This article is an embarrassment. He does not offer any evidence at all for his claim that the no-poaching agreement drove down programmers' salaries. It may have been anticompetitive enough for the participants to get sued, but how much did it actually decrease salaries? 10%? 1%? 0%? If this guy knows the answer, he doesn't tell us.
And I know for a fact that Facebook is not the only reason programmers' salaries are rising. There are a huge number of new startups all competing for the best people.
You can prove it had an effect on the employees that worked there, and to a fairly exact number. The agreement covered all employees, not just tech workers. Combined that makes up about 350k employees, and you can then extrapolate how many were engineers in silicon valley, or engineers in general. That means that there was at least a proportion of the population that was affected very inversely by these agreements. If you then realize that other companies use the larger more successful companies' salary policies as the model for their own, then it would have impacted other companies as well, and the employees that work there.
That alone disproves your implied statement: That these agreements did nothing to anyone. It definitely did something, and just the fact that those companies amount to tons of employees and tons of salaries is enough to show it had an effect.
You however are pulling the oldest trick in the book. You want conclusive proof that these completely illegal actions by corporations had a measurable effect before you'll admit that they were wrong. But, it was a conspiracy. You're not going to get exact numbers. It's not like these guys are keeping a ledger of money saved by not-poaching. You are basically demanding the impossible.
But, you are also demanding more than you give yourself. Where's your proof this did nothing? I know, you didn't explicitly say it did nothing, but that's your clever trick. You just imply your counter argument and then run around dumping links to logical fallacies on Wikipedia rather than give real numbers.
So, Paul, stop being a hypocrite and give us your bet on what impact this really had. If you demand exact socio-economic evaluations before you'll believe this did anything, then I demand that you give the same evidence it did nothing.
And your first task will be to disprove what's already been proven in court: That these agreements did impact all the employees at these companies, that it did reduce salaries or they wouldn't have bothered, that reducing salaries was their goal, and that they did fire employees over it.
Until then, your demand for exact evidence violates the standard in philosophical argument that your own argument can't be used against itself.
Your comment suggests you don't understand the distinction between saying something is false and saying there's no evidence either way.
I'm not saying the no poaching agreement had no effect. I'm
saying he offers no evidence for his claim that it had a big effect. Complaining that someone has offered no evidence for
a claim is not an "implied statement" that it's false. Do
you understand that distinction?
Totally agree with zedshaw here. When a long-running, highly-secret conspiracy with demonstrable follow-through (HR staff being fired) is uncovered, the onus is clearly on the conspirators to prove that their conspiracy was harmless, rather than on the uncoverers to quantify its harm. I understand why a venture capitalist would be nervous about the implications of this conspiracy's discovery (higher programmer salaries = shorter runway for investments), but trying to downplay (with no evidence) the impact of the no-poach on developers is disappointing to see on a site called Hacker News.
Can't we use New York software engineer salaries as an analog here? It's another market where good engineers are highly sought after. $200k for an engineer with 8 years work experience is not uncommon.
Hedge funds pay $250k consistently.
Wouldn't SV look something similar?
Your argument is technically right but It sounds like an excuse to punt on a practice that most of us find f$cking despicable.
I think I speak for everyone here when I say that we'd love to have your brain figure out how much of an effect this collusion could have had on engineer salaries or how to make sure this doesnt happen again.
It is very hard to give numbers. But I think it is hard to argue direction though. If there was a no-poaching agreement in place amongst a "cartel", would salaries of the employees of the cartel be higher than normal? Certainly not. They could be the same (assuming there is a sufficiently large number of non-cartel employers) or lower.
This sort of argument is often made in Economics, and is logically sound. Ceteris Paribus aka everything else being the same.
I agree the article smells of link-bait. I don't understand why Facebook's IPO is different from every other IPO that has happened in the past. Why does the author feel that this IPO will raise wages of developers in the valley?
To claim that it's unclear what the effect of a no poach cartel is on engineering salaries is at the same time to claim that it is not necessary that there should be legislation preventing no poach cartels from existing.
