s not for you then you can go off and do something else. So I think that’s why these fellowships are good, because it’s a way of getting your feet wet as opposed to saying, “Oh, I know a priori that this is the thing for me to do.”
Robert Wiblin: What do you think of a local or state government as either I guess a terminal role or as a stepping stone to federal role? Is that a common path?
Tom Kalil: In some cases, yeah, absolutely. I think the trade off is that you have a more immediate opportunity to see what impact you’re helping because you’re closer to the ground, so there’s a tighter, more immediate feedback loop between doing something and then seeing whether it worked or not, whereas at the federal government, you may be multiple steps removed from the policy to the impact of your policy on real people.
Robert Wiblin: If a listener wants to go into the US executive branch at some point in their career and perhaps they’re in their ’20s, maybe early ’30s, what are the first steps that they should think about taking now? Or is that just too broad a question?
Tom Kalil: I think determine whether there’s a particular issue that they’re interested in, to learn about that issue, to try to find out who are the people that are influential or understand that. To think about whether there’s a relationship that they can have with a mentor or a colleague that does have connectivity with the executive branch for the Congress, and can serve as an advocate for them.
Tom Kalil: So for example, the way in which Maya Shankar came to my attention, the woman who created the Social and Behavioral sciences team, she talked to someone and then that person said, “Oh you should talk to Cass.” And Cass said, “Oh, you should email Tom Kalil, he’d be interested in talking to you.” So I think the combination of having an idea that you’re at least interested in exploring and then finding out more senior people who have worked in that area and are willing to talk to you and then for example, if you’re starting off as a student and you were a research assistant for a professor who’s really knowledgeable about that area, they might be able to have the social capital to say, “Rob is a really hard worker, you should definitely take a chance on him.”
Robert Wiblin: How much is prestige and credentialing important in government? Is it more than other areas?
Tom Kalil: I think it does have an influence. In part because it’s a shortcut. So as much as we would like to say, “Well it shouldn’t matter at all.” You know that if someone went to Yale Law School, then a very small number of people got accepted to Yale and of those, only a tiny fraction are interested in careers in politics and policy. So you’re thinking, “Well they’re probably pretty good if they went to Yale Law School.”
Robert Wiblin: And it’s a lot faster to evaluate than having to talk to them at great length, right?
Tom Kalil: Yeah. Yeah, so I’m not saying this is a good thing. I think it would be better if we had other mechanisms but I would say that realistically, people will make certain assumptions about your raw level of intelligence on whether or not you went to, and successfully completed, some highly selective institution, or got some highly prestigious fellowship. Now, that might help get you in the door but then they’re going to ask all these other questions about you as well but it is a way of standing out.
Robert Wiblin: So it seems like it’s quite important to find people who can vouch for you-
Tom Kalil: Yes.
Robert Wiblin: … and recommend that you should talk to someone else that can-
Tom Kalil: Yes.
Robert Wiblin: … Do you have any advice for people doing that? Because it’s kind of networking but it’s a bit more than that, it’s kind of proving yourself to …
Tom Kalil: Yeah, for example, what’s something that you have when you’re young? You have time and energy, right? So there are many people in the world who have many more things that they would like to get done than they have time for. So the trade that you can make with someone like that is to say, “I will work really hard on a project that you have defined and the quid pro quo is if I really deliver the goods on that, you will say things about me that you believe to be truthful. Like you are a really hard worker, you were a quick study. You have a high general level of intelligence, you’re a good writer.” Whatever are the attributes that may be difficult for someone to observe just by looking at your resume, having someone vouch for you is important.
Robert Wiblin: What kind of mistakes do you see people make when they’re trying to build a career in government?
Tom Kalil: One thing is I think that sometimes there are people who are really good at managing up but have a bad relationship with their peers and ultimately I think that catches up with you because you will have a reputation for the type of person who is really good at managing up but other people dislike working with. So you can get some gains on that in the short term, if you could get promoted rapidly if you’re really good at managing up.
Robert Wiblin: But then some of those peers get promoted and then-
Tom Kalil: But some of those peers then will, the next time you’re up for a job-
Robert Wiblin: Kind of roll their eyes to-
Tom Kalil: … they’ll ask around and they’ll find out that person was good at kissing up but they kicked down. Or they had really sharp elbows, or something like that.
Robert Wiblin: Right, so is that how they’re not getting along with their peers? They’re just too competitive? Or not helpful enough?
