Time: 01:09:23
Size: 27.7 MB
Right-click or Option-click, and select "Save Link/Target As MP3.
Readings and Links related to this podcast
|
Podcast Readings
HIDE READINGS |
|---|
About this week's guest:
About ideas and people mentioned in this podcast:
|
Highlights
| Time |
Podcast Highlights
HIDE HIGHLIGHTS |
|---|---|
| 0:36 | Intro. [Recording date: March 29, 2012.] Russ: Topic: Economic status of the middle class. We hear a lot of talk about the middle class as disappearing. Colleague Tyler Cowen talks about the Great Stagnation. There's a view that the rich are getting richer; everyone else is either treading water, falling behind--so inequality is growing. And of course there are statistics to back these claims up. The median wage rate is growing very little, if at all, since the 1970s, and there does appear to be growing inequality. But it turns out a lot of these data are subject to interpretation and decisions have to be made by researchers on what they mean. And what we are going to talk about today is a paper you've done with Jeff Larrimore and Kosali Simon from the March issue of the National Tax Journal, which is called a "Second Opinion on the Economic Health of the American Middle Class." And what you show is that depending on how you define income and the unit of analysis, you can get a very different take on what is happening. So, start off by telling us what you did and how you came to the project. Guest: Sure. I've been looking at issues of income and poverty for the last 30 years, really. And my efforts in this regard actually go back to at least 10 years ago, when people were looking at what was happening in the 1980s. And that's when this issue first began--the issue of the middle class, the rich getting richer, the poor getting poorer, and the middle disappearing. Just confusing conventional methods in those days, using what the Census Bureau uses to measure income. We--that is, income from the current population survey, pre-tax, post-transfer income--that is, the income that people report when the Census Bureau guy knocks on your door and you say how much are your earnings, how much did you get in interest and dividends--those sorts of things. When you simply did that and looked at the distribution of household income of individuals and grade them from no income to the person with the highest income, you get a distribution. And we did something just very simple. We looked at this distribution--and it's kind of a Bell curve. There are two tails and a bunch of people in the middle, a little hump there; and we did it for 1979, which was the start of the 1980s business cycle; and we did it for 1990, which was the end of the business cycle; and we put those two distributions on top of each other. And we found this really kind of interesting thing. It was absolutely true that when you put the 1989 distribution on top of the 1979 distribution, the middle shrank. It disappeared. But where did all those people go? Well, they disproportionately became richer. So, we actually could see that you ended up with a very small portion of the tail that got poorer, about 10%, and of the mass that you pushed down from the middle, moved to the left and became a little poorer. But 90% became richer. So, this whole business that was talked about, about the middle disappearing--if you think about it mostly as the middle mass, where most of the people are--it is true that that disappeared. But most of it disappeared by people getting richer. So, I've always been suspect of people who make these claims just using the simple statistics that the government provides us. |
| 4:52 | Russ: Now, let me just clarify that. So, obviously if somebody asks you what you earn or what your income is, what your salary is, most people have in mind is what you said before--their pre-tax income. They don't think about what their after-tax income is. If they were asked, they might be able to get close to that. Most of us have some idea what we pay in taxes. Although I would argue that it is pretty imprecise except around the month of April. But when we look at government data on income, we are typically looking at the pre-tax income, to start with. And what you are talking about, this change over that decade of 1979-1989, you are talking about when you include non-earnings. You are adding in interest income, you are adding in government transfers? Is that what you are talking about, or are you still just talking about pre-tax income? Guest: No, I'm absolutely talking about that. That's right. This all gets very interesting when we decide, well, what is it that we are going to count. So, historically--well, what I should say is that empirical economists are really limited in what they can say by the data that they get. So, a major scientific innovation, beginning in the 1960s, was that in the 1967, for the first time, the Census Bureau became willing to go out and poll people in their homes; and we started to have an annual current population survey. In that survey, which is now the most important data set used to measure the income and employment of Americans, the surveyor goes into the door. They ask the principle person in the household to describe all of the sources of income of that person and all the other people who live in that household. And that is both market income--that is, returns to land, labor, and capital--and any transfers from the government. The technical term for that is pre-tax, post-transfer, in-cash income. Russ: So, it does not include in-kind benefits: It would not include food stamps; it would not include my vacation; it would not include my health insurance? Guest: That's right. Exactly. And also any subsidies you get for housing or many other in-kind-- Russ: non-cash-- Guest: that's right, exactly. Russ: Okay. So, starting in 1967, the Current Population Survey (CPS), gets that information. So, if I'm trying to figure out what's happening--I just want to make one other key point here, which is that in these debates, people often focus on the median, which has a very good thing about it. Which is, it's not sensitive to outliers. So, if the overall population is getting a lot, if national income is growing but it's growing because a few people at the very right-hand tail are gathering all the goodies, which is what some people think, then the average could go up, but the average person wouldn't be getting any of it. So to avoid that problem, the right-hand tail in distorting what's happening to the typical American people, uses the median. Now the median has its own problems, which we'll come back to later. But its best advantage is it avoids this problem of outliers to the right or to the left. Somewhat to the left, because zeros are obviously a little bit weird. But we'll put that aside. So, the advantage of the median is it isn't distorted by outliers. But one of the problems is what you count still