which is kuzey amo per questo every tardone any need soda questa for this brief delay on the occasion of the final event of the festival there are some controls underway there are people outside who are waiting to get in they will get in gradually but we cannot further delay the start of this lecture by professor Daniel McFadden it's very important because it draws a and they touches upon the main issues which have been discussed in these days Daniel McFadden is an econometric meaning that he's a scholar who deals with models abstract models of economics data and econometrics require is a work on models and Montelena and you certainly had the possibility of listening to presentations about econometrics in these days the reality is so complex that we have to simplify it and these models help us draw conclusions and then we need to test the consequences and implications of these models and see if they have a practical use and see whether these models are then confirmed in reality but the question is that information is often limited you have to change models so that you can match models with the reality and there is a continuous dialogue between data and reality so this job is fundamental for discipline like economics with the passing of time greater amounts of data and information has been made available and it is possible to examine this data pretty fast today and the work of the econometric person has become central and it is very rare today to find a scientific paper which doesn't have an empirical part in the past you often had just purely theoretical papers without the tests based on data analysis so the job of Daniel McFadden is particularly important and it is a type of job the all of us appreciate and value daniel McFadden received the Nobel Prize in 2004 the models of discrete choices these are models which match reality with empirical data and which consider situations where there are qualitative choices to make binary choices whether to choose an A or B whether to participate in the labor market or not we spoke a lot in these days about women's jobs and women's participation in the labor market so here there is a binary choice either you have women participating in the labor market or you don't have women participate in the labor market so the choices should be yes or no and the fact that choices are distributed in this way entails problems in terms of traditional measurements and analysis of data and the work of Professor McFadden enabled us to match data with reality and to infer implications from these data in his career he received barriers of knowledge wins and prizes I'll just mention a few the Jean Mackin and bento in 1975 and then he became fellow of the Academy of Arts and Sciences in 77 and then in 81 he became a member of the National Academy of Science and then in 1986 he received the bridge medal for the econometrics society and in 2000 he received the members price of economics or the Northwest University it's so important prizes and acknowledgement and then he graduated in physics and he also was a professor of physics and then he also dealt with psychology behavioral sciences so he has a very wide-ranging set of interests and if we look at his scientific CV which is really impressive we see that he made contributions in very different disciplines he published articles in the best international journals on production Theory transport and then of course econometrics and theory of econometrics and then he wrote articles on growth development pure economic theory mathematical theory and then he wrote about energy health and environment as well so people who have so much flexibility mental flexibility are really able to apply models to different disciplines and to different problems and issues this evening the title of his presentation is doing well for doing good how to reconcile private incentives profits and the common good so this is a central issue in the festival of these days and in this case we have to measure incentives measure performance and he will tell us about that I'd want to take away any further time and I give him straight away the floor and at the end of his lecture as usual we will have we will entertain some questions let me say first that this is my first visit to Trent oh and my wife Beverly and I have very much enjoyed the sights and shops and there's a pedestrian strolling in Trento lovely city second I want to apologize I do not speak Italian my wife's cousin in Tuscany once said to me how can you call yourself a professor when you don't speak Italian and it's true it's been a burden in my career but please indulge me then I will speak through a translator I'm going to pose initially a question which i think is a provocative one which is should politicians be paid for good social and economic performance by the state let me be specific suppose each member of the Italian parliament or offered a bonus of 10 million euros if difficult but realistic targets were met by the end of 2015 on growth unemployment and the government deficit would the possibility of achieving these targets increase substantially in the presence of this these incentives now this kind of scheme resembles the compensation packages offered to corporate executives these packages are intended to align their interests with those are the corporate stockholders and a question for you is this is it appropriate to run government like a business and compensate politicians like corporate executives after all many of their tasks are similar and we would all like them to work hard for our public interest as stockholders of the state one comment and that is that Benjamin Franklin the famous American statesman of the seven 18th century said that the key to success financial and in reputation is to do well by doing good if you are diligent and industrious in your work in an appropriate framework you can do well as a result now what about targets for the Italian economy and what could you possibly ask of the Italian parliament italy's growth rates have been pretty bad over the last five years there is now a slight improvement it appears that the economy will go up by about 1.1 percent next year with current oacd projections suppose we asked the parliament to do what they needed to do to increase that growth rate to two point one percent so on a 1% increase in the Italian growth rate that's by the way would not be out of line with what this economy has achieved in the past although it would be higher than it's been achieving for most of the last five years if that target were meant for matched then the Italian economy would have 13.7 billion euro gained in its domestic product compared to the OECD forecast so then let me come back to the question but it'd be worth it to offer the 315 members of the Italian parliament 10 million euros each if they are successful in achieving a target of 2.1 percent growth in 2015 without increasing government debt now if this target to reach this would mean that you would have 13.7 