Hacker News Reader

Comment by dvtrn | Hacker News Reader

I questioned the policy of paying people differently based on their location (more specifically, some algorithm's idea of the rent in their area) rather than the work they do

Interesting, there was a moment maybe a year and a half ago where I wanted to work for Gitlab, but upon learning precisely this I changed my mind. Not necessarily just that pay was location based, but reading through the Global Compensation page just felt very...I dunno, cold and dispassionate?

There's probably a better way to phrase the sensation, because I'm certain this wasn't their intent, but the right descriptor eludes me presently.

View on HN

I’m fine with the location-based pay.

Regarding the handbook though, I feel its entirety is cold and dispassionate. Devoid of anything human.

For me it's a combination of both, I think?

-- BEGIN: BIG ASS WALL OF TEXT --

Submitted for your thoughts: I toyed around with their salary calculator. Selected my job role and what I think is an appropriate guess of my own skill level with the options provided.

And then I got stuck. Because the "area" (my hometown in the rural southeast US, an economically depressed former mill town) I'm about to buy a house in isn't represented, it's not an option on the list. Neither was the city one county over, a city with a much more productive local economy. In fact, they don't even list the state Capitol, home to a pretty good business school and several engineering schools in the state. The nearest metro area that I could select to even calculate a potential salary is three hours away, separated by a state line with a completely different state tax rate than where I plan to reside.

It gets more complicated. The cost of living where I'd like to buy a house is objectively less than this city three hours away, separated by a state line. I know Gitlabbers love to comment and clarify things, so if one of you are reading this I'm curious to know how your salary sabermetrics adjust for this.

Then there's the fact that my significant other plan to start a family. If I buy a house in hometown, and I'm paid a lower salary than city three hours away, what's the discussion like if I get hired and two years later I have a child? That's not covered in the Compensation Handbook (if this is mentioned elsewhere in the handbook and I merely missed it, my apologies, I'll redact this complaint). Relocating requires involving a "written agreement" from my manager, and a recalculation of my salary.

On the one hand, it's VERY helpful I think to have all this laid out before applying, it's a level of transparency that more companies could stand to emulate, even while I'd say maybe don't emulate every practice employed by Gitlab...but the implementation of various policies just feel robotic, impersonal and quite rigid.

Again, I'll take full ownership of my words here if I've missed, or completely misread and misinterpreted aspects of this all, and I completely understand it's within Gitlab's prerogative to structure their company in a way that attracts the types of people for whom these things aren't considerations for their next job. It's just a shame for me personally, because I really like the product and really wanted to be part of solving some of the technical challenges their tools solve, having been a pretty vocal proponent of them in my current dev team.

Interesting. I have worked a bit on our calculator in the past (technical implementation).

Some clarifications, best of my knowledge, currently on mobile:

It would be impractical to calculate all the location factors in the world, and in your case it would be calculated. I agree with you that this is less than optimal, as it might stop people from applying if that data is missing. So this is an area where we could improve.

Internal discussions on the calculator have lead to changes, e.g. https://gitlab.com/gitlab-com/www-gitlab-com/issues/3396 So the calculator or location factors are also up for discussion and iteration.

When you move, higher location factor definitely needs approval, lower doesn’t afaik. To be honest often location changes with a big factor change are across state or country lines, which might mean change of Entities even in Europe. Compare that to a non remote job, if you are bound to an office location: you cannot move.

Personally I haven’t been involved in a location change of a team member, but as a manager I would definitely try to enable my team member to move let’s say from Manchester to London if they do wish.

Hello there, thanks for the response.

It would be impractical to calculate all the location factors in the world

On the one hand I absolutely agree with you here, however I searched for multiple cities in the state of consideration, two of them are international destinations on the Atlantic coastline with considerable economic impact to the state and region at large, and as mentioned one of them is the state Capitol with nationally recognized colleges, I'm just a bit surprised that not this city is represented in your calculation.

That said, I do hope that area is improved on at least for the sake of other potential applicants.

Cheers!

Did you see that "South Carolina" is an option in the drop-down? It has a value of 0.7, which seems pretty generous: it's higher than New Jersey (0.633), and is the same as Atlanta, Charlotte, and Connecticut (0.7), all of which have higher cost-of-living than most parts of South Carolina.

The fact that the urban areas of South Carolina aren't broken out actually seems better for anyone who would plan to live in a rural part of the state. By comparison, "Everywhere else, Georgia", "Everywhere else, North Carolina", and "Everywhere else, Virginia" all have a multiplier of 0.633. This seems to indicate that you'd get paid about ~10.5% more in rural SC than you would in other rural places in the SE.

