Mantra vs. Reality

by Russ Roberts on March 9, 2007

in Standard of Living

In this piece in the Washington Post, Harold Meyerson makes a common complaint about the American economy and our standard of living:

Over the past 35 years, the massive changes in the U.S. economy have largely condemned American workers to lives of economic insecurity. No longer can the worker count on a steady job for a single employer who provides a paycheck and health and retirement benefits, too. Over the past three decades, workers’ individual annual income fluctuations have consistently increased, while their aggregate income has stagnated. In the brave new economy of outsourced jobs and short-term gigs and on-again, off-again health coverage, American workers cannot rationally plan their economic futures.

Unfortunately, this mantra contradicts an earlier statement by Meyerson  on the facts of the matter. This earlier statement was not made four years earlier or four months earlier or four weeks earlier or four days earlier. The earlier statement was made FOUR PARAGRAPHS earlier in the SAME COLUMN. Here’s the quote, referring to an article in the Post by reporter Blaine Harden:

Taking into account all households, married couples with children are twice as likely to be in the top 20 percent of incomes, Harden reported. Their incomes have increased 59 percent over the past 30 years, while households overall have experienced just a 44 percent increase.

So over the last 30 years, married couples with children have seen their incomes increase 59%. The increase for all households is 44%. Both of these numbers suffer from composition bias. Over the last 30 years, the number of single-headed households has grown dramatically, dragging down the average for purely statistical reasons. So the overall change in household income of 44% is not representative of the average family. Increased immigration, the addition of households with lower than average incomes, also reduces measured household income and understates the gains to families who were already here.

But even with this bias, a 44% increase isn’t stagnation.

BTW, the Harden article doesn’t make it clear, but these are probably averages. A better number is the change in the median. This report from the Census Bureau shows that between 1967 and 2005, real median household income was up 31% for all households. Using 1970 as the starting point and the increase is about 25%. So even including the compositional problems I mentioned and even the flaws in the CPI (see here and here and here) which cause the increase in real income to be understated, the median household isn’t stagnating financially.

Be Sociable, Share!

Comments

comments

56 comments    Share Share    Print    Email

{ 28 comments }

Mathieu Bédard March 9, 2007 at 10:44 am

The only logic explanation is that these all happened in the first 5 years of his "past 35 years"… :)

How can someone get away writing such "mantra", as you put it, in an otherwise respectable publication? You don't need to be an economist to discard this paragraph as nonsense..

Jeff Singer March 9, 2007 at 1:47 pm

Here is why both paragraphs can make sense (and why Meyerson should be more careful to let his readers know what he is talking about):

- in the first paragraph, he is talking about households, many of which include TWO incomes;

- in the second paragraph referenced, he is talking about individuals' income (i.e. individual salaries)

One could argue that most of the household income gains of the past 35 years have come mainly as a result of two income couples.

Morgan March 9, 2007 at 1:55 pm

Actually, I think the usual explanation is that the gains in income come from having two income-earners in the household. I haven't looked into whether that's reasonable.

I also should note that the "decreasing stability" argument is in direct contradiction to data regarding initial filings for unemployment – the percent of people eligible to collect unemployment benefits who actually file in any given week has been dropping for the last 35 years, and currently stands just above the historic low it reached in early 2000. And the drop is huge – we're talking rates that are one half to one third what they were in the 1970s. There may be less "stability", but if so that number has to include separation from employers that does not involve unemployment – primarily that means changing jobs voluntarily.

Mike March 9, 2007 at 2:23 pm

Well, the empirical evidence certainly doesn't indicate that long-term employment relationships are going the way of the dinosaur.

Here is a recent paper by Ann Huff Stevens:
http://www.nber.org/digest/sep06/w11878.html

And here is something D. Acemoglu observed (http://econ-www.mit.edu/faculty/download_appen.php?id=114):

"despite the popular perception to the contrary, there has not been a large increase in employment instability. The tenure distribution of workers today looks quite similar to what it was 20 years ago."

Russ Roberts March 9, 2007 at 2:30 pm

Real median incomes for all kinds of families, families with two earners and families with one earner, have grown in real terms over the last 30-35 years. It's not just a matter of more two-earner families.

I will soon post a pdf file with this and similarly cheerful information.

ben March 9, 2007 at 6:56 pm

It seems to me economic security is not measured by the tenure of employment in any one position if it there is flexibility in the labour market and is straightforward to find new work.

