house property capital gain tax

sven beckert: ok,good afternoon. welcome. it's good to see somany of you here. my name is sven beckert,and i'm delighted to be able to welcome youhere today in my role as the co-chair of the programon the study of capitalism here at harvard, and also as afaculty associate of the center for european studies. i would also like to welcomeyou in the name of grzegorz

ekiert, who is thedirector of the center for european studies,and chris desan, who's sitting right here, who isthe co-director of the program on the study of capitalism. it is great to seeso many of you here. i know many who wouldhave liked to be here weren't able to join us. but i'm particularlyhonored to be able to welcome professorthomas piketty as our speaker

this afternoon. the program on thehistory of capitalism started more than 10years ago here at harvard. and when it started,an event like this seemed almost unimaginable. we had hoped to make morecentral matters of capitalism to debates in history andto bring other disciplines, including economics, into abroad discussion on the history and the present-daystate of capitalism.

and at first, such hopeseemed truly futile. our first seminarsattracted very few students. and when i began to lectureon the history of capitalism, many well-meaningcolleagues advised me that very few studentswould be interested in such an old-fashioned topic. things turned out tobe quite differently. and the study ofcapitalism has not only become extremelypopular here at harvard,

but has become one of themost hotly debated topics in contemporary culture, witheven pope francis and the queen chiming in on such debates. but no one has made the problemof capitalism more intriguing, more intellectuallyalive, and more urgent than today's guest,thomas piketty. when in the late summerof 2013, professor piketty published his 800-pageopus magnus in paris, few would havepredicted that it would

become a global bestseller. after all, here was abook that was chock-full of tables and charts, dealingwith decidedly yawn-inducing topics such as thehistory of taxation-- a book that, moreover,carried a decisively unsexy title-- namely capital inthe twenty-first century. and it was written in french. [laughter] few would have predicted that18 months later, the reading

of that very book would at leastlead one american ceo, mark bertolini of the aetnainsurance company, to raise the wages ofhis lowest paid workers, and that the presswould come to compare its author to a rock star. very few can claimthat their writings have had such an immediateimpact in the world. and it's a testament, not justto the brilliance of professor piketty's research butalso to his ability

to communicate thatresearch to a wider public that capital in thetwenty-first century has fertilizedpolitical debate and has brought issues of thedistributional effects of capitalism onthe political agenda in many differentparts of the world, including the united states. and of course,the book's success is also a sign of our own times.

thomas piketty isprofessor of economics at the paris school of economicsand at the ecole des hautes etudes en sciences sociales. he's the author of numerousarticles and books focusing on the interplay betweeneconomic development and the distributionof income and wealth over the longue duree. with a degree from theecole normale superieur, he entered a phd program atthe ecole des hautes etudes

and the londonschool of economics. and in 1993, he defendedhis doctorate dissertation on wealthredistribution-- a thesis that won the frencheconomic association's prize for the best thesisof that year. since then, he has taught atthe massachusetts institute of technology in thisfair city and worked as a researcher for thefrench national center for scientific research.

most recently, he has led thecreation of a massive database on world topincomes, which turned into the foundation for capitalin the twenty-first century. that book has now sold morethan 1.5 million copies. and it has become the mostwidely reviewed and discussed book of last year, sometimesin confusingly contradictory manner, such as when thedecisively anti-marxist journal the economist called professorpiketty "the modern marx," while self-identifiedmarxist david harvey

chided him for notbeing a marxist at all. and another journal, theamerican conservative, called him the anti-marx. beyond these confusions,piketty's work has been ofoverwhelming importance to reigniting discussions onthe distributional effects of capitalist development,the question of who gets what, and how that has changedover the past 200 years. in the course of providingpowerful sets of data

and interpreting them,professor piketty has created a newspace for discussing inequality and possiblepolitical responses to that inequality. in some ways, one couldargue that his most important contribution has been toallow us to think again about the future--the future not as an inescapable new liberalfaith that we must adjust to as best as we can, butinstead as something that we

can mold and shape politically. while his reading of thehistory of capitalism is often a dark one, his readingof the political possibilities of the human conditionis deeply optimistic. not only does heprovide political ideas on how to addresssharpening inequality, but he also givesus the mental space to allow us to think againabout the possibilities for alternative futures.

economic outcomes, hetells us, are fundamentally political outcomes. and he encouragesus to think not just about the run of necessitybut also about the run of democratic possibility. almost everycommentator has justly emphasized professorpiketty's contributions to the understanding of thelong history of inequality. but the pages of hisbeautifully written book

contain another contributionhiding in plain sight that is nearly as importantand that i, as an historian, appreciate in particular. he has rescued thinkingabout the economy from the monopoly of economists. professor piketty reachesout to other disciplines, and especially to thediscipline of history. and one can clearlysee the influence of such towering voices asfernand braudel and marc bloch.

his is not a past ofimmutable laws of development but a past that is political,a past that is contingent, and a past that is embeddedin cultural and social change. the trente glorieuses,for example, are not the result ofimmutable economic loss but the result of a confluenceof political factors and events. and thus, professorpiketty's work is grounds for optimismfor yet another reason--

that we might be ableto reignite a discussion on the economy, itspast and its future, by bringing in the voice ofhistorians, anthropologists, sociologists, and many others. without muchfurther ado, i would like to open the proceedings. professor piketty willspeak for about 45 minutes. and he will then befollowed by three very short, five-minuteresponses from members

of the harvard facultybefore we open up the discussion to the floor. and let me just introducequickly these three commentators before we move on. david kennedy, whowill respond first, is the hudson professor oflaw and faculty director of the institute forglobal law and policy at harvard law school,where he teaches on international law,international economic policy,

and also legal theory. he has published widelyon international law and just recently publishedglobal governance: new thinking aboutlaw and policy. he's also the mastermind behindthe institute for global law and policy, a projectthat brings together people across many disciplinesto think in new ways about globalization, itspower, and its challenges. stephen marglin,who will speak next,

holds the barker chair inthe department of economics at harvard university. his recent work focuses onthe foundational assumptions of economics and howthese assumptions make community invisibleto economists. this work reflected in thisbook the dismal science: how thinking like aneconomist undermines community attempts to counterthe aid and comfort these assumptionsgive to those who

would construct the world in theimage of economics-- ultimately a world without community. and last but not least, youwill hear from christine desan, who's not only the co-sponsorof the program on the history of capitalism here at harvardbut also, perhaps even mainly, the gottlieb professor oflaw at harvard law school. her work centers on theinstitutional underpinnings of capitalism so that we canopen them up to revision. she in particular exploresthe history of money

as a legal andpolitical project, and just published making money:coin currency and the coming of capitalism, abook that decodes the monetary architectureof capitalism. thank you again for joining us. and thank you again, professorpiketty, being with us today. professor piketty. [applause] thomas piketty: thank you.

thank you, sven. and i'm very glad to be here. i'm sorry that my englishsounds a lot like french, but i hope youcan understand me. i'm very glad to behere in cambridge. and my book firstcame out in french, and it came out in englishshortly after that. and now it has beentranslated in many languages. and out of about1.5 million copies,

i guess there's about 500,000in the english language, 500,000 in europeanlanguage-- french, italian, and german-- and about thesame, 500,000 in asia-- in japanese and chinese. but i think the english versionplayed a particular role. and i would reallylike to pay tribute to arthur goldhammer, who isright there in the first rank. there's no way icould have written such beautiful english and sucha beautiful book in english.

you can see my english. and i'm not particularlygood with foreign language. i do what i can. and my written frenchis much, much better than my written english. and there's no way icould have referred to the literatureor to express myself with the same clarity which artwas able to do by translating my book into english.

so i think this ishas been really-- it's more than a translator. i guess a translation isalways more than translation. and of course, there'sno [? czar ?] language for which i am able to checkthe translation was so good. so thanks a lot, art. so let me-- inthis presentation, i'm going to try to presentsome of the results of the book and reflect aboutsome of the debate,

also, that this book hasstimulated over the past year or so. so let me first say veryquickly that this book comes from a very collectiveproject of data collection. so i am trying, inthis book, to put the study of thedistribution back at the center ofpolitical economy and of the socialsciences more generally. but primarily what i havebeen doing in my research

is to collect a lotof historical data on income and wealth. and this is somethingthat i could never have done without severaldozens of co-authors from over 20 countries. so in particular, istarted collecting data on income and wealth infrance about 15 years ago. and then i was very fortunateto meet tony atkinson, emmanuel saez,gilles postel-vinay,

jean-laurent rosenthal, facundoalvarado, gabriel zucman, and many others which-- icannot quote everybody here. but it's clear thatthere is no way i would have beenable to collect all these data by myself. and this is an ongoing process. and so the book isjust photography of the kind ofhistorical material we had at one point in time.

but there will bemore data collection and more historical materialavailable in the future. so thanks to thisdata collection, we know a little bit morethan what we used to, but we still know too little. and in particular,there's absolutely no problem in my mindif people disagree with some of the historicalinterpretations that i give or the patterns that i find.

we are in the social sciences. there are different ways tointerpret historical evolution. and so the objective ofthe book is not to say, ok, this is the one way, theone law to interpret history. of course when you look at theevolution of income and wealth distribution over morethan 20 countries across three centuries, there cannot bejust one law or one principle. there are manydifferent institutions, policies, social control,political processes

that play a role. and i'm trying to illuminatesome of them in the book. but this is just the beginning. and there is a lot moreto do in this area. so in this presentation, and i'mgoing to present some results, mostly from part twoand three in the book so as to give you asense of the material. if you go to thiswebsite, you will find all the graphs and series.

