 Hello everyone, my name is John Palfrey and on behalf of the Harvard Law School welcome to the book talk in celebration of Yochai Bankler's brand-new exciting book The Penguin and the Leviathan. I am very glad to welcome you on behalf of multiple institutions that are eagerly hosting this event. One is the Berkman Center for Internet Society, another is the Harvard Law School Library. Michelle Pierce and others I see here from the library which has been hosting a series of books for book talk events for the Harvard Law School faculty and what a wonderful outpouring this is Yochai. There's somebody who knows how to take hard ideas and put them before us and fill the room at the same time which is such a wonderful wonderful trait. Let me give a few quick ministerial points and then quick introduction of Professor Bankler and I will turn it over. One is this event is not being webcast live at the moment but it is being recorded so during the Q&A should you wish to interact with Professor Bankler which I suspect you will please press the button it will light up and become red but you will be recorded just so you know of that. Second there'll be a reception after this event which is just outside. We'll take the last question around 7.15 or so and then we'll move out into the anti-chamber or whatever that area is out there for further celebration and last the coop is selling books in the back and I encourage you to buy them because it's a wonderful book and we all ought to have a copy and thank you to the coop for coming. So Yochai Bankler to this crowd needs very little if any introduction but I figure I'll embarrass him for a moment anyway. Yochai is the Berkman professor of entrepreneurial legal studies here at Harvard Law School and a co-director faculty co-director of the Berkman Center. He is a beloved colleague and an amazing scholar and someone who we love to celebrate and work with in this particular fashion. The book that he's written the Penguin and the Leviathan I think is an incredibly important work in lots of ways it is the culmination of a fair amount of work that he's been doing on the topic of cooperation series of rigorous studies and experiments that cross a huge number of different disciplines. This is also still a work in progress one of the exciting things about this book is that it's almost certainly a marker in time but not the end of his work on cooperation in ways that I'm excited to learn from and to hear about. Another key point I think about this particular book it builds on in incredibly interesting ways the wealth of networks his magisterial and very academically focused book I heard somebody walking in earlier this was one of those overheards as we walked in who said you know the wealth of networks was brilliant and it kind of helps to have a PhD to get all the way through it and really understand it. The Penguin and the Leviathan is just as brilliant but you don't need that PhD and it's completely accessible which I thought was an amazing compliment and I think this is true of Yu Yohai and your scholarship which is you take on the hard issues you present the most rigorous research and you reach a very broad swath of people. The wealth of networks was a book that won business book awards as well as being something that is cited consistently as the seminal work in the academic field I'm totally confident that the Penguin and the Leviathan will have similar effects on our scholarship but more broadly on society and it's just a thrill to have you here to talk about it more so with great thanks Yohai Bengler. You wouldn't think that at this point I'd get excited and by talking to a group of friends but I am and you're so generous and so wonderful to have made this introduction and thank you so much to you, to all the staff at Berkman Center, the library, Amar, thank you for putting this all together and to all of you to come and have this conversation about this book. As John said this is a very different kind of animal from wealth of networks in the sense that I very much put an emphasis on trying to get a set of ideas out there in a form that many people who aren't academically, who don't have the patience or the time could nonetheless access something that is grounded in a lot of serious work and as I was trying to present today I wanted to present this book because that's after all what I've said but on the other hand out there in our squares a movement is growing that is teaching us all in the making about cooperation about a set of values that's driving us so here and there I'll try to connect things that are in the book and speak to the moment that we're living without lying about what's in the book and and and bear with me on that. So this is actually from the introduction it's not a lie and it's and it goes roughly like this and it starts in some senses the reason for the book the reason for why we need this book. So this is October 23rd 2008 House Committee. We live in a world built around a mistaken model of human motivation. We have four decades of exquisite refinement of systems from our workplaces to our banking systems to our network structures that are all built around this core fundamental error that I learned Greenspan in a moment of truth was able to relate to and that basic error is not that we are sometimes self-interested that's correct that basic error is the idea that we can properly model and build our systems on assuming that we'll do well enough in designing our systems if we build them according to a model that assumes that part of rationality is self-interest and that if we approximate who we are by saying that we are more or less uniformly more or less self-interested we won't go too wrong whether it's in building structures of control that depend on reward and punishment whether it's building market systems that depends on getting the incentives right by which we mean getting the material incentives right in each of these cases we've built system across a broad range that depend on this model and that turn out in work throughout multiple fields and multiple areas of practice to simply be wrong and we stand today at a moment at which after somewhere between depending on how you start to count a literature of 25 or perhaps 15 years across all of these domains scientific selfishness is retreating and we need to learn how to adjust our systems design to this new model a lot of words let's try to tell some stories so one domain in which we recognize this very well is Gary Becker's classic idea from 1968 the way to model crime is to think of it as a payoff structure right you've got the benefit of the crime relative to the penalty times the probability of detection that's deterrents that's where you get models like a three-strikes law you up the penalty given that there's a budgetary limit on how much enforcement you can do and you should get more deterrence and less crime that's one domain another domain incredibly influential paper the Jensen and Murphy paper on performance pay and top management incentives right the idea that essentially the firm the company is structured such that every level is going to try to shirk and grab more and put it in its pocket relative to those above Oliver Williamson is the classic version of this model but what happened so the manager the mid-level managers look at the employees upper-level management at the mid-level and this agency problem is always a problem what do you do at the very top who looks at the top there you can get the same result by basically paying with stock options so the shareholders and the CEOs are aligned because they're just trying to improve the performance of the shares this worked fantastically for the CEOs this by the way is scaled to the 1986 compensation of GM's chair and Toyota's chair so this is a fantastic theory for the people who have the power to implement it it turns out in work that quite a few other people including not least Lucian Bebchuk here have done to actually worked pretty poorly for companies right so you have relatively high levels of return to share a relatively high levels of tax fraud relatively low return to shareholders their substantial work to suggest that actually what you get is misalignment from this model but it's a it's a model that has the same view of motivation everyone from the top down tries to shirk but they'll work well if we build the systems either