Friday, December 09, 2011

The Bed-Stuy Boomerang: how state officials gerrymandered a map to help Forest City Ratner recruit immigrant investors and save big (and how the EB-5 program is riddled with such practices)

Update 12/19/11: Note I've been covering the EB-5 issue at length for more than a year, including a 16-part series last year and coverage in the Huffington Post. Here's my critique of today's New York Times article on EB-5.

Public officials have done much to help developer Forest City Ratner (FCR) recruit Chinese investors to provide a $249 million low-interest loan in exchange for green cards--and now there's new evidence.

We knew that officials from New York City, New York State, and Brooklyn wrote letters to the United States Citizenship and Immigration Services (USCIS), the federal agency overseeing the EB-5 Immigrant Investor program, to get Atlantic Yards approved as an investment vehicle.

And we knew that Empire State Development Corporation official Peter Davidson joined a road show in China to hype the project before potential investors, misleadingly claiming that Atlantic Yards "will be the largest job-creating project in New York City in the last 20 years."

Now, evidence suggests that two New York State agencies helped gerrymander a map of Brooklyn unemployment--beginning at the Atlantic Yards site (in blue) in Prospect Heights, omitting more affluent census tracts nearby, and extending east to encompass poorer tracts in Bedford-Stuyvesant. (I'm dubbing the map "The Bed-Stuy Boomerang.")
The Bed-Stuy Boomerang/Graphic by Abby Weissman

The map ensured that the promoters of the EB-5 project could tell the needed 498 immigrant investors that the project was located in a Targeted Employment Area, featuring high unemployment. That meant investors had to put up only $500,000, rather than $1 million.

By getting this EB-5 project off the ground, the state helped FCR save more than $140 million, by my estimate, on a $249 million loan. (Neil deMause thinks it's under $100 million.)

And it could help the New York City Regional Center (NYCRC), a private investment pool federally authorized to attract purportedly job-creating investments, reap some $50 million. Of some 200 regional centers in operation, the NYCRC was the busiest this year in China, the biggest market internationally. (Graphic below from NYCRC web site.)

Can they get away with it?

Is this kosher? Well, it seems to violate the spirit of federal immigration law--especially since the targeted census tracts are not being helped much by the project--but likely not the letter.

And, as New York City promotes the EB-5 program via the NYCRC as a source of low-cost financing for other development projects, the state has apparently drawn other helpful maps.

For example, it attached the Brooklyn Navy Yard to a single adjacent census tract containing public housing, thus ensuring a zone of high unemployment. Also, evidence suggests that it has tethered the Battery Maritime Building in prosperous Lower Manhattan to another housing project. (I call these "The Navy Yard Appendage" and "The East River Stretch." Scroll down for more.)

To aid EB-5 projects nationally, it's common to gerrymander such maps, which reflect no legitimate geographical or political boundary. Some in the EB-5 field suggest that project sponsors get away with "statistical gymnastics" or make "egregious attempts... to cheat Congressional intent."

However, the issue seems below the radar: at a Senate hearing this week on permanently authorizing regional centers, there was no mention of TEAs, though there was discussion of other controversial issues.

The impact of federal deference

The federal United States Citizenship and Immigration Services (USCIS) defers to states in evaluating such maps, unwisely choosing to focus solely on whether the threshold of high unemployment is met.

Thus the federal agency ignores the incentive for states to draw irregular maps, thus helping favored projects get cheap financing, while enriching the middlemen.

And that detracts from the public interest behind the EB-5 program, in which each investor is supposed to create ten jobs, focusing on an area of high unemployment. And it disadvantages rival projects that play by the rules, delivering more of a public benefit.

Of course, the rationale behind EB-5 has already been diminished by federal policies that allow projects promoted by regional centers to calculate job creation via an economist's report rather than an actual head count.

The feds also allow immigrant investors to be credited with jobs created not merely by their investments but by the money they've invested combined with existing investments. That may make sense if their investments truly serve as seed money, but with Atlantic Yards, Forest City Ratner simply replaced a high-cost land loan with lower-cost financing, reaping the profits.

Why a TEA matters

New York State's questionable action concerns a threshold issue crucial to recruiting investors via regional centers: whether Atlantic Yards is located in a Targeted Employment Area (TEA), where the unemployment rate is at least 150% of the national unemployment rate.

For investments located in a TEA, potential immigrants using the EB-5 program to get green cards for themselves and their families must only put up $500,000 and park it for five years.

