|0:36||Intro. [Recording date: September 27, 2011.] Inequality and the standard of living of the average American. You hear a lot, in books, in the blogosphere, among pundits that the middle class is being hollowed out, the incomes for all but the top 1% or maybe the top 20% are stagnating; that the gap between rich and poor has grown inexorably; that the American dream is dead; that America is no longer the land of opportunity. Particularly alarming is the claim that these changes have occurred at a time--until recently--of great growth in national income: very healthy grown in the United States in per capita income between the early 1980s and the mid-2000s. Yet the claim is made that the rising tide did not lift all boats. Rather, it only lifted the big, rich yachts. The dinghies of the poor and the middle class did not share in that tide of improvement, share in that wellbeing. What are your thoughts on these claims? They are basically wrong. And that's what 10 years of research by myself and James Sullivan at Notre Dame has shown. If you measure incomes better, accounting properly for inflation, median incomes have gone up by about 50% since 1980. And if you look at consumption, it's gone up by a similar amount over that whole period--though exactly when it went up is slightly different from the income pattern. And you are going to argue that that is just a correction for a mistaken measure of inflation? Let's stick with the income side; we'll come to consumption later. Inflation is probably the most important reason, but also the picture looks much better if you look at what people are actually able to spend, which is the income that's left after taxes. So, taxes also matter quite a bit at the median. Particularly in the 2000s, if you account for taxes, the median has gone up much more than if you don't. And that is because of...? Cuts in income taxes in 2001 and 2003. But, Bruce: The Bush tax cuts all went to the rich! We all know that. No, they were pretty evenly spread across the distribution in percentage terms. In percent. A little secret that doesn't get talked about very much. I just thought that I'd just mention that. So, what I want to start with: There are of course other measures of despair that pessimists--I'll call them the pessimists, the people that think that the rich are getting everything and the average person is getting nothing or less, 25 years. So, if you look at hourly wages, corrected for inflation, they look pretty flat since about 1978. You hear the claim that median income is basically unchanged since 1973. Is your basic claim--I want to push you on what you said a minute ago--let's start with the correcting for inflation. What's wrong with those statements? They seem pretty powerful. They are corrected for inflation, of course. They are not nominal income. So if you look at hourly wages reported by the Bureau of Labor Statistics (BLS) and median income reported by some folks, I assume taken from the Bureau of the Census, it looks pretty dismal. Why is your story better? Well, back in 1996, the Boskin Commission concluded that our main rate of inflation, the CPI-U, overstates inflation by 1.1 percentage points per year at that time; and then historically they said it overstated inflation even more. And over time, we've made some corrections to how we've measured inflation, but the Commission members, when polled more recently, concluded that still the bias was about .7 to about .9 percentage points per year. Now, why do we not measure inflation well? Well, it's very hard to keep track of new goods and to adjust for the improvements in the quality of goods. Take, for example, cars. The car you drive you drive, that most people drive now, is nothing the car of 20 years ago. Anti-lock breaks, power windows, power steering, power breaks. More cup-holders. Everything's better. Everything is so much better. Braking, acceleration. |
|6:19||But doesn't the Bureau of Labor Statistics (BLS) try to correct for that using hedonic price indices? Or do they not? They do, but they don't do it adequately; and it's a very hard thing to do. And the problem is even more difficult when you think about new goods. If you think about DVD players, cell phones--it took the BLS 15 years to include cell phones after they were introduced. What do you mean? They weren't part of the Consumer Price Index (CPI) for a very long time. We should back up for a little bit. So, what the BLS does is it goes out regularly and it tries to measure the prices of a basket of goods where the weights on the basket are proportional to the spending that people do on these different goods. And when a new good comes along, it can't be in the basket the first day, obviously, because they haven't gone out and surveyed the price of that. But you are saying it took 15 years from the introduction of cell phones till where they were even in the basket? Exactly. And that's been historically true, that it's been very long before goods have been included. Most, historically, the vast majority of people had electric refrigerators before they were included. Used cars weren't included till, I think it was the 1940s or 1950s. Of course, in the beginning, very few people are buying these goods. But in recent times, the beginning doesn't last very long. So, leaving out cell phones for 15 years, you miss a big component of something that is actually falling in price rather dramatically, when it's 15 years, I assume. Absolutely. Now, the other issue they struggle with, which you mentioned, is quality. So, you have an iPod, which when I think it was introduced--I think it was around $500--the iPod that now sells for less than that in actually nominal terms holds more songs; the battery life is better; it's better, it's smaller, etc. Do they correct for that at all in the car example you mention? You said they did it poorly. Did they do it for a good like an iPod? I know they do it for big items; they try to. You know, I'm actually not sure about iPods. But in general, they are reluctant to do hedonic corrections because it's a lot of work and there's certainly some judgment involved. The preference is to accurately measure what's not quite the right thing. |