and you will have a bettersense of what material is included in the book. so in thispresentation, i'm going to focus on these three points. i will start withthe first point, which is about the long-rundynamics of income inequality. and then i will move to wealth. and to a large extent, whatmy book is trying to do is to shift the attentionfrom the issue of income

distribution to the issueof wealth and property concentration. but i will start withincome inequality. i will show you thenumber of results about the long-run dynamicsof income inequality. and some of the mainconclusions will be the end of the kuznetscurve, the end of universal law, and the idea thatcountry-specific institutions and policy are absolutelycritical if you

want to understand the long-runevolution of income inequality in the various countries. then point twowill be about what i call the return of patrimonialor wealth-based societies, particularly in europeand japan, where wealth-income ratios seemto be returning to very high levels in those countries. and i will stressalso the metamorphosis of capital, the transformationof the different forms

of property, which raise newchallenges for the future. and the third pointwill be about the future of wealth concentration andconcentration of property. and i will argue that oneof the important forces-- certainly not the only one, butone of the important forces-- for understanding the futureof wealth concentration is the gap betweenr and g, where r is the net oftax rate of return, particularly for alarge wealth portfolio,

and g is the growth rate. and to the extent that thiswill be higher in the future than what it wasin the past, this might contribute torising inequality. so let me startwith the first point so a big part of thedata on income inequality that i present in the bookcomes from this database, the world income database. so you can't see thecolors very well,

but let me just say thatthe countries in red-- well, say in black-- arealready in the database. and the countries in blue,like brazil or algeria, are about to enter the database. so what it means tobe in the database is that we try to collect allthe historical data on income that we have fora given country. so usually, the only sourceof information on income comes from theincome tax itself.

so it's important torealize-- but you all know this-- that taxationis more than taxation. taxation is also a wayto produce information. it's a way to producelegal categories and statistical categoriesso that society can also accumulate knowledgeabout itself. and so when you don'thave an income tax-- the income tax was createdin 1913 in the united states, in 1914 in france, a little bitearlier in japan and germany,

where it was createdaround 1880, 1890. in india, it was introduced in1922 by the british colonizer. so almost everywherebetween 1900 and 1920, you have the creation ofthe modern income tax. and this creates, also, a sourceof information about income. of course, that's not aperfect source of information. but it's better thanno information at all. so in the 19th century, youdon't have an income tax in most countries.

so, like in france, you havethe contribution des portes et fenetres, thecontribution on the number of doors and windows. so you have all sortsof beautiful statistics on the number ofdoors and windows by department, which is quiteinteresting, by the way. and this is anhistorical data source that is certainly underused. but it's less interesting,in a way, than income.

and so this is true for income. this is also true for wealth--the taxation of wealth, in particular the taxation ofinherited wealth, transmission of wealth at a timeof inheritance. it's also a way to registerproperty and to register wealth and to produceinformation about wealth. so the taxation of wealth,in particular of inheritance, as much older than thetaxation of income, because registering propertyis very important to organized

society in general. so you haveregistration of property in very ancient societies. and in particular, in france,with the french revolution, you have the introductionof an inheritance tax that was prettyuniversal for the time. and this is why we cango back and with fellow historians, postel-vinayand rosenthal, to the french archives,to the late 18th century

to study the distributionof wealth in france starting in thelate 18th century. now, for incomewe cannot do this, but we can start atleast around 1900, 1910. and so we can study theevolution of income inequality. i should mention that one ofthe very positive impacts for me of the publication andthe success of the book is that it allowed us toaccess historical fiscal data and current fiscalfiles in new countries

where governmentwere not so open. so now in brazil, in mexicoshould be in blue on the map. in taiwan andkorea, in chile, we are accessing thehistorical income data which we could not access before. and so there are morecountries in particular that we try to coverin the emerging world-- in latin america,in asia, also in africa. we put a lot of energy ontrying to use data from africa.

so the countries that arenot covered in the book, it's not that we don'twant to cover them. it's just that sometimeswe had no access to the historical sources. and we now havemore access to them. so let me show yousome examples of what we find in this database. so let me present threefacts about inequality in the long run-- oneabout income inequality.

and then i will show youabout wealth inequality and finally aboutwealth income ratio. and if you want tohave a short summary-- the book is very long, andi should apologize for that. so if you want to have-- thegood thing about science, about real science and realscientific people, is that they wantshort articles. so when we had to publishthis paper with my friend and colleague emmanuelsaez in science last year,

this had to be five pages long. so if you want afive-page summary, you can read the science articlecalled "inequality in the long run," where you will have thesethree facts about inequality in the long run that are exposedin a very condensed manner. but i think you willlose some of it. in particular, you will lose thebeautiful translation of art. this will just be my english,so that's less interesting. but at least the basicfacts will be presented

in a very condensed manner. so fact number one-- factnumber one is the following. in 1900, 1910, incomeinequality was higher in europe than in the united states,whereas in 2000, 2010, it is a lot higherin the united states. so that's an interestingreversal of inequality in the long run, because theway i will interpret this is that changing institution,changing policies can make a difference.

so it's not that somecountries are always more unequal than others. it depends how you organizeyourself in a country. there are different waysto organize capitalism. and there are different levelsof inequality that go with it. so let me show youone simple graph. so this is the share ofincome going to the top 10%. so you have europe, the us. so europe, as you know, isa complicated continent.

but here i am goingto simplify things. and so what i meanby europe here is the gdp-weighted averageof germany, france, britain, italy, sweden. so it's not quite europe,but it's a big part of western europe, at least. and the generalevolutions are very similar for the different--at least continental european countries.

so let's call iteurope for to simplify. and as you can see, in factin all european countries, income inequality was higherin 1900, 1910 than in the us. whereas today, youcan see-- so there was a big decline in inequalityfollowing world war i and particularly world warii, the great depression. and in the 1950s,this is what kuznets finds in his famous study. inequality is less inthe 1950s than in 1910.

so all what we'vedid in this research is to extend the workof kuznets to many more years in many more countries. and as you can see, thischanges a lot the perspective, because in the'50s, you could have this optimistic view ofdecline and then stabilization of inequality at a lower level. now in 2010, it's avery different story, because you've hadthis very big increase

of the share of total incomegoing to the top 10%, which in the us went fromabout one-third to about almost one-half. so when you go fromone-third to one-half of total incomegoing to the top 10, this is not just an issueof a few individuals getting very rich at the top,which after all, nobody cares if it was justa few individuals. but here we are talking aboutsignificant macroeconomic share

in total income. so here, it looksvery smooth, because i look at decennial averages, soeverything looks very smooth. and that's convenientin order to focus on the long-run pattern. if you look at annualseries for the us, you can see that ismuch less smooth. in particular, ifyou take into-- so this is the samecurve as before,

except that i onlyshow you the us. and i look at annual seriesrather than decennial averages. and you can see, in particularbecause of capital gains, that the stock market cycle hasa strong impact on inequality. so you can see inthe us, in 2007, you have a very highpoint in inequality. now in 2008 and '09, thisis clearly not a good time to exercise your capitalgains and not to cash a big bonus.

so you have a declinein inequality. but then in 2012, you areeven higher than 2007. so in 2012, you have51% going to the top. in 2013, it seems tobe a little bit lower. anyway, you haveshort-run variations, but if you take thelong run picture, it's pretty clear that youhave a spectacular increase in the share of totalincome going to the top 10%. so if you just takethe decennial averages

and you compareto europe, i think you can see how big thedifference between these two different groupsof rich countries-- the us on the one hand,and european countries on the other hand-- is. if you were to putjapan on the graph, this is what you would have. so japan would be inbetween europe and the us, closer to europe in many ways.

so it's interesting to putus and europe and japan, because these are like thethree parts of the rich world. and you can see that theyhave different experience. so that's important,because sometimes people want to explainrising inequality just by talking about globalization. and the story will be, ok,you have globalization. you have china enteringin the world labor market. therefore, this isputting a strong pressure

on the wages oflow-skill workers, and this is what createsrising inequality. and i'm not sayingthis is not important. but i'm just saying that if thiswas the only explanation, then you should have the samerising inequality everywhere, because globalizationhappened not only in the us, but also in japan and europe. so sometimes wehave debates which are very muchself-centered in the us,

but also in europe or in japan. but i think it's importantto look at other countries' experience, and to realize thatglobalization is important. but then there aredifferent institutions, different policies, that helpto organize globalization and to allow broadergroups of the population to benefit from globalization. so what are these policiesthat can make a difference? well, i tried to analyzethem in the book.

but let me summarizevery quickly. so the rising us inequalityin the recent decade is mostly due to risinginequality of labor income. also, rising concentrationof capital income and wealth is starting to bump inat the end of the period. and a recent study by saezand zucman emphasizes this. but most of the action so farcomes from rising inequality of labor income. this is due to amixture of reasons--

changing supply and demand forskills, rise between education and technology. globalization certainlyis part of the story. but i think there'smore than that. you need to havea story where you have more unequal accessto skills in the us than in europe or japan. so maybe raising tuition,insufficient public investment in education, thatcan explain why

you have higher inequality inskill acquisition and access to education in the us thanin the rest of the rich world. unprecedented rise of topmanagerial compensation in the us-- why did ithappen so much in the us? there are complicated,changing social norms and also changing incentives andchanging corporate governance. i tend to believe, on the basisof my research with emmanuel saez and stefanie stantcheva,that the very cut in top income tax rate that have occurred inthe us since the '70s, '80s,

have also played a role andhave transformed the incentives for very top managers to tryto bargain very aggressively, and try to get very high payincrease, which are often difficult to explain on thebasis of observed performance or observed productivity,and have more to do with the ability sometimesto put the right people in the rightcompensation committee in order to get pay increase. fall in minimum wage in the usalso played, clearly, a role.

so i don't ask you to agreewith the exact importance of each explanation. but i think we can allagree about the fact that education policy, corporategovernance, fiscal policy, labor market policy-- all thesedifferent institutions matter and explain why you can havedifferent rising inequality in these different countries. just let me show you theexample of the minimum wage. so this is a graph thatis taken from my book.

so this is the realvalue of the minimum wage in france and the us. so you see thatin the '50s, '60s, the minimum wage used tobe a lot higher in the us than in france, whereastoday it is a lot smaller. so some people infrance will say that it is too highin france, which i'm not going comment on this. it's a reasonable discussion.

