«Mixed Messages on Mixed incoMes Volume 15, Number 2 • 2013 U.S. Department of Housing and Urban Development | Office of Policy Development and ...»
Before the development of the Vacant Property Registration Ordinance Database, limited information on VPROs had been available. Little comprehensive data had been compiled on the coverage of the ordinances, their requirements, and the penalties for noncompliance. The Vacant Property Registration Ordinance Database provides the sort of information necessary to examine how ordinances vary. Ultimately, the database could be used to evaluate the effects of different types of ordinances on local housing market conditions.
Three fundamental types of VPROs exist: the Vacancy and Abandonment Model, the Foreclosure Model, and the Hybrid Model. The key difference among these models is the event that triggers the requirement to register properties and comply with the ordinance’s other requirements. Vacancy and Abandonment-Model ordinances require property owners to register properties after a certain length of vacancy. Foreclosure-Model ordinances are ordinances in which registration is triggered by a formal, state-required notice of default or intent to foreclose that is filed as a part of a judicial proceeding or advertised by the mortgagee or servicer as a part of a nonjudicial foreclosure process. One reason that this model was developed was that localities were finding that some properties where foreclosures had been initiated were being vacated well before the foreclosure sale was complete and the property became owned by the mortgagee or another new owner (Martin, 2010;
Schilling, 2009). Many more recently enacted ordinances share characteristics of the Vacancy and Abandonment Model and the Foreclosure Model, in that they can be triggered either by vacancy or by foreclosure-related actions. We classify such ordinances as following the Hybrid Model.
Within these three ordinance types (the Vacancy and Abandonment, Foreclosure, and Hybrid Models), the specific terms and requirements vary greatly. Coverage and exemptions vary, as do requirements for securing, maintaining, and insuring the property. Enforcement tools, although somewhat uniform in fundamental structure (the use of fines is the primary tool), also vary, with some localities specifying at least some violations as criminal (misdemeanor) offenses and other localities not. Maximum fine amounts also differ significantly. Another feature of some VPROs is the exemption of properties that are registered with industry databases.
260 Data Shop New Data on Local Vacant Property Registration Ordinances Developing the Vacant Property Registration Ordinance Database The initial list of VPROs came from the firm Safeguard Properties, Inc. (Safeguard), which has provided a frequently updated list of ordinances for several years. Safeguard is nationally recognized as a leading provider of asset management services for loan servicers and lenders. Using the Safeguard list, we identified 552 ordinances.1 Each ordinance—or, in a few instances, a summary of the ordinance—was then read and coded into more than 30 variables, described in exhibit 1.
An example of a typical database record is provided in the appendix.
We began with a list of 587 mandatory VPROs published by Safeguard, which we downloaded from http://www.
safeguardproperties.com/Resources/Vacant_Property_Registration.aspx on May 1, 2012. For a few ordinances in the Safeguard list, however, we were unable to find documentation of the ordinance (either a copy of the ordinance itself or, in a few cases, a summary of the ordinance). For 14 of the ordinances, the date of enactment was unclear, so they are not included in the time-on-enactment analysis. A significant undercount of ordinances adopted in the last few months of this period is likely, because we expect some (varying) lag between the date of enactment and the entry of the ordinance in the Safeguard database. The Safeguard list will likely expand somewhat to include ordinances enacted before May 2012 but not included in the list as of May 1, 2012.
Cityscape 261Lee, Terranova, and Immergluck
Hybrid-Model ordinances mushroomed, although major growth was still occurring in Vacancy and Abandonment-Model ordinances. After 2009, the number of new ordinances slowed a bit, but more than 200 ordinances were adopted from January 2010 to April 2012. The number of new Hybrid-Model ordinances slowed somewhat after 2009, the number of new Vacancy and Abandonment-Model ordinances held roughly constant, and the number of new ForeclosureModel ordinances increased.
