Friday, September 19, 2003

previous entry | main | next entry | TrackBack (4)

The great white whale of income inequality

My last Krugman post managed to generate a vigorous debate in the comments section while simultaneously confusing Donald Luskin. So it's worth focusing more closely on one of the points where Krugman's current analysis goes off the track -- his Ahab-like obsession with income inequality.

One of Krugman's biggest complaints about the trajectory of the American economy is the rise in income inequality. This rise was particularly acute during the Clinton era, and a constant refrain of his writing is that Bush's tax cuts will merely accelerate this trend, leading to more social frictions.

There are three big ways in which Krugman is wrong -- his emphasis on inequality in the first place, his failure to distinguish between the different causes for inequality, and his assumptions about the political effects of rising inequality.

1) Inequality is the wrong variable. I wrote a longish post over the summer about why the fears about income inequality are way overblown. To sum up -- a focus on inequality overlooks the high degree of income mobility in the United States, as well as the absolute improvements over time in the lives of the poorest Americans. For another refresher on this, go check out Todd Bass' more recent analysis on this point (link via Instapundit).

2) The sources of inequality matter. Take Krugman's concerns at face value. Are there moral reasons to oppose this rise in inequality? Anyone not completely blinded by ideology would at least acknowledge there are valid arguments against increasing inequality. However, a key question is the causes behind inequality. If the reason is increased social stratification due to the advantages accrued by inherited wealth, then I'm pretty sympathetic, since such stratification stifles growth. If the reason is increased opportunities for gain via entrepreneurial activity, then I'm pretty unsympathetic, because entrepreneurial activity promotes growth.

In Saving Capitalism from the Capitalists (p. 92), Raghuram Rajan and Luigi Zingales make an important point about the changing origins of American wealth:

One statistic best sums up the changes that have taken place: in 1929, 70 percent of the income of the top .01 percent of income earners in the United States came from holding of capital -- income such as dividends, interest, and rents. The rich were truly the idle rich. In 1998, wages and entrepreneurial income made up 80 percent of the income of the top .01 percent of income earners in the United States, and only 20 percent came from capital. Seen another way, in the 1890s the richest 10 percent of the population worked fewer hours than the poorest 10 percent. Today, the reverse is true. The idle rich have become the working rich!

Instead of an aristocracy of the merely rich, we are moving to an aristocracy of the capable and the rich.

Americans will not begrudge the rich getting richer if it's by dint of effort. [Krugman would respond by pointing to the astronomical rise in CEO pay--ed. No doubt, there are examples of malfeasance in matters of corporate governance. Suggesting a systemic problem, however, is a bit of an exaggeration, given the increase in asset prices of U.S. firms over the past twenty years. It's telling that Rajan and Zingales, who are sensitive to the issue of income distribution, are far more afraid of overreegulation in response to Enron-like episodes than underregulation]

For more on this, go read Thomas Piketty and Emmanuel Saez's NBER paper, "Income Inequality in the United States, 1913-1998" (updated in 2000).

3) Rising inequality does not lead to a breakdown in social cohesion. This is Krugman's core concern -- that inequality will lead to political and social instability. To repeat what he told Kevin Drum:

Is this the same country that we had in 1970? I think we have a much more polarized political system, a much more polarized social climate.

Krugman's reference to 1970 is interesting, since income inequality was much lower in 1970, the peak of the Great Society programs.

Despite the reduced level of inequality, society was more polarized back then. Anyone who believes that the country currently has a more socially polarizing climate now than in 1970 is, well, either lying or lost their grip on reality. Does Krugman really think that the debates about Iraq or affirmative action today even approximate the division and discord that Vietnam, Kent State or school busing generated thirty years ago?

Economic inequality has a far less significant effect on social instability relative to other factors -- the rate of absolute poverty, the method of raising armed forces, and the rate of economic growth and labor productivity. Krugman needs to worry about it less.

posted by Dan on 09.19.03 at 11:48 PM


I've worked in the corporate world for a long time and most of these executives are grossly overpaid for what they do. I can't remember how many times my colleagues and I have said that the company would be more productive if they paid the executives to stay home. Why should I support tax cuts for the wealthy? So that the people that have all the money can have even more money? Why not better services for all the people that have lost their jobs and their health insurance?

posted by: Scott K on 09.19.03 at 11:48 PM [permalink]

I agree that income inequality is the wrong variable when considering legislative measures because that may compromise liberty, as happened in the Marxian experiments.

However, income inequality is as troubling to a human (who has not masked their humanity) as perceived errors in logic. We are not only logic processors but we are also empaths (even Adam Smith based his analysis on that). The latter aspect is what most logical analysis avoids but democratic politics keeps dragging back.

The solution, in my opinion, is to make income inequality a prime subject in social science, while holding the logic of competition and liberty in place. Yes, failure to succeed for some people is necessary if they choose unproductive things to do. It is part of the social and individual learning process. But beyond that, what if it were possible to create institutions that taught people what they might do in order to be successful economically. My proposal is a Fannie Mae-type institution for adult retraining loans. So when farming or steel working becomes obsolete, loan officers will seek people out to fund their retraining and direct them toward schools that are good at it.

