A large company may have a handful of actually-very-special-role positions open for every hundred generic "our tech stack is this and isn’t going to change" developer roles.
They have the time to train someone into a role and typically don't have to worry about going to work for a competitor after they get out of AIT or A school.
While there's nothing wrong with those companies paying such, it drastically changes the incentives for the regular employee. Going to work in Big Tech can be like winning the lottery (in the dot com boom, I had about 6 million on paper... that I didn't sell before the crash).
That there is drastically different industries with correspondingly different levels of compensation has incentivized the person who can change jobs to move from a $60k job to a $120k job to a $250k job to do so.
In the 90s I was making $50k/year ($50k '95 -> $109k '26 https://www.bls.gov/data/inflation_calculator.htm ).
The companies where people used to work a decade or two or three are still there and in many cases, largely unchanged.
What's the difference between those job types? It's that software development and IT have a multimodal salary with much greater discrepancy between the companies that fall into the distribution around the lowest mode and the highest mode.
https://newsletter.pragmaticengineer.com/p/trimodal
If you switch from working at a local area company to Big Tech, you significantly increase your salary. For other careers, going from the local job to the national big company will likely give a much smaller boost.
Here are the stats for tenure: https://www.bls.gov/news.release/pdf/tenure.pdf
Table 6 (last page) shows the median tenure by job classification. Computer and mathematical occupations are the ones that were going down between 2014 and 2020. The others were holding fairly constant.
Table 2 shows 21% of people 35 and older have 10 years or more of tenure at their employer. At 45 years old, it's 45%.
I still hold that software development is the outlier for shorter tenure and the largest factor driving that is that you can quit your job and get a new one that pays twice as much.
Do you have an alternate explanation for this data or published data that contradicts this?
The US-style neoliberal attitude towards decimating the workforce at a drop of a hat for PR purposes does have a chilling effect on any employee's trust.
But don't fool yourself: one of the most important tools anyone has to improve their life and wellbeing is the ability to switch jobs. It can create career opportunities you would otherwise not have, and even unblock career progression by sidestepping your current job's corporate constraints.
no shit I work with multiple people that dipped out of their old gig, did 2 years somewhere else, and came back at a much more senior level to their old job because HR limits how often you can get promoted or increases inside a bracket.
jumping matters, as does the potential threat of jumping.
This shows some ignorance on the history of work relations. The bad old days of industrial guilds and company towns are not renowned for quality of life and independence. Your life was tied to the company, and all you could aspire to do in your life would be to toil in the same low paying job while mixing up your corporate structure with your life's social structure. If god forbid you lost your job, you and your family would lose your whole life.
Tell me how this is doing things right.
For those in the US to understand, see how your access to healthcare is tied to your employer, and the impact this have on your life of you are forced out of your job. Now extrapolate this to every single aspect of your life.
Common sense would dictate that a company with 10,000 people has a need for a diverse set of roles, and it would be completely stupid to recycle the same template pretending the role was always the same. In fact, repeating a stale advert suggests the company does in fact not have positions opened and either forgot to pull an old advert or some recruiter is trawling for applications to pretend to be productive.