endless. it adversely affects small firms and start-up firms. and large firms can grapple with the complexity but also calculating the provisions the cost of making calculations does not increase lynnkreescrease linearly. firms trying to launch their businesses or grow their businesses. >> i'm jim. the rate coalition. this is a terrific conference. and the title is the tax base. and that's obviously a critical issue, but i have also heard at least three of you up on this panel talk about the impact of the corporate tax rate which people know is the highest in the world. and i've even heard a discussion

about simplification and complexity. so those are three issues, therefore. the base, the rate and complexity/simplification. on the off chance that there's an impressionable staffer from the ways and means committee here, what might the four of you, and i really would like to get all four of you opinion on this like him or her, staffer that is to take back to chairman ryan and say, chief this is what we might accomplish in this congress. >> i'll start with that one. we'll go left to right. first thing is forget what can be accomplished. let's do that second. first thing to take back is the lesson that only people pay taxes. corporations are fctional entities. second then, the ideal rate, and

i use the word rate, is zero. that's the ideal rate. what do you take back to this congress? i think what you can get done is the narrative we're not competitive wither trading partners and that to be competitive we should be at or equal to. the rate should be no higher than 25%. and that is still higher than i'd like but at least that gets you to saying we're not being competitive by being around their average. we're not higher, lower or average. that seems to be a narrative i wound send. >> i'm going to disagree in one small sense. we want income to be taxed only one time. therefore, we shouldn't have both the corporate income and double tax on dividends and capital gains. theoretically you could move the corporate rate to zero and tax it at ordinary income tax rates. i think it's simpler to tax the income once at the corporate level and not have the double

taxation on individuals. there's one corporate taxpayer. that's easier than tracking down hundreds of thousands of shareholders. but six of one, half dozen of the other. it's a question of administration. in terms of your question jim, let me tell you an example. ireland is one of the few other countries besides the united states that has a worldwide tax system but no one xlaens because ireland's corporate tax rate is only 12.5%. if your marginal tax rates get low enough some of these distortions no longer have a big value. let's imagine we're going back to 1980 the jimmy carter top tax rate is still 70%. tax deduction was very valuable. you could low eryour tax bill. by the reagan years were you really going to go through as many gymnastics and hire as many

lawyers and akontants and tax planners to benefit 28 cents? you had much greater incentive to earn income and be productive without worrying about tax consequences. the rate is critically important. the whole purpose of this session is to focus on the fact we should fix the base. in reality, they are both important. have the lowest possible rate and make sure you have the base defined correctly because we don't want to penalize the saving and investing that's important for long run superiority. >> let's bring us back to what's the point of tax reform. it's primarily to grow the economy, increase the welfare of the american people. and potential gans from tax reform are in the order of 15% of gdp, which would make the country feel like it was in a

boom it hasn't experienced since the '80s. that would occur over a period of about ten years but be front-loaded. most of that potential again comes from improvement in business taxation or taxation of capital. so reducing marginal tax rates on corporate -- c corporations is critical but getting the tax base right is also critical. to the extent we're dropping margial rates by getting rid of wind energy credits or low income housing credits or employer-provided health insurance or takior pick it's pro growth. it improves the welfare of the american people an a whole. dropping corporate tax rates by further extending the period over which you have to deduct you know investments. whether it's pro growth or not is a very iffy thing.

and it probably isn't. so you lose all the growth effects by broadening the base and raising the cost of capital and further extending the capital cost recovery period. and that's one of the problems that the proposal had is a large portion of the revenue that was raised in that proposal was raised by lengthening capital cost recovery allowances or doing things like making advertising amortizable over roughly ten years. if you do things like that you're not doing something that's meaningfully pro growth. we have to buck up and, i think, understand that we are now way out of the main stream in the industrialized world. we tax our businesses too heavily. we need a business tax cut. and we're not going to biblee able

to drop the rates hard and get competitive. you get to the 25% average -- it's just an average -- by broadening the accident tax base. there's not enough tax preferences in the corporate code to do it. >> i have a question about whether or not you know good estimates of how much you get from improving the efficient allocation of capital as opposed to changes in the overall cost of capital. >> i think the answer to that is not very. most of the macroeconomic simulations looking at tax reform tend to look at the capital deepening effect. enlarging the capital stock. but, i mean to sort of get wonkish as dad put it, there is no doubt that you get gains from a more efficient allocation of

