13 ms·

It's not any kind of paradox. Structural unemployment happens when the skills of the work force don't match the needs of employers, so there is both unemploymen

by tlb 2y ago

It's not any kind of paradox. Structural unemployment happens when the skills of the work force don't match the needs of employers, so there is both unemployment and difficulty hiring. Structural unemployment is usually high when there's a rapid change in demand for skills, as of course there is in tech. It results in crazy high salaries too. People with machine learning experience are getting 7-figure offers, while people with jQuery experience can't find jobs. As an individual, you can both improve the economy AND make fat stacks by learning the skills that are in high demand. As an employer, you can do better by finding skill sets that aren't in high demand, with enough overlap with what you need that you can retrain. There are a lot of unemployed video game programmers right now, so if you can figure out how to use people with those skills you can hire some smart, energetic people at moderate salaries.

zeroonetwothree 2y ago
Structural unemployment is mostly overstated as a phenomenon. It’s really just a matching problem.
toomuchtodo 2y ago
You can have all the skills and education you want, if employers are unreasonable about their hiring and requirements, you have no control or recourse. You have to be lucky (including knowing someone, ie networking), not necessarily good.
- underdeserver 2y ago
  There is a very large number of employers. If you want to convince me they're all unreasonable, you'll need to show me some pretty compelling evidence.
  - toomuchtodo 2y ago
    What data would satisfy you that employers are being unnecessarily picky? Provide the criteria. Their costs are sunk, they can push candidates through as many cycles as they want until they can find the cheapest unicorns pushed through their pipelines. There is also strong evidence that some employers are cutting anyone with a developed world wage and pushing that work to cheap countries (Google, very publicly, but there are others). My hot take is the labor market was very tight, employers thought they were going to get a deal on folks with the layoffs [1] that took place over the last 2 years, but the labor market still remains tight so they continue to search for "diamonds in the rough" at lower costs (which leads to this mismatch at scale). Offshoring is cost optimization to continue to realize desired profits in a high capital cost macro, and there is probably some knock on effect from software development tax code changes [2]. [1] https://layoffs.fyi/ https://layoffs.fyi/ [2] https://news.ycombinator.com/item?id=39133028 https://news.ycombinator.com/item?id=39133028
  - CoolGuySteve 2y ago
    They're almost all cargo culting the same whiteboard algorithm hiring criteria for otherwise humdrum jobs so yeah, they're all kind of unreasonable. An HR fad can cause structural employment just as easily as bona fide requirements not being met.
    - noobermin 2y ago
      Look at the recent crypto bubble that popped. I'd point to the current potential hype cycle although that's contentious here. For the crypto fad, there was no demonstrated profit beyond speculation, yet lot's of money was poured into it. I have no explanation that can be explained by so-called market mechanisms, it's just chasing hype.
    - Mountain_Skies 2y ago
      The latest fad appears to be forcing all candidates to be vetted by an external recruiter, even if they are already known to people inside the company. Apparently, this is for reasons of "fairness", but it's yet another segment of an increasingly long pipeline full of holes that drop out candidates somewhat randomly rather than based on the actual required attributes to perform the job effectively. The HR filter was already bad enough but now having a recruiter filter before the HR filter just means even more random rejections of candidates who could have filled the role successfully.
    - golergka 2y ago
      I've gone through a lot of job seeking in the last year (two companies in a row I've worked for have closed). Didn't have a single leetcode challenge. And I'm looking for us companies who hire remotely from other countries but pay at the us level, so it's supposedly about the hardest difficulty setting possible.
    - cjbgkagh 2y ago
      They are cargo cutting, I was about to make the same point. I am more likely to get a leetcode quiz from a small company trying to emulate a FAANG company than I am from an interview with an actual FAANG company. Leetcode is part a costly signal given the study that has to be put into it, and part IQ test. I think part of the proliferation of leetcode is due to the illegality (edit: in effect even though not explicitly) of using IQ tests. But if someone could have a certified IQ test they could reuse that one test result for the entire job market improving liquidity. It would be no worse than leetcode as the IQ part of leetcode is in effect already an inefficient and arduous IQ test that people have to take repeatedly. The costly signal was supposed to be university but the academies have sold out their responsibilities. I think the effort to democratize university education, instead of lifting people up, has instead dragged universities down. The reason I think the solution is difficult is that we have to chose between a more fair world where just about anyone can get a degree and IQ tests are illegal but we have to keep taking these leetcode tests. Or a less fair world where an IQ test / SAT score and a university certificate is sufficient.
  - cratermoon 2y ago
    Not all of them, no. But consider: the hiring pipeline looks pretty much the same for all companies? Why is that? Well, they mostly use similar services to manage hiring, and those services all have similar features. But those features aren't great: they've narrowed the hiring pipeline to squeeze all candidates for all jobs down to the same sort of toothpaste that can be squeezed through the pipe. Are the employers unreasonable? No, they are following what is considered HR best practices for attracting and vetting talent. But that process is broken, therefore the majority of companies are struggling to find talent. The process is both homogenized across the industry and fundamentally geared towards preventing false positives - hiring the wrong person. That filters out many people that might have been hired, blowing up the number of false negatives - not hiring the right person. Companies can't find candidates and job seekers can find jobs because the tools and processes for connecting candidates with jobs is both broken and homogeneous across the industry.
  - Slava_Propanei 2y ago
    [dead]
