Krugman on income mobility

Krugman wrote:

True, Cowen isn’t advocating a complete caste society — but it’s actually not clear why, since he is suggesting that we’ll be happier as a society if people stay comfortably in the class into which they were born.

He compares my position to this poem:

The rich man in his castle,
The poor man at his gate,
God made them high and lowly,
And ordered their estate.

Uh, no.  What I wrote was that for fixed levels of income there was no gain per se from having a higher rate of mobility churn, if not accompanied by higher standards of living.  I also wrote that the best way to get useful upward mobility was to have a high rate of economic growth and spread across many income classes.  Think of it as the difference between positive-sum notions and metrics of mobility vs. zero- or negative-sum notions and metrics.  Read my first point:

If the general standard of living is rising (and I am more than willing to admit problems in this area for the United States), mobility takes care of itself over time.  I find it more useful to focus on slow growth, if indeed that is the case.  Just look at income growth for non-wealthy families and that is more useful than all the mobility measures put together.

Or read my follow-up post:

I see two big and very real problems: slow income growth for many income classes and a problem with excessively high returns to finance at the very top.

Let’s put it this way.  Paul Krugman is a great economist.  But of all the people in my RSS feed, in terms of his quality and skill as a reader, he is not in the top 90 percent.


Krugman was probably confused because this is the series of posts where you discussed "mood affiliation".

You could have also criticised Krugman for NOT acknowledging you also said in the same post: "Sweden deserves more praise,..."

Though, I think it would have been more damaging to you for him to have said you praised Sweden. seems like you've been more cross lately with your interlocutors, especially since the Crooked Timber exchange. It could be, of course, that your interlocutors are getting worse at reading, and so are misinterpreting you, which is making you cross. But it could also be that you're cross (for whatever reason) which is making you curt and enigmatic.

Put differently, I don't see the value in using writing as a test of who can read it well, any more than doing the same with math in an economics paper. Some difficult ideas require difficult prose and/or math to explain, certainly, but all the more reason to focus on clarity. Are you really trying as hard as you can to be understood?

No, Krugman was arguing against the point that he wish TC had made.

PK sees the world the way he chooses to see it. In other words, he has no conscience (and no science).

I think Krugman meant this, which you did write:

More upward — and thus downward — relative mobility probably means less aggregate happiness, due to habit formation and frame of reference effects.

The post says quite clearly the way to produce gains is to have economic growth produce new opportunities, and rising incomes for many different classes, not to play "Trading Places" for its own sake. That is hardly advocating a caste society or anything close to it.

I'm not sure that I buy the argument that intergenerational mobility is not inherently a good thing, regardless of growth. I think most people who debate about economic inequality and inequality of opportunity are worried about the fact that the game is becoming "rigged" for those in power.

In my opinion it seems that even with zero growth there would be still be a negative for one to be locked into the socioeconomic class of their parents regardless of ability, or the opportunity to grow one's ability. Which is exactly the virtue of "`Trading Places' for its own sake".

If your argument against this is that those kids of rich parents will be less happy because they can't achieve as much, which is what "relative mobility probably means less aggregate happiness" implies to me then that is very offensive.

The inherent virtue of social mobility and overall economic growth aren't exclusive of each other and I think some economists would argue even that stronger opportunity of mobility would encourage competitive behavior increase growth.

I think you are equating opportunity of mobility with actual mobility

A lot of people have trouble grasping the possibility that high opportunity of mobility doe not necessarily result in actual mobility. In fact, high opportunity of mobility results in low actual mobility over generations due to assortative mating, which create de facto castes.

"high opportunity of mobility results in low actual mobility over generations due to assortative mating"

So if what Miley says is true; is the converse valid: Low opportunity of mobility results in high actual mobility?

Rahul-- if you are being serious, nope--as high opportunity of mobility is a prerequisite for high actual mobility.


But is it logically consistent for "high opportunity to be a prerequisite for high actual mobility" and yet also simultaneously for "high opportunity of mobility to result in low actual mobility".

'But is it logically consistent for “high opportunity to be a prerequisite for high actual mobility” and yet also simultaneously for “high opportunity of mobility to result in low actual mobility”.'

The logic does not follow for the reason I mentioned. Additionally, the key and also necessary ingredient here is assortative mating.

"But is it logically consistent for “high opportunity to be a prerequisite for high actual mobility” and yet also simultaneously for “high opportunity of mobility to result in low actual mobility”."

Yes. I reckon that's a logically consistent result for all necessary but not sufficient conditions.

Was about to make a long comment but Mike P basically says the same thing above.

A failure to see lack of intergenerational mobility as a bad thing - even all by itself - reveals a tin ear for the harms of a caste system (whether facto or de facto). Krugman's read is uncharitable but it is not unreasonable.

Yes and Krugman does not say you are advocating for a caste system. I guess it is you who is having the reading comprehension problem. Or is it merely the fallacy of mood affiliation? LOL!