I agree that as long as the poorest do better than before, there is no evil going on. But what if we can do even better?


Karun Philip
Author: Zen and the Art of Funk Capitalism

posted by: Karun Philip on 09.19.03 at 11:48 PM [permalink]

Scott, if executives are such an albatross on company effectiveness, there's a real opportunity for somebody like you who really knows it all. Just quit your job and start a new company with those equally brilliant colleagues. If your company is as unproductive as you claim, it should be child's play for someone of your obvious genius to compete with it and kick their fat, leather-swaddled asses.

Or, check your premises.

Or, keep whining.

posted by: Don on 09.19.03 at 11:48 PM [permalink]

I share the setiments expressed in this post. Specifically, we should pay more attention to income mobility and that Krugman is fixated on this issue. However, he is a columnist entitled to a pointed world view, not a politician who needs a balenced one. I also feel that some key aspects of the story that Krugman is trying to tell are missing here.

The mobility story is inspiring, but limited. There is a huge amount of income mobility attributed to getting older and more experienced. My current situation is that I will likely earn less than $10,000 this year. However, I think I was in the top quintile of earners the previous year, and expect to return to that level in the next several years. The snapshot data is only a part of the story. More significant is the number of people who are entrenched in the lower income brackets with limited skills and ability to earn more. If you don't know what I am talking about read the New Yorker piece from about a month ago about the program to encourage marriage in a poor community in Oklahoma. It is those low-income people that are the issue, not me, but there is no statistical differentiation between us in most quoted data.

Second is that Krugman is not concerned with the top 1%, but the top .001%. It is not how those people acheived that position, but that they are able to exert an entirely disproportionate influence on our nation's policies. It is the increasing ability for these people to not contribute any part of their income or wealth to maintaining the public goods that make our nation that is the concern. To understand this point check out Warren Buffet's op-ed in the Washington Post a couple of months ago coming out against the Bush tax cut program. I personally am willing to accept that some people are getting absurdly wealthy, but it does bother me that at a time when economic trends are making this more common we are simultaneously passing policies that allow the super-rich to do less to support society.

Finally, the point that relating rising inequality and social cohesion. The point is not that reducing income inequality will alleviate all causes of social polarization and fragmentation. There are other potential causes of social discord that also should be managed and addressed, but income inequality is one of them. We are doing a good job paying attention to some of the other factors. To see the effect that extreme income inequality can have on society see the situation of Brazil, where there has been a tremendous rise in gated communities. Here I want to ask the question if those gates do more to imprison their residents than to keep people out. Income inequality is leading people to sacrafice their freedom of movement.

posted by: Rich on 09.19.03 at 11:48 PM [permalink]

I agree that a simplistic focus on inequality doesn't cut it. However, income inequality has historically followed trends in political polarization (and immigration) closely and there are important policy links. Howard Rosenthal has written an interesting and balanced paper evaluating the historical evidence:

posted by: Erik on 09.19.03 at 11:48 PM [permalink]

Krugman's main point is not that rising income inequality causes social instability. His point is that this level of inequality is unnecessary for maintaining a vibrant economy, and that the vast majority of the population would be better off if the wealthy paid more taxes not less taxes ( Krugman also includes the caveat that raising taxes too high on the wealthy would be counterproductive ). Why not take on Krugman's real argument instead of this straw man?

posted by: Michael Jones on 09.19.03 at 11:48 PM [permalink]

Since you mention it , here is a link to Krugman's real argument, at least as he presented it in the NY Times last fall. I suspect people will find Dr. D to be closer to addressing P. Krugman's real point than the previous commenter, but see for yourself.

For those with patience and a high pain threshold, here is a link to a topic-by-topic comparison of Krugman's article with the Pikkety/Saez paper to which he refers (and which is linked above).

Krugman does not take any time to explain why the period of the '50's to the '70's should be viewed as some sort of Golden Era in income inequality. Inequality was higher in the '20's than it is today. Why did inequality collapse? Krugman mentions the New Deal, but not, perhaps surprisingly, the Great Depression.

Krugman also does not dwell on the fact that in his anomalous Golden Era, our natural foreign competitiors were devastated by WWII and the US did not encourage immigration. A great era for US union workers, but was it sustainable?

If CEO pay was low in the '70's, that is probably the only good thing that can be said about US corporations of that era. Does anyone else remember Chrysler's bankruptcy, Ford's near-death and the rolling rubbish Detroit laughingly promoted as "automobiles"? The answer to everything was protection from foreign competition. Great.