Bill Conerly March 9, 2007 at 11:38 pm

We have this idealized notion that the "good old days" were characterized by lifetime employment. My father was downsized in 1957 (thanks to Eisenhower's military cuts) and again in 1970 (mutual fund industry went into a slide.) He got no pension or other retirement assets from either job. Growing up near Douglas Aircraft, periodic layoffs were common. The kids across the street had hard times when their dad would be laid off.

Martin March 10, 2007 at 9:17 am

"Increased immigration, the addition of households with lower than average incomes, also reduces measured household income and understates the gains to families who were already here."

So remind me why everyone should be in favour of more immigration?

python March 10, 2007 at 11:54 am

Martin,

Immigration lowers the "averages", not the individual's numbers. Immigrants come and the first jobs they obtain frequently pay less. The same is true for young people who are newly entering the workforce. This is one reason why averages are meaningless in dynamic societies.

Example: Nine people live in a town and all make $50,000. Then a tenth guy comes who only makes $30,000. The average has dropped from 50k to 48k, but no one made less money after the new guy came to town – immigrant, youth, or just a guy from another town who decided to move 30 miles away.

LisaMarie March 10, 2007 at 8:16 pm

So is he saying that if I cannot be sure that I will have the same job for the rest of my working life, I cannot rationally plan for the future? I wasn't aware that you needed absolute certainty before you could make any plans whatsoever for the future. Perhaps he has a crystal ball?

The Dirty Mac March 11, 2007 at 12:33 am

My parents were my age about thirty years ago. Around that time, our family downgraded from veal to chicken cutlets. If the food you're eating is not an economic indicator, I don't know what is. I'm not a rich guy, but I know I can go out and buy a large supply of veal cutlets tomorrow without much economic hardship. While doing so, I can listen to FM radio or play a CD in the car if I want to…

Tense Alcyoneus March 11, 2007 at 1:04 am

"It seems to me economic security is not measured by the tenure of employment in any one position if it there is flexibility in the labor market and is straightforward to find new work."

But it isn't. For example over the last ten years, almost an entire generation of IT professionals were relegated to fast food jobs. People are dissatisfied because their livelihood didn't disappear, it was taken by people from other countries.

Those foreigners are also conspicuous. Workers actually see them, all the time, doing much better than US workers. Worse, these workers discover that they cannot compete in the "flexible job market" because foreign workers are chosen a priori, and the US government has developed a private market for employers — one which bars domestic employees.

Even at our current full employment, people don't feel secure when the local short order cook has a PhD in Computer Science.

I was once a hard core capitalist. Now, I recognize that businessmen are the worst capitalists. It's a regulation war: businesses try to beat the workers with regulations, and workers try to beat businesses.

I'm now a dedicated socialist, and I vote straight Democrat. I still can't bring myself to join a union, though. But I'm working on it.

python March 11, 2007 at 3:21 am

"For example over the last ten years, almost an entire generation of IT professionals were relegated to fast food jobs. People are dissatisfied because their livelihood didn't disappear, it was taken by people from other countries."

This is an awesome paragraph. All the statistical relevance of an anecdote without an actual anecdote.

My anecdote is that I happen to have been in IT, know dozens of IT professionals in different levels of IT (support, programming, managerial) who work for different kinds of companies and not a single one lost a job to outsourcing.

Fortunately for me, I don't have the burden of proof to show that outsourcing has not stripped IT professionals of their jobs; it is Tense A. who has the burden to prove that it has. But I doubt he can bring stats like he brought that other stuff he typed.

Martin March 11, 2007 at 5:25 am

Python,

Define 'dynamic'.

python March 11, 2007 at 12:22 pm

Dynamic (from http://www.thefreedictionary.com/dynamic), definition adj 2.

"2. Characterized by continuous change, activity, or progress: a dynamic market."

In the context of this discussion, because our society has a "continuous change" of characters, taking averages is less than meaningful. Proper studies involve longitudinal methods that track individuals as they proceed through life. How's that tired line go? "When Shaquille O'Neal enters the room, the average height went up but no one got taller."

Of course there are many other reasons to be for or against immigration, but the fact that immigration lowers national income averages probably shouldn't be one of them.

