Semi-Daily Journal

I don't have them with me now, but if anyone is interested, I can post scare quotes by Jefferson, Hamilton, and even grouchy, conservative old John Adams on the dangers of the inequality of wealth for a republic, and republican virtue.

Adams is especially good, with visions of the corrupt aristocracy of wealth bearing down on the poor like hunters and hounds after a poorl little fox, and killing it just for sport. Adams even manages a good word for the demogoguic Caesers and Napoleans that rally the poor, since the mass of people will need them for protection, even if it means further corruption of society.

The wisdom of the Founders for today...

If indeed "a small group of wealthy but untalented children controls vast segments of the US economy", the negative consequences of that will go far beyond "a vast waste of human potential".

"American Dream" was a phenomenon of the growth stage of the Industrial Revolution in US, which is now nearing the end of its life cycle.

Indeed, goodbye, American Dream (which is not the fault of the Republicans, by the way).

And, when you take out American Dream, what do you have left of the "American Century"?

That's what I mean by consequences beyond waste of human potential.

Paul Krugman notes fiscal policy ways to shift after-tax income away from the poor and towards the rich, which is all true. But aren't even pretax incomes migrating from poor to the rich as in something (or things) in the market place is adding to this over time? Now if a cynical policy maker wanted to do this, the means would be to reduce the accumulation of capital (lower national savings) so its return rises while the return to labor falls (capital more scarce, labor relatively more abundant). Wait? That WAS Reagan fiscal policy and it IS Bush43 fiscal policy.

Harold McClure:

I think the cynical policy maker would rather simply step up immigration in order to make labor more abundant.

I've always wondered why economists, who surely believe in the importance of incentives more than any other group, don't usually take their arguments to the extreme. It is surely to be expected that, invisible hand notwithstanding, those with influence have an incentive to gather to themselves more influence. So Krugman asks:

"Suppose that you actually liked a caste society, and you were seeking ways to use your control of the government to further entrench the advantages of the haves against the have-nots. What would you do?"

and I think, what is that "suppose" doing there? Those at the top of a caste system have a strong incentive to further entrench their advantages. Which is what is happening. Why is anyone surprised?

Perhaps the main difference between liberal economists and other left wing folks is that liberal economists tend to believe that the invisible hand provide an effective counterbalancing force. Others believe that poor people get better off, collectively, because of their own efforts and struggles, and that those with influence have means, motive, and opportunity to cut the invisible hand off at the wrist.

Eeconomists really have to choose: incentives or markets, which do you really believe in?

Perhaps many of the good paying jobs have been exported, and Americans are being job shifted into working for Walmart. According to the WSJ, IBM plans to shift 4,700 tech jobs to India and China. The current employees will be told to train their replacements to add insult to injury. Will these people experience upward social mobility when they shift to new work?

How convenient to blame the Republicans for everything, the universal whipping boy for all the ills of society. Does Krugman think that states run by Democrats have the best schools? California must have a great school system.

And BTW you needn’t be a plutocrat to be living off dividends and capital gains. Suppose a University of California Professor decides to retire and take a lump-sum cash out instead of a monthly defined benefit. Does this transform him into a plutocrat?

Krugman's key factual claim about social or economic mobility -- not income inequality -- seems to be (from the linked Nation article):

"It is true, however, that America was once a place of substantial intergenerational mobility: Sons often did much better than their fathers. A classic 1978 survey found that among adult men whose fathers were in the bottom 25 percent of the population as ranked by social and economic status, 23 percent had made it into the top 25 percent. In other words, during the first thirty years or so after World War II, the American dream of upward mobility was a real experience for many people.

Now for the shocker: The Business Week piece cites a new survey of today's adult men, which finds that this number has dropped to only 10 percent. That is, over the past generation upward mobility has fallen drastically."

I'm not aware of which 'classic 1978 survey' this is. Does anybody know? Which contemporary study is he referring to? Knowing this would help the discussion a lot.

Two points, without knowing which studies Krugman is citing:

1) Studying intergenerational social and economic mobility requires data on parental social and economic status. This creates a difficult data problem since many/most people don't know what their parents actually earned (my parents sure didn't tell me -- since my father is a public employee I had to go look it up, but I don't think everybody can do this or actually does this). The way sociologists who study intergenerational mobility get around this data problem is that they ask for the father's 'profession'. I don't know off the top of my head what the categories are, but they are things like 'unskilled machine operator, skilled machine operator, clerical worker' etc. The sociologist then links these categories to 'socioeconomic status' -- there are standard ways of doing this. Intergenerational mobility then has to do with measures how many parents in bottom 25% occupational categories have kids in top 25% occupational categories. You can see the problems that arise here when making judgements about how mobility changes through time: the meaning of these occupational categories changes through time. What it meant to be an 'unskilled machine operator' in 1940 was different than in 1970 and is again different in 2000. Similarly, it may be hard to find equivalent yet different categories through time: what occupational group today is the social and economic status equivalent for 'unskilled machine operator' in 1940? You could get around this problem if children actually knew the wages or income of their parents, but they don't (or they are just wrong, even if they think they know). There are now studies that have been tracking families since the late 1960s and early 1970s that do have hard data -- the Michigan PSID in particular -- but these studies cannot tell us anything about income mobility from say 1950 to 1978, what Krugman seems to be trying to get at. If anybody knows of a study which actually achieves this I would very much like to hear about it.

2) If 23% of the top 25% by 'social and economic status' -- again, note this is not income or wealth, but to some degree self labeling by occupational class -- come from the bottom 25% in 1978, as claimed by Krugman, then this is pretty much total mixing (total mixing would be 25% of the top come from the bottom 25%). This just doesn't strike me as a credible figure even for the 1970s -- I just don't think the US was like this: for instance, had African-American achieved full intergenerational mobility by 1978? You'd need that to get a 23% figure like this.

This 23% figure would seem more credibly interpreted as at least in large part due to misreporting of parental occupation or some other measurement or survey error, due for instance to the shifting meanings of occupational labels -- this itself would be interesting, although not as immediately relevant politically.

Starting with references to the original article...

"America was a highly unequal society, and it stayed that way through the 1920s...

"During the 1930s and '40s, however, America experienced what the economic historians Claudia Goldin and Robert Margo have dubbed the Great Compression: a drastic narrowing of income gaps, probably as a result of New Deal policies"

New Deal policies?? Gee ... the Great Depression and World War II don't get even a mention?? Is this not a tad disingenuous? ;-)

"And the new economic order persisted for more than a generation:"

Well, that's about what one would expect for the consequences of the Great Depression and WWII generation, no? Which was an aberration, not the norm, after all -- and hardly brought about by any policy ... well, not by any policy we would want to emulate, certainly. But after that, back to normal and better times, as before.

"According to estimates by the economists Thomas Piketty and Emmanuel Saez--confirmed by data from the Congressional Budget Office--between 1973 and 2000 the average real income of the bottom 90 percent of American taxpayers actually fell by 7 percent."

Wow! But isn't it odd the Census didn't catch this? To have 90% of the population lose real income is something one would think the Census would spot!

I wonder why the Census reports that from 1973-2000 mean income in the lowest four income quintiles rose by 18%, 19%, 21% and 30% in inflation-adjusted dollars. It looks like 80% of the population was gaining real income while 90% was losing it!

And, of course, if the inflation measurement for that period is off by about 1% a year, as per Boskin, as both our host and Prof Krugman (elsewhere) have said is probably the case, then real gains were substantially higher.

"Now for the shocker: The Business Week piece cites a new survey of today's adult men, which finds that this number has dropped to only 10 percent...."

Though of course the fathers now are much richer than were the fathers of the Great Depression/WWII generation. So the fathers now naturally are harder to top -- and the children could do quite well as a group without surpassing quite so many of them.

However, what I find if not quite shocking then rather odd about this, is that a top economist would make what he calls a "shocker" claim just by citing a magazine citing only an anonymous "survey" as evidence of it. This seems conspicuously weak third-hand sourcing of authority, as top academics go. What survey? Why doesn't Professor K refer to it directly? Why isn't the methodology mentioned, or at least the source, if it is so persuasive? I can't get to it on the BW web site to find out for myself.

I do note however that the same issue of BW online has a couple nasty protectionist articles in it, bashing free trade. Does Prof K give equal authority to everything he reads in BW?

