Hacker News Reader

Comment by hn_throwaway_99 | Hacker News Reader

> Public pay information doesn’t require equality.

That's absolutely NOT how it works in practice. It is widely studied that there are HUGE differences in programmer productivity. While the theoretical "10x" programmer may not be that common, I see "2x" programmers all the time.

But you hardly ever see 2x differences in salaries, even though this would be the fair thing if you were looking at output. Once pay becomes public, management is basically constrained to keep people in relatively tight bands because it's trivial for anyone on the lower end of a pay scale to complain that Programmer X gets paid more money, even if Programmer X gets shit out the door 2x faster.

View on HN

The problem, I believe is in objectively measuring said "productivity". Yes, you say person A is 2X productive than person B. But that's just arbitrary from your POV. Someone in the same team may mark them as 1.5X productive. Heck, the same person you saw working 2X one week might not work the same amount the next. If you have any suggestion on ways of measuring productivity. That'll be cool to hear.

Some people consistently outperform most others in their clique. Fair compensation is a problem without a great solution there, but denial is not the answer.

I agree, measuring productivity is very difficult. And as you said, it can be a moving target as well.

But I don't think you are claiming there are not significant differences in productivity? Given that, it's not unreasonable for the more productive people to ask for more compensation, even if it is difficult to identify who they are.

It's definitely a hard problem. There is probably a lot of room for improvement in general for how we approach this.

I'm not disagreeing that is a problem, and the reason this is a thorny issue is that there is no reliable way to measure programming productivity: https://martinfowler.com/bliki/CannotMeasureProductivity.htm...

That said, while it can't be measured, I also don't think anyone would disagree that it does exist, in that I've easily see engineers that can produce reams of quality code in a fraction of the time other engineers can produce buggy code.

But yes, I do think there is an amount of judgment at the end of the day, which is why this is a tricky subject.

If 2x programmers are paid ~2x then the company is using some performance metric to influence pay, but then for pay to remain secret the performance metric also has to be secret.

You don't see differences in salaries perfectly correspond to differences in outputs for the same reasons that companies want to pay less for remote work. Pay in general doesn't correspond that great to value provided. Potential complaints about pay comparisons are way down the list of reasons.

If one can't make a compelling case to be in a higher "band," and therefore get paid more, then I'm failing to see how they could possibly make the case that they are actually 2x productive.

At most companies I worked for, I'm sure there is 2x-3x difference in "raw" dev productivity, or even 10x once we keep in mind the affects of the correct prioritization and technical solution, but the salary difference is almost always within 50% band. The most productive people could never negotiate 5x salaries.

I'm not saying there would be a 5x, or even 2x, salary. Only that if one cannot even negotiate for oneself, on the basis of some self-perceived productivity disparity, a higher salary (of literally any meaningful magnitude), then it seems quite dubious that such a productivity disparity even exists (especially at the perceived magnitude), given how flawed comparing "raw" productivity often is.

That tracks latency, but not throughput. Sure, a programmer may get a given ticket out the door 2x faster, but they may create 3x as much technical debt that winds up being addressed as separate tickets. That's a useful skill if you need one particular feature urgently for a deadline, but less useful for steady-state development.

While I agree that there can be or at least seem to be that overall difference in productivity between people, utilizing a single metric to make that conclusion is simplifying a much more complex issue.

Or a senior dev may wind up only getting a few tickets done, but it's because they're dealing with meetings and planning to determine the correct course of action for a project that may take a month or more. Or they may be a useful 'repository of knowledge' for the rest of the team, and spend a lot of their time enabling multiple other people to be more effective than they would be otherwise.

It is trivial to complain in general.

If a trivial complaint from a colleague who, in your theoretical here, is less productive than you can make your employer not pay you more then that means you are only trivially more productive or negotiating pay based on something other than productivity.

Labor is told repeatedly that wages are based on market forces and companies have to act at a sociopathic level of responding to economics.

Why does the narrative always flip to employers having to act emotionally whenever labor would get an up like decreasing the information asymmetry between employers and employees?

Let's assume that is true, what has that to do with public pay information. You say yourself you hardly see 2x differences in salary, and I assume you are not only talking about public pay companies (I certainly have not seen it much either even in non public pay companies).

At your large, in vogue, non-neflix companies, pay often has fairly well known bands, and approximately doubles (or more) every two levels (well not at Microsoft). This holds true at Amazon, Facebook, Google, Uber, Snap, Apple, Stripe, etc.

If you mean that, between people in the same role, there is usually huge differences in productivity, I honestly haven't seen that, except perhaps for brief periods where someone is late to a deserved promotion.

Reply on news.ycombinator.com