of punishment or of reward to get everybody aligned in the same direction and this is just across the street so I'm picking up my kids from their music lesson and as I'm getting into the car there's a woman with a maybe three-year-old trying to get the three-year-old perhaps two just because of the strong will into the car and she's saying okay get into the car get into the car or that's five cents off your allowance three two one okay that's five cents now get into the car it's another five cents off your allowance this model of who we are penetrates everywhere because every reasonably self-respecting educated person has gone through a system where the beauty of the models capable of being written to this simplifying assumption is so enormously persuasive and yet and yet we see this across multiple into the intellectual arc moves in multiple areas if you look at evolutionary biology for example in the 1950s we have all sorts of models of group selection why do starlings rise into the night these more or less get dismissed as false in the early to mid-60s with the rise of mathematical biology you move from group selections to selfish gene then to models of much more direct reciprocity with the classic being the evolution of cooperation and tit for tat something that's very easily reducible to pay off to the individual gene as the model you can translate cooperation into self interested and we're safe from that side we see it in economics I barely need to say anything but even there you see mechanism design and efficiency all built on the same model in political theory from down so also into hard and you see the same model of assuming the inability to come together to a shared model to a shared set of goals and instead these models of conflict and in management science organizational sociology from Frederick Taylor scientific management and they build the model of the organization and Schumpeter through Williamson that I've just described that's an evolutionary that's an intellectual art now in each of these fields obviously although in evolutionary biology a lot less than in others in each of these fields there are counter narratives but these become the not dominant narratives throughout these fields and yet she moves and yet we live in a world in which persistently we see examples of behaviors that don't fit any of this whether it's parents who will work at a Wall Street firm and live in a framework that completely dedicates them to self-interest telling their kids in the playground that they should share nicely with their toys because that's the decent thing to do all the way to models that say nothing will ever be written except for pay I mean my favorite story that that Terry was just telling me today about using heat in yoga as a potentially copyrighted claim of all the bizarre ways in which ideology completely denies doctrine in order to achieve a property like model and yet she moves and what we've seen and part of what I did in the work in wealth of networks and over the last longer period is how much online we've seen easily how we can cooperate things that shouldn't have worked work and of course now we've come to a point where it's become a solution space so whether you're talking about purely distributed systems like free software like certain kinds of free software wikipedia or shady whether you're talking about traditional NGOs leveraging social production whether you're talking about small organizations or IBM big organizations like IBM leveraging free software the whole notion that you can as a practical matter bring large collections of people effectively to cooperate in ways that have outcomes we care about moved from being bizarre and artistic nonsense 12 years ago trust me I know to hey let's just do it of course how else would you do it very quickly that's in the practice of life the theory hasn't necessarily caught up at least at this basic level of theory but we do see in each of these disciplines across these various years depending on different times the development of more complex models that allow for cooperation so in evolutionary biology the step the next step from direct reciprocity was indirect reciprocity what does it mean indirect reciprocity I scratch your back today someone else scratches my back some other day and if there's enough structure in the population so that it's not completely open for anyone to do anything it turns out cooperation emerges much more easily than was modeled 30 years ago across a whole set of domains whether it's the reemergent of the reemergence of something like group selection in the idea of multi-level selection and I talk a little bit in the book in one of the chapters about how that's developed or the application of evolutionary models to cultural practices so that evolution begin when we talk about human beings begins to be applied more in relationship to culture and breaking down the culture the nature nurture debate as one that allows us to model cooperation among human beings we're seeing tremendous amount of work there and I'll give you an example in a minute in economics the shift to experimental economics and modeling a way from self-interest the development of neuro of neuro economics or combining game experimental models with neuroscience and showing how people actually respond not because we like to assume one of the problems of course with economics is you can solve anything if you just assume the reason that I'm doing this is because I want to do it the problem of course is once you actually see that neurologically people actually do get pleasure from cooperating or trusting in a certain way under certain conditions introducing the assumption into the model becomes less of an act of picking friends from a crowd and more of an act of how do I model these things that I see in the world it becomes a much more grounded change we see it in political theory so Lynn Ostrom's work on the commons in some regard is a model but obviously there's a lot more in the shift to the possibility of cooperation and perhaps the earliest set of work on this was in organizational sociology beginning I know it's a bad time to talk about Toyota but Toyota production system and the shift that it introduced into production the work on networks by multiple sociologists again showing the possibility of cooperation the core insight of all of these fields each by the way independently sometimes fascinating to read a paper in two disciplines and they'll make the same points on may have long citation lists and they don't cross reference each other because they're developing them independently in their own literature and citation practices is that this initial model of uniform self-interest that is more or less of a good design model is abandoned or at least advanced and then there's the debate within each of these disciplines my favorite example just because it's crisp is the comparison of these two quotes and I hope you'll you'll permit me to read so in 1976 Richard Dawkins who is a preeminent biologist I don't need to tell you the book is more complex than what I'm describing but nonetheless this is the site guys that it captured this quote from the book be warned that if you wish as I do to build a society in which individuals cooperate generously and unselfish towards a common good you can expect little help from biological nature let us try to teach generosity and altruism because we are born selfish let us understand what our own selfish genes are up to because we may then at least have the chance to upset their designs something which no other species has ever aspired to do not ambiguous 30 years later Martin Novak from the program evolution dynamics just down the street perhaps the most remarkable aspect of evolution is its ability to generate cooperation in a competitive world thus we might add natural cooperation as a third fundamental principle of evolution beside mutation and natural selection that's a big shift within a discipline that we understand to be at the very core of science and it's a shift not because Dawkins was an outlier and Novak is an outlier but because as I said there was a progression of increasing sophistication in seeing just how much cooperation there is relative to the beautiful initial simple model and how much more likely we are to find cooperation occurring in nature given one of a set of mechanisms that I mentioned