is difficult to decide. Or sometimes you may have an axe to grind. So, if you want to make the growth in median income look small, you leave out in-kind benefits, and cash-benefits, outside of income, because they've got more important since 1979. Guest: That's right. And it's not only true that they've become more important because they've become an increasing component of the transfer system of government to low-income people. But even the compensation of workers has increasingly been in non-wage compensation. And the most-important one is the employer-provided health insurance. So, approximately 75% of all the workers in the United States either have their own employer providing them with compensation in health insurance, or they are the spouses of someone who is working and they are on their plan. So, this is a major part of the compensation program, and it is not included in measures of wage earnings or in income. Russ: Let me ask one more clarifying question, because you raised it when you said how the survey is administered. So, I'm sitting at home. The doorbell rings. It's the Census Bureau. And this is an annual survey, not the every-10 year, correct? Guest: Right. It's done each March. And while the CPS actually has people in it for a period of 16 months, it's only in March they actually ask the detailed questions about income; and the other months they ask questions about employment. Russ: So, the guy knocks on my door, and the unit of analysis--which we are going to come back to later--in this case, with the CPS, the unit of analysis is the household. So, if my father lives with me, and he receives Social Security, that would be included in post-transfer household income. Not, obviously, pre-transfer--it's not earnings. Just looking at the income of the household, my father doesn't work. But if he lives in my house and he gets Social Security, that would be included in the Survey--if I remember to tell them. Correct? Guest: That's correct. |
| 11:00 | Russ: What else would be in that household income that might be a little bit offbeat or might be a little bit unusual? Because I think it's important for people to remember: This is not a bunch of individuals who get surveyed. It's the household as a group. Guest: Well, that's right. And one of the major issues is that there are three kinds of sharing units, as we call them, that you can think about. The idea here is that if you are doing a study of wage rates or wage earnings, the natural unit to look that is the individual. And so we don't worry about what the individual does with its earnings. But once you recognize that people share things, and we do that in a natural way, when you are trying to look at people's economic wellbeing, you actually want to think about what their power is to consume. And what you recognize is that if you simply looked at the personal income of people who live in households, you'd get very strange results. You'd get that a dad who works fulltime in the marketplace and a mom who works fulltime at home raising their children, you'd end up if you used the individual for your analysis showing that the husband was making a lot of money and was in perfectly good shape. Russ: The wife would be dirt-poor. Guest: Yeah. The wife would have no income of her own and therefore would be counted as having, as being poor. And of course the children would be poor. But that's of course an odd way to think about things. In fact, we live and share in units, and we share income. So, you want to be sure to get the sharing unit correct. So, there are three ways of thinking about that. One is the household. All the people in the household, you assume share their incomes together. But, what we know is there could be more than one family in a household. A family is defined as anybody who is your blood relative or who you are married to. And it's possible that that's the sharing unit. And the third sharing unit is the tax unit. And the tax unit is who you put on your income tax form when you pay your money to the government. Russ: So, that would include my daughter, who is in college, who makes some money, working part time. That money by itself would make her look desperately poor. But of course, she's not poor, because she is still in my tax unit, and really still in my household even though she does not live here. Guest: That's right. The tax unit is--in our work, and our most recent work, we really think carefully about which is a more appropriate unit for the sharing unit. Should it be the sharing unit or should it be the household? And I think there are two really nice examples of a difference between those two. So, the first is two people who share everything except a marriage certificate. And increasingly in the United States in the generation of the people in their 20s, 30s, and more, people are testing out relationships before they officially marry. Russ: Right. Guest: And so you can get two people who are not officially married, who are not blood relatives and who are not married, who in fact share everything. But if you use the tax unit as the unit of analysis, you are assuming that they don't share anything in their individual units. If you use the household, you recognize you are sharing. That's one example. The other is the one that you mentioned, that if your father lives with you and is getting Social Security benefits, he's a tax unit, a separate tax unit from you and your family. Russ: He's not my dependent. Guest: That's right, he's not your dependent. But he is living in your household and he is probably sharing things with you. And likewise your child, your adult child, who started out--got his college degree but is a so called-boomerang kid and is living in your basement--he maybe works a little bit but maybe not a whole lot. He would be considered a separate tax unit, also. Russ: And of course, one of the issues here, which you have to deal with, is the expression that "two can live as cheaply as one." It's not quite true; but there are a lot of economies of scale, obviously, within the household, which is why you are talking about the sharing unit: you are sharing the food, you are sharing the responsibilities of upkeep on the house perhaps. A lot of these things mean that this group--looking at people as per capita within this unit isn't quite accurate. Guest: That's right. So, the next kind of building block, once you've chosen a sharing unit, then you have to recognize that if you are going to compare sharing units, and figuring out how much income they have and which is a better sharing unit, you have to recognize that a sharing unit that is one person and has $50,000 is going to be a lot better off than a unit that has four people and the same $50,000. So, what you really want to do is not look at the sharing units, or tax units, but look at the income of a household, the income available to people living within