billion dollars more now some of that would have to be collected as taxes to be to paid the Parliament to Parliament and in total would we see three point one five billion three point one five billion euros that but even after the public would have an additional ten point five billion euros in their pockets they we get 76% of the additional 1% of growth back now your reaction to this is probably this my reaction which is this this would be an insane and disgusting thing to do to pay politicians this much and a lot of people are already quite unhappy with the high compensation to corporate executives and I I understand the sentiments about this and we'll talk about that but the first question would be could part compartment actually do this because if you offered this kind of a strong incentive to the Parliament is there a possibility they could actually get an additional 1% growth out of the Italian economy and I think the answer is they probably could with the unemployment rate of twelve point six percent and they famously inefficient labor force Italy has a lot of room to expand its productivity and expand its income you could use the private sector by incentivizing firms to put people on to work in in productive ways or you could engage in public infrastructure such as expanded education Wi-Fi free Wi-Fi throughout the country so forth and I believe all these things put together organised by clever politicians that really work together to accomplish something for the youth in society rather than expending their energy trying to assure their election in the next round could they could accomplish quite a bit that they properly otherwise will not now in this talk I want to talk about these kinds of economic incentives and compare them with the incentives that Paulson's currently have and essentially ask the question whether sharpening incentives particularly economic incentives could improve performance there a series of topics that I'm going to cover slide by slide one the issue of democracy and how we how we are satisfied we are with it second lots of what are called principal-agent relationships and economic and social settings and legislators parliamentarians are our agents operating ostensibly on our behalf I'll talk about the economic theory of agency and performance-based compensation and then I will talk a little about the corporate experience with these kinds of schemes for giving people incentives to work hard I would say a little bit about the analogy between corporate and state governments aren't they're not the same and it's certainly not automatically true that what's what makes good businessman makes good government managers but we can talk about what the similarities are we'll talk about a one obvious issue in using these kinds of corporate schemes for government officials which is who gets to set the price how could you go about doing that and finally I'll just make a few comments about the Sociology performance if you have if you're a group has to do a task what's the best way to incentivize them to do that is it to give everyone an incentive is to give the whole group in incentive nobody gets it unless everyone succeeds and so forth that's where we'll go with this begin where the issue of democracy and discomfort with democracy don't grab government's is much of admired in theory but it often proves unresponsive and inefficient and practice it leads to public disdain for politics the traditional legal and constitutional approach to control of politicians is through approval by an informed and vigilant vigilant electorate and also by judicial enforcement of a code of conduct which is supposed to control corruption in practice approval or rejection at the ballot box is a blunt tool for regulating a political behavior voters are often poorly informed manipulable and capricious and politicians don't necessarily get reelected by performing for the public good second judicial checks to control corruption are somewhat even and quite slow so one comment before I go on I'm going to talk about how to make government more efficient but one point about democracy is that one of one of the features of democracy is it distributes power in such a way that it's very hard for any single individual to accumulate power and pretend perhaps abuse it and to some extent the inefficiency and petty corruption that occurs in democracy is the price price we pay for protection against abuse of power so so always keep in mind as we talk about how to think to make things more efficient we want to be careful not to lose the one of the great advantages of democracy which is to keep us free from abuse of power one question is do traditional ways of acute approving and compensating politicians make democracy less efficient and it needs to be and I think the answer is yes the way we fund political campaigns through parties and sponsors who have special interests to to focus the loyalty of politicians to their base rather than to the broader public the salaries that politicians received is true in the u.s. I think it's true here as well are relatively low compared to what many of these politicians could earn in the in the private sector and they do not have ties to perform so being a more effective legislator doesn't get you a higher salary as a result the politicians tend to be individuals who are more willing to seek their rewards through special dealing while in office and through connections that they have after they they leave office and it may discourage more constrained individuals from going into politics it now it should be in the public's interest to pay for and get politicians who are skilled and diligent and working on the public's behalf so the question of how politicians are paid should in the end be an economic question are we paying enough and our opinion in the right way to get the kind of performance from the current politicians that would benefit us all one general comment many politicians are dedicated public servants but is a mistake to think that dedication is enough moral fiber is strongest when not cut by self-interest now there are other arrangements than economic incentives that might improve governance I will mention a few I'll propose them as questions because these are not recommendations they're just questions and there's a tension between these and between the between these and the fundamental idea of democracy that it's a every one-person one-vote is he tend to be elitist and one of issues in any kind of schemes to improve the efficiency of the democratic process will often create a tension between the the ability perhaps the ability the elite to be more discriminating in their politicians and their own special interest moves majors it may give them an unfair advantage so what could you do you could require a more informed electorate you could license only voters who are literate on the candidates and issues an election you can require attendance that voters at debates and discussions before they