Tense Alcyoneus March 11, 2007 at 12:43 pm

http://www.infoworld.com/article/05/08/23/35OPreality_1.html
The cap for H-1B visas in 2006 is set at 58,200, but apparently, according to the USCIS Web site, 22,383 visas have already been approved and 29,556 are still pending. Berry says the DOL is “refusing to disclose the opening to U.S. citizens so that they may have equal opportunity to apply for and fill these U.S. jobs.”

http://www.eweek.com/article2/0,1895,1992897,00.asp
Berry reminded Carlson in the letter that "although LCAs are public records, U.S. workers do not have access to these records," and likened the records' lack of public availability to "choosing to reserve 65,000 U.S. jobs exclusively for foreign workers."

http://www.computerworld.com/careertopics/careers/labor/story/0,10801,72848,00.html
But Nobel economist Milton Friedman scoffs at the idea of the government stocking a farm system for the likes of Microsoft and Intel. "There is no doubt," he says, "that the [H-1B] program is a benefit to their employers, enabling them to get workers at a lower wage, and to that extent, it is a subsidy."
From free-market thinker Friedman, those are devastating words. The H-1B program is a subsidy that distorts the job market for IT talent.

Tense Alcyoneus March 11, 2007 at 12:57 pm

http://press.arrivenet.com/notforprofit/article.php/662012.html
Call for New Visa to Bring Employers Rather Than Workers
"U.S. employers are simply not creating enough jobs for skilled IT workers, causing upwards of 15% of IT professional to be displaced from the profession in the past four years," states Kim Berry, president of the Programmers Guild. "Therefore our organization urges Congress to create a visa category to attract the "best and brightest" foreign employers to the U.S. – on the condition that they exclusively hire unemployed U.S. workers for their IT slots."

This shortage of qualified employers is dissuading new college graduates. The ITAA study reports that "the popularity of computer science as a major among incoming freshmen dropped 59% from 2000 to 2004."

Martin March 11, 2007 at 1:20 pm

Python,

"In the context of this discussion, because our society has a "continuous change" of characters, taking averages is less than meaningful."

According to the definition you have stated, immigration does not form part of a dynamic, in that the character of American society was not affected by 'continuous change' until the enactment of the Immigration Act 1965.

Post 1965 mass immigration is therefore the result of a policy, not a process – being politically driven, it cannot therefore be naturally economic in character.

python March 11, 2007 at 2:15 pm

Martin,

Maybe we are talking about different things. Because of what you put in quotes, I thought your first comment was in regards to immigration lowering average income statistics,

I am talking from a strictly statistcal point of view that when "new" people come into a statistical pool, the averages change (except in the rare case that the new entrants match the pool's averags). But that doesn't mean that the people already in the pool have had anything change.

I'm not talking policy. To restate: in a changing dynamic society, when new people enter, averages will change. This is a mathematical fact, not an economic theory.

And as to the definition, immigration is exactly part of the definition of dynamic, in that it is always changing the sample group, as opposed to a static situation. "Characterized by continuous change…" Immigration continuously changes the players, as does young people maturing, old people dying, Americans leaving United States, etc.

Immigrants came to America before 1965, and therefore their impact on stats of Americans has been impacted since before that time frame. I don't see the confusion here.

Tense Alcyoneus March 11, 2007 at 2:24 pm

"Immigrants came to America before 1965, and therefore their impact on stats of Americans has been impacted since before that time frame. I don't see the confusion here."

Previous immigrants did not come here by participating in a private employment market created by the government.

ben March 11, 2007 at 6:57 pm

"But it isn't. For example over the last ten years, almost an entire generation of IT professionals were relegated to fast food jobs."

The number of IT jobs has more than recovered from the 2000 crash. The trend during the crash was from contract work to permanent employment, not from IT to fast food. I doubt more than 1% of workers formerly in IT work in the fast food industry as a result of the crash, but if you can point to some evidence I'll read it.

python March 11, 2007 at 7:19 pm

Tense A. says:

"This shortage of qualified employers is dissuading new college graduates. The ITAA study reports that 'the popularity of computer science as a major among incoming freshmen dropped 59% from 2000 to 2004.'"

Do you understand that 2000 was at the very height of "Internet boom"/Y2K? Of course, Computer Science was hot then. Trying to correlate that 59% drop to global outsourcing or H-1 Visas is going to be a tough task. How do incoming Computer Science freshman compare 1994 to 2004? Why cherry pick the very top? I think I know why.

In a few years, the number of people studying to be realtors will be much lower than it was in 2005. Is that proof that global considerations have quashed the aspirations of American realtors? No. It means that people will figure out how the market works, and that realtor jobs aren't as lucrative/easy as they once were.

Tense A. hasn't followed up on any of the other criticisms from his earlier posts, so I'm not sure why I fell for the bait and switch.

Tense Alcyoneus March 11, 2007 at 7:51 pm

"The number of IT jobs has more than recovered from the 2000 crash. The trend during the crash was from contract work to permanent employment, not from IT to fast food."

