4 ms·
If Gitlab wants to increase peoples' lifetime earning, why not start with its own employees? Didn't they layoff hundreds of people?
by rcme 4y ago
If Gitlab wants to increase peoples' lifetime earning, why not start with its own employees? Didn't they layoff hundreds of people?
- sosodev 4y agoYes, they let go 130 employees a couple of months ago. They're also a company whose remote pay is dependent on location. So they're unwilling to pay people what they're actually worth.
- 908B64B197 4y ago> They're also a company whose remote pay is dependent on location. So they're unwilling to pay people what they're actually worth. They are also paying significantly less than their competitors. No wonder they are falling behind. I have no clue what this "foundation" is, the wording is extremely vague.
- pmoriarty 4y ago"They're also a company whose remote pay is dependent on location." What is it with companies doing this? It never made sense to me that a company would pay me more or less depending on where I lived. If I'm doing the same work why should my location matter?
- rrgok 4y agoOh man, I agree with you. The typical answer is: market. But on a side note, this should make us blatantly clear that, all else being equal, our work has no inherent value.
- ingonealan3 4y agoTo take the point further: nothing, ultimately, has any inherent value.
- robopsychology 4y agoI use it to my advantage - if I'm remote working for the Bay Area I can double my local salary but still be significantly lower than a Bay Area local.
- geodel 4y agoYeah, just like you pay same price for house, food, daycare, school, medical treatment and so on. If it is same square foot why does it matter if house is in Cleveland, Ohio or San Francisco, CA.
- HatchedLake721 4y agoEvil GitLab is unwilling to pay people what they're actually worth. Poor GitLab employees worldwide cannot resist from being taken advantage of and have to work for unworthy amounts... Every time GitLab comes up someone will bring this topic up. San Francisco salaries are not a worldwide benchmark of people's worth. There are valid reasons why GitLab doesn't pay engineers in Kenya $200k. Should I also swing your logic the other way? If an engineer in Kenya for $50k does the same work and output as you, a $200k engineer in SF, can we now say you're being overpaid 4x what you're actually worth? If people don't understand why companies pays local labour market rates, either look at GitLab's income statement and multiply 2000 employees by your definition of "worth" or start your own business and show how everyone in the world can pay 8 billion people the same salaries. Good luck. https://about.gitlab.com/handbook/total-rewards/compensation/#paying-local-rates https://about.gitlab.com/handbook/total-rewards/compensation... https://about.gitlab.com/blog/2019/02/28/why-we-pay-local-rates/ https://about.gitlab.com/blog/2019/02/28/why-we-pay-local-ra...
- leereeves 4y ago> If an engineer in Kenya for $50k does the same work and output as you, a $200k engineer in SF, can we now say you're being overpaid 4x what you're actually worth? No. If GitLab thought the work was only worth $50k, they wouldn't pay anyone anywhere $200k for the job. If they do pay $200k for the job anywhere, they must expect to receive more than $200k in return, and pocket the difference.
- ojosilva 4y agoIt doesn't work like that. It's more like: there's a 250k budget. We got an ex-googler-ex-meta-kid for 200k, so that leaves us 50k for a... Kenya engineer. Then, once both are on-boarded, gee, the Kenya engineer is twice as productive as the googler kid. Jackpot!
- M95D 4y agoWhy do they need an ex-googler-ex-meta-kid then? Why not 5 Kenya engineers?
- prepend 4y agoPeople are not paid based on their value. They are paid based on what the labor market drives. There’s some relationship to value, but it depends on how many people can and are willing to do the job.
- wellthisisgreat 4y agoPeople are actually worth what they are able to get paid, and are willing to accept. Period. End of story.
- HopenHeyHi 4y ago[flagged]
- TheCoelacanth 4y agoThat would result in a $1 increase per dollar spent, which is 1% of the ROI they're aiming for.
- phailhaus 4y agoIncreasing people's lifetime earnings != "you shouldn't be allowed to fire anyone, ever." I mean come on.
- rcme 4y agoI understand your point, but Gitlab is a productivity company. They increase peoples' lifetime earnings by making them more productive, allowing them to increase their economic output and demand more compensation for their labor. Laying off people is an admission that Gitlab isn't able to deploy additional capital towards making people more productive anymore. So to then announce that Gitlab is establishing this foundation just seems like delusional management.
- phailhaus 4y agoThat doesn't follow. Laying off people is an admission that they hired more people than they need, that's it. It's like saying "wow Red Cross, why not just invest in your own employees before you talk about improving other people's lives????"
- rcme 4y ago> That doesn't follow. Laying off people is an admission that they hired more people than they need, that's it. Yes, exactly. Gitlab reached their ability to deploy capital and need to scale back head count. Using your Red Cross example, it's like the Red Cross announcing a new initiative outside of their area of expertise to help people while actively scaling back programs they had in place to help people.
- phailhaus 4y agoYes, and that would be valid if the new initiative is intended to help more people. That should be obvious! GitLab laid off 120 employees. 120 people, that's it. Don't you think that this new initiative is meant to help more than that?
- geodel 4y agoWhile we are at it why not double executive salary. It will increase their lifetime earning.
- apple4ever 4y agoIt would also help if they improved their product with better pricing.