|9:10||So, we might use the word hedonic. The idea of that is--I think it comes from work by Zvi Griliches originally--where the idea was to say, well, if it's better, some of the price is it's for the better good; to compare it to the old good, you have to take into account or somehow add a dollar figure for the improvement in quality. There are statistical techniques that help you try to do that. But as you point out earlier, it's very difficult to do. So, given all those changes; there's also been criticism of the BLS because they've often not surveyed some retailers such as Wal-Mart or super-Walmarts that were driving down prices and their baskets were not measured correctly. That's all true. I'm very sympathetic to that argument. What I'm less sympathetic to is how you mechanically go about logistically correcting the actual numbers. So, you said members of the actual Commission were surveyed; and they are an illustrious group of folks immersed in some very dull and dreary stuff, to be honest. And bless them for their diligence. But surveying them and saying, well, we think it's about .7 or .9--that's kind of just a guess, for most of the time. There are people like Mark Bils at the U. of Rochester who has done extremely careful work on manufactured goods showing an actual careful work on the bias on one small portion, an important portion, of goods. It's quite significant, and he's tried to correct for these quality changes. But how do you do that in a systematic way without just saying: Well, we know that real wages have grown more than the data show? So, there are two ways. You can put together a lot of careful studies like the Bils study for particular groups of goods over particular time periods and try and extrapolate from that. And that's basically what the Boskin Commission did in coming up with their numbers. Another way is to look at what you would have to adjust incomes by so that spending patterns over time at a given level of adjusted income stayed the same. Okay? Now that's a little bit harder to picture. You want to try that again? Sure. The idea is that: Think that if people's true incomes were at the same level over time, they would spend about the same share of their income on food, clothing, and leisure activities. So, you look for what level you would have to adjust income by so that spending patterns stay the same over time. And several people have done that--Dora Costa, Bruce Hamilton--and they come up with biases to the CPI that aren't that different from what the Boskin Commission concluded. The thing I find remarkable about this--and I just have to get this in because it's so extraordinary--one of my reasons for being skeptical of the mainstream, standard view now the average person has made no progress now for the last 30 years is I look around. Now looking around is dangerous. It can lead to very bad misperceptions. I live in the Washington, D.C. area; Washington, D.C. thrives through most recessions and it's thriving through this one. So, if I say: Well, the recession's not so bad; I went to the mall the other day and everybody was in there and it was packed and stuff was selling out the door. So, that's a stupid claim, to say because I looked around and things looked pretty good that this whole recession thing is exaggerated. But when I look back to 1973, when I was alive and perceptive, meaning conscious and aware of stuff, I was 19 years old--when I look around at how much the world has changed since 1973 and I see how much the car of the average person drives has changed, the house of the average person lives in, the health care that's available to so many more people at unbelievably higher quality; the innovation, the revolution in technology, the innovation; and you tell me that the average person is not any better off? I start to think maybe there really is something wrong with the data. And the thing that I find so remarkable is the pace of the improvement of daily life is what makes it hard to measure and gives you the impression--incorrectly--that the average person is falling behind. People say: Well, the 1950s and 1960s were more fair. Well, the 1950s and 1960s were pretty dull and pretty stagnant. And there wasn't a lot of innovation. Your car looked kind of similar at the end of those decades as it had at the beginning. And your refrigerator was kind of similar. Now, 10 years go by and everything's better. And so I think it's incredible that because of that people are led to the false conclusion that people aren't catching up, because they are mismeasuring these things such as inflation.