True, the number of jobs has recovered, especially in the last 12 months. But the most of the new jobs have not gone to US workers, but rather to foreign workers. Thus, many, many US workers feel underemployed. That's why confidence is low in a booming economy.

Tense Alcyoneus March 11, 2007 at 8:06 pm

"In a few years, the number of people studying to be realtors will be much lower than it was in 2005. Is that proof that global considerations have quashed the aspirations of American realtors? No. It means that people will figure out how the market works, and that realtor jobs aren't as lucrative/easy as they once were."

These hypothetical realtors will not be barred from competing in a free labor market, as have domestic IT workers.

python March 11, 2007 at 9:03 pm

"These hypothetical realtors will not be barred from competing in a free labor market, as have domestic IT workers."

Please explain how I and dozens of others that I know were barred from from a free labor market. I was in it from 1990-2005 in different settings and different job requirements. Please give stats for your hyperbole, which seems neverending. Occasionally, throw in a number to back up your words.

Perhaps you mean "Americans who hold IT jobs that don't require an on-site presence have felt pressure to compete with foreign-based employees." But then I would need you to quantify what percentage of IT employees can be based outside the company. There are a whole host of IT jobs that require you to be on-site: from software/hardware support to business managers to desktop configuration to most SysAdmins, just to name a few. Not all IT can be handled over the phone or from remote accessing. Go do a survey of how many companies under 1000 employees use foreign IT support for the majority of their support.

Here are some numbers: http://research.stlouisfed.org/fred2/categories/10

I suspected that IT curled over pretty fast in it's employment boom. Current levels are where they were June 1997. That data alone doesn't sound good, but look at how the other industries are doing in that same time frame (note the size of each sector relative to IT) (all numbers in thousands):

Information Services: 1997-6-1 = 2007-2-1 (both at 3087)
Financial Activities: 7156 -> 8450 (18%)
Profession and Business Services: 14318->17848 (20%)
Service Providing Industries: 98808->114935 (17%)

From it's peak on March 2001, IT has lost 600K jobs. Finanical Services alone has gained 500K jobs in the same time frame.

So a boom comes and lifts IT employment from 3.0 million to 3.7 million and back down, and you write that "almost an entire generation of IT professionals were relegated to fast food jobs." Hmmm. Something doesn't smell right.

As far as "That's why confidence is low in a booming economy.". The most trusted consumer confidence rating (UM) is currently at 91.3 which is near average for the past 30 years. The only time that rating was substantially higher (over 102) was in the tech bubble days. Perhaps you mean that confidence is low RELATIVE to the booming economy, because that is a common mistake amongst casual commentators.

Tense Alcyoneus March 11, 2007 at 9:10 pm

"Please explain how I and dozens of others that I know were barred from from a free labor market. I was in it from 1990-2005 in different settings and different job requirements. Please give stats for your hyperbole, which seems neverending."

I have provided links to articles on the DOL's practices. I won't further respond to you because you write your posts with the charateristic belligerence of a forum troll.

python March 11, 2007 at 10:59 pm

Tense A.,

The original topic regarded median household income. Your posts seem to be about how "foreigners are getting all the good IT jobs" You didn't really state any hard evidence for your opinions or why you think your tangent is relevant to the original post.

Here is my humble recap to clarify what I am referring to.

You have shown no support for your line "For example over the last ten years, almost an entire generation of IT professionals were relegated to fast food jobs." You back-tracked on this one but you added another comment with no data – "But the most of the new jobs have not gone to US workers, but rather to foreign workers.".

No support for "people don't feel secure when the local short order cook has a PhD in Computer Science."

65,000 Visas that subsidize corporations out of 150 million employee positions. Hmmm. Is this the thrust of your argument?

You (or the article you cite) cherry picked the time of highest IT demand (Y2K and Internet boom) as a baseline. That is not looked upon as an unbiased methodology.

You never showed how current IT professionals are barred from the free market. Your evidence is that 65K foreigners are getting first position in domestic jobs. Even if they were all in IT are they really manipulating the other 3 million IT jobs? How does your sentiment square with the data that shows IT jobs number the same amount as they did before the tech bubble, and are only 16% lower than at their peak? Obviously the other 84% aren't working fast food are they?

Please bring substantive data to an opinion that you post. It's part of the process of hashing out the topic. There is no belligerence. Saying something contains hyperbole is not a belligerent act. http://www.answers.com/topic/hyperbole

Kent Gatewood March 15, 2007 at 12:43 pm

Real Median Household Income (Census)

1967-2005 31% = 0.695 yearly
1970-2005 25% = 0.622 yearly

I was expecting something more dramatic, oh well.

Previous post:

Next post: