Examining Potential Impact of H-1B Reform with Data Visualizations


USCIS started accepting 2018 H-1B Visa applications in 2017. Explore potential impacts of the proposed H.R.670 bill on the tech industry and American economy.

This post may describe functionality for an old version of CARTO. Find out about the latest and cloud-native version here.
Examining Potential Impact of H-1B Reform with Data Visualizations

On April 3, 2017  the U. S. Citizenship and Immigration Services (USCIS) will begin accepting 2018 applications for H-1B Visas. The H-1B program permits U.S. employers to temporarily hire foreign workers with specialized skills each year. In January, however, members of the House of Representatives introduced H.R.670: High-Skilled Integrity and Fairness Act of 2017, a bill proposing significant changes to the current H-1B application process.

H-1B opponents have accused the bill of outsourcing American jobs at lower salaries to immigrant workers. Yet, this argument overlooks the fact that large segments of the American workforce lack necessary skills, and especially tech skills. The tech industry, which accounts for 12 percent of all jobs in the United States, is likely to feel the effects of H-1B reform immediately as H.R.670 would impose restrictions on recruitment and retainment of talent.

Let's explore possible ramifications for the tech industry, and the American economy more generally, should H.R.670 go into effect. Using open government data, spatial analysis, and data visualizations, let’s see what these changes may mean in the not so distant future.

The Current H-1B Application Process

To hire foreign workers through the H-1B program, prospective employers are required to submit a [Labor Condition Application (LCA)], and must designate themselves as either an H-1B Independent Employer or as an H-1B Dependent Employer. An employer's status is based upon a specific fraction that considers a company's size and the number of workers already employed on H-1B permit. One such fraction, for example, designates dependent status to a company employing 50 workers or more where at least 15 percent of those employees are on temporary work permits.

H-1B Dependent Employer applications are subject to extra scrutiny while under review by the Department of Labor, an added obstacle perhaps deterring employers from applying as a Dependent Employer. But this scrutiny can be avoided in one of two ways:

A pay wage exemption for applicants earning $60 000 or more each year.


An education exemption for applicants holding advanced degree (master’s or above) from U.S. institution of higher education.

Crucially, these exemptions level the playing field for both employer types. Of the total 85 000 visas granted each year, that is  20,000 are reserved for applicants meeting the education exemption requirements irrespective of employer type. Given the pay wage exemption, then, the probability of visa allocation for both H-1B Dependent and Independent Employers becomes the same, which yielded a probability of 33.33 percent last year presuming equal odds for the 236 000 applicants applying for one of the 85 000 visas available through the current lottery system.

The data visualization below is built with open data provided by the U.S. Department of Labor on H-1B visas applications received for fiscal year 2017  which totaled nearly 236 000.




This proportional symbol map, whose circles are proportionally sized to represent difference between number of visas applied for, by H-1B Dependent Employers (Red) and Independent Employers (Blue). The widget panel on the right hand side displays the mean average salary paid to employees on work permits for each employer type.H-1B Dependent Employers paid employees an average mean salary of $73 000 whereas H-1B Independent Employers paid an average mean salary of $89 000, which seems to confirm the suspicions of critics regarding lower wages.

Now  assuming equal statistical probability, let's see how the H-1B Dependent Employer exemptions play out within the current process in the data visualization below:





From the congressionally approved cap of 85,000 H-1B visas for fiscal year 2017, the lottery system allocated approximately 40,000 to Dependent Employer applicants and the remaining 45,000 to Independent Employer applicants.

The lottery-based system's random selection seems to promise fairness to applicants irrespective of employer type, which would ideally calm opponent’s suspicions regarding Dependent Employer applications.

And yet, H.R.670 would target H-1B Dependent Employers specifically, while also replacing the current lottery system with a merit-based system based upon market-value pay allocation.