"Put it this way: Suppose that you actually liked a caste society, and you were seeking ways to use your control of the government to further entrench the advantages of the haves against the have-nots. What would you do?.... you'd try to reduce tax rates on people with high incomes ..."

Sure, I'd be evil enough to do that. Although Prof K in his pre-2000 books, such as _Age of Diminished Expectations_, was fair and balanced enough to note that income inequality is a function of pre-tax income, and the tax code cannot be deemed either an effective cause or remedy of it. But that was then...

"Meanwhile, on the spending side, you'd cut back on healthcare for the poor..."

Yup, by enacting the biggest expansion of healthcare entitlements since LBJ created Medicare itself. ;-)

BTW, here we have yet another purported authoritative critique of the "income gap" that for unknown reasons fails to mention that measuring *consumption* by income level reduces the top-bottom gap by more than 75% -- and it is consumption that determines welfare.

But hey, let's not confuse people with a fuller picture. After all, 90% of the population haven't even gotten the message yet that they're poorer than they were 30 years ago.

by enacting the biggest expansion of healthcare entitlements since LBJ created Medicare itself

Correct me if I'm wrong, but the premium structure ($400/yr) and benefit payout (first $4000 out of $5000 out-of-pocket) isn't much of an entitlement for most people. If you can afford that, you aren't that needy.

History shows that the huge pool of wasted talent could be heavily armed and would overthrow the government run by the smaller number of wealthy, but untalented children, extract vengeance, and then start again.

Jim Glass: "Sure, I'd be evil enough to do that. Although Prof K in his pre-2000 books, such as _Age of Diminished Expectations_, was fair and balanced enough to note that income inequality is a function of pre-tax income, and the tax code cannot be deemed either an effective cause or remedy of it. But that was then..."

To which the only possible response is, "Huh?" Obviously increasing income transfers from the rich to the nonrich will reduce income inequality; the only valid question is whether it will also reduce total national productivity enough to do more harm than good.

Meanwhile, not a word from Glass about the fact that the elimination of the estate tax -- and the reduction in taxes on unearned income -- are indeed allowing the income of the current upper class (and their children) to snowball in a way far out of proportion with their actual efficiency as economic producers. Them that has gets more, to a very large degree, simply because they have more to invest -- not because they get more efficient at utilizing their investments to produce more. Which is particularly indefensible in the case of the estate tax.

And which makes Glass' comment that "measuring consumption by income level reduces the top- bottom gap by 75%" even more ridiculously irrelevant. Obviously Bill Gates doesn't spend all of those billions he makes per year -- only, at most, a few million per year -- which does absolutely nothing to alter the fact that less affluent people need those few million tremendously more, and that a fairer distribution of investable income (including money for schooling) could well allow them to gain such additional consumable income without reducing total national productivity enough to do more harm than good. For starters, they might be able to afford the sort of education that Bush Jr. threw away at Yale...

"History shows that the huge pool of wasted talent could be heavily armed and would overthrow the government run by the smaller number of wealthy, but untalented children, extract vengeance, and then start again."

Which seems a rather wasteful way to do it. In reality, in any democracy there will be an electoral reaction against excessive privilege at some point, precisely as there was during the Progressive Era (without the need for a Great Depression). The only question is how far inequities will grow before that backlash occurs -- which means that the one real issue of "unfairness" is the extent to which the wealthy monopolize campaign and political advertising funds, how seriously this has been worsened by the move toward broadcast media and away from print media, and what the best approach to reduce this is.

Hmm...

Here are some approaches I'd take to strengthen a hereditary caste system.

1) Emphasize the role of genetics in determining productivity and creativity, a la The Bell Curve. Make a genetic science of hereditary intellectual power and motivation the official position of the government.

2) Reform inheritance by making not just wealth, but debts hereditary. Introduce forms of debt bondage that would be permissible under the 13th amendment (with a compliant Supreme Court with a narrow definition of slavery).

3) Make it publically understood that to help ones family using mechanisms under your control not limited to wealth, such as patronage jobs from corporations or governments is not something perverse, but natural and good: the duty of a parent to help his/her child, as much so as education, clothing, etc.

4) Shift taxes from income to consumption. This could be publicly (and possibly rightly) said to be an effort to boost national savings. Uniform national sales taxes would be better than progressive consumption taxes in this respect.

5) Weaken regulations against environmental factors which would cause damage to lower-caste children, such as lead poisoning. This could help widen the real gap in abilities between lower-caste and upper-caste children.

6) Decriminalize or dramatically reduce sentences for white-collar offenses, such as bribing politicians, breaking contracts with workers, etc. Ensure that blue-collar offenses are broadly defined to include many activities commonly ignored if committed by the higher castes and are punished severely. Attempt to make blue-collar sentences break up nuclear families in blue-collar society, so as to create more single-family homes/orphans.

7) Reform public institutions of higher education by demanding that they stop letting students in for tuitions lower than the marginal cost of education, i.e. abolish scholarships if they mean that a student, as a whole, costs an institution money (a $15,000 scholarship on a $35,000 tuition would still be permissible if the cost of education was less than $20,000).

8) Reduce technological change: ban science if it's ethically questionable, and reduce public funding for research that isn't. This could promote stability in industry, preventing high-caste families in old industries from seeing their fortunes swept away by economic change (i.e. make sure that the son of a well-established buggy-whip manufacturer isn't impoverished by the ascendance of a Henry Ford).

9) Increase immigration, but abolish special preferential visas for skilled workers. Try to promote unskilled immigration.

10) Introduce barriers, such as licensing fees, spectrum broadcast permits, etc, to the media. Make publishing and broadcasting expensive. Introduce content-quality regulations into the internet, or urge ISPs to do so on their customers' behalf.

I'll see if I can think of more, but I achieved a ten-point caste reform plan, and since 10 is a nice number according to our arbitrary base ten system, that's what I'll settle on for now.

Let's see if this happens as time progresses.

Thoughts? What do you think is the best way to promote a more stable caste system?

Satire or no, Master Elson, you've given me a great idea for a role-playing campaign ;)

Elson: sounds uncomfortably like _Brave New World_.

Very keen observer, Julian Elson.

Good discussion here.

When the Soviet regime collapsed, the World Bank, which is really run by US, undertook a program of "democratization and liberalization" in Central and Eastern Europe.

The fundamental idea there was/is that market economy and democracy are inseparable.

Through markets, comes resource allocation, on the basis of vote by the wallet.

Through democracy, comes the rules by which the markets operate, on the basis of vote by the head.

This, of course, is the ideal picture and vote by the wallet does have an impact in making the rules.

Representative democracy facilitates vote by the wallet having an impact in making the rules -- even without trickery or subtle threat of using force.

I think technology and productivity have now reached a level in US and Europe and industrialized west in general for transition to direct democracy.

The Democractic Presidential candidate, therefore, should perhaps propose a fundamental policy of transition to direct democracy in industrialized west and expanding democratization and liberalization in the rest of the world.

If US could establish leadership in such a global movement and get it rolling, then that leadership role could make up for the disappearence of the "American Dream".

Your assumption that a democratic admin/congress would fix this is an act of faith with no clear basis in logic or history.

If Krugman (or Brad) had more than a thimbleful of grey matter between them then they would ask two very obvious questions:

a) Is this trend unique to the USA, or does it also occur in similar Western nations?

b) Has this trend only occurred under conservative government.

Particularly a). It is a critical question and anyone with a genuine spirit of inquiry (rather than the tiresome partisan hacks we see here) would be immediately asking it.

Oh well.

Stefan wrote, "If 23% of the top 25% by 'social and economic status' -- again, note this is not income or wealth, but to some degree self labeling by occupational class -- come from the bottom 25% in 1978, as claimed by Krugman, then this is pretty much total mixing (total mixing would be 25% of the top come from the bottom 25%). This just doesn't strike me as a credible figure even for the 1970s..."

It sounds like you think the 23% is 23 out of 100, which is certainly not plausible.

Clearly the reference is 23% *of* 25%, i.e., about 6%.

a wrote, "If Krugman (or Brad) had more than a thimbleful of grey matter between them then they would ask two very obvious questions:

"a) Is this trend unique to the USA, or does it also occur in similar Western nations?

"b) Has this trend only occurred under conservative government.

