Network News

X My Profile
View More Activity

Research desk investigates: How great is American income mobility?

By Dylan Matthews

Rossi1 asks:

Just how mobile are people when it comes to coming into or out of a certain annual income bracket? How far up the ladder do the poor typically go and how far down do the wealthiest fall?

The best study (PDF) of this in the U.S. alone comes from the Treasury Department, which tracked tax filers in 1996 and observed which income quintiles they landed in come 2005. The study found that 57.6 percent of filers in the lowest quintile in 1996 had moved to a higher one by 2005. The second and middle quintiles also showed overall upward mobility, while the fourth quintile saw about as many people earning less as earning more, and the highest quintile saw 30.7 percent of filers drop out of it.

International comparisons, however, show that the U.S. still has a long way to go toward matching other developed nations' income mobility. Anna C. d'Addio at the OECD calculated (PDF) "intergenerational earnings elasticity" figures for a number of OECD countries. The measure estimates how much a son's earnings reflect those of his father. A higher bar thus means less mobility, as it means your family's income better predicts your income. Here's d'Addio's chart, as reprinted in another OECD paper (PDF):

mobility_graph.png

The U.S. does not come out the worst here; Italy and Great Britain have sharper class divisions than we do. But most other countries do substantially better. This includes not just Scandinavian social democracies like Denmark, Norway, and Finland (Sweden, curiously, does a bit worse) but Anglophone states such as Canada and Australia, with which the U.S. has much more in common.

Lest you think this is an isolated finding, a number (PDF) of other studies (via Matt Zeitlin) have confirmed that the U.S. displays unusually low levels of income mobility across generations for a developed country.

By Dylan Matthews  |  August 3, 2010; 12:06 PM ET
Categories:  Inequality  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   Del.icio.us   StumbleUpon   Technorati   Google Buzz   Previous: If you read nothing else today
Next: Lunch Break

Comments

The NY Times ran a lengthy series a few years ago (David Cay Johnston?) and showed how American mobility had really become a myth in the last decade or so.

We have a largely frozen class structure that the Treasury figures don't really show. People can have a great deal of wealth but not report all that much income, or they may have a great year and then a bad year that makes them seem mobile by income but their overall wealth doesn't change so much.

The keys as I recall were education, where legacies have a leg up and wealthier kids just generally have more choices, and the parents' ability to finance things like buying a first home. The inability of most people in the bottom 3 quintiles to amass much wealth is a real problem that has only been getting worse.

Posted by: Mimikatz | August 3, 2010 12:35 PM | Report abuse

I must ask the question though on this analysis, if other coutries that are being used for comparison have a much less pronounced income gap divide what is the purpose in comparing mobility across income quintiles with those countries? Of course those countries will show more mobility across quintiles if movement between quintiles in real earnings is much less. Would love a response from the research desk to this question.

Chris

Posted by: chackney | August 3, 2010 12:59 PM | Report abuse

Would like to know how many people moved multiple tax bracket and how.

for example, I started my career in the 15% bracket. 10 years, one wedding, and one child later my family of three is in the 33% bracket.

Posted by: NoVAHockey | August 3, 2010 1:26 PM | Report abuse

"The study found that 57.6 percent of filers in the lowest quintile in 1996 had moved to a higher one by 2005. The second and middle quintiles also showed overall upward mobility, while the fourth quintile saw about as many people earning less as earning more, and the highest quintile saw 30.7 percent of filers drop out of it."

I'd say those numbers are pretty good. The intergenerational correlations are probably half due to a kind of inheritance the government hasn't yet figure out how to tax, combined with the additional genetic and cultural diversity, where we have cultural groups that highly value educational attainment and groups that do not.
http://www.springerlink.com/content/u7w47275777821m4/

Posted by: staticvars | August 3, 2010 1:40 PM | Report abuse

Chris--are you saying it would be better to compare income mobility in the US with other countries that are as unequal as we are, or that there is a better way to compare income mobility in the US to these OECD countries?

Or do you just think there is no use in trying to make such comparisons at all?

(I'm not protesting your question, just not sure what you are getting at.)

Posted by: julie18 | August 3, 2010 1:52 PM | Report abuse

When are we going to have a society where everyone is in the top fifth? Seriously, for someone to rise in this sort of study, doesn't someone also has to fall? So if we have lots of upward mobility, we also must have lots of downward mobility.

Seems to me to the extent we have career paths in professions and occupations, it would be expected for people to rise as they age. But I've never seen stats on the proportion of the labor force which is stuck in jobs with no such paths, nor on the amount of jumping among jobs.

Posted by: bharshaw | August 3, 2010 1:56 PM | Report abuse

"The U.S. does not come out the worst here; Italy and Great Britain have sharper class divisions than we do. But most other countries do substantially better."

Yet I suspect Italy, and I know Great Britain, have engaged in more efforts to redistribute wealth and provide generous entitlements, and even embrace multiculturalism, than America has. My take away from this is that the left's prescriptions to end class divisions has no effect; that class divisions transcend any well-meaning government efforts to abolish them. Even when certain countries engage in fairly progressive tax policies.