Exhibit 3 shows the growth of local VPROs for the nine states with the most local VPROs adopted through April 2012. These nine states account for 77 percent of VPROs, led by Florida and California, which each account for 17 percent (94 and 93 ordinances, respectively) of the VPROs.2 Illinois (61 ordinances, or 11 percent), Michigan (54 ordinances, or 9 percent), Ohio (37 ordinances, or 7 percent), Massachusetts (30 ordinances, or 5 percent), Minnesota, Georgia, and Missouri comprise the rest of the list. Many of these states have been among the leaders in foreclosure statistics during the prolonged U.S. housing crisis.
Connecticut passed a statewide vacant property registration statute in 2009, essentially imposing a vacant property registration requirement for properties across the state, although the statute allows for property owners to avoid registration with local governments if they register with a prescribed industry registration system. Connecticut localities are not included in any descriptive statistics here. For more discussion, see Immergluck, Lee, and Terranova (2012).
262 Data Shop New Data on Local Vacant Property Registration Ordinances When the national foreclosure crisis took hold in 2007, California was clearly the early leader in VPRO adoption, with 4 localities enacting ordinances in 2007 and another 44 localities enacting them in 2008. Ohio had seen a steady, if slower, increase in VPROs, with 3 new ordinances in 2007 and 5 in 2008. Other states saw a substantial increase in ordinances enacted in 2008, including Florida, Illinois, Massachusetts, and, to a lesser extent, Michigan and Missouri. Two states— Ohio and Georgia—saw the rate of VPRO adoption pick up markedly in 2011. Ohio localities had already been somewhat active in adopting VPROs, enacting 14 ordinances from 2008 to 2010. In 2011, 10 additional ordinances were enacted in the state. Before 2011, Georgia had seen a slow rate of VPRO adoption, with only 9 local laws enacted up through 2010. In 2011, 10 new ordinances were enacted throughout the state. In response to the surge in such ordinances, however, by the spring of 2012, opponents of local VPROs had gotten a state law passed essentially preempting all but relatively weak ordinances.3 The Foreclosure Crisis and VPRO Growth Examining the relationship between foreclosures and the adoption of VPROs, exhibit 4 plots the number of new VPROs in each state after 2007 against the increase in the quarterly foreclosure start rate at the beginning of the national foreclosure crisis. It shows a general positive association between these two variables, so that states with greater increases in foreclosure starts in 2006 and Exhibit 4 New Local VPROs (January 2008 to April 2012) Versus Increase in Quarterly Foreclosure Start Rate (Fourth Quarter of 2005 to Fourth Quarter of 2007) New VPROs after 2007 State of Georgia. House Bill 110. May 1, 2012. Available at http://www.legis.ga.gov/legislation/en-US/display/20112012/ HB/110 (accessed March 8, 2013).
2007 tended to experience more new, local VPROs after 2007. Exhibit 4 also shows that states with greater increases in foreclosure starts had substantial variation in the adoption of local VPROs, however. Indeed, many other factors are at play. For one thing, the sheer number of localities varies greatly across states. Beyond such very basic differences, another factor in determining local VPRO adoption is the authority that localities within a state possess to enact and implement VPROs. Some states, such as Nevada, are strong Dillon’s rule states, in which the authority to pass laws such as VPROs must be expressly granted by state statute. In other states, laws that limit vacant property registration practices at the local level or that require statewide registration may, in effect, actually discourage or prevent states from enacting their own ordinances. Differences in state property law, housing market and broader vacancy conditions, and local political environments are also likely to come into play in the extent to which local governments are likely to adopt VPROs.
Of particular note are the states in the lower right-hand portion of exhibit 4. These states, including Arizona and Nevada (two perennial leaders in foreclosure statistics during the crisis), saw very few VPROs adopted after 2007. Arizona had only one known VPRO (enacted in 2009), and Nevada had only three (enacted in 2006, 2010, and 2012). Again, state home-rule laws, state political climate, and the number of local governments are likely to be key factors here.
Potential Indicators of Ordinance Strength The complexity of VPROs makes it difficult to develop a simple measure of the strength or rigor of an ordinance. In fact, any concept of strength is likely to be somewhat subjective and to depend on a combination of a variety of characteristics, including coverage (which types of properties are covered or excluded), requirements (including registration fees, maintenance, security, insurance, and rehabilitation plans), and sanctions or penalties (fines, criminal penalties, liens, and so on).