- giantg2 2y ago
  "You have to be lucky (including knowing someone, ie networking), not necessarily good." Yes, and the network effects are largely detrimental to minority groups since networks tend to be biased toward similar people.
  - ryanbytes 2y ago
    [dead]
  - csa 2y ago
    > the network effects are largely detrimental to minority groups “… detrimental to low socioeconomic status groups, within which minorities are over-represented”. FTFY, not to be a pedant, but to highlight the fact that many white people from certain backgrounds also struggle with this, while well-connected minorities largely do not. In fact, connected minorities essentially have super-powered network effects since the demand for minorities who can jump through all of the hiring hoops greatly exceeds the supply.
    - giantg2 2y ago
      I'm not just talking about population level racial minorities. It can be things like disabilities, or even white people in areas/industries/ teams with a majority Asian population. It could even be somewhat based on political or religious beliefs and the other cultural-social background that forms what hobbies and other activities you might meet someone at. Yeah, certain minorities can benefit from programs that value them above others, but the above and beyond results you talk about come after the main connections are made. This doesn't extend to all minority groups either, like those with invisible disabilities (after all, it's mostly about how things look).
    - WarOnPrivacy 2y ago
      > In fact, connected minorities essentially have super-powered network... ...in isolated pockets where those things exist. Outside of those places, minorities have their slider set to something-other-than the easiest difficulty level. As a white guy in generally good condition, my slider was slid all the way to easy. My road was still super hard - but it was easier than it was for women, brown people, etc. In 1990 I moved to FL and ran headfirst into no connections=no work. I did the Make Your Own Luck thing. Hundreds of brief introductions led to a few relationships. A few of those became potential leads. It was a years long process. The eventual successes hinged on a few, key connections with people who saw something familiar. If I hadn't been familiar, I wouldn't have been remembered past the introduction.
- coldpie 2y ago
  > You have to be lucky (including knowing someone, ie networking) I don't think it's right to consider networking/knowing someone to be luck, exactly. Networking is a skill and a tool, and has consistently been presented as that for my entire career. As a result, I go out of my way to try to make those connections. I go in to the office, I have lunch and drinks with coworkers, I say Hi to people I recognize from chat at large meetings, I make friendly conversations on chat and email. When someone I have made a connection like that with leaves, I say goodbye and pass them my personal email address. It is difficult, I'm not an outgoing person and I have to make conscious effort to start those connections. But it has paid off, I've received several job pings over the years, and when I was ready to move on from a job recently I took up one of those pings on the offer. It wasn't luck, it was a skill and resource I specifically put work in to cultivate over years. Cold-applying is the absolute hardest route to getting hired. A big part of your career is doing what you can to make that your last resort when job hunting.
  - giantg2 2y ago
    "Cold-applying is the absolute hardest route to getting hired. A big part of your career is doing what you can to make that your last resort when job hunting." Cold applying should be the most fair and introduce the least bias. My entire life has been cold applying.
    - pdimitar 2y ago
      It should be, I agree, but my entire career of 22 years has shown me that it's not. And I've always cold-applied as well. Only had 1-2 contracts where former colleagues recommended me.
atoav 2y ago
As an employer you could as a first step even just check if the skill you look for could: A) be unrealistic. A need for 8 years of experience in a framework that existed for a year is a need zero people can fulfill. B) if your needs are so special that such an employee doesn't exist. I recall the days when companies would train people to do a job. C) The job description is realistic but the people who you seek to attract won't do the job for that salary in that location for that company.
- itronitron 2y ago
  I'd also advise employers to not require skills in any application that could be learned in a week or two of use on the job.
dbdoskey 2y ago
This is definitely it. The problem is that the job market has positioned it that coming with "gumption" and no real world skills (in past positions) there is a low chance that the company will take a risk on you. So doing these shifts are extremely hard.
- nicolas_t 2y ago
  My experience has always been that if you take 6 months to contribute to open source centered around the new thing you want to work on, you'll quickly meet people that will allow you to get a job.
  - ohthatsnotright 2y ago
    Unless you're looking to be hired in the US and live in a US territory, in which case you'll be hard pressed to find any employer who will hire you.
    - bishbosh 2y ago
      Interesting, what makes you say that?
      - ohthatsnotright 2y ago
        Personal experience trying to get hired as a US citizen in a US territory. Most companies simply reject without any sort of interview and when I ask (and get a reply) it's always because of location.
surajrmal 2y ago
It's also not helpful when employers refuse to train and demand folks with the skills they want. If the supply of folks you want doesn't't exist, you need to do your part to grow them.
- ralphc 2y ago
  I'll throw in my personal example from 2012 when I was laid off. Senior dev, about 30 years experience at the time. A friend knew of an opening somewhere but they wanted GWT experience, I didn't have it so they wouldn't talk to me. 6 weeks later they were still looking. Like I couldn't learn GWT in six weeks? If they hired any decent dev after a good general screening and gave him the docs they could self-learn GWT in less time than that.
  - shagie 2y ago
    The risk is that they hire someone who doesn't know GWT and six weeks later (and six months later) they still don't know it. The ability to do self directed learning of a new skill / technology is something that is difficult to find (a lot of developers that I know don't know how to learn a new version of Java or framework and are still writing code exactly as they did when they graduated college... and they could put down "Java developer - 5 years experience" and apply for senior positions.) There are a lot of devs who appear decent with current tech, but lack the ability to learn new tech without someone hand holding them for several months.