"Particularly a). It is a critical question and anyone with a genuine spirit of inquiry (rather than the tiresome partisan hacks we see here) would be immediately asking it."

Just off the top of my head, I think the answers are (a) not unique; (b) not just under conservative governments.

Speaking of "partisan hacks," what makes you think that BDL or Krugman *don't* think it's a critical question?

Broadly agree with Krugman and BDL. It is important to keep America as a Land of Opportunity, where social status and wealth are not automatically handed down.

I do think natural attrition plays an important role. Rich people in America can have goof-ball kids. Look at Paris Hilton. Whatever the government does, American culture is still such that you are defined by what you do - by your work - and people who don't work are frowned at if not despised. In such an environment the idle rich self-destruct - just give them time.

I also think he misunderestimates the role of universities, which after all globally have huge endowments and thus have a significant claim on America's present and future wealth. Unfortunately the wealthy will always have better access to the "best" universities, but it is far from exclusionary.

I don't know; 10% making it into the top 25% still seems decent to me, but I guess I'm expecting a much larger group to make it into the second quartile. Also would expect that a lot of the third quartile make it into the first (acually, my situation). If that didn't happen, I'd be very concerned.

Still again broadly in line with Krugman and BDL. Improving opportunity is good for everyone, including the idle rich. After all they need to sponge off somebody.

Jim: The reason for the disparity between Picketty's estimate and the Census numbers you quote is presumably encapsulated in the word "taxpayers". I presume that you speed-read and thus missed that "90% of taxpayers" does not mean "90% of the population", though I agree it has this implication and think it was wrong of Krugman to use this factoid.

Stephen - No, he has a good point. What he is saying is you are breaking the citizenry up into quarters:
1) bottom 25% of wage earners
2) lower middle 25% of wage earners
3) upper middle 25% of wage earners
4) upper 25% of wage earners

Now they are saying from one generation to the next, 23% of category #4 is from the previous category #1. You're right that this is only 6%(of the total), but it suggests that then 6% came from each of the previous categories(i.e. that someone who was a child from the previous #4 category was no more likely than someone who was a child from the previous #1 category to be in the current #4 category).

Now that said, I just thought of something - we are assuming that the fact that 1/4 came from the bottom rung means that the bottom rung was just as likely to be at the top in a generation as the old top rung was. What if(part of) the answer is instead simply that the bottom rung had far more children, and thus far more chance to get into the top rung. If say 40% of the children had parents in the bottom rung, and only 15% of the children had parents in the top rung, the numbers start sounding far more believable.

I'd still like to see the actual study though

Oops. Actually just re-read the piece. Krugman actually says "25 percent of the population as ranked by social and economic status, 23 percent had made it into the top 25 percent."

This means you can toss what I just said about it being explained by the poor having more kids. A truly bizarre claim, and one I find hard to believe is true, unless Krugman slipped up on the wording

As dsquared says the taxpayers bit does confuse. Essentially I expect as the tax base has been extended it has brought poorer people into the net, thus bringing the 'average' income down.

Essentially the bottom 90% took 56% of income in 2000, compared with 68.5% in 1970. The tax base though rose from something like (i've closed off the file so this is from recollection) 70m to 120m).

So one can't really make any comparisons without knowing the income of those not paying tax in 1970.

The libertarians who've taken over this thread are so wrapped up in their self-regarding Social Darwinism and distopian fantasies about Harrison Bergeron that they miss their own best argument. The data in Krugman's piece focuses on MEN. Women's incomes have risen continuously during the period in question.

The problem is that women's participation in the workforce has lowered wages overall (supply and demand), limited opportunity for working class men (because middle class women have better quals), and accentuated class differences (since middle class people marry one another).

Yes, policy choices do impact inequality. Title Seven of the Civil Rights Act was a monumentally-just law. But liberals need to admit it temporarily stalled social mobility for many men.

Two points.

1. A mountain of rhetoric ("caste society") on top of a molehill of evidence--two surveys. Other studies contradict those surveys. Also, if you just look at, say, the Forbes richest people, there are fewer heirs and heiresses than there were 25 years ago.

My guess is that if you were to do a complete survey of available evidence, you would not conclude that social and economic mobility in the U.S. has declined.

2. In terms of public policy, I think that the strongest factor holding down poor people is the public education system, run for the benefit of the teachers' union. The public schools where I send my children are almost completely dysfunctional. I believe (hope) that we make up for that in our household, but the children of less affluent, less well-educated parents who attend the same schools are going to suffer.

Gee, Arnold, attack Krugman for insufficient data, then present a vague, informal analysis of the tiny sample that is the Forbes 500 and your own rhetorically charged characterizations (they can't even qualify as anecdotes, since you offer no details).

Surely, two studies trump that stuff.

Tom Slee raises an interesting point, which reminds me of an old argument made by Milton Friedman who in turn credited Adam Smith for the insight. Often, government intervention aids the well to do by promoting monopoly power and other interferences with how the market would allocate resources. Of, by, and for the rich people, I guess.

Lot of interesting comments here but I wonder if we're missing the main thrust of Krugman's piece in the Nation, and perhaps, what BusinessWeek was trying to say. Two years ago my family's income was in the top 10%, now we're bottom 20%. The difference is surprisingly little when compared with those families in the 1% whose income rose 148% over the last 30 years and received significant help from the government these past two years. I suspect the thrust of Krugman's piece is right, upward mobility is taking a hit, income inequality is exploding, and our society down the road will be worse for it, but you're all right to want to dig into the data. Can't wait to see what you find.

Hence the tremendous opportunities for providers of lifelong learning and career services (check the capital markets for confirmation). In no small part because, over time, these providers will seek to grow by enlarging the size of the market through financing for customized LLCS programs...

So, as entrepreneurs are already racing to address the gruesome specter of wasted potential, don't despair too much...

Hey Opin, Arnold K did a fine quick comment on an easily checked factoid that clearly questions the two surveys as interpreted by PK.

The staus of women is very important. Also the number of poor folk in nuclear families vs broken homes -- I'd guess the vast majority of the prior bottom who make it to the top do so from homes where they live with both their parents.

Yes, income inequality is bad, and getting worse. But many gov't policies & actions are helping the rich stay rich with less effort, less risk, less benefit to others. -- yet the Left wants more power to the gov't!

When will the Left admit that gov't that is stronger, bigger, more controlling will more certainly be controlled by that same top 1%, who will insure they get massive, unfair, special perks? Just as regulators get "captured" by the industries they're supposed to regulate, politicians want to be liked and accepted by the top 1%.

Direct democracy could be implemented fairly quickly: 100% tax credits for donations to gov't agencies. Let the tax payers decide how much of their tax dollars go where.

Look at Europe, and France, and the bureau-hell decline, and be warned -- more gov't means more like France. Yechh.

Bush in Iraq is great. Most other Bush spending has been mediocre, or quite bad.

The Republican business model seems to be along the lines of parlaying contributions to the Republican Party into gaining direct access to the Federal treasury. Examples include the no-bid contracts awarded to Halliburton and the subsidized entree to Medicare awarded to the HMO's. I would doubt that many CEOs in innovative, truly competitive, job-creating industries are in fact Republicans.

Brad,

Gracious host. Resident Big Brain (meant respectfully, not mockingly).

What is your opinion on the piece? The commenters here seem to have raised some valid doubts about Krugman's sources and conclusions. I'd be interested to hear the opinion of an informed expert.

The "new survey" that Krugman (via Business Week) cites was done by David Wright, Earl Wysong, and Robert Perrucci. Frustratingly, it appears not to be available on the Web, and it has not been published anywhere. Here's a short article from Wichita State (where Wright teaches) about it: http://webs.wichita.edu/dt/insidewsu/show/article.asp?201. It's not clear from Krugman's piece that he's actually looked at the study, nor that he's actually read the 1978 study (which I assume is Featherman and Hauser's work, but I don't know) either. He calls it a "classic 1978 survey," while the Business Week writer calls it a "classic 1978 study," with neither citing the authors. That seems a little sketchy.

The Wright/Perrucci/Wyson study looked at 2749 father-and-son pairs. It's possible that one can extrapolate about the entire U.S. economy from 5500 people, but I'd like to see more evidence than that. More troublingly, the study was not looking purely at income mobility. Instead, it was looking at status mobility, looking at not just income but also, according to the New York Times, "occupational prestige." That seems to bring in a subjective dimension that compounds the already-difficult measurement problems.