Otherwise, they'd have to put up $1 million. Not only is that harder to raise, it represents a larger loss of potential investment income, since EB-5 investments generally pay little or no return. (The payoff is the green card.)

Thus, as indicated in the screenshot above from the NYCRC web site and the screenshot below from an NYCRC video, a TEA serves as a fundamental selling point. Nearly every regional center markets only TEA projects, which must be either areas of high unemployment or rural areas.

State holds the cards re TEA

The federal government gives much leeway to states to define TEAs. When asked at 6/30/11 EB-5 stakeholders meeting if they'd ever overturned a TEA designation, USCIS officials hedged, though one EB-5 practitioner told me the lore is that state-designated TEAs are sometimes not approved by USCIS.

Many states, that EB-5 practitioner told me, are like New York, allowing any contiguous combination of census tracts. Others may require the TEA to represent a "meaningful area," which is already defined or may have specific geographic limitations.

In the case of Atlantic Yards, New York State not only drew an irregular map, but seems to have offered special assistance.

As of August 2010, the time when the NYCRC sought TEA designation for Atlantic Yards, the state Department of Labor (DOL) seemed to be discouraging EB-5 project promoters.

It publicly stated (via the Internet Archive) that "no published areas" qualify as a TEA.

What's a "published area"? I queried the state but got no response. However, it likely represents a political jurisdiction; in February 2010 the DOL said that the city of Niagara Falls did qualify as a TEA. Today, the DOL says "no published areas" qualify.

What's the Atlantic Yards TEA?

Though there were "no published areas" at the time, the DOL in August 2010 quietly concocted an oddly shaped map that ensured that the Atlantic Yards site would qualify as part of a TEA.

The DOL does not publish unemployment rates for census tracts. But it apparently does offer assistance if the right people ask.

The Bed-Stuy Boomerang/Graphic by Abby Weissman
On 8/4/10, in a response to a request from Robert Scardamalia of the state Department of Economic Development (part of Empire State Development), DOL official Joseph Nardone wrote:
I identified an area of high unemployment with an annual average unemployment rate of 14.4% for calendar year 2009 that is composed of the adjoining census tracts 129.01, 129.02, 161, 163, 203, 223, 225, 227, 229, 247, 267, 269, 277, 279, 281, and 283 in Kings County. The unemployment rate for the United States for calendar year 2009 was 9.3% and thus the minimum 150% threshold level to qualify as a Targeted Employment Areaa is 14%.
Scardamalia, in turn, certified to the NYCRC principal George Olsen--who likely had earlier sent a query--that "the area described in the attached letter qualifies as a Targeted Employment Area." (The letters, at bottom, were discovered via a Freedom of Information Act request.)

What area? Those census tracts may be adjoining, but they represent no rational approach to the site, located in Prospect Heights.

One approach might be to expand in a concentric circle, encompassing adjacent neighborhood census tracts in Park Slope, Fort Greene, Prospect Heights, and Boerum Hill. But that would not be a zone of high unemployment.

Another might be to expand into Community Boards 2, 6, and 8, each of which include pieces of the project site.

Weighting the map to Bed-Stuy

Instead, of the 16 census tracts listed, nine are located within Community Board 3, in high-unemployment Bedford-Stuyvesant and not, as Neil deMause observes, in the gentrifying western segment. The map bears no relationship to any "geographic or political subdivision," which federal law, at least at in spirit, seems to require.

(Note that the western segment of census tract 227 should be filled in, as well.)

More poignantly, while presumably the requirement of a TEA is somewhat related to assisting the area defined as such, there's little evidence that Atlantic Yards has made a dent in the TEA outlined on the map.

I showed the map to the EB-5 practitioner, who observed that it was likely that the unemployment rates in the census tracts containing the project site were low, while they were high in most of the other tracts. (He was right, as described below.)

Source: the state's willing to help

New York state officials seem very accommodating regarding TEAs, according to another source, an attorney doing due diligence on EB-5 investments.

The attorney last year contacted the New York State Department of Economic Development regarding whether certain projects were in a TEA.

The attorney said he was told they didn't provide census tract level unemployment data but could get it for him for a fee. He also was directed to the Department of Labor web site, which indicated, as it does today (right), “For the rest of 2010 and until further notice, no published areas have an annual average rate that is high enough to qualify.”