Do you believe this to be the case? If so, do you really mean to say that engineering salaries should not be set by market forces?
PG is just saying that if it had been a competitive market we dont know the extent of how much it would have increased engineering salaries. Thats true.
However lets take the case of NEw York where you can get program trading jobs for $300K and I have friends who got these kind of jobs. Hedge funds routinely pay $250K plus.
I think this is what would have happened in the valley.
I don't think that at all makes sense. The finance business in general distorts pricing. There are a whole host of professions that pay significantly more in NY than they do anywhere else.
Trading companies. They're writing software in house to control stock trading (or that of any other financial instrument that's on the market. The software is so directly responsible for profit margins that the companies are forced to be very competitive with their compensation.
I was hoping for a more specific answer. I currently work at an algorithmic trading firm making software that trades securities, but I didn't find a position like this at $250K/year.
There is evidence that collusion in the job market does not necessarily have an effect on wages: http://ideas.repec.org/p/esx/essedp/626.html. Interestingly enough, the research "provide[s] strong evidence of a negative effect of collusion on labour productivity growth"
It doesn't seem to relate to this situation at all. I could only make it through about 15 pages due to the author's writing style.
Collusion for these firms was price-based collusion for products, not collusion to suppress the input costs of highly skilled labor. Manufacturers are said to have engaged in establishing minimum or fixed prices for their products.
Consumers will tend to absorb this form of market inefficiency. After the cartel agreements were broken, consumers reaped the benefits in the form of the "intensification of price competition" (lower prices).
It suggests that only in one case was there collusion in R&D, which might be a relevant example.
Additionally, the article cites outside competition from the USA as a factor. Competition from foreign sources on engineer wages such as India is still nascent. US manufacturing during this period was mature and robust.
> I don't understand why Facebook's IPO is different from every other IPO that has happened in the past.
its important because FB is the biggest social network on the internet. Believe it or not, many people still dont understand or get what "social web" means. Having FB stock publicly traded (company more transparent) and strong earnings would prove to the masses that web and social net is sustainable AND profitable AND here to stay. And it already works! look what happened with Zynga stock after FB IPO papers hit the daylight.
If we take into account that on Wall Street, there are engineers making $330K + for program trading jobs, then why would that not have happened here in the valley.
HEdge funds routinely pay $200+. With a fair market I think we should have seen similar numbers here in the valley.
There are many times more engineers in New York than in SV.
I was a trader at an investment bank's prop desk. My salary was $200k+bonus when I was 20. When I left a year later I was offered 350+150 guaranteed minimum bonus to stay. But I built algorithms from concept, testing, all the way through to design.
I had a team of coders and optimisers working for me whose salaries topped off at 150 all in, 200 for the one who could turn out assembly that ran in milliseconds.
I always saw my job as being finding in-efficiencies and irrationalities; they saw it as coding. I happened to code (and not all that brilliantly, hence the support). But the argument that pure developers should be compensated richly doesn't gel with the argument that programming is the new literacy.
This conversation is not about the "programming is the new literacy" - but I do agree on that part.
I'm trying to find an analogous system which could help us figure out what would have happened to developer salaries in SV given what tech companies pay in NYC.
How comparable would a web app developer and Wall Street dev be?
The basic financial literacy these guys are desired to have is higher than naught. Further, there are different technologies, paradigms, and cultures involved (though at a hedge fund the last part converges).
A test for this would be tech dev salaries in NYC versus SV.
The economics are different. I'm assuming you mean the guys writing the programs that run trading platforms. Those guys are super specialists and for their employers a small boost in performance means a huge boost in revenue for the day. Also from some first hand account I've read from here and on Reddit the job is very demanding. You can get called in at any time to write something that needs to be done by the next morning. It comes with its perks of pay but the trade off is freedom.
This gave me an interesting thought. You mention that those trading jobs are supposed to be very demanding, implicitly citing this as a reason for their higher salaries. Well, any time a subjective factor like that affects a salary, it is based purely on people's perception of the job. It seems to me that the typical perception of startup work is equally demanding. It's odd that instead of resulting in higher salaries, the benefit is diverted to the (statistically poor) hope of a big IPO or acquisition. This is ironic since the trading firms paying $250k+ to programmers are all about maximizing profits in an environment that could be even more volatile and risky than the startup world. It makes me wonder who has the smarter approach to risk...