Potential consequences of the High-Skilled Integrity and Fairness Act of 2017

H.R.670 proposes replacing the lottery system with a metric system where decisions would be determined on employee's merit, which the bill presumes would be reflected in annual pay wage reflecting market-based allocation. In addition, H.R.670 would change the Dependent Employer exemption in the following ways:First, the education exemption would be eliminated, which will lead to a significant decrease in the number of H-1B Dependent Employer visas and an increase in H-1B Independent Employer applications.Second, what the LCA terms the "prevailing wage" system exemption for dependent employers would be raised from $60,000 to $130,000.

The LCA’s current three tier "prevailing wage system" for all employers is also proposed to be readjusted to align annual wages for employees with specialized skill to market-based allocations for such positions. Although figures have yet to be revealed, analysts are confident that the minimum wage brackets for all professions would be raised  thus flooding this new bracket.

Subsequently, H-1B preference would be given to level three candidates with an annual salary equal to 200 percent (or more) of his or her American counterparts in that given profession. Each tier would follow suit as level two applicants would need to make 150 percent (or more) and level one applicants 100 percent (or more) in annual salaries than their respective American counterparts. The implicit logic behind H.R.670’s proposal to increase the pay wage exemption is that these above-market salaries demonstrate the invaluable expertise that an H-1B applicant offers, which  ironically enough  is what is now being valued according to market-based allocations.

Given the current level-three wages for all professions, a negligible fraction of applicants will meet either the 200 percent or 150 percent pay increase thresholds. As such, nearly all applicants will fall within a 100 percent of level-three wages.

Let's take a look at what this could mean in the data visualization below.






H.R.670 changes would restrict H-1B Dependent Employer eligibility to such an extent that only 928 visas would be issued.The H-1B Dependent Employer workforce would immediately decrease by nearly 39,000.These figures are based upon available open data and are liable to change in the near future, but nevertheless the information is alarming. The New York Times, for instance, reports that the recent fervor for immigration reform, of which this is a part, aptly titled the "Trump effect " has yielded a 40 percent decrease in international student applications to study at American universities.

As H.R.670 congressional debates near and as this type of decline spreads beyond academia in response, we are left with far more questions than answers: Will USCIS release open data for H-1B visa recipients post-2012? What will the proposed reforms to LCA's "prevailing wage system" look like in terms of dollars and cents? How will market-based figures for LCA "prevailing wage system" be determined? And, more over, how can these figures account for economic fluctuation within a globalized world? What measures are being taken to (re)train American workers in tech skills that will be needed should the bill pass? Cognizant Technology Solutions, as found in the data visualization above, could lose nearly 9,800 employees whereas Deloitte Consulting would gain close to 8,700 employees given my statistical assumptions.

The data visualization below reflects what these fluctuations could look like should the H-1B process change.






Companies will need to reassess allocation of resource because of external disruption. Although H.R.670 assumes that these measures will return the American workforce to greatness, but due to external pressures imposed within tight time frame companies will need to make decisions in company’s best interest to keep business afloat.

Are there any viable alternatives?

Since recruitment of new talent would be severely restricted following H.R.670 reforms, are there any options for companies to even retain current workers?

Hypothetically speaking, if companies were to exempt employees from proposed H.R.670 restrictions by reallocating excess revenue  how much would the companies need to spend?





The widget panel for the data visualization above displays buckets for each employer, the annual amount of capital needed in order to retain current employees, and the total percentage of annual revenue that retaining H-1B employees would cost. For nearly all employers the prohibitive costs do not provide a viable solution. Cognizant Technology would need to raise 1.6 billion to retain its temporary foreign workers, which would be nearly 11.6 percent of its annual revenue.


Although the proposed H-1B reform bill has yet to be ratified, the preceding exploration has raised several questions that need answers and soon. H.R.670's reforms will have long lasting effects that will be felt not only for temporary foreign workers whose lives will be disrupted, but also nearly every aspect of the American economy that deals with the tech industry.

So as the H-1B visa application pool opens in less than a week, let's demand more open data, and government transparency more generally, from elected officials before it's too late.

Happy Spatial Analysis