Eyes Above The Waves

A relative of mine gets a lot of work building iPhone, Android and RIM apps. He moved from Denver, CO to a city in Colima, MX and he gets to charge less, make more money, and keep a lot more of the money he makes. It is interesting to see that companies can easily move jobs from, for example, Flint, Michigan to Guadalajara, but where capital can often move across borders, there is often little support for labor to move across borders. He can only do it because he literally needs only the Internet to have a place to work. On the other hand, he points out that there are other costs to living someplace with a less developed infrastructure. If you get in a car accident in some parts of MX, you cannot assume there will be ambulance service. And you may need access to a pack of cash to ensure you will get quick medical attention, no matter what kind of insurance you have. There are benefits to living somewhere with a developed infrastructure. So who should pay for the infrastructure that holds our society up? Recent political events in the US would suggest we do not know the answer to that.

That's interesting, but it seems orthogonal to the question I'm asking. I'm certainly not suggesting that everyone does or should want to live where the cost of living is low.

This discussion reminds me of: http://blog.stackoverflow.com/2011/07/how-much-should-you-pay-developers/ I agree with the premise - it makes sense to incentivize people to live in expensive places if that's where your office is and that's where it pays to be competitive. I'm a different employee to Mozilla as a local Mountain View employee than if I were remote. At the same time, I don't think Mozilla should pay me more because I decide to work out of NYC vs. Minneapolis. It's my fault if I want to live somewhere expensive where we have no office. The biggest problem, is that we end up competing with a lot of different locales. E.g. if we hire someone from NYC, they have their NYC-based salary history which will be hire, and we'll want to make an offer in that ballpark. So perhaps there's a need to take into account location.

Cost-of-living based compensation of course is easily "hacked" by remote workers. All they need do is find the most expensive place to live while applying for jobs, then subsequently move to a cheaper location. Obviously there's difficulty with that strategy but it's definitely possible.

It's just supply and demand. If you paid people the same regardless of cost of living, that would mean one of two things. If you pay on par with cheaper places, you're not going to get people to work for you in expensive places, which limits your employee pool. If you pay on par with more expensive places, you're overpaying people who live in cheaper places (relative to local jobs), so you're wasting money: you could get more people to work for the same total cost. Simply scaling linearly by cost of living doesn't make that much sense either, though. An organization with a first-world budget can easily outbid local companies in third-world countries, so if it pays only on par with local companies, it's losing access to a lot of cheap talent. What makes the most sense is to individually negotiate salaries based on the employee's expectations and pay history, so you pay people as much as necessary to keep them happy and no more. This means you'll pay people less in cheaper places, but probably not linearly, because you have to compete with other multinationals that can outbid local companies. All this is from the point of view of the organization's best interests. If you want to weigh other considerations as well, like some conception of fairness, your results may vary. I don't find such considerations compelling here, because I don't think anyone has a right to any particular salary relative to anyone else.

Aryeh, that makes sense except for two things: -- In my experience organizations do not generally have a policy of paying people the minimum that it takes to get that person to take (or stay in) the job. Regardless of whether they should or not, they don't. -- Your explanation doesn't read as well if you treat location as a variable rather than a constant. Jeff's comment is apropos. Although really, we mostly agree. I say that one should ignore cost-of-living except for competitive effects, and so do you.

I suggested employers should pay people the minimum it takes to keep them happy, not just the minimum it takes to keep them on the job. For a sufficiently loose definition of "happy", I'd say that's what employers often do. Employees can get paid more just because they feel like they should earn more: e.g., having a higher salary at a previous job, or asking for more in salary negotiations, or having been at the same employer for longer and expecting more raises. The flip side is that employees who are happy with a smaller amount for whatever reason do often get paid less. Location is somewhat variable, but not fully. Lots of people will move to get higher pay, but a significant number won't. Few people would move to a third-world country just to benefit from lower cost of living. It's true that organizations don't want to give employees incentives to move to places where they'll have to be paid more for the same work, though, and that's a relevant consideration here. It argues in favor of not scaling pay in direct proportion to cost of living.

I suppose this comes down to Maslo's hierarchy of needs. Cost of living, monetary or otherwise, changes the pyramid and puts the company at higher risk of an unhappy employee. In the worst case they can loose the employee altogether. Although the cost of living adjustments leave room to be abused the question is why would "happy" employees ever do so? The relationship between employer and employee is the thing that will prevent abuse of priviledge.

Doesn't it really depend on whether the compensation is close to the cost of living in an area? You have to pay people enough to actually survive/thrive. Above that line, however, it seems that it's just a question of satisfying employee expecations/competitive needs. The cost of housing varies so greatly that while I can comfortably afford a home for 9 people here in Johnstown on our current income, trying to do the same thing anywhere near the office in California would be a huge stretch.

It's a tough question with no universal 'right' answer. Benjamin hits on a huge point: housing costs vary even more than generic "cost-of-living", and housing costs matter more to experienced older professionals than younger ones who may not have "settled down" yet (and who typically rent). Our property in the Philadelphia suburbs would cost maybe $4-7m anywhere in commuting distance to Mtn View, while what we could get for the same amount of money in Mtn View would probably be a 3-br postage stamp house around 1/2-1/3 the sq-footage and perhaps 1/50th the land. Because of how the housing market locks you into an area over time due to appreciation, it's very hard to consider relocating from a low or medium real-estate-cost area to a high-cost area after a certain point - you'll never catch up to those who got onto the housing treadmill in the high-value area when they were young. Perhaps this will end up changing if housing prices stagnate for years or a decade, but it's true now. And Philadelphia suburbs are not a low-cost area; they're above average nationwide -- in most of the country, even owning a 4-br house outright would barely get you a down payment on a *much* smaller place in Mtn View. That speaks more to relocation than remote work, of course. The cost to the company based on not doing cost-of-living salaries for remote workers is generally swamped by the cost savings of not having to provide an office and all the accoutrements (especially in high-cost areas), so it's easy to justify and the company gets to cherry-pick the best experienced remote workers. I'll note that most companies (almost all) do use locally-competitive salaries when they have a local facility, as the hiring pool and competition for the jobs is almost all local. Note that we're kind-of assuming that the "non-adjusted" salary is based on an expensive location, like Mtn View or NYC or SF. If the "baseline" salary is based on a more normal locale, it's easier to justify either paying more for people in expensive locations, or to convince people to relocate to a baseline-cost-of-living office. A company may still not pay less-than-baseline to remote workers in cheap areas.

Hi Robert, Old thread, but could please clarify if Mozilla takes in count cost of living for its salary policy? Another factor is local average salary - there are places with high cost of living but low salaries, just because people are used to it - so they just have weaker purchasing power. These places are quite common in Europe actually. So some comments from you about these things would be useful. Thank you, exim