Would be instructive I think to compare the average total compensation of startup engineers vs the average total compensation of hedge fund engineers. I wouldn't be surprised if the former were skewed a bit higher by the Microsoft, Google's, Facebooks.
I agree with this but it's tough to prove that there's a causation effect between the two (other than the one person who was fired, which by itself probably is enough to make a valid argument). But it seems like when the value add of a programmer is so high, it's a lack of mimicking the finance market to structure your compensation package that seems to be more closely correlated to the (low) average income. I say finance market because it's probably one of the few markets where compensation is directly and proportionally related to performance.
I was thinking that too. Though to be fair in this particular case they seem to be following the TechCrunch tradition of running linkbait guest posts on weekends (when there is less real news).
It's sad, given that so many good startups would love to get any kind of press coverage, weekend or not. Every time I see a worthless article on TC, I think, "There's 500 words that some good startup would have loved to have, and the audience probably would have enjoyed it more, too."
Would be interesting to see what percentage of posts covered the really big companies vs startups on there. Feels pretty skewed towards the ever small and often pointless detail of the big tech companies rather than startups.
Quoting another comment on this topic I made while inebriated to back up the idea that this is economics 101 material:
Libertarian views assume that free markets regulate themselves basically due to perfect competition. (Trying to keep this short and simple.)
Realistically, you do not always have perfect competition and you have problems with monopoly, oligopoly, etc. Firms are potent enough that their collective decisions can manipulate markets, and in this it would be labor markets. (Perfect competition assumes that firms cannot manipulate markets.)
Libertarians oppose a government presence in markets because they believe the result is inefficiently functioning markets, and we are all worse off.
In this case, the firms collude and are manipulating labor markets so that they behave inefficiently to the advantage of the manipulative firms and disadvantage of labor. (This is usually where people try to justify labor unions as a counter to the influence of firms.)
Agreed with most of what you said. However I find the cost of living in silicon valley to be an issue.
I think companies should do something really crazy. They should go and develop a new area which would eventually beat silicon valley. Where engineers can find very affordable rents, and very affordable modern deco houses. Where the weather is sunny, driving engineers who sit all day in offices to go outside and get some vitamin D. Where the gender ratio is balanced. Where your neighbors are good at other stuff and are willing to share there knowledge, like a real world quota.com.
A engineer, earning 120k from year 0 - 4, should very easily be able to save 250k and live comfortably in a 5 year timespan and not have to cut corners.
PG, there is a real problem here with silicon valley. I don't think the salaries are too low but I think the cost of living is too high.
If you truly want to be innovative and crazy, broaden your world beyond the web.
> I think companies should do something really crazy. They should go and develop a new area [...]
If they do that and it succeeds at all, here is what will quickly happen.
1. The area will be full of smart well paid people, and will be a place where those people want to live. House prices and rents will go up.
2. Since it was aimed at engineers, and prices are (now) expensive, it will tend to be occupied largely by engineers and other highly-paid people. Which, society being what it presently is, will lead to the same sort of sex ratio as you see in Silicon Valley.
End result: just one more SV-like area. And that's if it works, which it probably won't. (It's very easy to say that "companies should" develop an area with sunny weather and a balanced gender ratio and neighbours who are good at other stuff and willing to share their knowledge -- but not so easy to do it.)
Gender ratio is a consequence of our industry being gender imbalanced and tech being the superstar industry of the area. If you want to find all of that except for the climate, you can find it in Seattle. The wages are almost bay area levels, the housing significantly cheaper and better quality usually, no reams of crappy 1930s craftsman track houses everywhere. You have no state income tax and tech doesn't dominate as much, so it's significantly more gender balanced.
Austin also has that potential I hear too and it's a sunny place. Texas also doesn't have state income tax.
