10 ms·

Companies are lying about AI layoffs?

199 points by huijzer 1y ago

John23832 1y ago
Why hire an h1-b when you can shift the entire campus abroad? Google, Microsoft, and Amazon have definitely been doing that. This practice hurts all state side workers, citizen and h-1b alike.
- dboreham 1y ago
  I know there are stories of H1-B farms but my personal experience is H1 visas were only used to move people who otherwise would be living in a foreign country, being less productive (due to TZ skew) and not paying US taxes, employing people to cut their hair etc, to the US. Me and my former colleagues on those visa must have bought a few F35...
  - scarface_74 1y ago
    Well your personal experience doesn’t support the facts — see the WITCH companies.
- mapontosevenths 1y ago
  Interestingly, the folks I've actually seen replaced by AI the most have been offshore workers in entry level roles. Things like Indian call center roles are low hanging fruit for AI replacement in my recent experience.
- zajio1am 1y ago
  But helps abroad workers.
lordkrandel 1y ago
Association is no proof of cause. How many h1b were issued the years before AI? Has there been an increase? My bet is: no.
- huijzer 1y ago
  > How many h1b were issued the years before AI? Has there been an increase? That still wouldn't prove anything per your logic because, as you say, "Association is no proof of cause." But then the question is how could we ever proof anything like this? > How many h1b were issued the years before AI? The point is that companies are lying off people and saying it's AI and the economy while they still hire foreign workers. That suggests they are not telling the full story. Whether or not they are issuing more visas than before does not affect this conclusion.
  - lordkrandel 1y ago
    Who cares if you never find proof? This doesn't mean that what is unproven is true. Otherwise, Shiva god may be truely existing. You can't disprove that, right?
- MisterSandman 1y ago
  This is also assuming that 100% of layoffs are American citizens. A huge chunk of those layoffs are H1B workers themselves.
agentcoops 1y ago
I think this interpretation of the data rests on a misunderstanding of how a firm conceives of headcount. You saw this a lot during Covid layoffs when the press tried to point out the contradiction between hiring X number of people at the same time as firing Y number of people. Wasn't the firm just staying in place? Yet, in fact, the firm had planned to hire Z people that year, a number much greater than X, greater still than X+Y, and likely greater than Z_t-1, i.e. the number hired the previous year. A firm's headcount is a dynamic value that is reflective of a rate of growth: its headcount is going up by x% each year, ideally linear growth or at least not a higher rate of growth than revenues or free cash flow. It requires an accounting perspective on the inflows and outflows of an ongoing process and not just a small slice of data. If you just look at that firm's hiring and firing in a particular year, subtracting the one from the other, you would appear to have shown that the firm was more or less simply replacing the other with the one. Yet, the main question for actually measuring the (firm and investor expected) impact of AI is whether that rate of headcount growth has changed without the stock price going down. That is, is it now proposed by management and accepted by investors that the firm will henceforth require less human labor in order to deliver on the present expected value of future free cash flow. You cannot infer anything about this question from the data shown. There certainly is a separate question of whether a firm has changed its source of new hires, but that also isn't reflected in the data, which just shows approved visas against layoffs. For this, you would need to look at net hires year after year and prove that a growing percent thereof are H1-Bs.
- huijzer 1y ago
  > It requires an accounting perspective on the inflows and outflows of an ongoing process and not just a small slice of data. Correct, but we don't have that information so we have to rely on a best guess. We're not dealing with code here where everything is known exactly. > Yet, the main question for actually measuring the (firm and investor expected) impact of AI is whether that rate of headcount growth has changed without the stock price going down. Stock price is not necessarily a proxy for headcount either. Some companies have much lower headcounts per stock price (for example, Nvidia). > That is, is it now proposed by management and accepted by investors that the firm will henceforth require less human labor in order to deliver on the present expected value of future free cash flow. You are just talking without thinking now are you? It makes no sense. Human labor and free cash flow are probably correlated, but one isn't a proxy for the other. First you relate stock price to headcount and then you say "that is" and then you suddenly relate cash flow to headcount. Cash flow and stock price are correlated per industry, but it's also a very loose one. > You cannot infer anything about this question from the data shown. False. I only need to infer a tiny bit in order to falsify your extremely strong claim. > For this, you would need to look at net hires year after year and prove that a growing percent thereof are H1-Bs. Yes, this is something concrete that I can agree on. But how would you prove that given that we don't have access to that data?
nis0s 1y ago
The hard truth for Americans (I am talking about Americans because most of these are large cap US companies) is that other countries are cheaper for the following reasons: 1) Companies can find comparably trained and educated workers in non-US countries at cheaper rates because the U.S. education system doesn’t discriminate against less performant students. The education systems in many other countries are competitive at various stages and serve as strict filters against laggards, and there’s less consideration for meeting basic human needs in some countries. But in others there are different options where many students naturally find themselves (like less paying or blue collar work), and if that were to happen in the U.S. then inequality advocates will have a lot to say, and they wouldn’t be wrong. 2) There are a lot of realpolitiks at play when you have a multicultural society as people will often have dual allegiances. A CEO or hiring manager from country X or Y will want to favor their country of origin, naturally. It becomes easier to justify some decisions when the bottom line is also helped by said decisions by hiring cheaper labor. 3) The most expensive thing on the U.S. budget books for employers, employees and the government is social security and Medicare. Some countries simply deny their citizens and workers social security and access to welfare programs, which helps keep the cost of employing them down. 4) Even if a global corporate tax rate is set, it will still make labor in countries that don’t ask for social program contributions much cheaper. 