    - giantg2 2y ago
      The people I have meet who were slower to learn things generally had factors that could be overcome. Burnout is a huge one. If I'm being told to learn this new tech every 3-6 months and not given enough time to become an expert in the thing I previously learned, then why should I put in the effort? Its not a building block in my career. This is how it is for me (plus a disability). It never benefits me, so where's the incentive?
      - alephnerd 2y ago
banannaise 2y ago
Structural unemployment happens when the perceived skills of the workforce don't match the perceived needs of employers, so there is both unemployment and difficulty hiring. A lot of engineers will tell you that companies are often expecting both too much (precise types of experience) and the wrong things (leetcode-type challenges, unnecessarily specific knowledge...), which leads them to pass over perfectly qualified candidates.
- SideQuark 2y ago
  Quite often perception of reality matches actual reality. I've found it extremely rare to the point of almost being non-existent that what engineers think a company needs or should do is accurate compared to what a company needs or should do.
  - avmich 2y ago
    Probably to what company thinks the company needs or should do? It's rare - very - when a company really needs to do something unusual, which engineers wouldn't expect.
- pjc50 2y ago
  > Structural unemployment happens when the perceived skills of the workforce don't match the perceived needs of employers https://en.wikipedia.org/wiki/The_Market_for_Lemons https://en.wikipedia.org/wiki/The_Market_for_Lemons : when "buyers" cannot distinguish between good and bad due to informational issues, even the good products (or in this case employees) cannot find buyers. Market failure. It might be improvable by the relevant parties getting together and agreeing on a curriculum and exams so people only have to take a test once, for example. But I think that's a long way off. Of course, ChatGPT makes this a lot worse since it reduces the cost of fakes on both sides: fake applications and fake job adverts.
  - pseudocomposer 2y ago
    Unions have historically solved this problem. Non-union accreditation programs don’t seem to have the same effect, though in some industries they do have a positive effect on salary.
bugglebeetle 2y ago
> People with machine learning experience are getting 7-figure offers I’m paid fairly well, but not this much, so I think the threshold is a bit higher than this.
kerkeslager 2y ago
> Structural unemployment is usually high when there's a rapid change in demand for skills, as of course there is in tech. It results in crazy high salaries too. People with machine learning experience are getting 7-figure offers, while people with jQuery experience can't find jobs. Another cause of structural unemployment is workforces aging into retirement, and again there's stacks of money to be made filling this gap. This applies to Cobol developers. There are a ton of "too big to fail or rewrite" applications written in Cobol, particularly in finance. The people who wrote them are retiring. Finance is generally a great industry to have your hand in the pockets of anyway. A friend of mine has been doing this since shortly after he graduated college in 2010. He was able to start freelancing a lot earlier than I was, and his billing rate in 2012 was something like $125/hour (only slightly less than my current billing rate). Last time I talked to him he was billing $300/hour and had a 7 month backlog of work. I'm not good at keeping up with people, so we haven't talked in a few years, but it would not surprise me if he's billing $350/hour now. I've largely prioritized time/freedom over stability/money in recent years, but if I ever had a kid and needed stability/money again, I'd be transitioning toward that.
SJC_Hacker 2y ago
> People with machine learning experience are getting 7-figure offer This sounds like a gross exaggeration. Unless they did something like start a company. For an IC at least, even at FAANG it would be bit crazy to start at more than ~$400k. If a project lead, then its obviously more and you can make bank as you move up the ladder.
- nfriedly 2y ago
  They're probably including equity, not just cash. Equity can be worth significantly more than the base pay at higher levels.
logicallee 2y ago
>"it's not any kind of paradox" It is a paradox, because it is a market where people basically only do one salaried job at a time, meaning they have a fixed 40-60 hours to sell and it shouldn't take long to sell it if they want to move on to another job. (Here "selling" means you've been hired for your next job, since you just took the 40 hours you had available every week starting next month and sold it for money.) Within reason, people can be retrained to adjacent jobs or basically could do exactly what an employer wants within a low number of weeks. People know what kind of jobs they're capable of or it can be tested for quickly. Instead there are candidates who spend a year both training themselves and applying for jobs unsuccessfully, while employers spend MORE than the same amount of time on that, if you add up the amount of time recruiters are putting in on behalf of employers and HR puts in and employers put into interviews. It is a market failure for people to want to work, for employers to want to hire them, for them to be able to do the work, but for them not to actually be doing the work. It would be kind of like if people were desperate for any of either tomatoes, carrots, wheat, rice, potatoes, or really anything, whatever they get is fine for them since they can look up recipes for that and make it with a tiny investment of time (this is the demand side), so they are putting in ads for "Want any of: rice, carrots, wheat, potatoes, tomatoes" and willing to pay for it! Just put up a hundred ads, a hundred unfilled jobs. On the supply side meanwhile you have a farmer who owns all the necessary land (hours in the day) and has ALL the equipment (a computer, Internet, transportation, clothes, anything anyone needs to start working) and is really willing to produce anything and can do so on that arable land! There are no capital requirements. Great. So we have the demand side taken care of, people want quite a few different types of food and supply side taken care of, the farmer is willing to farm just about anything. But, for some reason, what ends up happening isn't that the fields are filled, it's that there are farmers spending all day every day selling and nobody is buying. And instead of saying, here try this sample, and