Tag Archives: high-skilled hacks

High-skilled hacks: a (very) brief overview of H-1Bs (more to follow)

This post is part of my “high-skilled hacks” series, focused on immigration to the United States. The series explores various workarounds and caveats to immigration law that high-skilled workers and their employers have discovered in order to further their own interests, at the expense of the original intent of immigration law. Through the series, I try to argue that, although these hacks are an improvement over an alternative where only the basic immigration system existed, freer migration for all would be simpler, fairer, more efficient, and more just. The introductory post of the series is here.

My primary purpose here is to provide information rather than push a viewpoint. Nonetheless, I do not claim to hide my perspective. Rather, the purpose here is to present information as seen from a particular viewpoint (something akin to historical revisionism). Later posts in this series may include more focused discussion of my viewpoint, as I delve deeper into the regulations.

In this respect, my writing here differs from the many Wikipedia pages related to the H-1B that I have created, where I strive to stick to facts and avoid revealing viewpoints. These pages include: Labor Condition Application, Form I-129, Premium Processing Service, H-1B-dependent employer, public access file, American Competitiveness and Workforce Improvement Act (ACWIA), American Competitiveness in the 21st Century Act, H-1B Visa Reform Act of 2004, and Employ American Workers Act. For full disclosure, I also paid for the creation of the pages on H-1A visa and H-1C visa.

The following topics related to the H-1B visa are covered in this post.

  • Steps to getting started in H-1B status. (more)
  • Relationship, similarity and differences between the H-1B and permanent immigration. (more)
  • Key stages in the legislative history of the H-1B. (more)
  • The H-1B annual cycle and cap. This is a very brief overview of the mechanics of the status. Workarounds, such as the use of Optional Practical Training and the use of other work visas as a temporary measure, are discussed. (more)
  • Comparing the stated purpose and real use of the H-1B program. Once again, this is a brief summary that will be elaborated on in future posts. (more)
  • How the H-1B compares with other options: L-1, TN-1, O-1, and H-2B (to name the most salient alternatives). (more)

The following are not covered in this post and will be the subject of future posts.

  • The role of H-1B1 and E-3, and their effects on the market of technology workers from Singapore, Chile, and Australia in the United States.
  • A deeper look at the different occupations, intended employment areas, and countries of origin for users of the H-1B program.
  • A full analysis of the educational credentialism in the H-1B system, contrasted with other temporary and permanent immigration categories. I discussed some of these points in an Open Borders Action Group post, that Bryan Capan reblogged on EconLog. But I intend to cover it in more depth.
  • A better overview of exactly how the H-1B lottery works. Basically, understanding what your chances are based on whether you have or don’t have a master’s degree.
  • The role of the H-4, the status for dependents of H-1B holders, and how the two H statuses interact.
  • Comparison with analogous statuses in other countries, i.e., with other generic temporary skilled work visa categories.
  • What the whole H-1B application process means for people who aspire to having a job (temporarily or permanently) in the United States, particularly people who are not from any of the treaty countries (Canada, Mexico, Australia, Singapore, and Chile). Note that the current post touches only briefly on the issue from the perspective of a job-seeker.
  • A detailed discussion of the Labor Condition Application and U.S. Department of Labor investigative authority. This will included discussion of H-1B-dependence, its current and past prevalence, and its implications for the mid-level tech workers who account for about 50% of H-1B use.
  • A full discussion of various criticisms leveled at the H-1B program, by people such as Norm Matloff, Michelle Malkin and John Miano in their book Sold Out, and labor unions and labor-advocacy think tanks such as the Economic Policy Institute (see for instance here). Either in the same post or in a different one, I will also look at the pro-H-1B rhetoric used by different groups ranging from immigration lawyers to liberal think tanks and advocacy centers to libertarian think tanks.

The list of things I don’t cover here also includes other stuff I haven’t yet even realized is important enough to be discussed! But the above list is already pretty daunting.

The H-1B: the most important work visa

One of the most important sources of high-skilled migration to the United States is the H-1B visa. The H-1B is significant in at least a few ways:

  • For many high-skilled workers who enter the United States for work, the H-1B is how they are able to first enter.
  • Even those who enter for work in other ways (such as the TN-visa, L-visa, or OPT on F student status) often transition to the H-1B when they get the opportunity.
  • While many who come to the US on H-1B status eventually leave, many others settle permanently in the United States. Some do so by applying for permanent residency through one of the EB visas (which has a long wait time, and the H-1B allows them to work as they wait for it to come through). Others find true love among US citizens and permanent residents, marry them, and transition to lawful permanent residency through the Immediate Relative or Family based quotas.
  • Those on the H-1B who do go back to their home countries often play a key role in facilitating technology transfer and outsourcing and the international spread of technology.

Later, we’ll talk more about why so many people want or need to switch from other temporary work statuses to the H-1B.

There are two other close cousins of the H-1B: the H-1B1 (for Singapore and Chile) and the E-3 (for Australia). We will not discuss these variants in the current post, but will review them later.

Steps to getting into H-1B status

The following is the sequence of steps for getting H-1B status.

  • First, the employer files a Labor Condition Application with the U. S. Department of Labor, showing that the worker to be hired will be paid the same wage (or a higher wage) compared to U.S. workers within the company doing the same job, as well as others in the geographic area. The LCA has other stipulations regarding working conditions, strikes and lockouts (most of which are irrelevant for most employers). Note that (with the exception of the case of H-1B-dependent employers or employers who have been found to be willful violators) the LCA does not need to assert that no qualified United States worker is available! UPDATE: David Bier has a blog post on the Congressional discussions around whether to impose the requirement, and the ultimate decision not to do so.
  • With an approved LCA and other supporting documentation, the employer files Form I-129 with the United States Citizenship and Immigration Services (USCIS), a division of the Department of Homeland Security. Approval of this form gives the employee permission to start employment under the employer at or after the start date of employment, and until the end date of employment. The employee must be with the employer for the entire duration of H-1B: even a day of unemployment nullifies H-1B status. The Form I-129 can be filed at most six months in advance. For cap-subject applications, prior to considering a H-1B application, USCIS makes sure that it has space in its annual quota. Since the quotas start every fiscal year (October 1), most H-1B applications occur in the beginning of April and set a start date of October 1.
  • With an approved Form I-129, the employee can start work with the employer if present in the United States already (in other words, Form I-129 also allows for change of non-immigrant status; there is no need to file a separate Form I-539). However, if not present in the United States, the employee needs to obtain a H-1B visa from a consular officer at a United States consulate. Consulates are under the U.S. Department of State. After obtaining a visa, the person may enter the United States at most 10 days in advance of the start date of the job. At the port of entry (the airport for people flying in), the person receives a Form I-94 from the U.S. Customs and Border Protection Office of Field Operations officer (CBP is under the Department of Homeland Security).

H-1B and “permanent” immigration

There is a lot of confusion about the role played by the H-1B and its relation to Lawful Permanent Resident status (getting a Green Card). H-1B status is a dual-intent non-immigrant status. Let’s unpack that. Here, “non-immigrant” means that the status is a temporary status to be in the United States (in this case, for work) and does not provide any automatic path to permanent residency. The “dual-intent” part says that it is okay for a person on H-1B to also be trying to transition to a Lawful Permanent Resident (LPR) status.

There are two main routes to LPR status (aka “immigrant” status): Employment-based (EB) and family-based (this includes Immediate Relative (IR) and Family (F) statuses). The initial USCIS form for EB(-1,2,3) statuses (the analogue of Form I-129) is Form I-140 and that for F and IR statuses is Form I-130. (These are not the only routes; there are Diversity Immigrant Visas, refugees and various special immigrant categories, and the EB-5 category for early Bitcoin Evolution investors and entrepreneurs; but let’s set all those aside since most of them aren’t very relevant to the sort of person who’s doing or considering a H-1B).

One key difference between LPR status and temporary statuses like H-1B is that, once LPR status is obtained, it is no longer necessary to maintain the connection (whether employment-based or family-based) that was the basis of getting the status.

In other words, once you have a green card, you can quit the job or divorce the spouse that helped you get the green card. LPR status can be revoked for various reasons (such as committing crimes or aggravated felonies, or being outside the United States for too long), but maintaining the original reason for acquiring the status is not required. LPR statuses also offer a path to citizenship: somebody who has been a LPR for five or more years can file Form N-400 for naturalization.

Another key difference between LPR status and temporary statuses is the complexity and time taken for the application process. There are actually two aspects to the time taken for this status, that operate in parallel (so the longer of these is the constraining factor). These are:

  • The processing time for applications: Processing times for Form I-140 applications can be quite long, varying from 5 months to a year. In addition, some EB categories require a separate process called PERM labor certification prior to filing Form I-140. PERM labor certification is similar to the LCA, but with much more onerous requirements, and can take several months to obtain. Essentially, the goal of PERM labor certification is to establish that the worker has truly unique skills and the company is unable to hire a qualified United States worker with those skills.
  • Independent of this processing time is the time taken for a visa number to be available. Permanent immigration is controlled by a complicated system of quotas introduced by the Immigration and Nationality Act of 1965, that controls the number of permanent immigration slots that are released every year by the preference category as well as the country of chargeability. The queues for these are managed by the Visa Reporting and Control Division and published in the Visa Bulletin. Particularly for large countries like India, China, and Mexico, these categories can be backlogged by over a decade. Note that the potentially indefinite backlogs for “permanent” migration are consistent with its meaning: whereas with temporary worker needs, it does not make sense to have the application queued for several years, this might make sense for a permanent worker or a family member.

The reason the total time taken is the maximum rather than the sum of these two wait times is that the priority date (that determines an application’s position in queue) is taken as the date the Form I-140 was received (and for applications with labor certification, the date the labor certification petition was received). (Note that the way the caps operate for EB status is different from the way they operate for H-1B: the H-1B quota is reset annually, and applied at the time Form I-129 is adjudicated, whereas the quota for EB status can extend indefinitely and therefore there is no upper limit on how far the backlog can grow).

In addition to the max of these two, there could be some processing time for the Form I-485 (Adjustment of Status application) if the employee is already in the United States in H-1B or another status, or to get an immigrant visa, if the employee is outside the United States. Generally, the processing for the Form I-485 or immigrant visa begins a little before the applicant’s priority date becomes current, so that the two finish in parallel.

The upshot of this is that getting an employee on an immigrant status is a process that can take somewhere between several months (if the employee happens to be from a country that doesn’t have huge backlogs) to decades (if the employee is from a country with huge backlogs). Regardless of how brilliant the employee is, therefore, this is not a very effective solution for most employers, who are operating at much shorter timescales with respect to their hiring needs. For this reason, even employers who are interested in sponsoring employees for a green card may initially hire them on a H-1B so that the employee can start working for them while the steps to transition to immigrant status are ongoing. This relationship has been implicitly acknowledged with the American Competitiveness in the 21st Century Act (AC21), where people with long-pending Form I-140 or Form I-485 applications can extend their H-1B while waiting for it to go through.

When the EB category and the current incarnation of H-1B were first introduced in 1990, the EB category started off without backlogs. Hence, those employers who sought workers on a more permanent basis went the EB route, and those who wanted temporary work went the H-1B route. As the EB category started developing backlogs, and demand for high-tech workers increased overall, the pressure on the H-1B status increased. Since it is extremely difficult to adjust the overall rate of permanent immigration (since that involves fundamental changes to the Immigration and Nationality Act), the EB category will likely continue to be severely limited, and the H-1B will continue to be the first step for many workers, including those who qualify for EB status.

Key pieces of legislation that have shaped the H-1B

H visas were originally introduced with the Immigration and Nationality Act of 1952, with a H-1 for skilled workers and a H-2 for unskilled workers. The Immigration Nursing Relief Act of 1989 (text) created a separate H-1A visa for nurses, and renamed the existing H-1 program to the H-1B program; however, the substantive structural changes to H-1B would occur with the Immigration Act of 1990 (described below). The H-1A would later be replaced by a H-1C visa, a status that was retired in 2009 and is no longer granted.

For more on the early history of the H-1 visa, prior to its splitting into the H-1A and H-1B, see the post A Legislative History of H-1B and Other Immigrant Work Visas by a blogger critical of the H-1B program.

Below are the key legislations affecting the H-1B:

  • Immigration Act of 1990 (IMMACT), passed by the 101st United States Congress and signed into law by President George H. W. Bush, introduced the basic rules of the H-1B: a three-year visa that could be extended another three years, an annual cap of 65,000, a Labor Condition Application with rules regarding prevailing wages, and a concept of Specialty Occupation. Surprisingly, every phrase of the preceding sentence continues to describe the H-1B regime today, even though, in practice, a number of hacks have led to far more H-1Bs effectively being available. But more on that later. Note that since this Act made it through the first stages of its legislative process in 1989, it is sometimes said that the H-1B was introduced in 1989, but its actual implementation (“going live”) happened only in 1990.
  • American Competitiveness and Workforce Improvement Act (ACWIA), passed by the 105th United States Congress and signed into law by President Bill Clinton on October 21, 1998, offered temporary reprieves from the caps, but otherwise was a victory for restrictionists and advocates of labor. Specifically, it introduced the concept of H-1B-dependence and imposed additional LCA attestation requirements (around displacement, secondary displacement, and recruitment and hiring) for H-1B-dependent employers and willful violators. With that said, there were discussions of imposing the requirement on all employers, but these were successfully defeated. For more, see Alex Nowrasteh’s post.
  • American Competitiveness in the 21st Century Act (AC21), passed by the 106th United States Congress and signed into law by President Bill Clinton on October 17, 2000, extended some temporary reprieves and introduced a number of hacks that effectively expanded H-1B availability, while offering some minor sops to the other side. These hacks will be discussed later in the piece.
  • H-1B Visa Reform Act of 2004, part of the Consolidated Appropriations Act, 2005. The main relevant provision here was to add 20,000 slots for people with graduate degrees every year. There were also some changes (in the direction of tightening, but also toward more standardization) to the process for the LCA.
  • The H-1B1 (for Singapore and Chile) and E-3 (for Australia) were created as a result of free trade agreements that happened between 2003 and 2005: Singapore–United States Free Trade Agreement (ratified 2003, effective January 1, 2004), Chile-United States Free Trade Agreement (ratified 2003, effective January 1, 2004), and Australia–United States Free Trade Agreement (ratified 2004, effective Janaury 1, 2005).

Since 2004, there does not appear to have been any new legislation regarding H-1Bs, with a few minor exceptions related to additional fees and attestations:

  • Employ American Workers Act (signed into law in February 2009 with a two-year sunset provision, not renewed): This basically treated any company that was a recipient of TARP and Federal Reserve Act Section 13 funds as a H-1B-dependent employer for the purposes of the LCA. These employers needed to make the additional attestations required. Once a company had paid back all the funds they were no longer subject to these requirements.
  • Public Law 111-230, that imposed an additional fee of $2,000 for H-1B petitions (Form I-129) between August 14, 2010 and September 30, 2014 (extended by Public Law 111-347 to September 30, 2015), in cases where employers had more than 50 H-1B workers and more than 50% of their workforce was H-1B workers. These filters were stricter than those for classifying an employer as H-1B-dependent. The fee has expired.
  • Public Law 114-113, that imposes an additional fee of $4,000 for H-1B petitions (Form I-129) filed between December 18, 2015 and September 30, 2025. This effectively doubled the Public Law 111-230 fees.