5) The rule of law situation in some of these countries is less than ideal, and so living there can be a hit or miss, which is why H1-Bs are a popular choice for both employers and non-U.S. employees. 6) But still it seems that regional security and stability in some regions has led to permissive environments for unrestricted work. This situation isn’t going to change for nuclear countries, unless they’re like Russia (screwed by economic sanctions and expansionist desires). So in the short-term, unless there’s some agreement to be made between countries for balancing their labor market loads, then economic sanctions seem like a terrible last resort. Addendum, I think the hard truth for large cap companies is that they never should have been allowed to get that big, and in many cases their growth was aided by direct investment by the U.S. government. There’s an interesting lesson here for policy makers about balancing power and investments. Maybe the U.S. should divert future government spending to only its research labs.
- deadbabe 1y ago
  Go ahead and hire exclusively cheap foreign engineers. See where it gets you.
DebtDeflation 1y ago
You also need to consider the impact of offshoring. Accenture's layoffs of 11,000 workers is listed in the chart, but a few days earlier they announced they were opening a new campus in India and hiring 12,000.
- steveBK123 1y ago
  Corporate America really thinks we are that stupid
  - sfn42 1y ago
    Looking at who you elected (twice!!) I don't know how wrong they are.
    - soulofmischief 1y ago
      Friendly reminder that America is not a single person, but a very large country filled with many different opinions.
      - zenlot 1y ago
        You realize that the large country you're mentioning, with many opinions, still elect 1 president, who represents that large country with different opinions?
        danaris 1y ago
        You realize that no more than a third of the people in the country actually voted for said president?
        zenlot 1y ago
        You realize, that it doesn't change the fact, that the elected president represents that large country with many opinions?
Havoc 1y ago
I don't see how the data presented necessarily leads to anything remotely close that conclusion with any real confidence. If anything the biggest H1B user - amazon - seems to have done negligible layoffs.
- huijzer 1y ago
  > I don't see how the data presented necessarily leads to anything remotely close that conclusion with any real confidence. You are not providing a counterargument apart from the fact that "you don't see". Well. Thanks for sharing I guess.
  - hshdhdhj4444 1y ago
    And OP is not claiming to provide a counter argument. They’re claiming, correctly, that the presented argument is invalid. They’re not even contesting the conclusions. Only that the information provided in the linked article does not justify the conclusion.
  - sm1k 1y ago
    This post is quite similar to tea leaf reading. There isn't a pattern explained by data, it doesn't exist. So there is no counter argument to provide apart from acknowledging the analysis presented in the post is totally bogus.
- jeremyjh 1y ago
  Yes, without knowing H1B beneficiaries in prior years its impossible to even conclude they've had a net add of them, since they have to be renewed every few years. I'm sure there are some net adds, but we have no idea how many.
dzonga 1y ago
worrying about H1B is looking at the wrong metric. Most companies are past H1B - the game is now offshore stuff. Look at the data: e.g a company like Rippling where a good portion of Software engineering jobs are in India. then Senior / Staff Roles are in the US. Most scale ups startups have similar dynamics. It could be India or Eastern Europe. That's the A.I eating software jobs.
kk0nrad 1y ago
yes also this video by techbutmakeitreal talks about this in great detail https://www.youtube.com/watch?v=OBX1YmJL0dw https://www.youtube.com/watch?v=OBX1YmJL0dw
pj_mukh 1y ago
I don’t know about people’s anecdata but this data does not support the claim it makes. Skills aren’t interchangeable and there is no universal hiring pool called “IT” under which you can hire anyone with an engineering degree and so an H1B swap becomes the only answer. There are a large number of pools of specializations that fall into and out of favor with time, with people transitioning between them (as they should!). AI has fallen into favor and some frontend/backend jobs have fallen out. Is this what this data shows? Maybe? We’d never know.
random9749832 1y ago
>There's a reason why 'AI = Actually Indians' is a meme everywhere on the internet, and it isn't racism, it's just people observing the reality around them." Both can be true. The Indian takeover of tech probably also fuels racism.
tonyhart7 1y ago
they fire engineer that have high experience + salary and replace them with engineer that accept lower salary
athrowaway3z 1y ago
I'm more of the belief that: - Layoffs would have happened anyway, natural cycle of shedding underperformers. - the current crop of graduates have too little AI experience from their education, and too little practical experience to be useful in an industry re-arranging itself. - the average developer quality (i.e. boot camps & curriculum) has dropped the value of "a developer" from a manager's perspective, up to the point that an average Actual Indian is competitive (even when accounting for the communication overhead/loss). Combined, that shows fewer people on the books, which they'll obviously frame in the best light; i.e. frugal & good management + innovative AI.
- whywhywhywhy 1y ago
  > the current crop of graduates have too little AI experience from their education Why would you need education to understand how to use the thing that means you don't need education
  - rimunroe 1y ago
    I see fairly few people claiming effectively using AI doesn’t require learning anything. I see a lot of people claiming it requires learning a lot less than traditional methods.
- Esophagus4 1y ago
  > the average developer quality (i.e. boot camps & curriculum) has dropped the value of "a developer" from a manager's perspective If that’s true, I would say it’s because there are simply more developers (therefore more below average developers), not because the good developers are getting less good. If anything, my experience is that the top developers are far better now. For an internship position I’m hiring for, I’m getting resumes with several previous internships, a top 20 university’s CS degree, research lab experience, side projects, (brief) exposure to a myriad of languages and tools, and certifications. For an internship, that’s far above what I or my peers had when we were interns.
- dgfitz 1y ago