I don't read Krugman because it tends to be not especially interesting, but he didn't say that you said it, he just gave you a very uncharitable read.

I do want to say, as a long time reader, it's extremely disappointing that you keep posting these "I am misread" pieces. If you are so misread maybe you should be more clear?

When you say "more upward ... relative mobility probably means less aggregate happiness...", how exactly do you want that read? It seems perfectly clear, and maybe your writing is the problem?

I agree this post is disappointing. However, it's also clear Krugman deliberately misread him. Asking Tyler to be clearer seems silly.

A better Cowen response would be to turn this misreadibg into an opportunity for smart analysis.

This exactly. Tyler, if I were you, I'd want to take those words back. The fact that they were twisted by knaves to make a trap for fools tells you everything you need to know about what Krugman has become.

He calls you un-American, you call him a great economist. And so it goes.

Worst ellipsis in commenting history. What part of "--and thus downward--" was worthy of excision, since the quote makes no sense without it?

Nongrowth churn must have equal parts up and down. Tyler's eminently reasonable point is that the down probably subtracts more than the up adds. Care to reconsider your point at all?

In terms of Andrew Smith's quality and skill as a reader, he is not in the top 99 percent.

Yes, that was the worst in history. OF the UNIVERSE! The very worst.

Tyler’s eminently reasonable point is that the down probably subtracts more than the up add

I can think of a million counter examples to this. If you took everyone who makes, say, a million dollars and put them to, say, $900,000, you could take a lot of people from $10K to $20K. Or whatever. Devil's in the details. The marginal utility of $1 is a lot to very poor people.

I wasn't trying to make the point you seem to think I want to make. Look, I'm not a communist. But if what Tyler was trying to say is "Nongrowth churn must have equal parts up and down" , which and I both agree is what it sounds like he said, then he was not being especially reasonable.

Not defending Krugman or anything, that guy sucks.

I should have said "Everyone's parents made a million..."

Andrew Smith,

You're just further demonstrating your abysmal reading skills. Tyler is talking about *mobility*, not inequality. Specifically, mobility between different income quintiles or deciles, as he clearly stated. By definition, there must be equal parts up and down. For each individual who moves up to a higher group, another individual must move down to a lower one.

How much of immobility is due to “inherited talent plus diminishing role for random circumstance”? Is not this cause of immobility very different — both practically and morally — from such factors as discrimination, bad schools, occupational licensing, etc.?

Are you serious?

Andrew Smith,

What? I didn't write the text you quote. I don't know where you got it from.

It's from the post Tyler (and Krugman, apparently) linked to. Tyler wrote it.

I think it only makes Tyler's quote more risible if you think relative declines in status are a cause of unhappiness. Or will my children be happy if society is more equal but they remain in the top quintile?

"Specifically, mobility between different income quintiles or deciles, as he clearly stated. By definition, there must be equal parts up and down. For each individual who moves up to a higher group, another individual must move down to a lower one."

Only if you're holding the populations of the strata constant. But if you do that, then you're almost certain to see changes in the boundaries of each strata. Conversely, if you hold the boundaries fixed (income levels), then the population size of each strata will change. In either case mobility will lead to changes in inequality - it just depends on whether you want to measure that inequality in the population size of each strata, or in the boundaries of equally sized populations.

If you try to hold the income boundaries fixed, then you don't get 1 for 1 changes in rank. If you try to hold the populations fixed, forcing 1 for 1 changes in rank, then 1 for 1 changes in rank produce changes in inequality between ranks.

Income mobility and income inequality can't be decoupled.

Only if you’re holding the populations of the strata constant.

The change at issue here is mobility, not population change or inequality. Andrew Smith's claim is about a change in inequality, not mobility. He either doesn't understand what Tyler wrote or he is intentionally misrepresenting him.

I think it only makes Tyler’s quote more risible if you think relative declines in status are a cause of unhappiness.

If you think you have evidence that relative declines in status are not a cause of unhappiness, produce it.

I'm not the one who said downward movements in status cause more unhappiness than upward movements in status, that's what you think Tyler said. I don't have to prove that's not true, it's clearly ridiculous on its face value.

I can understand that downward intergenerational movements in wealth or income can cause unhappiness, but you guys are explicitly saying that's not what Tyler meant.

You can't have this both ways. Either you have something completely ridiculous, ie, "I'm unhappy because my dad was in the top quintile but I'm in the 2nd quintile". Or you have something slightly indefensible "I'm unhappy because my dad made more money than I do." The second one seems to be implyed by the "inherited talent" quote from the original piece.

Maybe I do suck at reading (I think you are clearly confused by what I said, so maybe you also suck at reading?). Either way, the fact that it can be interpreted either way proves my original point: If you don't want to be misunderstood, write something coherent.

"The change at issue here is mobility, not population change or inequality. Andrew Smith’s claim is about a change in inequality, not mobility. He either doesn’t understand what Tyler wrote or he is intentionally misrepresenting him."