but it is certainlytoo low in the us. or at least-- inthe us right now, the federal minimum wageis $7.20 or $7.30 per hour. it used to be $10in the late 1960s. so it's quite unusual. so this is expressed in dollarsof today, in 2013 dollars. so the purchasing power of thefederal minimum wage in the us is today less than what itwas in the 1960s at a time when there was no moreunemployment than today.

so it's quiteunusual in a country to have a decline in the realvalue of the minimum wage over a 50-year period. and clearly this change inlabor market institutions, change in collectivebargaining, the role of unions also played a big role inthe evolution of inequality. let me show you anothergraph regarding education. so this is taken from avery interesting research by raj chetty, who teaches atharvard economic department,

with emmanuel saez,who is at berkeley. so you have-- on thelower horizontal axis, you have theparental income rank. so if you have 10,it means that you are in the bottom10% of the family distribution of incomein the us, and 90, you are in the top 10%. and this is your percentageattending college at 18, 21. so you get an almostperfect straight line.

and you go basicallyfrom 0% to 100%. so if your parents arepoor in this country, your probability toaccess higher education is a little more than 20%. and then if your parents arein the top 10%, it's 90%. so you don't go from 0 to 100,but you go from 20% to 90%, which is almost as spectacular. i find this graph very striking. so this means that you havea theoretical discourse

about meritocracy, equalopportunity, access to higher education, andyou have the reality. and the reality isa bit frightening. and i also-- i'm notsaying it's perfectly equal in other countries. i think there's alot of hypocrisy and a lot ofinequality in access to higher education everywhere. certainly in mycountry, in france,

there is sometimesa lot of hypocrisy in the ability ofpublic institutions to invest three times morein very elitist schools than in normal schools. but there is evidencethat inequality in access to higher educationis even higher in the us than in europe or in japan. and i give evidencefor this in the book. so anyway, this isjust to illustrate

that education policy,access to education, is very important, togetherwith the minimum wage, together withprogressive taxation. so it's a whole set ofinstitutional policies that matter. now let me moveto the second part of my lecture, whichis about the return of patrimonial society. so, so far i have focusedmostly on inequality of income

and particular labor income. now i want to move more towealth and capital, which is the main subject of the book. so of course, inequality andwealth and capital ownership is partly determined byinequality of labor income, because if you have moreunequal labor incomes, then you have more unequal resourcesto save and accumulate wealth and become owners oflarge pieces of property. but the inequality of wealthis a more complicated object

than the inequalityof labor income because it alsoinvolves inheritance. it involves naturalresources, which have been saved by no one. and the concentrationof property is always a lot higher than theconcentration of labor income, as i'm going to showyou in a minute. so i'm going tomake first a point. point number twoof my presentation

will be about thereturn of what i call patrimonial or wealth-basedsociety, particularly in europe and japan, where i willshow you that wealth income ratio seems to bereturning to very high levels in these countries. so what's particularin europe and japan is that you have verylow population growth-- and indeed, in fact, negativepopulation growth in japan or in some european countries.

and that's a very bigdifference with the us. and the basic intuition isthat in a slow-growth society, wealth accumulation inthe past can naturally become very important. and in the verylong run, this can be relevant for the entireworld, to the extent that population growth willstop everywhere at some point. so of course, if youhave a lot of migration, you can keep growing.

and i know thatthis is a country where immigration is important. in particular, universitieswould like all the rest of the planet to come here. but you know, if weare all in the us, that will not solve theproblem in the long run. you still have--population growth will depend on fertility. the projectionswe have so far is

that populationgrowth in the long run is going to be smaller thanwhat it was in the past, and possibly close to zero. so to the extentthat there will be a slow-down for otherparts of the planet than just europeand japan, this rise of wealth incomeratio and this return of what i call apatrimonial society will be relevantin more countries.

now this is not bad in itself,but this raises new challenges. and in particular, i willargue that the metamorphosis of capital assets call fornew forms of regulations, of property regulations,and that these will be very important issues in the future. and then i will moveto point number three. so regarding pointnumber two, let me first stress that wealthinequality is always a lot higher than income inequality.

so let me showyou-- let me first describe two important facts. so remember, fact number one wasthat income inequality is now higher in the us than in europe. fact number two is thatwealth inequality is always a lot higher thanincome inequality and that it is also nowhigher in the us than europe. but fact number three,wealth inequality is still less extreme todaythan what it was a century ago

in europe, where it wasreally very extreme, in spite of the fact that thetotal capitalization of private wealth relativeto national income has now recovered fromthe world war shocks. so it's important to distinguishbetween inequality on the one hand and total quantity,or total capitalization of wealth-- which is notquite the same thing-- on the other hand. so the fact that you have avery high total capitalization

of private wealthis not necessary bad if you have a largemiddle class which also would be part of total wealth. so it's important to distinguishinequality on the one hand and the wealth incomeratio on the other hand. so if we look atinequality-- so first look at the orders of magnitude. remember, for income inequality,the top 10% share was between one-third and one-half.

it used to beone-third in the us. it is now closer to one-half. now, wealth inequality, it'salways more than one-half. it goes between 60% to 90%in europe before world war i. so wealth is a lot moreconcentrated than income. for many people, thebottom 50% share in wealth is always less than 5%. so there's a very largegroup in the population for whom the very notion ofwealth or capital ownership

is quite abstract. and many people justhave a little savings in their accounts, orthey have a mortgage that is almost as big as theirreal estate property, so their net wealthis really quite small. so whatever does notbelong to the top 10% typically belongs to the middle40%-- what i called in my book the patrimonial middleclass, the people who are not in the bottom half,and who are not in the top 10%.

and so whatever does not belongto the top 10% on this graph belongs to this middle 40. so this means thatwhen europe goes from 90% of wealthfor the top 10 prior to world war i to about60% today, in between you have 30% of nationalwealth which used to belong tothe top 10 which now belongs to the middle 40. and this is this rise of apatrimonial middle class, which

i describe in my book asprobably the most important transformation in thelong run, because even though the middleclass still has less wealth than the rich--in spite of the fact that they are fourtimes more numerous-- it's still significant toown 30% of total wealth rather than 5% or 10%. this makes a big difference. now, in recent decades,the share going to the top

has started toincrease again, which means that the sharegoing to the middle class has started to decline. and one issue isto understand why. first, we're going to try tounderstand how the total value of wealth has changed. now, what is striking-- sothis is a different evolution. this is a wealth income ratio. and what's striking is thatfor wealth income ratio,

europe is actually above the us. so in europe, thereis less inequality of wealth than the us, butthe total value of wealth relative to nationalincome is higher. so that's why it'simportant to distinguish these two dimensions,because they don't need to move together. and so if you want tounderstand this big evolution of the wealth toincome ratio, you

can also decompose betweendifferent countries. so these are threeeuropean countries-- germany, france, uk. you can see thateverywhere you have a very high ratio of wealth toincome prior to world war i. so of course you havea lot of destruction and also lack of investmentbetween 1910 and 1950. also a lot of privatewealth was either nationalized after worldwar ii or its private value

was reduced by new sets ofinstitutions and policies, including rent controlfor housing values, so that the privatevalue of wealth in 1950 is extremely small byhistorical standards. and then it has started toincrease again in the past half century and is now notquite as large as what it used to be in the 19th century. but it's getting closer. now, how can we understandthis big evolution?

so the first pointis that there's nothing bad with highwealth to income ratio. and to a large extent, itcomes from a natural evolution with post-war reconstructionand the slow-down of growth. as i mentioned before, whenyou have a slow-down of growth in recent decades,you tend to accumulate more wealth relative to income. that's partly due to aging. and this is notnecessary bad in itself.

now, the problem is that thiscreates new policy challenges in terms of financialregulation, real estate bubbles, return ofinheritance, which is now for the new generationvery important in europe and also in japan. and in order toanalyze these issues, probably the most importantmessage of this presentation is that i really tryin my book to develop a multi-dimensional approachto the history of capital

and property relations. when you do this bigaddition and compute the total value of allassets, you are really doing something very abstract. and the reason why mybook is relatively long is because i actuallytry to analyze in a separate way eachseparate assets-- from land to business assets,to foreign assets, real estate, publicdebt, immaterial capital.

and all these differentkinds of property give rise to differentinstitutional challenges, to different negotiationbetween owners of capital and those who mostlyown their labor. and they really need to beanalyzed separately if we want to understand them properly. and it's more than justmaking the big addition. so let me take acouple of examples to illustrate this point.

first, if you look inthe very long run-- so this is a decompositionof the structure of property in the united kingdom. so you don't seethe color very well, but let me just say that thebottom part in the 18th century is agricultural land. so you can see that at thebeginning of the period, agricultural land is avery big part of wealth. the middle partis housing, which

was less importantin the 18th century but is now veryimportant, partly due to a very high price forhousing, maybe reflecting partly a bubble, but also thefact that many people want to live in the same place. and there could be somethingstructural in these very high housing prices. the third part isalso domestic capital, which is basicallybusiness assets, which

becomes very important withthe industrial revolution. and you have alittle white part, which is very important onthe eve of world war i, which is net foreign capital. so this is what britain ownedin the rest of the world. and you can see thatthis is very significant. you have almost twoyears of national income. almost one-third of everythingthe brits own in 1910 is they own part ofthe rest of the world.

and of course, that's relatedto their colonial empire. in france, there's asmaller colonial empire, but still, it'squite a significant. it's more than oneyear of national income in foreign assets. and now this disappearsentirely in france and britain during the 20th century. partly it's decolonization. but most of it is reallyworld war i and world war ii,

where, to finance the war,to pay for the war, many of the rich frenchand brits have to sell their foreignassets, buy some public bonds to their government. and then their publicbonds will be inflated away after 1945, so that in the end,they don't own much after that. so that's a bigpart of the process. and if you just look at thebig addition and you don't look separately at thesedifferent assets,

you cannot reallyunderstand the whole thing. so what's strikingin the long run is that you have metamorphosisof capital in the sense that agricultural landis not really playing a big role today. but the total the value ofassets is getting similar, like in britain in 2010 thanin the 19th or 18th century, but with a verydifferent form of assets. let me mention also that if youlook at annual series for more

countries-- so here youhave the same private wealth to national income ratioin more countries-- you can see thejapanese bubble in 1990. and generally speaking,what you can see is that bubbles onprice of real estate and very sharp movement andprice of other capital assets play a big role. so the history ofcapital is never quiet. it is full of crisis.