Tom Kalil: They’re too competitive. Yeah. It’s so clear what they’re interested in is-
Robert Wiblin: Themselves and their career.
Tom Kalil: … themselves and their career. So it’s the Adam Grant, I don’t know if you’ve read this book, Givers Versus Takers, right?
Robert Wiblin: Yeah.
Tom Kalil: So I think that if you are a taker, there are instances in which there are some short-term gains associated with that, but I think the reputational risk associated with being a taker, or just being unpleasant to work with, in many instances will catch up with you in the long run.
Robert Wiblin: Yeah, how much is a reputation for being likable important in DC?
Tom Kalil: I think it’s very important yeah, because this is a team sport. Getting anything done requires developing lots of relationships with people that are based on trust, mutual understanding, reciprocity and not being an aversive personality. If people are like, “So and so is really difficult to work with,” that will get around and people don’t … if they have the choice, they would rather work with people who are genuinely pleasant to interact with.
Robert Wiblin: I think some people have the perception that DC, or the policy world, is more conformist, more conservative perhaps, than some other areas that they might work in perhaps, like academia, or potentially entrepreneurship, or business, or Silicon Valley. Is that an accurate perception?
Tom Kalil: I think there’s plenty of room for creativity, but I can definitely see how people would view it as … particularly if you’re in the middle ranks of a very large hierarchical organization. So yeah, I think that’s not an unreasonable observation.
Robert Wiblin: Yeah, that’s something you potentially have to be somewhat comfortable with. Are there any other mistakes that you see people make?
Tom Kalil: I think one thing is people will sometimes underestimate how long it takes something to get done.
Robert Wiblin: And get frustrated if it doesn’t happen.
Tom Kalil: Yeah, I think that this is particularly the case if someone comes in from the business world and so they’re used to, “Hey, there’s someone in charge and once they make a decision then everyone is going to salute and we’re going to immediately go in that direction.” As opposed to this, “Hey, our founding fathers created this political system with lots of veto points in it.”
Robert Wiblin: And that has its benefits as well, I mean …
Tom Kalil: Yeah, but if you’re trying to get something done urgently, then the need to have lots of people agree can be frustrating to someone who’s like, “Well why can’t we just get everyone in a room and decide?”
Robert Wiblin: So you have to be willing to play the long game a bit?
Tom Kalil: Yeah, the other way in which I handled that was to have lots of projects going on, so if one was getting stuck then I could work on another project.
Robert Wiblin: Obviously DC is political in a sense, but office politics-y is it?
Tom Kalil: Depends on the organization, some organizations have more of that office politics associated with it. So yeah, I don’t know that there’s a universal answer to that that I can give.
Robert Wiblin: In the White House, or it seems like at least today, there’s a lot of jockeying for attention, influence, a lot of in-fighting, maybe that’s a bit untypical what’s going on right now? But yeah, do often have to find ways to route around people who are trying to stop things that people are working on?
Tom Kalil: Sure, yeah. I mean if something is important enough that you have to think about escalating it. So if you’re like, “Oh, these people on this level fundamentally disagree but we think that this is important enough so that it needs to go to a higher level and we need to repeat that.” So for example, that’s the way that the National Security Council is structured. So you might have a conversation that starts off in something called an inter-agency working group and then the issue is important enough, then it would get booted up to the Deputies Committee and those would be the deputy national security advisor, the deputy secretary of state, the deputy secretary of defense and then if it’s even more important, then it would go up to the principals. So that would be the secretary of defense, the secretary of state. Then if it’s even more important than that, then the president is directly involved.
Tom Kalil: So that process makes sure that you don’t get in a situation where nothing gets done because there’s disagreement at the cabinet level. So you say, “Well we can’t make a decision because the secretary of state and secretary of defense have a different position on this.” The national security advisor, if they’re doing their job, tries to resolve it at a lower level if that’s possible but if it’s not and if they believe the issue is important enough, they’re like, “The president decides.” So Obama used to complain that the only issues that he saw were the really gnarly ones-
Robert Wiblin: The hardest one, yeah.
Tom Kalil: … because if they were easier, they would have been resolved at a lower level.
Robert Wiblin: A lot of people, maybe including me, have this when you’re talking about inter-agency working groups and this committee and that committee and all those different positions, I guess it feels somewhat opaque and obviously you just … I don’t know, do you have to figure out which of these things actually cause things to happen and which of them are places where issues go to die?
Tom Kalil: Yes. Yes.
