Our lost decade in one graph

Total compensation, including the salary employers are providing in the form of health benefits, is the relevant metric. This graph does not tell us very much.

I'm curious if there's data as to why it jumped in 2005, after hitting a bottom there, and in what sectors we have growth in median income.

Any more graphs on that subject?

The health care bit is a good point. It would show how actual employeee salary is being converted by employers to payments to the health care industry.

Isn't it funny how corporate profits were going crazy at this time? I imagine you'd see a similar chart for the last two years, but without the tail... just a continuing dive in salary.

Adjusted for inflation, the typical family's health insurance premiums rose by about $3,500 from 2000 to 2007. The median working family probably has employer paid health insurance.

http://www.usatoday.com/money/industries/health/2009-09-15-insurance-costs_N.htm

Still not an impressive performance though, is it?

And the odds of lost decade #2 are still quite high.

Not to be sarcastic or anything, but what was the benefit of the tax cuts again?

After admitting that "This paper focuses on wage trends in non-inflation adjusted terms, what economists call nominal wages. Of course, living standards are set by wages adjusted for inflation, that is, real wages" the authors of the paper conclude that "one key policy would be to restore the minimum wage to half of the average wage and maintain that as a labor standard moving ahead." Thereafter appears an appeal for mandatory union membership (the "Employee Free Choice Act" in Obamaspeak) and a statement suggesting that the executive could achieve similar results more easily by simply usurping power.

Consider the asymptotic effect of the proposed policy. How well has such a policy worked for the (socialist) governments which have utilized it?

This is why I'm in favor of deflation - at least until the purchasing power of today's median wage is equal to the purchasing power of, oh, say 1973's median wage.

I'm only half kidding. Yes, yes, I know, we can't guarantee that it would end there, or that wages wouldn't also decline, but the fact remains that inflation in prices but not in wages is, essentially, what led to the debt crisis that caused the economic crisis.

It'd be nice if we could deflate our way out of the bind that caused so many Americans to overextend their credit. (I.e., keeping up with inflation on stagnant wages.)

A graph that includes 2008 data is available as Figure 1 in a Census Bureau report here:
http://www.census.gov/newsroom/releases/archives/income_wealth/cb09-141.html

Plotting the median income, rather than the average (or the technical term, mean) income is a curious choice.

It only tells you how people near the 50th percentile are doing, not the poor or the well-off.

As long as welfare encourages teenage girls to splinter off from their families by conceiving out of wedlock and collecting a check for a second poor family they now head up, the median income data are skewed to poorer and poorer classes of people.

It probably also neglects assistance programs, which increased significantly over the "lost" decade. (To be fair, I should've checked that in the link Klein provided.)

Those kind of arguments were also used to argue that "only the rich" benefitted during the Reagan era expansion. And they were weak arguments.

Use of per capita, or mean family, or welfare-adjusted incomes tends to show quite a different picture, and is also more thorough.

I'd love to know the post-2007 data. That's when we have Pelosi and Reid at the helm of Congress, and the president is either Obama or Bush in his last two years, when he was no fiscal conservative. In a few days John Boehner will go around saying "are you better off than you were four years ago?" and continue the strongly anti-Democrat trends in the polls.

Dear punditocracy: STOP USING PER-HOUSEHOLD DATA to point out trends (good or bad)! The composition of a "household" is constantly changing, and generally households are getting smaller as divorce rates rise, people marry later in life, and have fewer children. I realize that this timeframe is not a terribly long one, so differences in household size are not likely to be large from its beginning to end, but it's still vastly inferior to a graph plotting real median or average income per worker. Including benefits would also make for a better graph.

I just checked more Census Bureau data, and actually the family size and welfare payment arguments I was making above are no good: The average family size decreased from 2.62 to 2.61 from 2000 to 2007, and income due to government assistance is supposed to be included.

The good news is that Ezra Klein's main point is still wrong: That avergage - that is, mean - family income actually increased significantly during that period.

Table H-3 of the link below shows that the average household brought in $57,000 in 2000 and $68,000 in 2007.

http://tinyurl.com/243lpq5

The bad news is that inflation ate up the gains, and that the Pelosi/Reid Congress appears to have sent real incomes on a steep negative trajectory thereafter. We should concern ourselves with ending the Lost Half Decade of 2007-2012 by throwing the bums out and electing some fiscal conservatives for a change.

@angrydoug: Even if family income actually decreased during that period, I'm not sure that it's reasonable to expect that median income will constantly go up year over year, or to attribute the change in income, one way or another, to policy.

Could be just that more people moved from high-cost-of-living areas to lower-cost-of-living areas, where they made less money but enjoyed a higher standard of living. How's that plot on that kinda graph, huh?

angrydoug1 @ September 2, 2010 1:07 PM wrote "Plotting the median income, rather than the average (or the technical term, mean) income is a curious choice. It only tells you how people near the 50th percentile are doing, not the poor or the well-off."

Plotting the median is an accurate measure of the economic state. In this case, it is obvious that HALF THE EARNERS have been earning less and less throughout the decade.

Plotting the average (or mean) would tell us nothing of the sort. Afterall, there is a small sliver (about 2% of the population) that has been doing extremely well, and will lift the average up. Obviously they are doing very well at the expense of 50% of the population (being evaluated), which is what the median tells us.

Consider as an example these 3 cases:
(1) 4 each earn $1; 5 each earns $10; 1 earns $100: Median is $10, average is $15.4
(2) 4 each earn $1; 5 each earns $10; 1 earns $1000: Median is $10, average is $105.4
(3) 4 each earn $1; 5 each earns $10; 1 earns $10,000: Median is $10, average is $1005.4
If these show a trend, which do you see as the better measure of what is happening, the median or the average?