it is full ofvolatility, because it's difficult to put aprice of assets, also. putting a price on real estate,or putting a price on the stock market-- some peopledo this for business, and that's complicated. and so you have allthese big variations. if you look at the-- sothis is the spanish bubble over there, which is evenbigger than the japanese bubble. you can see thatin spain in 2007,

the total market valueof private wealth relative to nationalincome was eight years of national income--even more than in japan, where it was seven yearsof national income in 1990. so this creates newchallenges in terms of financial regulation. in 1970, the ratio of privatewealth to national income was only two to threeyears-- so these are the top eight or topnine developed economies

in the world. so everywhere it between twoand three and a half years. whereas today, it's betweenfour and seven or eight years. so of course, if youmake a 10% mistake on the price of your realestate in spain or japan, when you have aratio of six or seven years of gdp inprivate wealth, this is a mistake that can havevery big consequences, much bigger than when it's aratio of two or three.

so this creates newpolicy challenges. let me also mentionthat a significant part of the increase in privatewealth to gdp ratio is also a transfer frompublic to private capital. so you have privatizationof public assets. and you have anincrease in public debt. so public capital--so you can see-- well, again, you don't seethe color very well. but i just want tomention one example.

look at italy. italy is the bottompoint for public capital and actually is the toppoint for private capital. so bottom point forpublic capital-- look, italy is negativefor public capital. what does this mean? this means that even if theitalian government were selling all the public assets, allthe public buildings, schools, hospitals, financialassets-- they

don't have much,but assume they sell everything-- that willnot be enough to repay the public debt. they will still haveabout minus 60%, 70% of nationalincome in public debt. so i'm not sayingthey should do this, but it's important to realizethat-- many people would be shocked if we had to payrent to the private owners of the schools to whichwe send our children.

so i'm not sayingthey should do this. but it's important torealize that in a way, this is alreadywhat they are doing. because when you have apublic debt that's bigger then the value of yourpublic assets, in effect what you have to pay in interestpayment for your public debt can be higher than whatwould be the rental value of your public asset. so it's a bit moreabstract because it

goes through the financialintermediation system. but in the end,it's very concrete. right now in italy, theyare paying 5%, 6% of gdp each year in interest payment,whereas the total budget of their entire publicuniversity system is about 1% of gdp. and so is this the rightway to prepare the future for the new generation? it's not entirely clear.

so i try to put these issuesinto historical perspective. britain in the 19thcentury is a country where the holdersof public debt, the rentiers of publicdebt, are very powerful, not only in thenovels of jane austen but also in the realpolitics of the time. and they manageto get the country to repay duringan entire century 2%, 3% of gp in interest paymentand budget surplus each year,

which is more thanthe total education budget of britain at that time. and during an entirecentury, the country reduces the vast publicdebt of 200% of gdp that came from the napoleonicwar to 20%, 30% of gdp at the eve of world war i. so it worked, but ittook an entire century of repaying interest payment. and i would not like europe tomake the same mistake today.

but anyway, that's one ofthe things for which history can be useful. so let me alsomention, very quickly, some of the particular featuresof capital and inequality in the united states of america. and let me sayinequality in america has a differentstructure as in europe. and this was already likethis in the 19th century, where you have-- as we like tosay, the land of opportunity.

capital accumulated inthe past matters less than europe, partly because ofperpetual population growth, which is in a waya way to reduce the level of inherited wealth. and at the sametime, this is a land of slavery, which is in away the most extreme form of property relation. and i just finished yesterdayreading sven's book, and i regret thati couldn't read it

before when i wrote my book. but certainly, the importanceof slavery-- we all know after sven'sbook the importance it had in the development ofcapitalism in this country and in global history as well. i try also in mybook to illustrate this in the followingmanner-- if you look at the totalvalue of wealth, say, in britain inthe 18th century,

you have very high ratioof wealth to income. if you look in the us-- sothis is with the same scale-- you can see that the ratiois much lower, in particular because the valueof agricultural land is much lower. so land prices are very lowin the us in the 19th century. we all know this. tocqueville wrote aboutthis, and tocqueville thought one of the originsof the democratic spirit

of america was thateverybody could own land and everybody can access land. so if you just own land,you cannot be very rich, because there is so much landthat the price of land is very small. and generally speaking,the total value of everything there is to ownin the country is not very high. so you cannot be veryrich just by owning land, or you need toown a lot of land.

but of course, if youhave the clever idea to own not only the landbut also the people who work on the land, then youcan manage to be a lot richer. so here, this is the valueof slaves-- the market value of slaves at that time. and then you get towealth to income ratios that are much closerto the european level in the 19th century,with of course big variation betweenthe south of the us

and the north of the us. so that's why the us is, atthe same time in the north, the place where wealthaccumulated in the past is not very important,in particular is very cheap. whereas in the south, the slavesbasically more than compensate for the lower value of the land. and this corresponds toan inequality structure and a structure ofdomination based

on property which in many waysis much more violent than what you have in old europe. so you have this contrastbetween different parts of the us and the particularrelationship of the us with capital andwith inequality which i try to emphasize in my book. and again, this isa very extreme case where property rightsare socially historically determined.

and you cannot just takethem as given forever. they vary over time. they are a social construction. to give just anotherexample of this, taking an example that ismuch, much closer to us today, from germany. it's interesting. if you rememberwhen i showed you the graph with britain,france, and germany

that you havelower market values of capital assets in germany ascompared to france and britain. and one interestingquestion is why? so very often, people talkabout lower real estate prices in germany, whichcould be related also to different housing marketregulations in germany. but in fact, thebiggest part comes from lower stockmarket capitalization of corporations in germany.

and one interpretationis what we sometimes call stockholdercapitalism, which is that shareholders ingermany have to share power a little more than inother countries with worker representatives, sometimesregional government. so that at the end ofthe day, the market value is less than the bookvalue of a corporation. now, apparently this doesn'tprevent german companies from producing good cars.

so the fact that you have alower market value of companies is not bad in itself. and this exampleclearly illustrates that market and socialvalues of capital can be very differentand that more generally, property relations aresocially, legally, historically there are different waysto distribute power. so capital ownershipis about power. and there are different ways toregulate and to organize power.

i guess-- so this isa graph illustrating the ratio of market value andbook value of corporations. that's, again,taken from my book. you can find it online. but you can seethat for germany-- so according to the textbookmodel of perfect capital markets, this ratio shouldalways be equal to 100%. if you have the marketvalue, the book value should be the same.

in practice in germany, it'salways much less than 100%-- like, 60%, so themarket value is 60%-- whereas in the us andthe uk, it's more than 100%, at least in certain periods,or much closer to 100%. so the market and bookvalues of companies differ systematicallyacross countries. and this can beinterpreted as evidence for differentregulation of ownership and corporate power indifferent countries.

i'm going to movevery, very fast to the third point about thefuture of wealth inequality. this is where the gap betweenr and g is playing a role. so you can see that ineverything i have cited so far, r minus g doesn't play any role. you know, i have toconfess to you today that this wholething about r and g was sort of amarketing trick which apparently worked very well.

because many people thoughtthey could summarize the book with one equation,which of course, i don't believe thatyou can summarize 300 years ofhistorical evolution about income and wealthwith one equation. there are many differentinstitutions, policy, historical forcesthat play a role. there is one area where rminus g might be important, which is if youwant to understand

the long-run evolution of theconcentration of property. and here, the gap betweenr and g might be important. during most of humanhistory, the gap was large for asimple reason, which is that the growth ratein pre-industrial society was close to zero. and so the rate of return toland was typically 4% or 5%. it could vary with legalarrangements and feudalism, but it was certainly never 0%.

now, one of thefindings of my book is that the modernindustrial revolution did not change its basicrelationship between r and g as much as onemight have expected. it's only really the bigshocks of the 20th century that have completelyturned upside down this relation between at leastnet of tax, net of destruction, rate of return, and growthrate-- because of the shock due to the war, becauseof the very fast growth

of the post-war period. now, in the future, it looks asif growth rates, in particular for population, willbe less than what they were in the past. and competitionbetween countries to attract capital, maybealso financial deregulation, can contribute tohigh rate of return, in particular forhigh wealth portfolio. one of the problems isthat there's probably

too little transparencyabout global wealth dynamics, and particularlycross-border financial assets. and i think this is aproblem in rich countries. but this is even more aproblem in emerging countries, in particular china,latin america, africa, where a very big partof the capital stock is owned either bylocal or foreign elites through off-shore assetsand tax events, which makes it very difficultfor democratic regulation

of inequality. and that's a bigchallenge for the future. so i'm not going to presentthis in a detailed manner, just to mention that if youlook at data coming from wealth rankings-- so this is atable coming from the forbes' billionaire'swealth rankings-- it looks as if the very top wealthgroups at the world level are rising a lot fasterthan average wealth. let me make clear that this isnot particularly reliable data.

it's hard sometimesto know how the forbes people compute their ranking. but we live at atime-- it's a bit sad-- i would prefer to read theimf publication or eurostat publication or us governmentpublication to know about wealth dynamics. but you wouldn'tfind the data, partly because for lack of transparencyand automatic transmission of information aboutcross-border financial assets.

so we have to dowith what we have. so in my book, i am verypragmatic with data sources. there's no perfect data sources. and so i use a little bit thisforbes magazine at some point. and what you cansee is that here you have a major divergencebetween the average wealth at the top andthe average wealth in the world, which isnot so easy to explain. because sometimespeople say, well, ok,

but this is because you havea lot of new billionaires who make wealth. and indeed, thepeople at the top are not the samein 2013 and 1987. according to forbes, thepeople at the top in 1987 are japanese billionaires,which everybody has forgotten their name now. whereas in 2013, youhave carlos slim, you have bill gates, et cetera.

