Signaling in Policy Documents: The Case of Georgia QAP Incentives for Mixed-Income Projects

by Chris Thayer

Affordable housing is well-known as a powerful factor in economic development within cities, both directly and through enticing corporations’ headquarters to take advantage of the concentrated pool of employee talent, undisbursed to the less-expensive suburbs. One of the major tools for the provision of this affordable and workforce housing is the Low-Income Housing Tax Credit, as governed by each state’s Qualified Allocation Plan.

The Qualified Allocation Plan (QAP) is, necessarily, a policy document. Each state has its own version, in theory updated yearly, which covers the rules for the competitive allocation process for that state’s allowance of the federal Low-Income Housing Tax Credit, currently the largest supply-side affordable housing subsidy in the nation at approximately $6 billion per year in foregone tax revenue. The Credit is allocated to states on a population-based formula, whereupon each state organizes its own competitive application process, granting the Credits largely in accordance with its own preferences, rather than any federal mandate. Those preferences are laid out in the QAP, and typically change at least slightly year-to-year. Georgia is noted for having one of the lengthier and more detailed QAPs nationwide and a history of vigorous competition for the subsidy, as about half of projects that apply will be funded – granted the right to claim a specific number of Tax Credits. These Credits are sold to investors (called syndicators) in exchange for part of the funds needed to construct the proposed housing development, which will contain an agreed-upon number of long-term affordable-rent units. As the QAP’s contents are largely directed by the states, this allows for illuminating differences in policy approaches and results.

The Mixed-Income Project QAP Incentive
29 augAs part of a larger process of quantitative QAP change analysis for Georgia’s QAP between 2000 and 2016, it was observed that the provisions for incentivizing Mixed-Income Projects fluctuated substantially over time. The incentive appears in the first QAP on record, for 2000, as a sub-category of a larger scoring section, Tenancy Characteristics. It remained there through 2005, until in 2006 it became its own independent scoring category, perhaps reflecting the increased attention it merited in the policy development team’s priorities. It remained as such until 2009, when it dropped out of the QAP entirely and did not reappear until this year, 2016, as a sub-category of Stable Communities.

Recession & Reconsideration
This abrupt change in direction was in large part intentional. In a recent interview, Director Laurel Hart of the Housing Finance & Development Division at DCA (the Georgia Department of Community Affairs, Georgia’s allocating agency for Low-Income Housing Tax Credits), described the agency’s response to the 2008 financial crisis.

“We really changed our entire model after the crash… It really was a whole change in philosophy. We went towards less lending, we used options to increase cash flow, we changed our HOME requirements… We became much more risk-averse around that time. [The driving question became], ‘how do we make these properties sustainable?’”

This change in direction has persisted to this day thanks to the insights granted by existing properties’ experiences during the crash, which included elevated vacancy rates (due to renters’ inability to pay), unpaid water bills, and sometimes even foreclosure. DCA found that in many cases, nominally mixed-income properties featured market-rate units serving the same population as the 60% AMI (Area Median Income) affordable units, but without any form of subsidy to ease those renters’ burden–not mixing income groups at all. Director Hart emphasized the importance of integrating “what’s standard and what’s common in the industry” with an intimate understanding of Georgia economics in order to craft incentives that would let truly beneficial project rise, and prevent inefficiencies such as those seen with these earlier versions of the income-diverse properties. She also discussed the rebirth of the incentive in the 2016 QAP and the importance of its place within Stable Communities-winning (generally wealthier and more racially diverse) areas only, preventing the issue of too-low market rate rents from making the Mixed-Income Project points be essentially in vain.

Effects of QAP Incentive on Unit Mix
The appearance and disappearance of this incentive offers a unique opportunity to examine the relative effects of point incentives for Mixed-Income Projects on actual observed unit mix, both for applied and awarded projects, and to draw some potential conclusions for QAP development in the future. For the 2000-2016 period covered by available QAPs, full data was available for 2003-2016, and partial data for 2002, though there was nothing for the years prior. This data range, while less than ideal, is still robust enough to allow for correlational observations to be made about the units and incentives present. Upon analysis, this data pool reveals that the incentive has a very weak, possibly non-existent, effect on the competitive process itself.


The above figures demonstrate that there is not a meaningful difference in percent of market rate units, number of market rate units where present, or total units between the pool of all applicants and the applicants selected for an award. This means that any effect the Mixed-Income point incentive could be having rests not in the competitive process – one project edging out another for a part of the limited funding – but rather in the signal it might send to developers. But does it send such a signal?

Signaling Developers
29 aug 4To answer that question, the research took a slightly broader view, looking at each project’s status as either Mixed- or Single-Income – essentially, a Yes/No question of the project’s structure. Comparing the projects’ status with the incentive over time, it becomes clear that there is a strong signaling effect in the presence, or absence, of some mixed-income provision, evidently independent of the actual point value of its percentage of all possible points. This relationship is shown in the chart, and further analysis reveals that during incentivized periods, the average percentage of Mixed-Income projects applying for funds was 62, while in non-incentivized periods it was only 35, with a similar gap for awarded projects, clearly demonstrating the strong signaling effect of the QAP point incentive.

This research does bear several caveats. There were no 2000 or 2001 datasets available, and 2002 contained only a yes/no indicator for Mixed Income. 2007’s data was heavily cleaned due to a reliability issue, and is still likely to be somewhat skewed. Finally, this only reflects the experience of one state. It’s possible that the trends observed were caused by outside forces, such as the 2008 Recession, and more research is needed to support, or disprove, these conclusions.

However, the correlation between trends observed appears to be strong enough to merit some initial recommendations for QAP development. First, given that the Mixed-Income points themselves appear to serve as a signal, rather than a competitive factor, it could be useful to reduce the point amount given to only one, rather than the two it currently features. The point is symbolic, and a reduction here would make other categories – ones in which points awarded drive the competitive decision – have more impact by virtue of a smaller total number of points available. On a related note, the second recommendation this research suggests is a closer look at the total available points. This total has shifted wildly over time with no immediately apparent effort at consistency, making the relative value of a given section’s points vary in difficult-to-detect ways even if the section hasn’t been intentionally changed. Careful attention to this total and strategic point allocations thereof can somewhat regulate the swift-changing nature of the QAP, simplifying future fine-tuning efforts by having a relatively consistent base year-over-year.

Looking Forward
Reflecting on the future of mixed-income housing as a priority for DCA, Director Hart observed that its greatest potential successes are in cities, especially Atlanta, where the can be a meaningful differential between assisted and market rents. She also indicated that the recent resurgence is in part due to industry changes from the recent Supreme Court decision on Fair Housing, and said “Our properties shouldn’t just be warehousing people of a certain income, a certain color, into a certain area… I think that the Mixed Income [section] is coming back in part with that idea of ‘how do we make sure we’re not segregating the poor, segregating the people by color, into certain areas.’” This insight points to the power of mixed-income properties to encourage positive social developments. Indeed, the introduction of Mixed-Income Project incentives to competitive application processes could form a significant portion of initial Assessments of Fair Housing (AFH) that participating jurisdictions are now required to submit to HUD. As a feature of the new HUD Affirmatively Furthering Fair Housing rule, jurisdictions must submit their AFH plans to HUD for review, but HUD has notoriously scanty requirements or even suggestions for such plans. Developing and implementing a Mixed-Income incentive effort could prove a fruitful strategy for otherwise unsure jurisdictions, not only easing their regulatory burden but also increasing the presence of Mixed-Income communities, strengthening the affordable housing environment and promoting greater inclusivity nationwide.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s