Targeted Employment Areas from November 21

The EB-5 Immigrant Investor Program Modernization Regulation Final Rule took effect on November 21, 2019, and  changed USCIS Policy for Targeted Employment Area (TEA) definitions and process.  Rather than reacting with questions and complaints, I carefully review the specific content of current TEA policy, place changes in context, and address the theoretical background and practical implications. This simple post took a great deal of work and thought.

POST AGENDA

A. Who is affected by the new TEA rules?

B. What areas can now qualify as a TEA?

C. What data can now be used to qualify a TEA?

D. Who determines TEAs, and how and when?

DISCUSSION

A. Who is affected by the new TEA rules?

New TEA rules apply specifically and only to all I-526 petitions filed on or after November 21, 2019. (The final rule for the EB-5 regulation gave a 120-day implementation/transition period: that period started upon publication of the final rule on July 24, and ended when the rule took effect on November 21.)

“Applies to Form I-526 filed on or after Nov 21” is a hard and fast rule. This is very clear in the final rule text, and confirmed by subsequent comments.  The new TEA rules apply to every I-526 filed from 11/21 – no matter if the project had previous investors or an Exemplar I-526 approval pre-11/21, and no matter if the investor is seeking to retain a pre-11/21 priority date when filing the new I-526. The new TEA rules do not apply to any I-526 filed before 11/21, even if the investor funds had not been fully invested in the NCE or deployed to the JCE before 11/21. IPO Chief Sarah Kendall reassured the IIUSA conference that her staff have been trained to adjudicate each pending I-526 based on the rules in place at the time that I-526 was filed. People who filed I-526 before 11/21/2019 are only indirectly affected by the new TEA rules, to the extent that open offerings must now be amended. But policy specifies that such conforming amendments will not count as material change for past investors.  As always, TEA qualification is not an issue at the visa application or I-829 stages.

While the new EB-5 regulation applies to all I-526 filed going forward, it does not apply entirely new rules. Rural areas, for example, have the same definition before and after November 21. The standards for a high-employment MSA TEA are no different now than they were under previous policy. Data recommendations remain unchanged. This post goes on to review what is and is not new.

B. What areas can qualify as a TEA?

The old rules gave the states authority and flexibility to designate geographic areas for TEAs. The new rules instead specify a limited list of possible TEA areas defined by DHS. From now on, a job-creating entity is in a TEA if it is in one of the following defined areas:

1. A rural area, defined as an area that is not in a standard Metropolitan Statistical Area as defined by the Office of Management & Budget, and not within the outer boundary of any city or town having a population of over 20,000 or more based on the most recent decennial census; or
2. A high unemployment area, defined as an area that has experienced unemployment of at least 150 percent of the national average rate. For high unemployment, “area” can only mean:
   1. A Metropolitan Statistical Area (MSA)
   2. A county within an MSA
   3. A county that contains a city or town with 20,000+ population
   4. A city or town with population of 20,000+ or more which is outside an MSA
   5. A single census tract, and/or
   6. A group of census tracts comprising the census tract where the job-creating entity principally does business, plus any or all directly adjacent census tracts (PDF p. 11-12 of the NPRM illustrate specifically what DHS has in mind.)

Option 2.4 and 2.6 were revised by the regulations; other options match previous policy. The new list of geographies that can qualify excludes several areas that states were willing to designate as TEAs: census blocks, census block groups, and sprawling groups of census tracts.

As before, the EB-5-funded job-creating entity must principally do business and create jobs within the TEA area.

If you have an EB-5 project in mind, how can you find out the potentially qualifying “areas” to which it belongs?  You can get a quick sense of geography just by looking up the city/town name on Wikipedia, which will tell you to what county and MSA (if any) the place belongs, and give ballpark population data. From there I’d go and enter the project address in the government’s FFIEC mapping system, which will identify the census tract for that address, show the directly adjacent census tract numbers, and confirm whether or not the address is in an MSA. Once having identified the possible geographic areas for a TEA determination, you’re ready to think about data.