it's fine, you have a weird request for "What exactly had grown on these fields since the beginning of time" (this is called a resume) and then you have buyers inundated with answers to their bizarre requests for "what has grown on these fields since the beginning of time" but the buyers still aren't getting the food and the fields are empty, the farmers are trying unsuccessfully to sell all day and the buyers are somehow inundated with sales pitches but aren't actually buying anything, while burning time on putting up advertisements of all the food they want to buy. Why is this happening? How can people not be working but want to work, when they can train into being productive workers within a few weeks and are more than willing to do so? It would be like if shoppers spent days at a time at farmer's markets going from stand to stand and buying nothing, and farmers spent days at a time standing at the farmer's market with buyer after buyer asking them questions about the history of their land and not buying any damn potatoes. It's a potato lady you want it or not? If not step aside so someone who actually wants it can see it. p.s. half of the farmers are asking other farmers to endorse them on linkedin. Buyers ignore this noise completely. The job market is highly dysfunctional.
smsm42 2y ago
It's not the skills problem. Looking for a job, I know that my skillset would match the employer - in fact, thousands of employers likely - because I have successfully done such jobs in the past. But I have no way to prove it to a potential employer than going through an expensive and friction-laden process which could fail at any stage for a reason totally unrelated to the main question of qualifications - like some random interviewer not liking my answer to a weird question like "tell me about your three biggest failures" or getting some obscure pet detail of a language wrong when nobody actually uses that corner precisely because of how hard it to get right, or not solving irrelevant leetcode puzzle fast enough. Immense effort on both sides is spent to go through these dances - and almost all of it is wasted. I wish there were some good solution for it. So far there's none.
g9yuayon 2y ago
> Structural unemployment happens when the skills of the work force don't match the needs of employers, so there is both unemployment and difficulty hiring. Any study on why there would be this kind of mismatch in the tech industry? I'd imagine that tech industry is the most transparently competitive and least regulated sector so there should be a reasonable balance between the supply of skills and the demand. In addition, there are so much resource that teaches people all kinds of tech skills, so people should be able to move from one skill set to another if really needed.
- shagie 2y ago
  The easy problems are solved. Crud with the preferred language and framework is a known quantity and can be "ok here we are and go" and you can get a boot camp or new grad and have them do that. Let's say that we need to solve a new problem. The shop language is Scala. Who can learn how to program in Scala? Can the boot camp who learned how to do NodeJS be trained in Scala? If so, how long will it take? If we spend a year to get them to being an acceptable Scala dev, will they get a new job immediately? If we are paying a "we are training you low rates" can we get any applicants? If we are paying a "we assume you to be a skilled Scala dev in a year and we expect to get 2-3 years of productive work beyond training" rate and they leave immediately after getting them sufficiently trained... what then? As many developers see themselves as a "Java developer" or "React front end developer" rather than a software engineer, the mismatch is akin to asking "why can't we get plumbers to do electrical work?" The people who are Java developers or React front end developers are skilled in the trade rather than as a knowledge worker in software engineering.
  - g9yuayon 2y ago
    This sounds like the trend is good for the incumbents? That is, it's hard for the supply to catch up with the demand in broad scope, so people with solid experience will be marketable for a while.
    - shagie 2y ago
      It is good for people who can, as knowledge workers ( https://en.wikipedia.org/wiki/Knowledge_worker https://en.wikipedia.org/wiki/Knowledge_worker ) are able to continue to gain and market their ability to learn new things within the domain and apply their knowledge. There are incumbents who are not able (don't have the skill scaffolding to support a broader understanding) or are unwilling to spend time to continue to be able to work or are a "follow instructions". I've quipped that... A junior dev follows instructions A mid dev completes tasks A senior dev solves problems In this simplified model, I know a lot of junior devs (even with a decade of experience) who for whatever reason are unable to partition their software development tasks themselves to be able to be given a task to complete. Likewise, I know a number of devs who can be given tasks but if you give a broader "this is a problem that the organization needs to solve" are hopelessly lost and apply and reapply the patterns that they have learned to try to make the problem fit into a solution they know. I do not believe that everyone has the skills / aptitude to move from one to the next. Those that do will have the ability to navigate the new market. Those that lack the ability to move up will eventually be overtaken either by younger (cheaper) workers, or improving automation.
UncleOxidant 2y ago
> Structural unemployment happens when the skills of the work force don't match the needs of employers I know someone who was laid off from an AI group at a large chipmaker last year- his group was eliminated. But because he wasn't doing LLMs, he was doing CNNs for vision, it took him about a year to find a new gig teaching Machine Learning classes - not his preferred gig or level of pay. He could easily have transitioned from CNNs to LLMs but the algorithms sorting the resumes didn't seem to think so. Personally, I've worked on optimizing code to make machine learning algorithms run faster/more efficient - also compiler/LLVM work. When I was laid off over a year ago I'd apply for things and it was mostly crickets - had a few interviews which often ended up in ghosting. I gave up looking and am working on a startup with some other who've found themselves in a similar situation - no pay at this point, only equity. Anyway, I think your premise is flawed: AI/ML skills are in demand, but even people with those skills are having a hard time getting back into the industry. Maybe the companies doing AI/ML are spending so much money on GPUs now that they can't actually afford to hire people?

