Hacker News Reader

Comment by cletus | Hacker News Reader

Spot on.

(d) in particular is worth stressing. There is no difficulty in attracting and retaining talent. There is simply an unwillingness to pay them what they're worth and give them an environment in which they'd want to work.

Supply and demand. It's pretty simple.

Yet so many (typically business types) want to treat engineers as an interchangeable cost center. "5 years ago an engineer cost $100k so I refuse to pay more than $100k now to keep my costs in check".

The fact that it's taken this long for engineers to be recognized (financially) for the huge impact we have and the vast differences that can exist between individual engineers speaks volumes about us. As a whole, we're just not as good as the business guys at this compensation game and, a lot of the time, it's not our primary motivation. Give us something interesting to do and stay out of our way and we're happy.

View on HN

Hi! This is the Market Economics Fairy! I help supply and demand equilibrate everywhere! If nobody can find good programmers to hire, it means that programmer salaries are too low! Raise programmer salaries until...

...erm, actually, students are remarkably goddamn insensitive to their future salaries when it comes to picking a major. And the other way S&D could equilibrate here is by fewer companies even trying to hire programmers because they know they can't afford them. What's really needed - considering the Camel Has Two Humps effect where some people have the magic programming gear and some people don't - is a mass effort to take students in journalism studies or, heck, smart unemployed adults, and administer some simple tests to see if they have large amounts of hidden programming talent, screw the college degree. But that effort would benefit lots of companies equally, and that means it's a commons problem. Not sure the Market Economics Fairy can easily solve this one.

Just a note, there is a followup paper to "The Camel Has Two Humps". From the abstract:

Two years ago we appeared to have discovered an exciting and enigmatic new predictor of success in a first programming course. We now report that after six experiments, involving more than 500 students at six institutions in three countries, the predictive effect of our test has failed to live up to that early promise.

http://www.eis.mdx.ac.uk/research/PhDArea/saeed/paper3.pdf

Also, assuming this were not the case, it's hardly clear that searching for magic programming gear is a commons problem. Why couldn't Google/Facebook administer the simple tests and keep the results secret?

I forgot, they already try to do this, but it doesn't give them enough applicants: http://www.mentalfloss.com/wp-content/uploads/2008/02/Google... http://www.facebook.com/careers/puzzles.php

Or why couldn't a trusted third party, like Microsoft, Cisco or Red had charge journalism majors/unemployed adults to take the test?

I forgot, they already try to do this, but they just don't do a very good job and their tests have little predictive power: https://learningnetwork.cisco.com/community/certifications/c... http://www.microsoft.com/learning/en/us/certification/mcse.a... http://www.redhat.com/certification/rhce/

The new paper didn't say that the problem wasn't correct, just that their original solution (test) didn't work out. It's true that it seems to be hard to come up with a reliable test for this sort of thing, but that doesn't mean that it's impossible.

The magic market fairy can't help supply and demand equilibrate until the simple test exists, right?

My only point is that the problem isn't creation of a public good, the problem is that reliably identifying good people/good matches is really hard.

That was amusingly written but I'm not sure it rebuts the comment it responds to. Yes, you can also work to increase underlying supply instead of bidding up the price. And?

Smart companies already do things to skirt credentialized recruitment paths, because credentials (especially academic ones) are extremely inefficient, at least in software.

This isn't so much "insightful" as it is a basic obvious direct result of the law of supply and demand.

The Market Economics Fairy doesn't work here because there is such a large lead-in time to making money as a programmer. I have had 2 or 3 friends want to become programmers just because they want to make more money, and also because they see the opportunity to make their own apps etc.

I helped them as much as I could but none of them are now programmers because it's just too much work for no return to learn this stuff. You have to be interested in it, and supply and demand doesn't make more people actually interested in it...

If the salaries start getting higher and higher, and wages in other industries continue to stagnate, then more and more people are going to start getting interested in programming, natural inclinations be damned.

Let's put it this way: how many people who go into investment banking are actually interested in investment banking? How many fooled around with financial models in middle school, then realized they wanted to parlay that passion into a career at Goldman Sachs? Maybe the odd freakshow here and there, but by and large, people enter the field for the moolah.

Right now, programming is still dominated by people who became programmers because they enjoy it. That they're well compensated is nice, but many of these people would probably be programming even if it didn't pay handsomely.

The Market Economics Fairy would indicate, however, that that'll change eventually. More and more would-be business or marketing or even journalism majors are going to choose CS when the salary average disparity becomes wider and more widely known.

