The Old (Age 35) American Software Developer

A friend called my attention to an article, “The Plight of the Graying Tech Worker,” by Prof. Bill Kerr of HSB. Most economists writing on a controversial topic will be rather (or very) strongly on one side or the other. On H-1B, Bill is pro-, as evidenced by a string of published papers, as well as his recent book, The Gift of Global Talent: How Migration Shapes Business, Economy & Society (Stanford University Press, 2018).

I first “met” Bill in 2009, when he and I were virtual guests on Lou Dobbs’ TV show, following the publication of Bill’s first paper on H-1B. I disagreed with its finding (which in any case has been widely misinterpreted to mean that the H-1B program increases US innovation), but was rather harsh in saying so. This I regretted later when I met him in person, as I found him to be a warm, friendly guy of high intellect and a genuine desire to get at the truth. I’ve continued to appreciate these qualities ever since.

Yet I tend to find fault with some his analyses, as in this 2013 post to my old e-newsletter. (Be sure to see Bill’s rejoinder.) One point I made in particular is worth including here (“KKL” refers to the three authors — Bill, his wife, and a collaborator, Bill Lincoln; by the way, this 2013 working paper appears to be Reference 8 in the current article):

Congress and the press will take KKL’s research to have implications specific to the hiring of engineers on temporary work visas–even though KKL’s work does NOT do this. People in Congress and the press are not going to read the disclaimers in the paper.

Even more importantly, KKL omit what should be an obvious analysis. They spend 46 pages measuring what they claim to be the effect of hiring immigrants is on a firm’s subsequent hiring. The obvious question is then what the effect of hiring NATIVES is, compared to hiring immigrants. If, for example, the native impact on hiring is greater than the immigrant one, then policymakers should be reluctant to expand foreign skilled immigration in any long-term sense; indeed, in such a scenario, H-1B would bring net harm to the economy.

I asked Bill about this omission back in May. He replied, “The main rationale for pursuing the immigrant hiring calculation [but not] the native hiring is that the former is more proximate to a policy action and its consequences than the latter is. Given that the ‘creates 4 jobs’ statements are at the center of the policy debate, they will tend to get the most attention from economists.”

Proximate, schmoximate, I strongly disagree. The policy debate focuses in large part on the industry lobbyists’ claim that the foreign workers are especially innovative, “the best and the brightest,” and thus have special job-creating powers. It would thus be natural, indeed imperative, to compare the job-creating powers of the immigrants to the job-creating powers of the natives. Arguably, KKL’s failure to do so suggests an unconscious bias on their part.

The friend who brought this new article of Bill’s to my attention presented it as a “man bites dog” piece, quite out of character. Yet its theme, the adverse impact of the H-1B program on older US techies, is not new to Bill at all. Indeed, his 2013 paper cited above found such an impact.

I highly appreciate that previous work, and the current article, as I have over the years maintained that age is a, arguably THE, core issue in H-1B. Older workers cost more than younger ones, not only in wages but also in health care and other benefits. Since most H-1Bs (and almost all H-1Bs in the tech field) are young, basic economic principles apply — employers will prefer the young H-1B to the older (read 35+) US citizen or permanent resident.

So, good for Bill for addressing the issue, one that most critics of H-1B (writers, think tank people, researchers) studiously avoid.

Nice. But sorry, I don’t like Bill’s current article. First, he states that he is focusing on the adverse impact of H-1B on older Americans because young ones are not affected much. In fact, employers have strong incentives to prefer young H-1Bs over young Americans. One major benefit of hiring an H-1B is that she is a de facto indentured servant, trapped with her employer, who then need not worry that she might jump ship for another firm, a huge advantage. And Bill’s assurance that wage underpayment is not an issue, due to the H-1B statute’s prevailing wage requirement, is not correct; the legally-defined prevailing wage is typically well below the wage the H-1B would command on the open market.

Second, I am disappointed by Bill’s two main recommendations: (a) the government should dole out the H-1B visas in order of offered salary, and (b) the harm to older US workers should be ameliorated by retraining programs funded by H-1B user fees.