These agreements haven't gotten in the way that most engineers in the valley know there are a ton of other places they can work, including the "no poach" companies. Valley engineers switch jobs with an ease and frequency uncommon for the tech business in general.
Facebook alone certainly doesn't hire enough engineers for it to matter THAT much.
Compensation includes equity. One of the primary ways that programmers become millionaires is by gaining lots of stock grants and options by being early employees at quickly growing companies that go public. That whole system has been curtailed somewhat in recent years by SOX.
This article tries hard to give coders a warm, fuzzy feeling about how important they are.
As a coder working on a startup, I can tell you one thing I've learned: coding (at least for me) is not the hard part. If your dream is to be an entrepreneur, here's my advice: learn to code and learn to sell, or find a great cofounder who knows how to sell.
Yeah, that rubbed me the wrong way too. Engineers are completely overrated; your best bet is to hire no engineers and outsource to India or odesk, or if you absolutely must hire engineers, hire them for rock-bottom prices and fire them if they ask for more.
That is the dumbest piece of advice you could give a founder.
A start-up that outsources its core competency is doomed.
If you have an idea, and can't find a technical co-founder, learn to program yourself. PHP aint that hard.
Yeah, it must be a huge pile of valuable IP if someone without tech knowledge can hack it together with self-taught PHP in a few weeks...
Aren't startup people supposed to be all focused on getting results? If it takes paying someone else to hack up v1, rather than you spending time on it when you should be out selling or interviewing potential customers, etc, do that.
We never hear the stories from people who self-taught PHP or Rails for 6 months hacking on an idea only to then try to give it to users and have it rejected for multiple reasons. But they happen. Probably more than anyone realizes.
Skype outsourced it's development to Estonian developers, which as we know went very well. Though it's a prevalent opinion on HN, outsourcing is not inherently bad for a startup. What matters it the quality and the skills of the people, local as well as remote.
I think you're missing the point of the article. He's not talking about starting a company, he's talking about employee poaching (or lack thereof) by large tech companies, and its effects on engineering salaries.
But as an aside, I definitely agree about selling. There seems to be this disdain for MBA/business types that don't build. As an engineer I can tell you that I have the utmost respect for skilled salespeople and I would definitely want one on my side when starting out.
by time you will learn that coding and building sustainable software product are not same thing
there are ton of code monkeys but they will not produce product
Gibberish article, I’m afraid. As one of said engineers I don’t see how my employer is undercutting me by paying only twice the median, plus bennies.
The expense of Silicon Valley is largely down to growth policy, which is to say, they can’t build housing concomitant with demand. Similarly here in NYC.
Making very good money in an very expensive place is not a struggle, it’s a privilege of very few.
>The expense of Silicon Valley is largely down to growth policy, which is to say, they can’t build housing concomitant with demand.
I can't tell you how much this infuriates me. I actually heard a Mountain View city council member say, "Maybe we should tell Google that they can't hire any more people." Meanwhile, the city is full of termite-ridden rotting apartment buildings just waiting for the next earthquake to pulverize them.
The R3 zoning district (all that orange on the map) allows fairly high density - FAR 1.05 maximum and upto >40 DU's per acre. [Mountainview, Ca, Code of Ordinances, Chapter 36 Article XII available at www.municode.com]
However, the primary land subdivision pattern of many small lots (and consequently many owners) makes assembling parcels suitable for redevelopment at higher densities not just expensive, but difficult.
Furthermore, existing high rents mean that the value of the land based on existing cashflows may make redevelopment infeasible once the cost of new construction and land acquisition is amortized and the higher returns associated with the risk of development must be paid.
Obsolescent housing (and other structures) remain because they are profitable and can be valued on an existing income basis and thus traded with known risk.
Higher density projects may be possible, but the market may not be able to absorb the units if the rents are significantly higher - and of course, higher rents are also likely to be untenable for voters as well.
I don't think that's the issue. Look through the R4 zones on the sunnyvale zoning map. http://sunnyvale.ca.gov/Departments/CommunityDevelopment/Map... Almost every one of those is a beautiful new (tall) building that efficiently uses the space, and the plots aren't huge.