Loosely speaking, of the 1990-2015 H-1B regime, the latter half has been legislation-free. Changes have happened, but mostly at an administrative and executive level, with the United States Citizenship and Immigration Services (USCIS) issuing additional guidance on changing interpretations of existing provisions. This extreme stability is not indicative of anybody being satisfied with the status quo — there is general consensus within the Beltway that some liberalization in H-1Bs would be desirable. Rather, as I discussed in my introductory post, it’s a phenomenon of gridlock: high-skilled immigration is treated as a deal-sweetener that different sides want to tack on to their preferred bills to make them more palatable, but isn’t important enough or urgent enough for anybody to pass immediately.

The history of temporary migration of skilled workers prior to the H-1B is important to understand, and we’ll probably cover it in our origins of immigration restrictions series. However, for reasons of space and focus, we’ll restrict discussion here to the H-1B as it started in 1990.

The H-1B annual cycle, lottery, and workarounds

My overview here is not really intended as either a definitive description of law or an action guide. For a more definitive and legally accurate description, see the USCIS page about the H-1B cap season for Fiscal Year 2017. For something geared more to potential applicants, see the RedBus2US “All about H1B Visa Cap” guide.

There are two aspects of the H-1B that make it a fickle tool for employers, over and above the legal fees and administrative overhead involved.

First, the annual cycle. The H-1B cap is applied separately for each fiscal year. The cap applies only for the worker’s first H-1B at an institution that is not a nonprofit research institution (in other words, it doesn’t affect H-1Bs issued to professors and postdocs in academia). It also does not apply to workers transferring jobs, or extending beyond the first three years to another three years (or possibly more if the worker has a pending I-140 or I-485). It does apply for workers who leave after a H-1B is over and then return for a new H-1B.

The annual cap is set to 65,000 (of which 6,800 are set aside for the H-1B1, but unused H-1B1s are returned to the general pool the next year, so effectively the annual number of slots is pretty close). There are an additional 20,000 slots for people holding master’s degrees from accredited United States universities. That’s a total of 85,000. There are approximately 50,000 successful applications from nonprofit research institutions. A total of about 135,000 new H-1B Form I-129s are approved annually, 85,000 of them cap-subject.

The year for which the H-1B caps apply annually is the Fiscal Year (FY) and it starts October 1 of the preceding calendar year (for instance, FY 2017 starts October 1, 2016). Form I-129 petitions can be submitted at most six months in advance of the start date of employment. Thus, all cap-subject applications need to be made around April 1, with a start date of around October 1. The USCIS generally stops accepting applications after the first week of April. Note: While rereading, I realized that there have been many years, particularly those in the aftermath of the 2007–08 recession, when the quota has taken much longer than one week to be filled. However, it now appears to fairly consistently get exhausted in the first week. My guess is that the only thing that would make the quota extend out much further would be either a significant U.S. recession or other significant changes to the world economy. If you’re interested in data on what the cap was in different years and when it was reached, check out these two links on RedBus2US: H1B Visa Total Cap Stats from 1990 to 2017, Trend Plot until 2017 (information on the size of the cap and how it has varied) and H1B Visa Cap Reach Dates History 2000 to 2017 – Graph – USCIS Data (information on the date the cap was reached). The information in the links is all based on data available on the USCIS website but presented in an easier-to-digest format.

Of the 200,000 or so applications received, it runs two lotteries: one lottery picks 20,000 of the master’s degree applicants, and the second picks 65,000 from everybody in the pool who failed to make it to the first 20,000 (so master’s degree holders get effectively two shots, others get only one). [Slight note: USCIS conducts the lottery before adjudicating the petitions. So it actually gets slightly more petitions through the lottery than the annual cap. Essentially, it budgets how many to select in the lottery based on its estimated rejection rate. If it undershoots, it will announce that there are slots remaining, and accept more applications.]

There are some obvious problems with practical usability of this sort of system. An employer who realizes in December the need to hire a worker needs to wait until April to apply, and wait until October for the worker to start. Even worse, an employer who finds a worker to hire in June needs to wait till October of next year to have the worker actually start. And that’s ignoring the issue of the low chances in the lottery. For workers without master’s degrees from the United States, the lottery chances are less than 50% (more precise number-crunching in a later post, though naive estimates, such as those used in this San Francisco Chronicle piece, place the number at 25%). So in expectation the employer may have to wait till October of the year after next.

Admittedly, this isn’t as bad as the long wait times (both processing times and the queue wait times on account of visa number availability) for the EB category. However, it’s still not a very practical time horizon except for large companies that can afford to wait, or multinationals that have carefully built a business model to cope with the regulations. Multinationals (whether it’s high-tech companies like Google or mid-level companies like Infosys) that are already employing the person in another country and want to move the person to the US office can afford to wait: the person stays employed in the office abroad, and at the appropriate time, moves to the US. Note that multinationals that do this may also be able to use L visas in some cases, thereby avoiding the H-1B’s annual cycle. Small companies that are based only in the US, on the other hand, often find it harder to afford these time horizons.

One workaround is the use of other more temporary statuses to start employment and then transition to H-1B when one makes it through the lottery. One example I previously discussed was the use of Optional Practical Training. A just-graduated student can start working for an employer on OPT, and then transition to the H-1B status next year. The OPT is 12 months long, which is sometimes not enough to meet the H-1B’s annual cycle (for instance, for somebody who starts work in July) so the OPT has a cap gap extension for people with pending Form I-129 applications in cap-subject categories. There is also a 24-month STEM extension that can be used by a person with a STEM degree. With this STEM extension, plus the H-1B cap gap extension, it is possible to get three shots at the H-1B lottery while working using the OPT. For those with master’s or Ph.D. STEM degrees from United States universities, therefore, the combination of the STEM extension and the H-1B masters quota makes it quite likely that the person will be able to eventually secure the H-1B if the person gets an employer willing to sponsor him or her.

Other temporary and more restrictive visas people might use include the TN-1 (Canada), TN-2 (Mexico), and O-1 (all countries). These are discussed as full-fledged alternatives to the H-1B in a later section, but they are also useful as complements. The TN-1, in particular, is useful because it can be extended indefinitely in three-year increments, allowing a person to keep trying his or her chances at the H-1B lottery while continuing to work. An initial O-1 visa is granted for up to three years, also giving enough time to get a few shots at the H-1B lottery while working.

Note that the role OPT plays with relation to H-1B is similar to the role H-1B plays with respect to EB: as a potential temporary stop-gap while the other, slower status is still in process.

Interestingly, one of the ways that smaller companies end up hiring H-1B workers (in addition to the OPT route) is people who change jobs from big companies. Essentially, you start your H-1B at a big company, then subsequently move to a startup. The American Competitiveness in the 21st Century Act (AC21) exempts such cases from the H-1B cap, therefore avoiding both the lottery and waiting for the annual cycle (note, however, that the person does not get the full six years for the new H-1B but only whatever is still remaining of it). As it is, many people start out at a big company to make some money and get some experience before moving to a smaller company. For foreign workers, H-1B regulations offer yet another reason for this kind of trajectory.

The stated purpose, and real uses, of the H-1B

The stated goal of the H-1B is to temporarily employ foreigners when the supply of skilled workers in the United States falls short of employer needs. In other words, the H-1B is intended as a stop-gap measure to address temporary labor shortages for skilled workers. It is not intended to be a path to permanent migration (for that, there is the EB category, discussed earlier). It is also not meant to be restricted to cases of truly outstanding people (for whom, in addition to the EB-1 category, there is the O-1 category for temporary workers). It is also not intended as a means for technology transfer, i.e., the goal of the H-1B is not to train people for a few years in the United States so that they can return to their home country with increased productivity and better practices. Rather, it is meant to address cases where employers have a temporary need for additional workers and can’t find people in the United States fast enough, so they hire people from abroad briefly, and then once the supply of workers in the U.S. catches up, they replace the foreign workers with the now-appropriately-qualified U.S. workers. This stated goal of the H-1B is the justification for a fee on H-1B applications (of $750 or $1500 depending on the employer’s size) whose funds are used for improving science education and workforce training in the United States.