capital. basic price theory would lead you to that result. as far as i'm aware, the tradition of optimal tax analysis doesn't look at that issue either. it tends to just look at the relatively simple size of the capital stock uniform homogenous capital stock. and i think it would be good if we could get people to some degree to focus an the efficiency gains from a more efficient allocation of capital because that's certainly there. now the simplest way to get there and accomplish both results is expensing. alternatively you have to come up with some economic depreciation concept that is accurate. but that is i think conceptually difficult to impossible because that -- to know the actual decline in the future present discount, value of the asset you

have to have robust secondary markets and we don't. i don't think you can ever solve that problem in terms of getting it right, except by expensing. >> let me add one thing. it doesn't address your question of how you address that. every preference exclusion, credit that's put in the tax code is put in there to encourage people to make decisions they wouldn't otherwise make because they're not otherwise economically sensible. if we're having them make economically senseless sdgss decisions, there has to be a cost. we have to estimate how are those resources otherwise have been used presumably in ways better for the economy but it would be good if someone went out and measured that. maybe that will be a project for jason. >> let me add one thing to show how weak that literature is. as far as i know the only people that looked at this, other than, i think, some folks

within treasury that haven't published is the halton wycolt study in the early '80s. they looked at the length. in terms of the pattern of deductions, they -- in an important sense nobody has ever got empirical information on this because there is no emperical information because we don't have widget making machines. and, therefore, you have the proponents of an income tax have an inside theoretical problem. they'll always get it wrong because there's no information. they have to guess. the whole tax system is built on a guess. >> i would just add one more thing because this discussion is very enlightening. i would rather not force us to tax analysis of treasury or the hill to do gymnastics and try to

make something fit in the dynamic scoring window. we're seeing the corporate tax base erode because of the high corporate tax rate. so one thing i also would have staffers take back too their members, even on a static score if tax reform costs money, we should still do it because it will be better in the long term. we should just get this done. >> so this week senator paul announced his candidacy for running for president. and last week we saw senator cruz do that. looks like on sunday hillary clinton is going to put her name in and the week after -- >> that's what i heard on the news today. and next week senator rubio. what do we know about their tax plans, if anything.

>> well, senators lee and ruby will be speaking to the heritage foundation an april 15th about their plan. we released a paper analyzing it. it's very, very positive on the business side. and i think people should look at what it does because it's one of the more constructive plans. and then you may want to talk more about this, but as far as i know senator cruz and senator paul's plans are very 30,000 feet type flat tax plans without much detail. so i don't believe secretary of state clinton has given any details about what she would do on tax. >> i would add to david's point about senators lee and rubio their plan is probably the best one we've seen come out of capitol hill in a while on the

business side. they are also not trying to hamstring themselves by being revenue neutral. they're going for a more efficient tax code for economic growth, and i think that's important. >> one last one. all right. you were talking earlier about one of the goals of any tax reform would be complexity. and having worked on the income tax for years and the corporate tax specifically i can tell you that any -- that any income tax is going to be complex. what we -- it is so complex now that even if they cut it in half it would still be incredibly complex. tax people forget that if you talk to the average person about simple tax concepts, their eyes are going to spin over. they don't understand any of that, and even other attorneys that i talk to who aren't tax

attorneys, can't have a really meaningful discussion about taxes. my point is this idea of getting rid of complexity whot getting rid of the income tax to me is -- whatever you want to call it. income tax consumption tax. if you start off with income that determination is an abstraction and a very broad abstraction. >> well you definitely highlighted a big obstacle but i disagree with you. if you look at the hong kong flat tax, because it basically doesn't have any of the double taxation because it's a territorial tax system, no capital gains tax no death tax all those things, the hong kong flat tax which has been around 60 years so it's very durable. it's remarkably simple. is it as simple as the blue postcard? no. but other tax systems around the world, you find some of them