|14:55||I want to absolutely agree with what you just said. And I think that I need to go back a bit and emphasize that you were getting at something right when you said it's hard to measure how much we overstate inflation in our official price indices. I don't think people would universally agree that the Boskin Commission got it exactly right. There is a lot of uncertainty about what the exact bias in the CPI is. And it probably has varied over time. No doubt. But the way I can also point out that the price indices have to be overstating inflation is to look at the type of things you just mentioned. You can look at the types of houses, say the bottom 20% live in. The official measures say that the bottom 20% are worse off now than they were 30 years ago. The poverty rate now is quite a bit higher; and even two years ago it was quite a bit higher than the official poverty rate of 1980. But if you look at the bottom 20% of the income distribution, in terms of the house they live in, the car they drive, they are completely different. They have a dishwasher often, they have a washing machine; they have things that even an average person didn't have in 1980. In 1980--let me grab the numbers--27% of those who were in the bottom 20% of the distribution had central air. Whereas in 2009, it was 67%--almost three times the share. So, things have dramatically changed. And if you look at numbers for ownership of a dishwasher, a washer and dryer, those numbers have gone up similarly. The counter to that is: you are cherry picking a few toys, a few electronics; those have gotten cheaper but a lot of important things--housing, healthcare, education--crucial things--food's gotten cheaper, undeniably, but those big items which are hugely important have gotten a lot more expensive. So, just to point out that the poor have more goodies, that's an illusion about their improvement. Well, housing is really important. Housing is about 40% of consumption. For the poor, for the middle class, for people at high incomes. And the size of housing units that people live in has gone up dramatically. The other things we mentioned--air conditioning--that's gone up. You can look at lack of problems like leaks in the roof or in plumbing, and those have almost disappeared. And what about the education and health care argument? Healthcare is complicated. We try and stay away from it actually in our research because there is less agreement in the profession about how to handle health care. The quality changes there have been tremendous. Unlike education, where it's pretty much the same. The key issue is how to deal with quality. And if you look at some things like the access of children to a doctor, that has clearly increased over the last 30 years. So the share of children that have at least an annual checkup has gotten better. But if you look at other measures like the share of people at the median income who have insurance, that has gone down over time. Of course, insurance isn't all you care about. You care about health care. If a homeless person has a heart attack on the street in 1980, versus 2011, the medical care he is going to get in the emergency room--which is required by law, this is not a market phenomenon, though it could be if the law didn't require it; we don't know--the drugs, the treatment, the technology that that person is going to have access to is much, much better today. That's to me the fundamental question: what's your access to health care? The median--that's a little bit depressing. If the median person has less access to all technology--artificial knees, better treatment for cholesterol that reduces your chance of a heart attack--that would be alarming. Of course, it's gotten a lot more expensive per unit for a bunch of reasons, many of them driven by bad policy in my opinion; but that is something that would offset some of these gains we are talking about. That's something that we cannot speak to as much as we'd like. So, I'm going to have to leave that for another guest on your program to talk about--access of the average person to various treatments. Which I think would be a terrific topic. I'd like to do that. On the education point--my story is that the reason education has gotten a lot more expensive is that a lot more people have access to it and they've pushed up the price of something that has a relatively inelastic supply, where supply doesn't respond very much, so an increase in demand pushes up price a lot relative to quantity. So we have a lot more people going to college, to community college, to graduate school; so of course it's gotten a lot more expensive; but it's not like the average person can't afford it. It's expensive because they can afford it and they've pushed up the price. The counter to that is the only reason they can afford it is debt. Maybe we'll come to that in a little bit.|
|21:40||Let's go to the issue you brought up at the beginning about post-tax and pre-tax income. When we look at most of the charts that the Census Bureau puts out, or that are typically in newspapers, they are looking at pre-tax income, not post-tax income. They are certainly not looking at transfers that people receive in the form of welfare or other things. When we talk about the poverty rate, we are talking about the rate that people are in poverty based on their own earnings, not on what their standard of living is after they've received transfers from the government. How important are those factors in your conclusion that the average person's wellbeing has increased by 50%? How much of that roughly is an inflation correction and how much is the fact that we really want to look at post-tax income and consumption? Most of it is the inflation adjustment. But it is certainly true at the median that a substantial part of the increase, particularly in the 2000s, is due to tax cuts rather than just inflation. Now, if you look at those in the bottom 10% or 20%, taxes matter a lot more. Transfers, I assume. Oh, if you go back, because many of them got taken off the rolls. Because if you go back to the 1960s, we often were taxing people at a 15% marginal rate even at very low incomes. We've raised the exemption; we've added the