Those problems are also compounded by the fact that the authors of the study had, before the fact, a clear and explicit ideological stance on the matter. Wright, for instance, believes that no American should be legally allowed to make more than $80,000 a year. Obviously, you can be biased and still do serious work, but in a field as freighted and as complex as this one, it's not reassuring that the authors of this study found exactly what they not only expected to find but also wanted to find. And in any case, Wright, at least, seems to disagree with Krugman's argument that things have gotten significantly worse. He said: "There is not very much class mobility in the United States, and there never has been."

More puzzlingly, the very Business Week article upon which Krugman is relying for all his numbers also mentions (though not by author name) a study by Katherine Bradbury and Jane Katz of the Boston Fed. This, at least to my eyes, seems like a serious study: http://www.mindfully.org/Reform/2002/Incomes-More-UnequalSep02.htm.

Bradbury and Katz found that mobility was flat in the 1970s and 1980s, and "decreased slightly in the 1990s." Now, those numbers are skewed by the fact that they end in 1998, at which point the economy was really revving up, and the productivity boom was only a couple of years old. My suspicion is that if you ran the numbers through 2000, mobility would be about the same for all three decades. But in any case, there is zero sign of the "shocker" that Krugman describes. Bradbury and Katz's numbers say that if of the people in the poorest quintile (and by the way, measuring mobility by the poorest quintile's moves seems dubious to me) in 1969, 3% had made it to the top quintile by 1979. That's a long way from the other numbers PK cites. I think the way statistics are used in this Nation article is questionable at best.

Brad,

Gracious host. Resident Big Brain (meant respectfully, not mockingly).

What is your opinion on the piece? The commenters here seem to have raised some valid doubts about Krugman's sources and conclusions. I'd be interested to hear the opinion of an informed expert.

The "new survey" that Krugman (via Business Week) cites was done by David Wright, Earl Wysong, and Robert Perrucci. Frustratingly, it appears not to be available on the Web, and it has not been published anywhere. Here's a short article from Wichita State (where Wright teaches) about it: http://webs.wichita.edu/dt/insidewsu/show/article.asp?201. It's not clear from Krugman's piece that he's actually looked at the study, nor that he's actually read the 1978 study (which I assume is Featherman and Hauser's work, but I don't know) either. He calls it a "classic 1978 survey," while the Business Week writer calls it a "classic 1978 study," with neither citing the authors. That seems a little sketchy.

The Wright/Perrucci/Wyson study looked at 2749 father-and-son pairs. It's possible that one can extrapolate about the entire U.S. economy from 5500 people, but I'd like to see more evidence than that. More troublingly, the study was not looking purely at income mobility. Instead, it was looking at status mobility, looking at not just income but also, according to the New York Times, "occupational prestige." That seems to bring in a subjective dimension that compounds the already-difficult measurement problems.

Those problems are also compounded by the fact that the authors of the study had, before the fact, a clear and explicit ideological stance on the matter. Wright, for instance, believes that no American should be legally allowed to make more than $80,000 a year. Obviously, you can be biased and still do serious work, but in a field as freighted and as complex as this one, it's not reassuring that the authors of this study found exactly what they not only expected to find but also wanted to find. And in any case, Wright, at least, seems to disagree with Krugman's argument that things have gotten significantly worse. He said: "There is not very much class mobility in the United States, and there never has been."

More puzzlingly, the very Business Week article upon which Krugman is relying for all his numbers also mentions (though not by author name) a study by Katherine Bradbury and Jane Katz of the Boston Fed. This, at least to my eyes, seems like a serious study: http://www.mindfully.org/Reform/2002/Incomes-More-UnequalSep02.htm.

Bradbury and Katz found that mobility was flat in the 1970s and 1980s, and "decreased slightly in the 1990s." Now, those numbers are skewed by the fact that they end in 1998, at which point the economy was really revving up, and the productivity boom was only a couple of years old. My suspicion is that if you ran the numbers through 2000, mobility would be about the same for all three decades. But in any case, there is zero sign of the "shocker" that Krugman describes. Bradbury and Katz's numbers say that if of the people in the poorest quintile (and by the way, measuring mobility by the poorest quintile's moves seems dubious to me) in 1969, 3% had made it to the top quintile by 1979. That's a long way from the other numbers PK cites. I think the way statistics are used in this Nation article is questionable at best.

Brad,

Gracious host. Resident Big Brain (meant respectfully, not mockingly).

What is your opinion on the piece? The commenters here seem to have raised some valid doubts about Krugman's sources and conclusions. I'd be interested to hear the opinion of an informed expert.

Let's just see what we have really done to instill a caste system in the mid to latter part of the last century.
First, let's build a punitive income tax system that disenfranchises fully 50% of the people by having them pay less than 4% of the tax burden. What do they care what the people in Washington do?
Then, lets start with the "Great Society" whose only demonstrable effect has been to further the institutionalization of poverty greatly. In spite of that, we continue to build on its ideas having created a welfare system keeping generations in poverty.
Why not then devalue the importance of the educational system by pandering to the teachers' union to the detriment of the students. Turn the whole educational algorithm upside down to get votes from a special-interest group.
Finally, let's also devise an outrageous tort system to reward another special interest group of tort lawyers so that we can stifle innovation and invention for the advancement of medicine and other societal goods, thus assuring more votes to the political class. All this of course to the detriment of the lower economic strata who need these services.

An earlier thread: http://www.j-bradford-delong.net/movable_type/2003_archives/002838.html
has all kinds of data and links about inequality in America, and in particular about the question of whether things have, in fact, worsened significantly since the 1970s. (The debate in that thread was specifically about inequality and not mobility, but many of the numbers are relevant.) My reading is that they haven't, and that the biggest shift has been within the top 10%of the population, with the top 1% taking home a significantly larger share of income and wealth, which is arguably a bad thing, but not evidence that a once-equal society has become an oligarchy. But the numbers are open to interpretation, and certainly worth taking a look at.

Let's just see what we have really done to instill a caste system in the mid to latter part of the last century.
First, let's build a punitive income tax system that disenfranchises fully 50% of the people by having them pay less than 4% of the tax burden. What do they care what the people in Washington do?
Then, lets start with the "Great Society" whose only demonstrable effect has been to further the institutionalization of poverty greatly. In spite of that, we continue to build on its ideas having created a welfare system keeping generations in poverty.
Why not then devalue the importance of the educational system by pandering to the teachers' union to the detriment of the students. Turn the whole educational algorithm upside down to get votes from a special-interest group.
Finally, let's also devise an outrageous tort system to reward another special interest group of tort lawyers so that we can stifle innovation and invention for the advancement of medicine and other societal goods, thus assuring more votes to the political class. All this of course to the detriment of the lower economic strata who need these services.

An earlier thread: http://www.j-bradford-delong.net/movable_type/2003_archives/002838.html
has all kinds of data and links about inequality in America, and in particular about the question of whether things have, in fact, worsened significantly since the 1970s. (The debate in that thread was specifically about inequality and not mobility, but many of the numbers are relevant.) My reading is that they haven't, and that the biggest shift has been within the top 10%of the population, with the top 1% taking home a significantly larger share of income and wealth, which is arguably a bad thing, but not evidence that a once-equal society has become an oligarchy. But the numbers are open to interpretation, and certainly worth taking a look at.

Let's just see what we have really done to instill a caste system in the mid to latter part of the last century.
First, let's build a punitive income tax system that disenfranchises fully 50% of the people by having them pay less than 4% of the tax burden. What do they care what the people in Washington do?
Then, lets start with the "Great Society" whose only demonstrable effect has been to further the institutionalization of poverty greatly. In spite of that, we continue to build on its ideas having created a welfare system keeping generations in poverty.
Why not then devalue the importance of the educational system by pandering to the teachers' union to the detriment of the students. Turn the whole educational algorithm upside down to get votes from a special-interest group.
Finally, let's also devise an outrageous tort system to reward another special interest group of tort lawyers so that we can stifle innovation and invention for the advancement of medicine and other societal goods, thus assuring more votes to the political class. All this of course to the detriment of the lower economic strata who need these services.

"1. A mountain of rhetoric ("caste society") on top of a molehill of evidence--two surveys. Other studies contradict those surveys. Also, if you just look at, say, the Forbes richest people, there are fewer heirs and heiresses than there were 25 years ago."