vote you could allow internet voting on detailed issues so instead of or in addition to broad votes for candidates and parties people could express specific opinions on specific kinds of public project projects and initiatives second you could require more qualified candidates you could make a degree in public management of prerequisite for office you can require new Parliament trained parliamentarians to train and work together experimentally before they go to work for the public you could provide external audits and information on performance sort of like movie reviews of political performers you could also reform campaigning and campaign finance so that politicians are less concerned about the base from which they can get a financier there are campaigns finally you could you could change what politicians do I currently politicians will have to get elected they have to be executives who decide on broad policy and to a substantial extent they after the managers that are making decisions on day to day projects what to do and who's going to do them and so forth it's possible that some of those lasts more technical jobs should be handed off to professional managers and one way to do that to training trained cadres of highly qualified professional managers make them the operating officers and leave the politicians in the in the executive offer officer position now I'm going to talk about principle agent relationships and economics and social settings and apply those to government many human interactions economic and social involve a first party a principle engaged in with arranging for a second party or an agent to act on their behalf such agency arrangements are common when the agent has some specialized information or skills or time and thus has a comparative advantage in undertaking certain kinds of tasks virtually all commercial transactions have some elements of agency with each side in a transaction having some comparative advantage of special special knowledge here are examples of agency relationships the principle may be a landowner and the agent be a sharecropper in agriculture a very traditional arrangement the principal may be an employer and the agent employee the agent principle the patient the agent a physician an insurer and they people they insure a child and a parent who is an agent trying to work on behalf of the child a homeowner and the repairman the plumber a corporate shareholder and a corporate executive is working for the shareholders and finally politicians who are working for the voters those are all examples of agency relationships now there's some key features features of agency there may be one or more principles not necessarily all agreeing and there may be one or more agents again not necessarily all agreeing with each other contracts between principals and agents may be informal or formal principals and agents often differ in their objectives attitudes towards risk and time preference so one of the issues and how agents are compensated is going to be to balance those conflicts there are generally a symmetries and skills and information and in costs of monitoring and auditing and these mean that contracts between principals and agents must rely on common information and usually cannot perfectly align the interests of the parties now there's an additional feature and that is matters between principals and agents happen in markets that determine the qualifications and participation thresholds of the parties for example politicians compete in the in an election that's a kind of a market and the compensation they face is one of the things which determines whether someone enters that market or not there are two major issues in agency they're called moral hazard and adverse selection when an agent has information that the principal does not such as level or productivity of the agents effort there's a risk of what's called moral hazard where the agent undertakes actions that are in the agent's interest but in contract contrary to the principles interest such as careless or unnecessary work secondly when principals cannot observe all the qualifications of agents the market may unravel with more qualified agents unable to signal that they're better and worth more than the less qualified agents this is called adverse selection by the way these are very old ideas some of the ideas predate economics as a subject by millennia to the first recorded commentary on adverse selection was in the Aristophanes comedy the frogs in 405 BC which jokes about bad politicians driving good politicians out of the market for governance and in Athens and that adverse selection became an econ always Gresham's law which is that bad money that you can't tell from good money will tender a good good money out of the market now performance-based contracting historically principal-agent contract contained incentives for the agent to avoid moral hazards and to work diligently on behalf of the principal the one of the original forms of agency was sharecropping and that's an arrangement between a landlord and a tenant that landlord provides the land and the and the seeds and the tenant provides the labor the crop is shared between them and the sharing arrangement creates incentive for the tenant to work hard to succeed in getting a good crop now he cannot influence in compensation or a form of basic agency theory but incentives other than current compensation such as deferred incentives are important in practice particularly for agency relationships that extend over time also if risk preferences of the parties differ then risk sharing as well as incentives that suppress moral hazard are important elements in them in a contract one question which will come up generally an agency it's certainly an issue in the executive compensation that I think if you propose it for politicians it would be an issue they're his dependable agents get paid more under performance-based contracts than they would otherwise and I think the answer is simply there's no first of all there's no particular reason that people would get paid more under performance-based contract than they would under a flat payment but there are two two things which would tend to make that not quite true the first is that if there's a trade-off between a guaranteed salary and a bonus then you and you you pay a people more overall you may attract better people who can perform better and they'll get paid more and so as a result so that the presence of a performance-based contract may cost you more because you get more for your money secondly if agents are risk-averse then giving them a contract which requires them to take risk they don't get the payment unless they can accomplish particular targets then you have to compensate them for assuming that that risk so that will be an untraditional cost that the risk management if performance-based contracts are successful then both the principal and the agent received more than they would otherwise that's the whole point of that kind