> a mass effort to take . . . smart unemployed adults, and administer some simple tests to see if they have large amounts of hidden programming talent, screw the college degree.

This is the huge fucking source of programming talent that has remained largely untapped ever since CompSci became a recognized standard for employment qualification, due to the "monkey see, monkey do" best practices effect in businesses across the US. That blind best practices urge in hiring processes, in fact, keeps many actual programmers unemployed while firms in need of programming talent lobby congress to make it easier to import talent from other countries to fill the void created by their own hiring practices.

It's a perverse market right now.

The reason why business types see engineers as cost centers is because engineers have positioned themselves that way. Engineers pushed for higher salary in lieu of the variable upside.

What engineers really need to do is to put more skin in the game. Push for more equity, even if it means accepting a smaller salary. All too often I see programmers who just want a 9-5 that pays 6 figures even if the company is seeing hard times, but that won't improve the situation. Most of those business types you mention see engineers in the same way that engineers see themselves: wanting a large, stable paycheck. The business types shoot for the large end of year bonuses, which in general only get paid if the company does well (ignore the TBTF banks for a moment)

As an example of what I mean, most celebrity endorsements are lump sum payments plus a very small variable upside. William Shatner, taking a longer view, pushed Priceline to compensate him (for his celebrity appearance) in the form of equity, and now he's laughing his way to the bank. In this case, he was willing to take the risk.

"What engineers really need to do is to put more skin in the game. Push for more equity, even if it means accepting a smaller salary. All too often I see programmers who just want a 9-5 that pays 6 figures"

A lot of engineers I know are like this because they were badly burned earlier on in their careers with promises of equity (while working below-market paying job for that equity) that turned into dust (due to massive dilution and other factors) even when the company had a very successful exit.

One could argue there they only have themselves to blame for not being sure of the actual value of their position in all possible exit situations, but now you're arguing they shouldn't just be engineers but also financial/contract law experts, which is especially unrealistic when they are just getting out of school.

This isn't as simple as "push for equity". There are plenty of companies where equity makes no sense (say, a private consulting company, or a company with no plans to sell a la 37signals)

What other options are there? Guaranteed end of year bonuses? Commissions in the form of profit sharing?

It is funny that options are now so associated with a company exiting that we totally forget that their original purpose was as an efficient mechanism for profit sharing

I was thinking exactly the same thing, about the parallels between speculative bets on appreciation and dividends.

For private companies with no plans to sell, ideally it would be an arrangement where the engineer receives shares (in a C corp) or is written into the operating agreement as a partner (in an LLC or LP) but a profit sharing agreement would have the same effect.

In major law firms, the associates aspire to become partners in the firm, which is analogous to my interpretation when saying "equity".

Actually, setting up software firms where the best developers can "make partner" would be quite an interesting approach.

I'm not sure it would work, but if it did, it could be a perfect solution for keeping a great team together for the long haul.

See: Accenture.

It's a consulting company not a product company; it may be significantly different when "making partner" at a product company. However, from what I've heard, each partner at Accenture essentially runs his or her own business within the larger scale of the organization. Success is directly tied to the success of the internal mini-businesses. It is, however, an example of a company where IT professionals can become equity stakeholders in a private firm.

I imagine a product company would need to make the partners in charge of one product each (or maybe a couple of partners on one product) in order to avoid the "Too many cooks in the kitchen" problem.

Though your premise is correct, Accenture is public, not private. They (like many others) wanted to expand beyond pure consulting, and the partners wanted to cash out.

Accounting and law firms may be a better example, though even they aren't as entrepreneurial as you think. They have their own regulations and internal politics.

The amount of politics and backbiting and sabotage that gets mixed in with partner-track programs is horrifying. Find an incentive scheme that empowers first-year hires to do amazing things for your company.

That's sorta the YCombinator model, isn't it? If you do really amazing things, you get acquired for a few hundred million. If you do somewhat amazing things, you get acqhired for a million or so. If you don't do anything, you get nothing.

Seems to be largely working within the specific space of Silicon Valley YCombinator-funded startups, but I'm not sure how well it scales. There's a pretty sharp filter for getting into YC, and for someone without access to PG's connections, the chance of hitting it big or getting acqhired may not be worth the risk they take.

But YC invests in technically-competent businesspeople, not engineers

As someone who left software engineering for law, I have to say that partnership is a great motivator.

I don't want an incentive scheme to help make you money. Pay me 20-25% of the revenue I generate for the company and give me a non-zero chance to be a stakeholder and I'll work my ass of for you.