The attorney then contacted Kunimasa Akasaka, a Department of Economic Development staffer listed by the DOL as the contact person for TEA information. 

Akasaka, the attorney recalled, was accommodating: "He basically said to me to send him the census tract numbers and, in an encouraging manner, 'I'm sure they will qualify' or something like that." 

Any oversight?

Given that adjudication processes currently vary, some USCIS adjudicators might declare the "Bed-Stuy boomerang" gerrymandering, while others may not, the first EB-5 practitioner told me.

A preferable map from USCIS perspective, the practitioner said, would include the census tracts immediately bordering the project site, but the map as drawn does represent a legitimate interpretation of current roles.

"If that was the only way we could do it, I'm not going to tell somebody that the TEA can't be done," the practitioner said, "because by using a standard Bureau of Labor Statistics (BLS) estimating methodology, this contiguous geographic subdivision does meet the legitimate statutory and regulatory definitions of a high unemployment area."

Indeed, USCIS policy and statements, as noted below, give seemingly complete deference to the state.

Miami-based immigration lawyer José Latour, author of the Immigration Insider blog and commentator on EB-5 issues, told me, "I have been approached by several EB-5 projects sited in extremely posh areas where the state has issued TEA designation; in conducting the most preliminary due diligence, the surrounding census tracts are no different demographically and I steer clear."

In the Atlantic Yards case, an excursion beyond the immediate site managed to take in different demographics, and the state "identified" a TEA.

Looking at the numbers

Indeed, a closer look shows that the mapmakers deliberately omitted bordering census tracts with lower unemployment while creating a crescent with high unemployment statistics.

I couldn't get 2010 data, but did get 2000 unemployment rates from Community Boards 2, 3, 6, 8. (I assume the current data is roughly comparable, even if the overall numbers are likely now higher.)

The first four tracts on the list encompass the project site, with low unemployment: 129.01: 7.4%; 129.02: 5.2%; 161: 5.0%; 163: 3.1%.

The rest had much higher numbers: 203: 9.5%; 223: 23.4%; 225: 19.2%; 227: 12.2%; 229: 16.6%; 247: 21.0%; 267: 17.3%; 269: 15.3%; 277: 20.4%; 279: 19.4%; 281: 26.9%; 283: 29.6%.

Here are some contiguous census tracts--most much closer to the project site--that were left out. The 2000 unemployment rates were generally much lower:
35: 2.5%; 157: 3.1%; 159: 5.2%; 179: 8.5%; 199: 3.8%; 201: 4.8%; 205: 5.9%; 207: 5.0%; 215: 6.6%; 221: 33.1%; 231: 8.2%; 245: 6.9%; 249: 8.5%; 271.01: 5.7%; 315: 13.9%.

Law offers leeway on drawing the map

The law offers significant leeway regarding how to draw a "geographic subdivision."

Federal law (8 CFR 204.6) seems straightforward. It allows applicants provide their own evidence that "the metropolitan statistical area, the specific county within a metropolitan statistical area, or the county in which a city or town with a population of 20,000" has an unemployment rate of 150% the national average.

Alternatively, the law allows a letter from the state that "certifies that the geographic or political subdivision" including the new enterprise has a high unemployment rate.

That letter must meet the requirements of 8 CFR 204.6(i) , which states:
Evidence of such designation, including a description of the boundaries of the geographic or political subdivision and the method or methods by which the unemployment statistics were obtained, may be provided to a prospective alien entrepreneur for submission with Form I-526.
Also, a state official must ensure that unemployment rates are appropriately certified. But the law does not define "geographic subdivision."

What about gerrymandering?

Chapter 22 of the USCIS Adjudicator's Field Manual recognizes the potential for gerrymandering, but not when the state draws the map, only when the applicants do so, as is their option. The manual explains:
In some instances Form I-526 petitioners may claim high unemployment in only a portion or portions of a geographic area or political subdivision for which distinct unemployment data is not readily available to the general public from federal or state governmental sources. This may be indicative of an attempt by the petitioner to “gerrymander” a finding of high unemployment when in fact the area does not qualify as being a high unemployment area. Such a claim is not sufficient to establish that the area is a high unemployment area unless it is accompanied by a designation from an authorized authority of the state government
(Emphases added)

In other words, a state designation resolves the issue.

What if the state's behavior is suspect? 

