Mobility

by Russ Roberts on October 22, 2006

in Inequality

One of the most important on-going debates in America is the question of inequality and mobility. I have argued on this site that many comparisons across time are misleading because they are snapshots of different people and therefore mislead about the ability of people to improve their situation in America.

A new book is out from anthropologist Katherine Newman chronicling the lives of 300 applicants who applied for fast food jobs more than ten years ago in Harlem.

There are many objections to be made about such a study. Most importantly, it is only 300 people. But you learn something from such a study about the range of possibility. Especially when it focuses on the least skilled and least likely to succeed. From the New York Times book review:

In 1993, Katherine S. Newman, then a professor in the anthropology department at Columbia University,
began conducting interviews with 300 or so young men and women who had
applied for just about the least promising jobs you could think of:
flipping burgers and running registers at a fast-food franchise in
Harlem. Two hundred of them were paid minimum wage to do mind-numbing
work, and they were the lucky ones; the other 100 were turned down for
those same ill-paying, mind-numbing jobs. It was, Newman says, a
terrible time to be a low-wage worker in the inner city.

The reviewer, Times Magazine editor Paul Tough expected to find a dreary chronicle of failure:

Which is why it comes as such a shock when you read Newman’s
histories of people like Adam (a pseudonym, like all the names in the
book). The son of a mother on welfare, Adam dropped out of school after
10th grade, and he was turned down for a job at the restaurant. Doomed,
right? Well, no: he is now earning $70,000 a year, with full benefits,
as a union driver for an express delivery firm. Or Ebony, who was
working behind the counter doling out burgers when Newman met her and
is now a receptionist for a fancy law firm, studying to get her B.A. in
political science at night. Or Jamilla, who quit her job in the kitchen
to go on welfare, an unmarried mother raising her children alone, a
classic desperate case — until she completed her G.E.D. and worked her
way through culinary school. She is now a “stylish professional,”
Newman reports, with a well-paying job in a restaurant in Saks Fifth
Avenue. Shame? Try awe.

Newman doesn’t claim that these success
stories are typical. About a third of the 40 people she tracked down
and re-interviewed in 2002 were unemployed or still making the minimum
wage. But most had moved up, and almost a quarter were what she calls
“high fliers,” making $15.46 an hour or more. Newman’s fractions don’t
tell you a whole lot, as she herself admits; she’s an anthropologist,
not an economist, and her sample size is too small to prove much of
anything. Her book is valuable, though, as a collection of carefully
drawn portraits of people who got their start working at the bottom
rung of the American economy — in a lousy job, in a lousy neighborhood,
at the tail end of a recession — and in many cases managed to escape a
situation that seemed inescapable.

There is much that can be done to help low-skilled workers. Giving them the chance to acquire better skills in better schools is the right place to start. But it is good to know that even among the least-skilled Americans, success is possible.

The standard comparisons of average hourly earnings across time do not tell us what is really going on in people’s lives. Comparing average wages across time does not capture what is happening to the average or struggling worker.

Be Sociable, Share!

Comments

comments

Add a Comment    Share Share    Print    Email

{ 19 comments }

spencer October 23, 2006 at 10:15 am

Let's see. Her one success story was someone working in a union firm. But you keep arguing that unions are bad for employees. So what do I make of this?

Her quarter of what she called high flyers were making $15.46 vs $16.80 for average hourly earnings. Moreover, this is in New York where the cost of living is far above the national average.

A third were worse off then they were 13 years ago — either unemployeed or still earning the minimum wage that had fallen in real terms and as a percent of averge hourly earnings.

This leaves 42% (100 less one quarter less one third) somewhere inbetween no improvement and making less than averge hourly earnings.

You sure you want to use an example where roughly half of the sample are worse off then they were 13 years ago — the 33% plus roughly half of the 42% — and the other half are still below average to prove your case of how great the system is for the working poor?