earned income tax credit, which transfers a huge amount of money to people at the bottom. So the combination of tax cuts and expansions of the earned income tax credit have dramatically increased the after-tax incomes of the poor relative to the pre-tax incomes. Now, one other issue that comes up in these conversations that drives me crazy, and I don't know whether you've dealt with it in your work: When people compare the median in, say, 1973, to today or the bottom 20% or the top 1%, first thing is they are not the same people. People move in and out. There's a big debate in economics about how much movement there is. But when people talk about how much the top 1% get, they make it sound like somehow it's an exclusive club that somehow has found a way to extract money from the rest of us. Which some part of it is, in the form of subsidies to farmers and financial institution executives. But putting that to the side, they are not the same people in 1973 as today. People die; there's immigration; people are born and come into the workforce. But the most important thing that I feel is neglected is that starting in the 1970s--and it's not a coincidence when this stagnation starts--there's a big increase in the divorce rate. So, when you are doing household analysis which I know some of your work is based on and you are talking about the bottom quintile or the middle quintile, people in the 40-60% range of the income distribution, you are talking about radically different groups of people. In other words, who the median is can change radically if you add a whole bunch of new households due to divorce; and that median can go down dramatically even if the person who was the median before is doing better. This statistical paradox is usually ignored by people with an axe to grind who want to sell this story that there's been a lot of stagnation. Do you correct for any of that? We do worry about that quite a bit. What we do is we look at different types of households--married parent households, single parent households, single families without kids, married couples without kids, households that are headed by someone 65 or older. We also split up the population by race and whether or not the household head is employed or by education. And we look within each of those groups and see if the distribution has changed within those groups. And then we add up all those changes across the different types of families. And you get a different answer when you do it that way versus just lumping everyone together when you are doing this kind of quintiles stuff? The changes in the types of families that are in the United States over time don't play a big part in the changes in median incomes or poverty over time though you can see an effect of increased education, particularly if you look at measures of consumption because consumption seems to be a little more closely associated with improved education than income is. I'm surprised to hear you say that because on the poverty side at least because that's something I know just a little about explicitly. A study was done--it came out in the last few years, I've forgotten the authors--they looked at poverty rates by family structure, as you just said. They looked at what's the poverty rate for married people with both people are working, what's the poverty rate for when one person is working, for a couple with no children, for single man, single woman, with and without children. And what they found, and this is a paradox that people have trouble wrapping their heads around: every single demographic group a dramatic reductions in poverty over a long period of time, say I think it was through the 1970s through recently and yet the overall poverty rate went up. Every demographic group the poverty rate fell sometimes dramatically but the overall rate went up. Why? Because the group with the largest proportion with the highest poverty rate, which is single women with children, that group got bigger. So when you compare the modern poverty rate to the past the rate went up even though every single type of family had had a reduction. So I find it surprising you don't find a bigger effect. Well, the share of families headed by single mothers hasn't gone up that much in the last 10 years. Most of the change was before. 1970s and 1980s. So what you're describing could be true more historically. What we do see is that there are groups that have had their poverty decline much faster than others, so those households that are headed by someone 65 and older have had their poverty decline much faster. That's also true for single-mother-headed households. On the other hand, married couples with children, their poverty rate hasn't declined as fast as other groups. And that's even more true if you look at their spending patterns, at what they are consuming rather than their income. Over what time period is that, roughly? That's since 1980. Interesting. I have to dig deeper into your work on that. I'm looking forward to it. I would also like to take a careful look at the studies you have in mind. |
|30:19||Let's talk about consumption. The first punchline of your story is that when you correct for inflation, if you do it correctly, even when you are looking at different people over time rather than panel data where you are following people over time, the middle class stagnation that we hear about is not real. There's quite a bit of improvement. Why do you also look at consumption and what does that tell you? What are you looking for there and what are some of the findings? There are two reasons to look at consumption. There's the standard economic argument that one year's income is not really a good measure of a family's wellbeing because families save for a rainy day and they may have a period where their income is low but they are able to consume because they have saved. If you think about older Americans that idea becomes particularly important because the retired households will often have very little new income but maybe consuming out of their past savings. They may also own a house, a car; in fact, more than 80% of households headed by someone 65 and older own their own house, own at least one car. Some of them