Finally, Krugman makes much of US income inequality statistics (and other measures) in comparison with Europe. Since we accept (and absorb) immigrants, unlike our European counterparts, I wrote:

"The point - allowing lots of poor, unskilled, poorly educated people into [our] country is, I believe, a great thing for the people in question. I still believe in "Give me your tired, your poor, your huddled masses yearning to breathe free", as I am sure you do. However, it plays hell on statistics such as literacy, lowest quintile income, longevity, and health-related numbers. Far better, for presenting better such statistics, to manage your country like an exclusive, hereditary country club to which poor foreigners need not apply. For an example of such a country, how about Sweden, which you mention in your article? I understand their population is about 5% non-Scandanavian immigrants, and they find it to be a strain."

Oh, I was on a roll!

Anyway, stealing a theme from Jane Galt, and this is on the subject of "tax the rich", I also quipped that "if you advocated policies that pointed the guns at the identified target, you would be more convincing."

For an example torn from today's headlines (OK, website), here is Gen. Clark:

I would like to give more tax relief to people who are making, let's say, $100,000 a year and less, to working people with families who have responsibilities, and give less relief to people who are making, let's say, $300,000 and above because when you don't make as much money, you need to spend it on the necessities.

Uhh, people? $300,000 is a pretty good income, but it is not the "super-rich" Paul Krugman is screaming about.

Now, I think income mobility is the key to the story - it is not "the rich versus the rest of us" if your kids or grandkids (or you, quit reading blogs and get a job!) can switch teams. The state lotteries promote income inequality, but everyone plays. (Regressive tax and gambling opponents excepted).

That said, I worry about schools - access to a quality primary education is becoming tightly associated with income, and that is not good. Krugman ignores this, as he ignores the stranglehold the NEA has on his party of preference.

And (I am finishing up) I cannot take seriously the notion that we are more polarized now than in 1970. However, three things strike me as different and important:

- post Watergate campaign finance reform makes broad fundraising important and difficult. It is easier to get people to write checks opposing THE WORST PERSON EVER than it is to raise money in opposition to a candidate you disagree with. Hence, an increase in volume (and demonization) from Washington.

- politics as entertainment. Rush Limbaugh makes $20 million a year screaming about stupid Democrats; Al Franken makes a nice living doing the same on the other side. This makes it a bit harder to calmly discuss the issues. Bob Hope, for comparison, did political humor that was famously neutral at his peak.

- end of Cold War ended our faux unity. We are free to express our real opinion of the stupidity of the other side.

OK, that was my short summary. Ooops.

posted by: Tom Maguire on 09.19.03 at 11:48 PM [permalink]

Hey, how about a slightly off-subject, shallow musing? What if one of the big probs with extreme income inequality is that it creates all-too-attractive opportunities for demogogues (let alone newspaper columnists)? Not that that's a reason for passing laws or jiggering with tax codes. But maybe "the rich" might think about this a bit and do a little p-r for themselves -- show off less, be a little more humble, do good works, express gratitude to the system and thanks to everyone who pitched in -- do voluntarily at least some of what others might want to pass laws to force them to do. Might help take the edge off some of the free-floating resentment, and reduce the audience for people like Krugman.

posted by: Michael Blowhard on 09.19.03 at 11:48 PM [permalink]

Re, PR by the rich:

Emily Post Was Taking Notes

(From the NY Times)

Bill Gates, the chairman of Microsoft, announced $51.2 million in grants yesterday to create 67 small, theme-based public high schools in New York City.

...At a ceremony yesterday, Mr. Gates awkwardly shared the stage with Schools Chancellor Joel I. Klein, a former Justice Department official who led the government's aggressive antitrust prosecution of Microsoft.

Asked about their previous dealings, Mr. Gates nodded and barely cracked a smile. "I'm glad to be working on the same team," he said. Behind him, Mr. Klein flashed a sheepish grin.

The Gates gift is the largest ever to benefit the city's schools.

posted by: on 09.19.03 at 11:48 PM [permalink]


You write that "Anyone who believes that the country currently has a more socially polarizing climate now than in 1970 is, well, either lying or lost their grip on reality. ".


PK made clear in is article that he was basing his comments on research by several political scientists (Nolan McCarty, Howard Rosenthal, and Keith Poole) on this issue. PK, based on their work, says that party division is the most polarized since the 1920s. Do you disagree? Surely you are qualified to comment on the work by those political scientists. But simply dismissing them as liars or delusional is not a valid criticism.

Now Tom isn't a political scientist so he can get away with "I cannot take seriously the notion that we are more polarized now than in 1970". But you are. I'd be curious to read why you disagre with the research done on this area.

posted by: GT on 09.19.03 at 11:48 PM [permalink]

A reason why income inequality might concern even conservatives comes out of the literature on comparative political economy. In at least some circumstances, high inequality encourages higher levels of redistribution, i.e. taxation, as distance between the median voter's income and the average income increases (if I remember the literature correctly, which possibly I do not).

