Conservative claims about stimulus are wrong

The debate about the stimulus was fairly vacuous from the very start. Pundits with apparently little understanding — or worse an agenda to misrepresent things — would chime in on whether it worked. Of course, posing the question in terms of working or not working is already massively confused. Any serious discussion on the stimulus would center on the impact it had.

The Congressional Budget Office put out another report this week estimating the impact of the American Recovery and Reinvestment Act in the last quarter. The CBO estimates that in the second quarter of 2011, the stimulus increased real GDP by 0.8-2.5 percent, lowered the unemployment rate by 0.5-1.6 percent, and increased employment by 1.0-2.9 million people. There are of course other academic studies on the stimulus, the majority of which — as you would expect — point to some positive impact of the stimulus.

Conservative commentators deny this reality with predictably weak and undeveloped arguments. Admittedly, it may not be worthwhile to even engage the mainstream conservative pundits because they are almost certainly not arguing in good faith. Nonetheless, one thing that consistently pops up in their talking points is that the stimulus “did not work.” What is meant by this is somewhat unclear. At times, it appears that because unemployment continued to rise after the stimulus, they take that to be proof that it did not have a positive impact.

This sort of point is just analytically confused. The proper way to measure the impact — as the above studies do — is to estimate what would have occurred without the stimulus and what did occur with the stimulus. If unemployment would have risen by 3 percent due to the state of the economy, but only rose 2 percent with the stimulus, then any serious analysis counts that as a positive.

Other conservatives seem to take the question of whether it “worked” a bit too seriously, believing that the analysis called for in the case of the stimulus is truly binary. This of course is false. Decreasing umemployment by 3 percent is better than decreasing it by 2 percent which is better than decreasing it by 1 percent or not at all. There is no point at which one would say it worked or did not work; there is only bigger and smaller impacts.

What is odd about the conservative framing of this issue is that it is unnecessarily extreme. I realize from a public relations point that it is probably more compelling to send out messaging that talks about the stimulus having no impact or a negative impact. But given that it is undeniable that it had some positive impact relative to the counterfactual world in which no stimulus occurred, any serious right-wing critiques would focus on the trade-offs of the stimulus. Was 700 billion dollars of debt worth a 1 million jobs, a couple of percentage points for the GDP, and a slightly lower unemployment rate?

Debates could certainly be had on that front. But instead of taking that more honest approach to critiquing the stimulus, the right wing messaging continues to make claims that are demonstrably untrue. In a more informed society with better media outlets that sought truth instead of artificial balance and sensationalism, these misrepresentations and lies about the effects of the stimulus might harm the right wing. But in our present world of public relations dominated politics, they almost certainly do the opposite.

That is not to say that the stimulus was all that great. As most Keynesian economists at the time were pointing out, the size of it was far smaller than it should have been for the impact it was trying to achieve. Providing a Keynesian boost to aggregate demand that is half the size it should have been is kind of a strange policy. At the time, defenders of the administration remarked that the small size was not a problem because there was nothing limiting the injection of more stimulus later — a reassurance that has proven politically misguided.

With that said, as much as the left has rightly harangued the administration’s too weak stimulus policy, right wing commentators still manage to make their position on the stimulus even sillier. The reality is that the stimulus — whether it was too small or too expensive — has had a slight positive impact, all the hand-waiving, contrary assertions of the right wing notwithstanding.

Income tax claims are confused

The claim that half of Americans do not pay income taxes has been bandied about by Republican presidential candidates lately. The three front runners — Michelle Bachmann, Rick Perry, and Mitt Romney — have all thrown out the statistic in stump speeches in the past few weeks. It seems odd that Republicans would complain about individuals paying low taxes given that their opposition to taxation is probably the core feature of their party’s brand.

Of course, these candidates do not actually oppose taxation in general, just the taxation of the super-rich. Trotting out the claim that half of Americans do not pay taxes is no more than an effort to generate a counter-narrative that the wealthy are actually the ones being oppressed. Unsurprisingly, this attempted narrative is as confused as the income tax claims that it rests on.

Most of the commentators who have responded to the income tax talking point have rightly pointed out that income taxes are not the only taxes. The poor pay a higher percentage of their income in federal FICA taxes as well as state and local taxes. In fact, one of the motivations for the Earned Income Tax Credit — a credit which helps contribute to the low nominal income tax payments of the poor — was to counteract the regressive nature of the Social Security taxes.

But this rebuttal avoids the main driver of the income tax confusion: the total misunderstanding of what the tax code is. The reason citing numbers about the percentage of people who pay income taxes is so foolish is that the tax code serves a dual function: it is used both to generate revenue and to implement certain government programs.

For example, take the mortgage interest tax deduction. At some point, the Congress decided that it wanted to create a program to pay some of the loan interest for those who held mortgages, ostensibly to encourage home ownership. The government could have implemented this in one of two ways. It could have created a Mortgage Interest Agency (MIA) that sent checks to those paying a mortgage, or it could have allowed them to deduct from their taxes an equivalent amount of money. Fiscally and economically speaking, these two actions are identical.

