You might remember my post last week about [conservatives who can’t subtract][subtract]: in particular, about how a conservative blogger who goes by “Captain Ed” attacked John Edwards for saying there are 37 million people in poverty in the US. It turned out that good ol’ Ed wasn’t capable of doing simple subtraction.
You might also remember a post about [lying with statistics][liar], discussing an article by Tim Worstall, who quoted a newspaper piece about abortion rates, and tried to misuse the statistics to argue something about sexual education in the UK.
Well, Tim (the target of the second piece) was pretty ticked off at my criticism; and so now, he’s back – but not with a defense of his own piece (he tried that already), but with [a response to the criticism of the first piece][liar-subtracts]. Of course, he tries to defend our good captain not by defending his math – that is, by claiming that *the point that he made* was correct; but by moving the goalposts, and claiming that the *real* point about the piece wasn’t to call Edwards a liar, but to pretend that he was *really* making an economic argument about whether or not people below the povertly line are really correctly described as poor – because, lucky duckies that they are, they get some money from the earned income tax credit! So obviously they’re not *really* poor!
>”Thirty-seven million of our people, worried about feeding and clothing their
>children,” he said to his audience. “Aren’t we better than that?”
>
>and the link is to this table at the US Census Bureau which indeed states that
>there are some 37 million or so below the poverty line.
>
>Right, so that must mean that there really are 37 million poor people in the
>USA, right? So what’s Tim bitchin’ about? Well, how about the fact that those
>figures which show 37 million below the poverty line do not in fact show that
>there are 37 million poor people? Weird thought I know but nope, it ain’t true.
>
>For this reason:
>
>The official poverty definition uses money income before taxes and does not
>include capital gains or noncash benefits (such as public housing, Medicaid,
>and food stamps).
>
>What is being measured in the first definition of poverty is how many people
>there are below the poverty line before we try to do anything about it.
This is what those of us who know anything about logic refer to as a “non-sequiter”. That is, it’s a conclusion that has nothing to do with what came before it. It’s one of the oldest and sloppiest rhetorical tactics in the book. (Literally – I’ve got a book on classical rhetoric on my bookshelf, and it’s cited in there.)
Edwards was talking about the division of wealth in the US: we have people like the CEOs of big companies taking home unbelievable amounts of money, while at the same time, the income of the people in the middle class is declining slightly in real terms, and the income of the people at the bottom isn’t even approaching what they need to get by. There are 37 million people below the poverty line in this country in terms of their income. Some portion (not specified in the only source Tim and the captain cite) of those people are working, and despite working, are still not making enough money to get by. This is indisputable: there are many people in this country who are working, but who still require government assistance just to pay their bills. That’s what Edwards said.
What does that have to do with whether or not the government gives them some token assistance? The point is that our economic policies quite deliberately *refuse* to do anything to help the people on the bottom of the economic ladder become self-sufficient. Witness the recent refusal to even allow an open debate in congress on increasing the minimum wage, even while the members of congress gave themselves a raise. A person with a family, working full time for the minimum wage is left *below* the poverty line. But it’s not considered an important issue by the people currently running our government.
[subtract]: http://scienceblogs.com/goodmath/2006/07/subtraction_math_too_hard_for.php
[liar]: http://scienceblogs.com/goodmath/2006/07/lying_with_statistics_abortion.php
[liar-subtracts]: http://timworstall.typepad.com/timworstall/2006/07/good_maths_and_.html
Good Math/Bad Math
The beauty of math; the humor of stupidity.
I am startled to learn that “[t]he official poverty definition … does not include capital gains …”
But I can guess why — it’s so that, when Republicans cut the capital gains tax to help the poor, they won’t get any credit for it!
Now, we need an extra panel in Lucky Ducky comics in which LD gloats “Ha, ha! you only think I’m poor, while I’m actually rolling in non-cash benefits!”
And those who know anything about South Park call it “the Chewbacca Defense” 😉
I think one of the reasons why CEO compensation packages raise people’s ire is that oftentimes they are not warranted; the managers of poorly-performing corporations too often make as much as those managing successful companies. I don’t see a problem with an excellent manager at a successful company making a lot of money. You work hard, you get rich, right?
Mark
Your point about Edwards’ maths is fair enough, but Tim is presumably talking about the interpretation of those statistics. I can’t see why it is non-sequitur. It all depends how much those other factors such as capital gains and noncash benefits amount to.
I would add that is also essential to get a breakdown by age. Most of us start off our careers in the bottom decile or two of income.
It may be that there is a real problem with poverty in the USA but surely it is relevant to point out subtleties in the interpretation of the statistics?
Finally, while we are being pedantic, you seem to use “wealth” and “income” interchangeably above. They are, of coures, different things.
Mark,
You’ve completely left out the actual point of my post:
“Where this actually becomes important is here (from Publius again):”
Publius goes on to argue that the lower poverty rates of the 60s and early 70s were caused by the Great Society programs. Which are, as I point out, AFDC, welfare, HUD, Medicaid and (added later) the EITC. Only one of those, welfare, is actually a cash transfer before tax and so the others do not actually get included in our statistics when calculating the number in poverty. It is therefore impossible, using these statistics (it may well be with others, but not with these) to state that AFDC, HUD, EITC and Medicaid reduced the poverty level.