"My guess is that if you were to do a complete survey of available evidence, you would not conclude that social and economic mobility in the U.S. has declined."

Surely, as a bonafide economist, you can point to at least one concrete example, rather than rely on speculation?

"In terms of public policy, I think that the strongest factor holding down poor people is the public education system [...] children of less affluent, less well-educated parents who attend the same schools are going to suffer."

In other words, given equal educational opportunity, you assume educational outcomes are directly related to family affluence, and thereby confirm Krugman's point.

1. A mountain of rhetoric ("caste society") on top of a molehill of evidence--two surveys. Other studies contradict those surveys. Also, if you just look at, say, the Forbes richest people, there are fewer heirs and heiresses than there were 25 years ago.

My guess is that if you were to do a complete survey of available evidence, you would not conclude that social and economic mobility in the U.S. has declined.

Posted by: Arnold Kling on December 19, 2003 05:00 AM

Arnold, even if what you say is true, would this hold in the future with the removal of the estate tax and the removal of progressivity from our tax code? The answer is that no they would tend to induce less meritocracy in the future. So even if we were doing well now, we are making changes that could well be expected to undermine us in the future.

Jim Crockett wrote, "First, let's build a punitive income tax system that disenfranchises fully 50% of the people by having them pay less than 4% of the tax burden. What do they care what the people in Washington do?"

Huh? What about payroll taxes?

"Then, lets start with the 'Great Society' whose only demonstrable effect has been to further the institutionalization of poverty greatly."

Right. That's why poverty among the elderly has plummeted. Check.

"Finally, let's also devise an outrageous tort system to reward another special interest group of tort lawyers so that we can stifle innovation and invention for the advancement of medicine and other societal goods, thus assuring more votes to the political class." Uh huh. So why don't we restrict the rights of corporations to sue each other? More to the point, why should a doctor have limited liability for his practice when his profession refuses to be meaningfully regulated? (Meaning: ask yourself how many physicians in the US lose their licenses.) Finally, the system is stacked: if I were to create a website tomorrow collecting testimonials about egregious physician practices, I could be sued for libel, with, uh, what kind of limits on damages awarded?

Stan wrote, "Arnold, even if what you say is true, would this hold in the future with the removal of the estate tax and the removal of progressivity from our tax code? The answer is that no they would tend to induce less meritocracy in the future. So even if we were doing well now, we are making changes that could well be expected to undermine us in the future."

Agreed. It's one thing to argue that there's a limit to what government can or should do to increase social mobility and meritocracy. It's quite another to excuse policies that will make matters worse and which have little to recommend them otherwise.

Steve Carr:

Krugman cites the mobility of sons vs. fathers; Bradbury and Katz analyze that of families. I don't know if the difference in sample populations can be corrected for, but as it stands, you're comparing apples to oranges.

And the actual text of Bradbury and Katz is considerably more damning than the paraphrase you provide.

I used to really like reading Krugman.

But this paranoia about the evil tyrannical intentions of the Republicans is just too much.

The point about upward mobility is a good one, but doesn't he realize that the validity of the point is lost in the paranoid vitriol that tries to portray the Republicans as a secret society attempting to create a caste system?

I mean come on.

The point about class and wealth gap changes is valid AND very important IMO, but the more likely explanation is that it is social engineering incompetence (or even a perhaps erroneous belief that a rising tide floats all boats) rather than this secret conspiracy filled with malice that Krugman purports?

I've pretty much stopped taking Krugman seriously because he seems to have a political agenda - he's on a crusade to convince us all that Bush and the Republicans are evil monsters. I don't feel I can trust people with such an agenda to give me unbiased data.

It's sad really.

>.Now, those numbers are skewed by the fact that they end in 1998, at which point the economy was really revving up, and the productivity boom was only a couple of years old. My suspicion is that if you ran the numbers through 2000, mobility would be about the same for all three decades.

Almost certainly the other way round; the productivity miracle was accompanied by a rise in the profit share in GDP.

Also, I don't think James Surowiecki is being clear in his own mind about the meaning of the term "Oligarchy". It's precisely a domination of the economy by the very rich, not the top 10%.

"My reading is that they haven't [things worsened significantly], and that the biggest shift has been within the top 10%of the population, with the top 1% taking home a significantly larger share of income and wealth, which is arguably a bad thing, but not evidence that a once-equal society has become an oligarchy."

Sort of the point I was trying to make with my personal story; there could be lots of mobility within the bottom 99% but who cares, it's becoming less and less important to be in the bottom 9/10s of the top 10% and more important, for security and influence, to be in the top 1%. Not entirely true of course. It was much nicer having several months expenses in the bank versus living hand-to-mouth. A couple more generations of this and we'll get to see how well a very large democracy works with wealth so enormously concentrated.

Actually, the productivity miracle was not accompanied by a rise in the profit share of GDP. The "miracle" didn't get started, by all accounts, until 1995-1996. The profit share peaked in 1997 (assuming, of course, that these "profit" numbers are even reliable) and then fell for at least the next four years.(I'm not sure what happened in 2002.) In the nonfinancial corporate sector it fell even more sharply. (All these numbers from: http://www.findarticles.com/cf_dls/m1093/4_45/89871072/p1/article.jhtml?term=.)

In 2001, labor's share was the largest it's been in the postwar period: http://www.findarticles.com/cf_dls/m1093/4_45/89871073/p2/article.jhtml?term=. And the years from 1995-2001 saw sustained real wage growth for just about all American workers for the first time in two decades, which makes the idea that things have been getting steadily worse since 1973 a difficult argument to sustain.

Russell --

I'm not sure how the text of Bradbury and Katz is more damning than the paraphrase I provided. The question we were dealing with was whether mobility in the U.S. has changed significantly in the past three decades, as Krugman suggests it has. Bradbury and Katz say people in the bottom quintile are "a bit more stuck there," and that there was a "slight decrease in mobility" from 1989-1998. And their numbers speak for themselves. America was not especially mobile three decades ago, and it is not especially mobile now.

I realize the two studies are not looking at the same samples, but that's hardly a reason not to look at what Bradbury and Katz found, especially since the Wright study looks deeply flawed. And it would have been nice if we had a sense that Krugman had actually interrogated the numbers in these studies (including the 1978 survey) rather than simply using the Business Week article as a source.

I asked Krugman about "between 1973 and 2000 the average real income of the bottom 90 percent of American taxpayers actually fell by 7 percent" by email. He replied that he was quoting the Piketty-Saez paper. Problem is that the 2000 numbers are in an XLS file that updates the numbers from 1997/8 or whatever to 2000...and I don't have a copy of Excel. (And it crashes my version of StarOffice.)

Now how about that M$ monopoly...

Pragmatic Idealist wrote, "But this paranoia about the evil tyrannical intentions of the Republicans is just too much...The point about class and wealth gap changes is valid AND very important IMO, but the more likely explanation is that it is social engineering incompetence (or even a perhaps erroneous belief that a rising tide floats all boats) rather than this secret conspiracy filled with malice that Krugman purports?"

Huh? It's definitely malicious (speaking more broadly, not just on this question), but it's not a secret and it's not a conspiracy. Look at the evidence:
* Texas and Colorado redistricting moves;
* Out-and-out lying about the reasons for going to war in Iraq;
* Treasury Department refusing to release data on distributional impact of changes in the tax code;
and so forth.

About incompetence versus intentionally trying to transfer wealth upwards, I don't see any evidence for cutting them any slack, and there's lots of evidence to the contrary. (E.g. Paul O'Neill's quote that maybe it's reasonable to let people dodge taxes by moving their individual assets to tax shelters in the Caribbean.)

Quick, someone rush up into Canada to fetch some meds for Paul. I used to at least trust his fact-checking even if I'd disagree with what he had to say; didn't he tear a new one for Bill Greider for writing some nonsense about wages over time? What ever happened to his sincere defense of good economics? Compare his tone ten years ago about the same stuff to his tone today. Even if he's right, there's just no way of telling since it's hard to get past the venom.

Up until about November 2000: "We don't really know for sure what caused the increase in inequality, but we're sure it happened. The current best hypothesis seems to say that technological changes have been rewarding people with high levels of human capital. Trade might account for a few percent of the change, but it's really a bit of a mystery." This sort of thing was in response to some of the supply-side loons whose response is to deny that inequality has risen. Maybe he picked one of the more extreme positions as a strawman to rail against but he'd do it to illustrate a real point.