by tlb 2y ago

zeroonetwothree 2y ago

Structural unemployment is mostly overstated as a phenomenon. It’s really just a matching problem.

toomuchtodo 2y ago

You can have all the skills and education you want, if employers are unreasonable about their hiring and requirements, you have no control or recourse. You have to be lucky (including knowing someone, ie networking), not necessarily good.

underdeserver 2y ago
There is a very large number of employers. If you want to convince me they're all unreasonable, you'll need to show me some pretty compelling evidence.
- toomuchtodo 2y ago
  What data would satisfy you that employers are being unnecessarily picky? Provide the criteria. Their costs are sunk, they can push candidates through as many cycles as they want until they can find the cheapest unicorns pushed through their pipelines. There is also strong evidence that some employers are cutting anyone with a developed world wage and pushing that work to cheap countries (Google, very publicly, but there are others). My hot take is the labor market was very tight, employers thought they were going to get a deal on folks with the layoffs [1] that took place over the last 2 years, but the labor market still remains tight so they continue to search for "diamonds in the rough" at lower costs (which leads to this mismatch at scale). Offshoring is cost optimization to continue to realize desired profits in a high capital cost macro, and there is probably some knock on effect from software development tax code changes [2]. [1] https://layoffs.fyi/ https://layoffs.fyi/ [2] https://news.ycombinator.com/item?id=39133028 https://news.ycombinator.com/item?id=39133028
- CoolGuySteve 2y ago
  They're almost all cargo culting the same whiteboard algorithm hiring criteria for otherwise humdrum jobs so yeah, they're all kind of unreasonable. An HR fad can cause structural employment just as easily as bona fide requirements not being met.
  - noobermin 2y ago
    Look at the recent crypto bubble that popped. I'd point to the current potential hype cycle although that's contentious here. For the crypto fad, there was no demonstrated profit beyond speculation, yet lot's of money was poured into it. I have no explanation that can be explained by so-called market mechanisms, it's just chasing hype.
  - Mountain_Skies 2y ago
    The latest fad appears to be forcing all candidates to be vetted by an external recruiter, even if they are already known to people inside the company. Apparently, this is for reasons of "fairness", but it's yet another segment of an increasingly long pipeline full of holes that drop out candidates somewhat randomly rather than based on the actual required attributes to perform the job effectively. The HR filter was already bad enough but now having a recruiter filter before the HR filter just means even more random rejections of candidates who could have filled the role successfully.
  - golergka 2y ago
    I've gone through a lot of job seeking in the last year (two companies in a row I've worked for have closed). Didn't have a single leetcode challenge. And I'm looking for us companies who hire remotely from other countries but pay at the us level, so it's supposedly about the hardest difficulty setting possible.
  - cjbgkagh 2y ago
    They are cargo cutting, I was about to make the same point. I am more likely to get a leetcode quiz from a small company trying to emulate a FAANG company than I am from an interview with an actual FAANG company. Leetcode is part a costly signal given the study that has to be put into it, and part IQ test. I think part of the proliferation of leetcode is due to the illegality (edit: in effect even though not explicitly) of using IQ tests. But if someone could have a certified IQ test they could reuse that one test result for the entire job market improving liquidity. It would be no worse than leetcode as the IQ part of leetcode is in effect already an inefficient and arduous IQ test that people have to take repeatedly. The costly signal was supposed to be university but the academies have sold out their responsibilities. I think the effort to democratize university education, instead of lifting people up, has instead dragged universities down. The reason I think the solution is difficult is that we have to chose between a more fair world where just about anyone can get a degree and IQ tests are illegal but we have to keep taking these leetcode tests. Or a less fair world where an IQ test / SAT score and a university certificate is sufficient.
- cratermoon 2y ago
  Not all of them, no. But consider: the hiring pipeline looks pretty much the same for all companies? Why is that? Well, they mostly use similar services to manage hiring, and those services all have similar features. But those features aren't great: they've narrowed the hiring pipeline to squeeze all candidates for all jobs down to the same sort of toothpaste that can be squeezed through the pipe. Are the employers unreasonable? No, they are following what is considered HR best practices for attracting and vetting talent. But that process is broken, therefore the majority of companies are struggling to find talent. The process is both homogenized across the industry and fundamentally geared towards preventing false positives - hiring the wrong person. That filters out many people that might have been hired, blowing up the number of false negatives - not hiring the right person. Companies can't find candidates and job seekers can find jobs because the tools and processes for connecting candidates with jobs is both broken and homogeneous across the industry.
- Slava_Propanei 2y ago
  [dead]
giantg2 2y ago
"You have to be lucky (including knowing someone, ie networking), not necessarily good." Yes, and the network effects are largely detrimental to minority groups since networks tend to be biased toward similar people.
- ryanbytes 2y ago
  [dead]
- csa 2y ago
  > the network effects are largely detrimental to minority groups “… detrimental to low socioeconomic status groups, within which minorities are over-represented”. FTFY, not to be a pedant, but to highlight the fact that many white people from certain backgrounds also struggle with this, while well-connected minorities largely do not. In fact, connected minorities essentially have super-powered network effects since the demand for minorities who can jump through all of the hiring hoops greatly exceeds the supply.
  - giantg2 2y ago
    I'm not just talking about population level racial minorities. It can be things like disabilities, or even white people in areas/industries/ teams with a majority Asian population. It could even be somewhat based on political or religious beliefs and the other cultural-social background that forms what hobbies and other activities you might meet someone at. Yeah, certain minorities can benefit from programs that value them above others, but the above and beyond results you talk about come after the main connections are made. This doesn't extend to all minority groups either, like those with invisible disabilities (after all, it's mostly about how things look).
  - WarOnPrivacy 2y ago
    > In fact, connected minorities essentially have super-powered network... ...in isolated pockets where those things exist. Outside of those places, minorities have their slider set to something-other-than the easiest difficulty level. As a white guy in generally good condition, my slider was slid all the way to easy. My road was still super hard - but it was easier than it was for women, brown people, etc. In 1990 I moved to FL and ran headfirst into no connections=no work. I did the Make Your Own Luck thing. Hundreds of brief introductions led to a few relationships. A few of those became potential leads. It was a years long process. The eventual successes hinged on a few, key connections with people who saw something familiar. If I hadn't been familiar, I wouldn't have been remembered past the introduction.
coldpie 2y ago
> You have to be lucky (including knowing someone, ie networking) I don't think it's right to consider networking/knowing someone to be luck, exactly. Networking is a skill and a tool, and has consistently been presented as that for my entire career. As a result, I go out of my way to try to make those connections. I go in to the office, I have lunch and drinks with coworkers, I say Hi to people I recognize from chat at large meetings, I make friendly conversations on chat and email. When someone I have made a connection like that with leaves, I say goodbye and pass them my personal email address. It is difficult, I'm not an outgoing person and I have to make conscious effort to start those connections. But it has paid off, I've received several job pings over the years, and when I was ready to move on from a job recently I took up one of those pings on the offer. It wasn't luck, it was a skill and resource I specifically put work in to cultivate over years. Cold-applying is the absolute hardest route to getting hired. A big part of your career is doing what you can to make that your last resort when job hunting.
- giantg2 2y ago
  "Cold-applying is the absolute hardest route to getting hired. A big part of your career is doing what you can to make that your last resort when job hunting." Cold applying should be the most fair and introduce the least bias. My entire life has been cold applying.
  - pdimitar 2y ago
    It should be, I agree, but my entire career of 22 years has shown me that it's not. And I've always cold-applied as well. Only had 1-2 contracts where former colleagues recommended me.

