On Twitter the other day, David Anson asked,

If someone is working 100% remotely, why should their pay be tied to which city they are in? They produce exactly the same work if they are in a big city vs. a farm house. “Cost of living” adjustments are for when the job forces people to work somewhere; that’s not relevant here.

This sparked a lively debate.

remote pay

Several people suggested cost of living and local market rates warrant different pay based on a person’s city of residence. I disagreed. Yes, local market rates determine compensation today, but as remote work grows, local market rates for remote worker talent is not sustainable.

My belief is based on a few assumptions.

  1. We’re talking about information workers here, though I may use “developer” as a stand-in for information worker since developers are who I know.
  2. Information workers can produce the same quality of work no matter where they live as long as they have good internet access.
  3. Remote work will become more prevalent and even dominant in the future.

So if my assumptions are correct, in the long run, I expect more companies to disregard location when considering compensation. If my assumptions are incorrect, then, well you can disregard all this. I just saved you a lot of reading!

The reason for this trend towards geographic compensation equilibrium is described by an economic principle called the law of one price (LOOP). I’ll get back to that later.

How Companies Determine Compensation

Let’s start with an important question. How do companies determine compensation? Is it based on a moral idea of what a person needs or deserves? Is it based on cost of living? Or do they throw a dart at pieces of paper with salaries written on them?

The answer is that companies pay as little as they can get away with. This is not a cynical observation. There’s no value judgment in this statement. Nor am I making a point on what should companies pay. Rather, this is an observation of the overall tendency for companies given our current economic system. There may be exceptions, but they are few.

What a company can get away with depends on a number of factors, but the most significant one is how much other employers are paying. In other words, The Market.

A thought experiment may clarify this point.

But before we get into that, I need to state some assumptions for the pedants and the “Well, Actually” folks. For this post, when I compare compensation, please assume all other things being equal™. For example, a company with amazing perks may get away with paying under market compared to a company with shitty perks. Also, startups that are not yet profital won’t have the cash to compete with established companies in salary. But they can offer a larger stake in the company in the form of options than a profitable company would. Yes, that’s harder to compare to an established salary because it consists of a lot more risk, but also a lot more potential reward. For the purposes of this discussion, I assume a total compensation that factors in options, perks, risk, and other non-monetary factors. Also, when I compare developer compensation, assume developers are of the same type, level, capability, etc. I’m comparing apples to apples here. Thanks for giving me the benefit of the doubt. You may continue.

Imagine you’re a company in a city where the average compensation for a developer is $100K a year. You decide to pay $75,000 for developers. All other things being equal™, you will have a hard time hiring developers. Good developers will flock to the companies that pay $100,000 or more. This explains why the market puts a lower limit on pay. Companies cannot get away with paying too far below market rate.

Now suppose you decide to be generous and pay $125,000. You’ll have no trouble hiring developers. But there’s a downside. By paying so much above the market rate, you will run out of budget faster and can’t hire as many developers as you could otherwise. This might lead to having too few developers to staff your teams. If you paid closer to the market rate, you could hire more developers.

This is why there is a natural tendency to pay something close to the market rate as opposed to being too far above or below it.

Sometimes, companies get the bright idea to engage in collusion to keep wages low. It happened a few years ago with some prominent companies and they got caught.

Apple and Google have settled a massive lawsuit involving allegations that that they conspired with two other tech giants to keep wages low in Silicon Valley and to avoid hiring from within each others’ ranks…

Companies also leverage information asymmetry to keep wages low. Companies have better salary data than individual employees. Employees often don’t realize that their compensation hasn’t kept up with the market over time. Or that they are being paid less than their peers.

A calculation based on average raises vs salaries supports this claim.

Staying employed at the same company for over two years on average is going to make you earn less over your lifetime by about 50% or more.

Keep in mind that 50% is a conservative number at the lowest end of the spectrum. This is assuming that your career is only going to last 10 years. The longer you work, the greater the difference will become over your lifetime.

If companies tracked the market, then the difference between staying or leaving should be negligible over the course of ten years. If they paid above market, then employees would be rewarded for their loyalty. Instead, the evidence demonstrates that companies penalize loyalty. Why? Because they can get away with it due to information asymmetry.

How Remote Work Changes The Market

Again, while many factors come into play when determining compensation, the market is the most significant factor. Today, the market is by and large local. The compensation for developers in San Francisco are much higher than for those in Seattle. And those in Seattle are still higher than those in Columbus Ohio.

I don’t choose these cities to pick on them, but rather because I’ve been exposed to their Radford data in the past and managed employees in each of these cities and have seen first hand the compensation differences. Radford provides compensation data to a lot of companies who use that data to determine their compensation levels.

These differences persist because of market friction. The company in Columbus can continue to pay Columbus wages because developers in the area don’t have easy access to better options. It’s a lot of friction for a developer there to move to SF for higher pay. Likewise, it’s a lot of friction for SF companies to build a satellite office there to access cheaper talent. Wages in these cities match the local market.

This makes sense when companies require people to have their butts in seats at the company office. But what happens as more and more companies embrace remote work and allow their employees to live wherever they want? Hold onto your butts!

For companies that embrace a remote workforce, the market for talent expands beyond their current geographical boundary. This leads to a talent market that has less and less friction. The company in SF that is willing to have remote workers now has less friction to hire a developer in Columbus. They don’t have to build or lease office space in that city. There is a negligible difference in acquisition cost to hire a developer there than one back in SF. Though the compensation cost is quite different!