Carles Boix, a member of Professor Drezner's department, discusses this in his book 'Democracy and Redistribution'. So while in general growing income inequality may not be as big a deal as Krugman makes it out to be, there certainly is the possibility that it can have significant political consequences worthy of attention by both liberals and conservatives.

posted by: ps on 09.19.03 at 11:48 PM [permalink]

If I may save Dr. D for the heavy lifting, and take a stab at the question posed by GT: I strongly suspect, since he excerpted it a few days ago on his website, and repeated it in the current post, and mentioned Kevin Drum, that Dr. D is inspired by this punchline from the CalPundit interview with P Krugman:

Krugman on America's social cohesion: "Is this the same country that we had in 1970? I think we have a much more polarized political system, a much more polarized social climate. We certainly aren't the country of Franklin Roosevelt, and we're probably not the country of Richard Nixon either, so I think we have to take seriously the possibility that things won't work out this time."

I will forever paraphrase that as "we aren't the same country as when Nixon was President - and that is a bad thing". And I suspect I will always laugh when I say it.

In the Cal interview, Krugman goes on to describe gated communities in Brazil, rather than partisan voting in Congress. So I think that the point raised by GT is a bit of a non sequiteur.

But yes, I have glanced through the articles showing Congressional votes are more party-line these days. The southern Dems are now southern Reps, some Rockefeller Reps are now Dems, and so what? That may be what Krugman mentions in his article that I linked to, but it is not in the post Dr. Drezner linked to.

posted by: Tom Maguire on 09.19.03 at 11:48 PM [permalink]

A non sequitur is a conclusion that does not follow from the premises. Not clear how that applies to what I wrote.

In any case PK wrote a long article on income inequality with a whole section on its political implications. If you are a Political Science Phd, teaching at the University of Chicago and not, say, one of the many PK online stalkers that love to play gotcha word games, than I expect you will address the core arguments made by PK if you are going to criticize them, not base it on a a couple of comments part of a 20 minute hurried interview in turn part of a country wide book tour.

That's why I wrote your reaction is predictable Tom. But I was and am hoping that Dan will address the points made by PK and the politicsal science research they are based on.

This is not religion. Reasonings like "I don't believe X to be true" are not very useful. Was the research misquoted by PK? Is it wrong? is it irrelevant to the point at hand? All those are valid criticisms.

Dismissing it as either lying or delusion is, dare I say it?, Luskinism.

posted by: GT on 09.19.03 at 11:48 PM [permalink]

"...a focus on inequality overlooks the high degree of income mobility in the United States"

I'm sympathetic to the general argument of this
post, but is there really a "high degree of income mobility" in the US, comparatively speaking? I thought several recent studies had found that US income mobility was relatively mediocre when compared to European countries. Here's an example:

If there are contradicting studies I'd be interested to know of them.

posted by: rd on 09.19.03 at 11:48 PM [permalink]

Whatever else I may think of the idea of arranging things so that Warren Buffett doesn't have any more influence in Washington than I do, it seems to me that it would be a bit of a hollow triumph unless at the same time we arranged things so that Paul Krugman doesn't have any more influence in Washington than I do. I've a shrewd hunch, somehow, that that's not the plan. If Prof. K nevertheless wants to appeal to the ideal of equal influence, he could offer an earnest of his good intent by giving up his op-ed column to make room for more letters to the editor.

posted by: Paul Zrimsek on 09.19.03 at 11:48 PM [permalink]


There was less party-line voting in the 60's and 70's because of the issue of race. While members of the same party usually voted the same way on non-racial issues, they were likely to have voted differently on racial issues. It was not because back then there was less division. There was more division, but because the way the two sides were divided on racial issues was different from the way the two sides divided on non-race issue, this produced the statistical appearance of less disagreement. So ironically Congress now seems more divided (statistically speaking) even though in reality it is less.

posted by: Hei Lun Chan on 09.19.03 at 11:48 PM [permalink]


What evidence do you have for that? That's not what I recall reading.

posted by: GT on 09.19.03 at 11:48 PM [permalink]


The sacrifice of social bodies adopting a restrained moral distance when macro trends impinge on issues of morality and inequality is their Humanity. Life in the real world often does require hard decisions and commensurate action for the greater good. But this does not preclude sympathy for the plight of those affected. It is in exactly those circumstances where hard social choices must be made that one must be most conscious of a sympathetic awareness.

Perhaps “sympathetic” was a bad stylistic choice and not a moral position?

Agree that Americans on the whole do not begrudge anyone’s benefiting from invested labor. I propose that the disgruntlement with the current excess in C-level compensation is exactly from this perspective – compensation received is way out of whack with the effort put in to benefiting the corporation. Give a Dude his due, but eligibility for that “due” should be judged by the shared benefit to the entity and its members in Dude-ly charge. Covetousness without consideration of other’s welfare is bald-faced greed. And that just ain’t right.