  > the current crop of graduates have too little AI experience from their education They probably have more "AI" (read: LLM) experience than working professionals. Not like chatGPT came out last week, they've had it their entire undergrad. Also, that's like saying "they don't have enough experience using google" which is just absurd. You type shit into a text box and see what gets shat back out. If it doesn't work, you adapt it. Prompt engineering is not hard. At best, it takes days-to-weeks learning how to prompt. I think people who are proud of themselves for spending hours carefully crafting prompts are trying to justify the time spent. It isn't hard.
aussieguy1234 1y ago
I always thought a layoff/redundancy meant that not only does the employee loose their position, but the very position itself is eliminated. Which means if the company hires a replacement for the same position including through a sponsored visa, then it's not a genuine layoff. Nonetheless, I'm of the opinion that no single software engineer has yet lost their job to AI. It's simply not there yet.
- scubadude 1y ago
  100% agree. No direct developer job losses from AI. But if it makes you more efficient, there are indirect job losses. Still borderline as to how much nett efficiency it provides. I needed unit-tests pronto, it saved my bacon in crunch time. But then in scaffolding some code in a system I don't know, it's been nothing but trial and error, mostly error.
IshKebab 1y ago
> American employees are being replaced by cheaper H-1B visa workers. Ok someone explain to me exactly how they do that, because when I looked into working in America H-1B is a lottery (with 30% win rate) and it only happens once per year. How exactly do these companies hire people with those constraints?
shrubble 1y ago
I’ve worked with both H1B and offshore Indian workers and there is a huge amount of variance in the quality of their work. This is going to either surprise management or, more likely, have them obfuscate the work done in order to hide certain details of poor performance. The best workers will have to be compensated somehow, so there will be some opaque org chart setup to allow it.
- 4gotunameagain 1y ago
  There's also the huge cultural gap, where in Indian culture gaming the system is considered smart, while in other countries unethical.
- mrweasel 1y ago
  Didn't we already see this in like 15-20 years ago? Companies thought they could outsource project to India, the consulting companies put their best developers on design, prototyping, code samples, interviews, everything, once the contract was in place, the actual work got moved to the meeeh tier developers. Often we also saw management pick the cheapest outsource partner, because otherwise it didn't really make a ton of sense in terms of saving on cost. They then acted very surpised when they got the exact quality they paid for. In the current era, I'm concerned that development will be shifted to the cheapest option, and those people can barely code and will just type everything into an AI chatbot, not understanding the output. We had one consultant from an outsourcing partner come to our offices for two weeks during the start up fase. This dude would just rip through code yelling: No, no, no... NOOOO, this is shit! That was the point where we knew that we found the right outsourcing team.
shuckles 1y ago
This is bad data analysis. It has all kind of issues: using correlation to imply causation without any proposed causal mechanism; various selection bias; no context given for how the companies analyzed were selected; laughably large ranges for data points; etc. One of the worst parts, in my opinion, is how the author misunderstands the meaning of "Actually Indians," which is about the data labeling workforce, not the white collar technical workforce.
- Esophagus4 1y ago
  You know it’s poor analysis when supporting evidence is, “Just look at what people are saying on Blind and this one HN comment”
snowwrestler 1y ago
There is a national cap on the H-1B visas that is hit every single year. Therefore it is very hard to make the case that recent layoffs are being offset by some new surge in H-1Bs.
mig1 1y ago
How do they save money by hiring people in the US with H1B? Most of these companies have compensation policies so it’s not like they can hire a person at the same level and pay a lot less…
hiyer 1y ago
Are h1b workers really cheaper? Don't you have to pay them similar to what an American at the same level would be paid?
- huijzer 1y ago
  As I understand it, a benefit of H-1B visa's employees is that they are more incentivized to not talk back since they would not only lose their job but also lose their visa. I've heard a similar thing with student houses in the Netherlands. House owners often prefer international female students because they are less likely to know their rights or come up for their rights.
- 93po 1y ago
  yes, despite lots of anecdotes of them being paid more or similar. glassdoor has an article saying they're paid higher but their methodology is crap - it doesn't account for the size of the employer (which tend to be much larger on average if they're hiring lots of h1b, and also pay higher) and that h1b workers are likely later in career than entry level. also glassdoor's customers are companies, so of course they're going to toe the line. a better source is Economic Policy Institute, which is associated with left-wing people and so therefore presumably not being anti-immigrant, says: "DOL lets H-1B employers undercut local wages. Sixty percent of H-1B positions certified by the U.S. Department of Labor are assigned wage levels well below the local median wage for the occupation." https://www.epi.org/publication/h-1b-visas-and-prevailing-wage-levels/ https://www.epi.org/publication/h-1b-visas-and-prevailing-wa... and also supports my earlier claim too: "top 30 H-1B employers accounted for more than one in four of all 389,000 H-1B petitions" link has lots of other pretty damning evidence. i think the simplest argument is: 1. these companies have verifiably illegally engaged in wage suppression through gentlemen's agreements between CEOs (including apple/meta/google) 2. companies try as hard as possible to pay their employees as low as possible 3. why in the world would they hire h1bs that cost more when the two points above are true 4. ??? 5. profit
ProllyInfamous 1y ago
If you want to do your own local fact checking, check out http://www.h1bdata.info http://www.h1bdata.info .
catigula 1y ago
I found myself holding some critical feedback back recently of a poor management decision due to caution about my career. My co-worker, who is more matured in his career, with less need for caution, instead took up my complaints (a viewpoint he shared). I imagined how profoundly impossible it would have been for someone on an H1B visa to do this whose immigration status is contingent on their employment. Scary.
UmGuys 1y ago
Everyone who works in tech knows this is true and has been happening for longer than described in the article. Since 2020, it feels like I've trained half of Bangalore on how to do our jobs.
motbus3 1y ago
