Hacker News Reader

Comment by jrochkind1 | Hacker News Reader

If there was really a tech shortage -- and had been for years -- wouldn't the labor supply rise to meet demand? Wouldn't there be people willing to work less than the going rate (which is far above a living wage), wouldn't the going rate be dropping?

I don't know the answer, I think what is going on may actually be kind of complicated.

But that's the kind of question raised, you can't explain away the question of whether there's a shortage or not by saying "It's just that people aren't willing to pay the going rate."

View on HN

There are a few things going on here.

I don't think the world has fully realized this yet, but there's huge divergence in skill of developers. I know a few developers who are worth $500k/year, if not more. And I know plenty of others who are not worth that, but are easily worth $100k/year. With all of them under the same label of "developer", it can be hard to distinguish what you're getting.

Car example! Imagine wanting to buy a car, but not knowing anything about cars. So you call Kia, and ask how much a car costs, and they say, $15,000. Then you call Honda, they say $20,000. A few others, similar prices. Then you call Rolls Royce, and they say $500,000 and you think they're trying to pull a fast one on you. And to compound this, if you were to list everything that you want in a car, you'd probably come up with something closer to a Rolls Royce than a Kia. So you're shopping for a Rolls, and you've budgeted for a Kia. Also remember, you don't know anything about cars, so you don't know how to distinguish important features (ie. wheels, a functioning engine), from luxury ones (ie. a fancy sound system, heated seats).

In terms of people not being willing to pay the going rate - plenty are! All of the big west coast tech companies pay handsomely for developers. Outside of big cities though, things are slow to catch up. So I think that a lot of the sentiment comes from people not wanting to move to SF or NYC or Seattle, but also wanting to get a competitive rate. Eventually these companies will have to adapt, but it can take time, especially if there are enough people who don't want to leave the city they're in. There's also the quality divergence taking place - as much as we'd all like to think we're good, Google and co. are very selective.

In terms of working for less than the going rate (which as you said is generous) - for sure that's happening. But there's adverse selection happening there, in that people who are taking jobs below market are either desperate and will quit when they find something better, or are of low quality. Developers aren't paid well out of the goodness of the company's heart, they're paid because they make money for the company. And bad developers can have extremely negative productivity even ignoring their salary.

Finally, labor takes time to catch up, especially skilled labor. From no programming knowledge to junior developer at a top company takes typically five years - ie. an undergraduate degree in computer science and some tinkering on your own (yes there are exceptions). And we are seeing supply increase - enrollment in computer science programs is up across the country. Imagine we had a severe shortage of doctors - you can't just suddenly triple the supply of doctors over a couple of years. Now if you have a huge shortage of brain surgeons, which might take a decade or more to train you're in for a lot of trouble.

Kia has what most people want in a car. (Fuel efficiency, resilience to damage, cost-effectiveness, easy maintenance)

Rolls has what people imagine they want in the car. Sort of like those HR hiring managers imagining what they want in a developer.

Seriously, they all think they need Linus Torvalds.

Nah, he just doesn't have the skillset: not enough {Ruby,Node,my pet JS framework} experience.

I think your use of the word 'worth' highlights the problem. They are 'worth' whatever the market will pay them, that's why it's a job market.

People trying to suppress the market-forces that self-correct this type of disparity are the real problem.

>In terms of working for less than the going rate (which as you said is generous) - for sure that's happening. But there's adverse selection happening there, in that people who are taking jobs below market are either desperate and will quit when they find something better, or are of low quality. Developers aren't paid well out of the goodness of the company's heart, they're paid because they make money for the company. And bad developers can have extremely negative productivity even ignoring their salary.

I would add two more reasons to that.

1) Dunning-Kruger; the developer is really good but doesn't realise it.

2) The developer isn't confident or knowledgeable enough to know what the going rate is and ask for it. Sort of an overlap with desperate, but perhaps not as likely to move on.

Nit: Dunning-Kruger is actually when you're bad but don't realize it. I think you mean "Impostor syndrome," where you're good but feel like an unskilled fraud. :)

The real problem is that to compete in today's world, your website:

1. Needs to have great design 2. Needs to be rock-solid in terms of stability 3. Needs to be snappy to use 4. Needs to be able to handle a Reddit traffic swarm 5. Needs to have individualization and customization 6. Needs to play nicely on computers, tablets, and phones

The person (or people, more likely) who is good at all of this just isn't available at rates the companies want to pay. This is the important distinction. Hiring somebody for 100k/yr is pretty much the minimum ante for a good developer in a tech city. But these "entrepreneurs" and "founders" think that is too much money, and instead claim there's a shortage.