Since November 2000: "Bush and his right-wing cronies have a plan to turn this country into a Latin American craphole. They probably even have funny military uniforms with lots of brass buttons on order right now, in preparation for their coup. Picture Rumsfeld wearing really big sunglasses. The increase in inequality in the past few decades is a retroactive plot of theirs to enserf the majority of the population." Paul has confused making enemies with making a point.

OK, now a real econ question. Does anyone (Brad? Arnold? Daniel?) know of research about the effect of immigration on inequality (and maybe even total factor productivity, trying to tread lightly)? A casual reading of the evidence is that measured inequality is correlated with high levels of immigration since the US seems to attract disproportionately very rich or very poor people from other countries. I'd guess that there's a "significant" effect; has anyone actually measured how much? I'm looking for a research topic and I'd like to know if someone has already done this.

It could potentially explain a lot of things--there weren't that many immigrants in the 1940s through 60s when inequality was low, and there have been a lot since the 1960s, when inequality has been high. Ditto the 1880s.

Alternatively, it could be a spurious correlation and explain nothing at all.

Any ideas? Thanks.

Steve Carr,
Sure, I didn't say that your conflation of the sample populations proved Krugman right, I implied that because of this your claim that Krugman's cited data is bad did not hold.

"And their numbers speak for themselves. America was not especially mobile three decades ago, and it is not especially mobile now."

First, that's not the conclusion of the study.

Second, now you proceed to generalize from families to "America" using one study, but deny standing to the other, using the same logic as before, even though we agree that your initial logic is faulty. Assuming Krugman is not immolating his professional reputation, I suspect his conclusions are as valid as those of the Bradbury and Katz study, and the correct generalization to "America" would encompass both families and fathers/sons, i.e., the best representation about mobility in "America" is somewhere in the middle of the two views, and not the pessimal interpretation you prefer. But I agree with:

"And it would have been nice if we had a sense that Krugman had actually interrogated the numbers in these studies (including the 1978 survey) rather than simply using the Business Week article as a source"

The upshot of this is Krugman could still be wrong (we don't know yet), and if so, then I'd be much more in agreement with you.

Graciously, you've provided a link to the Bradbury and Katz study, so while I'm not going to worry this bone anymore, others may take a look for themselves.

All I can say is this: ignore Paul Krugman at your own risk. Call him paranoid, politically motivated, partisan, shrill, alarmist, a liar, a thief, a conspiracy to keep you poor and stupid, or whatever. Then go read Fuzzy Math. Maybe you don't like his style, but the guy IS a top-notch economist. I think most of his colleagues on the other side of the political spectrum would agree with that. And while he's not always right about everything, he IS NOT some partisan mouthpiece spouting the DNC talking points. Just keep that in mind when you read what he's got to say.

I read the Saez Excel table updating the stats through 2000 (you can find it here), but I can't find the 90% number Krugman was talking about. There is a number tracking the change in average real income (which I assume means median income), but that shows that average income rose 13% since 1973. More importantly, it shows that average real income is up 20% since 1993.

Similarly, this analysis of the CBO data, from a liberal think-tank: http://www.inequality.org/cbodata2.html#1, says that the real income of every quintile of the American population rose between 1979 and 2000. Now, these numbers are after-tax (I have no idea why), but the changes in the tax rates for the bottom 90% were not enough for their pre-tax income to have been negative. Similarly, Saez's data suggests there was only a slight decline in average real income between 1973 and 2000, so I don't think the CBO data can be reconciled with the assertion that the average income of the bottom 90% of taxpayers (or of Americans) fell between 1973 and 2000. And again the 1990s saw sharp increases (relative to the previous two decades) in real income for every quintile of the population.

I'm sure there is a table showing the performance of the 90%, but I can't seem to find it.

Good, now, what is the meaning of all this going to be when there is no such thing as "labor"? Things are headed in that direction and all this pain because we don't know yet how to deal with "no labor". How are we going to deal with it?

I find most alarming these comments from the libertarians here about public schools, which focus almost exclusively on an apparent all-powerful conspiracy of "the teachers' union."

This smacks of a ludicrously simple-minded view of public education in this country. I would wager that almost none of these libertarians actually have kids in public schools are have even set foot in one in years.

Public schools, at least in my area, are becoming extremely responsive to parental desires and to market forces. They are competing directly with private schools, and in my area at least, are matching up quite well with them.

Frankly, all this anti-public education rhetoric seems most dated.

Yeah, the knee-jerk anti-union crap gets old. How dare those teachers expect to be paid at anything but poverty wages! We've got the heirs of multi-millionairs to coddle instead!

Like Maher says, "Treat them like heros [cops, firefighters, soldiers, teachers], pay them like chumps."

"multi-millionaires" dammit!

http://businessweek.com/print/magazine/content/03_48/b3860067_mz021.htm
http://www.theonion.com/3948/news1.html

Julian,

Let's add the elimination of personal bankruptcy to your list. That makes your second suggestion more effective.

http://businessweek.com/print/magazine/content/03_48/b3860067_mz021.htm
http://www.theonion.com/3948/news1.html

http://businessweek.com/print/magazine/content/03_48/b3860067_mz021.htm
http://www.theonion.com/3948/news1.html

And that's the way it is going to be under representative democracy.

http://businessweek.com/print/magazine/content/03_48/b3860067_mz021.htm
http://www.theonion.com/3948/news1.html

James Surowiecki cites to this article:
http://www.findarticles.com/cf_dls/m1093/4_45/89871072/p1/article.jhtml?term=.)
He says that it supports his argument that the share of income attributable to labor has risen. I am having trouble understanding why that is so.
The author says

“…One representative comment asserts that ‘workers received 99% of the gains from faster productivity growth at non-financial corporations.’
“It is easy to demonstrate that this was not the case. Data from the U.S. Department of Commerce and the Labor Department show quite clearly that there was a redistribution from labor's share to capital's share (profits plus interest) during the decade--the first time this has been the case since the Vietnam War. During the rest of the post--World War II period, labor share has increased or at least held constant. The fact that the redistribution went from labor to capital--and not in the opposite direction, ….--means that capital received more than its share of the faster productivity growth in the 1990s.”

Further, the author notes that the share of income attributed to labor includes pay, bonuses, and option income to management. The numbers thus include the enormous salaries payable to the top 1% of income earners, which according to the Piketty-Saez data increased so greatly in the periods covered by the materials.

This also skews the second argument James S. makes, namely that “In 2001, labor's share was the largest it's been in the postwar period”. That is because in 2001, the profits of large corporations fell like a stone, and wages and salaries, including the monster takings of top management, didn’t fall as fast. The author of the original piece argues that he measured from the peak of one business cycle in 1988 to the peak of the next in 2000. The quote given by James S. relates to 2001, a bust year following the second peak. But the author’s figures already factor in the dip of the early 90s, so the relevance of 2001 to the point raised by the author is unclear.

Averaging in the massive salaries of top management into a serious discussion of income to the working class makes no sense. It seems to me that it makes more sense to say that if capital is willing to forego that money, we should assume that that top management took that money from the average worker. That puts top management into the position of capitalist, and that money, as a practical matter, should be treated as part of the return to capital.

Oops, I used the word Class. Sorry about the warfare.

I think it would be quite useful to measure the impact of inflation on upward mobility over time. When one must leverage oneself dramatically to obtain an education (pretty much a prerequisite now for upward mobility), even more dramatically to have a prayer at building up equity through homeownership, then pay extra to make sure the kids are adequately educated, is it any wonder that fewer people climb up the rungs?

Without the constant growth of money and credit, perhaps it would be easier for those with nothing to obtain something through merit.

There were actually two links in my post. The first one, which was from Dean Baker's response to Michael Mandel's piece, was intended to demonstrate that the corporate-profit share of GDP peaked in 1997 and fell during the productivity boom of the late 1990s. The second one had the statistic from Mandel that the labor share of income at nonfinancial corporations was higher in 2001 (85.7%) than at any other time in the postwar era.