before and what you see is a similar transfer transition in in several other disciplines the thing to remember about experimental economics is that most of the behavioral economics we read publicly and that gets a lot of attention has to do with what are perceived to be deviations from rationality not having preference orders that are trying to do any kind of instability offer ask problems these are irrationalities I'm not talking about that I'm talking about a particular branch within experimental economics that shows that we systematically behave as though we are not purely self-interested or not all of us are self-interested or not all of the time which is not has nothing to do with rationality and everything to do with what it is that we value we can rationally pursue non-self-interest non-self-interested goals that doesn't violate rationality in any way shape or form so the debate within evolution is not a new one I mean the classic round is is Herbert Spencer's survival of the fittest and social Darwinism versus go potkins mutual aid and the idea that if you actually go to Siberia instead of two other places you will find a lot of mutual aid among species but that of course was not what won the day a hundred years ago what won the day was the development of anthropology from France Boas and Margaret Mead and on to basically say no culture beats nature but in fact what we had in what we have in the new generation of work in biology is more a revival of the corpotkin response than a revival of the Boas and Mead approach that is to say an approach that so some of the work for example is work that you see up there some of the images showing empathy showing the brain imaging that shows the phenomenon of empathy through essentially one partner's brain lighting up in a way that's almost identical to the way that it's lit up when they themselves are shocked as when they see their partner shocked there's a physical reaction that isn't there but other processing of emotional processing areas light up in exactly the same way that they light up in the context of being actually shocked so I feel your pain turns out to be at least in some sense biologically correct or at least partly correct we see so the work the work for example on oxytocin the idea that essentially this is a one one economic study which basically ran a trust game which is a variant of a of an experimental economics game side one the experimenter gives one subject some of money that person can then give another player whom they don't see some portion of that that other player then gets three times that amount from the experimenter and can do whatever they want if they're fully trustworthy if it's a complete relationship of trust the first party will give everything to the second party the second party will get it all tripled that way the two of them get the most and then they'll split it equally and they both are best off in the classic self-interested model because the first person can't do anything to make the second person give back anything they won't give them anything so in a no-trust model the first person keeps it all in their pocket the second person goes home with anything and the experimenter keeps all the money one of the interesting things in this particular study is that when a particular a particular chemical called oxytocin is given in nasal spray to the subjects that will play number one they trust a lot more this turns out to be involved for example in the difference between monogamous and polygamous of all the various sorts and to be involved in all sorts of trust engagement the point is you're beginning to see work like Kropotkin's looking at the book trying to read the book of nature to tell us who we are and from that translating to the basic point that we care about trust we care about fairness etc that we'll see beyond why does this matter why does it matter if we build a system that relies on an idea of universal self-interest versus one that says actually we're quite different from each other there's a beautiful experiment by a social psychologist Lee Ross and collaborators that does the following it takes a standard finite prisoners dilemma very short seven rounds not hard for anyone to figure out what the last round the second the next the last round etc by all predictions everyone should defect in that prisoners dilemma and no one should cooperate you take now there's a lot of work across hundreds of studies showing that that's not how people behave in the lab that's important I describe it but that's not what I care about here what I care about here is that they took the exact same model of payoff the exact same game and they take it to gave it to groups of all American students to group of Israeli fighter pilots and in each case they distributed the groups and they said to one group you're not going to play the Wall Street game or they sold them you're not going to pay the community game they asked the commanders of the pilots and the teachers and and and dorm advisors of the students who would be cooperative and who wouldn't and those people couldn't really tell who would or wouldn't be predicted what did predict when you told people they were playing the community game 70% opened cooperatively and kept cooperating throughout the seven rounds when you told the group that they were playing the Wall Street game 30% opened cooperatively and declined over the course of the game which game and this is the simplest of interventions you can lie about this in a particular point although you can't lie for long enough once people realize in which of the two games it is but for me what this captures is that there is that whether we are in a population largely of cooperators or in a population largely of self-interested actors depends on how we build the system they inhabit sure there's a core even when we say community game there's a big minority 30% in this case who continue to behave as their self-interested they present a real challenge for any cooperative system that needs to be dealt with and needs to be understood and there is a small core again in this case interestingly 30% although in many other contexts it's probably somewhere close at the 20 or fewer percent who will be cooperators even in a context where they're not expected to be but there's a big middle who will be one or the other depending on the context they understand themselves to be and if that's correct then changing our systems design approach from one that assumes universal self-interest and designs for those self-interested actors to one that assumes diversity of motivation and tries to make sure that our systems are such that the big 40 or 50% in the middle treated as a cooperative enterprise treated as one in which being trustworthy doing the right thing being empathetic experiencing solidarity is the thing I'm supposed to do and the thing that I in fact will do becomes a critically important design target and when we think about controlling self-interest to then not come back and pull the rug out of the cooperative side not to end up making sure that only a minority end up being cooperative so that's the reason or the target so again in all of these systems we talked about you see the move from tough on crime to community policing is a model that changes multiple systems it changes technical systems from walking from being in a car to walking organization instead of 911 you have monthly community meetings agenda settings institutional having more room for discretion social humanization changing the us them boundary so that police officers don't only meet perpetrators and victims but also normal people changing across all of these systems to community policing unfortunately we don't have very good studies that tell us whether community policing is better or the new big board approaches that were developed in New York are better but we do know that this has become an a successful and be widely adopted in most communities of any decent size in the US and we see the same thing obviously in the stories about Toyota production system and systems product and team production that's been adopted very widely and obviously from Britannica to Wikipedia we see the fact that we have actual practices that are more cooperative and depend not on reward and punishment and monitoring but on cooperation and a set of design levers that I want to talk about to elicit