but the fact that you have somemobility-- the fact that you have all of these new peopleat the top-- this, in itself, does not explain why theaverage wealth in this top group should rise threeor four times faster than average wealthin the world. because the fact that youhave mobility-- in principle, you have somepeople who go down, you have some people who go up. if you were in an equilibriumof the world distribution

of wealth, these twoeffects should more or less compensate each other. and the average wealthin this top group should rise more orless at the same speed as average wealth in the world. it's ok to have very richpeople, very poor people, very middle people, as longas the different groups, in the long run, sort ofrise at the same speed. well, maybe not exactlyat the same speed,

but at speeds that areroughly comparable. but you cannot have foreveraverage wealth at the top rising three, four timesfaster than average wealth for the entire world economy. because if this wasto continue, the share of global wealthgoing to the top will rise to 100%, whicheverybody would argue is probably too much. and i'm not sayingit's going to go--

i'm not saying itwill go to 100%. i'm sure it willstop before that. but where exactly will it stop? nobody really knows. and this is an issue. so why is this happening? i think there's been a lot ofprivatization in recent decades that have produced, incertainly russian oligarchs who did not become richjust through saving.

they became rich bybecoming the owner of a large chunk of publicassets at very low prices. this also happen-- if you takeyour-- carlos slim did not invent the cell phone. so innovation is important. but in many cases, you cansee that at the origins of some of these large fortunes,you also have privatization. you have inheritance. you don't have thesame level of violence

which sven talks tous about in his book on the history of capitalism. but you-- it's farfrom being steady. it's a process wherebeing at the right time at the right placein order to acquire a big part of publicproperty at a very low price played a big rolein this time period. now, in the future--well, there's not much left to privatizein many of these countries.

so this is not goingto last forever. so that's onepossible regulation. also, by increasingthe public debt, you can have largenegative public wealth. and this can allow you to gofurther, even when there's not much left to privatize. i think part ofthe reason for this is also that financialderegulation has probably increased theinequality in access

to high financial returns. so that's good forplaces like harvard. let me concludes this with this. so this is the returnto harvard endowment. so the reason i'm showingyou university endowment is not to conclude with harvard,but rather because at least these people publish data. that's a good thingabout us universities, is that at least we know whatthey do with their portfolio.

we know the ways theyget the returns they get, which we don't knowfor the forbes people. so that's why i use these data. so you can see that you haveabout 850 us universities with capital endowments. they have done verywell-- 8.2% net of inflation, net ofadministrative costs between 1980 and 2010. and the higher theuniversity endowment,

the higher the return. so for harvard, you have10.2%, which is quite good. so i'm not saying it will beas high in the next decade. there were particularcircumstances. but still, this illustratesthe mechanisms through which inequality can feed itself. according to thetextbook economic model, what a perfect capitalmarket should do is that everybody should get thehighest return on the planet.

so you go to yourbank with $100,000. and your money gets investedin china the next morning. and you get the highestyield on the planet. but in the realworld, capital market don't seem to workexactly this way. and sometimes, if you are ableto access very sophisticated financial products-- sothe portfolio of harvard is made of privateequity, commodity derivatives, and complexfinancial products which

are very difficult toaccess to if you just go to your bank with $100,000--the management fees paid by harvard rightnow are 0.5% percent per year, which is very small. but if you have $40 billionon demand, and you pay 0.5%, you can still spend morethan $100 million each year to pay your group offinancial advisors. and so if this allows youto get 10% rather than 6%, then that's worth it.

that's interesting just toconclude-- to compare this with the amountof giving that was made to the harvard endowmentover the same time period. and over the sametime period, you have of the order-- it's alittle bit like the management fees now. it's of the order of 0.5% to1% of the endowment each year in giving, so as comparedby former students. so as compared tothe financial return,

this is really negligible. well, sometimesthat's useful if you want to get your childrenadmitted to harvard. but in terms of-- ifyou want to explain the rise of the endowment,the r is very important. so r versus g is an issue thatcan be quite significant when you look at large universityendowments, which can also be important, maybe, for theworld distribution of wealth, to try to explain why top wealthgroups are doing structurally

better than the middle class. let me stop there. sorry for being a bit too long. and thanks, again, forwelcoming me in your program. sven beckert: great. thank you so much, thomas. david kennedy:thank you very much. it's wonderful to be here. and i appreciatethe opportunity.

i'm honored by the opportunityto comment a bit on the talk. it's terrific to welcome youhere to the harvard law school, in part because yourbook, and also your talk, invites a kind of dialoguewith policy and legal people. so how did thiscome about and what can be done about it are thekinds of mechanical questions you invite us to consider. and there's kind of apuzzle about the book. it's an incrediblyinstitutionalist book

wrapped up in a formula. and the marketing trickthat you mentioned, again, may be somehow anexplanation there. but this is a book thatidentifies an enduring link between capitalismand inequality in the form of a historic normalin which returns to capital outpace economic growth. so this is the formulawith which you ended. and there would be afascinating conversation

to have, i think,about how differently in legal andeconomic specialties a word like capitalism,or even capital, comes to suggest a systemwithin which you can speak about historical regularitiesthat function as, in one way or another, logicor laws, and the way in which big stories getconstructed out of data and then need to be broken downin these institutionalist ways. but rather thanthat, i think what

i would do is justspeak very briefly about the legal andinternational dimensions of the story-- topick up the invitation that we engagewith the question, what does law haveto do with it? and so how do legal peoplethink about inequality? well, it's very common, i think,for legal people to accept a story about society thatcomes from economics-- and here professor piketty'sis a perfectly useful one--

and then try toconstruct a response out of the available policy tools. and to a certain extent, thekinds of tools one thinks of are tax and transfer tools. so we can-- the transfer couldhappen through schooling. the transfer could happeneven through a minimum wage. but the basic idea is,capitalism is what it is. and then if the statecan't respond in this way, we have to live withit, in some sense.

i want to propose asecond kind of legal frame that's opened up bymany of the examples in professor piketty'sbook, in which we get more down in the weeds where theactors and the structures that professorpiketty interprets as part of the story ofcapitalism get put together, and where they might seem tobe, in one or another way, more plastic. i mean, it's obvious,but easy to forget--

and he stressesthis repeatedly-- that the foundationsof economic life are legal-- capital,labor, money. law doesn't justregulate these things. it doesn't just tax them. it also creates them. and it could createthem in different ways with differentdistributional outcomes and potentiallydifferent trajectories

for society over time. so just as there arealternatives to capitalism, there are alsoalternatives within in. and a legal question wouldbe, how do we figure that out? what are the alternatives? and how does law set us onone path rather than another? so on the micro side, youcould imagine drilling down beneath aggregateinstitutional identities, like labor and capitaland state and so forth,

to think about people facingeach other in some kind of struggle over gains in whichtheir relative power vis-a-vis one another is a functionof their entitlements and their vulnerability. so in this sense,ricardo was right. rent is everything. economic gain arisesfrom a legal entitlement, whether it's a landor anything else, to exclude somebody elsefrom an element of value.

and a map of entitlementsat the micro level would be a map ofcoercive powers and the distributionof bargaining power, but also-- that makesit sound like we're going to have a discussion. the distribution ofthe coercive ability to force someone to forgogain that they might otherwise have hoped to enjoy. there are lots ofobvious examples--

intellectual property,inheritance law, and so forth. and you bring up a numberof them, also, in your talk. on the macro side,there'd be the process by which legal powers andvulnerabilities bunch. and here, i think, you werejust getting to that at the end with your story aboutthe harvard endowment. and how is it that, to thosewho have more and more accounts, what is the process bywhich the status of forces between groups, notbetween individuals,

is settled in society--between creditors and debtors, local workers andforeign investors, and so on? and how is it thatthe outcomes of past struggles bear downon current deals? so seeing it as apast settlement that arose through a bargainand a competition gives a different flavor to theidea of the weight of history than imagining it simply asan accumulation of money. and then there'd bea dynamic element.

virtuous andvicious cycles would get going as legalarrangements made it easier or more difficultfor differences to compound. so if we went at inequalitywith these legal ideas in mind, i think we might come up witha wider range of responses than transnationaltaxes or stronger federal regionalauthority, and so forth. a long march throughthe institutions could re-jigglethings in ways that

would strengthen the relativecoercive powers of people that we wanted tofavor and promote more virtuous cycles betweenthe haves and the have-nots, without a largetheory about how much inequality was goodor bad for capitalism or the development ofthe society as a whole, simply as an effortto strengthen the hands of those whosecoercive powers are now too few.

and it's not that difficultto figure out how, because people havebeen struggling intensely over legal regimesin these terms for a long time. you mention rent control asa very important struggle over the value and use potentialof some forms of real estate. but you could also altercontract and employer law to empower unions. you could change housingand local government law to link housing for the rich topublic services for the poor.

you could change the rules offinance and consumer protection to shift power fromcreditors to debtors and penalize exploitativefinancialization, and so on. indeed, you can shift the powerbetween identity groups-- men and women, blacks andwhites, old and young, future generations and currentgenerations-- in similar ways. so that's the legal background. there are lots ofways to address the dynamics of inequalitybeyond transfer payments

from wealthier to poorer. it's possible somehow tolink leading and lagging sectors or regions,either within a country or between countries, toone another productively. so a few words about theinternational dimensions of the story. the data that professorpiketty brings together is comparativenational data on inequality withinone country, which

he's extended to quitea number of countries and is now extending, aswe've heard, to even more. but these comparisons,as he acknowledges, are extremely difficult. and it's hard to say whatthis means about inequality at the global level. so does tax data, for example,exaggerate the inequality between countrieswhere countries have really differentways of understanding

income reporting anddifferent cultural practices with regard to taxation? but whatever the story isabout national inequality, it seems to me globalinequality is harder to assess or understandwhat we think about. so within a country,the existence of the 1% rankles when otherpeople's incomes are stagnant at some point. but imagine thefollowing-- imagine

it turned out that globalpoverty and inequality could be reduced if we empowereda super-rich class of financiers-- the0.01 percenters-- and we allowed wagesto fall in the richest quarter of countries, through acombination of financialization on the one hand andfactor price equalization for labor on the other hand. if we knew that that was true,professor piketty's findings about in-country inequalitymight look very different.

