Facebook, H-1B and Age

I’ve long emphasized that one of the major reasons the H-1B visa is so attractive to employers is that it enables them to avoid hiring older U.S. citizens and permanent residents. Other key reasons I’ve often cited are a desire for immobile labor (especially if the employer is also sponsoring the worker for a green card), and the convenience of simply recruiting at U.S. universities, where there are lots of foreign students.  In this post, I’ll focus on the age issue, though note that it ties in to the convenience issue as well, since most university students are young.  And I’ll use Facebook as my main example, both because Mark Zuckerberg’s FWD.us has been leading the fight in Congress for expansion of H-1B and because Senator Sessions called it out in his recent speech.

Though I generally place the age discrimination dividing line at 35, I’ve sometimes mentioned that it often occurs even earlier.  Just to make matters concrete before I get to the figures, let’s look at the example of my former student, whom I’ll call Tom.  He has skills of great interest to Facebook, and is one of the sharpest students I’ve had.  I certainly know people at Facebook who are not quite as sharp as Tom. Yet he didn’t even get a phone interview from Facebook when he applied.  This shocked his friends, who with similar backgrounds were in strong demand.

Tom’s problem, I believe, was that, in contrast to his younger friends, he was about 30 at the time he sought work at Facebook, as he had worked in the industry for a while before coming back for his Master’s degree. You might think that’s a plus, but it basically priced him out of the market.  I’m sure that Facebook would value his experience, BUT not so much as to pay him a salary appropriate to his years.

Keep that latter point in mind — experience is good in an absolute sense, but not when compared to the additional salary that goes with that experience.  To quantify that, consider this analysis of Facebook salaries.  The data there indicate that a Senior Software Engineer gets paid about $30,000 more than an ordinary Software Engineer, and more than $40,000 above what a new grad gets.  You do need to take these numbers with a grain of salt, first because it’s Glassdoor data and second because it doesn’t break down according to factors such as degree level (Bachelor’s vs. Master’s) and prestige of one’s university.  (Concerning the latter, two young men wearing Stanford sweatshirts are pictured as new grads, and data cited in my EPI paper indicate that a Stanford degree commands a premium of about 30%.) So, we shouldn’t focus on specific numbers, but one thing is quite clear:  That Senior title means a hefty differential in pay, compared to the ordinary Software Engineers.

Now, add to that another point I’ve shown repeatedly:  Even a Senior title typically means only 3-5 years of experience!  Now you can see why Tom was likely viewed as too expensive — at age 30!  Imagine how Facebook views a 35-year-old, let alone those over 40.  I’ve mentioned before a professor I know, who is over 70 but still active, and who after a visit to Facebook told me with his typical humor, “Most of their employees seem to be the age of my grandkids, a few the age of my kids and none close to my age.”

With that in mind, I looked at the PERM data (wage and other information for green card sponsorees) for Facebook, 2013.  Before I discuss this, though, a note on prevailing wage data (the legally required floor for H-1B and green cards):  I didn’t look at the measure I used in my Migration Letters paper, ratio of wage offered to prevailing wage, because of a major change in the latter.  The DOL has apparently replaced its Software Engineer category with a much broader Software Developers category — the latter having a prevailing wage about $20,000 cheaper than the former.  Knowing that the industry and the AILA are quite aggressive in pressuring regulators for favorable policies, I suspect that this category change came about at the behest of the industry.  In any case, though, I did not use that data here.  (Note by the way that these figures are for the hyperexpensive Bay Area, but it’s true that Facebook pays well.)

Instead, I tabulated the Prevailing Wage Level, I, II, III and IV in the DOL scheme.  John Miano has analyzed such data before, but here I wanted to tie it directly to the age issue.  The system is complex, but Level roughly corresponds to years of experience.  Note carefully that even Level II still is for the very young; if Tom had been an H-1B, he likely would have been at Level III.

Here are the results, among Software Developers:

22% 64% 2% 12%

Well, there you have it!  86% of Facebook’s foreign software developers are younger than Tom (age 30)!  This, I submit, is why Tom didn’t even get a phone interview from Facebook — the firm wants the young H-1Bs instead of him.  And, as noted, they are immobile too, unlike Tom, making them much more attractive to Facebook even if he had been younger.

Easy numbers to remember when you read FWD.us’ literature.


9 thoughts on “Facebook, H-1B and Age

  1. I’ve worked in hi-tech, and business owners love hiring Indentured Servants… I mean H1B visas. You forgot to mention when they are young and naive, they often have no family here, so they will happily work 12 hours a day, Saturday, and maybe Sunday, all for the allure of options or free soda / ping pong. They also are ignorant of their legal rights. Many startups are founded by immigrants, so they often feel more comfortable speaking their mother tongue, and will hire workers who are overwhelmingly from their country of origin.