of contractor now is performance-based compensation of politicians our remedy for self-dealing and corruption now first of all higher base salaries and performance incentives may make a political career attracted to people who are reluctant to engage in self-dealing and corruption so that may actually improve that aspect of politicians behavior but by and large incentives to do good doesn't mean that on the side you can't also do some bad and maybe those two things don't necessarily interfere with each other containing corruption that interferes with economic and social progress is important but it's possible that too much emphasis on limiting eliminating petty corruption rather than on promoting the public interest would actually interfere with effective governance I'll give you a couple of quotes from American politicians self-interest interested once there was a political boss named Hurley who said corruption is the lubricant of politics another political operator Jesse hundreth said money said money is the Mother's Milk of politics and there is this Chinese solution to corruption partly successful actually which is keep the rate of growth of corruption below the rate of growth of the economy so that it's too difficult to get rid of corruption but hopefully you can keep ahead of it now what shape of performance-based conversation is more effective here's some examples of different kinds of schedules in which you are relating the pay that you give to a an executive or to a a prime minister say as a function of some measure of output such as the growth rate of the economy one possibility is a linear scheme that's the second figure year but there are others the one that I started with would be like the third figure a a lump sum payment which is made only if a specific target is reached now these kinds of threshold bonuses can be quite effective if they're realistic but the problem is that if they become unrealistic realistic then they lose their effectiveness as incentives and often schemes like the the last of these will often be the most most effective now I'm going to not spend time on technical slides but effectively if you offer an agent a particular incentive schedule and the left-hand diagram in this slide is a straight line it's in a schedule they hire the output the more the the agent is paid and the agent dislikes putting in a lot of effort but certainly likes the income and would pick out an effort level which would maximize the agents utility given that schedule now it can change the schedule you would change the point of view of utility maximization and change the level of effort then what the principle would like to do is vary the schedule in this case say bytes varying the share of the extra output that goes to the agent in such a way that the payoff to the principal is maximized generally what will happen is that the agent is not adverse to risk then the optimal contract is always essentially to have the agent bear all the risk so that would mean that you for the in the case for parliamentarian the pol interion would would pay pay a great deal of money to be to enter the Parliament but then wouldn't have the opportunity if they could keep the economy going very vigorously to get that back on a great deal more but if the agents are risk adverse which most people are and have limited ability to absorb losses then you'll have some sharing arrangement like a share agricultural share cropping arrangement now what's the corporate experience with performance-based compensation it's an intensely controversial subject we see in the US and I think in other places a rapid rise in the compensation of managers relative to workers and that creates a lot of a lot of dissatisfaction that contributes to income inequality which is a a drag on economies both socially and economically and there are two major complaints about corporate executive compensation one is that the incentives are often poorly designed and ineffective we'll see whether that's true but I would say that in part this comet is correct executives are often too too short-sighted and that's because of the way the incentives are designed and they have an incentive to some have to manipulate the company books and that's also a problem the second complaint executives earn rents by capturing and manipulating boards of directors distorting contract incentives and performance measures and that's a that's a serious problem with the executive compensation and I think it leads in many cases to excessive compensation and if you think about applying it to politics there is a question of how in the world would you control this when the politicians we tend to have the ability to take over hey so I keep but before we address any more of that let me ask the question does is it in fact true that performance-based pay improves firm performance in the corporate setting by the way here's a slide this is taken from a paper by Ben lorica in 2006 median CEO compensation by my country I don't know if you can see these figures but the these numbers are very quite high and Italy surprisingly second only to the US and the level of which it compensates its its corporate executives how rude powder companies do if they pay or they prefer they provide very high performance incentives to their executives here's a here's a just a little example ten major US corporations drawn at random and this shows their return on investment as a percent on the vertical axis and their compensation and this is the part of the compensation that they wouldn't get unless they are that's it associated with achieving a high high return and what you see is that there is a positive relationship it's rather noisy and so our first question is does that kind of diagram indicate a causal path from increased pay to higher return on investment is it true that it if you give these executives incentives to increase the return on investment that it were they can actually do it they actually do do it now the trend line is by itself doesn't rule out the possibility that some of this is spirits spurious that Grant random variations in the fortunes of the firm could both increase pay and return on investment without the executive necessarily being responsible for it that's an issue there there are several issues in understanding the relationship between performance based pay schemes and their their payoff one is that there's heterogeneity and how how effective the targets are and second there's heterogeneity in the circumstances affirm some executives or have stationery firms that are in cruise control some are salvaging firms that have been in trouble and so forth and if you combine this all together you don't necessarily get a good picture of how well those executives are doing but there's a very substantial literature in economics on executive compensation and this literature by and large does a careful job of eliminating some of the spirits effects entry and identifying what the real impact of executive performance