This text reminds me much of tylervigen.com/spurious-correlation. The data presented is completely insufficient for the claim and based on cherry picked narratives. Anyone can slice and cut numbers the way they want to prove a point. The problem is that hiring an American is so expensive even for America. This will not change. People will need to accept to earn much less than they expected or to have the jobs taken elsewhere. The root cause of the problem is exactly why Chinese clone products and cheaper versions are on the run. They are in inferior quality many times BUT they are way cheaper. American people buy Chinese products because there is not enough money to spend everywhere, so we stretch our definitions of what is acceptable to cover more areas of our lives. These recent changes only seems to be effective but do effectively nothing to solve any of the problems. But let's say magically all immigrants disappeared and companies needed to find replacements, what would happen? - Companies will look to outsource to keep prices at same level - Companies will use AI for whatever is "good enough" - Companies will hire much less capable people and boost them with AI - Companies will hire capable people but pay less because they could have hired someone cheaper - Companies will be forced to pay more for Americans, mean average salary will increase so as every cost, which is basically inflation. The expedite use of "urgent/emergency" measurements is unsettling as-if US was under constant attack and panic and needs to defend themselves against evil outside world to a level where congress and laws can be totally ignored. This need of creating an enemy and the constant threat to solidify a heavy power that goes beyond any limits is in itself a threat. Even more now that a this power is being shared among those who can keep eyes and ears on you and your ears and eyes out of the real problems.
- anovikov 1y ago
  Well to keep salaries high, productivity will have to be increased, is that a surprise?
  - motbus3 1y ago
    If productivity was a problem, why accept paying less for a less capable worker? Same thing as for counterfeiting/cheap china products. They are good enough for the price. Increasing productivity does mean doing more, but doing more with less. Sadly American people already accepts being exploited by having no rights, no benefits and no work life balance (in many cases). The only thing that can get worse is to get less money. But to be honest, I don't think people will get less money. This is politically hard to explain. Things will get more expensive.
mmwako 1y ago
They weren't lying, they just meant AI = Affordable Indians
martin_drapeau 1y ago
A company replacing domestic workers by cheaper H1-B workers. As opposed to a company shutting down because foreign competitors took their marketshare. In either case, domestic company is not competitive. Protectionism won't make the domestic company more competitive.
yegle 1y ago
Genuine question: what's the relation between "beneficiaries approved" and "# of layoffs"? In Prometheus' term, the "beneficiaries approved" is neither a counter nor a gauge. It doesn't tell you the percentage of H1b holders amongst all full time employees and certainly don't include H1b contractors. It also don't include other temporary visa types like L1 or H4, both could be "cheap foreign labors". The "# of layoffs" also doesn't provide much information about how many H1b holders are part of the layoffs. The goal of combining the two columns seems to be giving reader a fuzzy feeling that they are correlated, to support the author's narrative.
bitbasher 1y ago
Considering this blog post author has a whole post talking about how Charlie Kirk had a blood pack under his shirt and the assassination was faked, really makes it difficult to pay attention to anything else they wrote.
thisisit 1y ago
There seems to a misunderstanding of H1B data. "Beneficiaries Approved" as a data point includes: 1. Extensions - H1B is approved for years. Once that period is complete employers need to extend it for another 3 years. 2. Amendments - This will include new job/employer. So, this does mean new H1B employee on the payroll but not exactly new H1B being brought from India. 3. New Hires - Completely new H1B hires. For example, Google's count might include a rolling set of employees requiring H1B extension before they file I-140 petition, H1Bs from Infosys jumping ship to Google for a handsome pay increase and new H1B employees. So, there is no 1:1 comparison. That said, I know Google departments being offshored but mostly due to the software tax code change in US. That has now been rolled back. So, hiring should pick up a bit. As pointed in the H1B fees threads - Corporations always look to maximize benefits. If Accenture and their US clients think that their contracts will be cheaper if they have 3 H1B employees instead of 10 and let rest 7 work out of India, then that is what they will do. H1B fees is just going to make things worse. There is no good way to stop this globalization. That said I think Accenture story is full of air. Its right there in the article > It is not immediately clear how much Accenture intends to invest in setting up the campus.
smik333 1y ago
The video link that's posted on the article is all over the place. IMO, it's hard to state what her conclusion was since it's not concrete at all. In the first half, similar to the post itself, the video argues that since there is H1B approvals for these companies which laid off workers, that means that citizens are being replaced by immigrants. In the second half it states that warehouse workers and other unskilled jobs are getting replaced by AI. And it (with a bit of snark) asks why if AI is so good , they still need to hire H1B workers. But that is exactly the point. H1B workers aren't workhouse workers. They are doing tech work. Saying companies are doing lay off and hiring H1Bs doesn't in any way establishes what kind of workers are getting laid off and what kind of workers are getting hired. As others have pointed it out, the data presented is so insufficient to such a degree that in this particular case, I have to wonder if this is genuine curiosity of the author of the post or if they are trying to justify their beliefs using bad data science.
pigpag 1y ago
[dead]
lisbbb 1y ago
My former employer, a consulting company, just bought out a small dev consulting firm in...wait for it...Guatemala! Yes, Guatemala. Because as you all well know, Guatemala has such a long and cherished history of providing skilled software developers to the world. Oh, oh, wait, no, sorry, it's because they're really, really cheap. The main advantage of Guatemala over India is the time zone, that's it. And $5/hr or whatever those chumps are getting. I don't even think Guatemala has electricity 24/7 in most parts of the country. It's all over for American tech dominance. MBAs sold us all out.
- anothereng 1y ago
  Im sure there are good engineers in Guatemala. Americans simply don't like competition and losing their jobs to foreigners I get that but don't pretend like you're better just because of the country you were born into.
Michael_Keller 1y ago
[dead]