a sharing unit; and that says that you take the total income, whatever it is that you've measured within the sharing unit and you divide by the number of people in that household or tax unit. But, as you said, the old saying is two can live as cheaply as one; that's not quite true or else you would just take that number. Russ: You just divide by the number of people in the household, but that's not quite right because you said, one person earning $50,000--two people earning each $50,000 and living together have a little more command over goods than that person. Even though the average in the house is $50,000, the two living together are going to have a little advantage in terms of their command over goods and services than the one person living on his own with $50,000, because of the economies of scale. Guest: That's right. You use the same amount of heat whether there is one person in the room or four; you can buy the economy-sized containers of milk and the large portions of meat, which is a little bit cheaper than buying the individual packs. Russ: A two-bedroom house isn't usually twice as expensive. A one-bedroom house where the two people sleep in the same bedroom is going to be the same price for the two. So, that's an even more dramatic example. Guest: Exactly right. So, for that reason the formula people usually use is take the income, divide it by the number of people, and then do it to a power; and if the power is 0, any number to the power of zero is 1; and that means that there are perfect economies of scale. That would mean that you love me so much that if I eat an apple, you feel satisfied. That's probably a bit much, so we don't want to say that a one-person household needs no more than a 10-person household. If you raise it to the power of 1, then of course that's per capita and you say there is no returns to scale. So what people do in the international literature, and I think this is a reasonable number, is they move it to the value of 0.5. I think if you move it to the power of .5 you get a very nice relationship. That is, a family of four needs exactly twice as much as a family of one to be at the same level of economic well-being. Russ: The power of .5 is just the square root. So, what you are saying is that if I've got a family of 4, which is, say, a traditional nuclear family of a husband, wife, and 2 kids, that family of 4--we could have many ways that that family could earn $100,000. Each parent could be working earning $50,000; one parent could be working earning $100,000 and the other working at home. But if the total command over goods and services is $100,000, that's equivalent to what of one person? Guest: $50,000. So, just take that 4, take the square root of it, which is 2. Russ: And divide that into $100,000. So, instead of taking literally per capita income, which would be $25,000, in the case of that family, you are really saying each family member is like living on $50,000, even though they don't each get $50,000 of spending money. Guest: That's correct. Okay. Keep going. Guest: Now, I'll just say that one of the nice things about that is that that simple calculation approximates the way that poverty lines are set traditionally in the United States. So, that elasticity, which is what we call that power, is very close to the elasticity that you would get if you looked at the poverty line for one person, a family of two persons, a family of 3 and 4, in the United States. Russ: So, analogously it doesn't just ramp up multiples of the number of people. It takes into account the fact that larger households can economize and get a larger per capita effective share. Guest: Exactly right. |
| 21:15 | Russ: So, the work we described before--if you are using household units as your measure of analysis, how does that play out when you are looking at, say, the growth in income from 1979 to 1989? So, there's two issues here. One is, you are saying that if you change from pre-tax to instead pre-tax/post-transfer, you get a different answer. How does that calculation we just went through--why is that relevant? Guest: Once you've done that, the simplest measure of how the typical American is doing is how the median person is doing. But how do you determine who the median American is? Russ: How do you line them up? Guest: Right. An interesting point is that if you ask about what's been happening to the median household, you are going to get a very different answer than if you ask what's been happening to the median person in the United States. Russ: Using household income. Guest: Right. It turns out that the median person has higher income than the median household does. Which seems like a Lake Wobegon statement that all of our kids are higher than average. But it's perfectly consistent to say that, because the majority of people live in households that are above the median. Why is that the case? It's because households with more income also have more people. This is why these seeming footnotes really matter. It turns out that if you just look at income last year, median income is about $50,000; but the median person in those households has income of about $60,000. Russ: So, I'm confused by that. Let's try to walk through that with maybe a stylized example. Because when you are saying the median person, are you saying using the calculation with the square root you just talked about? Guest: No, not even doing that. Russ: You are just taking all earners and lining them up. Is that correct? Guest: All persons. Yes. Russ: So, explain to me again why in that situation--maybe we could take an example of three households. That might be too hard to do. If I have three households were all six people are working--I'm not sure how to construct the example. Guest: This is the way to think about it: Take those three households. One household has $20,000 in it; the next has $40,000 in it; and the next one has $60,000 in it. Russ: Median household income is $40,000. Guest: Yes. But it turns out that the first household only has one person in it, the second has two people in it, and the third has three people in it. Russ: So the median is in the highest household. But that won't work--because the $60,000 has 3 people in it, but they are only earning $20,000. Guest: Yeah, but the household income of the median person is $60,000. Russ: Okay. I'm with you there. Now I understand. Great. Correct. If I went to you and you are in that large household and I said: What's your household's income? You'd say $60,000. Guest: Exactly. Russ: So, if I survey individuals and rank them, I am going to get a $60,000. So, I agree with you. So, richer families tend to have more income regardless of what the source of the causation is, there's a correlation there. Some of it is because they have more people working; but some of it is just for other reasons. Guest: I think the only point is you have to really think through