You are really proving the point that you are a poor reader.

Relative changes in income position (ordinal) are driven by absolute changes in income (scalar). Since people aren't just swapping ranks (we don't have a 52 card deck among which people just trade cards), but rather are gaining or losing income, by which we generate ranks, changes in rank (relative mobility) are almost always symptomatic of changes in inequality. Only a small set of the ways in which incomes can move up and down in a zero-sum fashion produce no change in inequality.

This, exactly! If Tyler really means to say that increasing the living standard is the real goal, and mobility is perhaps not a useful metric for social science, then he should just state that. Say "I don't believe that income mobility is a useful measurement because increasing utility is best done through increasing the standard of living". Now no one will misunderstand you. Now where is my PhD?

Andrew Smith,

I agree that Tyler's post is unclear, but if he had said it in plain English, he would have invited more pointed criticism because you would have recognized the refrain.

Let me translate Tyler's post into plain English.

Here goes:

Tyler: "...for fixed levels of income there was no gain per se from having a higher rate of mobility churn, if not accompanied by higher standards of living. ...the best way to get useful upward mobility was to have a high rate of economic growth and spread across many income classes. Think of it as the difference between positive-sum notions and metrics of mobility vs. zero- or negative-sum notions and metrics."

English Transaltion: If we grow the pie, everyone will have a bigger slice and mobility won't matter.

Whoopdee Do.

The problem is that the pie is not growing. People question whether their portion of the pie has been diminished by tax cuts (or elimination of estate taxes) for those who have bigger slices. They are also worried that government programs they rely upon to assist them in mobility are being cut, or that cuts in social insurance will further increase their risks of a smaller share of the pie.

That being said, Krugman did misrepresent Tyler. Tyler is not asking for a permanent caste system. But, he is not addressing the real issue either: that the pie is declining, the shares are being allocated to those with pie on their face with generous tax cuts, and that programs that assist in social mobility or reduce income insecurity are being curtailed.

Some would argue, as does Stiglitz, that increasing mobility of those below actually increases the pie. I would like to see some discussion of Stiglitz's position, not Krugmans, as much as everyone here like to talk about Krugman.

Pie metaphors are yummy. And delicious. Because that's what the economy is. A pie. No one knows where it comes from. It just shows up one morning, cooling on the windowsill. But rich people set their alarm clocks extra early, so they sneak out and get an extra-big slice.

It's like the language about increases in income "going to" the rich. Where did they come from? No one knows. We only know for sure that the rich stole them.

Tyler writes: "I see two big and very real problems: slow income growth for many income classes and a problem with excessively high returns to finance at the very top."

Why is there such a concern with slow income growth? Why can't people just be content and happy to have a job that they like that pays okay? Why does it always have to be "more, more, more"? It's not good to be so greedy. It seems that too many people are excessively selfish; they should be content being alive, enjoying life, hanging out with friends, working and earning a salary. I find this whole "keeping up with the Joneses" thing to be low-class. It's not healthy for our society for people to be so concerned about money and income. It's one of the reasons why I come into the office every Sunday to do work even though my boss isn't asking me to; I also don't bill overtime; I just enjoy working and don't think much about money.

Do any MR commenters have any thoughtful answers to some of my questions?

Yeah, who would want their life to get better? Seems totally crazy

Income growth = happy. Poor people are sad. That's where blues music comes from, poor, sad people. The average Roman citizen of the second century was a pauper compared to contemporary Americans and he was very unhappy. So unhappy he often stabbed the current emperor to death. Nineteenth century Eskimos, who didn't even own a steel knife and had never seen money of any kind literally wept the day away. And so it is even now for residents of the New Guinea highlands, who don't have 4G service for their magic phones. They never laugh or even tell jokes. Admit it, you've never seen a stand-up comic from New Guinea.

I think there is a lot more truth to your jokes than you believe. Maybe if we go all the way back to hunter-gatherer societies people really were just as happy, but returning to a hunter-gatherer lifestyle is not on the table.

Also, there is the question of HOW LONG you get to enjoy your life


They don't, they want their lives to be better RELATIVE to other people. Absolute income does not correlate with happiness at all.

Absolutely wrong

Andy Harless got it right on Twitter: "Sometimes Krugman really annoys me. His latest counterargument to Tyler Cowen is to make an obnoxious paraphrase & then say it's un-American"

You really seem irked. Your "Uh, no," sounds like the rejoinder of a teenage girl. And your ultimate resort to intelligence percentiles, as determined by you, is the refuge of some Dungeons & Dragons playing teenage boy. Not a good look, Tyler. Try to be more of an academic.

Intelligence percentiles?

Yes, intelligence percentiles. And we all know the real point of the jibe about the reading comprehension of Krugman relative to the rest of Tyler's RSS subscribers was to let us know that Krugman subscribes to Tyler. As if it were some sort of asymmetric relationship. Maybe Tyler should work on getting his own Nobel as a way of overcoming the feelings of inadequacy.

