New York Times, EB-5 Visa TEAs, and Gerrymandering: Part II

Following the publication of the New York Times article on the EB-5 visa program, Rules Stretched as Green Cards Go to Investors, and the follow-up editorial in the Times, EB5Info asked for and received comments from a number of economists and professionals whose job it is to interpret census tract data and submit requests to State officials for TEA designations to qualify their EB-5 visa projects for Regional Center designation by USCIS.  The following is a compilation of opinions and thoughts from that request. 

One of the first responses we received was by economist Scott Barnhart, whose complete commentary can be found here. He writes:

Like any large government program the legislation is loosely written and leaves much to be filled in by practitioners.  Will there be pushing of the envelope or even abuse of the regulations?  As with most government programs, more than likely the answer is yes, but should the program be axed because of this or should it be changed for the better?  I would challenge anyone to find even one government program where abuse is absent.  Given these circumstances, what issues raised in the article are valid and how can they be resolved?

To be clear, there is much ambiguity in the law and little guidance given by USCIS in applying it.  Is this the result of another bureaucratic government agency run amok?  Perhaps, but my limited experience through a number of RFE’s is they seem to be applying the code as best they can, but are no doubt over run with work and sorely underfunded.  If USCIS is like most state and federal government agencies these days, their funding has likely been cut to the bone in recent years and they are forced to do much more with much less.

Then what about the 150% unemployment and gerrymandering the census tracts?  Simply put, one must demonstrate that the unemployment rate is at least 150% of the national average.  The relevant code indicates what must be presented:

Evidence that the metropolitan statistical area … in which the new commercial enterprise is principally doing business has experienced an average unemployment rate of 150 percent of the national average rate.

In addition, a state may designate the area as one of high unemployment:

The state government of any state of the United States may designate a particular geographic or political subdivision located within a metropolitan statistical area or within a city or town having a population of 20,000 or more within such state as an area of high unemployment (at least 150 percent of the national average rate).

So much of the detail is left to the individual developer, regional center owner or consultant, and to the state. For example, which national average:

• The previous calendar year?
• The preceding 12 months from the current date?  
• Seasonally or not-seasonally adjusted data?  

In addition, considering local area unemployment has not been measured at the census tract level since the 2000 census, how does one measure census tract unemployment rates until the Census Bureau’s American Community Survey is reported, probably in 2015?  Census sharing methods are currently used in which 2000 census tract level employment ratios are applied to 2011 county level employment data. The approach obviously has its drawbacks. Finally, there is little guidance concerning how to collect census tracts surrounding the tract in which the project is located to use for unemployment calculations, hence the claim of gerrymandering. 

If USCIS would level the playing field for all states, any perception of whether one state is “unfairly” getting more projects than is justified would be eliminated.  This could be accomplished either by allowing states to claim any type of geographic region (or shape) as long as the resulting unemployment rate reaches the 150% rate or it can be done by setting very strict guidelines that all states must follow.  Either of the methods will work because all regional centers and state agencies will be on equal footing.  However, even the casual observer will recognize that although leveling the playing field would eliminate the perception that some states/projects have an unfair advantage, these two methods can have quite different outcomes; the former approach would foster regional center and job growth, while the second could inhibit regional center and job growth if the guidelines are too restrictive, which is clearly not the intent of the legislation.  The second approach also could be very difficult to implement, e.g., concentric areas surrounding the project tract, etc., considering census tracts are designed to be homogenous regarding population and economic status, but geographically are very diverse.

So what do the regulations say?

Joe Whalen, a former adjudicator with USCIS and someone who is intimately familiar with the program and process, references the statutes here:

The regulation at issue is found within 8 CFR § 204.6, specifically:

(i) State designation of a high unemployment area. The state government of any state of the United States may designate a particular geographic or political subdivision located within a metropolitan statistical area or within a city or town having a population of 20,000 or more within such state as an area of high unemployment (at least 150 percent of the national average rate). 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. Before any such designation is made, an official of the state must notify [USCIS] of the agency, board, or other appropriate governmental body of the state which shall be delegated the authority to certify that the geographic or political subdivision is a high unemployment area.