As you so forcefully pointed out before:
“So – what we’ve got here is a nut making an argument that’s trying to use statistics to justify his political stance. However, the real statistics, in context, don’t say what he wants them to say. So – as usual for a lying slimebag – he just selectively misquotes them to make it look like they say what he wants them to.”
I am saying the same thing, that Publius is selectively misquoting statistics in order to make them look like they are saying what he wants them to.
Why attack me for this? Shouldn’t I be getting a pat on the back from a non-partisan defender of the sanctity of numbers and truth?
Tim:
This post isn’t talking about what Publius did or did not say. I haven’t even *read* Publius’s post.
What I was replying to is specifically what you’re doing in your post. Let’s review briefly, shall we?
1. Captain Ed posted an article claiming that Edwards lied about the poverty statistics, because Ed was either too stupid to read a chart, or too stupid to to a subtraction correctly.
2. I responded, pointing out that Ed was a moron.
3. You posted a defense of Ed’s article, claiming that the article was still valid, even though it’s main point was that Edwards was lying; when in fact, Edwards was correct.
4. I posted a response to your defense of Ed, pointing out that your defense was a non-sequiter: Ed said Edwards was lying; he wasn’t. Arguing about the the correct interpretation of poverty statistics is a *different issue*. Either Ed was telling the truth the Edwards was lying about the poverty statistics; or Ed was wrong in saying Edwards was lying. The fact is, Cap’n Ed was wrong.
I’m not particularly interested in getting into a political argument about the merits of the Great Society programs; that’s an argument that inevitably involves more political philosophy than math, and I try to keep this blog as focused on the mathematical issues as possible.
But defending Ed’s assertion that Edwards is a liar for citing fabricated numbers by shifting issues from whether or not Edwards cited the real statistics, to whether or not those statistics mean what Edwards or Publius said they mean is a total non-sequiter. Ed called Edwards a liar. I, and several others, called him on it. He was wrong. Edwards wasn’t lying. He cited the numbers that he claimed to be citing correctly.
pg:
There are two things about CEO compensation packages that bother me.
One is the disproportionate nature of the it: the CEO and top-level executive staff are rewarded with mind-bogglingly large payments. When you look at the sheer size of their packages, and compare their rewards with the rewards of other people at the company, there’s a total disconnect. When the normal employee of the company – the people who produce the products and services that the company sell – get *less than* a cost-of-living salary increase, but the CEO receives a 20% increase in his compensation package, that’s wrong. The benefits should be shared. Many CEOs work hard, and they deserve to be rewarded for that. But they shouldn’t be rewarded for it *at the expense of* rewarding other employees at the company who work as hard.
The other problem with executive compensation is the disconnect between the performance of a company and the amount payed to the executives. Many CEOs get raises, bonuses, and stack options/grants every year, regardless of the performance of the company. To cite an example from a couple of years ago; Lucent Technologies was a highly profitable company when it first separated from AT&T. Then an idiot named McGinn became the CEO. McGinn ran the company into the ground – he made one utterly boneheaded decision after another. It’s arguable whether he made a single correct decision as the CEO of Lucent; as a result, Bell Labs is pretty much dead and gone, and Lucent’s stock price fell from the $80 range at its peak to *less than a dollar*; it’s since risen to hover between $2 and $5, mostly towards the lower end. McGinn *at his firing* was given a *twelve million* dollar severance payment, along with nearly a million dollars per year in cash pension. Lucent’s employees watched their pensions shrink, and their retirement benefits get cut to ribbons; but McGinn is *still* getting a million per year. That’s all too typical of how things work.
Mark: I appreciate that your main point was about the Edwards vs. Captain Ed “did he lie” question.
But then at the end you conclude with: “Witness the recent refusal to even allow an open debate in congress on increasing the minimum wage […] A person with a family, working full time for the minimum wage is left below the poverty line. But it’s not considered an important issue by the people currently running our government.”
The “minimum wage” concept is distinct from whether society cares about the poor. This is more of a political (and economic) issue than a math issue, but it’s unfair of you to conflate the two.
A different way of looking at it is that a minimum wage is a restriction on possible valid contracts that two adults might voluntarily choose to have between themselves. Almost any economist would tell you that such well-intentioned interference into an economy almost always make society worse off in the long run.
Another point to keep in mind is that not everyone affected by minimum wage laws is “a person [sole household breadwinner, presumably] with a family, working full time” These laws also affect high school students living at home, working at McDonald’s over the summer to earn some spending cash. It’s far less clear that society is better off by forcing those wages to be raised above the market-clearing price.
Finally, the real sadness of poverty is long-term lifetime lack of income/wealth (especially across generations). So it’s important to know how mobile that economic class is. Just reporting absolute counts from year to year is only the first crude look into whether this is a problem that needs to be fixed.
But even if so, the minimum wage is the wrong tool for the job.