you could still jigglethings to improve equality, but financializationmight not require so much super-charging of the rich. but if you thoughtthat it did and that overall global inequalitymight thereby be reduced, you might have a differentattitude towards it. moreover, theinternational situation is also a legal situation. the global, economic,and political space

is intensely legalized. and you can map legalarrangements globally that enable the captureof rent and contribute to center-peripherydynamics, just as you can at the domestic level. so there's a globalrent story-- who may coerce a greatershare of the gains from, let's say, naturalgas exploitation-- qatar or chevron?

the answer will lie inlegal arrangements-- sovereign powers,control over technology, know-how, monopolypower, the law regulating finance and corporatelaw-- all those will affect who gets the gainfrom the natural gas that's being pumped out in the gulf. what powers permit upgradingor force downgrading in a global value chain? the law governing the abilityto garner innovation rents,

antitrust law, and so forth. and there's a global story aboutcenter-periphery relations. what legal arrangements speedthe compounding of gains? the mobility of capital, thefragmentation of jurisdictions, the fact that privaterights travel more easily than public powers--all these things change the speed with whichsome wealth aggregates and others do not. just as the terms oftrade between detroit

and its suburbs--that's where i'm from-- is a function of everythingfrom jurisdictional boundaries to rules about school finance,so the transnational dualisms that we see betweenregions and countries are a function ofinternational economic law and international private law. so it's a complicated story. and in short, geographymatters and geography is a legal construct-- whocan do what in relationship

to whom with what force? so to sum up, i thinkprofessor piketty has identified a tendency. and to explore howthe tendency arises, how it's sustained,and might be changed, requires that wesomehow open up the hood and look insidemarket capitalism to see how it's put together. and it turns out the gluethat distributes is law.

thank you. sven beckert: thankyou so much, david. stephen marglin: i'm,like the other speakers here, honored to bepart of this program. this is a magnificent book. we all knew thatinequality was rising. what we lacked was any kind ofhistorical and international perspective to put this in. and his book hasdone this in spades.

it's really opened up awhole new field of research. and the economicsprofession should be in professor piketty'sdebt for some time to come. this book is a classic. it's right up therewith the other books that economists often talkabout but rarely have read. i could tell you astory about that, too, but i'm only allowed 10 minutes. sven beckert: yeah, five.

stephen marglin: five, he said. but he really meant 10. you know, there's beeninflation-- a lot of inflation. sven beckert: right,since yesterday. stephen marglin: thisfellow jordan ellenberg did scientific research onprofessor piketty's book. and he discoveredthat few people had got beyond page 26,which made it the new record. the record up tothat time, he said,

was held by stephen hawkingfor a book most owned and least read. and now that record haspassed to professor piketty. i have to say-- and ihope you'll forgive me for this, too-- i and theother speakers didn't know this is a 500-plus page book. we didn't know what part ofit professor piketty was going to emphasize, sowe had to prepare our remarks on the basis ofthe book and not on the talk.

i did not know that r greaterthan g was a marketing ploy. i, like many others,thought that that really was a theoreticalargument-- sorry, theoretical cumempirical argument-- that structured the book. so i'm actually goingto spend a lot of time-- i can't spend a lot of time. i'm going to spend alarge fraction of time on that particular point.

but i should say beforethat that much of what's of value in this bookis quite independent of that inequality--and a good thing, too, because i will suggestthat that inequality should not surprise anybody. it is not a contradictionof capitalism. it's exactly whatyou would expect. it's the norm. moreover, it is not, initself, destabilizing.

i'm writing it herewith a subscript, which you don't usually see-- thesubscript y to emphasize that this is about thegrowth of income and output, not the growth ofa capital stock, though those two are,obviously, related. here are three briefexcerpts from the book about the importanceof this inequality. the first one, i thinkreally is very important, and i think it'sabsolutely correct.

if you think about it, it's nota contradiction of capitalism. it's a contradiction ofcapitalism with democracy. we had a chat at lunch abouthow these two are related-- inconclusive as you might expectsuch a chat conversation to be. but this really is, i believe,a contradiction, but not a contradiction of capital, nota contradiction of capitalism, not a contradiction of growth. the last statement isjust flat-out wrong-- just flat-out wrong.

piketty's laws are correct. he has three of them. we only will emphasize one,but there are three laws. and they're absolutelyand invariably correct, because they arelaws of arithmetic. they're not laws of economics. they're not laws,certainly, of capital. here they are displayed. we don't have any time togo through them in detail.

but the notation--i'm sorry, i realize that this is inside baseball,because if you didn't get beyond page 26, then this isnot going to be crystal clear. but if you did get beyond page26, it should be familiar. but i'm going to pick them up. so the first one of his lawsis that the capital share is equal to the rateof return on capital times the capital output ratio. well, that's true, becausethey're both the same thing.

so that's fine. the second law-- andthere's two versions of it-- one is in terms of the rateof growth of the capital stock, which is why i made thedistinction in the first place. the other is in the rateof growth of income. but they both--whichever way you put it, they are both arithmetic laws. from those, you can deduce athird law which is not exactly piketty's law, but it givesthe condition under which--

how much time do i have? that's ok, we'll get there. it gives the conditionunder which r will exceed the rate of growth of output. and that condition issubject to a difference between the incrementalcapital output ratio and the averagecapital output ratio, a relationship between therate of saving in the economy and the capital share.

so basically, as long asthe capital share exceeds the rate of savings, thenr will be greater than g. it's as simple as that. normally, that's thecase under capitalism. it's not logically necessary,but it is normally the case. nor is it a reasonwhy the capital concentration must grow. because piketty'sargument is that, as in the 19th century,the likely course--

the course for the last 30,40 years and the likely course in the future-- is that theshare of profits, the share of capital, will grow. and it will grow as againsta constant rate of saving. and therefore, asthat ratio grows, as alpha becomes larger andlarger than the savings rate, r will grow relativeto g. as alpha grows, as the capital share grows,inequality will grow, because capital income is moreconcentrated than labor income.

that's the story. i have to say that this isnot an empirical argument. it's a theoretical argument. and the theory, i haveto say, in this book, does not match the empiricaloriginality, perseverance, intelligence thatthe book shows. the theory is, atbest, elementary. and it just doesn'tcome to the level that the empiricalarguments come to.

first question-- whydoes beta rise over time? why does the capitaloutput ratio-- i'm sorry, rise or fall over time? well, this is whatpiketty and his colleagues said in a companionpaper to this volume. they say it's-- inshort, capital is back, capital output ratio is growing,because low growth is back. well, that's true asan accounting identity. the two have to move together.

we see that from thefact that those equations are necessarily logically true. but it doesn't say anythingabout what's causing what. it doesn't say anythingabout what's causing what. two more minutes. so here is the samething-- that if you look at the rate of growthof the capital stock that, as i said, therate of return on capital will exceed the rate of growth,provided the share of capital

exceeds-- income exceedsthe savings rate. here is a diagram that indicatesa relationship between rate of return on capital, rate ofgrowth of the capital stock, and a very simple assumptionabout the economy-- that there are twoclasses of people. there are rentiers, patrimonialcapital which is passed down from generation to generation. and then there's a middle classof wage earners and savers. leave the poor out of this.

they don't save. they don't enterinto accumulation. and the solid line gives therelationship between-- i'm almost done-- between thegrowth of the capital stock and the rate ofreturn on capital. so yeah, in most of this,you're in a region where r exceeds the rate of growth. and you're notnecessarily there. here's a region where,at low rates of return,

the rate of growth willexceed the rate of return. so it's not logically necessary. but the main pointis that each of these corresponds to a stabledivision of capital. it does not necessarilyimply any kind of increasing concentration. so look at thispurple colored part. this is all a situation in whichpiketty's inequality holds. but wealth accumulatedin the past

is not growing more rapidlythan output and wages. they're growing atexactly the same rate. so there's no necessarytendency for income inequality to worsen. is this a theory? no, it's not a theory. it's an argument thatsays that a theory is missing from piketty's capitalin the twenty-first century. does that make the bookany less magnificent?

no, not in the slightest. because what pikettyhas done is give us the basis for thinkingabout, theorizing about, how the situationof the present is similar to the situationin the past, how the situation of the present islikely to evolve in the future. and for this, we'reall in your debt. sven beckert: thankyou so much, steve. christine desan: so i'vebeen asked-- can you hear me?

i've been asked to stayhere to be quicker. and that's ok, becausei feel like a justice. it's the only time in my lifei'll ever feel like that, on the court's podium. thomas piketty has changedthe debate over inequality. his arguments are arresting. they transformimpressions into trends. and then they confrontus with the trends. his data is compelling.

even those who qualifyone aspect or another accept the largerpicture as persuasive. his work is transparent. he makes his dataand his arguments very clear andavailable to all of us. and we are a large audience. there are overflowrooms in this case. i could continuewith the praise, but i'm supposed to be quick.

so i'm going toget to one theme. and i have a fewquestions wrapped into this discussion of oneof professor piketty's themes. so he concludes,his last chapter concludes, this way--if democracy is someday to regain controlof capitalism, it must start by recognizingthat the concrete institutions in which democracy andcapitalism are embodied need to be reinventedagain and again.

so his focus is the concreteinstitutions that embody democracy and capitalism. that's an invitation to themany of us whose lives and work revolves around institutions. and that's political scientists. it's sociologists. it's historians. but it's also membersof the public. and it's also lawstudents and lawyers.

i'm going to talkvery particularly to the latter-- lawstudents, legal historians, lawyers-- because our workis in institutional design. law puts politicaldeterminations into effect by institutionalizingthem through legislation, through regulations, throughcourt decisions, through norms, through the creationof categories. law implementspolitical conclusions. it builds theinstitutions we inhabit.