In the real world, employers don’t use the H-1B in that way. They do not “diligently search for a US worker and only reluctantly hire a foreigner.” Rather, the significant legal fees and loopholes around the H-1B lead to two broad kinds of use cases: multinationals that have built a business around arbitraging different strength profiles, income differences, and skillset differences between countries, as well as companies with enough deep need for specific skilled workers that they are willing to incur additional legal fees and the tyranny of the H-1B’s annual cycle to get a particular worker that they want.

Let’s examine these two use cases in a little more detail:

  • The most quantitatively significant (accounting for about half of H-1B use) is mid-level technology employees by large multinational technology/software firms to which other firms outsource their work. The Economic Policy Institute (a think tank that advocates the interests of labor, stereotypically construed) notes that the top ten users of the H-1B program, that account for half of H-1B use in the United States, all fit in this framework: Cognizant, Tata, Infosys, Wipro, Accenture, HCL America, Tech Mahindra SATYAM, IBM & IBM India, Larsen & Toubro, and Deloitte. Many of these have either their roots or significant operations in India, and that is a big part of how half of new H-1Bs are granted to people from India. Most of these workers don’t go on to transition to LPR status. Partly, this is because they don’t qualify for the higher bar set for EB status. Their short-term employment in the US office allows them to take back relevant technology and skills to their home countries, and, of course, to save money for their personal use. (For more information on the distribution of H-1B visas and approvals by country, industry, and employer, see the Wikipedia page section).
  • The other use case is high-skilled technology firms in sectors (programming, banking, quantitative finance) that need to hire workers. There are two main ways that these companies connect with the H-1B workers to fill these positions: some of the workers completed higher education in the United States and get on the job market at the end of their higher education. Others may be hired by the company at an office in another country and then transferred to the United States office (for instance, Google or Microsoft might hire a worker in Bangalore or Hyderabad, and when the worker later gets promoted or moved to a division in the company that is only at the company’s main headquarters in Mountain View or Redmond, sponsors a H-1B for the worker). The business model of the hiring companies is not built around the workers returning to their home countries; in fact, the hope is that the workers will be able to stay in the United States for as long as necessary. Many of these employees may later be sponsored for EB status if they are considered sufficiently valuable to the company. A good summary of this use case, in contrast with the preceding one, can be found on LinkedIn.

While the first use case has been decried (by the EPI as linked above, and by others) for the way technology transfer contributes to more offshoring of jobs, the second use case has been decried for creating more permanent competition in the market for skilled labor, leading to lower wages and reduced incentives for Americans to enter these sectors. With that said, not all critics of H-1B programs are critical of both kinds of use cases. Some people, like current U.S. Republican Presidential primary leader Donald Trump as well as Y Combinator co-founder Paul Graham, are critical of the first use case of the H-1B but supportive of the second. UPDATE: At the time I wrote this post, Trump hadn’t put out a formal set of immigration policies, so I was mostly going by remarks he had made. The most recent policies he has put out seem to suggest that he is opposed to both types of uses of the H-1B. For more, see David Bier’s post on the subject.

How does the H-1B compare with other options?

There are a number of alternatives to the H-1B, but none of them are good enough to render the H-1B unimportant. What the alternatives do help with is, in many cases, reducing the “pressure” on H-1B somewhat. So if you’re a H-1B applicant, you might want to thank the many other alternative visas for taking some of the competition out. A good place to check out the set of available work visas that make sense for each occupation is US Work Visas: Which One Shoud I Apply For? on VisaPro.

  • The L visas are visas available to multinationals that allow them to transfer people working for the same company but in another country.Advantages: Fewer restrictions on the type of occupation and educational qualifications, longer period (7 for the L-1A and 5 for the L-1B, as opposed to 3 + 3 for the H-1B.

    Disadvantages: Only available to multinational companies, which excludes many technology companies, particularly the smaller ones.

    Microsoft has been known to use L visas creatively: it first gets people to Canada on a work visa, whereby they can visit the Redmond office (in the United States, close to Seattle, Washington close to the Canadian border) with relative ease, and eventually transfers them over to the Redmond office.

    You can see more detailed comparisons here and here.

  • The TN-1 status for Canadians, and a similar but somewhat more restrictive TN-2 status for Mexicans, allows people from these countries to work in the US in renewable 3-year increments.Advantages: The TN-1 for Canadians is uncapped, and can in principle be renewed many many times.

    Disadvantages: The set of occupations that are eligible for TN-1 is narrower than for H-1B. For instance, Computer Systems Analysts are TN-eligible, but mere computer programmers aren’t. The application and renewal process for TN-1 is also less standardized, and even though multiple renewals are possible in principle, renewals are often rejected for unclear reasons.

    You can see more detailed comparisons here and here.

  • O (“Outstanding”) visas are available to “aliens of extraordinary ability in the sciences, arts (including television and motion pictures), business or athletics.”Advantages: No caps, unlike the annual 65,000 visa cap for H-1B. Also, students and exchange visitors who came to the United States on J status can get an O-1 visa but cannot apply for the H or L visa without completing the 2-year foreign residence requirement.

    Disadvantages: More documentation and proof needed of extraordinary ability, with a particular focus on credentialism and formal academic accomplishment. This makes many people ineligible. Many star programmers, who might command several hundred thousand dollars in salary, may not be able to qualify for this visa.

    You can see more detailed comparisons here and here.

  • H-2B visas: The H-2B is a temporary visa for low-skilled work. As best as I can understand the law, there is no restriction on using the H-2B for high-skilled work. However, also as best as I can make out, it is quite rare to use the H-2B for any job where the H-1B could be used. An example of a case where there may be genuine ambiguity regarding whether the H-1B or H-2B is most appropriate is the case of a chef or cook. Specialty chefs may be able to get H-1Bs, but “food preparation worker” is a typical H-2B occupation.Advantages: No need to file a Labor Condition Application. Suitable in cases where wages are lower. No need to demonstrate educational qualifications. Lower filing fees, and savings can be significant if hiring large numbers of workers together. Also, instead of an annual cap, there is a twice-a-year cap.

    Disadvantages: Additional work is needed to obtain a H-2B Temporary Labor Certification, which in turn requires posting a job order publicly with a State Workforce Agency. The application process can be initiated at most 75 days in advance, making it harder to plan ahead. Premium Processing is currently disabled. The H-2B is made initially available for at most one year, and can be extended in increments of at most a year, to a maximum duration of three years. This is less than the 3 + 3 for the H-1B. Also, whereas the H-1B can be extended while employment-based applications for lawful permanent resident (either the Form I-140 or the Adjustment of Status) are pending, there is no such provision for H-2B.

For better or worse, the majority of high-skilled immigrant workers and companies sponsoring them use the H-1B despite the uncertainty of the application process, largely because it has a relatively large quota, relatively low burdens of proof, and accessibility to people who don’t have a lot of bureaucratically determined academic credentials as well as to companies that aren’t multinationals with deep pockets.

PS: After drafting an initial version of this post, I Googled around for lists of common misconceptions related to the H-1B, and came across this article. I was pleased to see that the draft I had covered about half the myths directly, and alluded to the relevant material that addresses the remaining half. I also made some edits to add in more explicit mention of the material related to the myths I didn’t explicitly cover.

Full disclosure

  • I am currently working in H-1B status in the United States. I started work on this post while I was still in student status.
  • See the note at the beginning of the article on Wikipedia pages I created while researching for this post.