As one of said engineers I don’t see how my employer is undercutting me by paying only twice the median, plus bennies.
Your employer would probably pay you more if you had true freedom to take your labour to another company. Your employer seems to have agreed to a cartel to limit your ability to sell your labour.
If your employer is paying below market rate, he's screwing you, no matter how the market-rate for your skills compares to the median. Being a privilege to work somewhere doesn't make the screwing less wrong.
If they are suppressing the median for your job in that location then they are.
If its a very expensive place then your not making "good" money everything is relative -) and ok we re probably all better of than some black single mom in some decaying Rustbelt city it doesn't make it right for your employer to beak the law.
I keep thinking Dropbox is the next Facebook in terms of the next 100 billion IPO. Does anybody know what their option pool looks like? I know they want to be the hard drive of the internet, and will probably happen once their latency gets better (I recently built an App where I really, really wanted to use them but a minimum of 1.2 seconds latency just to _list_ the files in the directory was way too much).
Unfortunately for Dropbox, cloud storage doesn't really have the network effect that social networking does. Anyone can clone Dropbox and if the price is right, users can switch without suffering any hardship. This makes Dropbox look like something that will be bought rather than made public. Amazon can provide storage infrastructure for cheaper than Dropbox can, but they haven't written good client-side software. Buying Dropbox would fix that.
I understand why, the software works really well. But given the right amount of engineering resources, anyone could duplicate it. Nothing is keeping you on Dropbox other than the quality of the software, so if better quality software comes along, you'll switch to that and Dropbox will be a distant memory of the past.
Compare that to Facebook where you can't switch because your friends won't switch, and you only use Facebook to interact with your friends. Even if there is a better social networking site than Facebook, it won't matter because nobody will use it.
These are two different kinds of moats. Yes, Facebook has very strong network effects that create an advantage. However, it's a mistake to underestimate the strength of Dropbox's engineering complexity and strong product design. It wouldn't be simple for anyone - including Google or Microsoft - to just clone what Dropbox has done. Apple, Google, or Microsoft would have to decide that they were comfortable treating platforms other than their own as first-class citizens, and be able to defend the incredible emphasis on simplicity all the way through the product's lifecycle. Facebook would have to dedicate themselves to QA and quality-rather-than-velocity in a way they've shown they don't like doing. Anyone else, well, like I said: good luck.
Dropbox has shown the huge need people have for their services. Why wouldn't the big guys (Apple, Google, MS) be compelled to now get in the game and make the investment in treating platforms other than their own as first-class citizens?
One possibility is the amount of money to be gained. According to http://www.quora.com/What-is-Dropboxs-revenue, "revenue is on track to hit $240 million in 2011". Google, on the other hand, made $37,905 million in 2011. Microsoft made $69,943. Apple requires too much effort for me to get the numbers, but it's more money than either Microsoft or Google.
So my theory is, "why bother?" If one of those companies needed a few more million, there are easier ways of making the money.
There are articles on the subject, I can't recall them off the top of my head. The gist of it is that Dropbox is more than just cloud storage. They've put an unbelievable amount of work into the syncing client itself on all platforms so it's really robust, very cross platform, while at the same time uniquely tailored to each platform and completely seamless in operation.
And anyone can clone it, which is the point. If Google decides to launch Gdrive, it would dwarf Dropbox instantly. Whereas even if you were given all of Facebook's hardware and software and staff, but not user accounts, you would have a hard time competing with Facebook.
I'm not sure about that. Google Music's sync client is horrendous, as were Google Desktop and Gmail Notifier. I wouldn't trust Google to be able to make a client on par with that of Dropbox. (Remember, Dropbox engineers actually reverse engineered the Finder and performed code injection in early versions to get the user experience they wanted.)
A singular devotion to UX is not something that can be duplicated given "the right amount of engineering resources". (jrockway's thinking is totally emblematic of Google and Microsoft.) It requires the right, most product-focused people in charge.