I agree with you that society is less polarized today in terms of social consciousness, advantage and mores. Personally I wish it were more so. The trend that bothers me is the conglomeration of wealth at one pole without a clearly balancing opposite pole, coupled to the buying power of the wealthy end in terms of units of political and economic advantage. Self-interest dictates that a means with money will spend in its own preservation. When such preservation becomes overbearing as a whole on society, social change is affected. But only a polarized society can overcome the human inertia of resistance to change. It is exactly when the majority of society is not badly enough off that monopoly becomes more than a board game. I think that this is an historically unique socioeconomic mix that we have today and a major philosophical and practical challenge. We are a long way from the Globally-Wealthy Dominated Boogie-World of futuristic fiction, but like sympathy, freedom and equally require vigilant awareness and control. I think we have an interesting challenge coming in retaining open markets and basic rights in the face of an non-polarized society.

posted by: Jon on 09.19.03 at 11:48 PM [permalink]

Here's a link to the paper GT refers to in her comments. A few thoughts:

1) I'm not familiar with this part of the American politics literature, but from what I can see, the authors seem to assume that the ideological gap between the two parties remains constant over time. When talking about polarization, this is a whopper of an assumption. Casual empiricism suggests that the ideological differences between the two parties have narrowed over the past thirty years (they've both adopted more conservative policy portfolios, which I suspect is the source of Krugman's consternation). So, even if the parties are more ideologically homogeneous, if the gap between the parties is narrower, this fact becomes less important.

2) It's worth observing that Krugman's thesis about increased economic inequality leading to greater polarization does not hold up. From p. 7 of the linked paper:

The results show that almost all of the increase (in polarization) can be attributed to an increased effect of income on partisanship and changes in party allegiances of certain groups. Changes in the incomes of different groups and the widening income distribution do not play as large a role. (emphasis added)

So, part of what's going on is collapse of old-time Southern Democrats, and part of it is that increasingly, wealthier individuals do vote Republican (which, as a side note, strikes me as a contradiction of the Judis-Texieria thesis). The increase in income inequality is far less important a factor.

posted by: Dan on 09.19.03 at 11:48 PM [permalink]

Funny we all neglect the likliest source of income inequality, government spending. Income inequality shrank until the late 60s, when it began to increase, just at the time of the great society. Under Clinton, income inequality slowed its growth, according to the stats I reviewed, just as government was shrinking.

Government consumes 40% of the nation's income, and this puts pressure on private sector wages, causes mass immigration of poor immigrants, and favors outsourcing.

posted by: Matt Young on 09.19.03 at 11:48 PM [permalink]

Dr. D - you would make a more compelling case by linking to the updated, rather than preliminary version of the McCarty/Poole/Rosentahl paper. BTW, that is the very same paper that Krugman linked to at his website, so we are at least looking at the same thing.

Their updated abstract:

We find that partisanship has become more stratified by income. We argue that this trend is largely the consequence of polarization of the parties on economic issues and the development of a two-party system in the South. The trend is much less a reflection of increased economic inequality.

I.e,. rich southern Dems are now rich Southern Reps. And the relative prosperity of the South has risen, too.

posted by: Tom Maguire on 09.19.03 at 11:48 PM [permalink]

The Economist article reads as misleading in at least 3 or 4 places to me.

First, the poor Americans are better off than they used to be. The examples they give are all goods that technology has improved remarkbly in recent times. The cases would apply equally to subsuharan farmers as poor americans. Better would be to talk about access to such things as good food, a doctor (I would say health care, but there are technological improvements there also).

Are the poor better off than they used to be? Perhaps, but the standards used in the Economist would say every person on the planet is better off.

America is a remarkably mobile society? Two things in their evidence seem to deflate this conclusion somewhat. Firstly, their evidence is based on the bottom 20% moving up to (most likely) the second 20%. So what, the bottom 40% of people mix freely every 10 years or so. In other words, there isn't much trouble changing from poor to very poor and vice versa.

As some people have mentioned above, wealth is probably the better measure here anyway, and that provides the second piece of evidence that mobility is down. If income mobility is so high, why is there such a wealth concentration at the top (I mean, as a rough approximation, one is the derivative of the other)?

Combine the fact that wealth is the better measure, and we see that the bottom 20% is actually worse off than slaves in terms of net assets (because they are in net debt).

The US may spends the most on Education (having the most money, the US spends the most on Education, Health, Defense, Bureacracy, Guns, Drugs, etc.), but I suspect the best measure is as a %of GDP. I have the distinct feeling that the US would not be close to first on this measure.

The poor are better off than their peers in Britain and Australia, and almost as good as the Nordic countries? This sounds completely wrong to me, but not having seen Britain, I can only be sure of its falseness as regards the other places. Universal Health care is a great leveller, and there are almost no Australians/Scandinavians in net debt. I don't think I've seen a single homeless person in either place (I know they exist, but are very rare). In addition, the public schooling system in these countries means that plenty of graduate students at the top universities come from the bottom 20% of income (you could make a joke about this being true in the US too, because so many of the graduates are foreign, and come from poor countries).