And we all know the real point of the jibe about the reading comprehension of Krugman relative to the rest of Tyler’s RSS subscribers was to let us know that Krugman subscribes to Tyler.

What? That's not what an RSS reader is....

This was entertaining.

You must be using the royal "we all".

"Paul Krugman is a great economist." Is that 'great' in the common American sense of all right, OK, so-so, competent, ....?

It clearly isn't 'great' in the sense of Smith, Ricardo, Marx, Keynes... is it?

Perhaps Marx...

Ya, I've always thought that people look too much into things like income inequality and relative income mobility.

Are living standards improving across the board in an absolute manner?

That is what really matters to me.

I have read that France, Italy, Spain, etc all have both lower income inequality and higher income mobility than the USA.

But, how many people in the USA would trade our huge advantage in average income, disposable income, and simple material standard of living for more income mobility or less income inequality?

Agreed -- absolute is all that matters. I think the income inequality issue is used as a cudgel by the left to bash capitalism and justify increased government in the economy and higher taxes. It's a non-issue.

So why does Tyler care so much about stagnant median income since the 70s?

If median income is rising but average income is rising faster, income inequality is likely rising. I think that is OK.

However, if median income is flat but average income is rising, then that is probably an issue.

But, I think that median income is a MUCH more important measure than is any measure of income inequality.

I think that is a false choice. You can have both.

Sure you can.

But, some policies that are aimed at lowering inequality do tend to depress growth of average income.

For example, a relatively small, homogeneous country with communitarian could probably have relatively low inequality and high average income because it wouldn't require policies that harm growth to achieve inequality.

But, take a society like the USA. We have a large, culturally diverse country with individualistic values. The only way we could really achieve less inequality is by pursuing certain public policies. This includes things like high MTRs on higher incomes, very powerful unions, more labor regulations in general, probably more protectionism, etc.

These policies could reduce inequality but would almost certainly harm average income growth. I don't see that as a good trade-off.

for fixed levels of income there was no gain per se from having a higher rate of mobility churn
Right. So you didn't advocate a caste society, you just don't see a downside.

I think a caste society implies a lack of opportunity of mobility, not just a lack of actual mobility. I.e. there is something about the caste that actually makes it very difficult for you to leave even if you are very talented. That's a whole separate issue from inequality per se. Obviously unjust social restrictions would make people unhappy.

This is what the civil rights movement was meant to address. The states that were on the wrong side of that movement have switched from solidly Democratic to solidly Republican. Therefore many are suspicious of Republican motives when they say they want to focus on equality of opportunity rather than outcomes.

I know this is not a Republican vs. Democrat argument per se, but it's a more concrete construct than mood affiliation.

Why is this so hard for people to understand? Cowen's argument is that if society as a whole is static, there is no benefit from changing the roles. From that he argues that the primary focus should be on making society as whole non static. This is not an argument for a caste system, it is an argument that the absence of a caste system does not fix the problems inherent in a static society.

What seems to be throwing people off the track is an incidental reference to happiness studies that would seem to show that in a Static society, net happiness would actually decrease if people kept changing roles. I personally think that this is silly argument as I don't have much use for happiness studies. However, since happiness studies are often used to argue that inequality is a bad thing regardless of absolute factors, I think Cowen sees this as a "live by the sword, die by sword" type of argument. In other words, if you are going to use happiness studies as a reason for why inequality is bad, you should be consistent about where such studies lead you.

" if society as a whole is static, there is no benefit from changing the roles."

Heh... I think that for Krugman, as well as Mike P above, there is definitely a net benefit from changing roles. If I'd like to see the mighty pegged down a notch, social mobility is a net good. I wouldn't expect them to formulate it quite the same way, but it seems to me that that's the driving force behind all the harping about the 'lack of social mobility in the US'. The deeper reason, of course, it that they 'know' that the rich people are evil (that's why they need to be pulled down from their pedestal).

As far as Mike P goes, I don't have a problem with him as he seemed to get Cowen's point. I don't really have problem with people who want to argue with Cowen's point. It is just that so many people just don't seem to get or are deliberately miss-representing it.

The same people who complain that Cowen is not clear enough would be the first to complain that he was too wordy if he wrote a long post trying to make everything clear to the obtuse (or impossible to twist for the malicious). I guess it should not bother me. But the best thing about Marginal Revolution use to be the comment section as it used to seem like everyone seemed to be smart enough to get and also willing to argue in good faith. Now that does not seem to be the case anymore.

Must be too many beautiful woman showed up.

When was the last time someone complained about Tyler being too wordy?

Bill, in this thread, Rahul.


Given a fixed amount of wealth, relocation won't do anything but redistribution will. Both provoke upward and downward mobility.

@Ape Man
Cowen’s argument is that if society as a whole is static, there is no benefit from changing the roles.