    I’ve also read large portions of the labor code, and since these are considered highly paid positions, there is no overtime.


  2. > Though I generally place the age discrimination dividing line at 35, I’ve sometimes mentioned that it often occurs even earlier.

    Yes, I’ve noticed the same thing around the age of 35. If you look at the third graph and set of tables at http://econdataus.com/h1binfo.htm you’ll see that over 80 percent of initial H-1B hires are under 35. Likewise, the tables and graphs at http://econdataus.com/h1bage.htm show that the H-1B visas are being used to bring in very young workers. As you can see, the ages of non-citizens working in the industry spikes sharply in the 31 to 35 year range. As the author suggests, this demand is likely motivated by the fact that young workers are cheaper and immobile. Also, they are easier to take advantage of and can be dumped for a new batch of young workers when their visas run out and/or they start standing up for themselves. ALL workers should likely start looking over their shoulders as they approach their 50s, if not before.

    > Easy numbers to remember when you read FWD.us’ literature.

    Speaking of FWD.us’ literature, I’ve posted commentary on a recent FWD.us video titled “Know the Facts: H-1B Visas” at http://econdataus.com/fwdusvid1.htm . In any event, there is a discussion of this blog post on Reddit at http://www.reddit.com/r/Unemployed/comments/2hiqdd/facebook_h1b_and_age_the_major_reasons_the_h1b/ .


      • I agree. In fact, the more that I’ve thought about it the more that I’ve been bothered by that final claim that certain and/or all H-1B visa workers create many more jobs than they take on average. I’m seen that 2.62 jobs per worker number all over the web, even in a White House paper. I’ve been thinking a great deal on the topic and posted a response to just that claim at http://econdataus.com/h1bclaim1.htm . I wrote it quickly though and plan to go back and clean it up. I’d be very interested in hearing if you know of anyone who has verified or critiqued that 2.62 number. Thanks.


        • The general category of job creation claims was debunked in the Wall Street Journal, of all places. See my analysis of the WSJ article.

          The 2.62 figure is from Madeline Zavodny’s paper. This work was funded by an industry-related organization, and the author is a well-known supporter of the H-1B program. In this paper, her analysis relies on a state-by-state comparison of jobs and numbers of H-1Bs. This is an extremely risky approach, because the types of jobs taken by H-1Bs vary greatly from state to state, thus making comparisons highly questionable.


          • Thanks for the link. It’s good to see that people have been countering those claims. When I googled Madeline Zavodny’s paper, I found numerous places citing her 2.62 figure (including the top right of the page at https://www.facebook.com/competeamericaorg ) but nobody countering it. It really does seem that it would be a huge help if the media would subject all papers to some sort of standard to weed out all of this noise. It may be difficult to insure that each paper is reviewed by truly objective and competent people, as you mention to have been a problem with David Banks. But it would seem that we have to try. From my googling, I could find no evidence that Madeline Zavodny’s has been peer-reviewed or verified.

            What would also help would be to demand that all studies release their sources and their calculations in order to be taken seriously. As I mention at http://econdataus.com/h1bclaim1.htm , this was again revealed in the infamous Reinhart and Rogoff spreadsheet error. Madeline Zavodny’s paper gives links to many of her data sources and gives a general description of what she did. But, just as with the Reinhart and Rogoff paper, there is no realistic way that anyone could reproduce her results without her spreadsheet(s). I don’t expect for any of my conclusions to be accepted without showing my sources and work and don’t see why anyone else should expect that. If papers had to start supplying their sources and calculations in a way in which their results could be verified, I suspect that many of these amazing claims of job-creation would disappear. Those that remained could at least be verified.


  3. > 86% of Facebook’s foreign software developers are younger than Tom (age 30)!
    I know many folks (including my wife) who work at Facebook and 30+ years old. One data point is missing while crunching these numbers. Most of the companies including Facebook, sponsor green card for foreign workers. So if any foreign college grad joins any major firm right out of college, he/she will get green card in 3-4 years. So if you graduate at age of 25, you will have green card before you turn 30 (unless you are from India which has a longer waitlist). So company won’t have to file H1-B anymore. This makes you data and conclusion very biased.


    • No. What you are saying is irrelevant.

      My point about the relation between H-1B and age is (and always has been) that the employers want to hire younger (thus cheaper) workers, and thus benefit from the fact that the H-1Bs swell the young labor pool.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s