what it is that you are measuring, why it is that you are measuring it; because all these things that seem to sound the same turn out to lead to quite different measures. Now, it's not because there's mathematical errors here. It's that all of these measures of social success can lead to different levels and trends in what's happening in the United States. And it's why--I start this story out by talking about the 1970s. I talked about the 1970s and 1980s; I am old enough to have lived through the 1980s and in the 1990s when people were saying that the rich were getting richer, the poor were getting poorer, and the middle class was shrinking, that just didn't seem right. Because I looked at the United States and I thought we were much better off over the 1980s than the way these numbers were coming up. And it's what got me first into recognizing how the little details can really matter in terms of the perceptions of what's going on. Russ: Yeah, I have the same issue. It doesn't pass the sniff test with me. Of course, it's dangerous to think that way. It could be you just swim in the wrong circles, or very special circles. I live in the Washington, D.C. area and the economy looks pretty healthy; I shouldn't--that's not a very good indicator of how the economy is doing because Washington is very distinctive. But when people tell me that we have made no progress on average since 1970, 1975, 1974, 1979--I was alive then. I remember what cars looked like, what stores looked like; and the average person has a much richer material life now than they did 30, 40 years ago; and it's absurd. It seems absurd. Maybe I'm wrong. So, that's why you have to look at the data. But as you point out, it's not simple to look at the data. You have to make all kinds of decisions and assumptions; and your biases and ideology of course color what your assumptions are. |
| 27:47 | Guest: I think that's right. So, there are two issues. One are the things you just talked about. I think that's why it's very important from a political perspective to objectively look at these things. But I'll tell you the truth--and the truth is, and this is just the way we academics are, and I'll admit it: the puzzle to me is how can you get such apparently wildly different visions with the same data? And to bring us up to the paper we talk about today, it comes from a quest on our part to figure out two extraordinarily smart people--Piketty and Saez--could get such wildly different perspectives on what's happening in the United States using their data than I could using these traditional measures that I've been looking at for the last 30 years. So, I was getting some results that were much different from the vision you get looking at Piketty and Saez. And I knew these guys were really smart. So, either there was something wrong with the data they were using versus the data that we were using; or they were using different assumptions in the way they were measuring this data than we were; or one or the other of us had made mathematical errors. And it took us 4 years to actually figure this out. But the answer is: It wasn't the data. It wasn't that we were making mathematical errors. It was that we were fundamentally asking different questions, but using almost the same words to describe our findings. Russ: So, tell us what you did. Of course, now we are going to be looking at a longer span of time--1979 to when? Guest: We actually with them go back all the way to 1967 to 2007, and that paper is actually coming out in the Review of Economics and Statistics, this next issue; but it's also--that paper explains why we are doing different things. And then this paper that just came out in National Tax Journal argues that we think our way of thinking about it is better than their way of thinking about it. Russ: Let's do the National Tax Journal first, if you think that's appropriate, because you have a nice little table in there that shows dramatically how making slightly different assumptions that seem innocuous or relatively unimportant, like tax unit versus household unit, you get radically different answers for what's happened for, in this case, the last 20 or 30 years. Guest: Right. So, let's take the simplest measure of economic well-being of the typical American, and let's look at what's happened to median income between 1979 and 2007. Russ: This is pre-tax. Guest: Yes. So, first of all, this is using the CPS, which is not what Piketty and Saez use. They use IRS data. And this is a very important point. I started the conversation by saying economists are limited by the data they have available. Our questions really can only be answered in terms of the units we are talking about, beginning in 1967, in a consistent way, because that's when the CPS first started. Piketty and Saez were asking questions about: What's been happening to income back through the entire 20th century, and they used as their unit of analysis Internal Revenue Service (IRS) data on tax units. So, they looked at tax units; we looked at households. So, what we did, and this is what's kind of cool, is we took their assumptions about sharing unit--that is, the tax unit, which you can get in the CPS data. They can't get the household in the IRS data. But because the tax unit is a subset of a household, we can replicate their assumptions. So, we used a tax unit--that is, the units that people include in their tax filing units. We look at pre-tax, pre-transfer income, which is called market income. That's the return to land, labor, and capital. Russ: And that's really all they have as their income measure, because they have tax data. That's right. They don't have transfers, they don't have in-kind benefits, they don't have compensation that's non-monetary. Guest: They are asking a perfectly valid question: What's been happening to the median market income of tax units? That is, when you add up the returns to land, labor, and capital, what's been happening to median income? They don't adjust for tax-unit size, and as you said, they don't adjust for government transfers. They are just looking at the returns to land, labor, and capital. And when we use their measure in the CPS data from 1979-2007, we get a pretty discouraging result. The median income of these tax units has only increased by 3.2% over the entire 30 years. Russ: Which is close to zero. That's not annual; that's not per year. That's over the total--that's flat. Guest: That's stagnation. And that's what they call it: the middle class has stagnated over the last 30 years. And that statement is correct. But what does that statement mean, and how do you put it in the context of how the typical American has done in the last 30 years? Is it true that we haven't increased at all in the last 30 years in our ability to consume things? Russ: Or that if we have, it's all gone to the top 5% or 1% or tenth of 1%. The pie is clearly bigger. There's two issues here. One is one's absolute command over goods and services. And then you can discuss one's relative command and where you are in the distribution, or a group's command, a certain segment of the population--the middle fifth, say. But this is saying in absolute terms, if you just look at income and capital income--earnings plus dividends plus capital gains--correct? Guest: Yes. Russ: Then it's flat between 1979 and 2007. That's corrected for inflation. Once you correct for inflation, no gains to the median household. Median tax unit--excuse me. Guest: That's correct. Now, I want to be careful here: the CPS does not collect capital gains. So capital gains are not in here. And actually in the original paper that Piketty and Saez did, where all this stuff came from, they also didn't use capital gains in their principal measures. So that's a problem. Russ: But for the median it's not that big. It's pretty small. |
