The notice announcing the FY12 Proposed Fair Market Rents (FMRs) appeared in the Federal Register on August 19. Comments are due by September 19. FMRs are primarily used by agencies to determine payment standards in the Housing Choice Voucher (HCV) program and rents levels in some project-based programs.
If the proposed FMRs are implemented, many FMRs will be lower next year. Fully 3,318 counties would register declines and less than half that number (1421) would see increases. Twenty-six counties would not see a change. Furthermore, while the average decline would be $47, the average increase is just $21. The population-weighted national FMR, which accounts for the number of people affected by these increases and decreases, would decline to $916 from $932 in FY11.
While it is difficult to summarize the causes of these changes, both economic conditions and changes in how FMRs are calculated are implicated. It is perhaps most important to recognize that average rent levels were falling in many areas of the country in the latter part of the 2005 to 2009 period covered by the American Community Survey (ACS) data that HUD uses to set FMR levels. Furthermore, the Consumer Price Index (CPI), which HUD uses to project rents forward to the end of 2010, shows that rents continued to slide through December 2010 in many areas. HUD’s online Individual Area Proposed FY12 FMR Documentation system illustrates these factors in affected areas.
One change in how FMRs are calculated that affects all areas is related to HUD’s ongoing transition to the ACS. For most of the past decade, HUD has used a file from the “long form” survey that accompanied the 2000 census as its “base rent” file. It updated this file from year to year using the CPI and increasingly the ACS. The ACS is a relatively new survey; it was first fully implemented in 2005, and HUD has increasingly relied on it as the available data increased. This year HUD will rely entirely on the ACS and no longer use Census 2000 data as its starting point for calculating FMRs.
The transition to the ACS is inevitable since the 2010 Census did not collect rent data. However, while the ACS data are high quality and only lagged by a year, they also present some significant challenges in calculating FMRs not present with the 2000 data. The biggest difference between the ACS and the long form is that the ACS surveys fewer households in a single year, which limits the reliability of its estimates for smaller areas. To overcome this, the Census Bureau pools three or five years of data to generate ACS estimates for smaller places. While the long form reached approximately 1-in-6 addresses in 2000, over five years the ACS reaches about 1-in-8 addresses.
Data pooling, however, creates problems in calculating FMRs. This is because in calculating FMRs HUD only considers the rents paid by recent movers, defined as those who signed a lease within the past two years. These renters have completed a recent housing search and have negotiated a new contract and therefore are more likely to be paying “market rate” for their home. It makes little sense to use “recent mover” data from all five years of the pooled data, which in some instances may represent rent contracts signed over seven years before. But using just the most recent year of ACS data dramatically reduces the number of recent mover observations and in many areas, even many metropolitan areas, there are just too few observations to calculate reliable FMRs.
HUD proposes to overcome this problem by using rents for any standard quality rental home in all five years (adjusted for inflation) as the base rent and applying a “recent mover bonus.” In non-metropolitan counties, which have their own FMRs, the bonus is calculated based on recent movers in the all non-metropolitan areas of the state. In metropolitan FMR areas the recent mover bonus can be based on data from the metropolitan area itself or a larger area such as the combined metropolitan areas in the state or the entire state.
The other significant change this year relates to changes to specific areas that receive a boost in their FMRs to assist in meeting HUD’s goal of dispersing voucher holders to live outside poorer communities. Normally, FMRs are set where 40% of units rent for less than the FMR. In areas where voucher holders are concentrated in a few poor neighborhoods, an alternative policy, adopted in 2001, boosts the FMR to a level where 50% of the units fall below the FMR.
The higher rent level should assist voucher holders in renting units outside of these neighborhoods. As part of this “50th percentile FMR” policy, HUD must re-evaluate this boost very three years. Once areas meet the dispersal criteria, FMRs return to the 40th percentile.
Somewhat contradictorily, however, areas can also lose the boost if they show little progress or no progress at all. This year seven areas lost their 50th percentile FMR. Two of these areas, Richmond, VA and Milwaukee, were deemed to have exceeded the criteria. The other five areas (Albuquerque, Denver, Houston, Chicago, Kansas City, and Tacoma) all failed to show sufficient progress. In addition, 10 communities were reinstated as 50th percentile areas, after having lost that status for failure to meet the criteria by 2008. These areas are Austin; Honolulu; Orange County, CA; Riverside, CA; Virginia Beach; Forth Worth; Las Vegas; Phoenix; and Tuscon.
As with the rental market indicators, a look at HUD’s Individual Area Proposed FY12 FMR Documentation system can quickly reveal the influence that a change in an area’s 50th percentile status has on the FMR level. The influence that the transition to the ACS and the use of the recent mover boost has is more difficult to isolate.
The notice also discusses a change that is not being proposed this year. In a federal notice published on March 9, HUD sought information on changing how it projects rents forward from the most recent data to the middle of the next fiscal year in which the FMRs will be used. In this note, HUD provides responses to the comments it received but finds no consensus and postpones its decision until 2013.
In response to a number of unsolicited comments from NLIHC and others in their comments on trending, HUD does propose setting a date certain for the release of income limits. While FMRs have a statutory October 1 deadline, income limits have no such requirement and the date of their release can vary significantly from year to year. HUD suggests a date of either October 1 or December 1. Comments on this proposal are also due on September 19.
Finally, the notice reveals that HUD received just four applications to pilot the Small Area FMRs, and that information on that program will be released at a date later this year.
The Federal register notice, the Individual Area Proposed FY12 FMR Documentation system, and other related resources are available at: http://www.huduser.org/portal/datasets/fmr.html