This is where the law of one price I mentioned earlier comes into play.

The law of one price is an economic concept that states that the price of an identical asset or commodity will have the same price globally, regardless of location, when certain factors are considered.

The law of one price takes into account a frictionless market, where there are no transaction costs, transportation costs, or legal restrictions, the currency exchange rates are the same, and that there is no price manipulation by buyers or sellers. The law of one price exists because differences between asset prices in different locations would eventually be eliminated due to the arbitrage opportunity

Ah yes, described in the dry and detached language of an economist.

So how does this apply to developer compensation and remote work? Once again, let’s step into the magical world of hypotheticals.

Imagine a future in which remote work is the dominant mode of work in the United States. Developers can work for any company anywhere in the country. Yet developers are still compensated according to the local market in the city where they live. In this scenario, developers in SF make $200k and developers in Ohio make $100K. What would happen over time?

Well, smart companies in SF would realize they’re overpaying for developers. They will be tempted to dangle $125k salaries in front of the best developers in Ohio and poach them all. Faced with this competition, companies in Ohio would have to start raising wages in order to retain their best employees. At the same time, wages in SF would start to drop as SF developers are competing with developers in Ohio and across the country. The general trend would be for developer salaries across the country to reach something approaching equilibrium. This follows the law of one price.

The Wrench in the Market

Does that mean all developers of the same level and capability in the country will be paid the same amount. Will we reach full equilibrium in developer compensation?

No.

For starters, frictionless markets are a theoretical environment. There will always be some friction. For example, for many companies, time zone differences will be a source of friction. While I believe the discipline required to run successful teams across time zones produce many benefits, it remains a challenge for many companies.

One of the conditions for the law of one price is the lack of legal restrictions. Today, different states have different legal requirements and regulations for hiring employees within the state. That can be a blocker to reaching total equilibrium across the country. For example, I know some companies have a list of states they will not hire employees from. Over time, these companies tend to add more states as they navigate the legal challenges, but this does add friction.

Another conditions for the law of one price is,

there is no price manipulation by buyers or sellers.

As mentioned before, tech companies have engaged in collusion in the past. There’s no reason to believe they won’t try again in the future. But there’s also no reason to believe they’ll be able to maintain such collusion long term.

Racism and sexism also continue to result in pay disparity.

There will always be some friction in the developer market. But the friction due to geographic location will decline over time as companies embrace remote work. And that reduced friction will lead to a trend towards equilibrium between locations. The city you live in will not be a major factor in your compensation.

Is this a good thing?

I’m not a trained economist, but I’d like to play one on TV. And all those Planet Money podcasts I’ve listened to convinced me I can try and think like one. And so far, that’s what I’ve tried to do in this post. I’ve taken a dispassionate look at these trends to draw a conclusion on where they are headed.

Still, it’s important to ask the question, is all this is a good thing or not? The answer is, it’s hard to say. See, a great economist’s answer. Accurate but unsatisfying. Let me try harder.

On the growth of remote work, there are people who lament the lack of human touch and serendipitious interactions that comes with being physically present with coworkers. That is a valid concern, but it’s not as bad as you might expect. I spent seven years at GitHub as a remote worker and I formed deep bonds with my colleagues there. Shameless plug, I wrote a series about how to do remote work effectively based on those experiences.

Another concern is the lack of investment in local communites this may lead to. Many business owners take pride in running a business that employs people in their own communities. This may be mitigated by the fact that remote workers in that community may have more opportunities and better wages, and thus more disposable income to spend in the community.

Yet another concern is this will lead to companies in the United States shipping all their jobs to cheaper countries. I don’t doubt, many companies will try and move more work offshore, but that trend started a long time ago with mixed results.

However, contrary to popular perception, many businesses have had, at best, mixed results. According to several studies, half the organizations that shifted processes offshore failed to generate the financial benefits they expected to.

Hiring employees in other countries introduces new risks and all sorts of legal issues and regulations to consider. Also, there are cultural barriers and time zone challenges inherent as well. There is a lot of friction involved in sending jobs overseas. I don’t anticipate a massive exodus of job loss as a result of embracing remote work.

For developers, remote work provides lots of flexibility, especially for working parents. Remote workers can live where they please and have more control over their schedule. Developers in lower paying markets may see their wages increase as they join the larger market. Developers in the highest paying markets may see their pay decrease a bit, but they will have options to move to less expensive areas of the country.

There’s also a fairness argument to be made. If two developers provide the same value to a company, should they not be compensated the same? As a manager, I was in the situation where I had one employee making 50% more than another even though they were of the same level and skill simply because one lived in a Tier 1 city and the other in a Tier 3 city. For example, a company like Microsoft makes $273,000 profit per employee. Does that profit magically go down if an employee moves from SF to Columbus? Then why should their compensation?

There are many benefits to companies. Many companies will reduce the amount of expensive office space they lease. They will have access to more talent across the country. Some companies may struggle. For example, companies in the lowest paying markets may struggle as talent prices are driven up by larger and better funded companies. But that may be offset if they’re able to sell their product and services to a larger market as well.

In the end, whether it’s good or bad may be irrelevant. It’s inevitable. As more and more people and companies see the benefits of remote work, I don’t see a sustained backlash happening.

So if we accept that remote work is the future, it’s in our best interests to look at how we mitigate damage for those who are negatively affected and act accordingly. Good luck out there!