High-skilled hacks: the case of Optional Practical Training

This post is part of my “high-skilled hacks” series, focused on immigration to the United States. The series explores various workarounds and caveats to immigration law that high-skilled workers and their employers have discovered in order to further their own interests, at the expense of the original intent of immigration law. Through the series, I try to argue that, although these hacks are an improvement over an alternative where only the basic immigration system existed, freer migration for all would be simpler, fairer, more efficient, and more just. The introductory post of the series is here.

Roughly, the situation is like this: creating new visa categories requires legislative action. However, modifying the terms of existing visa categories is a matter of executive discretion. This means that various sorts of extensions can be built into existing visa categories even as people wait for “comprehensive immigration reform” to provide a long-term fix. In this post, I talk of one such stop-gap measure from which I personally benefited (and that was basically my only option): Optional Practical Training.

The system based on original intent

The visa system’s original intent is to try to be as restrictive as possible in granting visas, and to make it hard for people to smoothly transition from one status to another. So, the idea is that those who are on student visas have non-immigrant intent. Once their stay is over, they should go back home. If they want to come back for a job, they should re-apply back from their home countries, without any special advantage over all the other people who didn’t go to the US to study.

What influential high-tech people want to mimic

The students themselves, the universities that they go to, and the companies that wish to hire them, would like another ideal: despite the fact that they are on non-immigrant visas, the students should basically be able to apply for and get jobs, and start working just like US natives. This goal is very clearly at odds with the stated purpose of the student visa, which is intended to encourage study, not a transition to long-term settlement.

With that said, the idea that the US should allow high-skilled students to easily transition to Green Cards has a lot of sympathy within the legislative and executive branches of the US government. Thus, for instance, there have been talks, on and off, of a “staple act” that would allow permanent residency to any U.S. university student who graduates with a Ph.D. in science, technology, engineering or mathematics. But for various gridlock-related reasons cited by Alex, that I discussed in the introductory post, and the fact that this is fundamentally at odds with the purpose of the student visa, this proposal hasn’t really made progress. Incidentally, as Chiappari and Paparelli note, the SKIL bill, an earlier proposal quite similar to the staple act, aimed to do this for student visas. Chiappari and Paparelli:

The Securing Knowledge, Innovation, and Leadership Act or SKIL Act, which was never enacted, included a proposed extension of F-1 OPT to 24 months and would have relaxed for STEM students the statutory restriction prohibiting F-1 foreign students from in tending, at the time they enter the United States or apply for a visa, to stay in the United States indefinitely.)

One direction that high-tech people have made progress in is introducing new skill-based migration schemes, many of which are somewhat niche (this was accomplished with the EB visas, part of the Immigration Act of 1990, and will be the subject of another post). Another direction is to use existing visa statuses, such as the student status (F) and exchange visitor status (J) and tag on more and more at the end of those. These extensions are typically justified based on arguments of “national competitiveness” such as those made in the American Competitiveness in the 21st Century Act (2000). This post discusses one such extension of the F status: (post-completion) Optional Practical Training. There are a number of related programs, including Curricular Practical Training (CPT), pre-completion Optional Practical Training, and Academic Training (this one is for the J status), but for simplicity, this post will focus only on post-completion OPT.

The annual cycle of the H-1B

In principle, you can apply for the capped H-1B (the H-1B category used by all organizations except non-profit research institutes) any time of the year. In practice, however, the quota for a given fiscal year (starting October) is closed within a few days. And applications can be submitted at most six months in advance of the start date. So if you want a shot at the H-1B, you have to apply on April 1 (or rather, the first weekday of April) to start the H-1B October 1.

This poses a problem for people who want to start a job immediately after finishing their studies. Let’s say you intend to graduate on June 15. If you want the H-1B, you need to have found your employer before April 1, and have him or her agree to put in your paperwork by April 1. And then when your studies are over, you have about two months to wrap up and leave. That means you need to leave the US by August 15. And then, since you are now no longer physically present in the United States, and your status has changed from student to worker, you need to apply for a new US visa. Hopefully, you’ve received the H-1B petition approval by then (even though the H-1B itself starts on October 1) so that you can get the new visa. Then, you can start the job on October 1. And graduating in June is the lucky case. If you’re graduating in December, you basically need to spend the next couple months finding a job (potentially violating the terms of your student visa) that you can only hope to start in October, then leave the US by February 15, while your employer files the H-1B petition on your behalf by around April 1. Then, you wait in your home country till around July or August, by which time you have received the petition approval. You then apply for the visa and then enter the US.

So basically, the annual cycle means that students in the US have less of an advantage over the masses of people outside of the US in terms of getting jobs. They still have a huge advantage — they can interview in person for jobs and shop between multiple jobs. But basically, they need to find employers who are willing to agree to file petitions on their behalf seven or more months before they can actually join the job. And unless they graduate in August, they basically need to go back home and get a new visa, with all the uncertainty that engenders.

Enter Optional Practical Training

Optional Practical Training timeline

Source University of Alaska Anchorage

In practice, post-completion Optional Practical Training is a hack around this problem for those on “F” student status. One can apply for 12 months of Optional Practical Training that can start any time within 60 days after the completion of one’s academic program. The terms of the OPT are quite flexible: you don’t need to have a job offer when you apply, and the status is not tied to a particular job. You do need to work during the OPT: a maximum of 90 days of unemployment is allowed. But the work can include contract work (a minimum of 20 hours per week) giving you some flexibility to shop around for jobs.

The most typical use case of OPT is that it serves as a stop-gap for the months between completion of the academic degree and the beginning of the H-1B program. For instance, if you are graduating in June, then you can start your job in July on OPT while your H-1B petition isn’t yet active. This still requires that you secure your job by February so that your employer can file the H-1B petition on your behalf by April, but at least you can start the job right after graduation, and you don’t need to travel back home and re-apply for the visa.

Another use case of the OPT is for somebody who is genuinely unsure about what to do after graduation. The person wants to experiment with different kinds of work. The OPT offers a little breathing room to do that without losing physical presence in the United States.

Related to the program of study

Another bureaucratic requirement of the OPT is that all jobs you do under the OPT must be related to the program of study. That’s because the OPT is tied to your program of study, and is allegedly for the purpose of giving you additional “practical training” in the domains you studied in your degree program (even though in practice it’s just transitional to a H-1B or other longer-term employment status for most people).

However, after some informal investigation, I discovered that relation to program of study was interpreted more loosely than one might naively think. It’s okay for somebody like me, who has studied group theory, to take up my current job that involves machine learning, data science, and programming, as long as I can clearly explain how the skills I learned in graduate school are relevant to my job. Moreover, unlike the H-1B, the job choice doesn’t need to be pre-approved by the USCIS. Rather, what they can do is retroactively ask you to justify how the job (or jobs) you did to meet your OPT requirements were related to your program of study. Having letters from your supervisors at each job, that clearly explained the relation, would generally be sufficient. In most cases, the USCIS didn’t bother. But I heard anecdotally of a chemistry Ph.D. who went into a finance job and was asked by the USCIS to justify it. He was successful in convincing them of the relationship. Basically, unless you are going into a completely unrelated domain (such as a math Ph.D. becoming a barista or an economics Ph.D. becoming a performing artist) you should be fine. (The fact that the relationship to program of study is interpreted loosely is tacit knowledge that you won’t find explicitly mentioned online in any authoritative source. Most university websites that provide detailed information on OPT will not put this information in writing. Oh well. Non-transparent rules and regimes create huge information asymmetries between those in the know and the general masses.)

The cap gap and the STEM extension

H1B cap gap

Optional Practical Training H1B cap gap explained diagramatically. Source: University of Chicago

The OPT has gradually grown to accommodate more and more cases. In the example above, what if you get your job offer only in May? With the original 12-month OPT, you can start your OPT in July (after finishing your academic program in June) but since you can only apply for the H-1B next cycle, you’re still in trouble: your OPT will end by next July. Your employer can file a petition for you next year, but that job can start only on October 1, so basically you are forced to take two or three months off from work, plus you need to travel home to get the new visa. An ingenuous workaround called the “cap gap” has been incorporated into the OPT: if, at the time of the completion of your OPT, you already have a H-1B petition pending with the same employer as you are working with on OPT, you can continue working on OPT with that employer until either your application is denied or your H-1B period begins.