internal compliance and intrinsic motivation rather than external control or reward and punishment that's the that's the primary move and the point is this isn't just about online a big emphasis of this book for me is that we found it out in some sense powerfully online but that's just because the set of costs associated with coming together and being effective was lower it's much broader and it's something that's quite fundamental to everything we see and do so essentially we need an integrated approach technical organizational institutional and social for cooperative human systems design and we need to use the best evidence that we have in this regard John I think you're absolutely right that this is the beginning of a project rather than the end because it's an effort to throw out there here's the something we need to begin to develop in that regard it's probably also somewhat disappointing right there are no there's no here are the 10 steps in which to build the best framework it's it's an invitation for a continuing conversation more than anything else and there's a limit to how much we can get from evolution it's it's very fashionable it's incredibly exciting to talk about neurological studies and to talk about particularly evolution but it's very hard to actually design systems from these things we use that more as our modern religion to try to understand who we are by nature than we do to actually design systems instead what we see is a lot of work across many different disciplines that we can think of or at least one way in which I try to think of this in order to try to explain to us why at the end of the day isn't it just enough to add money okay so that all of these complex motivations fine but at the end of the day if we add some more money won't people do more of it and if we put a punishment won't people do less of it and if that's true so okay so you've got this black box with lots of stuff going on but at the end of the day we can keep following the model for which we've had 40 years of evidence that Alan Greenspan was talking about and things will turn out right and I think the basic answer to that is what is sometimes called a crowding out motivation crowding out in psychology literature and some of the economics literature and one way of thinking of this is just as as a set of forces you think us as being in some sense tied to four engines or four horses or whatever the source of power that you think of and four is a number that's manageable for me to talk about not because it's somehow true but if you think of our material interest as something we really care about we all know that in some sense we care about material interests we have to we have to keep body and soul together if for no other reason we care about a set of but we turns out we also care about morality we also care about doing what we think is right about being treated fairly and treating others fairly and being in situations that we understand as fair and that pulls us in a certain direction we have a variety of emotional needs and effective responses to different situations that pull us in a certain way because they tell us who we are we have a whole work in sociology on social motivations and connections some of it that looks very much like money and the work on social capital that tried to bridge between economics and sociology some of it in social network effects some of it in in questions of group of solidarity and group response so the basic question is if things don't align how do they work and the basic model you can think about the classic model is the question of blood donation and how it plays out so there's an all debate about the gift from 1970-71 where Richard Titmuss writes about the American and the UK blood donation system and the basic model if we try to map it using these four motivations is that if you have money for donating then your material interests are strong to donate your moral commitments it's still something good but perhaps not so powerful but if you're trying to signal I'm the kind of person who donates then it's not as good because you might be doing it for the pain if you're trying to tell yourself who am I what's my identity I'm the kind of person who gives blood again it doesn't do the job so it says don't donate blood go do something else this was a long a long debate classic debate and obviously if you eliminate the money then the material interests go towards don't give blood and everything else increases and moves in the other direction there's a nice study from 2008 in Sweden which is at the baseline a donation system that basically takes this and says okay now we're gonna pay you 50 Krona to register for this program what happens and essentially what happens is basically in the direction that fits motivation crowding out interestingly so was only statistically significant for women not for men and then they said well what about if this is a social signal or an emotional signal what if we let people give the money away then it won't affect and in fact when you allow the subjects to give the money they were getting for blood away to a children's health clinic everything moved back theoretically and practically it all moved back to where to where it started before the whole money exchange played out what does that tell us it tells us one important thing and one extremely important but frustrating the important thing that it tells us is that we can't just add money or punishment you read the motivation crowding out is a real phenomenon it you can map it onto these vectors of how you respond emotionally or socially and the social meaning of the act but essentially if you add money you can suppress an activity more than you actually add from those who care about the material interests that's important what's important and frustrating is that the response isn't uniform in the population you may have gender effects you may have specific sociological and cultural effects there is there are ways of collecting data there are ways of slowly learning how to design these systems but it's much harder than coming up with a simple mathematically tractable model that says here's my utility function give me more of it and I will work and I don't know what all these things let me just call it utiles or money and I'll move forward that comfort level we should have lost what's hard about it is of course we don't have the full answer beyond that so if we think about this idea of cooperative human systems we can think of it in conceptual terms in design terms and in politics and in terms of politics conceptually it's a move from modeling rationality as universal self-interest translatable intermaterial concerns to something that accepts a diversity of motivations and a preponderance of pro-social humanity sensitive to the conditions in which it is and we need to learn specifically how to break that up and build from a design perspective it means that we're looking for design that's based on behaviorally realistic evidence-based design not on models again it's very easy to just come up with a model where you just add a few things into the utility function and all works well one of the things that's making the work so slow is that you need evidence that we're comfortable with to be able to begin to shift those arguments in the utility function and to be able to give them weights and maybe measure them and then there are all the limitations of experiments and how realistic or unrealistic they are at center from a political perspective and this is where we begin to tie to the occupy movement I think quite directly I think we were all in some sense raised on and on on a model at least in part of our studies across these 30 or 40 years to imagine that if we're trying to increase welfare we need to get the incentives right first and then we can deal with fairness through taxation and redistribution and all sorts of other things but first of all make the pie bigger you couldn't possibly justify making the pie smaller and then distribute and I think one of the things that these at least at the micro level experimental studies show us is that you can't actually separate the two that people care about fairness as an intrinsic part of their motivations that you can't separate out the incentives from fairness and ethics and empathy and solidarity even as a matter of how effective and productive people will be