Miami-Dade Tax Credit Applications

Posted on

In December of 2017, the Florida Housing Finance Corporation received 27 applications for competitive 9% tax credits in Miami-Dade County. The locations of the associated development sites are mapped below.

Comparing the Sites to the Average Miami-Dade Neighborhood

Let's take a closer look at the communities where theses sites are located.

In the census tracts containing sites for which developers submitted applications for competetive tax-credit funding:

  • The average poverty rate is 31.3% (with a range of 11.5% to 54.0%);
  • The average rate of households receiving food assistance benefits is 43.2% (with a range of 14.3% to 55.7%);
  • The average adult labor force participation rate is 58.9% (with a range of 42.8% to 74.1%); and
  • The average unemployment rate (among individuals over 16 years of age) is 13.6%.

Let’s put that all in context. For all populated census tracts in Miami-Dade County:

  • The average poverty rate is 20.0%;
  • The average rate of households receiving food assistance benefits is 25.0%;
  • The average adult labor force particiaption rate is 62.2%; and
  • The average unemployment rate (among individuals over 16 years of age) is 8.6%.

That means that on average, applications for tax credits are originating from census tracts where the incidence of poverty is 56.5% greater than the typical Miami-Dade County census tract; dependence on food assistance is 72.8% greater. Applications are originating from tracts where labor force participation is meaningfully less; and among those in the work force, unemployment is 58% greater than in the typical Miami-Dade County census tract.

Comparing Sites to Each Other

Among the 17 applications indicating that the development will serve the family demographic (i.e occupancy will not be limited to older persons), 3 are in census tracts where the rate of poverty is less than the average Miami-Dade County census tract. Another 8 are located in census tracts with rates of poverty less than a full standard deviation above the mean.

There are 20 applications which supposedly include participation by not-for-profit organizations (including one in which the true principal is a for-profit developer subject to a deferred prosecution agreement for lying to the Corporation about construction costs in order to enrich himself). Of these, 11 indicate that the property will serve the family demographic. In all likelihood, the funded deals will be among these applications.

All of these applications come from competent developers who can make it through underwriting, complete construction, and place the buildings in service on time. The buildings will all be physically indistinguishable from market rate housing.  All of the proposals are for concrete new construction. The most significant difference between these sites is social geography. Simply put, most are located in neighborhoods where poverty and dependence on public assistance are normative. A few are located in neighborhoods where poverty is not normative and upward mobility is more likely. To put this all in perspective, the peak unemployment rate in Florida during the Great Recession was 11.2%. That means that employment prospects in the neighborhoods producing tax credit applications during normal – even good times – are worse than most people experienced during the bleakest economic period in living memory.

Looking Ahead - Fair Housing Implications

The Corporation is in the process of updating the rules which guide the allocation of competitive resources, as well as the Qualified Allocation Plan. These guidelines will be operationalized in future RFAs. If Florida is committed to fair housing, it must recognize that where a person lives, especially during childhood, has a a significant impact on her opportunities for upward mobility. Our housing production programs should not be focused on warehousing the poor and keeping them in a position of dependence. Instead, we should expand the options available to low-income households and use the tax credit program to create housing opportunities in otherwise inaccessible neighborhoods.

 

Sources:

Florida Housing Finance Corporation RFA 2017-112; RFA 2017-112 Applications Submitted Report

American Community Survey 2016 5-yr Average tables: S2301; S2201; S1501; S1701

 

 

What does ELI mean, and why is it different in Florida?

Posted on

What does ELI mean, and why is it different in Florida?

It means Extremely Low Income (ELI). Many housing programs utilize the income category. The U.S. Department of Housing and Urban Development (HUD) Office of Community Planning and Development (CPD), and Office of Public and Indian Housing (PIH), make frequent use of the income category in guidance issued to state and local government grantees, as well as Public Housing Authorities (PHAs).

For example, 100% of National Housing Trust Fund (NHTF) monies administered by the states must be used to serve ELI households in years when the total fund is less than $1 Billion. In public housing, no less than 40% of units which become available during the year must be made available to ELI households. When PHAs admit new families to the Housing Choice Voucher (HCV) program, at least 75% of the families admitted to the program during the course of a year must be ELI. As units become available in multifamily buildings assisted through project-based Section 8 programs (excluding PBVs), 40% must be made available to ELI families.