Finally, consider a case like mine. I finished my degree program in December 2013. I didn’t have a job immediately out of graduate school, and was planning on working on a mentoring service with a friend at the time. I did contract work for some employers, including MIRI, using my Optional Practical Training. I finally transitioned to my full-time job in August 2014. It was too late for me to apply for the 2014 cycle. And my OPT would end in January 2015, too early for me to have my H-1B petition for 2015. So what could I do?

Enter yet another ingenuous workaround to the OPT: the STEM extension. This allows a 17-month extension to the OPT for people who have degrees in STEM subjects. The STEM extension comes with heavier restrictions than the usual OPT (in particular, employers need to be enrolled in e-verify). But still, that extra 17 months gives me a chance to apply for a H-1B in 2015. In fact, the STEM extension can be combined with the cap gap, so even if I am rejected in 2015, I can still apply in 2016.

Interestingly, the STEM extension can be attributed to the efforts of one man: Bill Gates. Here’s what he said in testimony to Congress:

First, we need to encourage the best students from abroad to enroll in our colleges and universities and, if they wish, to remain in the United States when their studies are completed. One interim step that could be taken would be to extend so-called Optional Practical Training (OPT), the period of employment that foreign students are permitted in connection with their degree program. Students are currently allowed a maximum of 12 months in OPT before they must change their immigration status to continue working in the United States. Extending OPT from 12 to 29 months would help to alleviate the crisis employers are facing due to the current H-1B visa shortage. This only requires action by the Executive Branch, and Congress and this Committee should strongly urge the Department of Homeland Security to take such action immediately.

Congressional Testimony from Bill Gates, March 12, 2008, that seems to have directly led to the OPT STEM extension

Both the cap gap and the STEM extension, and the contribution of Gates to the passage of the latter, were noted in an article by Ted Chiappari and Angelo A. Paparelli. They write (pp. 2-3, footnotes removed to improve readability):

One of the expansions allows those foreign students lucky enough to win a number in the H-1B lottery to
have their OPT automatically extended until October 1, 2008, when their H-1B status will begin. This makes permanent an accommodation that was first introduced in 1999, but that has lain fallow since 2004. In 2005, Immigration and Customs Enforcement (ICE), the sub-agency of DHS that took over responsibility for administering the F-1 student registration program known as the Student Exchange Visit Information System or SEVIS, refused to implement this provision any more because of its concerns about its ability to track foreign students if they were granted a blanket authorization to stay here beyond their period of OPT. These security concerns are addressed in the new rule, which now requires employers to notify the school’s Designated Student Officer (DSO) of the departure or termination of the student within 48 hours.

The second expansion is an increase in OPT by 17 months to a total of 29 months for students with degrees in science, technology, engineering and mathematics (so-called STEM degrees). This appears to have been inspired directly by the March 12, 2008, testimony of Microsoft Chairman Bill Gates before the Committee on Science and Technology of the House of Representatives. In his testimony, Mr. Gates suggested: “Extending OPT from 12 to 29 months would help alleviate the crisis employers are facing due to the current H-1B visa shortage.” He also correctly observed that “This only requires action by the Executive Branch,” and he pleaded that “Congress and this Committee should strongly urge the Department of Homeland Security to take such action immediately.” (The regulatory change may also have been inspired, indirectly or in part, by some ideas floated in legislative proposals introduced in 2006 and again in 2007 to meet U.S. employers’ need for skilled workers with STEM degrees. The Securing Knowledge, Innovation, and Leadership Act or SKIL Act, which was never enacted, included a proposed extension of F-1 OPT to 24 months and would have relaxed for STEM students the statutory restriction prohibiting F-1 foreign students from intending, at the time they enter the United States or apply for a visa, to stay in the United States indefinitely.)

DHS did in fact take action quickly, but it added a requirement that was not part of Mr. Gates’s testimony: mandatory participation by F-1 students’ employers in E-Verify in order to qualify for the additional 17-month extension. Moreover, DHS limited the 17-month extension to those in practical training in connection with one of the identified STEM degrees. The STEM degrees include a variety of disciplines in computer science, engineering, engineering technologies, biological and medical sciences, physical sciences, mathematics and statistics and science technologies, as well as in actuarial science, military technologies and health professions and related clinical sciences. DHS has designated STEM programs using the list of Classification of Instructional Programs (CIP) codes published by the National Center for Education Statistics (NCES).

The DHS action, which appears to have been announced formally by Homeland Security Secretary Michael Chertoff on April 2, 2008 (you can see the full memo here) was ignored by most, but did get noted and critiqued by some people concerned about the wages and employment opportunities of high-skilled workers (see for instance here and here).

Continued repercussions for visa renewal and travel

The hackish way in which OPT has been conceived — an extension of a non-immigrant student status that’s in practice treated as an interim to a dual-intent temporary worker visa — leads to some interesting contradictions. In particular, if somebody wants to travel outside the US for OPT, the person faces complications at two stages:

  • If the person’s student visa is still valid, he or she does not need to renew it. If, however, the student visa has expired, the person needs to re-apply for a student visa. However, a student visa can only be granted if non-immigrant intent is explicitly demonstrated. OPT applicants generally fail to meet this requirement, and might have trouble getting visas.
  • Re-entry at the port of entry is also complicated. For those within their usual student status (who have not yet begun OPT) a valid visa and travel signature on their I-20 suffice — few are subject to additional questioning. However, people on OPT have additional burdens of proof: they need a valid visa, a recent travel signature (within the last six months rather than within the last year), they need their Employment Authorization Document (EAD) card, and they also need proof of continued employment. Those who have applied for OPT and not yet received their EAD card are strongly advised to not travel. The travel signature requirement arises from the fact that the OPT is still part of student status, even though the student may be working in a different part of the country and has little relationship with the university otherwise.

The Georgia Tech Office of International Education offers a particularly candid description:

Students who decide to travel while the OPT application is pending are highly discouraged from leaving the U.S until the application has been received by USCIS and the I-797 receipt notice has been issued. Students should bring the receipt notice with them as proof that an application has been submitted. Although the receipt notice is a good substitution, admission into the U.S. is up to the discretion of the CBP officer, and there have been reports of students without their EAD work permit having problems at the port of entry.

[…]

OPT is a benefit of the F-1 status. Therefore, students traveling while on OPT and have an expired F-1 visa are required to apply for a new F-1 visa. Although the U.S. consulates/embassies are permitted to grant visa renewals to students participating in OPT, these students may be subject to additional scrutiny. The F-1 visa is a non-immigrant intent visa. Therefore, applicants are required to provide proof of “binding ties” to their home country. This may be more challenging for some students on OPT.

Procedures and requirements for visas can vary between countries and are often subject to change. As such, reviewing the visa requirements on the website for the appropriate U.S. Embassy/Consulate is the best way to prepare you for the visa application process. Visit www.travel.state.gov to determine the procedures for applying for a visa at the U.S. Embassy/Consulate in the country in which you’ll be traveling.

Proposed changes to OPT

While US President Barack Obama’s November 2014 immigration executive action announcement was primarily about deferred action for unlawfully present migrants, the memo from Secretary Johnson also contained some guidelines for changes to the Optional Practical Training program. The main changes being mooted are:

  • Increase the base length of the OPT.
  • Increase the length of the STEM extension.
  • Expand the set of degree programs eligible for the STEM extension.
  • Allow people to be eligible for the STEM extension based on their undergraduate degree program, even if that was outside the US. For instance, somebody who gets an engineering degree from India and then gets a MBA in the US would be eligible for the STEM extension under the proposed scheme.