The States (including DC and territories) have been given specific authority within the EB-5 program within specific limits. The State has the authority to designate a geographic or political subdivision within that State as a high unemployment area (HUA) for EB-5 purposes. That State must provide a description of it methods in making that determination. However, no State is required to participate in the EB-5 program or to make any such determination.  Elsewhere in the regulations at 8 CFR 204.6(j)(6)(ii)(B) discussing “evidence”, there exits an option for a State to make the determination and issue a letter to the EB-5 alien investor for use in supporting his/her USCIS form I-526 petition. In the alternative, the EB-5 petitioner can do all the statistical analysis him/herself rather than ask the State to do it.  

In footnote #1 within the AAO non-precedent of September 21, 2010, (excerpt follows) which ultimately upheld the Service Center Director’s Denial of an I-526 filed by a Regional Center-affiliated EB-5 immigrant investor, USCIS’s Appellate Body-the Administrative Appeals Office (AAO)- found itself, as an arm of USCIS bound by the agency’s own regulations. In that the regulation in question, which the CSC director and AAO determined had been used to support an untenable outcome had, in fact, been properly promulgated pursuant to APA comment-and-notice rulemaking AAO lacked the authority to overrule it.

This is so because it was not a rule set by its own precedent but rather by the comment-and-notice process. After years of increased experience and developments in various key issues, AAO has expressed on behalf of USCIS and in hindsight that 8 C.F.R. § 204.6 (j) (6) (ii) (B) has been used to support an outcome that now appears to be in contravention to clear Congressional Intent. A Federal Court has the power to quash such a regulation as ultra vires, even through its broad interpretive authority, the agency does not have that power. If an agency’s Appellate Body attempted to declare as ultra vires, a longstanding rule that had been duly promulgated, then that action would itself be ultra vires. See 5 USC § 553(esp. (b)-(d)) and § 558 (b) “A sanction may not be imposed or a substantive rule or order issued except within jurisdiction delegated to the agency and as authorized by law.”  At first glance, it looks like it may be time for a regulatory change. Or is it?  As AAO has already stated:

1 The proposed investment will be wholly and entirely within Ward 2, a ward that is not itself suffering high unemployment in relation to the national unemployment rate. The director's conclusion that the investment will be within a targeted employment area is based on a designation by [REDACTED INFORMATION] for Planning and Economic Development, Washington, D.C. pursuant to 8 C.F.R. § 204.6(j)(6)(ii)(B). [REDACTED’s] designation includes Ward 2, but, of necessity, includes other wards and census tracts within D.C. to reach the necessary average unemployment rate. The director's conclusion that we must accept the designation is a reasonable interpretation of 8 C.F.R. § 204.6(j)(6)(ii)(B). That said, it is clear that the petitioner's investment of only $500,000 wholly within a ward that is not itself suffering high unemployment completely undermines the congressional intent underlying section 203(b)(5)(C)(ii) of the Act. Specifically, Congress intended that the reduced investment amount would encourage investment in areas that are truly suffering high unemployment. While we are bound by 8 C.F.R. § 204.6(j)(6)(ii)(B), it would appear that this regulation has produced unintended consequences that are clearly contrary to congressional intent.

Congressional intent and interpretation

This comes from an economist who is frequently consulted by regional centers regarding Requests for Evidence (RFEs) or denials from USCIS regarding their petitions and works with them to find workable solutions that would be accepted:

While gerrymandering can sometimes be a concern, there is nothing wrong with configuring an irregular geographical area if there is reasonable expectation that potential hiring opportunities will be available for residents of these high unemployment areas.

Often times, especially in inner city areas, there will be commercial areas with low unemployment where relatively few people live, but nearby will be economically distressed areas where residents can benefit by the employment opportunities created in easy to get to nearby locations.
 
I have found that in cities like Boston, New York, and Chicago, developers want to build in commercial zones where few people live and/or unemployment may be low, but they need to draw their workforce from these surrounding, more populous high unemployment areas.
 