Baker's assertions about labor's share in the 1990s are based on an eccentric definition of the 1990s and of the business cycle. He measures from 1988-2000. But what's the logic of that? We have an almost-perfect decade-long period, beginning with the trough of the recession in 1991 and ending with the trough of the recession in 2001, over which to measure economic performance. If you want to know what happened in the 1990s, that's the period to measure. And over that period, as Mandel argues, average labor share was higher than in the previous two decades. Even Baker later in that piece says that "capital share at least held constant" over the cycle.

As for excluding 2001, the logic of doing so escapes me. The billions of dollars that were lost in that year, to say nothing of the billions in write-downs that were taken, were just as real as the billions in profits earned by corporations in the late 1990s. If workers' wages continued to rise in 2001 -- as they did -- even as corporate profits vanished, surely that tells us something about the relative distribution of resources between the two.

The fact that labor includes top management does, of course, skew the data. But the data wasn't posted as part of an argument about the "working class," but about whether corporate profits rose during the late 1990s. They didn't. And it makes no sense to say that high-priced CEOs took money away from the "average worker." They took it from the average shareholder. Their wages and bonuses are not a return on capital and should not be treated as such. They are an (generally unearned) rent taken by an elite segment of the labor force.

In any case, the important second point is that even if you leave the gains of top management out, Baker says that real average wages rose briskly between 1995-2001, at the same time as productivity was booming. That has to at least complicate the idea that things have been getting steadily worse for most Americans since 1973.

Actually, Masaccio, I recall making the point to James S. in a thread a couple of weeks ago that control of capital rather than ownership of capital was nowadays probably a more relevant locus for determining the returns to capital vs. the returns to labor when comparing current statistics to those of bygone eras with all the immense structural changes that have occurred since, calling it a blurring of the agent/principal distinction. I didn't respond to his response because I worked a full shift and by the time I read his response, the thread was stale. But just to add to the point now, one of the main changes that has occurred is the much greater diffusion of ownership of capital assets, i.e. stocks and bonds, through pension funds and private household savings for retirement and tuition and the like. But these holdings are also likely to be managed by professional money managers in large aggregated funds, another agent/principal problem. Now why shouldn't the same sort of argument about the relative weakness of diffused individually smaller, but aggregately larger interests vs. the greater effectiveness of concentrated interests with insider access to decision makers that is used e.g. to explain the weakness of consumer interests vs. producer interests, whether discussing protectionism or regulatory policy, apply here?

" All I can say is this: ignore Paul Krugman at your own risk. Call him paranoid, politically motivated, partisan, shrill, alarmist, a liar, a thief, a conspiracy to keep you poor and stupid, or whatever. Then go read Fuzzy Math. Maybe you don't like his style, but the guy IS a top-notch economist."

No, he isn't. He's a very poor economist because he doesn't understand economics at its heart. If he did, he wouldn't make eggregious errors like the recent claim that news reports of Halliburton overcharging for gasoline delivered to the army in Iraq were definitely evidence of crony capitalism.

That is what Thomas Sowell calls the Physical Fallacy. In this case, now that the facts are out, we know that the army ordered Halliburton to get gasoline from Kuwait to deliver on an EMERGENCY basis to Basra. To avert rioting. That gasoline was billed at about $2.25 per gallon.

Then the army ordered more of this gasoline for other southern and central Iraqi cities from the same source. Again paying a premium for the emergency and danger of delivery.

Finally, when the army needed gas for northern Iraq, Halliburton objected to bringing it from Kuwait due to the danger and length of the trip. They found they could get it from Turkey for delivery in the north for about $1.18 per gallon.

Anyone with any experience of real world economics (and isn't that what we are actually interested in?) recognizes the "hindsight is 20-20" element in Halliburton being blamed for saving the taxpayer millions of dollars by finding gasoline in Turkey to deliver, at comparative leisure, in northern Iraq.

Anyone, excluding Krugman and most of the usual suspects here, that is.

Btw, as for the "7% Solution", I'm guessing Jim Glass is correct in his explanation (on another forum):

" Giving a quick scan Piketty and Saez, the closest I can figure is that they said the percentage share of all taxable income going to the top 10% rose by 7%. from about 26% to about 33%. Ergo [Kruman argues] the share of all taxble income going to the bottom 90% fell by 7%."

Which is entirely in character for Krugman.

James Surowiecki wrote, "I read the Saez Excel table updating the stats through 2000 (you can find it here), but I can't find the 90% number Krugman was talking about. There is a number tracking the change in average real income (which I assume means median income), but that shows that average income rose 13% since 1973. More importantly, it shows that average real income is up 20% since 1993."

It's "Table A4: Top fractiles income levels (excluding capital gains) in the United States, 1913-2000". Column P0-90 (7). Number for 1973: 28,540. Number for 26,616. That's a 6.7% decline. What it means in terms of how they constructed their time series, that I'm not so sure about.

James Surowiecki wrote, "Baker's assertions about labor's share in the 1990s are based on an eccentric definition of the 1990s and of the business cycle. He measures from 1988-2000. But what's the logic of that? We have an almost-perfect decade-long period, beginning with the trough of the recession in 1991 and ending with the trough of the recession in 2001, over which to measure economic performance."

I never see this class of numbers quoted trough-to-trough, but rather always peak-to-peak. Whether that means trough-to-trough is wrong is another matter.

Patrick Sullivan,

Your faith in NRO's accuracy is touching, but misplaced.

Stephen --

Thanks for that. When I flipped through Excel, the top of the table was cut off, and I didn't know what those numbers were. Stupid. Anyway, thanks.

There are a couple of things I'm confused by, though. First, the numbers in this table are different from those in Piketty and Saez's original paper. In that paper, the numbers for the 1970s were lower: 1973's total income was 27,492, instead of 28,540. And their numbers for the late 1990s were higher: 26,760 for 1998 rather than, as in the new tables, 25,902. The differences are not massive, but I'd like to know what happened. (The original table is at: www.nber.org/data-appendix/w8467/w8467-app.pdf)

More to the point, and this is just pure confusion, can the median income have risen from 1973 to 2000, as it has, while the average income of the bottom 90% fell? It seems like it would be possible if the bottom half of the bottom 90%'s income fell more than the top half of the bottom 90%'s income rose, but does that seem likely? And considering the poverty rate was just about the same in 2000 as it was in 1973, how do the numbers add up? I wonder if the difference is individual versus household income.

The NBER categorizes cycles both from trough to trough and from peak to peak: http://www.nber.org/cycles.html. There is no obvious theoretical difference between them. Here are a CBO study and a regional Fed study that use trough-to-trough:
http://www.cbo.gov/showdoc.cfm?index=4247&sequence=6 and www.rich.frb.org/pubs/eq/pdfs/winter2003/webb.pdf.

In any case, if you were going to go peak-to-peak, you'd go from the third quarter of 1990 (the last quarter before the recession) to the first quarter of 2001 (the last quarter before the recession). I have no idea where Baker got 1988 from.

"I asked Krugman about "between 1973 and 2000 the average real income of the bottom 90 percent of American taxpayers actually fell by 7 percent" by email. He replied that he was quoting the Piketty-Saez paper. Problem is that the 2000 numbers are in an XLS file that updates the numbers from 1997/8 or whatever to 2000...and I don't have a copy of Excel. (And it crashes my version of StarOffice.)"

Well, it took a while with the top process being openoffice.bin pegged at 96%, but openoffice 1.1 on FreeBSD 4.9 finally opened up after a minute or three, the spreadsheet. I saved in two different formats:

rcarter@quine [79] ls -lt
total 10560
-rw-r--r-- 1 rcarter www 6213632 Dec 19 20:29 PikettySaez2001.xls
-rw-r--r-- 1 rcarter www 4039680 Dec 19 20:29 PikettySaez2001.sdc
-rw-r--r-- 1 rcarter www 476524 Dec 19 20:29 PikettySaez2001.sxc

They're available here[1]:
http://quine.pinyon.org/coyoteblog/upload/

Opening up the .sxc is much much faster, and weird huh about the diskspace?

Anyway, so far I haven't seen anything to contradict JS or JG on the 7% issue, and I've applied a liberal helping of diligence. I'd like to understand better...

[1] Conditional on IMHO there being nothing wrong with this sharing, but if I am privy to any sort of dispute or contrary opinion, down and tossed into the bitbucket these files go.