Russ Roberts October 23, 2006 at 10:51 am

Spencer,

First of all, as I understand it, the sample is not quite the working poor, but the working and non-working poor. One third of the people she followed were turned down for fast-food employment.

Second, she finds that one-third makes 15.46 or more. Or more. Yes, average hourly earnings nationwide are a little over $16. But you're unimpressed that one-third of the people who were making the minimum wage in 1993 are now making roughly the average or greater. And many of them are making well above the average.

According to BLS data, Table B-47 in the Economic Report of the President, real average hourly earnings rose a pitiful 9% between 1993 and 2005. The standard interpretation of that number is that wages are stagnant, that only the rich are getting ahead, that the average worker is stuck on a treadmill with no hope for advancement.

I have written here many times why I think that interpretation is misleading—failure to account for benefits, failure to measure inflation correctly and a failure to look at the same people over time.

In this admittedly small sample of following the same people over time, one third of a sample of the poorest of the poor (not even the average) has a dramatic many-fold increase in their standard of living. To me, that is a very different finding from the standard story. You can argue that it's not representative. You can argue that the author is a liar and that she made up the data or the stories to sell books.

But I don't think you can argue that what she has found is a very different story from the standard one that is told about mobility in America.

Adam Malone October 23, 2006 at 10:58 am

Spencer,

You make a good point about everything that the die hard free marketers keep saying about "how great the system is for the working poor". You are right, poor people have it rough.

I am not going to say that our system in the USA is anywhere near perfect. However, I will say that is better than many systems in the world. While the results of this survey (which are statistically unimportant, as noted in the article) lead many to believe that "our" system has failed, what it leads me to wonder is what a similiar study would find in other countries. For example in Russia would we find that 28% of Moscow's poorest individuals were able to experience a marked improvement of their standard of living over a ten year period? What about in France or Germany?

Equally important to the question regarding the improvement of these workers wages would be a question regarding any changes in experience, education, or personal habits including hygeine, personal dress, and change in geography (movement within the borroughs).

I find it noteworthy that in your reading of the article YOU decided that the Union job was the most improved, ignoring the chef and the woman working to finish her degree.

Randy October 23, 2006 at 11:18 am

Exactly, Adam. In the progressive mind, reality always falls short of the ideal. Imagining a brave new world is easy – and quite common.

spencer October 23, 2006 at 11:27 am

Ok — you are right about overlooking the chef and student. I was wrong. But my basis point that from one-third to one-half of the sample are still worse off then they were 13 years ago is still valid. If the standard working life time of 40 years is used this means that they are one-third of their way through their working life with no improvement.

Russell argues that the averages do not reflect what happens over time. Yes, as people age their earning and real standard of living tend to improve on average. I agree completely. But the averages do incorporate this and reflect it.

Moreover, the data that adjusts for this is readily available at census. For example one of the series I look at is the real medium income for 35-44 years olds. this is the ten year age cohort that on average earns the most. It shows that for males real income using the cpi-rs has been flat and for females it has risen steadily. But if you taken a weighted average the real increase has been from $32,296 in 1973 to $32,628 in 2004 — a 5.7% increase.

Rather then saying the data does not reflect what you want it to why don't you go to census and get the data on income by age cohart and construct a series that reflects what you want it to. It would be very easy to do.

Then tell me what is wrong with the actual data that is reported.

Ryan Fuller October 23, 2006 at 12:35 pm

Spencer, Jerry Hausman has an excellent paper on CPI bias and its effects on inflation, linked from this web site. You can find the direct link here: http://econ-www.mit.edu/faculty/download_pdf.php?id=503

Ryan Fuller October 23, 2006 at 12:43 pm

Double posting since I didn't quite finish last time…

One of his findings is that systematic biases within the CPI result in inflation being overstated, producing real income growth figures that are much too low.