independent of the question of what sort of moral arguments you can make for how the pie needs to be divided and so bringing that back into the question of how we design our economic system becomes a critical political implication of this work at the much broader level so I talked to some extent about building blocks about communication about framing the situation about empathy and solidarity about what's right fair and normal about calculation about the various social dynamics these are all things that I try to provide the literature in the book in in a fairly accessible way about how we develop each of these as a potential design lever but it's it's moving on in time and I want to get an opportunity for a conversation so let me just pick out fairness ride through the other slides very quickly come to a conclusion and then if people want to ask about specific things I'll go back to them in the presentation rather than me continuing to talk so let me talk about fairness there's a tremendous amount of work now in experimental economics more specifically than in others about the question of fairness and by this we should mean essentially three big buckets that describe the word fairness of outcomes fairness of intentions and fairness of processes one of the things that we find out is that a sense of fairness is culturally contingent and diverse so what counts as fair in a given context can change in different societies and there's a beautiful set of studies particularly in small-scale societies collaboration between anthropologists and economists showing that once you move out of market societies there's quite a diverse set of what counts as a fair outcome in this game in ultimatum games and trust games some of the workhorses of experimental economics but one of the interesting things in those studies that actually in market societies a 50-50 division is a reasonable place where people agree on when they don't communicate even though again in the model there shouldn't be in these games anything like a fair distribution but this basic outcome in an ultimatum game of a 50-50 distribution as a major outcome and a 30 percent and a 30-70 distribution as the minimum before people just completely rejected and walk away is that it turns out that if you frame it in ways well you got lucky you flip the coin and that's why you got the money as opposed to because you're the first subject people are willing to let things lie where the luck is or if you filled out a short test so that somehow you have dessert in the money then that's fine you can keep more of it you don't have to give 50 for me before I blow up this experiment that's powerful because these are obviously stories we tell ourselves in modern market societies about inequality luck of the draw who I was born to dessert I worked these are the stories we tell ourselves and they turn out to play out in these experimental settings there's no single theory of justice this happens to be a nice study that's got very very nice studies from the cement industry this is about cement trucking and essentially looks at firms across that are that are regionally connected to each other and are occupying the same market and in principle should all converge on the same pay scale because labor should in principle be mobile from one to the other it turns out there's a lot of pay dispersion some fair some some firms pay the same rate for the same seniority others have wide disparities that's on the outcomes what they actually pay there's also a big disparity in how the firms say that they pay some pay some firms say oh we pay for performance others say we pay everyone equally it turns out that some firms correctly describe what they do they say pray pay for performance and they pay for performance and pay a little for a lot for a little performance or they say pay equally and they and they do in fact pay equally but it turns out oddly enough none of you will believe there are firms that don't do what they say they're doing and they say we pay equally but if somebody's a hard negotiator they'll pay them a little bit more maybe if they're somehow connected otherwise or there are same firms that say we pay for performance a bit they don't really monitor and they have very little pay dispersion they just say it well as it turns out when you measure in one case the accident frequency and then the other the out-of-service percentage that is a real performance measures you have fewer accidents and fewer out-of-service events if you're coherent if you say you're equal and you're equal or if you say you pay for performance and you pay for performance you do better than if you're misaligned between what you're say you're doing and whatnot and we can completely see that there are two theories of justice one is you work harder you should earn more you perform better the other is we're all doing roughly the same work we should all be paid the same the coherence between the fairness statement about how we ought to treat each other and be treated and the actual practice turns out to be more important than which of the two actually matters again what that does to a great extent is that because there's no single theory of justice we have to argue and negotiate I mean if you look at the 99 percent versus 53 percent debates over the Occupy movement it's a clear debate about what's the theory of justice that should prevail systemically in the United States is it one that has at least some minimal equality so that you don't have all of the benefits going to the 1% or is it one that basically said look you all of you who are getting and not paying you shouldn't be talking stop whining right it's just the people who pay taxes who should who should matter it's a different theory of justice and we can and the critical thing I think is that we can't resolve this technocratically there's no get the efficiency right first and that's the technocratically correct answer and that's the level of taxation and transfer and then let's talk about how how we do it there's no technocratic answer these are competing theories that in turn are internalized to people's willingness and to engage in a system in a way that is intrinsically engaged as opposed to constantly pushing back against it and that's a debate that is being had politically and now over which there's real then there are intentions people care about intentions as well as outcomes if in the same experiments you let people roll a dice or have a computer force them to make a choice about how much they're giving the others people are much more willing to just let the dice roll dice fall where they are with them when they think that the other person is treating them unfairly so much so that there's this there's a classic story about Herschel of Ostopoli who walks with a friend he's sort of the the classic Eastern Jewish joke character he walks down the street with a friend and they and they see a cookie and Herschel picks up the cookie and breaks it into and gives the small part to his friend and the friend looks at him and says Herschel what are you doing they said what do you want what would you have done if you've done the same thing what had I picked up the cookie and broken it into a small part and a big part I would have given you the big part and kept the small part for myself and Herschel says so what do you want agency and how we treat each other becomes absolutely central to what it is that we care about in fairness not only in effect and we care enormously about processes there's work from from criminologists like Tom Tyler about people's willingness to accept one's procedural justice is seen as acceptable I think one of the things that's very harsh to see in today's images is the militarized form of the police intervention vis-a-vis the occupiers that is a strict projection of repression and an effort to chill the speech in a way that is very powerful very expressive and at the same time is considered part of the normal police technocratic work that's a form of process of enforcement that's very different from what could be when I was in Israel in the summer during some of the demonstration there was a demonstration with about 300,000 people that is to say on the order of 5% of the country's population you could barely see a cop on the street where they were there they were relatively far away in simple uniforms and that was it that's a certain