I think the crucial point here is what's your system boundary. A static society with changing roles may result in no net benefit to society but if you include fairness to the constituent individuals then the "changing the roles" part matters a lot.

I am not really inclined to defend Cowen's point. I hate happiness studies. As a general rule, I think they are on very shaky ground methodologically and even if you could scientifically measure "Happiness" I think it is a very poor metric to use as guide.

So if you throw happiness studies out the window and if you posit a static society with no net change in economic capacity, then arguing about "mobility" in such a society is really only matter of arguing about who hears the voice of God most clearly. Person A says "I went up on to the mountain and God told me that fairness is the most important thing is a static society" and Person B says "I went out into the desert and God told me that a stable social order and clearly defined property rights are most important thing in a static society."

And if you want to make things even more needlessly complicated, you can start arguing about the composition of the static society. Is everyone in the static society truly equal in abilities and character? Maybe roles in the static order are a reflection of underlying genetic realities. But how can you be sure the genetic realities are really static? But really, since the entire static society is hypothetical, what is the point of arguing about it? It can be whatever you want.

Honestly, I find the substance of the argument to be pretty pointless. I would not have even gotten involved if it where not for the fact that I was honestly surprised by how many people jumped up in the comment section to say that Krugman was to right about what Cowen said. I think Cowen quite clearly had every right to be aggrieved about how Krugman twisted what he said. I don't think it was really profitable to complain about it, but sometimes whining is good for the soul even if does not accomplish much.

And I guess it was under that philosophy that I made my first comment.

To put it another way, don't confuse changing who occupies the roles with changing the roles.

You cut off the important part of the sentence
"if not accompanied by higher standards of living. "

I see a lot of purposeful misinterpretation here and especially by Krugman.

I agree (very strongly) with Cowen. One reason that a caste society is bad is because it purposefully keeps the pie from getting bigger, for the benefit a of few. Mobility is valued because it allows the pie to grow further and make everyone better. It's not about playing musical chairs; shifting deprivations onto others for it's own sake. Why a progressive would think that this sort of musical chairs would have value by itself, escapes me.

Imagine we were trapped in an insular world where the pie was fixed in size and there was no way to increase its size.

To me it still seems a valid goal to strive for churn. Wonder what people think. Should a fixed-output society necessarily be kept zero mobility?

While "shifting deprivations onto others" people forget that you're also simultaneously showering prosperity on to still others.

The question, though, is if it's just the pie that is fixed. Focusing on income mobility as relocation and swapping places implies that roles are also fixed. But you can get mobility not just through relocation, but also through repartition.

Paul Krugman *was* a skilled economist (I reserve the word "great" for, well, the great), when he did his seminal work on economic geography. Now he is a self-appointed social critic, not an economist at all, unless he is doing research that he doesn't write about in the Times.

How'd you classify Tyler?

Compared to Krugman? Honest and sincere are the words I would use for Cowen.

"But of all the people in my RSS feed, in terms of his quality and skill as a reader, Krugman is not in the top 90 percent."

it's worse than that. Krugman is not simply incompetent when he reads -- what matters more is what Krugman famously and proudly _refuses_ to read -- the great economists publishing before Krugman was in grad school who presented any of their work in prose.

Famously and notoriously, Krugman tells us he never read Bertil Ohlin's Interregional and International Trade until not so long ago he was invited to present a keynote lecture at a conference on Ohlin, see

And Krugman famously and proudly tells us he is unable to read ANY economic written in prose, esp. any economics written in prose and published before he entered grad school.

And just everything Krugman has written on every leading economic scientist writing before 1970 tells us that Krugman either hasn't read it, or was an incompetent reader when he did attempt to tackle the text.

Perhaps Krugman has a reading disability, but even if he doesn't, Krugman certainly has some sort of moral disability which allows him to happily misreport and mischaractize the scientific work of others with a giant smile on his face and a crazy gleam in his eyes.

" Krugman famously and proudly _refuses_ to read — the great economists publishing before Krugman was in grad school who presented any of their work in prose."

There are/were economists who published in verse?

Math is the verse of economists. Its pretty, its concise, and it gives an illusion of intellectual sophistication even when the foundations are just as shaking as when written out in prose.

It would generous to say Krugman has a problem with reading comprehension.

We are somewhere between "then they laugh at you" and "then they fight you."

And if there is zero-sum churn, the meritocracy meter was broken before even if it is right now. And (all relevant things being equal to make this work) if there is more churn tomorrow that means it is still broken today. I'm not sure why this is hard or controversial.

As I stated at the time, this is easily visualized by a diffusion/brownian motion model. Holding constant some things, it's math. It just IS. I think a turning of the back on positive economics is what trips some people up.

I think it is useful to point out that having stable boys and princes swapping places does not really make very much sense as a goal. The problem, though, is with having stable boys and princes, not with who is the stable boy and who is the prince. If Henri Albert de la Marche III is shoveling manure, that's a nice bit of schadenfreude, but it's still Schaden. The left concern with mobility can't be divorced from the left concern for equality. If mobility is nothing other than trading places style churn, then yes it doesn't do much beyond allow one to laugh at declassed rich folk.