C. What data can be used to qualify a TEA?

Since November 21, USCIS does not automatically approve any particular unemployment dataset for TEAs. Before November 21, USCIS also did not offer deference for unemployment data and methods. Regarding TEA data, the regulation simply repeats language that was introduced back in the May 30, 2013 EB-5 Policy Memo, and that has been included in (6)(G)(2)(A)(5) of each Policy Manual iteration since: “USCIS will review determinations of the unemployment rate” and “acceptable data sources for purposes of calculating unemployment include U.S. Census Bureau data (including data from the American Community Survey) and data from the Bureau of Labor Statistics (including data from the Local Area Unemployment Statistics).”  BLS data was specifically identified as acceptable in the December 2009 Neufeld Memo. By 2012, USCIS clarified that it would accept ACS data with census share methodology for subareas not covered by BLS. The point of this history lesson: we are not standing on new ground now, regarding data.  USCIS only changed its deference to state designations of TEA geographies — there never was deference for the data portion of TEA analysis, and suggested data sources remain unchanged. In fact, TEA requirements are, if anything, clearer now than they used to be.  To quote from discussion in the regulation final rule related to acceptable data:

• The regulation “does not provide one specific set of data from which petitioners can draw to demonstrate their investment is being made in a TEA. Rather, the burden is on the petitioner to provide DHS with evidence documenting that the area in which the petitioner has invested is a high unemployment area, and such evidence should be reliable and verifiable.” [Consistent with previous policy.]
• “The data necessary for the TEA designation determination is publicly available from the Bureau of Labor Statistics or U.S. Census Bureau. A TEA designation request alternatively can be supported with other data, public or private, provided that DHS can validate that data.” [Consistent with previous guidance.]
• “Regardless of which reliable and verifiable data petitioners choose to present to DHS, the data should be internally consistent. If petitioners rely on ACS data to determine the unemployment rate for the requested TEA, they should also rely on ACS data to determine the national unemployment area to which the TEA is compared.” If considering state data, the rule cautions that “petitioners may not be able to compare the state census tract data to a national unemployment rate that utilizes the same methodology.”
• To calculate the weighted average for a group of census tracts, the Final Rule opts to keep the cumbersome method described in Footnote 41 of the NPRM, except specifying that civilian labor force rather than total labor force should be used: (1) divide the labor force of a census tract by the labor force of the entire TEA area; (2) multiply this figure by the unemployment rate of that census tract to calculate a weighted unemployment rate for that tract; (3) repeat Steps 1-2 for each tract in the TEA area; (4) sum the weighted unemployment rates for all tracts in the group to calculate a total that can then be compared with the national unemployment rate.

The final rule optimistically states that the TEA process can be “easily navigated by any petitioner–whether associated with a regional center or not–for little or no cost,” because “unemployment data is readily available by which they can determine if an investment in a particular area satisfies applicable TEA designation requirements.”

The person who wrote the rule clearly never tried to pick an address, venture online, and find and interpret appropriate unemployment data for that location at the MSA, county, city, and census tract levels. It’s not easy.  In practice, most people will have to pay qualified consultants to help with the data portion of TEA determinations. But if you still want a sense of what’s available to the public, a few links:

• Guidance for Labor Force Statistics Data Users, published by the U.S. Census Bureau, reviews the types and sources of unemployment data available for different types of geographic areas. The EB-5 regulation merely acknowledges that “no one dataset is perfect for every scenario”; Census Bureau guidance explains which dataset to use for which scenario.
• The Bureau of Labor Statistics publishes monthly and annual unemployment data for the nation, MSAs, and counties. TEA designations have traditionally referenced the annual data –  one doesn’t want to update the TEA analysis every month, and annual data facilitates apples-to-apples comparisons across geographies. To find the annual average employment rate for an MSA or county, open the BLS Local Area Unemployment Statistics page to the section on Tables and Maps Created by BLS. Within that section, scroll down to the “Annual Average” subsection, and within that subsection to “Metropolitan Area Data” and “County Data.” (This link jumps directly to annual average county data.) Alternatively, perform a search using the Featured LAU Searchable Database. Either way, you will be directed to a table crammed with data that’s ugly and not convenient to print and share, but reliable and verifiable.  The nationwide annual average employment rate, for comparison, is on this page. Monthly data is less workable for TEA purposes, but has a benefit of coming in focused and print-friendly reports.  For county data, I like the BLS reports linked to the Geographic Information > Economic Summaries page. They’re in PDF format, and handily compare county unemployment with nationwide unemployment, as required for TEA designation. If my project were in a clear high-unemployment county covered by one of these reports, I’d consider this resource.  (Just keeping in mind that BLS refreshes these reports every month, and does not archive older versions, so they’re not directly verifiable over time. Archives of monthly MSA data can be found here, but monthly county data archives are tougher to locate.)
• The U.S. Census Bureau’s American Community Survey comes in because BLS does not collect or report unemployment data at the census tract level, or for cities outside MSAs. One can search for ACS data for employment by geography, including at the census tract level, using the advanced search function in the old factfinder.census.gov or the new data.census.gov. The census bureau search functions are not friendly to casual human users, and their employment data is relatively outdated. State TEA designations would frequently update ACS employment data for census tracts with reference to the more recent BLS unemployment data at the county level using a method called census share (as described here by BLS and here in the EB-5 context, for example). But I don’t recommend trying this at home. You’ll want an experienced professional to crunch data for any TEA below the county level. But in the meantime, to get a preliminary sense of unemployment at the census tract level, try using one of the free mapping tools for EB-5, such as by IIUSA and Impact DataSource.
• State workforce agencies also publish labor market information, as part of a nationally designed LMI infrastructure that connects BLS, the Census Bureau, and each state. Such state-reported data should also be acceptable, as it’s linked to the BLS and ACS data specifically name-checked by DHS as evidence that “should be reliable and verifiable.” The challenge is to determine which national unemployment rate is comparable to the state unemployment rate. (For example, should use the BLS national rate if the state is referencing BLS data or ACS data updated with census share methodology, the ACS 2017 5-year estimate unemployment rate if the state’s numbers are based on ACS 2017 5-year estimates, etc.)

D. Who determines TEAs, and how and when?

We’d gotten comfortable thinking about TEAs determined in advance by state agencies, via designation letters.  That TEA letter comfort was useful for marketing, but somewhat of an illusion.  In fact, TEA determinations were never fixed as of the date of a letter, because policy has required TEA status to be determined for each investor based on the date of investment or I-526 filing (whichever came first). As discussed above, state letters were not granted automatic deference; USCIS reserved the right to question the timeliness, data, and methods. We’ve long had to work with a degree of uncertainty and case-by-case discretion by USCIS when it comes to TEAs.  The new situation is not necessarily more ambiguous, just different.

Determining the geography component of TEAS

The regulations depart from previous practice primarily by eliminating state designation of TEA geography. The power to designate an “area” now lies with DHS, and DHS has made the geography determination once and for all in advance by specifying a limited and strictly defined list of possible areas in the final rule.  Petitioners just have to pick one of the defined area types (see the list in Section B above), and provide unemployment data for that area.

DHS intended for the new reg to eliminate ambiguity and individual discretion from the geography element of TEAs, and apparently succeeded.  There’s no need for anyone to “designate” the geography portion of a TEA; a list of acceptable geographic areas has already been defined.

Determining the unemployment data component of TEAs

As discussed above, the process for data remains unchanged in theory. Whoever provided the TEA data, USCIS has always reviewed and assessed that data in context of each investor petition, and determined as part of I-526 adjudication whether TEA requirements were met.