image you project to a crowd as opposed to the riot gear and the response to the degree of acceptance of that policing is commensurate so let me just finish up there's lots of other stuff but let me just finish up with a few points and then open it up I think what we've been seeing and part of what I tried to describe in this book is the retreat of scientific selfishness the retreat of the idea that scientific policymaking pushes us away from what we know and what we've been raised on in terms of a decency towards one another a sharing and a collaboration we've seen one decade's worth of actual practices in the network environment and increasingly visibly over the last 25 years in various businesses reviving this sharing nicely idea the broad pro social educational bent that we infuse our children with we're seeing diverse business and social production models begin to challenge the efficiency efficacy and growth oriented effects of scientific selfishness essentially science begins to push back with theoretical experimental and observation and observational work it turns out from all of this that we aren't universally self-interested the different people respond differently to empathy or solidarity the ethics and fairness to material payoffs sometimes it's each one of us at different times in different concepts will respond differently to each of these and I think most of us have that memory within us at one level as an academic community we need a new field of cooperative human systems design that accounts for the diversity of motivation and for the inability to add just money to make things work but at another level I think one of the things we need to do to recover from this moment that Greenspan's testimony captures so beautifully is a new view of a shared humanity that we have that in some sense we've all grown up with and know perhaps if not all most and at the same time as we grow through our educational system we're taught in some sense to suppress it to understand to be somewhat cynical to be somewhat cautious to know that the right scientific rigorous way to think about it is more in self-interest and we need to overcome that and regain that capacity primarily and first and foremost to say flip a coin more likely than not in this case better than a flip of a coin the person across from you is somebody with whom you can cooperate and that's a big fundamental difference from a model that assumes we're all self-interested and need to cabin ourselves in these systems that will force us to work in coordination if not in genuine cooperation that's the project of trying to begin to help us leverage from science to a sense of shared humanity okay thank you that was fantastic and for those who haven't yet read the book still do even though you told some of the stories that are in the book and the green span and the dock and so those are in there you left a lot on the table in terms of stories in particular so anybody watching the video still ought to read the book here all right I can imagine there are lots of hands ready to go up and do you mind please press your button if you're willing to tell us who you are on the record we'd love it I'm visiting at the law school you hi this is fascinating and I can't wait to read the book but let me give you two knee jerk reactions on on the theory behind it one is that the passage from the domain of science in particular the issue of biology and evolution to the social sciences has long been discredited because we are talking about two different things that we cannot extrapolate from to explain society and the other dimension is that going back to the green span issue no doubt there are two aspects of each one of us and we are being coerced into a design that has been dominantly leviathanized but the question is that there is an objective limit to this and the argument that the reason why people are in the street to criticize the way the banking system has served them so poorly can be responded to by saying let's have more cooperation is too simple because the material interests that the bankers have are not defined by a theory that says you have to follow leviathan and market but it's supported by an obvious self-interest that transforms the battle really in one which marks has long under underlined as one that cannot be solved that simply by moving from one design to the other so so cheaply that's as always incisive and and hard the first part easier I think than the second so the first argument you made was this whole sociobiology stuff yeah who cares it's old and it's been discredited I think that's a description of the state of various disciplines that is true of the early to mid 80s but is no longer descriptively true in terms of just how much you are seeing more social sciences economics in particular trying to move to and as well as some aspects of anthropology trying to move to evolutionary modeling and evolutionary models to explain social sciences so in some sense the the richness and complexity and ability to remain agnostic to the outcome like economics want but without assuming universal self-interest and at the same time a rigorous mathematical modeling has become extremely attractive to try to model economic and social interactions in evolutionary terms and so what was in the early 80s resolved seems even in more of the social sciences less result the second point is and I struggled with this exact problem in in writing this chapter of the book and in using some of the materials later on is that we use science and particularly evolution we use evolutionary sciences a way to negotiate our conception all the it's a religion it's a way of understanding the order of the universe and our place in it so that's why I start by talking about how the discipline has changed because it's a religious disputation about ourselves in the universe and our place in it but then say but it's not where we learn the design leave it's from for that we need to go to the real social sciences and yet there are components particularly in the neuroscience work that do have direct feed-ins and in the models that transfer to to to cultural evolution the second question the second question was yeah cooperation sure but bankers self-select because they care about money they control this system it's in their self-interest to keep it as it is how are you going to change that there's a real answer which is well that's a lot of work but that's not an interesting answer so I'll answer anyway one of the problems that happens when you have a system that's optimized for money so that money also becomes the social signal right if I make a million and a buck or a million and three bucks doesn't matter to me on the material interest side as much as the fact that the fact that that person got a million and three means that he's more value I want a million and four it becomes the way in which we translate even the social motivations but it obviously also self-selects for people who are driven by that the population there's a lot of people in the population who care about that so if you're trying to look at interventions what would happen so so so Bruno Frey and Margit and Margit Ostolo have a beautiful paper called why executives should be paid like bureaucrats that is say what would happen if you externally politically forced payments to be more bureaucratic and much more related to worker pay so you know the differences between when when Jensen and Murphy in the 80s Japanese ratios of CEO paid to worker pay was about one to 26 the US if I remember this off the top of my head was around 1 to 40 1 to 50 Europe was somewhere in the middle after the Jensen and Murphy piece gave rise to this sure you should pay in stock options the US went to 1 to 500 white to 600 without Japan and Europe following to the same extent at all so you got a different set of people so that's an intervention that basically says don't pull these people do you or don't you introduce ethics into and professional ethics into business we as lawyers have professional ethics all the time we know their limitations but we at least have something to touch against when we try to train people for mediation so in the communication part I talk about some of the handbooks and differences between litigation and mediation we try to teach different values to get people not to my side wins but let's figure out where so you train