Even though these two approaches are identical in all relevant ways, the deduction-oriented route leads to lower nominal tax payments than the MIA route. No one could seriously maintain that this distinction matters, but it is precisely this kind of distinction that the claims about half of the population not paying taxes rests on. What is actually going on in all of these cases is that government programs being implemented through the tax code — deductions for the elderly, exemptions for parents and the poor, and other various tax credits — transfer the same amount or more money than the households pay in taxes.

The objection to the non-paying of taxes then is totally confused. Is the objection really just that nominal tax payments are zero? If that is the case, perhaps we ought to just create an Elderly Benefits Agency, Earned Income Agency, Child Benefits Agency, and so on to replace the more efficient way we currently enact cash transfers through the tax code. Shifting cash transfers to check-writing agencies would ensure that everyone paid their exact marginal rate.

Presumably these Republican candidates are not just objecting to the way in which the programs are carried out, but something more substantive than that. However, phrasing their objection in terms of zero tax payments leaves their actual objection completely unclear.

As mentioned before, what is going on in these cases they are objecting to is that poor and old people are receiving more from government programs than they are paying in taxes. Is the objection that a person should not benefit more from government programs than they pay in taxes? That is the only thing remotely coherent, but it is so extreme a view that I am hesitant to attribute it to even figures as right wing as Bachmann and Perry.

The logical consequence of this sort of view would make even a flat tax in which everyone paid an identical percentage objectionable. After all, in a flat tax, low income earners still necessarily benefit more from government services than they pay in, at least in a superficial sense. Someone making 100x what someone else made would pay 100x more while receiving the exact same benefit from military services, police services, emergency services, infrastructure projects, and so on. Even though this kind of spending does not manifest itself as cash payments like tax expenditures do, that does not make it any different.

Even if all individuals paid a flat tax and received the same government services, the bottom half of income earners would still be paying less in taxes than they get back in government services. So, under the only coherent interpretation of the no-tax objection, this ideal Republican world would still be equally objectionable.

No matter how you try to cut up this talking point, it winds up empty or confused. These candidates either do not understand how the tax code functions, are being intentionally misleading, or have embraced full-fledged Randian insanity more than I have given them credit for. In any case, the bizarre focus on zero nominal tax payments is so confused that it is difficult to even grasp what if any real point is being made.

Study provides more evidence against the bad school hypothesis

A new study was released in the last week by MIT and Duke researchers about the efficacy of exclusive, magnet schools. In the working paper, the researchers compared the achievements of students who narrowly qualified for magnet schools in New York and Boston against those who narrowly failed to qualify for the same schools. The researchers found that — with the exception of a couple of marginal cases — the magnet schools had little effect on student achievement.

The clever methodological approach of this paper finally provides an apples-to-apples comparison to evaluate the purported effects magnet schools have on learning. It is usually taken for granted that magnet schools are better schools, and then efforts are made to explain why that is the case. Those explanations typically revolve around claims that magnet schools seek out and attract better teachers, and that the schools benefit from the positive peer effects of more motivated and better students.

The underlying assumption of these discussions — that magnet schools are actually better — has never been properly demonstrated. This assumption of superiority has typically been based on crude comparisons between the average achievement of students at magnet schools and regular schools, comparisons that are basically useless. Obviously magnet schools will on average do better than regular schools; magnet schools intentionally select students who on average do better than other students. This study cuts through that useless comparisons of averages, and shows more than any other study I have seen that better students make magnet schools better, not the other way around.

By itself, this study does not appear that significant, but when you consider the battle raging about education policy, it actually has major significance. This study provides more support against the idea that bad and good schools are the primary drivers of educational failure and success. If even magnet schools are performing no better than the often-criticized New York City public schools, the bad school hypothesis becomes even more implausible.

Despite its increasing implausibility, it is the bad school hypothesis — and the derivative bad teacher hypothesis — that is the operating assumption of the entire education reform movement. By modifying the way schools are administered and who teaches in them, education reformers try to argue that we can close the educational achievement gap without ending the socioeconomic inequality that has generated it.

Although this aim is somewhat noble and plays into the trope of the intrepid teacher getting through to disadvantaged kids, if the hypothesis upon which it is based is false, the efforts are ultimately in vain. We hurt poor students by funneling them into gimmicky schools that try to ameliorate the educational problems inherent in poverty without actually getting rid of poverty.

In some ways I understand the urge to find a way to achieve educational egalitarianism without achieving economic egalitarianism. After all, doing so might finally get us to a point where we can reach that elusive American Dream stage in which wretched poverty is acceptable because no one has a disproportionate chance of being the one who winds up living in it. That does not strike me as a terribly noble goal, but it at least strikes me as one that fits with dominant American cultural norms.

However, as much as reformers might not want it to be the case, it appears to be true that the leftists were right all along: equality of opportunity is incompatible with massive inequality of wealth and income. Unfortunately (not really) we cannot pay a janitor poverty wages without also negatively affecting her child’s life chances.

The sooner we realize this and abandon the bad school hypothesis, the sooner we can begin to move attention back to the more serious critiques of social and economic structure that might actually yield real solutions.