I keep hearing about the plight of the middle class—it's shrinking, it's worse off economically, it's stagnating, it's backsliding, etc. And there never seems to be a shortage of anecdotes to back up those assertions. [UPDATE: See my correction to this article dated 6/14/07.]
But I remember receiving some good advice a long time ago: Never argue by anecdote, because there's always an anecdote to back up any position one wants to take. [In critical thinking terms, it's a logical fallacy called hasty generalization, as follows: John Doe is middle class. John Doe is backsliding economically. Therefore, the middle class is backsliding economically.]
Some politicians don't even bother selecting anecdotes; they just state it as if it's fact. Here's a recent example:
[Hillary] Clinton argued that Democrats proved to be good economic stewards and protectors of the middle class during the administration of her husband, Bill Clinton, but said that, over the past six years, the Republican leadership has eroded the economic and income gains that many workers experienced during the 1990s.
Should I accept the premise? Did the middle class lose ground in the last six years? Somehow that didn't square with the admirable economic results I've been tracking, but I've been wrong before, so I decided to check the numbers at the US Census Bureau's website. Here's a chart showing the change, from 2000-2005, in the number of families at various income levels:
[To check the data, go to this page at the USCB, then choose "Family" for any of the income years 2000 through 2005, then choose "FINC-07: Income Distribution to $250,000 or More for Families."]
In 2005, the latest year for which these numbers are available, there were 77.4 million families in total, 5 million than there were in 2000. The distribution of families within income ranges shifted towards higher incomes during that time: there were 1.1 million fewer families in the $0-50K ranges, and 6.2 million extra families in the ranges above $50K. And that's "money income," which doesn't include the redistributive effects of taxation or government spending.
In short, the middle class appears to be moving up into higher income brackets [this implication still holds; see 6/14/07 article], according to the US Census Bureau. Why would anyone think that's a problem, and want to "fix" it? Why not propose growth-friendly policies designed to accelerate what's happening?
=============
End note:
I had to make an assumption about which levels make up the "middle class." [Politicians, you may have noticed, usually forget to define that term—let alone the word "rich"—so I made an educated guess that "middle class" is at and slightly above the median family income of $54,000.] I chose $25K increments; notice, however, that $50K increments are all that's available starting at the $100K bracket. [...I revised this assumption; see 6/14 article.]
I also decided not to attempt adjusting each year's money incomes for inflation [...a bad decision, see 6/14 article]; as noted in the graphic below, it wouldn't have affected the results much at all [...or so I'd thought], but it would have made my numbers much harder for anyone to check against the USCB source I used.
The first chart above showed what changed between 2000 and 2005. If you'd like to see the aggregate distributions for those two years, including the whole curve's shift towards higher income brackets, click on the following thumbnail.
Notice that, if the $25K increment were available in the $100K bracket, the curve would taper downward more smoothly.
Good work, but somehow I don't think Hillary is looking for better data!
Posted by: Kevin | 13 June 2007 at 06:49
Steve,
What percent of the population would you guess possesses just rudimentary critical thinking skills?
Fred Thompson made a statement I latched on to:
"Our problems are getting larger and politicians are getting smaller."
Posted by: Bob | 13 June 2007 at 07:01
The longer I study the issue, the more I agree with the libertarians on this ground: The best thing government can do to promote prosperity among all income levels is nothing.
Posted by: pawnking | 13 June 2007 at 07:12
Current dollars, lol.
Posted by: VG | 13 June 2007 at 07:34
If you'd like to see other data, I've recently been looking at how the distribution of income has shifted from 1995 to 2005:
http://tinyurl.com/2g858b
Going by age group, this post graphs out how many income earners there are at every income level from $0 to $95,000 (in 2004 US dollars) for both 1995 and 2005. Generally speaking, you see a large shift away from the lowest incomes toward higher annual incomes.
http://tinyurl.com/youp2x
When looking at age groups spanning ten years each over a 10 year period, you can see how the income earners of 1995 progressed to be the 10-year older group in 2005.
Once again, you see the distribution of income earners shift away from lower incomes to higher incomes. The only exception coincides with the age groups that are moving into retirement (where incomes are typically less than in one's working life.)
I should note that even with this change, 2005's retirees are well ahead of their 1995 counterparts.
http://tinyurl.com/26og9p
For those who would like to see the changes in percentages!
Finally, if you want to run numbers for these years yourself, you can using our comparison tool:
http://tinyurl.com/2gxshj
Posted by: Ironman | 13 June 2007 at 08:23
"I had to make an assumption about which levels make up the "middle class.""