In the past, we’ve used letters from state agencies as a vehicle for presenting unemployment data to USCIS. Nothing in the regs would prevent us from continuing to do this. DHS has relieved state agencies of the extraneous responsibility of drawing boundaries for EB-5 incentive areas. DHS has not stripped state workforce agencies of their own mandate to supply workforce data.  State agencies may or may not be amenable to continued requests from EB-5 users for unemployment reports customized to DHS-defined areas. But state letters are a tidy and convenient vehicle for reliable unemployment data, and it doesn’t hurt to ask. State workforce agencies are subject to uniform, nationally-designed standards for Labor Market Information (LMI) reporting, so USCIS couldn’t suspect the agencies of being idiosyncratic or inventive with the data portion of a TEA determination.  At least, I would try the state workforce agency, before downloading hundred-column spreadsheets myself from the internet, and before requesting unemployment analysis from some former Uber driver Joe Smith now d/b/a TEA Designations, LLC.

In the past, we’ve used consultants, particularly EB-5 experienced economists, to help identify TEAs and approach states for letters. Now, we can ask those same consultants to prepare letters with unemployment analysis to present to investors and USCIS. We should demand that the consultant’s work product meet these standards: (1) define the geographic area with specific references to the latest EB-5 policy/regs, (2) identify the sources for population and employment data with sufficient specificity to allow the reader to go online and find the publicly-available data referenced, (3) show all the steps in any calculation, (4) explain, with references to the EB-5 regulation and BLS and/or Census Bureau guidance, why the analysis is reasonable. If you, as a reader, can verify the data and see that the analysis aligns with authoritative guidance, odds are the USCIS adjudicator will likewise find it reliable and verifiable. I’d demand more detail and footnotes from a consultant report than from a state letter. Compared with the Georgia Department of Labor, Joe Smith has a hurdle to prove his data and methods.

Whoever wrote the regulation seems to think that people can easily go online and get appropriate unemployment data to print out as evidence.  As briefly discussed above, BLS and ACS data is not that easy to navigate or interpret (or even print, for that matter), and info from third party mapping programs and other sources may or may not be up-to-date, reliable, and verifiable. It takes some expertise even to accomplish a simple task like choosing a national unemployment rate that’s internally consistent with a given local area unemployment rate. And it takes considerable expertise to bolster a TEA analysis with references and explanations that leave no crack for USCIS questions.  So I think we’re still in a world of securing TEAs using letters and reports – the only question is: who prepares them.

Some wondered whether DHS itself could start providing TEA designations in advance of investor petitions. The regulation states that “this rule does not establish a separate application or process for obtaining TEA designation from USCIS prior to filing the EB-5 immigrant petition and USCIS will not issue separate TEA designation letters for areas of high unemployment.” The regulation offers that a regional center may seek TEA determination by filing an exemplar petition, and “If the exemplar application is approved, the approval (including the TEA determination) will receive deference in individual investor petition filings associated with that exemplar in accordance with existing USCIS policy (for example, absent a material change in facts affecting the underlying favorable determination or its applicability to eligibility for the individual investor).” However, this offer is 100% useless and void, unless USCIS can start providing exemplar approvals in less than the time that it takes unemployment data to expire, and thus become inapplicable to individual investor eligibility.  The currently posted I-924 processing time is 62 to 115 months. No investor can claim TEA status at the time of investment or I-526 filing based on a TEA determination calculated five to ten years previously.

Regarding timing, the regulations do not imply a change from past practice.  A TEA determination has always needed to be valid at the time of an EB-5 investor’s investment or I-526 filing, whichever comes first. A TEA determination has always been valid so long as the underlying data is the most current available. Most state letters were effective for up to a year because they calculated unemployment rates from annual average data that is, naturally, updated just once a year. The regulations do not change what unemployment data is available, or when BLS and the Census Bureau publish updates. The regs do not suggest that DHS had a problem with the unemployment data and methods that states have used all these years, only a problem with how states were willing to gerrymander geographies. So I do not see any new policy or new ambiguity, when it comes to timing of TEA determinations.   When a consultant creates a TEA analysis, just be sure to specify the validity period for the underlying data, and point out that this defines the shelf life of the TEA determination.