I would like to ask why you feel a job should be rewarded differently based on location when the deliverables and other gains from the employee are the same?

E.g I don’t see why I should take a pay cut moving from the US to the EU. Looking at Basecamp, they give everyone the US salary and they arrange their culture around remote work that way.

Note how it doesn’t even work this way between countries, through the ‘most favoured nation’ laws.

We talk about our reasoning for paying local rates at https://about.gitlab.com/handbook/people-group/global-compen...

Your wages aren't even competitive for SF.

We're always adjusting our calculations based on market research. Are you including the value of stock options as well?

Options aren't worth much and may even end up being a liability. Current advice says to value options at $0. At least grant double binding RSUs.

It allows them to capture extra value from employing people in areas with less competition from employers.

They're not going to say that because a lot of people have odd ideas about compensation that they're very emotionally attached to. There are people that would take a job with them for $X when told its about cost of living fairness that would not take the job for $X if they were told it's about capturing surplus value.

This is the same reason it's dangerous to share your compensation with coworkers as an employee. They don't just all say thank you for telling me and negotiate for a higher salary while acting discreetly with regard to your conversation, or start job searching; some will get real mad and irrational and there's a very good chance you'll be caught up in the fallout.

Piling on here...”pay me more because I live in San Francisco” carries about as much weight as “pay me more because I’ve got twelve kids have to pay alimony to two ex wives.”. It’s one thing when there is an office in San Francisco where you need to show up most days. When you are working from a spare bedroom, it really doesn’t matter where that bedroom is.

It does matter, because payment isn't necessarily a reflection of delivered value, payment follows a supply and demand curve.

If you don't pay SF rates to somebody living in a SF bedroom, then you can't hire that person, unless they accept leaving money on the table and nobody wants that, especially because wages usually reflect the cost of living in the area.

So if you want to pay everyone the same wage, you either stop hiring from SF altogether or you give everyone SF salaries.

The later option isn't feasible because if you pay everyone SF wages, then you might as well open an SF office and hire only from SF. Is SF overcrowded by companies? Plenty of tech hubs to choose from, you only need a decent university nearby and developers don't need to be in the same location as the sales team.

As a matter of fact one reason companies hire outside of big tech hubs like SF is cost efficiency. Because remote work has disadvantages, like poor communications and the need for employees that can self manage well. You take cost savings off the table and remote work is much less appealing.

Doesn't matter where they bedroom is?

That's a socialist concept. Not in the world we live in.

> So if you want to pay everyone the same wage, you either stop hiring from SF altogether or you give everyone SF salaries.

The first option is a perfectly reasonable position to take. There's less than a million people in SF. There's over seven billion people everywhere else. The people in SF aren't magical; people outside of SF can do the job just as well.

> if you pay everyone SF wages, then you might as well open an SF office and hire only from SF.

This statement only makes sense if the only thing you are considering is cost. Even if the cost is identical, it's still advantageous to hire remotely. It's easier to hire, there's a much larger pool of candidates, it lets you hire people who would otherwise be inaccessible to you for other reasons, it increases loyalty, there's no commute draining people's energy, etc.

> Because remote work has disadvantages, like poor communications

In my experience, fully remote teams are much better at communicating than mixed or onsite-only teams. Why do you think remote work brings poor communications?

I think taken to this logical extreme, the only conclusion you can come to is that they value having people from more expensive areas. I mean, from a business perspective, it seems insane to hire anybody from SV if you can fill your roster with equivalent people from poorer areas. Some areas are absolutely chock full of experienced developers used to working for a tenth of the salary that a SV developer can demand. So why would any rational company choose to pay up to 10 times more than they have to for workers?

I think the obvious reason is because there is more to having a high tech worker than the work they produce. Remember that Gitlabs is still a VC backed company. In addition they still have to exit sometime to pay back all those VC investors. As much as I hate to say it, a startup staffed by people from eastern Europe or south east Asia doesn't have anywhere near the cache of a startup staffed from people in SV.

Right now Gitlabs has paying customers, but it's important to understand that its bottom line could depend as much on their investors (and potential investors) than its sales. Not only that, but the buzz you generate from being a SV startup leads to other SV startups using your service.

How many companies would (very unfortunately) give Gitlab a pass if it was staffed entirely out of India? I'm extremely sorry to say that I think it would be quite a few.