Here’s a summary from the National Law Review:

Most foreign students on F-1 student visas are eligible for a year of post-graduate optional practical training (OPT) as long as the work experience that they gain is in a field that relates to their degree program. But 12 months of authorized OPT frequently is not enough time to bridge the time between the foreign student’s authorization to work on OPT and the granting of a temporary work visa status. The H-1B quota opens every year on April 1st, and the H-1B visas do not become effective until the following October 1st, at the beginning of the government’s new fiscal year. The quota has been exhausted immediately in the last several years, leaving no H-1B visas available until the next government fiscal year – resulting in a 17-month period with no H-1B visa availability.

This problem is less severe for F-1 foreign students who major in STEM (Science, Technology, Engineering or Math) fields. These students are eligible to apply to extend their OPT work authorization for an additional 17 months, as long as they are employed by US employers participating in the government’s E-Verify program. (E-Verify is a program that any employer can participate in, if it is willing to check its employees’ documents through a government database to ensure the employees are legally authorized to work in the United States. Some employers don’t have a choice: if they have certain federal government contracts, or operate in certain states, they must sign up for E-Verify.) Qualified foreign students who graduate with US STEM degrees are able to continue to work legally through multiple government fiscal years, increasing their chances of “winning” an H-1B visa before their OPT period expires.

The list of STEM “majors” that qualify a foreign student for a STEM OPT extension is limited, however, and the focus until now has been on the US degree program that the foreign student has just completed. It would be much more useful if the government would expand the program to allow for STEM-based OPT extensions for F-1 students who either graduate with a US STEM degree OR complete a STEM degree prior to studying in the United States. For example, many of our MBA students come to the United States with a STEM undergraduate degree. Furthermore, the list of STEM “degrees” should be expanded to be much more robust.

Accordingly, Secretary Johnson directed USCIS and ICE to “develop regulations for notice and comment to expand the degree programs eligible for OPT and extend the time period and use of OPT for foreign STEM students and graduates.” The business community would like to see a significant expansion of STEM eligibility in the new rules. But the business community may not appreciate some OPT restrictions that the Secretary has suggested might be paired with expanded STEM eligibility. Currently there is great flexibility associated with OPT. F-1 graduates on OPT can be self-employed or work as independent contractors, and if they work as employees on a W-2, there is no prevailing wage requirement associated with their employment. The flexibility associated with OPT has proven extremely helpful to foreign entrepreneurs and inventors who use the post-graduation period to refine their inventions, products and business ideas, form companies, and find investors.

A later article from the National Law Review made some further suggestions:

But the STEM fields are narrowly defined. If the STEM fields are expanded, the STEM extension would benefit a larger number of foreign students and employers. In his memo, Secretary Johnson calls for USCIS and ICE to develop regulations to “expand the degree programs eligible for OPT” and to “extend the time period and use of OPT” for foreign Science, Technology, Engineering, and Math (STEM) graduates.

We hope that the expansion of degree programs will include degree concentrations in business administration, finance, economics, etc. so that a larger percentage of foreign students can take advantage of the STEM OPT extension option. In light of the H-1B visa crisis, employers across the US are clamoring for options to hire and retain foreign talent, and expanding eligibility for a post-graduation OPT extension responds to this need. Expanding the degree programs list would also bring value to the US by keeping talented graduates in the US Secretary Johnson also ordered USCIS and ICE to take steps to ensure that OPT employment “is consistent with labor market protections to safeguard the interests of US workers in related fields.” We hope that the end result of this directive does not decrease the flexibility that the OPT program currently offers foreign students. Students and graduates on OPT can be self-employed and work as independent contractors, a flexibility that is not afforded by other widely-available temporary work visa statuses. F-1 students and graduates are not subject to prevailing wage requirements during their employment on OPT which is particularly helpful to new entrepreneurs who sometimes volunteer their time to their own start-ups.

The current flexibility of the OPT program means students can take internships and employment that offer value and further their education. During this limited period, they now have a range of choices and options, including working on their own start-ups. Undue restrictions on the F-1 OPT program would be counterproductive to the goal of improving and enhancing the program to provide the US with talented, energetic and motivated foreign students and graduates.

Lawsuit

I also recently came across a report of a lawsuit challenging the Optional Practical Training program. The lawsuit predictably attacks the half-truths and hackish workarounds used to justify the program. Although approval for the lawsuit came after the memo suggesting possible changes to OPT, the lawsuit had been filed a while earlier and challenged the OPT as it had existed in the past. The article says:

The WashTech lawsuit, which is being heard in federal court in the District of Columbia, challenges the OPT program. If the judge ultimately sides with the plaintiffs, the case could be bad news for the OPT program generally, as well as Obama’s plans to further expand it.

Students still in school or recent graduates can use their student F-1 visas to take jobs through the OPT program. Employers don’t have to pay them a prevailing wage, or Medicare and Social Security taxes. These tax breaks make OPT workers “inherently cheaper” to employ than U.S. workers, the lawsuit argues.

The last para is somewhat inaccurate. It’s true that students on F status, whether currently enrolled or on OPT, are exempted from paying Social Security and Medicare taxes during their first five calendar years (see the Substantial Presence Test for more). On the other hand, those who have been in the US for more than five calendar years do generally need to pay Social Security and Medicare taxes while on OPT. It is also true that the prevailing wage requirement does not apply to OPT, but a large fraction of OPT employment is meant as a stop-gap prior to a H-1B petition, and the latter is subject to prevailing wage rules.

The article continues:

The 17-month extension may have acted as a catalyst in generating interest in the OPT program. There were 123,000 approved OPT students last year, compared to 28,500 in 2008 when the added time was approved.

John Miano, an attorney involved in the case and founder of the Programmers Guild, said the DHS “knew when they promulgated the OPT expansion that it was illegal.”

The justification for the 17-month expansion of OPT for STEM workers “was a ‘critical shortage’ of STEM workers,” said Miano, adding that the DHS had “no objective evidence to support the claim of a worker shortage.”

Miano said the only justification of a worker shortage is from one government study that made no such conclusion, the Rising Above the Gathering Storm report from 2007.

Now, the government will have to explain “where in the cited reported does it say there is a critical shortage of STEM regulations?” said Miano. And with that, “the regulations fall apart.”

There had also been an earlier lawsuit from the Immigration Reform Law Institute, back on May 29, 2008, challenging the STEM extension shortly after its announcement. However, the lawsuit was thrown out by a New Jersey district court judge (you can see the full proceedings here).

Conclusion

I’m personally quite grateful for this series of hacks. Without them, I wouldn’t have been able to stay in the United States. Or I might have been able to stay but would have been much more constrained in my choices. Without the STEM extension, I could not in good faith have accepted my current job. Just a few weeks ago, I was a little worried, because my case status for the STEM extension wasn’t showing up on the USCIS website. If my case had been lost, this would have been terrible for me and my employer. However, after a long phone conversation with USCIS representatives, I learned that the reason my case status wasn’t displaying online was that Obama’s deferred action announcement had required some system upgrades so they weren’t up-to-date. I finally got information that my case was proceeding normally, and my application was approved last week.

But, while these hacks have served me quite well, I also believe that the system as a whole doesn’t make sense. I wish that the same effective freedom that I enjoy were available to people around the world, high-skilled and low-skilled. People will always be constrained by their personal limitations and the limitations of the world around them. But I believe that many of the limitations imposed by migration law are an unnecessary additional set that should be dispensed with.

PS: For simplicity, I have omitted discussion of many aspects of the OPT, including something called pre-completion OPT. For a more comprehensive overview, see the Wikipedia page [FULL DISCLOSURE: I significantly expanded and restructured the page in a series of edits in September 2013, while researching it in preparation for applying for it. I made another series of relatively minor edits shortly after the publication of this blog post, adding some material I discovered in the course of research for the blog post.]