To be sure, I myself support (a). However, the subtext of Bill’s support seems to be the “Intels Good, Infosyses Bad” idea that I have criticized so much — the thinking that the Indian outsourcing firms are the main abusers of the program, while the US mainstream firms use it responsibly. Bill is missing the boat here, and I believe it colors his other analyses.

Bill’s advocacy of (b) is puzzling. Doesn’t he know that this has been tried before (in the 1998 H-1B bill) resulting in failure? Even more importantly, it was pointed out — beforehand by me and in retrospect by Sun Microsystems — that such a program CANNOT work. An older programmer, say, who had been passed over by employers because he was too expensive will STILL be too expensive (and even older) after retraining. Actually, most software developers do have modern skill sets; they tend to be the types who love tinkering and learning new technologies. But employers won’t touch them, due to their perceived salary requirements, not to mention their health care costs and so on.

Again, all this is old news. See for example my comments in another post to my old e-newsletter, in which I said regarding Microsoft,

More subtle, but equally important, is what the article says MS will do in expanded MS operations in Vancouver: “This month [MS] announced plans to more than double its roughly 300-employee office in Vancouver, where video games have been the focus. There, Microsoft will hire and train 400 software developers from around the world to work on mobile and cloud projects.” What? “Hire AND TRAIN”? Bear with me while I explain.

The training issue–a phony one, for reasons I’ll discuss shortly–has been central to the H-1B debate ever since the industry’s first push to expand H-1B in 1998. The legislation that was enacted included an H-1B user fee, so that employers would fund training that would allow Americans to do these jobs in the future. I stated at the time that this would never work, because (1) good programmers can learn new skills on the job, very quickly, without formal training, and (2) employers wouldn’t hire them even with their newly acquired skills, as the employers wanted cheap foreign workers.

As Bill’s book title makes clear, he does indeed view the “Intels” as good economic citizens, and views America’s world tech dominance as due in substantial part to H-1Bs hired as foreign students from US universities. But no, sadly the “Intels” tend NOT to be  good economic citizens. I’ve written, for instance, that Google openly told a group of us researchers that it prefers to hire foreign workers due to their “trapped” nature, and need I mention the name “Facebook”? And though I’ve always strongly supported hiring “the best and the brightest” from around the world, the vast majority of foreign-student H-1Bs hired by the Intels are ordinary people doing ordinary work. Without understanding these key points, it’s impossible to formulate good solutions to the age problem, even if it were politically feasible to do so.



27 thoughts on “The Old (Age 35) American Software Developer

  1. I have created a new page at Keep America At Work where I address all of these issues in 3 short videos.
    I urge you to view them and the associated links found below the videos on youtube as they show that we are not creating jobs at all in the computer and mathematical occupation group (which is where the majority of H-1B usage is).

    They also address the other issues mentioned in this article.


    • For what it’s worth, the 737 Max problem seems to be the result of corporate decisions rather than sloppy software development. As I understand it, there was a decision that the control system would *forcefully* override commands by the pilot, and another decision not to include redundancy in the sensors.Those decisions were probably made to make the engine upgrade appear seamless.

      By comparison, sloppy development will usually manifest in reports of spaghetti code, user reports of significant errors that don’t get fixed and excessive project delays that fail to converge. When these characteristics emerge, you almost always find that the company involved is a big user of H1-Bs or their equivalents.


  2. Norm, would you know where this “creates 4 jobs” comes from? Report? News article? I’ve never seen the source of this.

    As for visa indenture, it is the only legal form of 360 degree non-compete for employers. And subservient, as they aren’t here for the job, they’re here for the potential green card.


      • To be even more precise, it’s 2.62 jobs. With two decimal places, it’s got to be accurate! Seriously, it is totally fake as you say. I reproduced that result and posted the main problem with it at . The model that came up with a 2.62 job gain was looking at 2000-2007. But if you slightly change the first or last year, the result changes radically. For example, if you just push the span forward two years, to 2002-2009, the 2.62 job gain becomes a 1.21 job LOSS. Any reasonable person would conclude that such a model is useless. Yet the 2.62 number continues to be cited as can be seen in the listing at .