where it's a page back and forth. so if you are dealing with the tax base correctly as i said before, that automatically eliminates so much complication. if you are a business and don't have to figure out depreciation. you put, these are my gross receipts, the wages i paid, raw material costs investment expenditures. what's left is your taxable income, you can actually have a dramatically simple system. i don't think the definition of income is complicated if you have a consumption basis. if you're going to mix it with 102 years of micromanagement you get to the system that you correctly describe which is a big giant mess. but it can be solved. whether it will be, of course we'll be up here when we're 110 giving the same presentations. >> the sources of complexity and income tax goes away with lee/rubio or the flat tax or new

flat tax. it's really a cash flow tax. and a sales tax. one is no deprishiable lives. just deduct capital expenses. the third is accounting inventory. it's a mess if you take the tax law seriously and have to capitalize in various costs. before you know if you've employed an army of accountants. then international, all the income sourcing, expense allocation rules separate baskets, all goes away. in hall/rabuscko you no longer have to take into account capital gains or bank accounts or interest expenses because interest is neither deductible nor taxable. all the major sources of complex irt in income tax fall away with

any of these plans because they are really consumption taxes. [ inaudible ] >> in a flat tax you would. in a sales tax it would be irrelevant. >> it also doesn't matter too much with that. >> we'll end with that. thank you, everybody, for coming. i'll do one housekeeping thing. an friday, april 18th at cato institute, we'll have another event called should the gao audit the federal reserve. hopefully you'll come out for that if you are interested in monetary policy. thank you, everyone. let's thank our speakers.

the new hampshire republican party kicks off its first in the nation leadership summit this weekend with speeches from some potential candidates friday morning at 10:50 a.m. chris christie, rick perry, senator marco rubio and jeb bush. that's live an our companion network c-span. the conference continues saturday morning with remarks from senators rand paul ted cruz and lindsey graham. also speeches from wisconsin governor scott walker, ohio

governor john kasich, mike huckabee and donald trump saturday morning at 10:30 a.m. eastern on c-span. this weekend the c-span cities tour partnered with comcast to learn about the history and literary life of st. augustine, florida. >> a lot of people have said he was out for additional property for the king of spain and colonization colonization. we do know that ponce de leone came ashore after searching for good harbor. took on water and wood. this area presents one of the new freshwater springs in the area around 30 degrees eight minutes and is the location of the 1565 first settlement of st.

augustine before jamestown was founded and before the pilgrims landed on plymouth rock. >> the ponce de leone hotel was built by flagler. he's very little known outside of florida but he was one of the wealthiest men. he was a co-founder of standard oil company with john d. rockefeller. a man who always wanted to undertake some great enterprise. and as it turned out, florida was it. he realized that he needed to own the railroad between jacksonville and st. augustine to ensure that guests could get to get to his hotel conveniently. clearly the dream was beginning

to grow on flagler. he had big dreams. he was a visionary. >> watch all of our events from st. augustine saturday at noon eastern on c-span 2's book tv and sunday at 2:00 on c-span3, american history tv. the brookings institution hosted a session on how technological advancements help the workforce. robert rubin opened the forum. okay. i think we'll get under way. good morning. i'm bob rubin. and on behalf of my colleagues at the hamilton project, i look forward to talking about the machine age. before i lay out some of the

issues, let me say a few words about the hamilton project. we started about nine years ago. we are not an institution, but rather, we're a small partnership of policy experts, former government officials, academics and business leaders organized as an advisory council. and our architecture is totally open. when we have policy proposals, they are commissioned from leading experts around the country, and then they are peer reviewed rather than coming from internal staff. our purpose is to support policy development and to support serious as a purpose of discussion, debate and dialogue. we believe that is particularly important at this time when, unfortunately, the public policy debate has become so affected by politics, by ideology, and by opinion that is not grounded in facts or objective analysis.