199 points by huijzer 1y ago

Why hire an h1-b when you can shift the entire campus abroad? Google, Microsoft, and Amazon have definitely been doing that. This practice hurts all state side workers, citizen and h-1b alike.

dboreham 1y ago
I know there are stories of H1-B farms but my personal experience is H1 visas were only used to move people who otherwise would be living in a foreign country, being less productive (due to TZ skew) and not paying US taxes, employing people to cut their hair etc, to the US. Me and my former colleagues on those visa must have bought a few F35...
- scarface_74 1y ago
  Well your personal experience doesn’t support the facts — see the WITCH companies.
mapontosevenths 1y ago
Interestingly, the folks I've actually seen replaced by AI the most have been offshore workers in entry level roles. Things like Indian call center roles are low hanging fruit for AI replacement in my recent experience.
zajio1am 1y ago
But helps abroad workers.

lordkrandel 1y ago

Association is no proof of cause. How many h1b were issued the years before AI? Has there been an increase? My bet is: no.

huijzer 1y ago
> How many h1b were issued the years before AI? Has there been an increase? That still wouldn't prove anything per your logic because, as you say, "Association is no proof of cause." But then the question is how could we ever proof anything like this? > How many h1b were issued the years before AI? The point is that companies are lying off people and saying it's AI and the economy while they still hire foreign workers. That suggests they are not telling the full story. Whether or not they are issuing more visas than before does not affect this conclusion.
- lordkrandel 1y ago
  Who cares if you never find proof? This doesn't mean that what is unproven is true. Otherwise, Shiva god may be truely existing. You can't disprove that, right?
MisterSandman 1y ago
This is also assuming that 100% of layoffs are American citizens. A huge chunk of those layoffs are H1B workers themselves.

agentcoops 1y ago

I think this interpretation of the data rests on a misunderstanding of how a firm conceives of headcount. You saw this a lot during Covid layoffs when the press tried to point out the contradiction between hiring X number of people at the same time as firing Y number of people. Wasn't the firm just staying in place? Yet, in fact, the firm had planned to hire Z people that year, a number much greater than X, greater still than X+Y, and likely greater than Z_t-1, i.e. the number hired the previous year. A firm's headcount is a dynamic value that is reflective of a rate of growth: its headcount is going up by x% each year, ideally linear growth or at least not a higher rate of growth than revenues or free cash flow. It requires an accounting perspective on the inflows and outflows of an ongoing process and not just a small slice of data. If you just look at that firm's hiring and firing in a particular year, subtracting the one from the other, you would appear to have shown that the firm was more or less simply replacing the other with the one. Yet, the main question for actually measuring the (firm and investor expected) impact of AI is whether that rate of headcount growth has changed without the stock price going down. That is, is it now proposed by management and accepted by investors that the firm will henceforth require less human labor in order to deliver on the present expected value of future free cash flow. You cannot infer anything about this question from the data shown. There certainly is a separate question of whether a firm has changed its source of new hires, but that also isn't reflected in the data, which just shows approved visas against layoffs. For this, you would need to look at net hires year after year and prove that a growing percent thereof are H1-Bs.

huijzer 1y ago
> It requires an accounting perspective on the inflows and outflows of an ongoing process and not just a small slice of data. Correct, but we don't have that information so we have to rely on a best guess. We're not dealing with code here where everything is known exactly. > Yet, the main question for actually measuring the (firm and investor expected) impact of AI is whether that rate of headcount growth has changed without the stock price going down. Stock price is not necessarily a proxy for headcount either. Some companies have much lower headcounts per stock price (for example, Nvidia). > That is, is it now proposed by management and accepted by investors that the firm will henceforth require less human labor in order to deliver on the present expected value of future free cash flow. You are just talking without thinking now are you? It makes no sense. Human labor and free cash flow are probably correlated, but one isn't a proxy for the other. First you relate stock price to headcount and then you say "that is" and then you suddenly relate cash flow to headcount. Cash flow and stock price are correlated per industry, but it's also a very loose one. > You cannot infer anything about this question from the data shown. False. I only need to infer a tiny bit in order to falsify your extremely strong claim. > For this, you would need to look at net hires year after year and prove that a growing percent thereof are H1-Bs. Yes, this is something concrete that I can agree on. But how would you prove that given that we don't have access to that data?

nis0s 1y ago

The hard truth for Americans (I am talking about Americans because most of these are large cap US companies) is that other countries are cheaper for the following reasons: 1) Companies can find comparably trained and educated workers in non-US countries at cheaper rates because the U.S. education system doesn’t discriminate against less performant students. The education systems in many other countries are competitive at various stages and serve as strict filters against laggards, and there’s less consideration for meeting basic human needs in some countries. But in others there are different options where many students naturally find themselves (like less paying or blue collar work), and if that were to happen in the U.S. then inequality advocates will have a lot to say, and they wouldn’t be wrong. 2) There are a lot of realpolitiks at play when you have a multicultural society as people will often have dual allegiances. A CEO or hiring manager from country X or Y will want to favor their country of origin, naturally. It becomes easier to justify some decisions when the bottom line is also helped by said decisions by hiring cheaper labor. 3) The most expensive thing on the U.S. budget books for employers, employees and the government is social security and Medicare. Some countries simply deny their citizens and workers social security and access to welfare programs, which helps keep the cost of employing them down. 4) Even if a global corporate tax rate is set, it will still make labor in countries that don’t ask for social program contributions much cheaper. 5) The rule of law situation in some of these countries is less than ideal, and so living there can be a hit or miss, which is why H1-Bs are a popular choice for both employers and non-U.S. employees. 6) But still it seems that regional security and stability in some regions has led to permissive environments for unrestricted work. This situation isn’t going to change for nuclear countries, unless they’re like Russia (screwed by economic sanctions and expansionist desires). So in the short-term, unless there’s some agreement to be made between countries for balancing their labor market loads, then economic sanctions seem like a terrible last resort. Addendum, I think the hard truth for large cap companies is that they never should have been allowed to get that big, and in many cases their growth was aided by direct investment by the U.S. government. There’s an interesting lesson here for policy makers about balancing power and investments. Maybe the U.S. should divert future government spending to only its research labs.

deadbabe 1y ago
Go ahead and hire exclusively cheap foreign engineers. See where it gets you.