It's not too long, and I found it a great read. It also tells you what is wrong with the actual data that is reported, as per your request. :)

kebko October 23, 2006 at 1:42 pm

I can't stand it when journalists & social critics describe certain jobs as "mind numbing". It's condescending, demeaning, and wrong. There is nothing about working at McDonalds that is any more mind numbing than any number of "better" jobs. I'd like to see anybody get a job there, make some attempt to enjoy & trade banter with their co-workers, make it through the lunch rush where they face & serve dozens of people & their families & constantly juggle customer orders, special requests, etc. and dare to call it mind numbing. I consider those kind of comments to be a signal that the speaker or writer is the type of shallow leftist that assumes that anybody with a job he can't have is a robber baron and anybody with a job he wouldn't take is an affront to human dignity.

python October 23, 2006 at 2:06 pm

Spencer,

"But my basis point that from one-third to one-half of the sample are still worse off then they were 13 years ago is still valid." What does "still worse off" mean? "Worse off" than what? Do you mean, relative to others, or relative to where they were 13 years ago? If you mean relative to others are you surprised that only half the people can move out of a low income group? By the very numbers you show later, you make it sound like they all should be exactly where they were with 0% moving up, and 50% to 66% is a lot bigger than zero the last time I checked. And I can't see how you could come to the conclusion that they were worse off compared to themselves 13 years ago.

So, are you admitting that over half of the people are rising out of their income category? I'm not sure what your point is. Because I would be baffled to hear that any society can elevate a much larger percentage of the bottom tier.

And I guess you still don't understand Dr. Roberts meaning when he says "Comparing average wages across time does not capture what is happening to the average or struggling worker." The stats that you quote don't illuminate anything on that topic. They don't show what change has occurred to INDIVIDUALS as they progress through life. Your data and all data like it is skewed by many things including immigration. And you are also looking at income, not personal wealth or other real measurements. (Savings, home values, and other investments can rise even when income appears flat, and that doesn't apply just to rich people.) People immigrate to the US at all ages and when the average immigrant arrives they make substantially lower incomes than others. The only true studies of how to track how the mobility of individuals is to track them as individuals by doing longitudinal studies. "When Shaquille O'Neal comes into the room, the average height goes up, but no one got taller."

anon October 23, 2006 at 3:15 pm

"Let's see. Her one success story was someone working in a union firm. But you keep arguing that unions are bad for employees. So what do I make of this?"

I only read here occasionally, but has Russ ever said that? I would think he'd admit that unions are great for employees lucky enough to be in the union (provided their demands don't force the company under). He'd also argue, of course, that they're bad for customers and for unemployed people who are willing to work for less money than unionized workers.

spencer October 23, 2006 at 4:29 pm

I have been thinking about your argument that the average data does not reflect the point that as people age they earn more.

So I looked at the data over the last 25
years and what I see is that the baby
boomers aging appears to have significantly biased the average data higher rather then lower as you argue.

If you look at the share of the labor force
since the 1980s you can see the baby
boomers impact. For example, the 16 to 19 year olds share of the labor force
fell from 8.8% in 1980 to 4.8% in 2004. The share for 20 to 24 years olds also fell from
14.9% to 10.3% over the same time.

share of civilian labor force
……..16 to…20 to..25to…. 44 to…46to
……..19……24…..34…….45……54
…….years years years years years
1980…..8.8….14.9….27.3….19.1….15.8
1990…..6.2….11.7….28.6….25.5….16.1
1995…..5.9….10.3….25.8….27.0….19.1
2000…..5.8….10.0….23.0….26.3….21.8
2004…..4.8….10.3….21.8….24.5….22.9

You see that the 25-34 year cohort peaked around 1990 at 28.6% and the next corort peaked at 27% in 1995 and the 46 to 54 age
cohort just reached 22.9% in 2004.

If you take your quite valid point that as people age their income grows and applying it to the distribution as the above table
shows if becomes quite apparent that the baby boomers moving through the labor force over the last quarter century has created
a significant upward bais to the average earnings data.