I have spent far too much time today on this thread (especially since, though one might not guess it, I think the Bush tax cuts were a bad idea and repealing the estate tax is crazy). But I have to say that this kind of detailed discussion is one of the reasons why the Net is awesome, and Russell uploading the tables so everyone can look at them is another.

I talkd to Emannuel Saez himself last night about the disparity between his graphs and the Census' of changes in real income of the lower 95% of adults between 1973 and 2000 -- his graphs show their real income dropping while the US Census' show them rising. Lo and behold, it turns out that on this point Glass is entirely right and Krugman wrong: "Census income includes all government transfers such as Social security, unemployment benefits, welfare, etc. whereas the income tax estimates [which Saez used in his charts] include only market income (and hence
exclude all transfers). Transfers have grown pretty fast since the 1970s." And any fair estimate of actual income among different groups, of course, DOES include government income transfers from one group to another. So once again Krugman has very seriously overstated the case through misleading languge -- taxable income is not the same thing as actual income.

That being said, Saez does agree that the disparity between income (of all types) among the top 1% and the lower 99% is growing MUCH faster than average income in this country, and that it can certainly be shrunk by hiking taxes on the super-rich without producing more harm than good. And all these economic figures, of course, are pre-Bush -- it will be very interesting to see what the 2010 census shows. The fact remains that Krugman has once again dishonestly overstated his case.

Patrick Sullivan's account of Halliburton's activities in Iraq (which supposedly saved the government $164 million) turns out to be taken word for word from the company's own response ads in newspapers. However, the Sacramento Bee reported yesterday morning that the Pentagon itself takes another view of the company: "The Pentagon has launched a sweeping review of two pending Iraqi oil field contracts, pulling the $2 billion deals out of the normal contracting process in a highly unusual move and preventing the Army Corps of Engineers from making new awards... The new contracts to rebuild and maintain 22 oil fields in southern and northern Iraq were originally supposed to have been awarded in August...

"[A] Pentagon spokesman declined to provide details about the scope or duration of the review -- or, most important, who ordered it. He did say the review is 'in light of the report that came out from the Defense Contract Audit Agency' about Halliburton's past work in Iraq...'

Former Reagan asst. defense secretary Lawrence Korb -- who is not a kneejerk rightist -- suggests that Halliburton may perhaps really be blameless and that this review may be a response to domestic political pressure on the White House. (Ironically, the review -- by delaying the new competitive contracts for five months -- has provided Halliburton with $1 billion more in work contracts for that period.) But: " 'We have found some issues of serious concern that are worthy of immediate attention," said William H. Reed, the [DCAA's] director. He said last week that the agency was awaiting response from Kellogg, Brown & Root, the Halliburton subsidiary." So we'll see.

crayz wrote "Stephen - No, he has a good point. What he is saying is you are breaking the citizenry up into quarters:
1) bottom 25% of wage earners
2) lower middle 25% of wage earners
3) upper middle 25% of wage earners
4) upper 25% of wage earners"

You're right--my bad. Though the data are only *consistent* with total mixing.

I'm curious about what impact the bursting of the stock-market bubble and the (relative) reduction in the number of stock options companies are handing out to CEOs will have on these numbers. (Since Saez's data runs only through 2000, we don't know.) A significant chunk of the massive income gains of the very rich in the 1990s came from options, and it seems unlikely that there will be similar gains to be enjoyed in this decade, especially if, as seems possible, companies have to expense options. That would, I think, put a significant dent (well, if you can dent multi-millionaires) in the income of the top 1%.

-----------quote--------------
Patrick Sullivan,

Your faith in NRO's accuracy is touching, but misplaced.

Posted by Bernard Yomtov at December 19, 2003 05:00 PM
---------endquote---------

Apparently my faith is nothing compared to Bernard's in his own credibility. Did he notice that I cited my own experience in government contracting to bolster Halliburton's side of the story?

And I'm pleased that Bruce Moomaw was honest enough to eventually get around to admitting: "Halliburton may perhaps really be blameless and that this review may be a response to domestic political pressure"

But why introduce the red herring: "The Pentagon has launched a sweeping review of two pending Iraqi oil field contracts..."?

I was specifically talking about a claim in two Krugman columns. A claim which appears to be sheer baloney (or wishful thinking).

Paul Krugman seems to elicit radical defensiveness at times. For instance, it was enscribed above:

He's a very poor economist because he doesn't understand economics at its heart.

We really need to get in touch with the J. B. Clark medal award people and explain why they should revoke Krugman's. Maybe.

"I talkd to Emannuel Saez himself last night ... Lo and behold, it turns out that on this point Glass is entirely right and Krugman wrong... Krugman has once again dishonestly overstated his case."

Hey, folks, you've gotta learn that you can trust an honest lawyer over data-twisting, dissembling economists any time. ;-)

(No, the revenue lost to the Bush tax cuts was not "more than enough to fund Social Security and Medicare for the next 75 years" ... and the Internet Tax Moratorium had nothing at all to do with blocking sales taxes on retail sales, so John McCain was not "confused or a panderer" [another overdue apology owed added to the long list there] etc. and so on)

Now that this has been straightened out, let's see if Prof. K. will be fair and balanced enough in some future presentation to compare his income gap to the "consumption gap". E.g., by mentioning the BLS numbers that show an "income gap" between the highest and lowest quintiles of about 14 to 1 -- the highest quintile with 14 times the income of the lowest -- but a "consumption gap" between highest and lowest income quintiles of only about 4 to 1 ... 70% smaller. With the consumption gap *not* increasing over time as his income gap has been.

I mean, just *mention* it as something very relevant to the story. After that, rationalize or hand-wave it away, if he wants. After all, if he is really going to call himself "the lonely voice of truth" as he does, he ought to be able to bravely mention to his constituency the *bad news* that there is some reason to believe inequality might not be as bad as they are wont to believe.

Maybe then he'd also get around to mentioning the notion that in bringing about the "great contraction" of income inequality during the 1930s-40s, New Deal policies might possibly have been supplemented by the Great Depression and World War II. ;-)

And remember, all, that public dissembling is a field just like any other, you want a professional -- a lawyer -- to separate the wheat from the chaff for you. Trust your barrister.

A consumption gap of 4 to 1, lest we forget, is still goddamn big. And if the GOP monopoly of government -- which, lest we also forget, we haven't had since the halcyon days of the Depression, except for two brief years in the Fifties with a relatively moderate GOP President -- continues much longer, it's already abundantly clear that both the income gap and the consumption gap are going to get a lot bigger, thanks to the coming wholesale chokeoff of economic transfers from the more wealthy to the less wealthy once the GOP can no longer ignore the exploding deficit. Which is most of Krugman's point and (as far as I can tell) all of Delong's point.

Of course, this also means that at some point the chickens will come flapping home and start dropping political eggs on the heads of the Republicans -- as would already have happened by now if the Democrats weren't faced with the difficult task of quickly reversing their image as the Dove Party which they had been working feverishly to cement since 1969. As in the Grant and Harding Administrations, I imagine a lot of the people associated with the current GOP regime are well aware of this, and are simply trying to grab everything they can get before the political roof falls in in 2008.

Footnote: it is, after all, Piketty and Saez themselves (as quoted by Delong) who say that the coming further growth of income inequality under the GOP will NOT increase the economic growth rate; it will simply increase the extent to which the wealthy are rewarded even if they happen to be incompetent boobs.

There is, though, a paradox in using Piketty and Saez's work to argue for the emergence of a rentier class. Although Piketty does predict -- and it is a prediction, not a statement based on concrete data -- that in a "few decades" bad policies could lead to the incompetent boobs running everything, what the original 1998 paper documented was that far more of the income of the top 1% than ever before was coming from wages and salaries rather than capital income. Now, that doesn't mean the high salaries were justified -- CEO compensation is obviously absurdly high, in no small part because compensation for CEOs has been mostly set by other CEOs (serving as corporate directors). But the top 1% in the 1990s was not, for the most part, a group of "wealthy but untalented children." (In fact, there is (or at least was in the 1980s, and given the importance of options in the 1990s I'd be surprised if this had changed) tremendous mobility into and out of the top 1%.) Piketty and Saez end their paper by writing, "In the United States . . . the coupon-clipping rentiers have been overtaken by the working rich." That might be at least worth mentioning in a piece that uses their work as the foundation of a theory about a new rentier class.