The sought after variety of mobility is one that compresses relative inequality of position and leads to convergence. And that means, yes, increased absolute upward movement for the poor. But it also typically means increased absolute downward movement for the rich, insofar as the rich are too rich - e.g. profligate, vainglorious, ostentatious, domineeringly powerful, &etc relative to everyone else. That's not swapping positions, it's reconstituting them. A world with neither stable boys nor princes (to the glue factory with you, Rafalca). And if you can't get rid of the stable boys, well you very well can get rid of the princes (and let the stable boys eat their victuals - it's "eat the rich," not "trade places with the rich," after all).

"Paul Krugman is a great economist. But of all the people in my RSS feed, in terms of his quality and skill as a reader, he is not in the top 90 percent." Tyler, the fact that you acknowledge the greatness of an economist who distorts your views shows you are a very balanced thinker. I look forward to a comprehensive book on economic thought authored by you. It will have a libertarian slant but you will present the ideas of those you disapprove of in a fair manner

Has any one noticed a hilarious feature of Paul Krugman's attacks on Tyler?

Two examples:

* Brad DeLong revisits Tyler Cowen’s remarkable defense of lack of social mobility.
* John Quiggin sees Tyler Cowen making the transition from #2 to #3, and does a systematic demolition job.

Evidently the Sultan of Sneer is in a panic that someone might think that he is slumming by visiting Marginal Revolution, so he is always careful to signal that he learned what Tyler had to say through other bloggers!.

See above: We are somewhere between “then they laugh at you” and “then they fight you.”

I utterly lack the gene for pussy footing. I wish it was as obvious to everyone else (and it probably will be in a few years more, I'm learning that the recognition process is painfully long) but everything about PK is calculated to lower the status as those he perceives as his opposition.

Imagine oil and water. The oil is on top and you want it there because it won the density meritocracy fair and square. It's not that way because it is a "caste," because you like the caste, or you want it that way. It is that way because of the fundamental driving forces between the molecules. Maybe there are a few water errant drops still on top of the oil. You are impatient. Now you stir the shit out of it. What have you done? For the time being, you've made it much worse. Now, if that's not the kind of system we are talking about, then show THAT.



Professor Krugman is head and shoulders above you in the quality of his analysis, writing and humanity.

You need to get over thinking you are somehow his equal.

That's all you got? What a waste of a comment.

I think it was a joke...

That is what I had to say that had not already been said, and I said it in all seriousness.

Tyler appears to consider himself some sort of libertarian iconoclast - he's not half clever enough to pull it off.

What a waste of a comment.

Well, I do find Krugman's humanity to be quite high quality. But I have to disagree with 2/3.

Tyler should stick to listing what he's reading from the free books he gets. His theories are so hard to defend, he has to use language complications that Wittgenstine warned about to try to defend them. It's clear that Krugman's ability to put things in terms that the masses can understand gets under the skin of Tyler and his Chicago School types.

Or, there is the possibility of a non-zero probability that life is a little more complicated than "you are either with or against Paul Krugman."

Seems there is some sort of misunderstanding between the castes of economists, those in the private sector and those in the public. Is it fair to say then that Krugman and Cowen are having a "Passage to India" moment?

Bad writing. Worse reading.

Can't both sides just take a Mulligan on this one?

I wonder why this invective against Tyler on the part of some readers when he is is being polite. Joe Smith says Krugman is a greater economist than Tyler. Tyler will no doubt would agree and no where does imply that he is Krugman's equal. What he is saying is that Krugman distorted his views.And being a greater economist does not give you the right to do this.

"why this invective against Tyler"

Because Tyler (1) tends to have a mean edge to his positions as he shills for the libertarians; and (2) on this one Professor Krugman is right and Tyler is wrong.

I'm guessing this comment came in with an IP address in the New Jersey area? "mean edge", pot, kettle, ...?

Eric - It did not come from New Jersey or anywhere on the Eastern seaboard.

"Cowen’s argument is that if society as a whole is static, there is no benefit from changing the roles."

What happened to the meritocracy? If your understanding of Cowen is correct, he is indeed saying there is no problem with an ossified economic class system that is mostly hereditary, i.e., a caste system.

Rags to riches and back again in three generations - a society where that is possible is the epitome of capitalism. We put a safety net so that the rags are metaphorical only (and the riches are not so Croesusian) and we are set.

Again, see above. Just increasing churn doesn't mean you are enforcing meritocracy. You could just as easily be breaking it.

The simplest way to increase Brownian motion might be to increase temperature. This increases randomness. Assuming some level of meritocracy in the beginning, randomness doesn't necessarily improve on that. It depends, but JUST the randomness (and it could be much worse than random of course) isn't what improves the distribution. One has to assume, as a liberal might, that the current meritocracy is so broken that even randomness would be an improvement. Assume a caste, anyone disagreeing with you must be pro-caste.