Robert Wiblin: That’s just something that you have to learn through experience? Like, “Am I talking to the right people?”
Tom Kalil: Yes.
Robert Wiblin: So an issue that comes up with quite a lot of people we talk to is, they’re interested in doing something in government but obviously it’s so large, we’re talking about millions of positions here and maybe only a few hundred thousand of those come up … are advertised each year but how does one go about a job search to actually try to figure out … How do you find the positions? The really good ones you want to apply for in a sea of just so many different roles that you could conceivably go for?
Tom Kalil: Talk to someone who has deep institutional knowledge of the particular domain that you’re interested in. So for example, let’s take the Office of Science and Technology Policy. When we entered, when the Obama administration began, there were say 40 people at OSTP. By the time we left, we had significantly increased the number of people that were working at OSTP so that it was over 100. So the way in which the office was structured was that we had … it was headed up by the science advisor, there was a chief technology officer who had similar rank to the science advisor. There were a number of different divisions, so there was one group working on national security and international affairs, one group working on energy and environment, one group working on science and the group that I headed up, which was working on technology and innovation. Then I would have 20 people that reported to me.
Tom Kalil: So if someone was interested in working on a particular issue, they’d want to know where are the parts in the government that have a significant impact on that issue and how do they impact on that? So for example, there would be … Let’s say that you were interested in pathogens, or improving our ability to respond to the next pandemic. There might be … people are looking at that from a policy point of view, someone who is at the National Security Council, who has all of WMD, weapons of mass destruction, that’s something that they’re responsible for and then maybe they have one person each that is responsible for different types of WMDs and then there might be a handful of people working on that at OSTP.
Tom Kalil: Then there are people working on that in different parts of HHS, like the Centers for Disease Control, or BARDA, so there’s not a substitute for talking to someone who can give you a sense for the organizational landscape on a particular issue that you’re interested in and explaining what their roles and responsibilities are and the types of things that you would be doing day to day if you were to work there, either in an implementation role, or a policy setting role and what are the types of backgrounds that people who have those roles, you know, traditionally have?
Robert Wiblin: Yeah but it seems like so much of the information you need to get, it has to come from talking to people individually. I guess that’s common in a lot of the world, but it seems like it’s particularly common here?
Tom Kalil: Well I mean, how do you find out how to raise a Series A round? If you are in the Bay Area? You talk to other people who do that, right?
Robert Wiblin: Yeah, yeah.
Tom Kalil: I mean the other thing is that you can find out some information by reading things.
Robert Wiblin: Only takes you so far.
Tom Kalil: Yeah, it only takes you so far, but for example if you wanted to find out how DARPA works, you can look at the professional background of all 130 program managers and you can look at what they did before they became a DARPA program manager and what programs each of those DARPA program managers is running and what the technical goals are. You can look at the equivalent of the request for proposals that tells you exactly the problem that you’re trying to solve. So it’s not like these entities are totally opaque but that you would want to couple looking at public information and the relevant literature with talking with people who are either in those roles or have had those roles in the past, or are deeply knowledgeable about the institutional landscape.
Robert Wiblin: From some of your interviews, it sounded like you had to work pretty crazy hours at times. I think one that stuck in my head was you had the saying, what was it? Friday means it’s only two days until Monday?
Tom Kalil: Yes, yeah.
Robert Wiblin: How much is the work fun and, I guess, how much do you just have to love it to be able to handle that?
Tom Kalil: Well I found it to be, over the long term, to be very rewarding. So what I found to be rewarding about it was the experience of either having an idea myself or more frequently, having people inside or outside the government bring me an idea or a problem that I thought was important, and then engaging in all the activities that were necessary to manifest that idea. So to go from a conversation with an interdisciplinary group of researchers, to there being a brain initiative that NIH and DARPA and the NSF and IARPA are funding and the congress making a decision to support the NIH component of it for 10 years, which is almost unheard of.
Tom Kalil: Or seeing a field like nanoscale science and engineering take off, or recruiting at any one time, 20 people to the federal government and not only seeing what they’d been able to accomplish while they were in the White House, but knowing that a lot of them are going to come back and engage in public service at some point in their career. So I found it very personally rewarding and also very intellectually engaging because of the range of issues that I got to work on.
Tom Kalil: So I had some people who were working on going to Mars and other people who were working on undernutrition in developing countries. So I personally liked the range of issues that I got an opportunity to work on and help advance.