pay is and the finding is that in terms of realized return our firm growth there's a strong and positive relationship between managerial pay and the performance of the firm and as you you give a performance higher performance pay incentives and these the executives will tend to do better for the firm's a few other comments from this literature outside directives directors in a corporation improve the tuning of performance pay contracts and are more effective in getting rid of underperforming managers equity based compensation seems to do better than bonuses rachet standards where your compensation is based on doing better than last year tend to be tend to blow up they tend to be counterproductive and finally in a one study of Italian compensation it compares family-controlled firms and non family controlled firms in family control firms in Italy the exact executive compensation is higher but firm insists no better so there's a case where rents are being taken now what about applying any of this to government I mean there are issues and even even using it for corporations but now the question is what about using it for government officials there certainly are issues one while the interests of stock holes are at least roughly aligned and focused on maximizing expected discounted cash flow the general public is heterogeneous in its preferences and for not only for the outcomes of state governments but also for the ideals and form of governance what's true is that at Parliament is the analogue of a Board of Directors and there's no obvious mechanism under which a disinterested party body could establish and monitor compensation for politicians in Parliament now incentivizing agents to work together in a legislative body involves group incentives and social dynamics that are not covered in the standard theory of agency there are two areas of economics to have to say something about this one is the public theory of public goods provision which gives some idea of how you might determine how to compensate politicians and finally behavioral economics may show some of the effects of social interaction on the effectiveness of various performance schemes so question one who sets the price when you have multiple objective principles one possibility is using what I call economic juries I'll describe this in in some sequence suppose people disagree on the priorities that should be given to public projects and actions including performance Mitch marks and incentives given to politicians to induce effective governance is there any way to elicit honest valuations and combine them to make social judgments and the answer is yes in principle but most of these schemes are theoretical schemes of economists that do not have a lot of practical experience I'm going to describe one first the slide is difficult to read but I'm just going to describe it in words it's a method called Dubin method and the idea is you have K projects the project might be alternative and scented shares for the politicians for example then the state must choose exactly one so then you ask each voter to write down a euro value are willing to pay for each of the alternatives that are available with under the following rules what they write down can be either positive or negative for any given project but there's some overall the project's has to be non-negative I should say in in total they're willing to pay for at least one of these projects has to be summed up has to be greater than equal to zero if you then array all of the people in a big array you know in columns and you put all the projects in rows you have an array in which all of the column sums are non-negative an implication of that is that at least one growth sum has to be non-negative that means that for at least one project there's a weighted for all the people who have positive negative willing to pay for that project can be compensated by the ones that have positive willings to pay there is a way to redistribute people's stated values in such a way that everyone is is that the is satisfied with that outcome so that is a possible way of making a social decision now that particular scheme is doesn't does not have a property that economists call incentive compatibility it's possible that you would try to game that voting system by if you think there's a one of the projects is sure to win and you actually think it's a very good project you might act they give it a negative vote because you think it's going to win anyway and you get paid a little bit as a result so that's a strategic manipulation and if you have too much of that going on it will overturn this and it will not work right but there are other schemes that are actually more effective in controlling these problems of manipulation let me define first of all what's called an economic jury a jury is a body of experts that's utilized to reach some kind of evaluation finding our verdict examples their committees used to judge wines juries and panel to determine remedies and civil litigation elected legislators would be an example of an economic jury and subjects and surveys their experiments who are asked to rank or value public projects such as an anti anti-smoking campaign would be an example of an economic jury not as the design of mechanisms to ensure representative recruitment and accurate reporting of values by jurors is a principal agent problem now you could ask an economic jury to evaluate and judge an economic matter and choose among public projects and what the question then is is how can you give the jurors incentives so that their interests are aligned with those of the general public or at least our alignment such a way that they will be truthful in their valuations so here's how a jury any kind of a jury would work you would randomly select a jury would say Capital J members from a population of voters the question of the jury jurors than the value they're placing on a public project each individual giving his or her own value you'd use the incentive compatible mechanism I'll explain what that means and they're asked to give an honest value for what they think a project is worth and the question is how does how does this actually work first of all what's an incentive compatible mechanism and solicitation method and which under reasonable expectations and behavior each juror has a self-interest in responding honestly so that's in the Senate competitive I don't mechanism and various mechanism or or less incentive compatible it's it's actually impossible to design one that's perfectly innocent incompatible but at least in jurors that are isolated from each other you can do pretty well I'll mention two and I'm actually going to in interest of time simply just tell you what these are without going through them one is a mechanism cultic becker DeGroot mechanism and it's essentially a voting mechanism which people write down the value which they would they would vote for this project or not and it turns out that under some reasonable