Often times we have a skills mismatch.  For example, if a hotel is being built in an affluent area of downtown Manhattan, can we expect that the service workers will be drawn from that affluent census tract?  No, the potential labor pool will most likely come from the not too distant higher unemployment areas.  The same type of example holds with a manufacturing plant that may be built in a rather affluent town but needs to draw workers from the neighboring blue collar town.
 
It is not the wealth of the businesses in an area that is important, but the economic status of the people who reside in that and nearby areas.  Usually when unemployment in a census tract is zero or very low there is a very small labor force in the area due to the fact that the area may be mostly commercial, a largely undeveloped parcel of land, or a waterfront area.
 
The reference to the tract that had zero unemployment for 5 years is misleading.  As the numbers are based on a 2000 census share methodology, it will always be estimated as zero until a new data source is used.
 
It’s ridiculous to consider any two tract combination as gerrymandering.  In the mid-town Manhattan reference, each census tract is probably one square block.  If the first tract had zero unemployment and had the same labor force size as the second, than unemployment in the second tract would need to be at least 28.8% just to reach a combined qualifying rate of 14.4%.

If the second tract was smaller, it would need to have even a higher rate than 28.8%.  In the most likely scenario where the labor force in the first tract is probably smaller, the second tract would still need a rate in excess of 14.4%.  For example, tract one (LF 100 and unemployment 0) and tract two (LF 900 and unemployment 150 or 16.7%).  The combined numbers would be (LF 1,000 and unemployment 150 or 15.0%).  I can’t imagine anyone questioning any of these scenarios as gerrymandering and implying that the people who live one block away would not benefit from the new employment opportunities.

How do states interpret the regulations?

From one economist:

There are differences between the states, but for the most part the TEA certifications need to follow the calculation guidelines for the Census Share Method that is provided by the US DOL and has been consistently been updated (about every 3-5 years).  That being said, there are some states that are more permissive than others in terms of geographic composition (the positive interpretation is that it is “flexibility”).  There are some cases where it can involve “gerrymandering,” but NY is really not one of the “bad guys.

Another economist familiar with program writes:

I think some states do a better job of monitoring (not limiting) how TEAs are set up.  They may approve a large geographic area that could wrongly be seen as gerrymandering but fits into a logical jurisdiction, while at the same time they may reject a smaller configuration because it  doesn’t show any reasonable economic linkage.  The ones who keep the TEA designations as outlined and in line with the law are actually the ones that can be questioned for gerrymandering.

Most states, like New York and Florida, will allow anything within the rules and regulations, i.e. contiguous geographic or political subdivisions using approved BLS methodologies.  Some of the states that have deviated from the TEA regulations have been influenced by USCIS in their attempts to eliminate what they consider to be gerrymandering.  Some other states have just taken it upon themselves to simplify their own workload by arbitrarily deciding for instance to just allow any two tract combination or in one case, only individual tracts (this state actually uses outdated data from about three years ago and feels no obligation to update).

One such state influenced by USCIS is California.  On the one hand they are to be complemented for thoroughly reviewing each request in terms of the information they require, the potential economic impact, and the legitimacy of the area being requested (i.e. no odd configurations or gerrymandering).  They do their due diligence.  On the other hand, while they adhere to the regulations that specify contiguous geographic or political subdivisions, they have incorrectly interpreted that census tracts are not geographic subdivisions (the Census Bureau specifically defines census tracts as geographic subdivisions).

As a result, they will not designate any individual or even small groups of contiguous tracts, even if for example they all had unemployment rates of over 30%.  Yet, you can carve out a city council district in a large city that meets the definition of a political subdivision, put together the 50-60 census tracts that make up the area, and if the rate happens to qualify, you may be able to get your wealthy census tract in a TEA.  This may seem to make no sense, but then again, it makes just as much sense as a city or MSA that automatically qualifies.  Why is it OK to put a TEA project in a wealthy town in a large qualifying MSA or a wealthy census tract in a qualifying city but not OK to put a TEA in a wealthy census tract in NYC with high unemployment areas not far away?  