DebtDeflation 1y ago

You also need to consider the impact of offshoring. Accenture's layoffs of 11,000 workers is listed in the chart, but a few days earlier they announced they were opening a new campus in India and hiring 12,000.

steveBK123 1y ago
Corporate America really thinks we are that stupid
- sfn42 1y ago
  Looking at who you elected (twice!!) I don't know how wrong they are.
  - soulofmischief 1y ago
    Friendly reminder that America is not a single person, but a very large country filled with many different opinions.
    - zenlot 1y ago
      You realize that the large country you're mentioning, with many opinions, still elect 1 president, who represents that large country with different opinions?
      - danaris 1y ago
        You realize that no more than a third of the people in the country actually voted for said president?
        zenlot 1y ago
        You realize, that it doesn't change the fact, that the elected president represents that large country with many opinions?

Havoc 1y ago

I don't see how the data presented necessarily leads to anything remotely close that conclusion with any real confidence. If anything the biggest H1B user - amazon - seems to have done negligible layoffs.

huijzer 1y ago
> I don't see how the data presented necessarily leads to anything remotely close that conclusion with any real confidence. You are not providing a counterargument apart from the fact that "you don't see". Well. Thanks for sharing I guess.
- hshdhdhj4444 1y ago
  And OP is not claiming to provide a counter argument. They’re claiming, correctly, that the presented argument is invalid. They’re not even contesting the conclusions. Only that the information provided in the linked article does not justify the conclusion.
- sm1k 1y ago
  This post is quite similar to tea leaf reading. There isn't a pattern explained by data, it doesn't exist. So there is no counter argument to provide apart from acknowledging the analysis presented in the post is totally bogus.
jeremyjh 1y ago
Yes, without knowing H1B beneficiaries in prior years its impossible to even conclude they've had a net add of them, since they have to be renewed every few years. I'm sure there are some net adds, but we have no idea how many.

dzonga 1y ago

worrying about H1B is looking at the wrong metric. Most companies are past H1B - the game is now offshore stuff. Look at the data: e.g a company like Rippling where a good portion of Software engineering jobs are in India. then Senior / Staff Roles are in the US. Most scale ups startups have similar dynamics. It could be India or Eastern Europe. That's the A.I eating software jobs.

kk0nrad 1y ago

yes also this video by techbutmakeitreal talks about this in great detail https://www.youtube.com/watch?v=OBX1YmJL0dw https://www.youtube.com/watch?v=OBX1YmJL0dw

pj_mukh 1y ago

I don’t know about people’s anecdata but this data does not support the claim it makes. Skills aren’t interchangeable and there is no universal hiring pool called “IT” under which you can hire anyone with an engineering degree and so an H1B swap becomes the only answer. There are a large number of pools of specializations that fall into and out of favor with time, with people transitioning between them (as they should!). AI has fallen into favor and some frontend/backend jobs have fallen out. Is this what this data shows? Maybe? We’d never know.

random9749832 1y ago

>There's a reason why 'AI = Actually Indians' is a meme everywhere on the internet, and it isn't racism, it's just people observing the reality around them." Both can be true. The Indian takeover of tech probably also fuels racism.

tonyhart7 1y ago

they fire engineer that have high experience + salary and replace them with engineer that accept lower salary

athrowaway3z 1y ago

I'm more of the belief that: - Layoffs would have happened anyway, natural cycle of shedding underperformers. - the current crop of graduates have too little AI experience from their education, and too little practical experience to be useful in an industry re-arranging itself. - the average developer quality (i.e. boot camps & curriculum) has dropped the value of "a developer" from a manager's perspective, up to the point that an average Actual Indian is competitive (even when accounting for the communication overhead/loss). Combined, that shows fewer people on the books, which they'll obviously frame in the best light; i.e. frugal & good management + innovative AI.

whywhywhywhy 1y ago
> the current crop of graduates have too little AI experience from their education Why would you need education to understand how to use the thing that means you don't need education
- rimunroe 1y ago
  I see fairly few people claiming effectively using AI doesn’t require learning anything. I see a lot of people claiming it requires learning a lot less than traditional methods.
Esophagus4 1y ago
> the average developer quality (i.e. boot camps & curriculum) has dropped the value of "a developer" from a manager's perspective If that’s true, I would say it’s because there are simply more developers (therefore more below average developers), not because the good developers are getting less good. If anything, my experience is that the top developers are far better now. For an internship position I’m hiring for, I’m getting resumes with several previous internships, a top 20 university’s CS degree, research lab experience, side projects, (brief) exposure to a myriad of languages and tools, and certifications. For an internship, that’s far above what I or my peers had when we were interns.
dgfitz 1y ago
> the current crop of graduates have too little AI experience from their education They probably have more "AI" (read: LLM) experience than working professionals. Not like chatGPT came out last week, they've had it their entire undergrad. Also, that's like saying "they don't have enough experience using google" which is just absurd. You type shit into a text box and see what gets shat back out. If it doesn't work, you adapt it. Prompt engineering is not hard. At best, it takes days-to-weeks learning how to prompt. I think people who are proud of themselves for spending hours carefully crafting prompts are trying to justify the time spent. It isn't hard.