atoav 2y ago

As an employer you could as a first step even just check if the skill you look for could: A) be unrealistic. A need for 8 years of experience in a framework that existed for a year is a need zero people can fulfill. B) if your needs are so special that such an employee doesn't exist. I recall the days when companies would train people to do a job. C) The job description is realistic but the people who you seek to attract won't do the job for that salary in that location for that company.

itronitron 2y ago
I'd also advise employers to not require skills in any application that could be learned in a week or two of use on the job.

dbdoskey 2y ago

This is definitely it. The problem is that the job market has positioned it that coming with "gumption" and no real world skills (in past positions) there is a low chance that the company will take a risk on you. So doing these shifts are extremely hard.

nicolas_t 2y ago
My experience has always been that if you take 6 months to contribute to open source centered around the new thing you want to work on, you'll quickly meet people that will allow you to get a job.
- ohthatsnotright 2y ago
  Unless you're looking to be hired in the US and live in a US territory, in which case you'll be hard pressed to find any employer who will hire you.
  - bishbosh 2y ago
    Interesting, what makes you say that?
    - ohthatsnotright 2y ago
      Personal experience trying to get hired as a US citizen in a US territory. Most companies simply reject without any sort of interview and when I ask (and get a reply) it's always because of location.

surajrmal 2y ago

It's also not helpful when employers refuse to train and demand folks with the skills they want. If the supply of folks you want doesn't't exist, you need to do your part to grow them.

ralphc 2y ago
I'll throw in my personal example from 2012 when I was laid off. Senior dev, about 30 years experience at the time. A friend knew of an opening somewhere but they wanted GWT experience, I didn't have it so they wouldn't talk to me. 6 weeks later they were still looking. Like I couldn't learn GWT in six weeks? If they hired any decent dev after a good general screening and gave him the docs they could self-learn GWT in less time than that.
- shagie 2y ago
  The risk is that they hire someone who doesn't know GWT and six weeks later (and six months later) they still don't know it. The ability to do self directed learning of a new skill / technology is something that is difficult to find (a lot of developers that I know don't know how to learn a new version of Java or framework and are still writing code exactly as they did when they graduated college... and they could put down "Java developer - 5 years experience" and apply for senior positions.) There are a lot of devs who appear decent with current tech, but lack the ability to learn new tech without someone hand holding them for several months.
  - giantg2 2y ago
    The people I have meet who were slower to learn things generally had factors that could be overcome. Burnout is a huge one. If I'm being told to learn this new tech every 3-6 months and not given enough time to become an expert in the thing I previously learned, then why should I put in the effort? Its not a building block in my career. This is how it is for me (plus a disability). It never benefits me, so where's the incentive?
    - alephnerd 2y ago

banannaise 2y ago

Structural unemployment happens when the perceived skills of the workforce don't match the perceived needs of employers, so there is both unemployment and difficulty hiring. A lot of engineers will tell you that companies are often expecting both too much (precise types of experience) and the wrong things (leetcode-type challenges, unnecessarily specific knowledge...), which leads them to pass over perfectly qualified candidates.