Conclusion: Irrespective of the conclusions, it's disappointing to see the slanted reasoning that the Economist is using. My personal experience would be that the poor in America are considerably worse off than the poor in Australia and Scandinavia. The upper middle class and above are probably better off here, the rich certainly so.

posted by: Jason on 09.19.03 at 11:48 PM [permalink]

A brief comment on the methodology used to measure political polarization: the authors DO NOT assume that the ideological gap between the two parties remains constant over time. Rather, they use the fact that members of Congress are not replaced all at once to anchor a common ideological space. What they have found among others is that while congressmen generally do not change their positions much (with the exception of party-switchers), there has been a change, especially due to the replacement of moderate republicans by more conservative republicans. This (among others) has led to a wider gap between the parties. The measures Poole and Rosenthal have developed are now the standard in the American politics literature.

I should note that their findings reflect conflict on a mostly socio-economic liberal-conservative dimension. Much of the social conflict due to race in the 1960s was captured by a second dimension, which significance has declined. I linked to a paper by Howard Rosenthal in a comment above that explains how income inequality may be related to political polarization (by the way, Poole and Rosenthal are not liberal academics).

posted by: Erik on 09.19.03 at 11:48 PM [permalink]

Couple of things:

I notice that you're kind of assuming that if income inequality is due to differing labor wages, that somehow sweeps that whole thing under the rug ("inequality isn't due to the idle rich anymore, it's due to entrepeurship", to paraphrase). "Americans will not begrudge the rich getting richer if it's by dint of effort." They may not begrude them as *much*, but that doesn't make the point moot; it's not like Gates is working a million hours a week. Lots of people reasonable suspect that a large amount of it is luck.

On income mobility, apparently we have the same level of mobility as Scandanavia, as you yourself say here, Daniel.

The thing that chiefly confuses me about mobility discussions is that they're all measured in terms of year-to-year, life-of-one-person, instead of "income mobility compared to parents."

Theoretically, if we lived in a perfect society where income was determined entirely by hard work, you'd expect to see the income of children end up completely independent of their parent's income. That's not what actually happens; Google gives results in the range of .4 correlation.

posted by: Jason McCullough on 09.19.03 at 11:48 PM [permalink]

Jason, there is simply no question whatsoever that the poor are better off today, in material terms, than they were in the 1950's, even on the basic goods that you describe. In the 1950's and 1960's, hunger and malnutrition still stalked the poor, a significant percentage of them lacked not only electricity, but also running water, more poor people were crammed into any unit of housing, health care was not merely technologically, but qualitatively worse (when was the last time you saw anyone on an open ward outside of ICU or the emergency room?), obtaining sufficient clothing to keep warm was a major problem for a significant subset of the poor . . . on any measure you care to choose, today's poor are better off. *And* there are a lot fewer of them: in 1950, one in five Americans lived under the poverty line. One subset, the homeless, isn't better off, but the homeless are a result of mental patient deinstitutionalization and the housing code's ban on flophouses, not the structure of our economy.

posted by: Jane Galt on 09.19.03 at 11:48 PM [permalink]

Re: The Perfect World, where "income was determined entirely by hard work, [and] you'd expect to see the income of children end up completely independent of their parent's income."

Well, not really. Determinants of income seem to be include genetic endowment (IQ is hereditable). Also high income parents leave their kids more than money - they are more likely to raise them in more stable households, read to them, value education, and teach habits that are likely to lead to success.

Now, obviously wealth has direct benefits as well - they also have have access to better education, better health care, and leave their kids cash. But even in the perfect world, there would be a correlation from one generation to the next.

posted by: Tom Maguire on 09.19.03 at 11:48 PM [permalink]


I wasn't trying to claim the poor are worse off than they were in the 1950's, and I suspect you are right that even in the goods I would use as a measure, that the poor are better off than they were (although my father likes to say the throughout his lifespan the amount of hours work for a construction worker to afford a pint at the pub has remained constant).

My point was, that the article didn't choose to use those goods. Rather, it chose to use goods from which you can draw the conclusion that everyone, everywhere is better off than they were in the 50's. Perhaps true, but that would allow that America might have changed from Switzerland to Kenya in terms of living standards for the poor while still improving the lot of the poor.

As a result, it reads like spin. They picked some statistics that supported their view and quoted them, rather than saying "What is a good measure of income inequality?" and then applying the measure.

As for access to health care, I'm not a good judge for how much it's improved for the poor here. I do know that the access is better for the poor in social democracies.