so i want to beconcrete about that. i'll just givetwo examples-- one very stark and one more sweepingas an institutional matter. the stark one picks up verymuch on what david kennedy said. so professor piketty givesus many aggregates-- capital stock, economic growth--these things only have meaning because oftheir legal components, components likewages and assets. so david's beautifuldescription of legal assets

as depending on the settlementof previous bargaining processes-- i was goingto come up with an-- i have an example of an asset. but in fact, professorpiketty's presentation made me cross that out. slavery-- slaves arethe most obvious example of an asset the value of which,insofar as people had a market value, it was becauseslavery was legal. so an asset depends onits legal definition.

so that's the stark example. i want to move to the moresweeping institutional example. and this is to show the impactof legal design on capitalism, or to look at capitalismas an institutional form. so i'm going to preface itwith a question to professor piketty. what do you mean by capitalism? do the concreteinstitutions that make up capitalismhang together somehow?

can we understandcapitalism as something? why does capitalism matterif returns to capital are higher than economicgrowth in earlier ages, as well as our own? i think there areanswers to this question. i think they matter. i think we need to havea coherent definition of capitalism if we'regoing to take it on, if we're going to rethink it.

so here's an answer. i'm going to give avery specific answer about legal architecture orinstitutional architecture of capitalism. in 1700, capitalismwas redesigned. we get a whole seriesof new institutions. we get new forms ofmoney and finance. they depend on each other. they re-engineerwealth and production.

they create a newarchitecture for the market. and that architecture,that infrastructure, is what we call capitalism. so i'm going to identifyjust four elements to the new architecture. first, we get theinvention of circulating, credible public debt, which asprofessor piketty points out is extremely important. public bonds lookboring, but they

are the firstfinancial instrument pitched to the general public. a public bond isan instrument that asks people,regular individuals, to lend to the public,lend to the government, and it gives them areward for doing so. that's an innovation, onethat has legal import. so it changes the way peoplethink about material interest. it changes the way theythink about self-interest.

it makes it acceptable,even patriotic to follow one's self-interest. and governments soongrant public creditors a right to repayment. this is the financial instrumentthat jane austen's gentlemen are holding-- for good reason. this is in professorpiketty's lovely description. second, we get theinnovation of modern cash. the innovation ofmodern cash follows

the innovation of public bonds. public bonds furnish thecollateral, the security, against which nationalbanknotes are issued. so banknotes are issued againstpublic bonds by national banks, and they travel widelyagainst public bonds. cash backed by public bondsbreaks england and later european societies out ofthe monetary scarcities of the middle ages. and those scarcitieswere scathing.

modern money-- modernbanknotes-- by contrast issues much more readilythan silver coin. it issues when national bankinvestors take bonds and issue bank notes to the governmentsthat are borrowing from them. and here, just as inthe case of public debt, there's a conceptualinnovation-- a change. the new design puts a premiumon profit-oriented calculation. that calculation becomesthe pump for money creation legally-- legal change.

third, after public debtand national bank cash come capital markets. so public debt andnational bank cash together create capitalmarkets and provide the money to fuel them. after all, public debt is alegally transferable security. people get together. they start to exchange it. we know the coffee housesin exchange alley in london

where they start to do this. they represent acritical mass of traders trading a stable security. and given that startingpoint, private securities begin to get traded in the sameplaces after public securities. increasingly, the money usedto invest in capital markets will be modern cash, modern bankmoney-- the more copious cash of the modern world, notthe old, awkward commodity money of the middle ages.

so notice the synergy betweenpublic debt, public bonds, national bank notes, and capitalmarkets that we're building. again, there's aconceptual component. capital marketsdirect investment in ways that can be tracked. they transform speculation,long despised, into something else-- or so thoughtjustice holmes. so here's holmes. when he sees thecumulative effect

of speculation conducted on anexchange, he calls it a market. so he says in a landmarkcase about futures trading that futures tradingwas not gambling, because speculation, as heputs it, by competent men is the self-adjustment ofsociety to the probable. its value is well known. the speculationvia justice holmes is ordained beneficial,informative, revelatory in law,which would otherwise

terminate it as gambling. a last development--after public debt, national bank money,and capital markets-- all new, legallyengineered institutions, we get commercial banks-- a hugeexpansion of commercial banks. they can lend in a unit. everyone recognizes that'snational bank money. they can borrow from each other. they lend into thecapital markets,

often holding publicbonds as security. they effectivelylend more than they hold because they can berescued by the national banks, now known as central banks, whenthey fail-- which is regularly, given their highlyleveraged structure. so the asset bubblesthat professor piketty is talking about come from thestructure of commercial banks. as a result, because of theexpansion of commercial banks, the money supply expandsmore than 65-fold

in the 19th century. the industry of bankinggains the profits on money creation denominatedin the national currency, in those national bank notes. taxpayers go into debtto rescue and revive the system when it fails. the cost of rescuingbanks, according to reinhart and rogoffalmost doubles public debt on average in the three yearsfollowing a banking crisis.

so here, too, is a conceptualdimension-- a shift. money appears to besomething privately created once it comes fromthese commercial banks. and 97% of the moneysupply in the modern world comes from commercialbanks, not the government. my point-- this architectureis modern capitalism. these institutions--public debt, modern cash, capital markets,commercial banks, i could add moneymarkets if we had more

time-- they produce the market. in this market, it ispossible to accumulate wealth to a prodigious extent. so the wealth that follows fromcapital investment and that takes financial form-- this isthe modern structure of capital that professor piketty showed. from agricultural wemove to business capital to financial capital. this was his slide4.6, i noticed.

this is the newcomposition of capital. these are the forms ofwealth that he flags for us. and that wealth iswrapped up in legal forms. it's enabled. it is, basically, afunction of these forms. ironically, many peoplemiss the public architecture of capitalism altogether. they see powerful markets. they see cash.

they see capital. they see securities. they see banks. they fail to see thatthese are institutions that we have built witha public infrastructure, with public funds, and with law. so why does thatoversight matter-- that is, the failure tosee the architecture? if you overlook itspublic engineering, then

the architecture,the engineering appears to be anatural development, the market appears to be aproduct of autonomous forces, and the wealth produced on thatmarket appears to be merited, or at least inevitable giventhe power of those forces. that impression is reinforcedby the conceptual components, the conceptualmomentum emphasized by the new institutions. so acts in one'sown self-interest,

profit-drivencalculation, the tendency to see decentralizedmarket decisions as the revelationof probability as opposed to speculation--these dynamics gain stature as if they were morebasic to human activity than public coordinationand community. by contrast, we can makevisible the institutional, the legal components of thearchitecture we've created. we can revise it.

we can find new waysto structure capital. we could impose amore progressive tax. i totally agree withprofessor piketty. there are many otherrevisions we can make. that's a third-- alast question to him. i think he agrees, becausefocusing on a wealth tax alone leaves the rest of thearchitecture uncontested when it's possible thatthat architecture-- it seems to me thatarchitecture-- is

at the center of the problem. so to return to professorpiketty's injunction to us, if democracy issomeday to regain control of capitalism, it muststart by recognizing that the concrete institutionsin which democracy and capitalism areembodied need to be reinvented again and again. i couldn't agree more. sven beckert: thankyou so much, christine.

we now have-- we will be ableto go a little bit over time. so we can probably gountil 10 past 4:00. and i hope you're goingto be able to stay here. and we want to doso to also give you a chance to participatein this discussion. so i ask professor pikettynot to respond right away to these comments,but i want you to have an opportunity to askquestions or make comments. and there are two microphonesat the center of the room.

so if you want to asksomething, please go there. and keep yourquestion very short. and whatever you say, itshould have a question mark at the end, ok? and we have already twobig topics on the table, which we probablycannot resolve here. the one is the theory,and the other one is the legal politicalconstruction of markets. audience: hello, myname is sabata lapontes.

i'm a collegestudent from brazil. it's a great honor to be here. and my questionis, when you talk about a country like brazil,who is in the middle income trap, which means that wedo have very low wages, but we also don'thave high technology. you still would prescribethe recommendations you have in the book? audience: hello,my name is patrick.

thank you so much for beinghere, professor piketty. so i'm going to ask a moreunconventional question. if you could have dinnerwith anyone in history, who would it be? sven beckert: one more. audience: my name'sjulian duran. i'm a first-yearstudent at the college. and my question is,a lot of your critics say that your proposedcapital tax would end up

taxing innovation, whichwould, of course, harm economic growth. so i'm just wonderinghow you would respond to criticisms such as those. sven beckert: ok, so why don'twe take these questions now, and then we'll move on. so please wait for a moment. thomas. thomas piketty: oh, ok.

well, let me say a fewwords, also, to david and steve and christine. thanks a lot for your comments,which make me think a lot. one or two quickremarks-- i really liked, david, what yousaid about geography as a legal construct andthe international dimension of this discussionabout the legal system. this made me see--i wrote recently a paper that's notincluded in my book

about trying tomeasure inequality at the level of themiddle east as a whole-- so forgetting about thenational boundaries, and taking together this250, 300 million inhabitants from egypt to iran. and so if you do that, basicallywhat-- you find something which is very of use, butwhich is important, which is that you have a levelof inequality of incomes that is much higher than anyother region in the world,

and in particular, muchhigher than in latin america or thatcountry like brazil, which are usually viewed asthe most unequal countries. but given theinternational inequality between-- at some point inmy book, i mention egypt. the total education budgetfor all schools in egypt, which is a country withalmost 100 million people, is 100 times lessthan all resources going to countrieswith no population

a few hundred kilometersaway in the gulf. so the level of internationalinequality that you have here is such that if you put allthese countries together, even though withincountry inequality is not necessarily very high, thetotal inequality is much higher. so i think playing aroundwith these boundaries and trying to look atinequality beyond the existing boundaries-- which, ofcourse, are legal constructs. so in the case of themiddle east, some of them

are being redrawn right now. so these are legalconstructs, and these are changing legal constructs. and also, some of themhave been constructed by the west, of course, andhave been supported by the west. i was very influencedas a student by two major political events. one was the fall of theberlin wall, of course. i was 18 in 1989.