The field manual does not address the possibility of the state gerrymandering the map, but urges deference:
The designation of high unemployment areas are within the purview of each U.S. state governor, or if applicable, his or her designee. USCIS personnel have no substantive authority to question or challenge such high unemployment designations....
Others have raised questions about gerrymandering. In a 3/17/11 USCIS stakeholder presentation, the agency was asked:
What is the meaning of a "geographic subdivision" in 8 CFR § 204.6(i)?  What are the limitations in creating and designating such a special area?... Yet, current practice by applicants interprets it to be any area carved out at will, irrespective of geographical and political boundaries.  It goes as far as creating an area by "statistical gymnastics" that yields a qualifying rate for high unemployment area designation and bears no relationship to any economic or employment effect of the prospective business.
(Emphasis added)

The USCIS deferred to the states:
With respect to geographic and political subdivisions of this size, however, the Service believes that the enterprise of assembling and evaluating the data necessary to select targeted areas, and particularly the enterprise of defining the boundaries of such areas, should not be conducted exclusively at the Federal level without providing some opportunity for participation from state or local government.

If a state government is presented with a request to designate an area as a TEA that the state does not feel is appropriate, then the state may in its discretion refuse to issue the TEA designation.
Querying USCIS

I  queried USCIS: what's to stop a state from agreeing to designate an unrealistic, irregular area as a TEA?

Nothing, apparently. Spokesman Bill Wright stated:
Just to reemphasize, pursuant to the regulation, the state designates the geographic boundaries and we confirm whether the unemployment rate meets the regulatory requirement.  
The lawyers want to lock it in

Immigration lawyers, who can make big money from EB-5, want to let the states keep easing projects down the road. According to official USCIS notes from an 8/10/11 meeting between USCIS officials and representatives of the American Immigration Lawyers Association's (AILA) EB-5 Committee:
AILA is seeking confirmation that a state’s designation of a geographic area as a TEA will be given deference and not questioned by USCIS. AILA asserted that USCIS’s analysis of the TEA should focus solely on the required unemployment level; the state designation would define the boundaries while investors are required to present the methodology in support of the unemployment level.
Indeed, according to a USCIS Draft Policy Memorandum on EB-5 (the comment period ends today), the agency is focused only on confirming unemployment statistics.

The Brooklyn Navy Yard example

The Navy Yard Appendage
Thus, federal officials seem to be ignoring map-making services such as that New York State performed for the New York City Regional Center in March 2009, when the regional center was preparing for two projects at the Brooklyn Navy Yard.

The sequence of three letters (bottom, read from bottom up) began with a request from the NYCRC's Olsen.

In a 3/24/09 letter, he informed Scardamalia, the Department of Economic Development official, that the Brooklyn Navy Yard is in census tract 543. He then asked the agency to "determine if 63 Flushing Avenue, Brooklyn, NY in census tract 543 is a TEA."

The location of the Navy Yard was duly designated a TEA, but not because census tract 543 itself represented a TEA. After all, it's a non-residential industrial zone.

Scardamalia apparently contacted the DOL's Nardone, who, in a 3/26/09 response, duly "identified an area of high unemployment... that is composed of the adjoining census tracts 543 and 23 in Kings County."

In doing so, the state chose one of many adjacent tracts to graft onto the tract with the Navy Yard. (I've dubbed this one "The Navy Yard Appendage.")

Census tract 23 contains the New York City Housing Authority's Farragut Houses (left), with a high unemployment rate.

(While I couldn't get 2008 data, in 2000, census tract 543 had a population of zero, while tract 23 had a population of 4314 and an unemployment rate of 24.1%.)

Scardamalia on 3/26/09 passed on Nardone's letter to Olsen, certifying the TEA. The NYCRC then proceeded to market its project, including Scardamalia's letter in its video (see screenshot above, at 2:05 of the video).

The Battery Maritime example

An even more creative solution seems to  have been achieved for an EB-5 project in Lower Manhattan, America's financial capital. The Wall Street Journal reported 2/2/11:
The New York City Regional Center, the intermediary that arranged the financing for the Nets arena, is also planning to seek through EB-5 about $77 million for a stalled $90 million-plus boutique hotel and retail project in Lower Manhattan, people involved said. That project, the redevelopment of the historic Battery Maritime Building that is linked to a larger waterfront revitalization, was first planned by developer Dermot Co. in 2007 but was shelved when the credit markets dried up the following year.

The East River Stretch
Last year, still unable to find traditional private financing, Dermot turned to EB-5 to fill a financing gap.
So, how did the NYCRC and the state map a TEA?