BTW, what is the "poverty line" figure that people quote and how is it derived? It certainly can't be based on family net assets which is IMO one of the best measures of poverty.

posted by: Jason on 09.19.03 at 11:48 PM [permalink]

On the issue of party line voting in Congress. I haven't looked at this myself, so I don't know how true it is, but, assuming it is, I think I remember reading discussions of gerrymandering which suggested that it led to increased party-line voting. In any case, I would be more inclined to believe that partisan splits were a result of gerrymandering rather than a fundamental increase in the polarization of our society. ( And anyone who takes this as an opportunity to criticize the gerrymandering in Texas, fine, but be prepared to equally criticize the gerrymandering by the Democratic establishment in California.)

posted by: Bill on 09.19.03 at 11:48 PM [permalink]

some thoughts

The first comment by Tom Maguire is interesting and generates a lot of hypotheses worth further investigation.

yes, the composition of the income stream is important. data divide it between 'earned' (wages and salaries) and 'unearned' (dividends, interest, capital gains, retirement pay). the point that a greater proportion of the income of the upper levels is earned is important. would that there would be a way to measure the source of 'unearned' income. i think it matters a hell of a lot wheere that comes from. while one may despise Bill Gates for his monopolization of the desktop, one has to admit that the fact that his obscene income comes from dividends and capital gains from his youthful entrepeneurship and foresight makes it easier to take. same goes for Warren Buffet and his uncanny ability to pick stocks as a source of income. that is a lot different from being born into a wealthy land-owning family or an industrial dynasty.

the age structure of the population must also be considered. the 'golden age' of income distribution was a time when society was much younger than today. other things being equal, older people have higher incomes than younger people. what effect does that have on the Gini coefficient?

bottom line, though, is that much of this discussion is feel good stuff. sure we all hate the overpaid CEO and the idle rich; but can anyone come up with a tax system that hits them and only them, whilst preserving the incentives offered to entrepeneurs?

posted by: Jim Linnane on 09.19.03 at 11:48 PM [permalink]

1) How about a min. tax on top-line corp. revenue (say 1% -- effects similar to a sales tax, unless corp. income tax is higher). Should reduce tax avoidance.
2) An excessive exec pay surtax on corporations. From 20% up to 60% on the pay of execs/ partners making 2x (20%) more than the US pres. in the preceeding rule; 3x (30%), 4x (40%), 5x (50%), 6x(60%).
I think the (real) disincentives to work won't result in any statistically significant effect on corp. profitability; especially if expensed share options are excluded (but, if the options become worthless and are "revalued", such a gift would not be excluded).

A blog is not a place to discuss any rule detail.

posted by: Tom Grey on 09.19.03 at 11:48 PM [permalink]

Jason asks:

BTW, what is the "poverty line" figure that people quote and how is it derived? It certainly can't be based on family net assets which is IMO one of the best measures of poverty.

It's a government statistic, kept by HHS. In 1963, Mollie Orshansky, a statistician in the Social Security Administration, tried to estimate the minimum amount an impoverished family would need to maintain a decent standard of living.

She came up with the number after looking at a survey of 1955 consumption patterns, which found that lower income people spent about 1/3 of their income on food. She learned from another survey what a subsistence diet would cost, and multiplied the total by 3 to determine the poverty line.

The government has since updated this number only by changes in the Consumer Price Index (inflation).

The basis for the initial calculation, and technical updates, has generated considerable conflict and quite a number of academic papers. This is as you might expect; one purpose of the figure is to determine eligibility for government benefits. Another purpose is to serve as a kind of benchmark for levels of poverty. Both, obviously, generate considerable political heat.

To touch on just one point, one that particularly concerns me: the poverty threshold does not move upward as real income moves upward. Because it is indexed only to inflation, it states the level of purchasing power that seemed adequate to Orshansky in the 1960s. Over time, it is very likely that a smaller and smaller number of people will fall below the threshold (and that "poverty" will be reduced ) even if a smaller and smaller share of the national income goes to people at the bottom of the distribution. As a marker of victory it seems hollow to me.

Put it another way. If the "poverty line" slips from, say, 40% of median income to 20%, and the number of "poor" measured by that line declines, what does it tell us? To my mind, not much.

Now I know that the Pareto optimists out there will say that if most people's incomes do not budge, and those at the top move up smartly, there is no problem for public policy to address. Some are better off, and none are worse off, in absolute terms. But I, personally, don't feel that way. I would gladly suffer small decreases in growth if I could make substantial reductions in inequality.

posted by: on 09.19.03 at 11:48 PM [permalink]

Oops. Forgot to put my name on the above post.

For a fuller story on the poverty thresholds, see

It's pretty good.

posted by: TedL on 09.19.03 at 11:48 PM [permalink]

Thanks a lot Ted!

I'm not American, so my experience with CPI may not be valid here, but growing up, I noticed that the CPI was a relatively biased (low) estimator of the increase in living expenses, largely because it wouldn't recognize when a substitute product raised the cost of living.

Just a quick thought as to poverty calculation - here's an idea. If a lower income bracket has a lower expected life than the brackets above it, then they are in poverty. They aren't receiving all their needs, let alone luxuries. Of course, the poor have different lifestyles, and it may even be that expected lifespan is negatively correlated with wealth (although I doubt it).

posted by: Jason on 09.19.03 at 11:48 PM [permalink]


I would be extremely surprised if lifespan is not positively correlated with wealth. >65% of Americans are obese, >30% morbidly so, and the top six causes of death in the US are elective diseases.