aussieguy1234 1y ago

I always thought a layoff/redundancy meant that not only does the employee loose their position, but the very position itself is eliminated. Which means if the company hires a replacement for the same position including through a sponsored visa, then it's not a genuine layoff. Nonetheless, I'm of the opinion that no single software engineer has yet lost their job to AI. It's simply not there yet.

scubadude 1y ago
100% agree. No direct developer job losses from AI. But if it makes you more efficient, there are indirect job losses. Still borderline as to how much nett efficiency it provides. I needed unit-tests pronto, it saved my bacon in crunch time. But then in scaffolding some code in a system I don't know, it's been nothing but trial and error, mostly error.

IshKebab 1y ago

> American employees are being replaced by cheaper H-1B visa workers. Ok someone explain to me exactly how they do that, because when I looked into working in America H-1B is a lottery (with 30% win rate) and it only happens once per year. How exactly do these companies hire people with those constraints?

shrubble 1y ago

I’ve worked with both H1B and offshore Indian workers and there is a huge amount of variance in the quality of their work. This is going to either surprise management or, more likely, have them obfuscate the work done in order to hide certain details of poor performance. The best workers will have to be compensated somehow, so there will be some opaque org chart setup to allow it.

4gotunameagain 1y ago
There's also the huge cultural gap, where in Indian culture gaming the system is considered smart, while in other countries unethical.
mrweasel 1y ago
Didn't we already see this in like 15-20 years ago? Companies thought they could outsource project to India, the consulting companies put their best developers on design, prototyping, code samples, interviews, everything, once the contract was in place, the actual work got moved to the meeeh tier developers. Often we also saw management pick the cheapest outsource partner, because otherwise it didn't really make a ton of sense in terms of saving on cost. They then acted very surpised when they got the exact quality they paid for. In the current era, I'm concerned that development will be shifted to the cheapest option, and those people can barely code and will just type everything into an AI chatbot, not understanding the output. We had one consultant from an outsourcing partner come to our offices for two weeks during the start up fase. This dude would just rip through code yelling: No, no, no... NOOOO, this is shit! That was the point where we knew that we found the right outsourcing team.

shuckles 1y ago

This is bad data analysis. It has all kind of issues: using correlation to imply causation without any proposed causal mechanism; various selection bias; no context given for how the companies analyzed were selected; laughably large ranges for data points; etc. One of the worst parts, in my opinion, is how the author misunderstands the meaning of "Actually Indians," which is about the data labeling workforce, not the white collar technical workforce.

Esophagus4 1y ago
You know it’s poor analysis when supporting evidence is, “Just look at what people are saying on Blind and this one HN comment”

snowwrestler 1y ago

There is a national cap on the H-1B visas that is hit every single year. Therefore it is very hard to make the case that recent layoffs are being offset by some new surge in H-1Bs.

mig1 1y ago

How do they save money by hiring people in the US with H1B? Most of these companies have compensation policies so it’s not like they can hire a person at the same level and pay a lot less…

hiyer 1y ago

Are h1b workers really cheaper? Don't you have to pay them similar to what an American at the same level would be paid?

huijzer 1y ago
As I understand it, a benefit of H-1B visa's employees is that they are more incentivized to not talk back since they would not only lose their job but also lose their visa. I've heard a similar thing with student houses in the Netherlands. House owners often prefer international female students because they are less likely to know their rights or come up for their rights.
93po 1y ago
yes, despite lots of anecdotes of them being paid more or similar. glassdoor has an article saying they're paid higher but their methodology is crap - it doesn't account for the size of the employer (which tend to be much larger on average if they're hiring lots of h1b, and also pay higher) and that h1b workers are likely later in career than entry level. also glassdoor's customers are companies, so of course they're going to toe the line. a better source is Economic Policy Institute, which is associated with left-wing people and so therefore presumably not being anti-immigrant, says: "DOL lets H-1B employers undercut local wages. Sixty percent of H-1B positions certified by the U.S. Department of Labor are assigned wage levels well below the local median wage for the occupation." https://www.epi.org/publication/h-1b-visas-and-prevailing-wage-levels/ https://www.epi.org/publication/h-1b-visas-and-prevailing-wa... and also supports my earlier claim too: "top 30 H-1B employers accounted for more than one in four of all 389,000 H-1B petitions" link has lots of other pretty damning evidence. i think the simplest argument is: 1. these companies have verifiably illegally engaged in wage suppression through gentlemen's agreements between CEOs (including apple/meta/google) 2. companies try as hard as possible to pay their employees as low as possible 3. why in the world would they hire h1bs that cost more when the two points above are true 4. ??? 5. profit

ProllyInfamous 1y ago

If you want to do your own local fact checking, check out http://www.h1bdata.info http://www.h1bdata.info .

catigula 1y ago

I found myself holding some critical feedback back recently of a poor management decision due to caution about my career. My co-worker, who is more matured in his career, with less need for caution, instead took up my complaints (a viewpoint he shared). I imagined how profoundly impossible it would have been for someone on an H1B visa to do this whose immigration status is contingent on their employment. Scary.