people differently that's a long-term intervention is it hard sure it's hard when you talk to some of the businesses I don't know like IBM who are trying to interface with the networked with the networked environment and trying to connect to say for example open source projects one of the problems is the kind of person who can be an effective participant in open source development for IBM is a very different kind of person from the person who can weave their way up in a bureaucratic hierarchy an organization has to begin to learn who the right people are for itself so if you're lucky and you have an organization somebody relatively at the top understands that they begin to find these ways and develop processes to identify different talent that responds differently say valuing the fact that people go out and were part of the p score as opposed to ran straight from business school those are the kinds of interventions they're not a from today for tomorrow but I think that they do tie to the specific sources of insight as opposed to just waving my arms and saying do these things which is really the effort of trying to translate what we have from the evidence which is incomplete into a set of at least first-line interventions which are also incomplete to obviously great questions with lots to respond session so just to push you a little bit further on this I mean I love this work as you know but so today I'm a new professor at MIT and today my department was visited by someone from the licensing office so they gave us a very nice PowerPoint about the model through which anything any ideas that we develop can easily be copyrighted and patented and and so on and when I asked him about alternative licensing models he did have an answer which was that yes we certainly do that there's the MIT license we have open licensing but it was really a yes but answer and when I tried to push him a little bit further and talk about well it would be interesting since there is such a broad transformation in so many domains of knowledge production and invention to actually have resources in your office devoted to helping people learn about these other modes his answer was well you know talk to the top of the hierarchy you know my office does what it does and and so I guess the question is how do we I guess sort of normalize or infiltrate or generate spaces where these forms or modes of distributed and shared a networked production can be system system ties across the whole educational system especially in a context where those who are responsible for doing that work are responding to the imperatives given to them by those who were very good at climbing to the top of of the hierarchy based on the previous model so I think that's again a fascinating and very real world problem and in this room there are many people who are working on it in different ways as with any other context to some extent you fight to some extent you argue and sometimes you even persuade the question of university licensing is a is is a combination of all of these so if you look for example at the at at at university at university open access publication so John you're part of this major project here Stuart Schieber who's one of the directors of the Berkman Center and of the computer science department Bob Darden the the library pushes forward in the faculty we we use this is what's interesting we use our power in the hierarchy of universities to raise a flag and say we care and we can do it this way that is to say we can all commit in these faculties and this law school as well to distribute our work freely and that allows somebody else to come and say hey look at what they did it's feasible I think there was a lot of this business of people doing their own releasing and then getting other people you know after I released wealth of networks freely on the web the first day that it came out and that was the first university press sort of book type thing that came out I got emails from all sorts of people who are trying to negotiate what did you do what did you say well get another offer start getting them to feel like they want to and negotiate hard and ultimately point to see what happened with the sales of that thing or what happened with the sales of the other and then you have to and then there's there's there's politics so if you've got the student organization University Allied for essential medicines it starts out with patenting for essential medicines and becomes more generally a student organization that's trying to pull universities in there's no question universities should in principle be easier because in some sense our business model does not depend on this thing in the least though there are people for whom their metrics of success is succeeding that's their job following from a statute that says yeah just just really quickly on that he one of his slides showed that MIT sort of held up as you know one of the most successful examples of the licensing to 2 to 3 percent of the income of the university you know came from this license yeah so this is a paper that I when I tried to measure there about 10 years ago when I did a paper on this you know it ranged from half a percent in University of California to that's all it's not a rounding error but it's but for many universities it is for many universities actually the cost of the office doesn't cover the revenues but it's the thing to do and that's part of what I'm trying to do with this book is get people to stop thinking that it's the thing to do because science tells us so no science doesn't tell us so I will note that both your book at Yale University Press who was open and Jonathan Zittrin's who was in the back I saw him before he got wise you did the future dinner both were available open access and sold for academic books were blockbusters right I mean they really that is two amazing data points both from Yale University Press notably how about someone in the back yes sir right in front of the microphone in the camera yo hi I'm completely convinced a Nobel laureate in economics once said it's safe to assume that people are self-interest with guy and clearly science does not support this so Oliver Williamson may be a Nobel laureate in economics but he was probably wrong on this what I hope to get from you is what are the burning policy implications of this shift in the way we view the world so if you had a magic wand and you could just fly around and make people see the light make people understand that we should conceptualize humans not as self-interest with guy but as cooperators where do you think we would get the biggest bang for the buck where should we start should we make Congress more cooperative should we change the patent system should we change the way organizations work who should pay attention to date who should be the foremost people that should pay attention to these findings so thanks that's a real and hard question in the interest of full disclosure I don't answer it in this book here and there from the examples it begins to point in the direction I think for example that a more formal assessment of the relative success of community policing and big board approaches is an important intervention in the day-to-day control of large populations and the degree to which they are or are not free that's a level again look at the squares look at the parks the basic question of whether or not we continue to build a society that is built on maximizing total growth or society that trades off some form of total growth growth with a more equitable distribution of outcomes with some minimal set of capability am I able to back out of this anything like whether or not the model is living wage or a certain level of educational tax subsidies or something else I can't each what I can and what I have tried to say is that these kinds of interventions and purposefully I'm not going to patent I'm not going to open access I'm not going to networks although there I am much more competent to speak specifically because I want to make it very clear that the point is much broader and deeper and that I think that the we might call it Washington consensus if that's what we want to call it the particular model that has typified essentially the last 20 years to or really since the 70s of deregulation is one that is built on an erroneous model of interest that has contributed at least to some extent to this