As a higher income earner, I am highly opposed to anyone confiscating my income to defray the natural expenses of anyone's lack of lifestyle control.

posted by: Jonathan on 09.19.03 at 11:48 PM [permalink]

Crooked Timber has a somewhat related post.

posted by: GT on 09.19.03 at 11:48 PM [permalink]

Does that mean you aren't opposed to subsidies on healthy food and gym membership to the level that everyone can get access to them? :)

One of the reasons the poor are obese is that fat is a cheap source of calories (not the only reason, admittedly).

posted by: Jason on 09.19.03 at 11:48 PM [permalink]

I disagree with that. A bowl of carrots with a pat of butter and an apple is cheaper than a big mac or a bag of fritos. This is about poor decision making. Is it so politically incorrect to argue that one of the key predictors of long term poverty is poor decision making? Is it a further stretch to assume that not being responsible with your diet could be an indicator of irresponsibility in other important areas? The statistics show that being below the poverty level in America is more often than not a temporary state. Society can provide the opportunity for upward mobility is, but can simply not guarantee success. Maybe its crass, but the reality is that the 30-50% of those stuck below the poverty line for an extended period are there largely do to their own poor decision making (and having children out of wedlock is a huge one).

posted by: Mark Buehner on 09.19.03 at 11:48 PM [permalink]

I am no economist, but isn't income inequality correlated to the distribution of political power? i.e won't the rich and very rich try to maintain their status by influencing the political process as well? Also, I think the real issue is one of mobility into the higher economic strata -i.e into and out of the top 10% or so of the economy. Is there any way of determining this? And won't the wish of rich and very rich people to maintain their status at some time go against maintaining mobility?

posted by: vkr on 09.19.03 at 11:48 PM [permalink]

Scott K:

A bit of gentler correction, here. I used to feel as you did about corporate execs. But after watching them for a couple of decades, I've come to a couple of conclusions:

1) They are willing to do the jobs they do. I'm not.

2) Apparently, they generate sufficient return for the company to justify their pay. If the ability to generate profit for the company were commonplace, we could hire a half dozen at less than $100k each.

3) Although these guys are dumb in ways I'm not, I've come to the conclusion that they're smart in completely different ways that I'm not. And that I'm not interested in, either.

Hope this helps. The ability to generate income in a particular area is a function of both interest and competency (which, admittedly, is also a function of interest). If you think you can do better for less, you should pursue that. I'd like nothing more than to see the compensation packages for corporate execs to get pared down a bit. I just can't see that it makes sense for any law to dictate that.

posted by: David Perron on 09.19.03 at 11:48 PM [permalink]

The Consumer Price Index is based on the price of a "Standard Market Basket" whose contents has shifted over time. When the price of cell phones and microwave ovens drops from $99 to $89, this is reflected in a drop in the CPI. Neither of these items were in the market basket when they cost $1500. When the average price of an automobile goes up, so does the CPI, but this does not reflect the fact that today's automobile includes antilock brakes, which added $5000 dollars to the price of a Jensen Interceptor at the time that a new BMW sports sedan cost $3500.

posted by: triticale on 09.19.03 at 11:48 PM [permalink]

I'm not an economist by trade, but I studied the subject in college. I have a question about the wealth measurement that people were touting above. Namely, what effect would longer lifespans have one the accumulation of wealth at the top percentiles, however you wish to cut them up? It would seem to me that with older people living longer, and thus holding on to their gains longer, there would be fewer wealth transfers to the young and penniless -- those who may have income but have not yet accumulated wealth.

I am in that category. I have a good job and a good income, but no wealth accumulation, as I just graduated from law school last year. My grandparents are still alive -- in their late 80s. They have accumulated wealth, although they have fixed incomes now. Hopefully my parents will enjoy good health for a long time to come -- they have both accumulated wealth and relatively high income. But this longevity would seemingly affect any historical comparison of a "wealth" measurement of economic mobility. In the 50s, my grandparents would statistically have a high likelihood of not having made it this far, and my parents might not have had very long to go until their wealth was distributed to myself and my siblings via inheritence.

What say the experts? Could the the driving factor behind a seeming stagnation of mobility of wealth be increased lifespans?

Also, relatedly, how does income mobility compare with age? Does it generally correlate, with younger people falling into lower income percentiles generally than their elders who have similar education and other qualifications (other than experience of course)?

posted by: Chris on 09.19.03 at 11:48 PM [permalink]

I was just wondering if anyone has a personal story about their life regarding income inequality because I am a writing an essay for a contest regarding income inequality. I am looking for someone who has a story about how they hardly made any money and then they eventually rose in the income categories and is doing well today. If you do you can send your reply to and you can remain anonymous for your own sake.

Thanks to anyone who chooses to respond. It will be most helpful in writing my essay.


posted by: Jason on 09.19.03 at 11:48 PM [permalink]

Post a Comment:


Email Address:



Remember your info?