The Congressional Budget Office (CBO) has a revealing report out about hidden spending through the U.S. tax code. Instead of simply lowering rates, the U.S. tax code tries to encourage and discourage certain behaviors by writing preferences into the tax code. So rather than giving people money to buy a house or buy health insurance, the tax code actually gives people a tax break on their mortgage or their employer a tax break if they provide health care.
These tax expenditures are extremely important for two reasons. First, they are worth a whole lot of money, about 900 billion a year according to the CBO. Second, unlike appropriations spending, Congress doesn’t actually control how much money is spent subsidizing mortgages, healthcare, charity, and so forth. This also means they are a less transparent and less accountable method of governing.
This leads to some interesting outcomes. For example, as Ezra Klein at points out at Wonkblog: “Among the more mind-blowing facts about the health-care system is that the tax break we give to employer-provided insurance dwarfs the cost of the entire Affordable Care Act — and, if you want to take the concept a bit further, this means those of us who don’t get insurance from our employers are being forced, even mandated, to pay for those of us who are. But this break is largely uncontroversial in American politics, while subsidies to help people who can’t afford health insurance are extremely controversial.”
As it turns out, it’s not only health insurance where the tax code favors the wealthy. Overall, 17 percent of tax expenditures (153 billion dollars a year!) go to the top one percent. That’s more than twice as much (8 percent, 72 billion dollars) as goes to the entire bottom 20 percent. All in all, 51 percent of tax expenditures benefit the top 20 percent of individuals (459 billion dollars).
Now, it can be difficult to put all of these billions of dollars in context, but that 459 billion dollars of tax breaks that are given to the top 20 percent is more money than we spend on housing, food, unemployment, heating, and other income security services that make up the social safety net.
Besides the inequality in tax expenditures, there’s a few other things that need to be pointed out. First, this is a reminder that trying to judge the size of government by how much it spends is a terrible way to measure government. Tax expenditures involve just as much government as appropriations spending. In fact, since behavioral economics suggest human beings are (on average) more motivated by fear of losing money than by the offer of additional money, tax expenditures may actually be more manipulative of human behavior. From an accounting standpoint there’s no real difference between offering someone 100 dollars to do something and giving them 100 dollars off of their taxes, but right now we treat those two approaches to governing very differently.
Until we start thinking and talking about tax expenditures as actual spending (remember, if we didn’t have any tax expenditures we could lower tax rates across the board the same way we could if we spent less) they are going to proliferate, as it tends to be easier to adjust the tax code and add a tax break (for either the wealthy or the poor) than to start a new spending program through the traditional budgeting process. This leads to unaccountable governance, because as difficult as the budget can be to understand, the tax code is essentially indecipherable to non-specialists.
Finally, a caveat on the CBO report. The CBO estimate of 900 billion assumes that individuals don’t change spending behavior based on the tax expenditures. In actuality, if we got rid of all tax expenditures we wouldn’t raise a full 900 billion as people (on average) would buy fewer of the goods and services we had been subsidizing. It’s extremely difficult to estimate how much it would raise (which is why the CBO didn’t try), but it’s tough to imagine it being less than 500 billion.
Note: I have also used these two graphs for the NCC Poverty Initiative, but they were created by me. I’ve simply used them in more than one place.