Economic Mythmaking: Tax Cuts and Income

Matt Yglesias approvingly posts Jason Furman's testimony before the Ways and Means Committee.

While Furman's testimony deals with a variety of critiques of the Bush administration tax cuts (some of which I agree with), one of the most dramatic claims he makes is that if the tax cuts were offset by reduced spending, the bottom 4 quintiles of the population would actually experience a reduction in income. This is a rhetorically powerful claim and is likely to become commonly cited in critiques of tax cuts. It also happens to be based on some very unrealistic assumptions which dramatically increase the negative income effects projected.

Table 3 from Furman's testimony (a copy of Matt Yglesias's copy I have posted here) details his projections broken down by quintile and separates out the top 1%. The change in income is determined by 3 factors - the Static Tax Cut consisting of direct reduction in taxes from the cuts, the Income Change, representing the expected change in income based on higher economic output due to lowered taxes, and a Finance Cost, dealing with the cost of offsetting the revenue reduction. And in this final factor (the sole source of negative income effects) lies the flaw in Furman's analysis.

Furman arrives at this number by assuming that every dollar cut from spending is a dollar reduced from someone's income received as a result of a transfer. Not only does this ignore the distinction between consumption and transfers that is made in the Treasury Department's analyses (as Furman admits in footnote 13) , it would require that all of the spending cut come from transfers (and none from obvious non-transfer items like military spending, administrative spending, foreign aid, etc) and that the transfers in question would have no overhead costs.

A more realistic assumption regarding how much the spending cut reduces transfers neutralizes Furman's claim. If transfers were reduced by 40% of the total spending cut (as would happen if half the cut were from transfers and the transfers had a 20% overhead on average - a far more realistic assumption than Furman's and still very charitable to his case, given the composition of the federal budget and the political difficulties of cutting transfers), every quartile except the bottom one experiences an increase in after tax income, including reduced transfers. This fits with the intuitive expectation that spending cuts that included transfers would free up more income for most people but would reduce the incomes of people at the bottom of the income distribution who utilize "safety net" programs and pay relatively little in taxes. This contrasts sharply with the narrative Furman and Yglesias are promoting, where the majority experiences a net income loss as a result of tax cuts offset by spending cuts.

The irony of this all is that Yglesias uses Furman's testimony as part of a series of posts haranguing conservatives willingness to use shoddy economic analysis to reach ideologically simpatico conclusions when buying into the myth that we are on the right side of the maximum of the Laffer Curve (more on how that particular illusion is constructed later).

Or a good weekend of binge drinking

3 IQ points, that is. The statistically significant, but not quite practically, significant difference in the IQs between eldest Norwegian men and their siblings. Jake Young says so what here.

To put this into perspective, the traditional standard deviation of IQ is 15 points. In the standard deviation for height in inches in the USA for 19 year olds, is about 5 inches. So, 3 IQ points can be compared to to one inch of height for American 19 year olds.

And now a significantly less snarky post about Prof DeLong's Plan

The plan is here if you haven't seen it. Aside from disagreements over how much health paternalism the government sound engage in to keep down costs (Brad favors A LOT), the funding scheme also raises some issues.

First off, 20% of your income can be a lot of money depending on your situation. This doesn't matter that much for people who do have insurance who will recoup most of the cost of their previous insurance plans. However, this will be a major drain on the liquidity of those who don't currently have insurance who will have to fork over 20% of their current income up front. Three quarters of this (15% of total) goes into a HSA, but for someone who is unlikely to need much non-emergency care (young males) or will be mostly covered by the free preventative care (young females), this money is effectively tied up until the next tax rebate.

For most people who have insurance, it will be a step down in coverage. This may actually drive savings by making costs more transparent to the consumers, but on the other hand, it means that there will still be an unsatisfied demand for insurance due to risk adversity (15% of most people's income is a lot for them), so unless it is outlawed, the demand for some private insurance will still be out there. I'd also like to point out that insurance companies are probably more efficient drivers of cost suppression than individuals, since assessing medical necessity requires specialized knowledge and they both have a financial incentive.

Finally, health care spending, the distribution of which is neatly displayed in Figure 1 here is dominated by the top quintile, which accounts for 80% of spending. Accounting for a mere 3.4% of costs, the bottom 50% would mostly be using just the preventative care and will have a bunch of money pointlessly locked up in an HSA. On the other hand, the top 5%, where 49.2% of the costs lie, will tend to be well past the maximum out of pocket. The area where the least cost-effective interventions lie is also where government rationing is going to be the dominant factor, which you may take in whatever direction your presumably strong preexisting opinion of the efficiency of government health care rationing guides you. This is also unlikely to be covered by the 5% of income allocated for both it and the free preventative care (aside: Free preventative care isn't that expensive and probably isn't a horrible idea if you're going single payer. Of course, insurance companies tend to cover it well too, so there's nothing special about single payer here either.), so we'd be on the hook for some general budget tax increases on top of the 5-20% extra on income.

So in other words, the Pew Internet & American Life Project is calling me a geek.

Commenter MP raises an interesting question "then why the gun fetish? are kids more likely to be killed by gun accidents, or criminals?"

It turns out Matt Yglesias (as roughly represented by the cohort of white males dying in the District of Columbia 18 to 34 years old between 1999 and 2003) has an equal probability of dying of unintentional injuries (presumably including not only gun accidents but car accidents) and homicide (thanks CDC) - 18.2 in 100,000.

So, while I am strongly against the arming of suburban children, someone in Matt Y's situation may indeed have a pragmatic interest in exercising their 2nd amendment rights.