Robert Wiblin: Yeah, if you’re trying to build a career in central government, how important is it to move to DC, maybe even before you actually have a role in government, just so you can meet all of these people and get recommended by them?
Tom Kalil: I think it’s probably useful but there are certainly people who did it without that but I think that they had, even if they weren’t in DC, they had a set of relationships where again, someone would say, “Oh, this is the person that you need to talk to if you’re interested in this.”
Robert Wiblin: Yeah, do you have any advice for people who are just setting out and building a network in this area? How do you first meet … Initially you don’t have very much to offer, because you’ve got no one to vouch for you and you don’t necessarily have any experience. I think for many listeners who are interested in this, they don’t really have much experience, they don’t have many people who they know in DC or involved in government. How do you get off the ground floor?
Tom Kalil: Well I guess the question is, you may not have anyone who can vouch for you who’s in DC but the question is, can you have someone who can vouch for you who has some set of relationships with people in DC? Or obviously what some people do is they start working on the Hill, so there are certainly jobs that you can get working on the Hill straight out of college. They may not be particularly glamorous jobs but it’s a way for you to get your foot in the door. There may be positions that you could get at a think tank straight out of a college, or after a Masters degree.
Tom Kalil: There are entry level positions that you can get … So there’s something called the PMF program, that’s something that gives you an opportunity to rotate between different executive branch agencies that you can get right out of a Masters in public policy, for example.
Robert Wiblin: What are some possibly important things that people need to know, or differences between the US executive and legislative branch? What kinds of people go into each one of them, or is that the wrong way of thinking about it?
Tom Kalil: Well there’s just different roles, so the president proposes a budget but it’s ultimately up to the congress in terms of what they appropriate, so that’s a very important role that they play. So if you want to do something that is going to require resources, then having the congress on board in the appropriations process is very important. That’s number one.
Tom Kalil: Number two is if there’s something that you want to get done that requires a change of law, obviously you cannot do that without the legislative branch and then number three, they can engage in this oversight role, which is really important, to say, “We think that the executive branch is not doing a very good job in this particular area and we’re going to hold hearings and try to get to the bottom of why they’ve been less successful.”
Robert Wiblin: How much do you feel formal education taught you, useful things and how to get things done versus learning by doing in the actual job?
Tom Kalil: I think it was more learning by doing. I think there were probably some things that I learned that were valuable, but a whole lot of it was just-
Robert Wiblin: Through experience.
Tom Kalil: Yeah, experience.
Robert Wiblin: I guess a lot of that … are there any particular courses of study that you think are useful? Or universities that you would particularly recommend that people go to, that might be underrated? Or perhaps it’s just the course of study is not the key question here?
Tom Kalil: I think it depends whether you want to be a generalist or a specialist. So a lot of the generalists were people who had a law degree from Yale. That’s pretty good if you want to be a generalist. I had other people on my team, Robbie Barbero, who was responsible for the biotech portfolio. He had a PhD from MIT in bioengineering and had done several biotech startups. Then he applied through the AAAS Fellowship and someone told me that he was doing this and that I should scoop him up and so I did.
Robert Wiblin: Is there more demand for specialist than generalists? Do you have any view on whether more people should one or the other, or how to choose?
Tom Kalil: No.
Robert Wiblin: Or maybe they … No. Takes both, or …
Tom Kalil: Yeah. Yeah, I’ve been a generalist, so I think that there are values to being a generalist but me being a successful generalist depends on there being this very broad ecosystem of specialists.
Robert Wiblin: Yeah, in our conversation earlier, you talked about the Intergovernmental Personnel Act.
Tom Kalil: Yes.
Robert Wiblin: Could you just explain how that works and how that can really allow people to move upwards quite quickly?
Tom Kalil: Sure. So the Intergovernmental Personnel Act is a very underappreciated law. What it allows is for a federal agency to say, “I want to move Rob from 80,000 hours to OSTP.” So basically what the government can do, someone in the government agency, is to move someone who is currently in a university, state and local government, or a nonprofit organization that has something about advising the government in its charter and they need to have been in that organization for 90 days. So if you look at an organization like the National Science Foundation, their leadership are university faculty members who are doing this tour of duty through the National Science Foundation. So the person who runs the computer science division of NSF, which has a budget of a billion dollars, is not staffed by a NSF career employee, it is staffed by a leading computer science professor, who is on loan from that university for some period of time, like four years.
Tom Kalil: So it’s a way in which the government can recruit someone who is in a nonprofit, state and local government, or academia and do so in a very easy and straightforward way.
