”In light of other tech companies laying off their workers, Y's reluctance to slim down its workforce demonstrates Y's failure to reduce its dependence on employees. With a high level of uncertainty surrounding today's labour market and Y's limited ability to lower its fixed costs, our confidence in Y delivering competitive returns to its shareholders in the medium to long term is diminished, leading to our revised, lower rating of Y."
From that viewpoint, it's better to overhire when other companies are hiring and do layoffs when other companies are doing layoffs vs. just hiring the right number of people in the first place.
Which is completely insane, and is both terrible to workers and leads to more recessions.
How do we get investors to value companies making correct calls over just following the industry?
The class and jobs that suffered during the Pandy have rebounded and the people who benefiting greatly during it are now taking their haircut.
I was an essential employee during those years and switched to tech so I got the bad with the bad.
Also generational and geopolitical factors play a large factor as well. Inflation will become wage inflation for the next decade. If I were a mega corp I would be trying to retain talent now rather than have to pay even more to get freshers from a smaller and smaller pool.
Pandemic delusions and overhiring seem like the main issue here. Tech companies saw the level of success the pandemic brought (thanks to the pandemic lockdowns, lack of alternatives and rise in remote work), and assumed that was the new status quo. They went big, hired a lot of people, took advantage of government policies to ease the effects of the pandemic, etc, and then got caught unaware* when things went back to normal.
Adding to this, it's likely the effects were made worse by people getting bored or burnt out by their products/services too. When you use social media or other tech products for a few hours a day (or perhaps a few minutes at a time at various intervals), they're great. When you have nothing to do but use them for like, 12+ hours a day due to lockdowns, the appeal wears down fast.
So in addition to the 'well, we have nothing else to do' crowd ditching these products the minute the real world opened up again, a lot of the existing crowd likely drifted away too. People who were regulars on Twitter/Instagram/Reddit/Discord/YouTube/whatever become far less active post lockdowns, and it seems like these companies are feeling that.
* Well maybe not unaware. More suffered the curse of short term thinking all public companies seem to suffer from, since the accepted attitude seems to be about chasing gold rushes and doing what everyone else in your industry is doing, no matter how irrational or unsustainable it might be.
CEO's are driven by compensation packages that are heavily dependant on stock prices and peer pressure. Retaining and retraining good people are just not a factor in their decision making process, neither is social responsibility.
These mega tech companies are now so big and so few that a cartel-effect has definitely emerged and it is resulting in this collective push towards the AI hype train and ofcourse layoffs. Most of this AI training data was scraped surreptitiously, labelled by the poorest of the poor, and then sold as original art which is just not in any meaningful way.
It might be time for us as a society to stop arguing over gender/race/socio-economic/gun politics, or being distracted by AI psuedo prophets saying that AI has some mighty power that we must regulate(to lock-in their 1st mover advantage) and other topics loudly shouted at us by large media companies(which are frequently owned by wealthy interests) and start thinking about how we can reclaim our privacy, freedoms and agency in the halls of our capitals and courts.
Even this, an article that tries to answer the question of why seems to be unable to mention how many people these companies added in the recent years. Is that data hard to get or are they just incompetent?
It’s more convenient to pretend that these “tech workers” didn’t have to have packed their ears full of cement to not know what rollercoaster they were getting on by working for these companies.
It's easy to say tech companies "just over hired" during the pandemic. But for some companies that just not the case.
Google's 2019 revenue was 161.85; it's 2022 revenue was 282.83. Some of these companies legitimately just grew a lot during covid (and probably would've grown anyways).
Of course some companies like Twitter were hemorrhaging money; but if your revenue per employee is in the millions it's a pretty good question of why are you firing anybody.
Well; I think you're going to have a great deal of trouble showing that the 6% layoff in 2023 was targeted to "incredibly unprofitable departments" considering basically every department had fired employees.
But also Google then spent like 8 months of salary per fired employee (severance & etc). Surely it's cheaper to figure out how to move the employee from the unnamed "incredibly unprofitable department" to a profitable department. Like Google had 186,779 employees end of 2022 Q3 which drops down to 182,381 in 2023 Q3.
---
To get back onto the article, it does make the claim of "tech company workforces have largely returned to pre-pandemic levels".
> Nearly 25,000 tech workers were laid off in the first weeks of 2024. Why?
They are being replaced by AI. (I heard that AI learned to code and, taking a look at the masterpieces Windows 10 and Windows 11, this makes a lot of sense). /s
So executives are just playing a big game of "monkey see, monkey do". Why do they get paid so much?