conditions that actually in a jury a reasonable size will will get the real incentive to be to give accurate responses because if if they're if they're pivotal in the sense that they could make a difference in the vote and in a relatively small jury many people are often pivotal then they actually will personally suffer from giving an incorrect response it turns out that under this mechanism it's as strong as a strongly dominant strategy for each person to be honest in their response a second mechanism and a more actually a more famous one is called the Clark groves mechanism in the form I will use it it's modified by Jerry green and by jean-jacques Lafont so the CG GL mechanism and the idea here is you pick a jury and effectively you tell each juror you will get a payment which depends on the net benefit that all the other jurors other than yourself see and now you get to tell us you're you're willing to pay for this and it turns out with considerable technical analysis that once again you can't as an individual juror you cannot gain by Miss Roselli misrepresenting your your value so that this is a what's called a provision point mechanism and there's a slide I'm definitely going to go over quickly there's you can demonstrate that that in that kind of mechanism there no one has a positive incentive to cheat and give a giving a false answer so conclusions on economic juries a relatively small economic jury is adequate to determine with very small cost of error whether projects are desirable for the public and there's actually two reasons why are relatively small jurors jury economic juries will do a good job one is that under very general conditions you get a very rapid statistical convergence so that you think well it's a sample that's our one over here one over J sampling where there J jurors is actually you get better convergence than that statistically second you can think of the problem in this way if it's very very clear that a project is worth doing or not worth doing you can find out that out very quickly with very few people by asking very few people about it it's only in the case that the that the project is just at them just in the margin where there's it's really hard to determine whether it's desirable or not that you would need a lot of people to actually make the choice so the implication is you don't no reason to use a big jury because a small jury is going to be able to decide the easy projects and it can with the hard projects it's not going to matter too much the mistakes you make are not going to be so serious in any case one example this is related to my own research I've done an example in which I could consider a project that has a cost of a 20 million euros and a in a city of 240,000 voters a little bit larger than Trento a little smaller than Trentino how many jurors do you need to make a error of very low error decision about whether certain projects should take place in the city and the province of Trentino say such as putting in Wi-Fi throughout the the system we're having bicycle stands on every corner and the answer is if you take a journey of 125 people will do this with a very high degree of accuracy less than 2/10 of 1% possible cost of error so the the implication is that you can actually use the public instead of them requiring everything to go through politicians one way to get better governance is to ask politicians to do more of the executive tasks and actually go back to the to the public to economic jurors and to technicians to do more of the operations and technical tasks one final comment what about when you are trying to compensate politicians as groups at Parliament as opposed to individuals but he can always tell you about that well there's one little economic game very simple game that economists have studied at in some depth and that gives some very interesting implications for how people behave in society as opposed to behaving according to very traditional individualistic economic rationality rules here's the way the ultimatum game works - there are two players generally strangers and they're offered a reward if the first player can make an offer of some division to the second player which the second player then accepts and if the second player accepts then they get the reward in that division and if the second player says no they neither get anything now that's a game where rationality says the second player has nothing to lose if he's offered any positive amount so you should always say yes to any positive amount and if the second player is going to do that then the first player might as well offer the smallest positive amount that's not the way people play the way people actually play in many cultures is that the first player will offer about 40% and that will usually be accepted and if the first player offers less than about 20% that will often be rejected that's sometimes called altruistic punishment so what is the one of the implications for for politics the implications are that you might you may be able to use performance incentives in a way in which you essentially playing a version of the ultimatum game in your Parliament you you get the first player I guess the Prime Minister a very very large reward if the problem is successful in serving the public interests interest and then the then there's a potentially ultimatum play in which each parliamentarian would accept to reject the distribution that's offered by the Prime Minister no would that what could you say about such a scheme first of all in many traditional democratic organizations such operations are often denigrated as buying votes and they're often even illegal but if you're trying to design economic incentives to optimize the outcomes of governance for the public such operations should be considered legitimate forms of contracting to improve performance a comment social pressure from performance paid to groups can be intense and incentive schemes that reward information sharing and collaboration often more effective than those of isolate pivotal groups finally this is an area of economics has not been much explored that is to playing ultimatum games within groups and for possible political applications it would be interesting as a research topic inclusion the control of Politico and the interest the public is a principal agent problem economists have found that in agency problems carefully designed performance-based compensation schemes can substantially increase payoffs for principals the most experience with performance-based compensations for corporate executives where these schemes are controversial and widely criticized and often seem to allow excess rents to executives but nevertheless contribute to shareholder value there are issues in applying corporate governance models to the state nevertheless it can be argued that we get what we pay for in our compensation of politicians and it's worth considering using some of the tools of agency theory to create economic and social incentives for