Isn't the "middle class" usually defined as the middle quintile of income (+/- one decile about the median)? If so, the question basically boils down to how did median family real income do over that period.
The answer to that question is given by the census bureau page http://www.census.gov/hhes/www/income/histinc/h11ar.html
which shows that median income, in 2005 dollars, declined from 47,599 in 2000 to 46,326 in 2005. In which case, Clinton's sort of right, though for any longer period, the median family did better.
Posted by: Bret | 13 June 2007 at 09:48
Let's cut to the chase. All of the reasoning, rational thought and critical thinking amounts to this:
If the Democrats get control of the White House and maintain control of Congress they are going to raise the income tax rate on whomever they think is not paying enough. Facts do not matter one iota.
Not one.
Posted by: Bob | 13 June 2007 at 10:10
Ironman:
Excellent analyses; thanks for the links.
Bret:
First, if you have links to the government's official definition of "middle class" for persons, for families, and for households, I'd welcome the info, and will add them to my bookmarks.
Second, "Household" quintiles tell us less than most apparently think. For example, if a husband-wife two-earner household brings in the median income of $50,000, all they have to do to decrease the median is to get divorced, thereby turning themselves into two single-earner households averaging $25,000 each.
That's why I chose family data instead of household data -- even though family data still has some similar drawbacks.
Bob:
I fear you might be correct. A Hillary-Obama ticket would be a formidable class warfare emotion machine; if the Republicans compound that disadvantage by turning themselves into a permanent minority party by committing suicide on the immigration issue, we can look forward to at least 16 years of growth-unfriendly presidents dispensing goodies to their chosen recipients, such as corporate welfare for auto companies. It will be difficult for me to remain optimistic. Maybe a new party could emerge in the middle somehow.
Posted by: Steve | 13 June 2007 at 11:35
Steve - I agree that the Republican party seems to be trying to permanently give the Latino vote to the Democrats, which is very unfortunate.
Unfortunately, our plurality voting system makes it very difficult for a viable third party to emerge, since (1) most people don't want to "waste their vote" by voting for a minor party candidate (2) if a minor party becomes large enough, it will just split the vote for the major party it most resembles ideologically. For instance, if 10% of the population voted for the Green Party candidates, it would enormously weaken the Democrats.
We could empower the minor parties by switching to another voting system, such as approval voting, where people vote for as many candidates as they want. That way someone could vote for both Kerry AND Nader if they're so inclined, giving a better representation of people's real preferences.
Posted by: Ariah | 13 June 2007 at 12:42
Bret:
Steve's points regarding household data are why I opted for individual data over the household data as well. When you drill all the way down, income is always distributed from its source to specific individuals, and not households or families.
Regardless, the trend has been away from low incomes toward higher incomes, and especially so at the lowest end of the income spectrum toward the middle.
Posted by: Ironman | 13 June 2007 at 14:01
Ariah,
I'm not sure a third party is doomed. Frankly, I could see some real conservatives move away from the Republicans and form something new.
Steve,
16 years under the developing agenda of this batch of Democrats is unthinkable. You wouldn't recognize the country should that happen. Corporate welfare would be the least of you concerns.
If Hillary, Obama, Pelosi, Harry Reid and Murtha are running the show for 4 years I'll wager they will do enough damage voters won't want them back. But that four years will be real tough.
In either case you may want to talk to your granddaughters parents about keeping her out of the public school system. It will get a lot worse even if we only endure 4 years of Democratic rule.
And get ready for single payer health care. That's a sure bet if Hill gets in.
Posted by: Bob | 13 June 2007 at 19:36
Bob -
I don't think that any particular President can do a great deal of damage domestically. Lately we had Clinton (who the conservatives said would destroy the nation) and then Bush (who the liberals claimed would destroy the nation) and both turned out to be very moderate. Even if one had been an extremist, it wouldn't have been that important since everything has to pass through Congress anyway.
Posted by: Ariah | 13 June 2007 at 20:35
I am wondering if any data is available on the average number of people working in a "household" to earn the income. For example, in 2000, were there 1.3 workers/household, vs. 1.5 today?
Posted by: Matt | 13 June 2007 at 22:38
Ariah,
I agree about one President. My premise is based on a Democratic President AND retention of the current Democratic Congress.
They aren't JFK Democrats, sir.
Posted by: Ariah | 14 June 2007 at 06:30
Steve,
I loved your comments on Mr. Wizard. He led me to study science as well and inspired our generation of techically savvy leaders. We'll miss him.
Tricia
Posted by: Tricia | 21 June 2007 at 20:38