probably shouldn't be driving that car. Fair enough. My Dad's 81; he still drives. I'm glad. A little scary, though. Leaving that issue aside, which is real: It's clear that the incomes of those who are retired is not a good measure of their wellbeing because of the ability to draw on savings; because they get a flow of resources from assets they own like homes and cars. Now, that's the standard reason for looking at consumption, but when you are looking at people at the bottom, which we do in quite a few of our studies, there incomes are very complicated. There's a wonderful study of single mothers by Eden and Lane that talks about their various sources of income; there's formal work, informal work, money that is given to them from family, friends, boyfriends, fathers of children that may no longer be boyfriends. And in addition money from government transfers. With all those sources, it's much harder to respond in a survey to what your income is. On the other hand, these same people do a pretty good job of reporting what their rent is, what they are spending in a given week on food. So that looking at consumption can give you a better idea of what people at the bottom are able to do in terms of their living standard. There also is the issue that our surveys have to try and get people to give out information that they are not always so eager to give out. When you ask people about their income, they often won't respond to the survey if they know they are going to be asked about income questions; when you get to income questions, they are less willing to answer those questions. And they often don't give you a very accurate answer. I've spent a fair amount of time looking at the accuracy of the responses in the main survey that we use to collect information on income and inequality, which is called the population survey. And if you match the data from this survey to the government reports of how much they've paid out in food stamps, for example, the survey only captures about half of what's paid out. Interesting. It's not that people necessarily are ashamed of the fact that they receive food stamps. It's just that it makes it a lot quicker to respond to the survey if you just don't start to talk about all of the benefits you receive from the government and how much you received and what months you received them. So, our surveys are doing a poor job of capturing some of the government transfers that people at the bottom receive. So, you try to correct for that, and what's the implication? What do you find that might be easy to summarize? That if you look at trends in consumption over the last 10 years at the bottom, the rate of poverty indicated by what people are consuming has gone down a lot faster than a rate based on income. In fact, the official income measure says that poverty has gone up over the last 10 years. Now, admittedly the last 3 years have just been awful and we would expect poverty to have gone up. But over the longer term looking at other measures like the types of housing people live in, the types of cars they drive, it suggests that people at the bottom have seen an improvement in their living standard over the long term.
|37:04||What is the role of debt? It's one of the things, a meme that's circulating that's common now: the reason that we had such a bad set of times over the last ten years is that people saw their income stagnating and they had to borrow to live better. Usually that's not a very good strategy. It's hard to get people to lend you money. The last 10-15 years the U.S. government has tried to make it as easy as possible to borrow money to buy a house; and that I think was a disastrous public policy we've talked about many times on the program. But when we think about consumers going into debt to raise their standard of living--how common a phenomenon is that, and how would that distort attempts to use consumption as a measure of material wellbeing? I think that's a real issue, particularly when you are talking about median incomes. When you look at people near the poverty line, they are not doing a lot of borrowing. So, debt is not a big part of the increase in consumption at the bottom. For the median I think that's more of a real issue. You can see that in the last couple of years consumption has gone down, as people have had to pay back debts. They've also been worried about the state of the economy, and their incomes have fallen. So, you see a decline in the last few years. My take on this is: You could look at income, you could look at consumption; but consumption shows you what people were consuming in the good times. And it's in the bad times they have to repay debts; if they are worried about the future then they are not going to be consuming. And you also want to look at that. We've been talking about the average level of how people are doing, whether they are above or below poverty, how well has the median done over time. I want to turn to inequality, but before we do I just want to mention that the study we've been talking about on poverty is by Hoynes, Page, and Stevens, in the Journal of Economic Perspectives. What they found that was so remarkable is that every single demographic group, their poverty rate fell by double digits in percentage terms; and other than single men, it fell by 20% or more. But the overall poverty rate fell by about 4%, and that's because the group with the highest poverty rate, single women, had grown. This is 1967-2003, which is capturing this era of alleged stagnation. For me, the crucial question is, does a rising tide lift all boats? Between 1967 and 2003, GDP per capita, any measure of our economy, grew dramatically. The idea that it wouldn't reduce poverty would be very depressing about the state of the American economy. I think it's somewhat distorted by the demographic changes--dramatically distorted, actually--that happened over that time period. |