In this case I must speculate, since I haven't found documents listing the census tracts at issue. However,  a Korean web site marketing the project offers an annotated map (right) that seems to point the way.

The outline in red encompasses the building slated for development, then moves uptown via the narrow FDR Drive.

(Update: the project includes a promenade along the East River to Brooklyn Bridge, so the map--also available on the NYCRC's Chinese web site--encompasses the full project site, not the TEA.

The Brooklyn Bridge marks the boundary of Community District 1. So I bet the TEA includes the Alfred E. Smith Houses, a 12-building public housing complex in Community District 3, on the other side of the bridge.)

So, would it be legal to create the TEA I've dubbed "The East River Stretch"?

Apparently, though it strains credulity.

Census tract 319 of Community District 1, which includes the Battery Maritime Building, also encompasses the FDR Drive. It's thus contiguous, if not significantly adjacent, to census tract 25 of Community District 3. The latter contains the housing project.

In 2000, Census tract 319, which had a minuscule population of 202 people over 16 years old, had a 0% unemployment rate. (Nearby census tracts 9, 7, 15.02, 15.01, and 29 had unemployment rates of 5.1%, 3.9%, 3.4%, 10.2%, and 6.3%, respectively.)

By contrast,  census tract 25, which had a population of 1643 people over 16 years old, had an unemployment rate of 17.2% in 2000. Assuming such statistics are similar, that would suffice for a TEA.

How to calculate a TEA

How does it work elsewhere? Florida immigration lawyer Latour told me, "The State of Florida, where we primarily work, has a very firm grasp of census share disaggregation (CSD) as permitted by the [U.S. Department of Labor]. If they can make it a TEA, they’ll help you, but when they can’t, they flat out tell you 'sorry, no can do' and that’s that."

But Latour acknowledges that's not the case everywhere. "Because of the political pressure on state governments to create new jobs (i.e., facilitate foreign investment), the theory of 'statistical gymnastics' may apply in other states," he said. "I’ve spoken to more than one state designated official in other jurisdictions whose response to my request for a CSD assessment when the proposed census tract was clearly not in a TEA, their response has been “census share WHAT??”

That suggests that some states do not even produce the rudimentary statistics to calculate local unemployment, as do Florida, New York, and other states.

Even a CSD assessment, which the Florida Agency for Workforce Innovation does, may produce a somewhat irregular map, though not necessarily on the scale of the "Bed-Stuy Boomerang." Above right are highlighted census tracts in and around Hollywood, FL that qualified a TEA for a project of the Hollywood Beach Regional Center.

Examples in Michigan and California


Some states establish TEAs in a more transparent manner than does New York. Consider the processes used by the states of Michigan and California.

Unlike New York State, Michigan publishes the unemployment rate of qualifying counties and census tracts.

California does not list census tracts, but rather qualifying metropolitan statistical areas, counties, cities, town, and census designated places ("a geographic entity that serves as the statistical counterpart of an incorporated place for the purpose of presenting census data for an area with a concentration of population, housing, and commercial structures that is identifiable by name").

Both seem to take a more transparent, rational approach when faced with a request for a "special designation of a smaller area within the otherwise non-qualifying area."

Michigan and California ask applicants for such designation to describe the boundaries of the geographic subdivision ("an area carved out based on the physical features of the Earth’s surface) or political subdivision ("a division of a state that exists primarily to discharge some function of local government").

To help the states evaluate impacts of such a designation, applicants should "also provide a logical economic reasoning," including information "necessary to justify the claim that the establishment of the new business will have a positive and meaningful effect on unemployment in the desired area."

That may all be boilerplate, but in theory it would preclude such designations as "The Bed-Stuy Boomerang." There is, of course, little proof that establishment of the Atlantic Yards project "will have a positive and meaningful effect on unemployment" in the TEA designated.

A better definition?

Some in the field think TEAs could be better defined. The 10/15/09 Bender's Immigration Bulletin suggests that existing Empowerment and Enterprise Zones, aimed at areas of disadvantage, should be integrated into the TEA definition:
The current definition of "Targeted Employment Area" is unique to the EB-5 program, and does not include other federal or state economic-incentive designations designed to promote economic  development. These designations already carry tax incentives, wage credits, or other investment incentives. Why not  immigration incentives?
Atlantic Yards TEA Letters


BNY TEA Letter

No comments:

Post a Comment