  3. You support (a), I support (a). So let’s talk about (b). What if, say, big fat employers like Microsoft, would actually be very pleased to take (some) older workers, put them through six months of full-time retraining, and then keep them on after all? Frankly it’s nonsense, but if it were nonsense that would change Microsoft’s behavior, then let’s have more nonsense!

    The major effect of the H-1B program is to lower the entire wage scale tremendously, by 50% or more, by clever misuse of the prevailing wage rules, as you always say.

    But I’m wondering if that even matters anymore. The IT/coder communities at this point are so overpopulated, and the jobs are so trivialized, that I think the dynamics are rather different now than ten or twenty years ago. And these communities are overpopulated because so many other job categories in the US (and world) have been decimated by globalization and automation. So many of the jobs are deskilled, expectations are horrendously low, and pay follows correspondingly.

    I look at LinkedIn a lot to try to judge what’s going on – it has proved almost useless in job search for me, I assume exactly because I am not looking for one of these homogenized, commoditized positions. For anyone who still thinks of STEM as a profession, at least in computing, I suggest my favorite cliché now on LinkedIn: “Jobs are obsolete”. Take “jobs” for a few years, if you can get them, as sort of a post-grad. Then start your own company. Requires additional skills, so is not for everyone, but it’s much more urgent to do this today, and there are many more opportunities for doing it, than twenty years ago.

    Liked by 1 person

  4. Kerr doesn’t really understand skills. The article he links to is mostly talking about outsiders learning basic skills in order to change careers, not experienced professionals.

    More importantly, experienced programmers are expert learners and update their skills without external training courses. Knowledge of particular languages or frameworks is only a tiny part of programming.Economists don’t understand or account for that.

    As well, he presumes that what’s good for individual hiring managers is good for their companies and good for America, which is not necessarily true. Excessive access to visas hides the fact that many hiring managers are themselves a bit shaky on identifying skills, and it also protects inefficient companies.

    For example, many hirers default to hiring people from the same backgrounds as themselves. That’s how companies like Facebook become thick.

    Making it easy for outsourcer executives to hire lots of young workers lets those outsourcer companies displace expert smaller firms who, in bidding for big government contracts, would do a better job at a fraction of the cost. Economists presume hirers and government make perfect decisions which is just not true. Government project managers think a firm with 1,000 “coders” must be better than a team of 10 in the suburbs, which is also not necessarily true.

    We are finally seeing policy makers reexamine their relationships with Facebook, Google and the others. It’s time for economists to get busy too.


    • “Making it easy for outsourcer executives to hire lots of young workers lets those outsourcer companies displace expert smaller firms who, in bidding for big government contracts, would do a better job at a fraction of the cost. ”

      That is exactly right. I did some quick-fix contract work for our county government on a project that was way behind schedule, and I learned of a nightmare scenario. That project was to take two years and cost ten million dollars, and was outsourced. On various techie boards, I would relate the mess they made – over $100 million and ten years, it still wasn’t done. Every single time I told that story, others would post similar stories from all over the country.

      They were all the result of have non-techies managing large contracts they were technically incapable of understanding. In our case, this was a year or two after firing almost half the experience professional programmer/analysts.

      It’s kind of like in the 60s, “made in Japan” meant soddy workmanship, it took a generation or two, but eventually Japan took over the consumer electronics industry,and drove American TV, auto and audio companies out of business.

      In the 80’s Chinese manufacturing was famous for making cheap toys and tools. That the toys broke was not important but the crappy tools irritated a lot of people. It took a generaion or two for the Chinese manaufactures to mature, but that happened.

      Now, software has to go through that same maturing situations. Unfortunately for us, the generatin or two that will be required for the developers to mature will be suffered by taxpayers and users, with no recourse.


  5. Trump’s election came largely from displaced workers both young and old. My unemployed millennial pulled the lever for Trump because the priorities of the other side didn’t benefit him. Bernie was his first choice, but Bernie sold out and will not be trusted again. The immigration issue will dominate politics for a long time in America and the resentment is deep. The media ignored it at their peril.