|40:44||But let's turn to inequality, which is the other issue you hear about, that the gap between the top and the bottom is growing. Your claim is that the argument that the top got all the gains of the last 30 years, which you hear all the time--that between 1980 and the present or 1990 and the present, the average person got none of the gains; they all went to the rich; the rich's share grew--your claim is that those statements are wrong because mis-measured inflation and a focus perhaps on pre-tax income. But what about the inequality ratio between the top 90% and the bottom 10% or the top 1%'s share of income? What has your work been on that? We've focused on the difference between the 90th percentile and the median, and the median and the 10% percentile. We haven't looked at the top 1%, so I can't speak to that. What we find is that in the last 30 years, there was this period in the 1980s where there was a very big increase in inequality no matter how you measure it. Whether you look at income, consumption, pre-tax or after-tax income, the late 1970s and early 1980s were bad for inequality. But since then, the picture, when you measure things better, when you look at after-tax income or better yet when you look at consumption, things aren't that bad for bulk of the population. So, the difference between the 90th percentile and the 10th has grown a little bit over time since the late 1980s but not a lot. And if you look at the bottom half of the distribution, the 10th percentile relative to the 50th, the 10th has grown relative to the 50th over that period of time. When you say grown, you mean their standard of living has gotten closer to the median. Yes. And that's particularly true in the last ten years. So, there has been an improvement in the living standards of the 10th percentile relative to the 50th. Now, there has still been some slight worsening of the median relative to the 90th percentile in recent years, but it's been slow. My claim is that the average person has no idea unless they look at the data unless they look at the data where they are exactly and whether they should be depressed or whether it went up a percentage point or two. What I found striking about your papers is how small the changes are compared to what you hear, and I think it would be useful for people to check those out and see what those magnitudes are, because they are relatively small. A big part of that, at the bottom, as I mentioned earlier: our tax system has moved toward not taxing people with very low incomes who work, but instead providing them a work subsidy through the earned income tax credit. And that has meant a big increase in the 10th percentile. Now, that's true--the income tax--but they do pay a healthy amount of payroll tax as a percentage of their income, because it's a flat tax; and those rates have risen over time. So, you take that into account? We do. We take into account Federal and state income taxes and payroll taxes. |
|45:01||Now, you said that the late 1970s and 1980s there was a relatively large change in income and an increase in income inequality. Do you want to speculate about why that is? Most people just talk about inequality as if it is some inexorable fact of life, but the level of inequality emerges from a complex process. Some of the reasons it emerges are good: example of Sergey Brin and Larry Page--they founded Google and they make our lives better and vaulted from being graduate students and unemployed to being in the top 1%. More power to 'em, I say--well, not power but more wellbeing to them--because they've made my life and others' much better off. They've captured nearly all of it. And then there is the not-so-good, which is, to me, bailouts of Wall Street or rich farmers. There's other things going on at the same time. Any thoughts on those large trends? There's been some work on the increase in inequality, and a fair amount of it has been attributed to changes in the demand for high-skilled labor and the failure of us to supply enough high-skilled labor at the times when demand for it was increasing. There are also other stories. Around that time there was little increase in the minimum wage, which can decrease inequality in certain circumstances. If you don't measure unemployed people as 0s, which we don't in the data, it can be a little bit misleading. If you are looking at income then you are capturing that, so there is some evidence to suggest that even though minimum wage will have disemployment effects and other negative consequences, they are associated with some improvement in inequality in some circumstances. Fair enough. And at that time we also had very high inflation. The early 1980s, some people have suggested that that could be part of the story. I don't think we fully know what seems to be behind the inequality, that seems to be so focused on those years. A couple of things I could mention. The other part is globalization. You mention a change in relative demand for high-skill versus low-skill. One of the reasons that low-skill had been in less demand in the United States is that many of the things they had produced were being produced overseas. They now are competing with workers who are cheaper than they are and that puts downward pressure on their wages. There is also technological change, which has made low-skilled workers--they are in competition with machines. And so, when people talk about loss of manufacturing jobs, they often forget that a lot of that comes from productivity increases. We don't need as many people to make stuff any more. Which is a good thing; as is globalization--both the same thing, both forms of productivity. But our school system is so mediocre, our K-12, it's perhaps not surprising in the face of that tremendous change, globalization, it hasn't responded very vigorously in how it's responded to people who go on to college. If anything, it's done a worse job for them. So, that's part of the problem. Over most of this period, we don't see much of an increase in inequality. I don't think that globalization or immigration are the threats to our standard of living that some people have claimed. And you are saying that because during this time of increased globalization, immigration--it's the earlier data. Yes. It's the late 1970s, early 1980s where there was a sharp increase in inequality, not in the last 20 years certainly. When the pace of globalization and immigration have gotten more intense. That's interesting, very interesting point.