| 35:32 | Russ: Okay, so carry on. But then you made a different assumption, so you got a different result. Guest: Okay; so let's--and this is the fun part of the puzzle. Let's no longer take the tax unit as the unit of analysis. Now let's look at the household. And you go from a gain of 3.2% over that period to a gain of 15.2%. And that happens because the number of tax units per household has been rising over that 30-year period. There are a lot more people living together and sharing everything except a marriage certificate. Russ: Fascinating. Let's stop there for a second. So, that, just changing the unit of analysis from tax unit to a household unit, taking account of the fact that there are people who are living together who are called separate for tax purposes but actually can have some economies of scale--that changes the growth rate by a multiple of almost 5. Guest: It's actually different even than that. We haven't even brought in economies of scale yet. In the sense of dividing. Russ: Oh, it's just the unit. Guest: Strictly going from a tax unit to a household unit changes that because in these tax units, you have people--their assumptions about tax units are: they care very much about what the top 1% or 5% of tax units are. But they also want to include not only the people who pay taxes, but the people who don't file taxes. So they have to make some assumptions about who is in these tax units. And their assumptions are that anyone over the age of 20 who is not married is an individual tax unit. So, they get a lot more individuals, low numbers than you would when you recognize that most of these people who are above the age of 20 but aren't married are connected in some way with other individuals in households. So you get fewer. Russ: So, they've got a bigger left-hand tail. They get a lot more units, total. And since they are below the median, typically, they are pulling the median down. Guest: That's correct. Now, I want to be clear on this. They can't do this. That's why they didn't. The advantage of what we can do is we can actually get a median because we know what the distribution is below these top units. They don't know that. They make assumptions about that. Piketty and Saez can't tell you what the median tax unit's pre-tax, pre-transfer income is: they just don't have that data. We do because we have the CPS and we can put it into smaller tax units. And the important thing about the recent paper that we've done is we can virtually get their results using the CPS. That's what is really cool about that paper. Russ: Just to make that clearer to people who find that confusing: You are using a different data set than they are using. But you can replicate their results because you have a lot of the data that they are effectively using. They have more of it; yours is just a sample. But it doesn't matter. Because you can mimic what they've done by taking the same assumptions in your data. So that's a way to check the sensitivity of the results to the assumptions. If you start with their assumptions with your data, you get a similar result even though you are not using the exact same data set. Guest: That's exactly right. And here is the really important intellectual point here: No one has been able to do this before. So, people have argued that you can't actually use the CPS to look at the tails of the distribution, and therefore, Piketty and Saez, even though their data has problems, those problems are less than the problems in the CPS. The thing that took us 4 years to do was to show people: No, that's not true. We can actually get very close to the levels and trends that they get in the IRS data using the CPS data. That's what took a long time. Once that is established. Once you grant me that the CPS can get approximately what the IRS is getting, then you can write a National Tax Journal article that can actually show how sensitive the assumptions that they are forced to take, because of the limits of their data, matter if you go beyond what it is that they are asking. That's what this paper is I think interesting about, and is so profound--this one table, as we go through it, is, I think, so shocking that you can get such dramatically different numbers for a simple concept, like median income. So, we've gone from tax units, where there's an increase between 1979 and 2007 of 3.2%, simply from going from the tax unit to the household unit, we get 15.2%. |
| 41:06 | Russ: Almost a five-fold increase. But still, 15.2% over that length of time--it's 28 years--it's not great. It's like a half a percentage point a year. It's fair; it's pretty good. It's not stagnant. But it's not great. Guest: So now, when you go to household, size-adjusted, pre-tax, post-transfer income--so now we are going to the usual measure that I've been using, the Census Bureau's measure: pre-tax, post-transfer income and the unit is the household; and we are adjusting by the number of people in the household to the 0.5. You go from 15.2% to 23.6%. So, now, instead of 3.2% for a tax unit, pre-tax, pre-transfer, you now have 23.6% for a household sharing unit, adjusted by the number of people in it; and we are including government transfers. Russ: Is it pre-tax, post-transfer? Is that correct? Guest: Yes, that's right. Russ: It's not post-tax, post-transfer. Guest: That's correct. We haven't looked at income taxes and sales taxes yet. Russ: Okay. Guest: Okay, then, what you need to recognize is: Not only does the government transfer money from high-income to low-income people through government transfers like Social Security benefits, disability benefits, Temporary Assistance for Needy Families (TANF) benefits, and those sorts of things, but it also does it through a progressive income tax. So, when we then use the National Bureau of Economic Research's tax simulator, with the CPS data, we can simulate the amount of income taxes, state taxes, and payroll taxes that people pay. Russ: Because you know a