That being said, our income mobility is, by the charts given, actually quite good as a percentage. I can't tell how it compares in real dollars, but I suspect we perform fairly well there, as well. Just, several other nations do better.

Posted by: Kevin_Willis | August 3, 2010 2:33 PM | Report abuse

"Yet I suspect Italy, and I know Great Britain, have engaged in more efforts to redistribute wealth and provide generous entitlements, and even embrace multiculturalism, than America has. My take away from this is that the left's prescriptions to end class divisions has no effect..."

Exactly. Because all those other countries doing so much better in this regard are well-known for their rightwing economic policies.

Posted by: slag | August 3, 2010 3:12 PM | Report abuse

"I must ask the question though on this analysis, if other coutries that are being used for comparison have a much less pronounced income gap divide what is the purpose in comparing mobility across income quintiles with those countries? Of course those countries will show more mobility across quintiles if movement between quintiles in real earnings is much less. Would love a response from the research desk to this question.

Posted by: chackney"

My question, exactly. It probably doesn't require much of a change in income to move up and down a quintile or two in a society with a very flat income structure.

Also, it seems that the report Ezra cited on US income mobility was not intergenerational at all. It measured individual mobility. And tends to substantiate the claims of those who say that poverty statistics disregard the fact that the poor are not the same group of people across decades. Individuals move up and down income levels during their life cycles.

Posted by: bgmma50 | August 3, 2010 6:15 PM | Report abuse

Julie18...thanks for the thoughtful response. I'm actually quite interested in analysis of how attainable the "American Dream" is, which income mobility is one proxy for. My issue was with the means of looking at comparative mobility across nations as a gauge of whether we're doing well in this regard. While I appreciate the attempt here, I really don't think its the right basis for such analysis and was just questioning that. I would be more curious to see the incremental "real income gains in absolute terms (ie real dollars in the US case" that was attained by each quintile. This measure would give you a better picture of real gains for each class than movement across quintiles would. My suspicion is that although you have move movement across quintiles in societies with less of an income gap, you will see more real income growth in each US quintile over lifetimes which would be a more accurate portrayal of "opportunity across each starting point in life"....which I believe is what the point of the exercise really is. Last note...I do know that "real income growth" has stagnated in the US in the last few years, but I would guess that that stagnation has been even more pronounced in other nations over the same time period (you could also get into a discussion around what constitutes "real income" and does that include just pay or pay plus benefits...the latter of which would show more growth).

By the way, the only thing I would take issue with in your post is when you say "unequal as we are"...I would say there's a big difference in inequality of opportunity versus inequality of outcome. I think this analysis is trying to measure the formal which is why the "income gap" in this country should be taken into account when using quintiles as a gauge.

Posted by: chackney | August 3, 2010 7:33 PM | Report abuse

Julie18...thanks for the thoughtful response. I'm actually quite interested in analysis of how attainable the "American Dream" is, which income mobility is one proxy for. My issue was with the means of looking at comparative mobility across nations as a gauge of whether we're doing well in this regard. While I appreciate the attempt here, I really don't think its the right basis for such analysis and was just questioning that. I would be more curious to see the incremental "real income gains in absolute terms (ie real dollars in the US case" that was attained by each quintile. This measure would give you a better picture of real gains for each class than movement across quintiles would. My suspicion is that although you have move movement across quintiles in societies with less of an income gap, you will see more real income growth in each US quintile over lifetimes which would be a more accurate portrayal of "opportunity across each starting point in life"....which I believe is what the point of the exercise really is. Last note...I do know that "real income growth" has stagnated in the US in the last few years, but I would guess that that stagnation has been even more pronounced in other nations over the same time period (you could also get into a discussion around what constitutes "real income" and does that include just pay or pay plus benefits...the latter of which would show more growth).

By the way, the only thing I would take issue with in your post is when you say "unequal as we are"...I would say there's a big difference in inequality of opportunity versus inequality of outcome. I think this analysis is trying to measure the former which is why the "income gap" in this country should be taken into account when using quintiles as a gauge.

Posted by: chackney | August 3, 2010 7:34 PM | Report abuse

"My take away from this is that the left's prescriptions to end class divisions has no effect; that class divisions transcend any well-meaning government efforts to abolish them."

You'd almost think that center-right governments had been in power in the UK and Italy for most of the post-WW2 years.

Oh.

Just because you consider something redistributionist by your warped American standards doesn't make it so.

Posted by: pseudonymousinnc | August 4, 2010 1:10 AM | Report abuse

Has anyone correlated income mobility with family structure? In other words, perhaps children from two-parent families have greater upward income mobility than do children from single-parent families. It seems to me that we have become statistically less mobile as the percentage of single-parent families have grown.

Posted by: Frankca | August 4, 2010 9:34 PM | Report abuse

"10 years, one wedding, and one child later my family of three is in the 33% bracket."

I'm pretty sure it's changes in your income, rather than any of those other things, that moved you into the 33% bracket.

Posted by: theorajones1 | August 10, 2010 5:36 PM | Report abuse

The comments to this entry are closed.

 
 
RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company