politicians to work more diligence tells only for the benefit of the public to do well by doing good thank you comedy several in its atemi as i told you in the beginning the topics that were covered by daniel McFadden ranged through many things that were discussed at the festival and I refer to the alignment for instance between governance schemes and compensation and corporate situations we have listened to many theoretical observations but there is a direct repercussion of all this in reality actually we have received many important practical inputs today in Italy there was no mention about schemes to change the compensation of Italian members of parliament our members of parliament are very well paid rather the opposite is asked to reduce their compensation and when attempts were made to link the performance of MPs to their to something actually the proposal was made to work on the inputs that is for instance to pay them more if they are physically present in Parliament and if they vote there and no mention was made of the output of their work but much has been debated actually also in Italy about using a private cooperation scheme sir for public administrators and we have discussed about this in the framework of the festival as well by the way yesterday a bill was mentioned by Prime Minister Renzi he said that soon there is going to be one bill and one law decree about this about the fact that in public administration there will be schemes similar to corporations and that would could even be extended to linking the income of state employees to the GDP so this is something which is absolutely in line with the things that were said that Professor McFadden personally I do not think that state employees can have an impact on the GDP growth in Italy why should give give money to people based on that I don't think that the GDP is dependent on the work of the civil servant anyway let's now take a couple of questions to three questions not more than that because then we have to also to draw conclusions of the overall festival but now the floor is open to questions down there there is one person asking for the floor there thank you good evening can you hear me first of all I would like to thank wholeheartedly the organizers of the festival and the speaker of course for his contribution thank you for this lecture I would also like to tell you a bit about my life you see because I'm 86 years old and I've worked a lot in the volunteer sector in the organization for organ donors and then the Red Cross and then blood donors associations so it is correct to speak about economics obviously but we should also talk about the relationship between children and parents and young people and the elderly thank you but we did so we had a whole festival on intergenerational relations I would like to have questions on the lecture well of course it is important to speak about economics but you should also talk about the family because families are thank you thank you we understood and labor because you see labor has changed so much in the past we worked in the countryside and at the end of the day we went back home and we were fine and would have time to spend with our family things now are not like that anymore thank you so much for your contribution we have another person asking for the floor here I don't know whether there are other people who wish to take the floor your legend it seems that you take for granted that the outcome of all these games and strategies are actually measurable but in fact as you said there's an information asymmetry between principal and agent so how can you deal with these problems so how is are these performance actually measurable since the people that actually talk to the public about those outcomes may modify information according to their interest to the question the question you posed is a a critical one actually and I think that we know it's speculation style worked in politics but I think we already see in the corporate sector that there is a there are serious issues in what you measure and what the what the targets for the executive would be and I don't have I don't have a simple answer I think that right now many corporate compensation schemes do rather poorly in terms of the long run benefit of the of the corporations and I think that if you try to apply these performance schemes to politicians you're going to have many of the same problems but I still think that we should think about it I think we should think about how how what kind of incentives if politicians have now and to what extent the incentives they have now distort their their behavior so that it does not work in our interests and simply ask the question is could we do could we modify it in any way that would make it better that the corporate model is one possible model but I think it would be very good for a Congress public becomes to think seriously about schemes somewhere somewhere in the whole spectrum of possible possibilities Damona demanda we say there is one question here I see no other raised hand sir well yes there is one person there up there please stand up so that we can see you good even professor boy Eric good evening professor McFadden it is very interesting to apply the game of theory to the possibility of a better governance but I think that it is not practically possible in terms of having for instance a compensation for people judging of the programs of politicians I don't think that it would be democratic applying money to politics this is what I mean I think it's not democratic actually and that is one reason an additional reason is that it is difficult for people then to pass an independent judgement about programs because they are influenced by political parties and political parties established beforehand a priori what should be important and they tell people what should be important for them I agree I agree with the difficulties but what I would say just is that it's true that you would like an ideally have a political system that is not driven by money but even without considering new performance-based compensation schemes we have politicians that are influenced by money money is one of the factors that enters and it enters in a way which I think works to the public disadvantage when it mostly comes from special interests and and I think that we need to think seriously about how to replace money special interest money with money that would be directed to the politicians if they perform for the benefit of the whole demand across sue McGarry there is a question up there I don't know how you can get the microphone shall we try interpreters cannot hear unless the question is asked in a microphone interpreters do not hear unless the question is asked in the microphone if it is in English we can't translate that into Italian thank you there's any incentive scheme that's based on output has the possibility that you will compensate your manager your agent not because of the effort they put out but because of chance