lot of their characteristics. So you are going to try to approximate what their income tax form would be filled out to yield, if you could see it. You can't see it, but you know a lot about it. You know how many kids they have, what kind of income; you don't know everything about them because you don't know their charitable deductions, all their medical deductions. Do you know the value of their house--do you have their mortgage? You might have that, right? Guest: We don't know about mortgage payments. Russ: So, it's a crude estimate. So, what do you find, then? Guest: So, then we go from 23.6%, which is the median income without taxes, to 29.3% growth. Now you might say: How can you get median income growing faster when you are subtracting things out? And the answer is that we've actually been paying, at the median anyway, a smaller share of our income into taxes; and that's why you get the growth is 29.3% rather than 23.6%. Russ: Okay. And then there's one last change. Guest: Yes, and then the last change is to talk about this growing share of earnings that comes in non-wage compensation, and also the fact that Medicare and Medicaid have been growing also. So, as an example of how important it is to think about in-kind transfers as well as in-cash transfers, and to think about the value of health insurance for employees as well as their wages, we are able to estimate the employer share of employer-provided health insurance to workers and the insurance value of Medicare and Medicaid to lower-income people who are getting those benefits. And when you do that, it goes from 29.3% to 36.7%. So, we are really talking about a difference from 3.2% in the way Piketty and Saez are thinking about things in tax units, to 36.7% if we simply include the value of health insurance as well as do the other things that we've talked about. Russ: It's a tenfold increase. So, I think there are two obvious thoughts here. One is: Boy, it sure makes a difference, what you can assume. And as you point out, there are different justifications for what you might look at. But if you are looking at economic wellbeing, it's hard to argue that that first measure is the right measure. You'd want to include things like transfers and health insurance. And you've left out a bunch of stuff, by the way. There are at least two things you've left out. One is: there's all kinds of other non-monetary compensation that isn't in the CPS that I suspect has been growing. Small things, but not zero. Dental benefits, vacation; I'm not sure it gets bonuses. Does it get non-regular bonuses, in the March CPS? Guest: In principle it does, but how well that does I don't know. Russ: But they've become more important over the last 30 years in compensation. The point is: one, as you said, it's a beautiful example of how assumptions matter and data analysis. There's "lies, damn lies, and statistics." It's kind of amazing that range, from 3.2% to 36.7%. But the other thing that some people will be asking, and this is the kind of thing I hear sometimes from students: So, which one is right? Because they can't both be right. One's really big and one's really small. So, which one's right? Because they want to write it down on the exam. When the exam comes and says, how is the median income person doing, you've got to put in an answer. And there's A. really well, B. pretty well, C. not so well, D. lousy. What's the right answer, Professor Burkhauser? Guest: Well, I think the right answer depends on what the right question is. So, if you are asking what's been happening to market income, I think there's no question that wage rates have become more unequal. But once you adjust for health insurance and other things, it's less unequal. And that in terms of real compensation, it's clearly rising. So, I think that even there, returns to work have been rising. The notion that we as a society are not doing as well as we were 30 years ago, I think by virtually any reasonable measure, is just false. And I think that's the main statement. The issue of distribution is a little harder. It's a little harder to argue. We have become somewhat more unequal. But even there, what we find, in the CPS data, is between 1992 and 1993, there was a change in the ability of the Census to capture exotic incomes of the top 1% of the income distribution. So, if people look at the CPS data and don't recognize that in 1993 we suddenly were able to better capture income, they will get the false notion that income inequality increased between 1992 and 1993. It didn't. It just was in our ability to capture that income. And we actually adjust for that here. But what it says is that when you get people telling you things that just don't seem to be consistent with reality, you really do need to look very carefully at the assumptions they are making. |
| 49:38 | Russ: So, what kind of reaction have you gotten from these two papers? Oh, by the way, we haven't talked about the Review of Economics and Statistics paper. Give me a punchline on that one. Guest: Sure. Well, the punch-line on that one is that when we sent that paper to the American Economic Review (AER)--we sent a paper to the AER that made the following statement: that income inequality in the United States rose substantially between 1967 and 1992, but hasn't increased very much since 1993. And we did that using the measure I just talked about--household size-adjusted pre-tax, post-transfer income. And we got back a referee report from the AER that said: This clearly can't be true because Piketty and Saez have found that income inequality has risen dramatically in the last 20 years. So, we were just minding our own business, using the traditional measures, and hadn't paid a lot of attention to Piketty and Saez in that literature. So, in order for us to publish, clearly, in traditional ways of measuring things, we had to demonstrate to referees that in fact you could get those two results. So, that's when we looked more carefully at what Piketty and Saez did. We then took--the other thing we had that was really cool was that the CPS public use data limits what you can know about top-income people, because in the top codes, each of the 23 types of income that people get--by top-coding I mean that they want to protect the confidentiality of people, so that if you have a lot of income and a particular source of income, they-- Russ: They truncate it. Guest: Right-- Russ: They just call it a million, even though it might be $87 million, or whatever is the truncation. Guest: That's right. So, good economists, recognizing that, stopped using the CPS to measure inequality, because the top parts of the distribution were non-systematically lopped off. And they started using 90-10 ratios, and those sorts of things: looking at persons in the 90th percentile and the 10th percentile. So, we actually gained access--and this is kind of a cool story in itself--to