SideQuark 2y ago
Quite often perception of reality matches actual reality. I've found it extremely rare to the point of almost being non-existent that what engineers think a company needs or should do is accurate compared to what a company needs or should do.
- avmich 2y ago
  Probably to what company thinks the company needs or should do? It's rare - very - when a company really needs to do something unusual, which engineers wouldn't expect.
pjc50 2y ago
> Structural unemployment happens when the perceived skills of the workforce don't match the perceived needs of employers https://en.wikipedia.org/wiki/The_Market_for_Lemons https://en.wikipedia.org/wiki/The_Market_for_Lemons : when "buyers" cannot distinguish between good and bad due to informational issues, even the good products (or in this case employees) cannot find buyers. Market failure. It might be improvable by the relevant parties getting together and agreeing on a curriculum and exams so people only have to take a test once, for example. But I think that's a long way off. Of course, ChatGPT makes this a lot worse since it reduces the cost of fakes on both sides: fake applications and fake job adverts.
- pseudocomposer 2y ago
  Unions have historically solved this problem. Non-union accreditation programs don’t seem to have the same effect, though in some industries they do have a positive effect on salary.

bugglebeetle 2y ago

> People with machine learning experience are getting 7-figure offers I’m paid fairly well, but not this much, so I think the threshold is a bit higher than this.

kerkeslager 2y ago

> Structural unemployment is usually high when there's a rapid change in demand for skills, as of course there is in tech. It results in crazy high salaries too. People with machine learning experience are getting 7-figure offers, while people with jQuery experience can't find jobs. Another cause of structural unemployment is workforces aging into retirement, and again there's stacks of money to be made filling this gap. This applies to Cobol developers. There are a ton of "too big to fail or rewrite" applications written in Cobol, particularly in finance. The people who wrote them are retiring. Finance is generally a great industry to have your hand in the pockets of anyway. A friend of mine has been doing this since shortly after he graduated college in 2010. He was able to start freelancing a lot earlier than I was, and his billing rate in 2012 was something like $125/hour (only slightly less than my current billing rate). Last time I talked to him he was billing $300/hour and had a 7 month backlog of work. I'm not good at keeping up with people, so we haven't talked in a few years, but it would not surprise me if he's billing $350/hour now. I've largely prioritized time/freedom over stability/money in recent years, but if I ever had a kid and needed stability/money again, I'd be transitioning toward that.

SJC_Hacker 2y ago

> People with machine learning experience are getting 7-figure offer This sounds like a gross exaggeration. Unless they did something like start a company. For an IC at least, even at FAANG it would be bit crazy to start at more than ~$400k. If a project lead, then its obviously more and you can make bank as you move up the ladder.

nfriedly 2y ago
They're probably including equity, not just cash. Equity can be worth significantly more than the base pay at higher levels.

logicallee 2y ago

>"it's not any kind of paradox" It is a paradox, because it is a market where people basically only do one salaried job at a time, meaning they have a fixed 40-60 hours to sell and it shouldn't take long to sell it if they want to move on to another job. (Here "selling" means you've been hired for your next job, since you just took the 40 hours you had available every week starting next month and sold it for money.) Within reason, people can be retrained to adjacent jobs or basically could do exactly what an employer wants within a low number of weeks. People know what kind of jobs they're capable of or it can be tested for quickly. Instead there are candidates who spend a year both training themselves and applying for jobs unsuccessfully, while employers spend MORE than the same amount of time on that, if you add up the amount of time recruiters are putting in on behalf of employers and HR puts in and employers put into interviews. It is a market failure for people to want to work, for employers to want to hire them, for them to be able to do the work, but for them not to actually be doing the work. It would be kind of like if people were desperate for any of either tomatoes, carrots, wheat, rice, potatoes, or really anything, whatever they get is fine for them since they can look up recipes for that and make it with a tiny investment of time (this is the demand side), so they are putting in ads for "Want any of: rice, carrots, wheat, potatoes, tomatoes" and willing to pay for it! Just put up a hundred ads, a hundred unfilled jobs. On the supply side meanwhile you have a farmer who owns all the necessary land (hours in the day) and has ALL the equipment (a computer, Internet, transportation, clothes, anything anyone needs to start working) and is really willing to produce anything and can do so on that arable land! There are no capital requirements. Great. So we have the demand side taken care of, people want quite a few different types of food and supply side taken care of, the farmer is willing to farm just about anything. But, for some reason, what ends up happening isn't that the fields are filled, it's that there are farmers spending all day every day selling and nobody is buying. And instead of saying, here try this sample, and it's fine, you have a weird request for "What exactly had grown on these fields since the beginning of time" (this is called a resume) and then you have buyers inundated with answers to their bizarre requests for "what has grown on these fields since the beginning of time" but the buyers still aren't getting the food and the fields are empty, the farmers are trying unsuccessfully to sell all day and the buyers are somehow inundated with sales pitches but aren't actually buying anything, while burning time on putting up advertisements of all the food they want to buy. Why is this happening? How can people not be working but want to work, when they can train into being productive workers within a few weeks and are more than willing to do so? It would be like if shoppers spent days at a time at farmer's markets going from stand to stand and buying nothing, and farmers spent days at a time standing at the farmer's market with buyer after buyer asking them questions about the history of their land and not buying any damn potatoes. It's a potato lady you want it or not? If not step aside so someone who actually wants it can see it. p.s. half of the farmers are asking other farmers to endorse them on linkedin. Buyers ignore this noise completely. The job market is highly dysfunctional.