Most traditional firms, to make partner, you have to be personally responsible for a large amount of business.

At Andersen, at one point, I heard the number was $10M/yr of revenue.

It would be harder to figure out how much revenue is generated by an individual contributor (even if they are a superb coder).

Also, partnership usually involves leadership decisions

Private consulting companies have all sorts of var comp plans; the most common is profit sharing.

Presumably, equity in these companies would be like being a "partner" in a law/accountancy firm.

I hope not. If you talk to folks in law, in particular, the partnership system is broken. It basically results in a situation where it's in the interest of the existing "partners" to have as many "non-partner" employees as possible, dangle the carrot of partnership in front of them so they work really hard, but in reality only promote 1% or so to partner. That maximizes the return to existing equity holders but it means life sucks for everyone junior.

Yet the partner/director carrot in law/banking/consulting convinces tons of Harvard, Yale, etc, grads each year to work 80 hour weeks for their 20s for a shot at equity (which realistically is more like 5-10% at big NYC type places and upwards of 30% in secondary markets).

This is really mostly a law thing - banks and consulting firms, by and large, aren't partnership models any more and haven't been in a long time. Any large bank or consultancy is public now, which means they give equity to employees out of a huge pool and it has relatively little effect on the comp of their managers. That's different from a law firm, where even in fairly large ones the addition of new equity partners can have a material impact on the existing partners' comp.

As for making partner in a BigLaw firm, folks I've talked to describe it as "a longshot" to "nearly impossible." A firm like Skadden will hire 500+ new associates a year (worldwide) and name maybe 4-6 partners, not all of whom started as associates. Unlikely the chance of going from associate to partner breaks 1%.

(Updated #s after asking someone who worked for Skadden.)

McKinsey and Bain are incorporated partnerships, and BCG is still an actual partnership. Goldman Sachs, JPM, etc, are public companies, but their investment banking and trading divisions operate very differently from a typical public company. GS, for example, pays out 50% of revenues as compensation, with MDs and PMDs (partner managing director) sharing from a huge profit pool. Internally they function very much like partnerships.

As for the chances of making partner, it depends heavily on the firm and the year. Skadden is something of an outlier[1]. The other big NYC firms bring aboard about ~100-120 associates per year, and promote between 5-15 depending on the economy. The last few years it's been around 5 and during the boom it was around 15. At medium-sized firms outside NYC, the %-age is more like 20-35%, though profits-per-partner at those firms are more like $300-500k/year, versus $2m+ at Skadden.

[1] Though your numbers seem grossly exaggerated. Worldwide Skadden brought aboard ~200 this year, which was a big crop. The two years before that were ~100 put together. And over the last few years, the partners promoted counts have been: 2007 (29), 2008 (25), 2009 (8), 2010 (6), 2011 (8).

I read in an interview that at the end of the year, 37signals takes most of its profits, and pays them as a dividend to its shareholders. I don't know if lay employees get a chance to own stock, but if they did it could be a great incentive.

Having worked on the Priceline campaign and with Mr. Shatner, I can tell you he laughed significantly less than halfway to the bank. He cashed in his shares ten years ago, well before Priceline's recent stock blast-off. Your point, however, is valid: he was paid with shares.

structuring compensation as base / bonus could also alleviate this. at least that's the carrot finance industry is using to get most performance out of developers even though the bonus is usually exactly what promised/paid last year + n%. but it seems to work (especially around 3rd/4th quarter). but the higher you go, the higher % of your pay is bonus and supposedly tied to co performance. though this is not as good as equity as there is still no real ownership incentive.

"structuring compensation as base / bonus could also alleviate this" <-- I am not convinced. If you knew that your bonus was a percentage of your salary, the only way that it would motivate you is if your salary was low (or if your bonus was extremely large). And for developers, the bonus is generally in the 10-30% of base salary range. There really has to be a variable component depending on the firm profits that isn't capped at a percentage of salary.

"though this is not as good as equity as there is still no real ownership incentive." <-- this was the largest complaint to the investment banks transforming from partnerships to corporations (in the former case, the partners were personally liable for losses)

In fact, I think this problem is even worse in the CEO world due to the legacy MBA culture of greed.

You're talking about The Pardo Axiom:

"There's no shortage of smart, hardworking engineers. There's a shortage of smart, hardworking engineers willing to work for very little money." ~ David "Pardo" Keppel

If you're having trouble hiring, it's probably because you're not paying enough. It turns out that talented people are worth paying a lot for.

Reply on news.ycombinator.com