Texas operates in a world of their own.  It is the only state where authorization was not given to a state agency but to each individual Mayor.  It is supposed to be given to a state agency but apparently USCIS has decided to let them continue in this manner (I really think USCIS should address this inconsistency before anything else).  Just as each state can establish their own TEA procedures, each Texas city can do likewise.  Compounding matters is when a TEA crosses multiple jurisdictions.
 
In about half the states, the Agency authorized by the Governor to do the designations, is not the labor market agency that compiles the data and does the calculations.  For example, in NY, the agency that prepares the letters is Empire State Development but that person, Leonard Gaines, is not the Agency spokesperson quoted in the article.  The NY Department of Labor is the one that does the calculations but you have to go through Empire State Development.  In what may seem to you as odd after reading the article, I have probably had more difficulty in getting areas to qualify as TEAs in NY than in any other state.  They have been cooperative but I have worked mainly with upstate areas where there are many pockets of high unemployment but the numbers are just not high enough to work into TEAs.
 
California also has a different agency that does the designations from the ones that produce the data but I believe there is a group effort in setting their policy decisions.

Joe Whalen makes this note about California:

I have made no secret that I admire the way that California has addressed its role for that limited part of the EB-5 program within its purview. The following is found on its website here.

The Secretary of the Business, Transportation and Housing Agency, or the Secretary’s designee, may provide certifications for applicants that a proposed new business falls into an area (MSA, county, CDP, or city) that qualifies as a High Unemployment Area (HUA). In order to be classified as a High Unemployment Area, the MSA, county, CDP, or city must have experienced an unemployment rate that was equal to or exceeded 150% of the 2010 national employment rate. In 2010, 150% of the 9.6% average national unemployment rate was equal to 14.40%. (Requesting such a certification letter may not be necessary as the web site publishing this information contains a self certification option.

There is, of course, much more available there. Any State that chooses not to follow California’s example could easily do worse, but I doubt any could do better.

Scott Barnhart summarizes the question with the following comments:

Will jobs be lost in Nevada, California or other states because a 34 floor glass tower is built in New York?  Perhaps, but currently there is great demand for US citizenship and over 200 regional centers from which to choose.  Moreover, given the current state of the US economy, the economic benefits of projects rejected based on the 150% unemployment target may simply be lost forever as locations in other area or other states are not close substitutes.

For example, if the 34 floor tower typically used for retail, office space and/or residential purposes did not qualify in New York, one can be assured that states with the highest unemployment levels are not likely close substitutes for a Manhattan address for either the developer or prospective investors, so this project would likely be shelved.  Similarly, a large condominium in Florida will not sell if located in a high unemployment area away from the coast instead of a lower unemployment area on the coast, yet the labor will be imported to the site.

Both projects though will provide large economic benefits to their regions.  So given the current robust demand for citizenship in the US, this is likely not an issue at the present, but this may change in the future.  Again, leveling the playing field would come a long way in resolving any perceived or real issue.

Solutions?

Several economists offered solutions, one noted the following:

I certainly can see how projects can be manipulated to approve destinations that may otherwise not qualify but I don’t see this as fraud as it is within the TEA rules and regulations. The fraud comes into play with the false promises to potential investors. Sure, there could be abuse with TEAs but that would have to do with having no intention of offering employment opportunities to people in these areas."

I think a better way of controlling abuse would be to monitor the residencies of the people hired. In other words, if 20 people were hired, did they all come from the project location with very low unemployment or did at least some come from high unemployment areas within the TEA geography? Is it OK to include an Indian reservation just as any other high unemployment area so maybe there should be an expectation to hire people from these areas? At least this way they will have some employment opportunities whereas if you don’t allow the high area to be included, no opportunities at all will exist.

In our final article on this subject, we will profile a response from the State of Florida, which has the second largest concentration of EB-5 Regional Centers in the country and has many projects that are being promoted as located in targeted employment areas.