the hamilton project works in partnership with the brookings institution. brookings contributes enormously to our intellectual vitality. since launching the hamilton project, our view has been that the objectives of economic policy should be growth, broad-based participation in that growth, and economic security. we believe that these objectives can be mutually reinforcing. for example, widespread income gains promote growth by increasing demand, by increasing the ability of workers to access education, nutrition, housing and so many inputs and factors that contribute to productivity and by increasing support, political support, public and political support for growth enhancing policies. we support market-based economics, and equally we support a strong role for government to perform the functions of markets that by their very nature will not perform. that takes us to today's

subject, the future of work in the age of the machine. technological development and globalization are keys to increasing productivity and growth. but they put pressure on job creation and on wages. over the past few decades, as technological development has increased at a rapid rate and global development has increased, wages have been in many cases stagnant and inequality has increased substantially. the exception was the second half of the 1990s when tight labor markets increased incomes at all levels. today we're going to talk about how to think about that tension between the growth-enhancing effects of technology and globalization on the one hand and the effects of technology

and globalization on wages and on job creation for lower and middle income workers. this forum is a continuation of a long line of programs that we've had at the hamilton project focusing on middle income and lower income workers. growth is necessary but not sufficient for the purpose of aiding and enhancing the economic position of middle income and lower income workers. growth creates tighter markets, labor markets, as happened in the mid to late '90s and it increases the pie. but we do need a broader perspective. for example, policies that focus on education and policies that focus on job creation and on productivity through infrastructure investment, basic research and so much else, both promote growth and directly improve the position of the american worker. with this frame in mind, i'll pose a number of questions that

today's discussions may address. is technological development likely to continue moving forward at a rapid rate and with great economic significance? or as some argue, will its pace slow and its significance decrease? relatedly, is the dynamism intact or has it declined? if you don't have dynamism, then the rate will not be applied. the rate of business in the united states has increased significantly in very recent times. is that relevant to the question of dynamism in our society? and if it is relevant, is that a cyclical phenomenon or has something more fundamental changed? productivity has clearly fallen to low levels in the last few years. again, is that a cyclical phenomenon or is something more secular and fundamental happening?

if labor and technology does move at a rapid pace and if dynamism continues such that technology is deployed, will new industries and new jobs develop that will replace those that have been lost, and will those new jobs be well paid? in other words, what will the net effect of all this be on job creation and on wages for middle income and lower income workers? to go further, are there trends in the workforce that aren't yet adequately understood that may relate to these questions? for example, would the nature of jobs themselves change? with fewer employees of companies and more independent contractors. with the increase in the number of functions performed by independent contractors being a function of the enabling power of technology. for example, what's the future of clerical help when you can get clerical help on an online basis on demand?

and that takes us to policy. policies that could help address the pressures from technology and globalization, if those pressures continue, aside from improving the ability of workers through the many facets of education and training to succeed in this new world are going to need an enormous amount of creative focus. for example, we may need an increase in the earned income tax credit, not only for those who receive it at the present time but perhaps much further up the income scale. measures that facilitate collective bargaining can result in a broader participation in the benefits of productivity and growth. and there are enormous number of other possibilities and potentials we should consider in the policy arena.

moving further, i think there may be a more fundamental question that's going to have to be answered at some future time, and that may be a distant future time. but if we have ever more rapid technological development and it is labor displacing, at some point in the future -- as i say, that may be some distant point in the future -- should that lead to some basic change in our lifestyles with less work, more leisure and a richer, more robust use of that leisure? and if the forces of technology and globalization continue to create risingin ekwaultd even if that rising inequality is acopped by growth in addition to everything that needs to be done to enhance growth and tight labor markets and to improve the position of middle

and lower income workers, should there be increased redistribution to accomplish the broad objectives of our society? and if there is to be increased redistribution, how does that get done without impeding growth? the united states has tremendous strengths, and i think we are well positioned to succeed over time, but we need an effective government. one that can deal with the usually consequential policy challenges that we face such as sound, intered myiat and longer term fiscal conditions. and reform on immigration k through 12 education, energy and so much else. in that context, technological development and globalization, raise the thorny issues i just mentioned, and i'm sure many others that will come up in the course of these discussions. we will begin our program with framing remarks from erik brynjolfsson, professor of management of science and

director, center of digital business, m.i.t. school of management and andrew mcafee principle research scientist at the center for business at m.i.t. they will be followed by the founder and chairman of evercore. our first roundtable is entitled "the future of jobs." participants are erik brynjolfsson, david autor, aneesh chopra, and larry summers and charles w. elliott. elizabeth carney will be the moderator. director of the hamilton project. the second panel is the future of business innovation. in addition to andy mcafee, whom i've already introduced, the