Robert Wiblin: Yeah, you got most of your staff from this, is that right?
Tom Kalil: I got a number of my staff from this, yeah.
Robert Wiblin: Yeah, so how can someone use this knowledge? Can they approach people in government and be like, “Oh you could hire me using this mechanism.”
Tom Kalil: Yes, yeah. So someone who doesn’t know about this might say, “Oh, I would have to hire you through the civil service and there would need to be this open search and 150 people would apply and I might not be involved in the decision making and so I might get some random person that really doesn’t have the necessary skills that I need. But if I know, “Oh, Rob is the guy, he has exactly the right background,” and you’re in a nonprofit that has something about advising the government in its charter, or you’re in a university, then the government agency can then just cover your salary and then you are detailed to the government for up to four years.
Robert Wiblin: How come this isn’t more widely known? It seems like an amazing backdoor way of …
Tom Kalil: I don’t know. People think there are two types of people, there’s the political people and then there’s the people who are going to be there for 30-40 years and actually it turns out that there are a set of flexible hiring authorities that are on the books that allow the government to bring in people in a more flexible way, as long as it is term limited.
Tom Kalil: So you couldn’t bring someone on and say, “You can work for the government forever.” You could say, “You could work for the government for up to four years.” For a lot of people that’s great and then they’ll go do something else.
Robert Wiblin: Yeah, are there any people you can’t get … Can you get people just from any business?
Tom Kalil: You can’t do an IPA if someone is in a for-profit company.
Robert Wiblin: Okay. Right, right, I see.
Tom Kalil: Yeah, so if someone said, “I’m working at IBM,” you wouldn’t be able to do this.
Robert Wiblin: I guess, is that in part to avoid conflicts of interest?
Tom Kalil: Yeah, yeah.
Robert Wiblin: Interesting, makes sense. Are there any jobs available at the moment that you’re aware of that people can actually apply for? Do you regularly hear about positions and do you go scouting for people who could potentially go work in the government?
Tom Kalil: Yeah, or like … Jason Matheny is hiring.
Robert Wiblin: Yeah, right. Oh the Center for Security and Emerging Technology?
Tom Kalil: Yes.
Robert Wiblin: At Georgetown?
Tom Kalil: Yeah.
Robert Wiblin: What do you think about that project?
Tom Kalil: Yeah, I think it’s really interesting and I had an opportunity to work with Jason when it was the director of IARPA and I think very highly of him, so I’m really excited to see what he’s going to be able to get done.
Robert Wiblin: Are there any red flags that you think might show up for someone who’s not suitable for work in government or policy?
Tom Kalil: Yeah, I mean someone who dislikes working with other people, someone who is really impatient. So, I think you can be impatient inside but if you communicate-
Robert Wiblin: You have to control that.
Tom Kalil: … you have to be able to control that. So I think you have to be willing to be persistent and take the long view but yeah, those are some of the things.
Robert Wiblin: Are there any people who helped you a lot to advance in your career and were you able to figure out who they were ahead of time and get conversation with them?
Tom Kalil: Yeah, so someone who was very helpful to me and someone who was the best man at my wedding. So he and I worked together on the ’88 Dukakis for president campaign. He then went down to Little Rock and was the person who said, “Well why don’t you come down to Little Rock and work on President Clinton’s position papers on science and technology?” His name was Gene Sperling and so when Bob Rubin asked Gene, “Who should I hire?” He said, “Oh, you should hire Tom Kalil.” So both Gene Sperling and Sylvia Matthews were very influential in me getting the job working for Bob Rubin on the National Economic Council and the reason that I wound up working for President Obama was because I’d worked for President Clinton. So people knew some of the things that I’d been able to get done when I worked for President Clinton.
Tom Kalil: Then there was someone who was one of my early recruits on the Office of Science and Technology Policy, Kumar Garg, and he and I have worked together since 2009 and are continuing to work together and I wouldn’t have been able to accomplish half the things I’d done without a deep and long-lasting friendship and partnership with Kumar.
Robert Wiblin: Yeah, having worked in White House for 16 years, you must have some interesting or funny stories from your experiences there, different perhaps than what people expect? Can you share one of them?