smsm42 2y ago

It's not the skills problem. Looking for a job, I know that my skillset would match the employer - in fact, thousands of employers likely - because I have successfully done such jobs in the past. But I have no way to prove it to a potential employer than going through an expensive and friction-laden process which could fail at any stage for a reason totally unrelated to the main question of qualifications - like some random interviewer not liking my answer to a weird question like "tell me about your three biggest failures" or getting some obscure pet detail of a language wrong when nobody actually uses that corner precisely because of how hard it to get right, or not solving irrelevant leetcode puzzle fast enough. Immense effort on both sides is spent to go through these dances - and almost all of it is wasted. I wish there were some good solution for it. So far there's none.

g9yuayon 2y ago

> Structural unemployment happens when the skills of the work force don't match the needs of employers, so there is both unemployment and difficulty hiring. Any study on why there would be this kind of mismatch in the tech industry? I'd imagine that tech industry is the most transparently competitive and least regulated sector so there should be a reasonable balance between the supply of skills and the demand. In addition, there are so much resource that teaches people all kinds of tech skills, so people should be able to move from one skill set to another if really needed.

shagie 2y ago
The easy problems are solved. Crud with the preferred language and framework is a known quantity and can be "ok here we are and go" and you can get a boot camp or new grad and have them do that. Let's say that we need to solve a new problem. The shop language is Scala. Who can learn how to program in Scala? Can the boot camp who learned how to do NodeJS be trained in Scala? If so, how long will it take? If we spend a year to get them to being an acceptable Scala dev, will they get a new job immediately? If we are paying a "we are training you low rates" can we get any applicants? If we are paying a "we assume you to be a skilled Scala dev in a year and we expect to get 2-3 years of productive work beyond training" rate and they leave immediately after getting them sufficiently trained... what then? As many developers see themselves as a "Java developer" or "React front end developer" rather than a software engineer, the mismatch is akin to asking "why can't we get plumbers to do electrical work?" The people who are Java developers or React front end developers are skilled in the trade rather than as a knowledge worker in software engineering.
- g9yuayon 2y ago
  This sounds like the trend is good for the incumbents? That is, it's hard for the supply to catch up with the demand in broad scope, so people with solid experience will be marketable for a while.
  - shagie 2y ago
    It is good for people who can, as knowledge workers ( https://en.wikipedia.org/wiki/Knowledge_worker https://en.wikipedia.org/wiki/Knowledge_worker ) are able to continue to gain and market their ability to learn new things within the domain and apply their knowledge. There are incumbents who are not able (don't have the skill scaffolding to support a broader understanding) or are unwilling to spend time to continue to be able to work or are a "follow instructions". I've quipped that... A junior dev follows instructions A mid dev completes tasks A senior dev solves problems In this simplified model, I know a lot of junior devs (even with a decade of experience) who for whatever reason are unable to partition their software development tasks themselves to be able to be given a task to complete. Likewise, I know a number of devs who can be given tasks but if you give a broader "this is a problem that the organization needs to solve" are hopelessly lost and apply and reapply the patterns that they have learned to try to make the problem fit into a solution they know. I do not believe that everyone has the skills / aptitude to move from one to the next. Those that do will have the ability to navigate the new market. Those that lack the ability to move up will eventually be overtaken either by younger (cheaper) workers, or improving automation.

UncleOxidant 2y ago

> Structural unemployment happens when the skills of the work force don't match the needs of employers I know someone who was laid off from an AI group at a large chipmaker last year- his group was eliminated. But because he wasn't doing LLMs, he was doing CNNs for vision, it took him about a year to find a new gig teaching Machine Learning classes - not his preferred gig or level of pay. He could easily have transitioned from CNNs to LLMs but the algorithms sorting the resumes didn't seem to think so. Personally, I've worked on optimizing code to make machine learning algorithms run faster/more efficient - also compiler/LLVM work. When I was laid off over a year ago I'd apply for things and it was mostly crickets - had a few interviews which often ended up in ghosting. I gave up looking and am working on a startup with some other who've found themselves in a similar situation - no pay at this point, only equity. Anyway, I think your premise is flawed: AI/ML skills are in demand, but even people with those skills are having a hard time getting back into the industry. Maybe the companies doing AI/ML are spending so much money on GPUs now that they can't actually afford to hire people?