pattern of internal ratcheting up of the degree to which people who are self-interested are running certain organizations and systems leading to their failures and that's the those are the biggest policy questions but in some sense they're also the most distant because the work that I use is all done at the micro level right none of this work says an organization isn't self-interested I don't know anything about that because we don't have work of this kind the work is all done at the micro in some sense as the book plays the role of who we are and what is our place in nature to plug into those debates rather than the immediate translation there are immediate translation so if you talk about copyright in music the idea that musicians really need these payoffs is simply false descriptively we all know it from actual observations of the market the idea that the only way you'll get people to pay for their musician to support their musicians is if you force them criminally that's false we have empirical data for that that's something that I do talk about a little bit in the book so there are places where the relationship between the micro motivations and the and the institutional intervention are very close and are relatively easy to make and there are places that are much more fundamental and I try in the book to play a little bit between both of those levels but not do much of the translation I think you okay you sell yourself a little short chapter 10 on how to raise a penguin seems to be in fact maybe some of that description although that's more about organizational structures and processes than it is about policy question which is what Shane was asking about that sounds right so we've had a couple questions or comments that all seem pretty friendly and agreeable I wonder is there anyone who really it's kind of in a fundamental way disagrees who wants to crack it and then there'll be one final question yes sir you're on with the disagreement Bob's Astro I worked in corporate America for most of my life and also as a federal cross prosecutor one comment I've looked at pay inequality in Japan for a while and in the United States and if you look at that in relation to corporate compliance the last thing that I think makes a difference is the degree of gap between the manager and the worker for example you look at antitrust compliance in Japan vis-a-vis America it is much higher in the United States because the deterrent regimes are very very effective and they're very well enforced if you look at America for example 50 or 60 years ago when my granddad ran railway express and I think he made $25,000 a year or something like that and compared just a level of compliance on things we all care about which are civil rights you know environmental protection antitrust I wouldn't say disclosure to consumers but you know a whole set of things compliance is you know much better in the current view and it's much better in America vis-a-vis many other countries you know price basically cartels for example don't have meetings in America because they know the FBI may be knocking at the door so do you have any other prescriptions for reforming corporate America in areas where we all are concerned about other than just compressing salaries so several things that's a rich and and and good set of questions several things the first is the once you're talking about the behavior of a corporation or organization it's we don't have the same kind of data on what happens within a corporate decision-making when people see their role as agents of someone else so they're not acting for themselves you begin to have conflicts about the ethics of the role and the ethics of the individual sometimes doing something that's skirting the edge for the good of my shareholders my community my whatever becomes morally acceptable given the framework of what my role is my fiduciary duty etc and so then you need the external enforcement no I'll actually do them poorly so it's very hard for me to say at the level of corporate enforcement what to do except to then begin to look at are you talking about individual are you talking about corporate incentives how do you are you able to essentially change the perception of these corporate managers about what they're supposed to do I think a lot of people would would agree that the one thing that really does make a huge difference is jail time for the executive and the probability of getting caught which was a huge problem in our securities debacle and if you compare the record of securities enforcement and antitrust enforcement for example in the United States you'll see that it is relatively certain jail time if you're convicted is a huge deterrent and is very effective how would they how would you then explain the findings for example that pay disparity or pay ratio predicts probability of tax fraud which where you keep the baseline so this is one of the one of the early 2000s studies that tries to test Jensen and Murphy same baseline enforcement on the background but a real contributor is the extent to which you've got stock options are not the pay ratio but whether or not you have stock options predicts tax fraud outcomes I really can't I don't know I'm just coming with my examples so the ones that I'm familiar with I would urge you though really to take a look at the record of antitrust enforcement over time and also transnationally because I think that would be a counter example so my point though is that there's no question that adding money or adding punishment has effects on behavior the question in each context and this is why I was trying to be very careful about talking about specific context and specific evidence based on what we can actually observe and measure is whether or not the intervention at the margin of enforcement has effects that are beneficial or not beneficial on the behavior relative to interventions that are more cooperative in nature would that be education would that be limiting pay those are those are the levers with which at the moment we have a little bit of data but that's all we have at the moment fair point you're ready for one more final question sure please thank you my name is Emily Dexter and if a principal of a high school came to you and the high school served a very you know multi-ethnic multi-income diverse population so there are a lot of different interests and a lot of different people trying to get different things out of their educational system and this was sort of your typical high school with different types of courses and teachers and all those sorts of things what kinds of ideas would you want to see that how could your ideas influence secondary school for example so I'm very fortunate to be the son of someone who founded three schools in Israel over the course of her life actually now it's two schools and one volunteer organization so so dealing with exactly this question in populations of extremely poor parts of the city as well as in a magnet school that brought people from all over the city and in that regard the lessons are relatively straightforward which is to say higher degrees of autonomy higher degrees of trust structure as opposed with clearly set goals but at the same time a good bit of teamwork and cooperation between the students less hierarchical and less distant relationship between teachers and students high stakes testing doesn't seem to me to have been particularly effective on the side of motivating teachers at least that's what I think we're beginning to see here in the results here so in that regard because you're talking again at the micro level of what works with kids and getting them to want to learn to want to cooperate to want to behave that's the set of solutions how easy it is when you're talking about a population that comes from a very low trust environment that has a long personal history of being treated aggressively is a very different question but I think in that domain the application is is much more direct and there's literature within the education literature that talks about somewhat more open with greater trust models and the performance effects relative to much more rigid architectures and that and that's in some sense some of the primary materials that I see rather than where I would what I would learn from rather than what I would try to teach okay this was a great gift thank you so much this was a wonderful