but that's what that's because in in the nature of this relationship a chance is going to enter the the output determination I think that if you want to use these schemes to induce more careful management than you then one of the prices is that sometimes the executives will get money for it's not really justified by their own individual effort now you raise the question of whether there are clear targets that would be in the public interest and I would say it's true it's accomplished this is a complicated business and it needs to be it needs to be flexible but looking say at the Italian economy I would say that if the Parliament had a target which said that a great deal of their money would be taken away if the government deficit went outside target ranges our unemployment when above target ranges for example I believe that would focus their attention and perhaps replace some of the intra-party squabbling with an effort to try to make sure that they don't all get great deal income loss from poor performance and courting the demand we take another question you can create the incentives for all of the politicians and and to align them but if they still need to be reelected in order to make that pay incentive in the next period and it's very expensive to be elected in terms of raising a lot of money isn't there still a tremendous distortion in the system and secondly don't the special interest gusto bribe the juries or the rule makers who make the incentives for the politicians and the game just is carried on on a different frontier well unfortunately I'm afraid you're right about the second that if it it's very nice to think about these things theoretically it's of course very very challenging to make them insulate them in practice from the people who are very skilled at manipulating schemes like this benecrine okay see a motor a la primera daddy mask in the first presentation by Professor Maskin at the beginning now you know that all the presentations of the festival are in the website I'm saying that also to all those who are watching us online now this is the last conference of this festival just a couple of words to draw some conclusions to strike a balance of these days well I can tell you that data on participation are satisfactory you yourselves have seen how many people participated in the festival there are figures on visitors to the website which are extraordinary indeed which will be published in the next few days and then quality is important yesterday we asked the people who were in the whole santa chiara before the meeting with the Prime Minister Renzi we asked them how many students were there were three people out of four we're students so we appreciate we're very happy about such a larger share of students this morning there was the prize ceremony of students and they spoke about their experience in the festival and it was very nice and then the public when too many debates also the most difficult ones the most technical ones the most abstract ones I have to say that also the questions made during conferences are indicative yesterday for example there were questions to the prime minister unfortunately he could answer only though three of those questions but he took notes of those questions and slowly he will answer those questions and he will publish them in the website of the government and then a second reason for satisfaction has to do with the impact now judging based on the coverage the media coverage of the festival well today for example we had a large coverage in Germany in the German media we have many journals many newspapers participating but then there are other indicators which perhaps are less direct to measure the impact and then the impact is measured by the presence of so many people here and the public certainly represents a new ruling class many of the people who have been here in these days are people who work our people for whom time is money and we've had managers executives who accepted to stay longer then expected so they came they delivered their speech but then they stayed longer and this is a measure of the interest and then in terms of content we had a lot of suggestions of course we cannot expect a festival to produce definitive answers but I think that some food for thought was presented in terms of women in the ruling classes and also in terms of the elections legislation some suggestions were made and then university contests and competitions so I think that many ideas which were discussed Judis during this festival will be useful contributions to the national debate now there are various reasons for satisfaction and then also with reference to professor McFadden the slides the texts will remain in the archive of the site of the festival so you can go to the site and visit all the various sections now I like to invite here to the stage all the people who have made this festival possible starting from the president of the provincial government Rossi the Dean of university director of university misses the practice and then the scientific committee with its president cheap or late to say Bella Terra and Paolo Collini and then many other people who have worked hard to make this and wonderful we cannot invite them all on the stage there are many people who have made an extraordinary work and job and as representative I'd like to invite medina de Francesco sí sí seguro for coming many bonus expressing some things Teterboro our University Giuseppe litera and Marilyn a different choice go who has taken on the responsibility of this work I'd like to thank the public the audience the people who have participated in the festival I'd like to thank a professor McFadden and by thanking him I thank all the speakers who have honored us and hopefully we will draw some useful conclusions and implications also in Trentino in relation to the compensation of policymakers this is a very hot topic I appreciate it I like very much so the price award in ceremony this morning and there was a young lady a young woman this morning who said I thank you because in this festival I heard things that normally I do not hear well I don't know if we can call it a success but certainly I breathe it mayor of success an era of attention we tried to improve the spin-off also in relation to the german-speaking community and speak German speaking world I think that never as this year the festival has been able to combine the contents of the presentations with the ability to provide opportunities for discussion and debate now we've just had the European elections and the electoral campaign let me say that has not really been able to go in-depth into the topics while the festival has been very useful in my opinion and I like to think that many of the youths participating in this festival can draw some useful conclusions and will be perhaps a spurred and encouraged to become the ruling classes of the future for the for the common good of course this is our hope for the future thank you and see you next year you