|50:30||You said you don't look at lot at the top 1%. We are not going to look at why they are in the news, but we are in a time in September of 2011 when people are talking about raising taxes on the top 1% because they earn so much money. And they do. For good reasons sometimes, and sometimes for not so good reasons. But haven't focused on the 90-50, the comparison of the 90th percentile to the median, you are arguing that these standard stories that there has been this big gap in wellbeing between the average person and the people at the upper end has been exaggerated. Is that a fair summary? It is absolutely fair. So, what are those other studies doing wrong? If you look at the 90-50, for one, and if you look at earnings even in the last 20 years, you don't see that big an increase in inequality. You see compression in the bottom half of the distribution. So, the 10th percentile is rising a bit relative to the 50th. So part of it is that people are focusing on the top 1%. Another things is that people are focusing on the earlier years, the 1980s. It's true that if you look after tax, things look better, but still even pre-tax the increase in inequality is not that sharp in the last 20 years, leaving aside the top 1% which I'm not going to really be able to speak to. The top 1%--a lot of them are entertainers, movie stars, sports figures--people who have been able to leverage their skills in dramatic ways because of improvements in technology and they are more valuable than they used to be; which is just fine with me. It includes people who are great entrepreneurs. It includes people who earn money because they are being subsidized. Our industry, education, is highly subsidized in direct and indirect ways. When I get depressed about Wall Street, I am slightly consoled by the fact that the subsidies to borrowing and lending on Wall Street by large financial institutions increase the demand for academic economists; and you and I are probably the beneficiaries of that. I'm not proud of it, but it helps me sleep better at night. The top 1%, there's a lot of drama there inevitably. But your point is when you take a more sober look at the non-rarified air, the glass is half full. I encourage people to look at the diagrams on the web. It's hard to describe. The chart is better than 1000 words. But, do you want to try to give a feel for what the magnitudes are when you say it hasn't increased very much or it's very flat. If you look at after-tax income inequality, say the 90-10 ratio, it's about the same now, well, in 2009, as it was in 1993. That's surprising. It did go down a little bit in the 1990s and then came up a little bit in the recession of 2001-2002, but it's almost the same level as it was in the early 1990s. Which I think would surprise many people. And that's accounting for taxes. The significant part of that is that the 10th percentile when you account for taxes has done quite well. The top half of the distribution, that ratio of the 90th to the 50th, that still has gone up over the last 20 years. By how much? The ratio of the 90th to the 50th stood at around 2.5 around 1990 and it's now more like 2.75. I call that a relatively small increase. Yes. Not what I think the average person would say, given what they hear all the time. It's very different from what you hear about the top 1% or the top tenth of 1%. Or the CEOs of the Fortune 500 companies. They run much larger organizations than they used to.
|56:44||We've talked a lot about numbers. You want to talk about what you think might be done to make America a more healthy place to seek opportunity--which is what I care about? The numbers are important, in debate all the time, they matter. But what I care about ultimately is whether my children and your children and grandchildren and somebody who is not doing as well as we are, their children and grandchildren, whether they are going to get ahead or not? Whether they have a chance to dream and achieve. What policies or issues might be relevant? I think we want to focus on the fact that long-term growth has been pretty widely distributed, and that we should keep our eye on the ball on what will continue to keep economic growth high. And we want to distinguish between policies that may be appropriate for a recession and policies that are focused on long-term growth. Because those may be different policies, and we don't want to justify certain things because of false claims that the growth rate over the last 30 years in incomes has been small. Incomes have risen when properly measured, a lot over the last 30 years; and we shouldn't use false claims of stagnation to justify things like closing the borders to goods and immigrants.