Tom Kalil: Sure. This is a story that happened in 1995 and 1996, so as I mentioned, Vice President Gore was really interested in this idea of the information superhighway and one of his goals was, what if we could connect every classroom to the internet? So I would tell people about the vice president’s interest in this issue and someone who I’d gotten to know, John Gage, who was at a company by the name of Sun Microsystems said, “Oh, I’ve got this idea called Net Day. The idea is, what if on a single day, tens of thousands of engineers showed up in schools all across California and started the process of wiring California classrooms to the internet?” I said, “Well great.” He said, “You know, I’ve got a web page of what this would look like if it actually happened.” So he emailed it to me and I gave it to the vice president and the vice president thought it was a done deal.
Tom Kalil: So at this point, it was just in the fevered imagination of John Gage. So the vice president has a weekly lunch with the president and so he said, “Mr. President, we have Sun, we have Apple, we have HP, we have IBM, we have Pacific Gas and Electric and they have all agreed that they’re going to wire thousands of classrooms and schools all across California.” The president was like, “Great. Let’s announce it.” So it turned out that they were going to be in the Bay Area anyway so they decided, we’re going to announce Net Day. So I called up John Gage and I said, “They’re going to come out and announce this.” So he and I spent the next week calling in every favor that we had to get these CEOs to show up and announce that they were for us. They were a little sketchy on the details of what it was.
Tom Kalil: What John did was he developed a website, which was a clickable map of California, that allowed you to zoom all the way down to the street level, all 12,000 public and private K through 12 schools had their own homepage. You could indicate your level of expertise from, “I am an experienced network engineer.” To, “I will bring coffee and donuts.” All the schools were color-coded red, yellow and green depending on how many volunteers had signed up. So we could look at the map and figure out which communities were getting onboard and which needed some positive reinforcement.
Tom Kalil: So they announced that not only were they supporting it but they were going to come back and personally participate in it. So by the time they did, we actually had tens of thousands of people who had volunteered, so it was this positive, self-fulfilling prophecy because they said, “Oh, there’s going to be a Net Day.” In fact, there was a Net Day and tens of thousands of engineers showed up to wire the schools and many parents showed up to wire the schools, but they discovered the windows were broken and the bathrooms didn’t work, so a lot of them got more engaged in the schools as a result.
Tom Kalil: Many states decided they were going to do this and entire countries decided that they were going to have a Net Day as well. So it was this experience, a couple of things that I took away from it, one is that you could create this positive self-fulfilling prophecy, even though that was a very nerve-wracking period of time for me personally because I’d committed to the president and vice president to do something-
Robert Wiblin: To announce the thing.
Tom Kalil: … and announce something that didn’t really exist yet. Right?
Robert Wiblin: Yeah, sounds a little bit like an episode of Veep.
Tom Kalil: Then it was sort of applying massive parallels to this problem. So as opposed to saying, “How are we going to wire 10,000 schools?” The question was, how could you get every community to take responsibility for one school? So it was just very interesting of the experience that I had of going from something being a complete fantasy to actually seeing it happen.
Robert Wiblin: So government can get things done.
Tom Kalil: Yes.
Robert Wiblin: Just in sometimes peculiar manner.
Tom Kalil: Yes, exactly.
Robert Wiblin: My guest today has been Tom Kalil, thanks so much for coming on the show, Tom.
Tom Kalil: Thank you, I’ve really enjoyed it and I’m a big fan of the podcast.
Robert Wiblin: So in the episode we talked about the new Georgetown Center for Security and Emerging Technology. If you’ve made it this far, you might be interested to know that they are hiring. They’re looking for people to fill a wide range of roles, including:
- Research Fellow
- AI/ML Fellow
- Staff Researcher
- Research Analyst
- Data Scientist
- Senior Software Engineer
We’ll stick up a link to their vacancies page, and our own job board which lists many other policy related roles in the show notes.
The 80,000 Hours Podcast is produced by Keiran Harris.
Thanks for joining, talk to you in a week or two.
Related episodes
About the show
The 80,000 Hours Podcast features unusually in-depth conversations about the world's most pressing problems and how you can use your career to solve them. We invite guests pursuing a wide range of career paths — from academics and activists to entrepreneurs and policymakers — to analyse the case for and against working on different issues and which approaches are best for solving them.
The 80,000 Hours Podcast is produced and edited by Keiran Harris. Get in touch with feedback or guest suggestions by emailing [email protected].
What should I listen to first?
We've carefully selected 10 episodes we think it could make sense to listen to first, on a separate podcast feed:
Check out 'Effective Altruism: An Introduction'
Subscribe here, or anywhere you get podcasts:
If you're new, see the podcast homepage for ideas on where to start, or browse our full episode archive.