In reality, if you're competing for top talent, 150k/yr should be your ante, and be prepared to go well into the 200s. This shouldn't be a surprise. To get even average talent from a consulting/contracting company, you're paying about 125/hr (of which 60-70 goes to the developer). That's 250k/yr. So why are companies freaking out at paying over 100k/yr for a full-time person who they can invest in? Hell, cut out the contractors and pay your people 200k/yr, and watch the good people come talk to you.

This is what Amazon and the like are doing now. I'm hearing 22 year old kids who haven't coed a line of code in a professional capacity, getting offers of 120k/yr with a 20k signing bonus. Netflix pays a significant chunk of its engineers over 200k.

The only real problem is that it's getting more expensive to do business in a tech world. But compare the skill sets that used to draw 100k/yr, and the skill sets now required to get 145k/yr (100k/yr in 1998 dollars = 144k in 2014 dollars). Before, you could just know HTML and some shell scripting. Now, you need to know multiple frameworks, how to deal with scalability, multiple data stores, "full stack," etc.

If anything, the good talent in this industry is still dramatically under-paid. A common metric is that the company should make from you about 7x your salary to justify your position. So if a company is making 2.1 million, you should be getting 300k. But we all know that's not what's happening. Instead, our wages are being capped and forced down due to illegal collusion between the top tech employers.

This all comes down to the people who own the businesses wanting to treat tech people as interchangeable cogs, despite all the evidence to the contrary.

The thing is, it's not usually easy to make money at that rate as a small company, especially on the consumer side. That is objectively a lot of money.

I'm not saying that good devs aren't worth a lot, but they're typically only worth that much when applied to a problem that impacts a huge number of people, as at Google, FB, Netflix, etc. The same dev working on a project that a small company launches may only have a $20k/yr impact on the bottom line. It takes a serious amount of usage to create the $2M/dev revenue figure you're quoting, and at that point, how much of the marginal impact is due to the dev vs. the corporate machine that enables him to have that large an impact?

You have to pay if you want to play.

If your expected return on a project is less than your planned costs to complete the project, why would you start at ANY price?

Because there's something you want to make? And it's possible in the long term to ramp up to that point, it's just that these salary expectations make the first few years rather difficult - almost dependent on raising money at the absurd valuations that are currently the norm.

Just saying, something's gotta give.

>>The thing is, it's not usually easy to make money at that rate as a small company, especially on the consumer side. That is objectively a lot of money.

This is like saying you want a F1 car, but won't driving fast at all. But you need it anyway. Much of this making money thing depends on knowing how to use people to make money. If you can't use a rockstar or don't even need one why expect them to work for you.

>>they're typically only worth that much when applied to a problem that impacts a huge number of people, as at Google, FB, Netflix, etc.

Ironically, the chances of a somebody good ending up working on something trivial are high in such places. You can easily get 'placed' into some place where your skills can just rot without having any hope of moving out.

So why are companies freaking out at paying over 100k/yr for a full-time person who they can invest in?

That's a good question, but I'm afraid there's a fatal flaw therein. The assumption is that companies want to invest in training employees, but it seems that they don't. Startups don't have time or money to do it, and established big companies that are engaging in collusion on salaries certainly aren't going to spend money on training (or even mentoring) since they are collectively clutching their strand of pearls at the horrific thought of their best employees getting hired away and taking their newly-minted skills to a more competitive business.

what i m hearing is 'greed is the root of all evil'.

That's what standard products are for, starting with hosted Wordpress and going up from there. Why are you building a web site from scratch?

The exact opposite would happen.

If there is a tech labour shortage, that is to say, too many jobs and not enough employees to meet those jobs, then we expect the price of labour to rise until enough people are incentivized to work. Think the logic behind Uber surge pricing

If there are too many jobs and not enough employees, then you would not at all expect people to be willing to work less than the going rate.

Additionally, you would expect the price to rise first, as the most productive/profitable firms can pay a higher salary and still make money. You would expect the price to rise until a) Less profitable companies stop looking for labour because they won't make money on it; and b) enough extra people, motivated by riches, enter the job market.

The fact that there seems to be chronically unfilled positions at every tech company I've ever seen, would predict that salaries should rise until they're filled. But they don't. We can make a few hypotheses about why this is. For funsies, here's some:

1) Government immigration caps. There is plenty of developer supply out there.... outside the borders. We aren't letting enough of them in

2) Greedy corporations. They simply don't want to raise salaries, because they're soulless MBABots.

