Nearly one-third of all federal tax expenditures--$384 billion in 2013 alone-- is aimed at various forms of asset building, such as retirement savings, higher education, and home ownership. Yet, according to research by several of my Tax Policy Center and Urban Institute colleagues, these tax breaks do little to help low- and middle-income households build wealth. Incentives for homeownership represent a bit more than half of these subsidies, special tax treatment for retirement savings nearly 40 percent, and higher education subsidies about 8 percent. Together, they are so big that they’ll inevitably be part of any broad-based tax reform effort. Because they are such a high percentage of tax expenditures, it is effectively impossible to achieve significant tax rate reduction without scaling them back. That’s why they were such an important element in House Ways & Means Committee Chairman Dave Camp’s tax reform plan. In both a long paper and in a newly issued fact sheet, Gene Steuerle, Ben Harris, Signe-Mary McKernan, Caleb Quackenbush, and Caroline Ratcliffe conclude that these asset building subsidies are ripe for reform, whether through a broad tax code rewrite or on their own. Because high income households receive the lion’s share of the benefits of these subsidies, they raise important distributional issues. But, more importantly, they fail to achieve a key goal: To help households boost their assets. They simply don’t work very well. Take subsidies for retirement savings. While they were worth nearly $200 billion in 2013, there is little evidence that they increase savings by much at all. Only about half of workers participate in 401(k)-type plans and those that do tend to contribute far less than they could. When Raj Chetty and colleagues tried to measure the “bang for the buck” from these tax breaks, they figured that net new savings was equal to only about 1 percent of the value of the subsidies. The distribution of these tax breaks helps explain why they add so little to savings. The top 20 percent of taxpayers get about two-thirds of the benefits. The middle 20 percent get only about 10 percent and the bottom 20 percent—who need to save the most--get almost nothing at all. As a result, retirement savings incentives drain the public fisc while often subsidizing those who shift money from taxable accounts to tax-subsidized portfolios. Similarly, existing subsidies for homeownership encourage high-income households to buy bigger houses or take out more tax-deductible mortgage debt. But they do little to help many middle-income families buy a home. Higher education tax breaks do appear to be somewhat better targeted. Nearly 90 percent of the Lifetime Learning Credit, for instance, goes to middle-income taxpayers. But it is not clear that these subsidies encourage people who otherwise would not enroll to go to college (and, more importantly, to graduate). There is no shortage of ideas to improve these preferences. The mortgage deduction could be turned into a tax credit, which would better target the subsidy to middle-income households. For example, a few years ago Phil Swagel, Bob Carroll, and John O’Hare suggested replacing all federal housing subsidies with a single $3,700 credit. Similarly, tax-advantaged retirement accounts could be aimed at those who are most likely to increase savings. For instance, contributions could be capped at, say, 20 percent of income or $20,000 (an idea backed by the Bowles-Simpson deficit commission). However, Gene, Ben, Signe-Mary and their colleagues would go further and provide new incentives to employers that enroll more low- and moderate-income workers. Sadly, nearly all pols (Camp is the notable exception) still say they’d finance rate cutting tax reform by closing “loopholes.” They can’t. It requires big cuts in popular preferences such as those for housing and retirement savings. The good news is that we do such a poor job subsidizing asset building that there is room to both do better and perhaps save some money at the same time.