Friday, March 26, 2010

Lies and Damn Lies

One of my pet peeves is when people, especially those in the media, play loose with statistics just to make a point. For example: today's front page story in the Metro. A Chinese Canadian's life was saved by a stem cell donation, all well and good. But the real moral of the story is how it's hard for Chinese Canadians to find compatible donors.
It’s a “miracle” he found two matches in three months, Chu said, because the Chinese population is dramatically underrepresented on the national stem cell donor database, OneMatch. [...] Chinese donors make up only 2 per cent of those registered on OneMatch, compared to 82 per cent of registered Caucasian donors.
And therein lies the problem. The Chinese are "dramatically underrepresented" because only 2% of registered donors are Chinese. Meanwhile, an astonishing 82% of registered donors are Caucasians! What a travesty!

To merit the claim that the Chinese are "dramatically underrepresented," you need to have some context.
“For (Chinese Canadians) to represent such a small piece of the pie, that’s just embarrassing, considering how many Chinese people there are in Vancouver.” Chu added the under representation is likely due to old world values.
Does anyone else see the problem in comparing Chinese representation in the national stem cell donor database to the Chinese population of one of the most Chinese cities in North America?

According to Canada's 2006 Census, 83.8% of the population are "non-visible-minority" (Canadianese for Caucasian), and 3.7% of the population are Chinese. Compare that to their representation in the donor database, and both groups come up short about 1.7-1.8%. That would be within the margin of error of any random selection of Canadians.

In other words: The statistic is meaningless, with just a little bit of context. Don't get me wrong, I'm glad the guy lived. But he can rest easy knowing he doesn't need any scapegoats like "old world values" to explain his nonexistant "dramatic underrepresentation."


Thursday, March 4, 2010

Repeal the Individual Health Insurance Tax

The number of uninsured in the United States is estimated by the Census to be around 47 million people. Many of them are only uninsured because they lost their job, and their health insurance was tied to that job. When someone gets sick, they're more likely to lose their job but are in much greater need of health insurance. It's remarkable that so many Americans take that risk of losing their insurance right when they need it the most. Even so, about 93% of Americans who have private health insurance have that insurance through their employer. Now why would they do that when it's such a risk?

It's actually pretty straightforward. Back in the 50s, the federal government began giving Americans a tax break for purchasing health insurance through their employer. Any income you spend on employer-based health insurance is exempt from the income tax. However, if you want to purchase an insurance plan not provided by your employer, you have to pay the full tax on that income. Depending on your tax bracket, for most people that's a 20-25% penalty fee just for wanting to buy insurance that won't disappear if you lose your job.

The solution? Just extend the tax break. If you spend any portion of your income on health care or health insurance, you are refunded all tax paid on that income.

How much of a difference would that make? For one thing, the tens of millions of people who are uninsured simply because they're between jobs would be able to afford insurance in the interim. That could be as much as one-third to one-half of the uninsured at any given time. It's not a silver bullet solution, but I think it's as close as you can get on an issue this complex. This one tiny change could cover at least a third of the uninsured. I think it's a great place to start.

So what would it cost? The biggest cost would be to the government in lost tax revenue, but even that is not very much. There's about 12 million Americans currently paying this tax. The average annual premium for one-person non-employer insurance is about $3,000, and the average marginal tax rate in the US is about 22%. [Sources: insurance, taxes] That works out to about $8 billion in tax revenue that the government would be giving up with this policy, or $26 per American. That's pocketchange to a government that's running deficits hundreds of times this amount.

So why isn't it being considered? Well, I don't know. Maybe someone reading this can provide a counterargument? It's worth noting that the only play this has gotten in the Democrats' plan has been suggestions to eliminate the tax break entirely, and tax the entire country on their health plans (but that's only to pay for the trillion-plus cost of the rest of the program). That would be much more devastating to the average American, although it would eliminate the incentive for employer-based insurance.