Ezra's graph correlates pretty neatly with what we know about illegal immigration numbers during those years.

Kevin, I don't believe that cost-of-living dynamic changes much, not on a nationwide scale. If lots of people move from pricey areas to cheap ones, then the pricey areas become a little cheaper and the cheap farmland or whatever will become more expensive, probably a net wash. The nationwide cost of living rate of change is important; I'm trying to post data relative to the CPI which is supposed to reflect that.

You question whether the year-to-year changes in income are fairly attributed to policy, and I agree; there are numerous factors at work. (Though a 7-year period is better.) But what do you think Klein's point WAS? When a known Bush hater and biased liberal posts a 'negative' data trend over the period 2000-2007, is it not implied the bad data is due to the author's favorite villain?

I am posting data that I believe are more informative than Klein's. The connection between my data posts and policy may not be so strong, but the same goes for Klein's data then.

To wit, here's some more good data. I was wondering whether Klein's data from Census included or excluded taxes. That matters; Bush's tax cuts naturally improved family incomes, but pre-tax data would conceal that. I couldn't find a definition of income in Klein's Census Bureau data that clarified whether posted income data had taxes removed. But the CBO, hardly a GOP servant, publishes its own income estimates, and makes them quite clear about taxation.

See http://tinyurl.com/22w9g6j for CBO data.

The CBO reports that the average family income rose from $67,000 to $72,000 from 2000-2006 in 2006 dollars. 2007 data weren't available, and this is the "All Quintiles" average column of after-tax data. The before-tax data reflect a somewhat less positive trend.

The only factor left out, far as I can tell, is the cash value of benefits, which I suspect also went up, as employers have to pay a lot more for workers' health care premiums than they used to.

There you have it, despite 9/11 and the residue of the internet bubble popping, the average family was better off late in the Bush years than at the beginning. The extent to which Dubya himself CAUSED those improvements is perpetually debatable. But American families did better under him, not worse, not at least until Pelosi rose to power. And Klein should know better than to post misleading trends.

angrydoug1 @ September 2, 2010 1:44 PM &
Kevin_Willis @September 2, 2010 1:56 PM:

angrydoug1: I believe that I just proved to you that the average is a rather meaningless measure of how well things are going. That makes you desire to blame anything on 'the Pelosi/Reid Congress' is just the petty posturing of a political hack.

Kevin_Willis: If you follow the link that adonsig @ September 2, 2010 1:03 PM posted you will realize that noone moved to low wage areas for a higher standard of living. Nice try, but an irrelevant comment.

"Bush in his last two years, when he was no fiscal conservative."

Oh, that was good for a laugh! Bush was never a fiscal conservative. He was always an irresponsible spendthrift who wanted to cut taxes.

AMviennaA, median statistics do NOT tell you how bad the poorest of the poor have it, nor how well the richest of the rich have it. They only tell you about the people smack dab in the middle.

Mean income statistics usually do skew towards the rich, but they do include the poorest of the poor, and are a logical starting point.

If you're curious about the distribution of income, and how you think an evil GOP regime might have changed it, note the CBO values I referenced, quintile by quintile. Folks in all five quintiles did better between 2000 and 2006, after taxes and inflation. You see, Ezra Klein's doom-and-gloom chart is misleading, indicating the opposite trend.

angrydoug1, you need to take a basic stats course.

angrydoug1 @ September 2, 2010 3:25 PM: As steveh46 pointed out, you need to take a statistics course.

As for "the distribution of income", I find it fascinating that although the absolute earning of quintiles (except the top) was reduced (even prior to adjusting for inflation), you maintain that all quintiles did better. If you look at the trajectory of the median, that tells you automatically that at least 3 of the 5 quintiles did worse!

I believe that you need more than just a course in statistics.

Could someone provide an authentic source to the information on the chart?
You cited a progressive think tank.
The lack of credibility of your source renders that chart about as authentic as Michael Mann's hockey-stick graph, which has already been debunked.

However, I can provide data from the Carter years and the Reagan years that will confirm that household incomes rose by $4000 during Reagan's term.

Now that's impressive....and true

This graph is misleading for two reasons.

1) It does not take into account the number of people per household. That number has been steadily declining. If you were to look wages per individual and not per household, you see a different result.

2) It doesn't take into account total compensation (wages + compensation). Compensation has actually risen steadily this past decades as it did in the 1990s. The only problem is that most our extra money our employers were throwing at us were instead going towards the rising cost of health care.

bradpeterson @ September 2, 2010 7:19 PM wrote "2) It doesn't take into account total compensation (wages + compensation). Compensation has actually risen steadily this past decades as it did in the 1990s. The only problem is that most our extra money our employers were throwing at us were instead going towards the rising cost of health care."

Firstly, not all employees receive health care coverage from employers (which means that it was paid from earnings or not at all). Secondly, even those employers who provided coverage, on average paid 35% of the cost (which means that the employee's costs went up). (By the way, the employee can claim only the amount that exceeds 7% of earnings on the tax return).

Therefore: with compensation declining, and extras costing more, I see no reason to be greatful for the 'prosperity' of the Republican policies.

This graph makes me wonder if the current resesinon was inevitable or preventable. I personally have heard more arguments for the later and in some ways this graph can confirm this, but in the other side it also is a very subtle detail that may have been an early warning sign or just an irregular statistic that foreshadowed what was to come.

Notice the word household, what the real data did not support the point that you wanted to make?