the internal CPS data. And were able to get around these top codes. And to see much more clearly what was going on. At least the top codes, in the interim, they were much more in the range of the top 99.2 or 99.3, rather than lower down, which is what the other ones were doing. And we observed one important thing. In 1995-1996, it appears in the public use data that income inequality gigantically increases. Well, it turns out the only reason it gigantically increased was instead of using top codes, in 1996 starting in the public use data, it began using cell-means of the top parts of the distribution. That is, rather than lopping things off at a million dollars and giving everyone a million dollars as the value, they looked at all the values above a million dollars, found out what the mean of those values was, and gave you that. Say, $2.5 million dollars. So people naively used the public data. Suddenly they saw all these $2.5 million millionaires, that were the $1.0 millionaires they had the previous year. So, all these kinds of things. Russ: And it always makes the NY Times, and I always look at those data and I want to say: They changed the definition or it's a coding error. The world doesn't change like that in a year. But that never stops them. And maybe I'm being unfair to the NYTimes. But I guarantee, I'm very confident that someone wrote an embarrassing article assuming those numbers were meaningful. Guest: Absolutely. You'll still see those pictures in the NY Times and even the Wall Street Journal, surprisingly enough. So anyway, we got around all that. We got the internal data. And that's when we found this kind of important point: that while income inequality has increased slightly since 1993, it hasn't increased by all that much. We couldn't get that published, because people said that was inconsistent with Piketty and Saez. Then that led us to see that Piketty and Saez were really measuring market income, not the total income. They were looking at tax units rather than the households. They weren't looking at government transfers. And we were able, then, to ask the question: What's been happening to the trends in the top 1% of tax units? How much of total tax market income do they hold? What about the top 5%? 10%? And we were able to show that for the 90-95th percentile, when we used the CPS data but their definitions, we got their trends dead on. For the 90th, 95th percentile. For the 95th-99th percentile, we got their trends dead on. It was only in the top 1% that we had some differences. The two biggest differences--and this is kind of cool--is that in 1992, 1993, when I told you that they changed the ability in the Census to get these exotic numbers, there is a big jump in our data that's clearly an error--not an error, it's a change in definition-ology in the CPS. We get a big uptick, where they are actually getting a downtick. So, their numbers are better there, and that's clearly wrong. And that said that we should lift all of our previous years up so that there is no break. But they also have a tick. The Reagan Administration worked with the Democrats in Congress in the 1980s to adjust the top taxes of the top personal income taxes, and in the 1986 reforms, for the first time made the highest personal income tax rate lower than the corporate rate. And when that happened, not surprisingly, there was a dramatic increase in the amount of market income recorded on people's taxes. Russ: Because they used to take it as distributed profits in partnerships and other businesses. Guest: Exactly right. Yeah. So, in their data, they have this big uptick in the share held by the top 1% that in our data there is no change. So, I would argue that that is the exact same problem. When you adjust for those two sorts of things, then the trends are actually pretty close. |
| 56:41 | Russ: Okay. So then you were then able to look at a different trend, when you include--I assume you are then going to do something similar to what you do in the National Tax Journal piece. Guest: Actually, we didn't even do that. All we did in that was show we could get their results. Russ: Oh, okay. Guest: But that's important because it says--it's important for them and important for us--we both verify that if you are asking the question that they are asking, they are correct, and we can verify it. And if we are asking the question we are asking, we are correct. And it's not inconsistent with what they are doing. But it goes back to this question. The ultimate question is: What's the question? To me, here's how Piketty and Saez can really be misused. There's a book out that uses the Piketty and Saez numbers and it's by a political scientist. And he's arguing that look, we are back to the Gilded Era in the United States; the share of income held by the top 1% is as great today as it was in the 1920s. Well, that is true of market income. But let's think about the 1920s and think about today. In the 1920s, there was no Social Security System. If you weren't earning market income, you were in bad shape. We do now have a Social Security System, and all the people over the age of 65 who have no market income are not starving in the streets. But that transfer income is not included in that comparison. So, to make comparisons with what's been going on with market income and imply something about the economic wellbeing of people in the United States, is to completely misuse Piketty and Saez's numbers. Russ: I just want to mention to listeners that Pitketty and Saez's work--we'll put links up to it--they make their data available online. If you are playing with it, you can see it. They are very transparent about their data. You can create your own spreadsheet or cut the numbers a little differently than they've done it. But as you point out, they are asking a different question. What I think the real issue is, is people don't care what question they are asking. They are going to take their answer and they are going to use it to answer their own question, which is: How are people doing? How is the country doing? I think to give them the benefit of the doubt--and as my listeners know I'm very skeptical about these theories that the average person is doing poorly, so I'm very amenable to your kind of improvement--what I would call an improvement. But let me play devil's advocate for a minute. If you look at, say, men's income or men's wages, it is remarkably flat. Now, you have to be careful because--and I'm talking about individual workers now--one of the mistakes that people make, it's a very subtle point, is that if you are taking an increasing share of your income in the form of health care insurance from your employer, the Consumer Price Index (CPI) that you are using then to deflate your market earnings is misleading. Because that CPI has been going up somewhat dramatically partly because health care costs are going up. But if you have health care insurance you are insulated from that somewhat. You are taking your income in the form of that insurance, but you can't then use the entire CPI to deflate your market income. It's not the correct measure because you don't have to use that market income to buy health care. So, that's a subtle point; I don't know if people listening can grasp it; maybe I spoke too quickly there. But the point is that on the surface the stagnation of market income is not encouraging. It's not that comforting to be told: Oh, well, don't worry, you are getting it in the form of health care insurance. Well, yeah, but what I have left over isn't changing much. Part of what you have left over is actually a little bigger than it appears because of the way the CPI is calculated and this fact that you are getting this in-kind transfer. But still, it's not the cheeriest conclusion to say, well, the middle class is doing better because if you include transfers it turns out it's pretty good. That is slightly alarming. Guest: Well, I think that the more important point is that what is driving these kinds of changes and what can you do about it. I think that's where all of this goes to. These are social success indicators that hopefully give us some idea of what's happening. So, I think that's what happening is that the returns in the market place are now more unequal, that there is a great premium on education and training. And you want to ask yourself: Why is that happening? The sort of simple-minded answer to that, which some people would argue, is because the greedy top 1% are getting all this in profits and exploiting the working man. I just think it's a much more complex issue than that, and we need to get beyond that and think about what's actually happening here. And what I think most economists are arguing is we're having technological change which is skill-biased, and if you have skills you actually are doing pretty well. The return on education has been rising even though the number of people going to school and getting higher educations has been rising. But if you don't have skills today you are really competing with the unskilled labor around the world. And what are we going to do about that? Again, it depends on how you think markets work. I think markets work--markets are like mirrors; they show you the way the real world is working. And you can either deny that or you can adjust your practices so you are able to do better in that world. So what it says is, if we are going to be in a world of competition, a world where we allow the free movement of goods across borders--and if you have international trade that means we have to be skillful. And if we have workers, we have to invest in education and we have to recognize that if you don't do that, our return is not going to be that great. So, my view is, for people who don't have those skills, have low income, are poor, I'm absolutely in favor of transferring income to those folks; but I want to do it in a way that doesn't kill the golden goose that allows us to be productive--and that is by allowing people to get a fair return on their investments. So, this class warfare discussion that seems to come from this rather arcane discussion that we are having, I think is very serious. And what I am suggesting is that the numbers that I'm showing is that yes, things are more unequal in the marketplace, but they've been offset substantially by 50 or 60 years' of programs that have worried about redistribution. But worried about it with the recognition that you can't do simple things like raise the marginal tax rates on the top 1% and not expect that to have some impact on productivity and growth. |
| 1:05:02 | Russ: Well, I guess, two thoughts, and I'd like to get your reaction. One is, if you do raise taxes on the top 1%, you are going to change the pre-tax income of the top 1%, not just the post-tax income. So, there is sort of a presumption of all parties that we'll just take this money and give it to these folks and then pre-tax income will be more equal. Well, it won't be, because you are playing with the returns to the labor market, and supply and demand are going to push the pre-tax compensation of those folks up partially to compensate for their lost income in the form of taxes. So, in fact, what you are going to do is often make what appears to be the gap between rich and poor actually bigger. If you tax the top 1%, or the top 25%, or whatever it is, you are actually going to raise their pre-tax income through economic forces, market forces; and it's going to look like you've made it worse. Your post-tax is going to be closer together, but your pre-tax is going to be bigger. The second thought I have is that that 50 and 60 years you are talking about of transfer income, and I thought you were going to allude to this earlier, that affects your market income. Obviously our ability and generosity through the political system to make it easier for people to get by with less work makes it easier for people to work less. And so you can't look at the independently. You can debate how big those effects are, the disincentives to work from transfer programs and implicit high marginal tax rates of those programs. But you can't just look at the market system's distribution of wages based on skill independently of that transfer system, because it affects it. Guest: Well, of course that's exactly right. And that comes to the major problem of all evaluators of policy--what's the counterfactual? What would things really have looked like if you hadn't done the policy? And that goes back to this 1929 comparison I was talking about. In 1929, people over the age of 65 had far more market income than they do today. That's not because they can't work. It's because they don't have to work, because we have these transfer systems that allow people to not work after the age of 62. Russ: And if they do work, they lose some of those benefits, so they are discouraged from working. It's not just that they are comfortable. Guest: That's right. And actually a wonderful example of that is the one group whose market income has dramatically increased since 1996 are single mothers. Well, why has their market income increased so much? It's increased so much because we went from the Aid to Families with Dependent Children (AFDC) program, which completely discouraged work, to the new TANF program, which says that if you are a single mom, we will allow you to get benefits, but only for 5 years; but if you work, we will, even if it's in a low wage job, we'll provide you with earned income tax credits which will supplement and subsidize work rather than leisure. So, in that sense, if all you want to do is increase market income, I suppose you could stop providing all transfers, and then we'd have a lot more market income. But do we really want to do that? Russ: It's a question for another time. |