Some randomness superimposed on top of meritocracy might help by preventing people from being trapped in local maxima. Sometimes a small anti-meritocratic move might be in the best interests of achieving a larger meritocratic target.

Samuel Bowles' calculations (inspired by the arguments in "The Bell Curve") suggest that churn is below what it should be even given strong assortative mating and a strong genetic component of IQ.

So the fact that we don't observe the predicted level of churn means either that nobody (including Charles Murray, Rushton, etc.) has a good enough model of the relationship between genes, IQ and earnings or that America isn't much of a meritocracy. Or both.

The point isn't that random churning is a good thing. Rather, the point is that there is fairly good evidence that non-random churning that is predicted to be happening isn't and this may well be a socially undesirable state of affairs.

" Samuel Bowles’ calculations (inspired by the arguments in “The Bell Curve”) suggest that churn is below what it should be even given strong assortative mating and a strong genetic component of IQ."

Thinking of how every adopted kid I knew wound up (all raised in a handful of wealthy, high IQ suburbs, every one is now on a path to be permanently working class), I can't decide whether that surprises me or not.

There is a difference between opportunity for mobility and realized mobility

If you take a static society as a given, then in theory nothing you can do will change this society from being static. So what is the point of being meritocracy if it does not improve society any more than being a "caste based system"?

I think the problem you are having is that you are failing to understand the difference between saying "Given A, b is true" and saying "A is true". By positing an absurd world where change is not possible and then arguing from that world Cowen's was trying to prove the central problem is to keep society as a whole from stagnating as opposed to making peoples relative position in society keeps changing. But that is not the same as saying you can have an innovative society while still having fixed roles for individuals.

Anyone who has read Cowen writings knows he favors a meritocracy because it allows for growth and benefits society as a whole. The whole point of his argument was that in the absence of growth/innovation, what does a meritocracy have over a caste system? Cowen's question is "If everything is static, is it really more fair for me to do better than you if I was born with more ability than you as opposed to just being born in the right caste?" In Cowen's eyes, the only reason meritocracy is better than a cast system is if everyone ultimately benefits. You can argue that point if you want to. But you should not try to claim that Cowen does not believe in a meritocracy. He obviously does as even a cursory reading of his writings would show.

To see a practical outworking of Cowen's philosophy in the real world, consider his support for immigration. Cowen acknowledges that immigration increases inequality. However, he still supports immigration because he feels that it increases innovation and growth in society as a whole. At the same time, Cowen opposes regulations that he feels makes certain classes of people immune from competition and thus increase the static nature of society as whole (and making certain people wealthy without "earning it"). In both cases Cowen's focus is on keeping society from being static. In one case this focus would increase inequality where as in another case this focus might reduce it.

That's what you get for writing so cryptically TC :P

"3. For a given level of income, if some are moving up others are moving down. Do you take theories of wage rigidity seriously? If so, you might favor less relative mobility, other things remaining equal. More upward — and thus downward — relative mobility probably means less aggregate happiness, due to habit formation and frame of reference effects."

I believe that is what you wrote. It explicitly says that the aggregate happiness everyone will be reduced by rich people falling from on high (habit formation, frame of reference effect) than poor people rising up ... i.e. aggregate happiness will be at a higher level if "people stay comfortably in the class into which they were born."

This is the sentence the left is on about:

"More upward — and thus downward — relative mobility probably means less aggregate happiness, due to habit formation and frame of reference effects."

Krugman and the rest of left read this sentence perfectly. To make it more concise: More downward relative mobility means a worse world due to rich people being used to their comfortable lives.

In the original comment, does Tyler endorse this view? It's not clear to me, but the argument is a whole other conversation.

He's saying that this is a view some people may take. In particular, he may be thinking of a fashionable "happiness" study argument, informed by behavioral insights on "loss aversion", the current darling of the economic left. So, something of a "hoist by their own petard" angle?

I think it's a dumb argument, and, even in a "static" economy, most people, particularly those of a Libertarian bent, would prefer that there by lots of scope for rising and falling, and would, ceterus paribus, regard mobility in relative outcomes as a sign of health.

If you think that taking one sentence out of context is reading perfectly it explains a lot about a fair portion of the comments on this post. If "mobility" only meant people trading relative positions with no other befits, then yes, Cowen would be against it. But he would also be against immigration, innovation, and lots of other things that cause mobility.

On this very blog Cowen has argued that people are too reluctant to uproot themselves and move to improve their economic position (an argument I disagree with by the way). And yet people still think blithely go along arguing that he is against economic mobility with out even stopping to think about what he is saying.

I'm surprised that there hasn't been more explicit discussion of this in terms of loss aversion. In particular, note the following sentence from Wiki ( "[s]ome studies suggest that losses are twice as powerful, psychologically, as gains." Note also that "habit formation" and "frame of reference effects" probably contribute quite a bit to the disproportionate favouring of status quo over potential loss (and wage rigidity may, in part, be a result of it).