    Now the media asks, where is the socialist undertone in politics coming from? It is coming from a generation that watched their parents lose everything and landed up with nothing. The focus on Americans vs outsiders will be a huge issue that will affect the millennial generation in the same way the great depression affected that generation. Politicians ignore it at their peril.

    The arguments about H1Bs and closing borders has already been decided in the minds of a generation. The govt can act or not act but there will continue to be ramifications on the ground.

    Liked by 1 person

  6. Regarding ” the harm to older US workers should be ameliorated by retraining programs funded by H-1B user fees.”

    I don’t need retraining.

    First, at the age of 40, I was ordered by corp mgmt to train my foreign replacements. I left mid-training when I found work with a former client. 17 years later I’m still developing applications.

    Liked by 1 person

    • Right, I often point this out too. With Americans training their foreign replacements, who’s the one who needs training, the Americans or the foreigners? It’s a good, quick sound byte, but more relevant to the Infosyses than the Intels.


      • Mike is so spot on about the retraining.
        I’ve had to teach myself javascript, google maps, php, mysql, how to make videos, how to research government data and thousands of other things since I got forced out back in 2003, all in the quest to get rehired, which at 62 isn’t going to happen no matter how hard I try.

        My point very simply is, we can acquire whatever skills are necessary if we are given the option and we want too.


  7. It seems there is concerted effort to drive Americans out of several tech jobs. And the new immigrants are playing up this game very well, the LA times article about a certain Chinese national “Leo Wang” who couldn’t get H1B was written by a fellow Asian, his story re-appeared in Business Insider after a while. Then a recent story about certain Indian “Viraj Parikh” on CNBC was written by 2 Indians. I’m not posting links to these articles here as they can be found easily. No wonder the 2016 election turned out the way it had.


    • Needless to say neither of the two subjects in these articles are “geniuses”, rather someone trained to get that 100k Cali H1b job and immigration benefits along with it. As millions of people in Asia graduate into the middle income bracket and able to buy plane tickets, it seems the open border lobbyists want them to come and immigrate here.


      • And another article piled on about reverse brain drain yada yada yada …. “Indian Expats Are Leaving SoCal To Find A Better Life — In India”. The subject of this article is an utter garbage based on his LinkedIn profile, an American high schooler would do better than him. And yet the author wants to fight for him and make him a hero or sorts.


  8. I found the following excerpt from your critique of Kerr’s paper interesting:

    > IVs are quite popular among economists, but are rarely used elsewhere. Even among economists IVs are controversial.

    As it happens, I just finished a discussion regarding Giovanni Peri’s use of IVs in a paper at (search for Peri). Most of the posters there seemed to be great admirers of Peri and critics of George Borjas. They had brought up an early look at a Peri study that I had done where I did an OLS regression. Of course, this was beyond forgiveness and they extolled the superiority of IV regression. I did in fact use IV regression in a replication of the study at and still found problems with the paper. Of course, nobody there had any interest in those problems and only wanted to extol the virtue of IVs and Peri’s superiority over Borjas.

    I have been interested in promoting the replication of studies for several years. However, it seems that the most important studies to replicate are those whose findings continue to be referenced and likely affect public policy. This would include the study the claims the 2.62 job creation and another claim I’ve seen that there will be 1 million unfilled computer jobs by 2020. It is likely also important to replicate studies that are often referenced by other economists in other studies.


  9. Norm, like I’ve said in previous posts, I spent over 35 years in IT, the vast majority in private enterprise. Over a decade at my last firm, we were downsized to the max and I survived at least 10 layoffs. Seeing the writing on the wall, I took an IT job with the State of Oregon. While I can’t say that I lost any job due to the H-1B visa, I do agree that it’s depressed my wages over the years.

    In my 3 years in government employment, I’ve met many co-workers that are from private industry. Their reasons are many – laid-off, H-1B took job, jobs offshored, desire for pension, worried about losing benefits/health care, want union protection, etc.