Moreover, tradeoffs may exist between characteristics. As an example, localities may specify higher maximum fines, but this increase may be partly related to their exclusion of more property types.
Although no one variable in the database will provide a comprehensive measure of ordinance strength, one might expect a set of indicators to be closely associated with overall ordinance strength. The database will enable researchers to develop their own measures of ordinance strength or, potentially, to test the effect of a particular ordinance characteristic on housing market outcomes.
Conclusion More than 5 years after the beginning of the foreclosure crisis, localities continue to adopt VPROs at a substantial pace, but the rate of growth has slowed somewhat since the peak of the crisis. The Vacant Property Registration Ordinance Database can be updated to reflect this growth. The database is expected to help researchers and practitioners understand the nature and variation of these ordinances across many characteristics. For a fuller description of the database and a detailed data dictionary, see Immergluck, Lee, and Terranova (2012). To obtain a copy of the database, contact Dan Immergluck at email@example.com.
Acknowledgments The authors thank the Federal Reserve Bank of Atlanta’s Center for Real Estate Analytics for supporting this research.
Authors Yun Sang Lee is a recent Ph.D. graduate of the School of City and Regional Planning at Georgia Institute of Technology.
Patrick Terranova is a master’s degree student in the School of City and Regional Planning at Georgia Institute of Technology.
Dan Immergluck is a professor in the School of City and Regional Planning at Georgia Institute of Technology.
References Immergluck, Dan, Yun Sang Lee, and Patrick Terranova. 2012. “Local Vacant Property Registration Ordinances in the U.S.: An Analysis of Growth, Regional Trends, and Some Key Characteristics.” Available at http://ssrn.com/abstract=2130775 (accessed October 20, 2012).
Martin, Benton. 2010. “Vacant Property Registration Ordinances,” Real Estate Law Journal 39 (1):
Schilling, Joseph. 2009. “Code Enforcement and Community Stabilization: The Forgotten First Responders to Vacant and Foreclosed Homes,” Albany Government Law Review 2 (1): 101–162.
266 Data Shop Graphic Detail Geographic Information Systems (GIS) organize and clarify the patterns of human activities on the Earth’s surface and their interaction with each other. GIS data, in the form of maps, can quickly and powerfully convey relationships to policymakers and the public.
This department of Cityscape includes maps that convey important housing or community development policy issues or solutions. If you have made such a map and are willing to share it in a future issue of Cityscape, please contact firstname.lastname@example.org.
Visualizing Same-Sex Couple Household Data With Linked Micromaps Brent D. Mast U.S. Department of Housing and Urban Development The views expressed in this article are those of the author and do not represent the official positions or policies of the Office of Policy Development and Research or the U.S. Department of Housing and Urban Development.
In this article, I demonstrate how using linked micromaps (Carr and Pickle, 2010) can improve mapping of same-sex couple (SSC) household data. Micromaps display multiple maps on the same exhibit and highlight different geographic units in each map. Linked micromaps display columns of data next to micromaps.
I improve on typical census data mapping (for example, Lofquist, 2011) in several ways, most importantly by providing context for interpretation.1 Typical choropleth maps provide no context to help the reader understand why, for example, Washington, D.C. (hereafter, D.C.), has such a high percentage of same-sex couples compared with the SSC percentage of the 50 states. Linked micromaps allow for state total estimates to be reported along with estimates by metropolitan status. When areas within states with the same metropolitan status are compared, D.C. is no longer such a significant outlier. I also improve on typical census mapping by presenting SSC estimates in descending order, which puts similar states into smaller perceptual subgroups for easier comprehension. I also report confidence limits, which help the reader gauge the relative precision of the mean estimates.
For Census Bureau data on SSC households, see http://www.census.gov/hhes/samesex/.
To compute custom estimates by state and metropolitan status, I used American Community Survey (ACS) Public Use Microdata Sample (PUMS) data. To increase precision, I analyzed 5 years of PUMS data (2006 through 2010), which I downloaded from the University of Minnesota’s IPUMS-USA database (Ruggles et al., 2010).