From 1998 through 2014, HUD defined ELI simply: families with incomes at or below 30% of Area Median Income (AMI)[1]. However, as part of its annual appropriations act of 2014, Congress directed HUD to update the definition and modify its ELI targeting in some programs. The change, describe in the Federal Register on June 25, 2014, defines an ELI family as one whose income does not exceed:

  • the Federal poverty level; or
  • 30% of AMI.[2]

The new definition helps families who live in places where the AMI itself is extremely low. As a relative measure, the previous definition punished households living in impoverished areas. The poverty level is a measure of absolute, rather than relative, need. Conceptually, a family living below the poverty level does not have sufficient income to sustain itself. The fact that the median household in the same geographic area is also poor should not exclude such families from housing assistance programs.

Why is it different in Florida?

Because, everything is different in Florida! The serious answer is that the Florida Housing Finance Corporation uses an adjusted ELI level to establish set-aside requirements. When the Corporation provides capital subsidies to developers, it often requires that a portion of the units be “deeply targeted” to ELI households. The rationale for this is straightforward. Rent levels at properties financed through the Low Income Housing Tax Credit (LIHTC) program are generally set at 30% of the household-size-adjusted maximum income allowed under the program.[3] By definition, an ELI household has an income half that of a household at the maximum of LIHTC eligibility. An ELI family faces a potential rent burden equal to 60% of its total income in a tax credit unit. Therefore, in order to get the credit, or additional subsidies through the State Apartment Incentive Loan (SAIL) program, the Corporation requires developers to set-aside some of its units for ELI households at more affordable rents.

But why is the ELI level different?  

The Corporation requires developers to set-aside units for households at an ELI level which varies by county. The level can be greater than or less than 30% of AMI. Florida Statute 420.0004(9) adopts HUD’s ELI definition, but empowers the Corporation to adjust the threshold by rule. In 67-48.002(39) of the Florida Administrative Code, the Corporation repeats the definition. In actual practice, the Corporation determines the full-time income of a person earning the current minimum wage as an approximate proportion of the median income for each county. The results vary widely and reflect the economic geography of the state. For example, the ELI level in Broward County is 28% of AMI; in rural North Florida (e.g. Union County, Levy County, and Jackson County) it is 45% of AMI.

This is an area in which Florida has been a leader in housing policy. The Corporation’s approach to allocating competitive resources has long recognized economic variation across the state. The county-adjusted ELI level is perhaps the most elegant illustration of this forward thinking. At the federal level, HUD has only recently begun to utilize more fine-grained geographies through Small Area Fair Market Rents (SAFMRs) and Small Area Difficult Development Areas (SDDAs).  The county is a very coarse geography for this purpose. Hopefully the Corporation will update its approach, now that it is using more advanced GIS tools.

Thank you for reading. Please feel free to leave a comment below.

 

 

 

 

 

[1] Already in use in academic literature, Congress established the new income standard as part of the Quality Housing and Work Responsibility Act (HWRA) of 1998. The new standard more narrowly focused housing assistance within the Very Low Income (VLI) category introduced as part of Housing and Community Development Act of 1974.

[2] For the curious:  Area Median Income, as we are using the phrase here, is determined by HUD’s Office of Policy Development and Research (PD&R)  using the geographic area definitions developed by the White House Office of Management and Budget (OMB) and data from the American Community Survey (ACS). The poverty guidelines refered to here are developed by the U.S. Department of Health and Human Services (HHS).

[3] Under Section 42 of the Internal Revenue Code, LIHTC properties must serve households with incomes at or below 60% of AMI.

Who wins the most competitive tax credit awards in Florida?

Posted on

Who wins the most 9% tax credit awards in Florida? The question proved difficult to answer and some major caveats are in order. During the three-year period of 2014, 2015, and 2016, the Florida Housing Finance Corporation received 702 applications for 9% tax credits. The Florida Housing Board of Directors approved preliminary awards for 85 of these. Subsequent protests led to final orders and settlements which resulted in a few additional awards, as well as some of the 85 awarded deals being deemed ineligible. I used applications and preliminary awards for the information below, I’ll correct for additional awards and ineligible applications another time. The award amount comes from preliminary awards (based on figures provided by applicants), not from final credit underwriting reports.