PS2: This didn’t fit well in the main post, but might be of interest anyway: employers have historically been reluctant to hire people on OPT if the annual cycle for those people means they can’t have a H-1B petition in by then. A National Law Review article notes that it has been ruled legal (i.e., not afoul of anti-discrimination law) for employers to choose to forgo hiring workers on OPT for this reason:

In a Technical Assistance Letter (TAL), the Office of Special Counsel has stated that an employer can disqualify an applicant because of future OPT expiration without running afoul of the Immigration and Nationality Act. The INA’s anti-discrimination provisions apply only to U.S. citizens, lawful permanent residents who are not yet eligible to apply for naturalization or who have applied within six months of eligibility, asylees, and refugees. “Accordingly, F-1 visa holders are not protected from citizenship status discrimination,” the Office of Special Counsel states. Therefore, an employer who inquires as to whether the candidate requires sponsorship now or in the future is not participating in discriminatory practices.

This makes conditional sense: the private “discrimination” here isn’t discrimination as much as a government-forced constraint on hiring. My co-blogger Nathan has argued that private discrimination should be permitted even as governments get out of the way, but in this case (and many other cases) the private discrimination is a rational response necessitated by government policy.

Related reading

High-skilled hacks: why the US immigration system needs serious refactoring

This post is the introduction to the series of “high-skilled hacks” posts, focused on immigration to the United States. The series explores various workarounds and caveats to immigration law that high-skilled immigrantworkers and their employers have discovered and engineered in order to further their own interests, at the expense of the original intent of immigration law. Through the series, I try to argue that, although these hacks are an improvement over an alternative where only the basic immigration system existed, freer migration for all would be simpler, fairer, more efficient, and more just.

Recently, I posted the following puzzle to the Open Borders Action Group:

In a comment on an OBAG post, Alex says of liberalizing high-skilled immigration to the United States: “Most folks in the policy debate want those reforms or something similar to them. They haven’t happened because high tech reform is intentionally bundled with the rest of the less popular elements of the comprehensive immigration reform package to “sweeten” the whole deal. If high-skilled reform becomes law separately, the rest of the portions of immigration reform become less likely to pass. Those who want the less popular items hold enough power to block individual pieces of the reform from becoming law on their own.

That’s the dominant theory that has failed to produce immigration reform just about every year since 2002.”

Does this accord with your impressions? Do you think people calculated correctly? As an open borders advocate, would you try to reduce the chances of a standalone high-skilled immigration passing in the hope that it could later be added to other, more comprehensive, bills, to sweeten the deal and increase the probability of their passage?

The comments were generally supportive of Alex’s view. I’ll have more to say this later (but check out the comments in the meantime) but in this post I want to consider an auxiliary puzzle: why haven’t high-skilled workers (in the technology sector, academia, and elsewhere), a generally influential population segment where even the natives who “compete” with the migrants tend to be more favorable to migration of their competitors than the general population, been more successful at getting their way? The answer is that they have been able to get their way, but in hackish fashion. They’ve basically taken an immigration system that isn’t designed to be helpful to them at all and made a bunch of changes to it here and there that somewhat approximate a migration regime that is, in practice, a lot more liberal than it looks on paper. Indeed, even a lot of criticism of existing skilled migration regimes generally accepts the broad premises of the status quo and only argues for special treatment of their favored groups. An example of this sort is Paul Graham’s recent article, that I recently critiqued. Yet another example may be high-skilled migration proponent Vivek Wadhwa, whose somewhat confused position at the Intelligence Squared Debate was critiqued by my co-blogger John. Relatedly, the advocacy efforts of FWD.us appear to have been at least partly responsible for the inclusion of H-1B liberalization suggestions in Obama’s November 2014 immigration executive action announcement.

The clever workarounds come at a price. The price is simplicity and fairness. Despite that, the system, with all its hacks, is probably vastly better for the high-skilled workers, the United States, and the world at large (through greater economic growth and more innovation), than the system as originally intended and designed.

Why have high-skilled immigrant workers and the groups that lobby for them been able to engineer so many hacks, and yet not fundamentally changed the system? To understand this, we need to understand the division of labor between different branches of government with respect to immigration law. Some types of changes to immigration law (the creation of new visa categories or the expansion or elimination of quotas for some capped visa categories) require legislative action by the US Congress. Any changes that need explicit legislative action are subject to legislative gridlock for the reasons Alex highlighted. On the other hand, there are other changes, including changes to the scope and time limits of existing visas, as well as enforcement details and the types of evidence that need to be provided, that are a matter of executive discretion. Some of these are decided at the very top, by the US President, but others can be decided at lower levels, by the DHS branches (USCIS, CBP, and ICE) working in conjunction. It is changes of the latter sort where interest groups are best able to move fast. For instance, as I’ll describe in my forthcoming post on Optional Practical Training, Bill Gates’ Congressional testimony suggesting that the length of the Optional Practical Training period be extended led directly to the creation of the OPT STEM extension through direct action at the executive level without the need for new legislation. As Gates himself observed:

This only requires action by the Executive Branch, and Congress and this Committee should strongly urge the Department of Homeland Security to take such action immediately.

In a similar vein, co-blogger Michelangelo has proposed that Obama work to expand the scope of NAFTA’s labor provisions (the TN visa):

Expansion of the TN status should be an attractive route and it is surprising that both successive Presidents and open border advocates have ignored it. The TN status is already part of the US code (Title 8 Section 214.6) and no further enabling legislation from Congress would be necessary. The President’s administration would not be creating a new status using executive order, it would merely be easing the administration process of a well established aspect of US immigration law.

Guest blogger Fabio Rojas has noted that there are many “publics” on any given issue, and on the issue of immigration, immigration lawyers and government agents form a public that is easier to influence at the margin than the general public. Fabio was focused on courts, but the same point applies to enforcement agencies:

This is why open borders advocates should directly target the legal public because of its special position in society. Courts are responsible for enforcing and interpreting the law. Contrary to popular opinion, judges do not apply the law in the same way that a baseball umpire calls “balls and strikes.” Instead, judges are influenced by what they learned in law school, in their time as practicing attorneys, and by political and social trends. That is why the open borders movement should target the legal public. If we can introduce a legal theory that supports open borders, then it will be easier for courts to uphold open borders policy and side with immigrants.

The “hacks” to expand the scope of immigration law are not limited to high-skilled workers. Similar hacks can be found in family immigration law, humanitarian statuses, low-skilled temporary work, and the regularization of people who entered unlawfully. But this series will focus on high-skilled hacks, which have the most “above-board” legal status, partly because of the greater political influence of high-skilled workers and partly because high-skilled immigration meets with less nativist opposition.

On this blog, we’ve been somewhat critical of the “high-skill only” focus of some immigration reform efforts (see for instance Nathan’s post). We’ve also been critical of high-skill-focused groups such as FWD.us. I think our criticism is generally valid. But I recently had a chance to look at the FWD.us video section, and I was somewhat favorably impressed at their use of high-skilled immigration as a way to open a conversation on migration policy. They don’t go as far as I’d like, but they offer important information and provoke interesting thoughts. In this series, I hope to go further and deeper, helping people both understand the current status of immigration law and the direction in which it should be changed.

PS: For those curious about the terminology in the title, code refactoring is a common practice in software engineering. A piece of software originally designed for a particular context, that gets successively rewritten to handle a number of special cases, can get very messy over time. Occasional refactoring to clean up the code can be helpful. The FWD.us homepage had a quote to a similar effect. As John Lee pointed out in a comment on the OBAG post, the “outdated” message has been used for quite a while, at least since the time of Harry Truman. My focus here, though, is not so much the “outdated” message as the “messy” message. There’s a lot of copy-and-paste code and clever workarounds in immigration law. Let’s think of a simpler way of accomplishing morally and practically appropriate goals: radically freeing migration.