My understanding of economics and game theory lead me to believe that the free market will eventually take care of this "no-poach" cartel (looks like it's already happening). If I recall correctly, this agreement can be reduced to a "prisoner's dilemma" game: the rational decision for a given company is to cheat the agreement or simply not take part in it.
Perhaps economics and game theory will in the long run but here and now the Department of Justice and the class-action plaintiffs are doing a good job taking down the cartel.
Most of todays internet tech companies don't have the shelf life of Microsoft. Many will be has-beens in 10 years time (or even 5!). So you want to take the cash upfront.
(Disclaimer: Not a real survey.) Doing a spot-check of big tech IPOs in 2004 (first post-crash year with a significant number), there do seem to be pretty inconsistent outcomes.
Big IPOs that year: Google, DreamWorks Animation, InPhonic, Shopping.com, SalesForce.com, Atheros, PortalPlayer, Shanda.
Of those, Google, DreamWorks Animation, SalesForce.com, and Shanda continue to be successful. Shopping.com was acquired by eBay within a year at a modest premium over the IPO price (15%), and Atheros was acquired by Qualcomm after 7 years, for about 3x the IPO price (also around 15% annual return). PortalPlayer was acquired by Nvidia after 3 years for 20% below the IPO price (and 45% below the first-day closing price). InPhonic went bankrupt after 3 years.
Most of the companies back then didn't have that kind of shelf life either. Making an investment in an early stage company is inherently risky.
If your goal is to get rich via investment opportunities I would not recommend trying to win the stock option lottery via the employee option pool as the best strategy to achieve that goal.
Depends if Facebook is the main way to socialize for the next century. Imagine having that much data on that many people and owning that channel! That's worth a lot of money.
Yes, considering the previous Facebook didn't nearly last as long. I'm thinking that social networks don't have nearly as high of a shelf life than you think.
When it comes to investing though you invest in a management team and not a product. The rate of change is increasing, all the products on the market today will be well and truly obsolete in a few years, even the market opportunities themselves could change
It's things like Facebook putting 'the hacker way' into their DNA and being able to retain top talent that makes the company worth anything. If you took the facebook source code and all their customers away they'd probably have a pretty good chance of making something else that's quite good. Today anyway, but if tomorrow they can't it'll be because they somehow lost their way and let the MBA's in. Kodak had a great place in the market but they just couldn't successfully adapt
I haven't looked at the price but if I was in the USA I'd probably buy some shares, the google float of $100 seemed outrageous at the time, and unless they get further involved with Microsoft I'm pretty bullish that they'll keep going up
In my mind, that's the big gamble with Facebook. It could either be THE source for socializing for 100 years (and hence very valuable), or it could be another MySpace (etc.), and hence not worth a lot.
The cynic in me, would say that people who own stock now don't think it would last and are getting out when it's at it's top.
Not even the entire Bell system would be worth $1 trillion today, and unlike Facebook, people actually had to use the Bell system, it wasn't just a source of idle entertainment.
Waaah, waaaah! Are these self-entitled vermin seriously trying to claim poor house?
It's not anyone else's fault you choose to live amidst a hyperinflated housing bubble, so quit bitching that it's so expensive. The rest of us don't whine about not being able to afford Manhattan penthouses.
This type of sob story only reinforces the external prejudice that the Valley is a phenomenally insular echo chamber, wholly disconnected from all semblance of national or global reality.
As Silicon Valley salaries rise, they will eventually reach the point where more employers will decide to spend their personnel dollars elsewhere. Engineers in Austin or Ann Arbor or Cambridge (either one) can work out of local offices, or straight out of their homes, and produce more lines of useful code per dollar than their peers in Palo Alto, Mountain View, or Sunnyvale. Go ahead, call it "inshoring" and look down your nose at those cheaper programmers if you like. They'll still be taking your jobs, and you'll be left living in a very expensive place after the salary bubble collapses.
The only question is when. I think the process has already begun. Some might predict another year or two. Nobody sane would expect the ride to last longer than that before supply and demand reassert themselves over mere fashion.
And I know for a fact that Facebook is not the only reason programmers' salaries are rising. There are a huge number of new startups all competing for the best people.