3) Skills gap. As many people are pointing out in this thread: Most journalists' treatments of "Silicon Valley worker shortage" seems to be of the formula "There are X unfilled jobs at software companies. There are Y citizens in the United States with certifications A, B, and C. NOW KISS". And yet this is pretty simplistic. Perhaps there is an abundance of (say) people able to do grunt-level IT work, but skilled software architects are in short supply

4) The B Word. Maybe all of these startups and brand name tech companies are unprofitable, and only surviving on investor storytime. Maybe they legitimately can't afford to pay developers more. Maybe we're all going to be out of jobs in a year or two

5) Location gap: maybe there's plenty of talented engineers elsewhere, and they can't or won't relocate to California.

I've seen a lot of 5) where I work. The company that owned our business prior to the current owners had a policy of intentionally hiring local engineering graduates and building their skills. This has held wages in the area below average, I believe, because we have a solid local talent pool who don't want to leave home sweet home (myself included - it's a nice place to live). Part of this is also cost of living - you can get a very nice, new-ish 4000+ sqft house on two acres here for just a bit over $300k.

The reasons tend to vary by region, but they still often come down to people playing games with job postings to try to get someone to fill a role that is bigger than what they are willing to pay for. I've had two employers give me a job title that didn't reflect the work I was doing, and allowed them to pay me less than they would with the correct title.

Job postings in my region tend to either have a pay range that looks like it belongs on a posting which requires 5 fewer years of experience, or they put something into the requirements as a "must-have" skill that would normally bump the rate.

Another part of the problem is that much of the educational system has put itself into the position of job training instead of educating when it comes to development and systems/network administration. It is often difficult or impossible to find a good curriculums for administration. Computer Science curriculums often choose languages based on the insistence of employers that those languages are necessary in the job market (though the average lifespan of some languages in the job market is less than the average student spends getting a BS in Computer Science).

Meanwhile, employers often mistake experience with a given language as a higher qualifier than overall development experience and ability to learn and adapt what you have learned in one language to working in another. This seems like it should be especially relevant in web development, since full-stack developers have to be familiar with front- and back-end languages, markup, stylesheets, and often at least a little SQL.

Finally, there's high turn-over and the ever-present revolving door positions. I've met far too many people who go back to school to either get their MBA and move into management or to get a degree in something else just to get out of the field completely. People get stressed out, treated terribly, and often feel like they have to completely re-learn their job every few years to maintain basic skills. After 5 or 10 years they may reach the salary cap for their job title, and they may have no opportunities for upward movement in a technical position. A new person might come on board to do nearly the same job, sometimes with the same (or more) pay, while being less qualified and less capable.

All of these things still trend in one direction, though: employers undervalue an applicant's overall experience while over-emphasizing a combination of skills which isn't often found in one person, all because they not only aren't willing to pay the going rate, but don't even want to hire enough people to do the job.

> All of these things still trend in one direction, though: employers undervalue an applicant's overall experience while over-emphasizing a combination of skills which isn't often found in one person, all because they not only aren't willing to pay the going rate, but don't even want to hire enough people to do the job.

Great summary of the situation. This "shortage" is self-imposed.

This is a great comment. I think the part about employees reaching the salary cap and then going to business schools is particularly true. In fact, I've seen it go 3 ways: 1) smart person reaches salary cap and gets tired of software, goes to business school and comes back as either product lead/CETO/VP etc. 2) smart person reaches salary cap and has tremendous amount of software engineering/design experience, uses experience to get a VP/CTO position in another company with better prospects. 3) smart person does not take time to update his skills, has done lots of work but still does not have insight to architect/design complex systems. Faces unemployment. (Sadly, this was one of my former co-workers)

> Wouldn't there be people willing to work less than the going rate (which is far above a living wage), wouldn't the going rate be dropping?

Possibly, except that there are a lot of pressures on people to move out of programming. Some examples: folks over 35 aren't welcome a lot of places; financial incentives often push programmers out of programming into e.g. project management. So you have a lot of people constantly leaving the field, somewhat by the design of the roles offered.

Oh, and every programmer is reading HN trying to figure out how to start their own gig. Higher wages and other job fixes (e.g. security, status, professionalism) might bring a lot of these people back into the recruiting pool.

Luckily, 50 is the new 35, now that 35 year olds have figured out how to dress like 25 year olds at the office.

I'd argue that it is rising - in the form of code schools and accelerators. But, as others have pointed out here, that is feeding dev talent into startups, it isn't fixing the lack of IT talent fed into corporate America. Corporate America reaped what it sowed on that though. And, really these code schools are producing supply, but not necessarily quality.

> If there was really a tech shortage -- and had been for years -- wouldn't the labor supply rise to meet demand?

Much of the demand for skilled labour is in the USA, much of the supply of it is not in the USA.

Reply on news.ycombinator.com