During the period evaluated (2014, 2015, and 2016) it was often difficult to tell exactly who was applying for tax credits. Actual applicant entities are created for a specific property; similarly, the names of developer entities listed in applications often don’t reflect the parent organization. For this reason, I used the contact person listed in the application. One person who is especially active (and successful) changed organizations during the time-period evaluated. I decided to combine his applications from both employers, which include the same PHA co-developer. His former employer submitted additional applications, but did not win any preliminary awards. Obviously, in this case, the person and not the organization is critical.

Another name appeared as the contact for several applications, some of which were successful. The name is listed on CAHP’s website as a member associated with an organization with which I am unfamiliar. Searching for the applicant and developer names on Sunbiz, I found that McCormack Baron Salazar is part of the ownership structure for each of them. Accordingly, I added that organization to the developer column for those applications.

A few organizations submit applications under different contact names; I consolidated these into a single record for the rankings below. There are some practical difficulties gleaning this information from Florida Housing’s website. Several applicants write their names differently on different applications. For example, “Matthew” is sometimes “Matt”; “Elizabeth” is sometime “Liz”; and “Kimberly” is sometimes “Kim”. Similarly, “Jr.” and “Sr.” appear inconsistently and sometimes wihtout the period. As I imported these names into Excel, I made the names consistent, removed all periods, and trimmed excessive spaces (e.g. “John Q.   Doe” became “John Q Doe”). To keep things simple, I limited the evaluation to the Geographic RFAs, Preservation RFAs, and Revitalization RFAs.

The following RFAs were evaluated:

RFA 2014-104, RFA 2014-106, RFA 2014-114, RFA 2014-115, RFA 2014-116

RFA 2015-104, RFA 2015-106, RFA 2015-107, RFA 2015-108, RFA 2015-111, RFA 2015-113

RFA 2016-110, RFA 2016-113, RFA 2016-114, RFA 2016-116

I did not include solicitations which involved non-competitive funding, credits paired with SAIL, or other more narrowly focused RFAs (e.g. housing for persons with a disabling condition). In order to focus in on regular competitors, I filtered out all applicants who had submitted fewer than 4 applications over the three-year period.

The results are presented below in three ways: 1) highest percentage of wins (preliminary awards as a percentage of applications submitted); 2) total number of preliminary awards; and 3) total amount of the preliminary awards (annual tax credit requested in application).

Attention: The internal data of table “4” is corrupted!
Attention: The internal data of table “3” is corrupted!
Attention: The internal data of table “2” is corrupted!

A quick summary of applications reveals a few important pieces of information.

First, there are very few active organizations consistently pursuing tax-credit financing in Florida. Only 37 organizations submitted 4 or more tax credit applications during 2014, 2015, and 2016. Of those, only 28 won a single preliminary award. Among those 28 organizations, the median developer submitted 13 applications over the three-year period. Those applications yielded the median developer 2 preliminary awards; a success rate of 15.4%.

Second, those few active professional developers utilize very different strategies. HTG submitted 26 applications (55%) more than Southport, but ended up with the same number of preliminary awards. However, HTG’s deals are much larger, and the total amount of its preliminary awards is 63.7% larger than that of Southport. (I’ll evaluate the geographic dispersion of applications and wins another time. Presumably there is a relationship between the firms’ performance and the areas where they operate). HTG stands apart in its successful use of what I’ll call a “brute force” strategy. Although all of the top ten developers – by any of the measures above – submit many more applications than their competitors, HTG submitted 5x the median number of applications. Although the company’s success rate is well below median, the sheer volume of applications yielded a full pipeline of deals (without even counting any 4%/SAIL awards). Conversely, some organizations, especially those which frequently partner with PHAs, have success rates 2-3x that of the median competitor.  Norstar won 6 preliminary awards from just 20 applications. Roughly half of the regular competitors submitted around 10 applications during the three-year period, winning 1 or 2 preliminary awards. Again, an important caveat is that the protest process can lead to additional funded deals. Therefore, this quick evaluation does not reflect each and every 9% tax credit award in Florida.

Third, some developers are more successful than others. Given the role of the “lottery” as a tie-breaker, as well as the impact of the county award tally, it may be more accurate to say that some developers are luckier than others. Either way, some developers persevere in the face of loss over and over. One developer submitted 30 applications during the three-year period without winning a single preliminary award. That means someone spent $90,000 in application fees, did not win, and continues to compete for tax credit funding. Conversely, Atlantic Pacific submitted just 23 applications, which yielded more than $100,000,000 in tax credit equity.

Thank you for reading. Feel free to check out the data I used. If you see any errors, please leave a comment below.