participants will be john hotelwanger, distinguished university professor university of maryland, and art d. probiker, director of the defense advanced research agency known to all of us as darpa. the moderator will be laura tyson, professor of business administration economics at the berkeley haas school of business and member of the advisory council of the hamilton project. let me close by extending particular thanks to melissa carney, who i already mentioned, our director, kristin mcintosh, our managing director, and brad hirschbine, the visiting fellow with the hamilton project providing the intellectual construct for this session and putting together truly a remarkable program. let me thank the members of the staff of the hamilton project whose thoughtful hard work is central to everything we do at our project. with that, let me turn the

podium over to roger altman. roger. >> morning, everyone. i'll be real brief. i think that the upcoming framing remarks and the two panels we're about to have are going to fit anyone's description of provocative. i, too, want to thank melissa and kristen and the entire hamilton staff for organizing an event as rich as this and as substantive as this. as bob said, we're going to start with framing remarks from erik and andy, both professors at m.i.t. and the sloane school of management there. erik runs the m.i.t. initiative on the digital economy, and andy

is a principle research scientist at m.i.t. and the sun school in his field of research is the impact of digital technologies on business, the economy and on society. so we really could not have better framers, including because they've written a profound book many in this room, i'm sure have read it. "the second machine age." i took away three points from that book. one, we're at an inflection point on the pace of digital technological advance, that it is accelerating, and that it will produce unexpected and transformative effects.

two, that these effects will be, on the whole, positive. more choice, more freedom, more wealth. and, three, getting to our focus today, that these effects also will produce considerable economic disruption. in particular the premiums which labor markets have increasingly been placing on education and skills will rise. rise sharply. and by implication, the wage pressures and lack of employment opportunity for those workers who don't possess those skills will worsen. bob reviewed a series of questions we want to debate today that stem from the book. i'm really just going to add one to his very good list.

the past 20 years have already seen labor markets place a big premium on education. most people in this room are familiar with the ubiquitous charts that detail the returns to education. so this trend has been under way for some time. and it's an absolutely central element in the outlook for american society. and that it's been under way for some time, even apart from this prospective acceleration of technology and its impacts is why larry katz and claudia, for example, describe the challenge as a race between technology and education. and the obvious question is, is it imaginable that we will raise

education levels in this country in proportion to these rising skills premiums and in proportion to the acceleration and the pace of digital technology and software and its impacts that eric and andy are now going to discuss. over to you. >> roger, good morning. great to be here. america has never been richer. private wealth is over a trillion and private income is also at record levels. american workers have never been more productive than right now, and the reason for this bounty is because of recent advances in technology. there's also a paradox. as bob mentioned, median income

has stagnated, lower now than it was 15, 20 years ago. andy and i -- let's see if we can get the chart up here. andy and i call this the great decoupling. there it is. the share of the workforce that's employed has also fallen. if you look back for much of the 20th century, there was a rising tide that lifted all boats, an implicit social contract people would participate in that. recently, that's become somewhat unraveled. again, a number of reasons for that. the great recession didn't help. as you can see, this decoupling really started before the financial crash of a few years ago. there have been changes in tax policy and globalization, there's some measurement issues. we're not counting some of the

free goods, like wikipedia or free apps that give us driving directions. that's not really enough to close this gap. a lot of it has to do with changes in the nature of technology. if you look at the broader sweep, much of the wealth creation has -- can be traced to some amazing improvements in technology going back to say 200 years ago when a powerful general purpose technology, the watt steam engine, helped ignite the industrial revolution and successive general purpose technologies replaced a lot of muscle work with machines. and by and large, broadly, these machines were complementary to human labor. wages grew, and more people were working. but we're now in, i think, the early stages of what andy and i call a second machine age, where machines are also beginning to supplant minds as well as muscles and do a lot of the control functions that used to be integral and only done by