but the other one in 1990,1991, was the first gulf war where, in effect,we were protecting a given set of frontiers. and these are legal constructs. and this goes together withthe level of inequality that is clearly part of thestory of what's going on. so probably in the book, itake too much the nation state as given, and i don't go enoughbeyond the existing boundaries to study inequality.

but this is somethingthat should be done more. let me say a few words tosteve about the marketing trick and r bigger than g. sowhat i mean by marketing trick is that there is a lotmore-- many other mechanisms, institutions, policiesthat play a big role in the book independentlyof r bigger than g. now, let me also mentionthat in itself, you can have r biggerthan g in a world with perfect equality--no inequality at all.

all what this means isif the growth rate is 1% and the rate of returnon average is 5%, all this means is thatyou could be in a society with perfectly equalproperty of capital. and all this implies is thatthe owner of-- each family needs to reinvest 20%of its capital income and can consumethe other 80% so as to ensure that its wealthrises at the same speed as the size of the economy.

so you could have aperfect egalitarian model-- a representativeagent model, as economists say--where r is bigger than g. you have perfect equality, andeverything is perfectly fine. so in a way, yes, thisis a normal situation, because if r wassmaller than g, then it means that you wouldneed to reinvest even more than the returnto your capital in order to ensurethat your capital grows

as fast as the sizeof the economy, which would be really stupid. what's the pointof being an owner if you need to reinvestmore than the return to your ownership? so this is reallythe very least you can ask from capitalownership, which is that r is bigger than g.otherwise, what's the point? so in itself, r biggerthan g is not-- there's

no problem with it. it could come with aperfectly egalitarian society. [inaudible] now, the problem is that inpractice, there are many forces that make wealth unequal. you have many shocksin the life of families and in the process of wealthaccumulation and wealth transmission. so different families havedifferent numbers of children.

they die at different ages. some make very good investments. some go bankrupt. some have very high wages. some have low wages. the point is that fora given variance of all these other shocks, abigger gap between r and g will tend to amplifythis variance, and will tend to getthe economy to converge

to a level of wealthconcentrations that will be a steeply risingfunction of of r bigger than g. so this is the basic theoreticalmodel that i have in mind. this is exposed in the technicalappendix to chapter 10. that's available online. you can go look. that's exposed in thescience paper as well. this is a well-knowntheoretical model, but here i try totake it seriously,

which is that a smalldifference in r bigger than g can indeed magnify other shocks. but you need other shocks. if you don't have all theseother shocks to rate of return, to labor income, todemographics-- in itself, r bigger than g doesnot produce inequality. but if you have allthese other shocks, then it will tend tomagnify inequality. and i think that's partof the explanation for why

wealth concentration isso high in pretty much every society up until worldwar i, is because of that. in particular, why iswealth concentrated almost as much or evermore in france in 1914 than in france in 1789? which the elite atthe time in france did not want to believe this. they thought, we have modernproperty, modern growth, equality and property regimes.

we are not in theancien regime anymore, so we should have equality. well, except that in fact,the concentration of wealth was just as large as acentury and a half before. and i think part of it isbecause the gap between r and g did not decline becauseof modern capitalism and because of theindustrial revolution. which brings me to thevery important question raised by christine-- what'smy definition of capitalism?

well, i like your definition. it's a set of institutions youdescribe as very important. probably sven would addthat the development of new appropriation ofland in the colonies, in the new world,the invention-- or at least thedevelopment-- of slavery at a scale that was unknownbefore played a big role. so i think all thesedifferent institutions and legal institutionand coercive powers--

to take the terms of david--are critical to the development but one of the messages ofmy book is that in the end, modern capitalist institutionand modern industrial revolution did not affectthis basic inequality between the rate of return andthe growth rate as much as one might have expectedin the long run. so of course, thegrowth rate increased from 0% in preindustrialsociety to 1% or 2% in the long run in modernindustrial societies.

but the rate ofreturn also increased. and so the gap between thetwo did not change that much. of course, it dependson legal regimes. the legal rules canaffect this gap a lot. so this is not-- i don'ttake it as an act of god. it is a rule of law,and this can be changed by different legal regimes. but by and large,this big gap was there before industrial capitalism,and it will probably be there

after industrial capitalism. so you have metamorphosesof property structure, but this structuralrelation is still there. let me move to the questionsthat were raised after that. brazil is-- do youhave the same policy recommendations for brazilthan for the us or france? i think no. each country has its own setof institutions and policy to develop.

each country has its ownparticular intimate history so brazil, ofcourse, is a country with a lot of inequality,which a lot of it comes from itsparticular history. this was the last countrywhere slavery was abolished. there was one-thirdof the population that was slave in 1880 yearsas compared to 10% in the us. so slavery took a dimension. so there are historicalroutes to inequality

that are different indifferent countries. now, still some of thesolution that i mention in the book i think can be ofinterest even if each country has to find its own way. in terms ofprogressive taxation, brazil is a country where youhave very large indirect tax, consumption tax. so you pay your electricitybill, you pay a 30% tax. but then the inheritance taxright now in brazil is 3%.

so if you inherit millions, tensof millions, you pay 3% tax. and i think probably youcould reduce the first one and increase the second one. so there are issuesabout taxation in order to financebetter public services and public education in brazil,where i think there are things to learn from other countries. but that doesn't mean thatthere's a one-size-fits-all policy.

i will skip thedinner questions, because it's-- i don't quiteknow what to answer to this. jesus christ would have been-- anyway, so the lastquestions-- i'm sorry that i didn't keep track. yes, progressivetaxation and innovation. the historical evidenceis that it really depends how you structureyour level of progressivity. if you had very hightax rate on everybody,

including people whoare just starting new accumulation of wealth,then that's probably not good for innovation. but if you have veryhigh progressivity only at the very top end,the evidence-- in particular from this country,in the us-- the us is a country where between1930 and 1980, on average the top income tax ratewas 82%, which is really as high as it can possibly be.

well, there was state incometax in addition to that. but this applied tovery high income levels, typically above $1million in annual incomes. and if anything, theproductivity growth of the us economy-- well,apparently this didn't kill americancapitalism, otherwise we would have noticed itin a 50-year period. and if anything,productivity growth was higher in the '50s,'60s, '70s than what it has

been since the reagan years. probably becausepaying top managers, as i say during mytalk, $10 million per year rather than $1million is not so useful. and so i think theevidence is that, of course, if you have 80%tax rate on anybody who's making $100,000 or$200,000, probably you will have a different effect. so it really depends how youstructure your progressive tax

system. sven beckert: ok, let's taketwo more questions, very short, and thenwe'll unfortunately, have to come to an end. audience: hello,professor piketty. my question is alsopretty connected to that case ofendogenous growth that was the previousquestion about. have you tried toestimate in some way what

part of the change in thenet worth of the richest individuals came either frombuying new assets, change of prices of assetsthey already have, or from creating entirelynew kinds of assets? like, for example, building acell phone network in mexico. sven beckert: thank you. audience: if capitaltax were to be imposed, if somebody owns assets whichdo not produce cash flow, does that mean that that personwould have to sell assets

in order to pay their taxes? and would this lead to amarket crash in that case? thomas piketty: hm. well, just to startwith the last question. the property tax-- there isalready a wealth tax in the us and in most countries. this is called the property tax. it's just that it's based onyour real estate property. but this is already halfof household wealth,

so this is half of awealth tax, if you will. and i think the main reasonwhy it's half of a wealth tax is because this was createda long time ago-- 200 years ago in the us, or actuallyin britain or in france, [? taxe financiere ?], which isthe equivalent of the property tax, that was created bythe revolution-- at a time where wealth was mostlyeither land or real estate. and there was limited financialwealth and financial liability. so that's why thistax was created,

as a tax of real property. but the way it worksis that indeed, if you have a lot ofproperty but you don't get any income out of it-- so ifyou have a secondary residence everywhere in the country,but you don't rent them, you don't do anythingwith them, and you just spend one night per monthin each of them-- then your property tax willbe more than your income, in the extreme case whereyou have zero income

and you have property there. and so yes, you will have tosell some of your property to pay your property tax. but i don't thinkanybody's asking that such people should beexempt from property tax. i've never heardanybody asking for this. so i think the whole point ofa property tax or wealth tax is that, indeed, if you have alot of wealth but very little income, and you're just sittingon your properties, then, well,

you will have tosell some of it so that other people who knowwhat to do with the wealth will do something with it. and that's the whole logicof property taxation. it's always been like this. so that's why this issueof how much you should tax the stock of propertyversus the flow of new income is complicated. in france, a big proponentof wealth taxation

was maurice allais,who was a nobel prize winner in economics. and i can tell you, hewas not left wing at all. he was a very right-wing person. but he thought that we shouldtax the stock of property and not at all theincome flow, so as to put incentives onpeople to get a high return on their property. now, it's too extreme,because sometimes the rate

of return you get is justpartly due to bad luck. so if you are a company,you're making losses-- maybe if you keep paying atax on the stock of wealth, you will pay as muchtax as a company making huge profits which maybewill put you in bankruptcy, so that's probablynot a good idea. so you want to find abalance between how much you tax the stock and howmuch you tax the flow. and each tax system hascomponents of the two.

the problem is thatsometimes these tax systems were set up 200years ago and did not adapt to the structureof wealth today. but i think we-- both kindsof taxation are useful. we have to findthe right balance. sven beckert: ok,and on this note, unfortunately, we have tocome to an end, not least because professorpiketty is going to have to give another talk in45 minutes elsewhere in boston.

but we could go on,obviously, for much longer. and in some ways, we shouldgo on for much longer. this should be aninvitation to continue to this debate on these issues. and we should thankprofessor piketty, not just for being here--which, of course, we thank you forvery much-- but also for instigating thisreally important debate. and thank you for joining us.