UmGuys 1y ago

Everyone who works in tech knows this is true and has been happening for longer than described in the article. Since 2020, it feels like I've trained half of Bangalore on how to do our jobs.

motbus3 1y ago

This text reminds me much of tylervigen.com/spurious-correlation. The data presented is completely insufficient for the claim and based on cherry picked narratives. Anyone can slice and cut numbers the way they want to prove a point. The problem is that hiring an American is so expensive even for America. This will not change. People will need to accept to earn much less than they expected or to have the jobs taken elsewhere. The root cause of the problem is exactly why Chinese clone products and cheaper versions are on the run. They are in inferior quality many times BUT they are way cheaper. American people buy Chinese products because there is not enough money to spend everywhere, so we stretch our definitions of what is acceptable to cover more areas of our lives. These recent changes only seems to be effective but do effectively nothing to solve any of the problems. But let's say magically all immigrants disappeared and companies needed to find replacements, what would happen? - Companies will look to outsource to keep prices at same level - Companies will use AI for whatever is "good enough" - Companies will hire much less capable people and boost them with AI - Companies will hire capable people but pay less because they could have hired someone cheaper - Companies will be forced to pay more for Americans, mean average salary will increase so as every cost, which is basically inflation. The expedite use of "urgent/emergency" measurements is unsettling as-if US was under constant attack and panic and needs to defend themselves against evil outside world to a level where congress and laws can be totally ignored. This need of creating an enemy and the constant threat to solidify a heavy power that goes beyond any limits is in itself a threat. Even more now that a this power is being shared among those who can keep eyes and ears on you and your ears and eyes out of the real problems.

anovikov 1y ago
Well to keep salaries high, productivity will have to be increased, is that a surprise?
- motbus3 1y ago
  If productivity was a problem, why accept paying less for a less capable worker? Same thing as for counterfeiting/cheap china products. They are good enough for the price. Increasing productivity does mean doing more, but doing more with less. Sadly American people already accepts being exploited by having no rights, no benefits and no work life balance (in many cases). The only thing that can get worse is to get less money. But to be honest, I don't think people will get less money. This is politically hard to explain. Things will get more expensive.

mmwako 1y ago

They weren't lying, they just meant AI = Affordable Indians

martin_drapeau 1y ago

A company replacing domestic workers by cheaper H1-B workers. As opposed to a company shutting down because foreign competitors took their marketshare. In either case, domestic company is not competitive. Protectionism won't make the domestic company more competitive.

yegle 1y ago

Genuine question: what's the relation between "beneficiaries approved" and "# of layoffs"? In Prometheus' term, the "beneficiaries approved" is neither a counter nor a gauge. It doesn't tell you the percentage of H1b holders amongst all full time employees and certainly don't include H1b contractors. It also don't include other temporary visa types like L1 or H4, both could be "cheap foreign labors". The "# of layoffs" also doesn't provide much information about how many H1b holders are part of the layoffs. The goal of combining the two columns seems to be giving reader a fuzzy feeling that they are correlated, to support the author's narrative.

bitbasher 1y ago

Considering this blog post author has a whole post talking about how Charlie Kirk had a blood pack under his shirt and the assassination was faked, really makes it difficult to pay attention to anything else they wrote.

thisisit 1y ago

There seems to a misunderstanding of H1B data. "Beneficiaries Approved" as a data point includes: 1. Extensions - H1B is approved for years. Once that period is complete employers need to extend it for another 3 years. 2. Amendments - This will include new job/employer. So, this does mean new H1B employee on the payroll but not exactly new H1B being brought from India. 3. New Hires - Completely new H1B hires. For example, Google's count might include a rolling set of employees requiring H1B extension before they file I-140 petition, H1Bs from Infosys jumping ship to Google for a handsome pay increase and new H1B employees. So, there is no 1:1 comparison. That said, I know Google departments being offshored but mostly due to the software tax code change in US. That has now been rolled back. So, hiring should pick up a bit. As pointed in the H1B fees threads - Corporations always look to maximize benefits. If Accenture and their US clients think that their contracts will be cheaper if they have 3 H1B employees instead of 10 and let rest 7 work out of India, then that is what they will do. H1B fees is just going to make things worse. There is no good way to stop this globalization. That said I think Accenture story is full of air. Its right there in the article > It is not immediately clear how much Accenture intends to invest in setting up the campus.

smik333 1y ago

The video link that's posted on the article is all over the place. IMO, it's hard to state what her conclusion was since it's not concrete at all. In the first half, similar to the post itself, the video argues that since there is H1B approvals for these companies which laid off workers, that means that citizens are being replaced by immigrants. In the second half it states that warehouse workers and other unskilled jobs are getting replaced by AI. And it (with a bit of snark) asks why if AI is so good , they still need to hire H1B workers. But that is exactly the point. H1B workers aren't workhouse workers. They are doing tech work. Saying companies are doing lay off and hiring H1Bs doesn't in any way establishes what kind of workers are getting laid off and what kind of workers are getting hired. As others have pointed it out, the data presented is so insufficient to such a degree that in this particular case, I have to wonder if this is genuine curiosity of the author of the post or if they are trying to justify their beliefs using bad data science.

pigpag 1y ago

[dead]

lisbbb 1y ago

My former employer, a consulting company, just bought out a small dev consulting firm in...wait for it...Guatemala! Yes, Guatemala. Because as you all well know, Guatemala has such a long and cherished history of providing skilled software developers to the world. Oh, oh, wait, no, sorry, it's because they're really, really cheap. The main advantage of Guatemala over India is the time zone, that's it. And $5/hr or whatever those chumps are getting. I don't even think Guatemala has electricity 24/7 in most parts of the country. It's all over for American tech dominance. MBAs sold us all out.

anothereng 1y ago
Im sure there are good engineers in Guatemala. Americans simply don't like competition and losing their jobs to foreigners I get that but don't pretend like you're better just because of the country you were born into.

Michael_Keller 1y ago

[dead]