humans. in fact, about 10 years ago, many of us thought that there were a number of categories that humans were uniquely good at and humans were not -- the machines were not very good at substituting for in areas like dexterity, language, unstructured problem solving. in recent years, there have been big improvements in machine intelligence in all these areas, catching some of us off guard even. there have been improvements in robotics, robots like baxter here are working in more and more factories doing all sorts of manipulation. improvements in mobility. baxter, according to rod brooks, works for about $4 an hour and a bunch of simple tasks like this, people all over the world, people in massachusetts and throughout the world are doing tasks like this. machines have made huge advances in language which used to be a uniquely human capability. these days, if you see somebody talking on their phone, there's

a good chance they're actually talking to a machine, not to another human, and expecting the machine to understand. of course, they're not real good yet. we're in the middle, i think, of a 10-year period where we went from machines not being able to understand what we're saying to us being expecting the machines to understand what we're telling them and answer our questions and carry out instructions. by any measure that's a remarkable milestone. machines are translating between languages. skype will let you speak in english and speak in some version of german or french and chinese to other people. they're writing simple stories. a story about apple and a byline written by somebody with the name narrative science. that's a machine, and they write thousands and thousands of these stories about sports, earnings reports, lots of other topics.

my favorite example of unstructured problem solving what happened with jeopardy, where you have all sorts of questions and variety of different topics of human knowledge whether sports or geography or science or current events. and the father of watson came to my class and showed me this incredible chart i want to share with you. these dots are the charts of the human jeopardy. most are pretty good getting a large percent of the answers correct. when watson came on, he couldn't answer the questions all that well. but watson had something the humans didn't and that was the ability to learn at aer if er ifer if er ifer if ferocious rate and learned from wikipedia and there's watson

that got better and better and then they went on national tv and played ken jennings, the champion of jeopardy. watson won, as you may know, not just $75,000, but now watson is being used in all sorts of other applications. there's a call center in south africa that answers questions when people call in, and watson is powering that question/answer system. there are legal versions of watson, runs in the cloud now. a banking version. my students actually -- a team of students of mine work with ibm to create a version of watson that has read the dodd-frank rules and helps explain them to companies and apparently there's billions of dollars at stake there. versions of watson in a medical diagnosis. describe your symptoms close to english and it does a good job diagnosing, however obscure it may be. if watson is not today's best medical diagnostician, i expect it will be in five years and be available on the cloud.

you may very well be going to your first or second opinion to versions of watson or other related machines. this is great news in many way because it is creating all this wealth i mentioned. it was a puzzle to me when we saw the stats on stagnating median income and didn't really understand how that could be. then we were reminded there's no economic law that says that technology automatically, even if it grows the pie, that everybody is going to benefit evenly. some people could be left behind. manufacturers when cars came in or there could be potentially a majority of people left behind who do routine information processing work or basic manual skills. there's nothing in economic theory that says that can't happen, and we recently have had various flavors of technical change. in the book we describe three of them, although there are more.

skill-based technical change and my colleagues at m.i.t. made a very nice one that illustrates the fanning out of skill biased technical change leveling out towards the end. capital and labor getting different shares. the share going to workers has been falling in the united states and other countries quite precipitously, which may be some evidence machines aren't just complementary as they once were to human labor. superstars are getting a bigger and bigger share. there are a number of reasons. one of the reason is the nature of technology. digital technologies are quite different. you can take a process and codify it. once you codify it, you can digitize it. once you digitize it, you can make a copy or 10 copies or 100 million copies. each of those copies have three very interesting characteristics. they can be made at almost zero cost. they're perfect replicas of the

original. and they can be transmitted anywhere on the planet, more-or-less, instantaneously. free, perfect and instant are three adjectives we didn't use to describe most goods and services historically, but they're standard for digital goods, and they lead to some weird and sometimes wonderful economics. they can lead to a lot of bounty, but they can also lead to winner take most markets. if you codify tax preparation, you don't need hundreds of thousands of human tax preparers each serving a local market. a few good tax programs, maybe one or a few can cover a big chunk of the market. of course, as marc-andreason said this isn't just in a few obscure corners of the economy. software is eating the world. it's coming to retailing, to finance, manufacturing, to media, more and more parts of the industry. so these economics are coming to more and more parts of the economy. ultimately, this can have