Now, consider the following statement:

"For a given level of income, where upward mobility of X generates satisfaction in proportion to X and downward mobility of X generates dissatisfaction in proportion to (X <= Y <= 2X), relative mobility probably means less aggregate happiness."

Krugman, of course, theatrically interprets this as Tyler calling for a caste system, to maintain the status of the ultrarich and keep the poor at heel. It does not seem to even occur to him - or to you - that this lopsided dissatisfaction would hold true for a drop from the 4th to 5th quintile just as it would as a drop from the first to the second.

Now, I would actually plump for the Churn society in Alex's other post, on liberty/justice/deserts grounds. And I think it's pretty evident that the Growth component is paramount. But if you're strictly considering aggregate happiness under growth constraints, as Tyler was - i.e. giving equal weight to the happiness of every person in society, rather than allowing the nebulous hostility exhibited in your last sentence to cloud your preferences - then the above statement is entirely reasonable.


Don't forget that for a rich person, with all needs satisfied who is only saving, the marginal utility for the extra dollar is lower than for the poor person sitting under a bridge without food. If you take away a dollar from the rich person, or the last dollar from the poor person, the loss of marginal utility is different to both.

I don't have a problem with the behavioral observation you make, but don't forget marginal utility theory.

I honestly don't understand why anyone pays any attention to Krugman. He can not present an argument without logical fallacy and is obviously no longer an Economist, he is a political commentator.

"he is a political commentator."

Indeed. The fact that his economic "analysis" primarily appears in the OPINION pages of the New York Times next to other political commentators such as Maureen Dowd and David Brooks and is similarly "data-light" makes this clear.

What is truly remarkable is how a guy who publishes his opinions that regularly in the mass media has a reputation that has escaped a truly remarkable track record of being wrong. A reading from his work in the early and middle part of the last decade can almost be read today as comedy.

Wrong? I don't think you've really researched that too well. Krugman may be obnoxious sometimes, but his blog is hardly data-light and his predictions for the last few years have been spot on.

Like most political commentators, he specializes in telling you where you are and spinning it to sound as if it's where he said you'd be, while dissembling from his actual predictions.

He cannot be read with a critical eye over the last decade, especially near turning points such as 2001 or 2007, to have had any great foresight. Telling me today that Greece will likely leave the EU today for example is not going on a limb and does not qualify as accurate, though he will claim it as a prediction when it happens. Actually highlighting the extreme likelihood when the EU was formed (as many did while Krugman literally scoffed), would be worthy of respect, or even just advancing it as a looming issue a couple of years ago before it was a fullblown crisis would also qualify.

For his recent predictions, he has simply been correct in extrapolating existing economic weakness. That has not been a contrarian view, many other economists and strategists have had similar/more accurate projections for this weak growth. Many in fact that have completely opposite views on the causes and on the solutions. Those solutions by the way have seen far less implementation than Krugman's, yet he has the audacity to use the results as evidence of his theories!

Now I will admit to almost exclusively being exposed to his work in the oped pages of the NYT and do not visit his blog. Those predictions however are the most widely-read and when I see what passes for an economic analysis in them (a favorite is using a non-inflation adjusted graph for periods two decades apart b/c it suits his point by compressing earlier growth), I am not inspired to read more from him. His credibility as an analyst is shot from those columns and political opinion pages is certainly where they belong.

Krugman sees everything through a political prism. He wants the government sector to be large, and so no matter what the question is, that is always the answer. Wherever that prescription is not tried to his satisfaction, it is the reason for the weakness and wherever it is tried and unsuccessful, it just wasn't enough. He conveniently explains away the poor results in places that have implemented his grand solution as much as possible, and pretends that it hasn't been the overriding philosophy tried here since the bursting of the bubble (that he didn't see forming).

Krugman is a smart, well-read, articulate man, but he is highly political and lacks objectivity. While he is not alone amongst his profession, he lacks the open-mindedness and humility that anyone should have who is pontificating on a science that is so uncertain as to be closer to art.

Sorry, Dr. Cowan. I read your piece exactly the way Dr. Krugman did, and had exactly the same reaction.

To be frank, it seems to me that you have a surprisingly thin skin. You don't much like it when people disagree with you, do you?

Filed under "a huge percentage of tyler's readers don't read carefully enough to realize he doesn't have a nonstandard spelling of his name."

How often does Krugman get called "Kruegman" or something like that?

Krugman is a hack, Krugman will accuse everyone who disagrees with him as being a hack, DeLong will agree with Krugman, and Tyler should write more clearly. The end.


Meritocracy, not mobility, should be the goal. Under a meritocratic system, there will be no mobility as long as smart parents have smart kids.

Unless it's smart to be poor and dumb to be rich.

Comments for this post are closed