But It's Really No Laughing Matter!

Greetings from the Left Coast - and Happy New Year!

A couple of weeks ago, when referring to the New York Sun article about discovering that Bill Clinton had blocked the release of some 2,600 documents from his presidency, when advisor Bruce Lindsey had said in a written statement, "Bill Clinton has not blocked the release of a single document," I made the comment that sometimes you just have to shake your head and laugh. But the fact is that it's really no laughing matter when politicians go out of their way to make statements that are technically true, but totally misleading.

A great example of this is the hand-wringing over "tax cuts for the rich" that I wrote about a couple of months ago. Since the "wealthy" pay the majority of the income tax that's collected (households making over $157,176 paid 58% of all income taxes collected in 2004), it would be impossible to have any kind of meaningful tax cut that wouldn't disproportionately favor those peolple - because they're paying a disproportionate amount of the tax! So, is it true that most tax cuts disproportionately favor the wealthy? Yes. Is that statement extremely misleading? Yes. And, by the way, these hand-wringers are the same people who don't think that allowing the "Bush tax cuts" to expire should count as a "tax increase," and will talk about a "reduction in government spending" when what they really mean is a reduction in the rate of increase in government spending.

Here's another one for you. As the campaign season progresses, expect to hear more and more from the Democratic candidates about the woes of the shrinking Middle Class in this country. This will, of course, happen against the media drumbeat of how we are facing an impending recession because of rising oil prices, a disappointing Christmas shopping season, and the subprime mortgage crisis. (And, before you take me to task over my media-bashing, please answer this: A recent Gallup poll found that 84% of Americans are satisfied with their personal lives, and 59% say they are "very satisfied." Yet only 27% were satisfied with the way things were going in the country as a whole. How can you explain that disparity other than to lay it at the feet of those in the media who, day in and day out, tell us how terrible things are going?)

Is the Middle Class shrinking? Yes, it is, depending on how you define "Middle Class." But the reason might surprise you. One definition of "Middle Class" is households making between $30,000 and $75,000 per year. In 1979, about 53% of American households fell into that category. Roughly 21% made less than $30,000 per year, and roughly 26% made more than $75,000. Between 1979 and 2004, the percentage of households making between $30,000 and $75,000 per year dropped by roughly 13%! Wow! But did they make less money? No - the percentage of households making between $75,000 and $100,000 per year went up by about 1%, and the percentage of households making over $100,000 per year went up by roughly 13%! The percentage of households making below $30,000 per year was basically unchanged.

(This information, by the way, was in an article by Stephen Rose that was published back in September, 2006, by The American Prospect, which is not a particularly conservative publication.)

So, is it true that the Middle Class (as defined above) is shrinking? Yes, it is. Is it also extremely misleading when used, as it usually is, to imply that Americans today are worse off than they were a few years ago? Yes.

And lest you think that the numbers are just being skewed by the dollar amounts involved, the median household income in 2004 for families beween the ages of 25 and 59 was $63,300. If you just consider married households, the median income jumped to roughly $70,000, and if you consider two-earner households, it was nearly $80,000.

"Median," by the way, means that there were just as many households above that number as there were below that number...which brings me to another point: Be very careful when you hear someone use the word "average." There are four different definitions of "average" - the "mean," the "median," the "mode," and the "midrange." Consider the following set of numbers: 2, 2, 3, 9, 15, 63, and 75.

• The "mean," or, more accurately, "arithmetic mean," is what most people think of when they hear the word "average." That means you add up the values and divide that total by the number of values in the set. So, for the example above, the "mean" would be 2 + 2 + 3 + 9 + 15 + 63 + 75 (which equals 169) divided by 7, because there were 7 numbers in the example. That works out to 24 point something.
• The "median" means the value in the set of numbers where there are just as many values higher as there are lower. So the median value of the set is 9, because there are three numbers in our example that are greater than 9, and three that are less than 9. The "median" is a good value to consider when looking at things like income distribution, as in the example above, because it gives you the "midpoint" of the range of values. (Before you ask, if there is an even number of values in the set, the median value is the mean of the middle two values.)
• The "mode" is the value that appears most frequently. So the mode of our example is 2, because the number 2 appears twice in the example, and all of the other numbers appear only once. If no value appears more than once, there is no mode, and if multiple values are duplicated, a mathemetician would say that the set is "multi-modal."
• The "midrange" is the arithmetic mean of the highest and lowest values in the set. So the midrange of our example would be (2 + 75)/2 = 38.5. This can be extremely misleading, because the term "midrange" would suggest that it refers to a value somewhere in the middle of the range of numbers, when in fact it can be even more skewed by extreme values than the arithmetic mean of the entire set, as we see in our example here.

So, a statistician could legitimately state that the "average" value of our example is either 2, 9, 24 or 38.5, and be literally correct. Now, when we're talking about a fairly small sequence of numbers, as we are here, your intuition would probably say, "Wait a minute..." if you heard someone say that the average value is 2. But if they're talking about things like average income, or average home prices, you're probably not going to intuitively know which "average" they're talking about, because you don't have the raw data at your fingertips, and even if you did, the number of data elements is so massive that it still wouldn't be intuitive. So it's important to know what definition of "average" is being used, particularly when the number is being used to justify someone's point of view, even more particularly when that someone is a politician.

It brings to mind the old cliche: "Figures don't lie, but liars can figure!" And so can politicians.

Thanks for listening.