Congress Loves the IRS, and We Should, Too

-- Posted by Neil H. Buchanan

In my FindLaw column this week, "Is the IRS the Most Trustworthy Agency in the Country? Even Republicans Seem to Think So," I move from defense to offense in discussing the role in society of the IRS and its employees. Two weeks ago, in my FindLaw column and Dorf on Law post, I responded to the February 18 suicide attack on the IRS building in Austin by saying, in essence: "Why aren't we more shocked? The IRS is wrongly vilified, and its employees are unfairly demonized for political gain. They are good people who deserve better." This week, my counter-intuitive point is that we have strong evidence not only that the IRS is staffed with many of our best and most professional citizens, but Congress -- no matter which party or ideology is in charge -- relies on the high quality of our tax enforcers to carry out its policies.

I will return to that argument momentarily. First, however, I thought I would share a few interesting facts that I have learned in following this story. When the subject is politics in Texas, things are never boring.

News reports shortly after the terrorist attack focused on the leadership of Lloyd Doggett, a Democratic Congressman who was identified as representing Austin. I checked the map of Congressional districts in the area, and it turns out that Austin has been sliced into three parts, with the other sections represented in Congress by two Republicans, Lamar Smith and Michael McCaul. This configuration, in turn, was part of the result of the absurd mid-decade redistricting that was engineered by former House Majority Leader Tom Delay. If you want to see a gerrymandered map, you will love Texas's congressional districts.

The major target of that redistricting plan in Austin, in fact, was Rep. Doggett, who nonetheless won re-election. When Doggett spoke out publicly against the attack on the IRS building, he was joined only by McCaul, with their third colleague notably missing. It is difficult to decide which was more surprising -- that one Republican would allow himself to be seen saying something positive about anything IRS-related, or that another could not bring himself to join in condemning a murderous attack on IRS employees in his backyard.

Perhaps more interesting, when Rep. Doggett introduced a resolution in the House condemning the attack, it passed by a vote of 408-2. Again, do we focus on the 408 women and men who were willing to say "violence is bad," or the two men who could not bring themselves to do so? The resolution, "Expressing concern regarding the suicide plane attack on Internal Revenue Service employees in Austin, Texas," honors the man who was killed (Vernon Hunter, a 68-year-old career public servant and Vietnam veteran), the first responders, and "commends Internal Revenue Service employees for their dedication and public service."

That last item was too much for one Congressman (whom I will not bother to name), who decided that the wording of the resolution implies that the tax code is "moral" when it is not. Even more bizarrely, the other "no" vote came from a Congressman who decided that the phrase "suicide plane attack" was the first step on a slippery slope toward stricter regulations on private aircraft. (I am not making this up.) Not all of these guys are playing with a full deck.

More than the two strange "no" votes on a resolution that should have been completely uncontroversial, the bigger worry is that this attack has been minimized and nearly forgotten so quickly. In fairness, the muted response is probably driven in part by the good luck that the deranged attacker was not able to kill more people. Still, it is difficult not to sense that the target of the attack is just too politically unpopular for anyone to care about. Other than (some of) the local Congressmen, no one could be bothered to focus on this tragedy.

All of which brings me back to this week's FindLaw column. Economists have a term called "revealed preference." Because traditional economic theory relies so heavily on the concept of "utility functions," which are mathematical expressions of people's preferences and desires, it would be great if we could determine each person's utility function. The problem is, we know that people will often lie if we directly ask them questions like: "If I took away 10% of your income, how many movies would you attend in a year?" The desire to convince someone NOT to take away 10% of one's income will be a strong incentive to overstate one's likely response. Moreover, even if people would not lie, they might not know how they would respond.

Because of the unreliability of this type of information gathering, it is sometimes helpful to ignore people's stated preferences and look at their revealed preferences. Although the definition of revealed preference is quite technical, the intuition is obvious: No matter what people say they prefer, we can (with some important limitations) see what they prefer by what they do. As so often happens in economics, a lot of mathematical firepower has been trained on the wisdom that we can derive from an old saying, in this case: "Actions speak louder than words."

In this case, we have politicians talking endlessly about how much they do not like taxes and the IRS. Many politicians specifically target the IRS and its employees as arrogant, incompetent, and out of control. If they really meant that, then we should expect to see those politicians taking actions consistent with their stated views. Instead, we have the opposite: Whereas every tax preference in existence could be administered by non-tax agencies (Health and Human Services enforcing the rules on health savings accounts, for example), we have every member of Congress repeatedly and enthusiastically voting for "tax cuts" that create the need for the IRS to enforce and administer yet more social policy through the tax code.

There is a reason, for example, that it became necessary a few years ago for Congress to adopt a consistent definition of "child" for tax purposes. Because of the adoption of various kid-friendly tax credits over the years, there were five inconsistent and overlapping definitions of child in the tax code. If Congress wants to, for example, make it less expensive to purchase child care, it enacts a tax credit for that purpose. We still have many different ways that we subsidize child rearing through the tax code, but at least we now have a consistent definition. (Until the definition was clarified by Congress, of course, the IRS was blamed for the complexity of their rules.)

This revealed preference for using the IRS to administer social policy is not a choice made in a vacuum. Congress uses the Code because its members like to call things "tax cuts." Even so, we at least have evidence that Congress does not hate the IRS enough to give up on this political gravy train.

As I argue in my column, however, this reliance on the IRS is a good thing. It is true that we could put the different social programs in different boxes on the big organizational chart, but doing so would squander the professionalism and expertise that we have gathered in the Internal Revenue Service. This is affirmatively a good way to run social policy, not an unfortunate side effect of Congressional opportunism.