For example if you assume that the 16-19 years old make $1/hr, the 20-34 $2.00,
etc, etc, and the 45-54 years olds make $5.00 in every years this generates a weighted average of$55.18 in 1980 and a weighed averge of $60.71 in 2004. So with no change in wages and just aging the labor force from 1980 to 2004 generates a 10% increase in average wages — just the opposite of your argument.

spencer October 23, 2006 at 4:54 pm

Hw do I evaluate the Jerry Hausman article when the first fact I encounter is wrong.

Cell phone servie was rolled out in most cities in 1985 according to him and I think he is correct. But he says it took the BLS some 15 years to include it in the cpi which would be the year 2000.

But I go to the bls and find that cell phone service has a weight in the cpi of 0.039% in 1993-95.

Something does not add up here.

spencer October 23, 2006 at 4:54 pm

Hw do I evaluate the Jerry Hausman article when the first fact I encounter is wrong.

Cell phone servie was rolled out in most cities in 1985 according to him and I think he is correct. But he says it took the BLS some 15 years to include it in the cpi which would be the year 2000.

But I go to the bls and find that cell phone service has a weight in the cpi of 0.039% in 1993-95.

Something does not add up here.

python October 23, 2006 at 8:12 pm

Spencer,

You are completely missing the point of using income data. You can not use averages of entire populations to see how individuals are doing. There are always young people and immigrants of all ages skewing the data. You have to track individuals as individuals. "When an unemployed guy enters the room the average income just dropped, but no one got poorer."

As the article says: about 1/3 of the applicants were in the same status in life, but a majority had moved up, and a quarter were up to average or higher. That sounds like improvement to me. No one is claiming that everyone will go from poor to above average in their lives, just that it is far more common than is frequently reported when discussing "the poor."

What percentage of those job applicants do you want to see improvements in before you are happy with our economic system? What era in time or georgraphic location were a significantly larger portion than shown in this study able to move up from far lower than average to average in 13 years????

spencer October 24, 2006 at 8:08 am

I'm very happy with our economic system.

I'm very unhappy with poor economic analysis.

Press October 24, 2006 at 9:22 am

I am curious about the effect of education on the inequality data. Do we as a society expect that compensation levels to remain relatively constant across education levels across time? Is it normal and expected that the compensation of an illiterate would decline against all others over the last 50 years. Is that inequality that we as a society should be concerned about? Or should we just acknowledge that it is a normal occurrence for a advancing society and we need to ensure a good education system? And then how much of the changes seen in the inequality data is education related?

Press October 24, 2006 at 9:23 am

I am curious about the effect of education on the inequality data. Do we as a society expect that compensation levels to remain relatively constant across education levels across time? Is it normal and expected that the compensation of an illiterate would decline against all others over the last 50 years. Is that inequality that we as a society should be concerned about? Or should we just acknowledge that it is a normal occurrence for a advancing society and we need to ensure a good education system? And then how much of the changes seen in the inequality data is education related?

Russell Nelson October 25, 2006 at 1:42 am

Spencer, would you please pay attention? Russ is pointing out that averages don't reflect individuals. Individuals move from one average group to another average group. Thus, comparing an average over times doesn't say much about actual people. It just says something about the average person — which most intelligent commentators agree doesn't exist.

(hint: I mean to leave you out of the latter group, but by all means get smarter and join us.)

happyjuggler0 October 25, 2006 at 8:18 pm

I still think that adding a bunch of low skill immigrants at the bottom of the scale distorts the median even when everyone else moves up the ladder.

Looking at US stats without subtracting immigrants is a serious mistake.

I also think we have more or less reverted to the pre-WWII (maybe pre-depression too) income distribution curve between the highly educated, the sorta educated, and the poorly educated. WWII, its preceding time frime, and its aftermath was an anomalous event that artificially raised the incomes of the relatively uneducated in the US. Eventually the high human capital part of the world rebuilt its factories and physical infrastructure, often better than before, and education again had increasing returns compared to the 50's and 60's. Increase the returns to education relative to the relatively uneducated, and you get a set of data that looks an awful lot like the data we have going back to before WWII.

Previous post:

Next post: