Over the past three decades, housing vouchers have become one of the primary ways that the federal government subsidizes the rents of low-income households. Providing poor families with government assistance so that they can afford to live where they want will, theoretically, allow them to move to better neighborhoods with more opportunities that help them climb out of poverty.

It doesn’t always work out that way. When voucher amounts are all the same, regardless of the neighborhood’s quality, many voucher recipients opt for areas with worse crime and poverty. After all, the cheaper rents mean that they have a lot more affordable inventory from which to choose.

However, there can also be an unintended impact on the market. Landlords in high-demand areas may just raise rents on voucher households, with the increase largely paid for by the federal government. When this happens, the net effect of the policy is essentially to benefit landlords.

A paper in the May issue of the American Economic Journal: Economic Policy examines how the program’s design influences whether a voucher holder moves to a higher quality neighborhood.

Figure 10 from Collinson et al. (2018)

Researchers Robert Collinson and Peter Ganong looked at what happened when housing authorities in Dallas, Texas, switched from using a single metro-wide limit on the value of the voucher to one that was more generous in pricier ZIP codes (and less generous vouchers in low-rent ZIP codes). They then compared it to voucher holders in neighboring Fort Worth.

The figure above plots their results. Neighborhood quality — which takes into account poverty rate, test scores, unemployment rate, the share of kids with single mothers, and violent crime —  is plotted along the y-axis and the years before and after the policy are on the horizontal axis. The vertical line marks the last year before the policy change went into effect in Dallas.

Voucher holders in Dallas were more likely to move to safer and less impoverished neighborhoods after the policy change. Not only was the policy in Dallas more effective, it came at zero net cost to the government.

Not that all families moved to the nicest parts of Dallas. Far from it. The quality of the neighborhoods was still somewhat low, according to the authors. However, their findings underscore the importance of designing housing assistance programs to benefit the people policymakers intend to help.