So, as terrible as I think it sounds (especially as a developer who works out of rural Japan), I can completely see the value of paying more for developers from certain areas. SF, LA, New York, London, Paris and even Tokyo are just much more impressive places for your staff to live than We-don't-even-have-a-train kuso-inaka Japan.

> "The people in SF aren't magical; people outside of SF can do the job just as well."

We agree. But if there's expertise in SF that you can't get anywhere else, then you end up hiring from SF too.

In my experience this is exactly what happens with companies working with remote workers, they only hire from SF or NYC or other expensive cities when they don't have a better choice.

---

> "In my experience, fully remote teams are much better at communicating than mixed or onsite-only teams. Why do you think remote work brings poor communications?"

I have more than 10 years of experience working remotely and sometimes I hate it. Functional remote teams are better at communicating, yes, but that's survivorship bias [1] in action ... the people themselves are better, due to the selection process.

Communication skills can be honed of course, but the people have to have an inclination for it and they have to be capable of self management. I have no idea if better education and better processes can indeed make a difference, but I've seen people and companies failing hard at remote work and the companies that succeed in working remotely are those with a strong culture for it, a culture that's not easy to build. And their hiring process does tend to select for those capable to function well remotely.

N.b. I know this is a total anecdote, sometimes I wish we used the scientific method in advancing this field, but alas we don't.

[1]: https://en.wikipedia.org/wiki/Survivorship_bias

> Functional remote teams are better at communicating, yes, but that's survivorship bias

I agree with this, I just don't think that factor has any practical relevance to this particular discussion. Remote companies that aren't good at communicating are dead companies, and dead companies don't hire anybody. So what's the point in even considering them in this type of discussion? In this case survivorship bias is just filtering out irrelevant noise. If a remote company is hiring, then you can be pretty confident they are good communicators. With the exception of extremely early-stage companies of course, in which case all bets are off.

Ah and regarding planning a family, that is under the benefits and we currently upped our parental leave benefit from 3 to 4 months:

https://about.gitlab.com/handbook/benefits/#parental-leave

Well the 'child' inquiry was more related to how having a child can dramatically alter a family's cost of living, and how or even if this is considered with equal earnest in salary maths--less so a critique about the absence or presence of parental leave (which admittedly, yours is impressive).

That makes sense. Yes, children are not factored into the location calculator as this would be very hard to do.

I could imagine that this is even protected information which your employer isn’t allowed to know in some countries.

Not sure I follow, admittedly I'm not entirely versed in employment law well enough to know better, but what would be the protected information that precludes a family planning calculation in the salary calculator but not parental leave?

Asked a better way: if, per your Handbook employees who relocate go through a process "to stay aligned with our compensation principlea" and "understand the impact to compensation or your role at GitLab"[0], do new or expectant parents not go through a similar process?

[0] https://about.gitlab.com/handbook/people-group/code-of-condu...

Not associated with gitlab or employment law in any way, but employers in my country may not ask if you intend to have kids. The reasoning is that will be less likely to hire people that will be absent on parental leave, have more sick days, won't be as likely to do overtime, etc.

Oh for sure, we have those laws too and I'm definitely aware of them. I guess my question has existing employees baked in mind. An existing employee who wants to move to a higher COL environment seemingly has to go through a process that existing employees who are expecting a child don't, even though having a child dramatically changes the day to day expenses of living for a household. Or at least, that is a situation that isn't discussed in the handbook.

I could have clarified that inquiry a bit better, me thinks.

You seen to be conflating a household's actual expenses with a city's relative cost of living. Gitlab (along with my own employer and most remote-friendly companies I know of) adjust's pay based on the latter, not the former.

If I wanted to move to San Francisco I'd have to get my manager's approval, and I'd get a raise.

If I decided to have more kids, buy a bigger house, eat out more, start collecting vintage cars, etc I don't need my manager'd approval, but I won't get a raise.

You seen to be conflating a household's actual expenses with a city's relative cost of living.

I mean, I'd argue there are several things involved with having a family that have costs relative to where you live (access to schools/primary education for example), so I don't really see it as disparately.

> Handbooks are devoid of anything human.

Heh, that's probably by design. Pause for a while and think: this handbook would have been code executed by a soulless machine. And the business is that machine, detached from any humans that are necessarily in it.

cold and dispassionate

I recall them rationalizing how it is beneficial for employees getting payed low salaries, so it does not disrupt their local markets ...

Reply on news.ycombinator.com