    What I’ve come to conclude is, we need to come to a realistic solution that we can win. I think everyone involved, and on both sides of the H-1B argument, needs to inform young IT workers and students:

    • Work in private IT industry until you’re 35 or 40
    • Then
    – Change industries OR
    – Work for the government until retirement

    I don’t see any other way out of this. We need more laws that give preference to citizens for local, state and federal IT jobs – especially workers 40 and older. Perhaps if we could make a deal with Silicon Valley and their co-horts, maybe they would back such a plan. They could have their hordes of young, dependent foreigner workers that will willingly work weekends, holidays and after-hours for low pay – as long as we have a place to go for employment. Add in restrictions on the number of H-1B contractors that the government can contract out to or hire.

    Believe me – I’m not discounting our worth. But with Silicon Valley’s billions in lobbyist money, I don’t see how we can fight it.

    One last point – I’m in agreement with those that have the capital and the will to start their own firms. There are companies and managers that appreciate great quality software written by experienced, seasoned IT workers. But most of us don’t have the means to start such businesses.


    • Wholeheartedly agree with pablo

      Problem I’m finding with the Department of Veterans Affairs is they don’t believe any of this is happening.

      I’ve now worked there twice as a janitor or housekeeping aid begging everybody for the opportunity to prove my IT skills with no luck.

      I’ve been unable to find anybody that was willing to run opposition for me even though I have written everybody at every level between myself and our senate and congress veterans committees.

      I will be 62 in December.
      I would like to at least have 5 good years before I die, but I sure don’t know how to break through this bureaucracy.

      For what its worth, I’m fixing to try a third time as I should be going to work on the evening shift in a few weeks and will continue to work my day job for as long as it lasts.

      I bring this up because people need to know that there can be safety in government service IF you can get in there.

      What is sad is the IT person that supported the Kerrville VA retired in Feb.
      I have the skills to take his job.
      He tried to get them to hire me to take his job.

      So far, nobody has been hired to take his job and it is my understanding that they only send somebody from austin twice per week as they don’t think there is a need for a support person in a facility where they have close to 500 boxes.

      Sadly, I just read that a veteran committed suicide at the Austin VA hospital in the waiting room.

      I feel for him and his family, but I can understand the frustration that led to that terrible decision.


      • Virgil, I wish you could move to Oregon. We have a lot of workers and contractors that are 50+ and some that are 60+ – I’m one year from 60 myself. The rumor is that many state workers will retire in the next coming years due to PERS pension changes. Here the state is willing to take anyone that can do the job, regardless of age.


  10. Have you ever wondered the cause of ageism in tech industry as well as in corporate America? Here is my thought (based on the experience I had when I lived there) – American society is highly age-segregated as opposed to Asian / Indian societies. Note that in Asian and Indian cultures old people live with their children and garner a lot more respect in society, family as well as sought for their wisdom in Industry. On the contrary in western culture, old people / parents etc. are not respected in the society and when these dudes grow up and work in corporate America they pursue the same cultural traits. And this translates into ageism in Corporate America. So, if you want to fix this, you should look at changing your culture.


  11. Having worked now worked something like 36 years in high tech, my experience is that when in the last 10+ years that I have seen layoffs, the people being laid off were NOT behind in the latest “technologies”. The whole presumption of “re-training” people who were laid off pushes the idea that they were inadequate in their skills After I myself got laid off, I realized that there is an entire cottage industry of HR reps, job consultants, writes,, “job visionaries”, etc who are part of a cottate industry justifying the laying off and firing of older workers. Bill Kerr and people like him are part of that profitable cottage industry.

    After two years I lucked out and found a job with a company taken over by Europeans who were not ageist in hiring. Let me repeat lucky. Two of the last people I knew who were laid off and who I thought were very good ended up as shuttle/bus drivers. The whole idea of re-training older workers is an utter scam.

    Liked by 2 people

    • Maybe if you laid off your govt representatives regularly their attitudes towards your real world problems would improve.


  12. It does not help when economists, including economists employed by the Federal government, refer to “prime working age”. Prime working age is 25 to 54.

    Age 55 to 64 is “those passing the peak of their career and approaching retirement”.

    Better to get rid of the 35 year old so that he/she will not have protection under age discrimination laws that kick in at age 40.

    Yes, health costs can be a major reason for getting rid of older workers. One more reason we need healthcare reform.


Leave a Reply

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

You are commenting using your 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