profound effects, as roger is saying, not just on the bounty but distribution of income. really, it's how we use this technology. not the technology per se, that does this, how it interacts with our organizations, skills, institutions. at the end of the day, the most important thing to remember is technology is and always has been merely a tool. we have more powerful tools than we ever have had before and they have the potential to create enormous wealth with much less need for work. some people see that as a bug. i think we should see it as a feature. it should be good news. i think, shame on us, if we aren't using these amazing tools to create more shared prosperity. so, at the end of the day, ultimately, what's going to determine how we distribute this bounty is our choices, our choices in tax policy, in education, in health and

welfare. ultimately, technology doesn't determine the distribution, it's our own choices. thanks. let me turn it over to andy, who has a few additional comments on that. >> before we get to the panelists, i would just like to say two things, building on what my colleague and co-author erik just talked about. first thank you to our host this morning. fantastic for brookings and the hamilton project to convene this conversation and bring such a room of fantastic participants together for this. i'm deeply grateful. in particular, i'm extremely appreciative of the focus that the hamilton project has. to my eye, a lot of the debate about the trends in the economy, not that it's pointless or misplaced. it's missing the really important story. we're arguing about the 1% and the 1% of the 1%. that's a valid conversation. a much more important one is what's happening at the 50th

percentile of the american workforce, what's happening at the 25th percentile of income and earnings? these are the people as we look around really facing job and wage challenges. hamilton's focus on those portions of our population and our workforce seems to be exactly the right focus for us to take. the second thing i'd like to do is congratulate our host, particularly bob and roger, for bringing together a really diverse crowd this morning. erik and i have a fantastic partnership and have a bit of a problem now that we're finishing each other's sentences, and we see technology under every rock. this morning, you will hear from people who don't quite look at the world that way and have done really the best work in many areas and will bring a variety of viewpoints and perspectives and stories about what's actually going on in the economy that will be extremely valuable for all of us to listen to. i know i will learn a lot.

our colleagues agree we're living in deeply interesting and somewhat weird times. we better figure out exactly what's driving these changes so we can figure out what levers to pull on. as i said, erik and i think technology is one of the big probably under appreciated levers. the second point i'd like to make before we turn it over to the panels, if that story is anywhere near accurate, hold onto your hats, everybody, we honestly, we ain't seen nothing yet when it comes to technological progress. we heard this idea of an inflection point and erik talked about how surprised we have been by examples of technological progress. even after writing our book, the two of us still get surprised because it appears objects in the future are closer than they appear. i want to tell you three quick stories to make that point. these are things we learned not since publishing the book in january 2014. these are things we learned this year so far. in our conversations so far in 2015, erik and i had a chance to talk to an entrepreneur whose

name would probably be familiar to you, whose pretty well-known for making really cool fast cars. and we were talking to him about the amount of automation in his vehicles. we said, when will you have the capability for a fully autonomous car. he essentially said yesterday. we said, i'm sorry. yeah, i believe our cars are about as good as the average human driver in fully automatic mode. >> i said, why haven't you turned on that mode and made that available to us? are you worried about regulators or liability? he said that's not the main factor. when the time comes, we will deal with insurance issues and liability and regulation issues. the main reason we haven't enabled that fully automatic mode yet on our vehicles, we want to wait until we are confident our technologies are ten times better than the average human driver, not just at parity with them. we asked, of course, when does this happen? he said, look, i have stopped trying to make that prediction

because i kept noticing the date kept going like this, kept marching forward in time. so the driverless car, erik and i have ridden in one version of it. i think more robust versions are coming more quickly than even he and i anticipated. the second example i want to give is of a task we've been trying to get computers to be good at almost as long as we've had computers, honestly 40 or 50 years of progress on this with unbelievably poor results. how many of us have heard of the asian board game, "go"? is this familiar to people here? if you're a strategy geek, "go" is the purest expression of a form of strategy for a geek. geek is a form of praise. you and i take turns putting our black and white stones on it. when i surround your stones, i take your stones off the board. a sentence to say lifetime to master kinds of games.