Don,
While not wanting to turn this thread into a argument on minimum wage, there are a few points to be made in favor of it.
First, one of Keynes great insights was that consumption rises with income. If you want people to purchase more goods, getting them more money will do so. This results in an increased demand, leading to a growing, stronger economy and the market value of the labor will settle at the higher minumum wage.
Second, labor, prior to organization and goverment regulation, has historically been at a surplus to any level of demand. What this means in layman’s terms is that the market level of labor (unskilled) is always lower than the consumption level which would drive demand. (Skilled labor is another animal and is not important to minimum wage arguments because the market value of skilled labor is always higher than unskilled labor.)
Third, in an industrial economy, most job positions for unskilled labor are rarely based on the labor market but are instead based on system needs. An assembly line needs X number of laborers, X-1 laborer means the line stops, X+1 laborers does not make the line go faster. This is true in fast-food as well. You never see a fast food resturant saying that they will hire until the wages account is empty, instead they will hire until the shift is filled.
Forth, not all increases in minimum wage result in descreased employment, as classical economics suggests. A recent example comparing farm wages in Kansas and a neighboring state found that when the minimum wage increased in Kansas as farmers actually hired more laborers because their yields increased more rapidly than their cost in wages increased. This seems to discount point three, but remember point 3 was about jobs where the labor is specialized rather than farm labor where the labor is general.
Fifth, many jobs are not included in all minimum wage laws. I well remember that the farm work I did as a teenager was not restricted by minimum wage. The recent passage a an increase in the Michigan minimum wage excluded a wide variety of laborers. I don’t necessarily agree with what the Michigan legislature has done, but it does refute your claim that a minimum wage law affects everyone.
While there may be problems with minimum wages from a pure classical economic view, in a modern understanding of economics there are more things to consider than classical supply and demand.
Cheers,
-Flex
“1. Captain Ed posted an article claiming that Edwards lied about the poverty statistics, because Ed was either too stupid to read a chart, or too stupid to to a subtraction correctly.
2. I responded, pointing out that Ed was a moron.”
And I said that he couldn’t do subtraction and described this as “silly, silly”. That is hardly a ringing endorsement of Capt Ed.
“3. You posted a defense of Ed’s article, claiming that the article was still valid, even though it’s main point was that Edwards was lying; when in fact, Edwards was correct.”
No, I did not try to defend Ed’s article. I didn’t even read it, only that part that you and Publius quoted. I was much more interested in using all three pieces to illuminate an ecnomic point, that the method by which poverty is generally measured in the US (and which Edwards was quoting from) is a pre-amelioration number. That’s a point which many don’t actually know and is worthy of wider dissemination.
“4. I posted a response to your defense of Ed, pointing out that your defense was a non-sequiter: Ed said Edwards was lying; he wasn’t. Arguing about the the correct interpretation of poverty statistics is a *different issue*.”
That is the issue my post was about.
“Either Ed was telling the truth the Edwards was lying about the poverty statistics; or Ed was wrong in saying Edwards was lying. The fact is, Cap’n Ed was wrong.”
Indeed, as I said in my introduction, Ed was wrong. Then I go on to point out why the statistics Edwards is using can be misleading (no, not are, just can be) and then, referring to Publius, give an example of the use where they are indeed misleading.
About the Great Society programs. I agree, more political philosophy (or perhaps economics) than maths. However, in any debate, as I’m sure you would agree, the maths has to be right before those other subjects are delved into. In this specific case one cannot use that specific definition of poverty from the Census to prove (or disprove) the validity of the Great Society programs at diminishing poverty. For they do not include the effect of the Great Society programs upon poverty. That is a math point and one I thought, you being interested in the math, you might appreciate someone pointing out.
Flex, do you have a link to that Kansas study? The only one I’m aware of that gives similar results is Card and Krueger and that was about fast food workers in NJ and Pennsylvania. As that result is counter-intuitive (from the point of view of classical economics) I’d love to see the paper itself.
Tim,
I don’t have a link as it was discussed in one of the business economics courses I took last year. I will try to find details on the study for you. I have e-mailed my professor (whom I have been discussing other economics questions with for some time) for additional referances. When I get home from work, I’ll also dig up my notebook and find the referance in there.
It is possible that I got a garbled version of Card and Krueger. I don’t think so since we discussed that one on a different occasion, but I’m relying on memory without my notes in from of me.
Either way, even if I have made a mistake on the study the Card and Kruegar study supports my point that the effects of minimum wage adjustements are not entirely explicable with classical economics.
Cheers,
-Flex
Tim, and others here,
I apologise for the mistake I made earlier. I was mistaken about a Kansas study on the unemployment effects of raising the minimum wage. I can find nothing in my notes, and my Professor doesn’t recall mentioning one.
So my point five is invalid and it’s entirly due to my missunderstanding.
Apparently I was being reminded of the Card and Kruegar study. But my memory was faulty.
-Flex
Sorry to be pedantic about spelling, but it’s really bugging me — it’s “non sequitur”, not “non sequiter”. (Blame my comics page for my knowing that one.)