people spend decades playing the game and trying to understand how to play it well. the computer geeks saw this, great, strategy game, let's try to program computers to be good at it. they have made unbelievably little progress at that for two main reasons. one is the game is just too complex for brute force simulation methods to work. you can't come close to simulating all the possible "go" games before the sun burns itself out so the brute force we have with computers is not that useful in this context. then the geeks would say, okay why don't we teach the computer the right strategies and program in the strategies and refine them over time and beat the best human players that way. the main problem there is when you go ask the best human players how they knew what move to make, they go, i don't know. i've done this for 30 years, i understand the pattern. some part of my brain gets it. that move just felt right.

they cannot articulate the strategies that they're using to play the game at a high level. so our brute force methods won't work. understanding strategies doesn't work. it feels like a little bit of a dead end. just this past year, a team of geeks said let's try a different approach. let's just configure a system and show it a bunch of examples of games played at a very, very high level. we have a library of "go" games played at a high level. let's show the computer a bunch of examples of those games, that's it. we won't try to elicit the strategies or point out the patterns most salient here. we will show it a bunch. they showed it high level games in midstream and they said, hey, what's the smart next move here? they're at the point right now where that system is able to come up with the exact same move as the human expert more than

50% of the time. this is after six months or less work on this problem. i made a bet on twitter platform for all deep thoughts, i made a bet the world's best "go" player will no longer be a human being. with a couple of examples like that, it becomes more and more clear to me and erik that the future is coming at us more than the experts are predicting. that means the economic consequences are coming more quickly than a lot of us are expecting and makes the discussions for today all the more important. thanks very much. [ applause ] in the second part of the discussion on technology and the workforce, we hear from larry summers, as well as the chief technology officer from president obama's first term.

>> thank you all for joining us this morning. my name is melissa carney. i have the privilege of moderating our first panel discussion this morning. this panel is going to take the premise that andy and erik laid out for us, that there has been rapid technological advance, particularly in the information sector. we will ask the question, what does that imply for the future of work, the future of workers and the nature of employment in this country in particular? as we i tried to lay out in our hamilton project framing paper, there are a lot of views on this topic. in particular on whether this is going to be good or bad on the for the whole of society. fortunately this morning we have a really expert group to discuss these issues with us. truly, i would say some of the

leading minds in the world on these very questions. you have their full bios in your program. i won't run through them in detail but i'll just briefly introduce our panel. to my left is david autor professor of economics at m.i.t. one of the nation's leading economists who has probably contributed more to the nation's leading trends and labor market than any. we have larry summers, university professor, president emeritus at harvard university and served in a number of senior policy positions including secretary of the treasury of the united states and director of the national economic council. aneesh chopra served as our nation's first technology officer appointed by president obama and served as the virginia secretary of technology. he's the director of hunch

analytics. and erik brynjolfsson has already been introduced. and still a professor at m.i.t. the way we will do this, i will pose an opening question to each of our panelists and we will move to a moderated free flowing discussion and leave the final 10 minutes for audience q&a. we will be collecting your questions on note cards which then will be brought up to the panel. david, i'm going to open it up with a question for you. you have written extensively about the nuanced relationship between technology and computers and workers, particularly noting that there are certain things that computers can do that substitute for tasks historically or traditionally performed by humans and other things computers do that complement tasks performed by humans. so, in light of your research and the framework that erik and andy have laid out for us, how do you see this all shaking out for workers?

>> that's a great question. i'm honored to be part of this discussion and really like the work they've written. i'm glad this topic is getting the thoughtful discussion it deserves. 15, 20 years ago, erik and i started talking about this when i was a graduate student and erik was assistant professor. at that time we felt people weren't taking this issue seriously. if anything, i thought people should not panic at this point. i think there are a number of remarks i could make. i think there's reason for some skepticism about how fast things are actually moving and a lot of aggregate data that don't support the idea the labor market is changing or economy changing as rapidly as the story so the premium for higher

education has plateaued over the last 10 years and we see evidence highly skilled workers are moving -- have less rapid career directories, moving into less skilled areas. an important part of the puzzle and productivity not moving rapidly and a lot of growth has been in relatively low education service occupations. captioning performed by vitac captions copyright national cable satellite corp. 2008