Hacker News Reader

I agree that there were larger, diffuse consequences for a great many people, but the mass tort system isn't really designed to remedy that kind of thing.

The most appropriate mechanism for that is government action. There was such an investigation, and it did find wrongdoing, but the ultimate response was so feeble as to be meaningless. The DOJ sued the tech companies under the Sherman Antitrust act which reads in part:

Every person who shall make any contract or engage in any combination or conspiracy hereby declared to be illegal shall be deemed guilty of a felony, and, on conviction thereof, shall be punished by fine not exceeding $100,000,000 if a corporation, or, if any other person, $1,000,000, or by imprisonment not exceeding 10 years, or by both said punishments, in the discretion of the court.

After filing suit, the government settled the case in exchange for a promise not to violate the law for five years. Given that they were already obligated to follow the law, and not just for five years, one wonders what the point was.

> the government settled the case in exchange for a promise not to violate the law for five years.

Wonder if we apply that to a bank robber who took $1M. Just make them pay back $50K, and promise to not to rob banks for 5 years.

> but the mass tort system isn't really designed to remedy that kind of thing.

I'd say prison could be a great remedy for that sort of thing.

I was very disappointed not to see prison sentences for executives and other fiduciaries in the 2008 financial scandals. Weren't those used in the 80s S&L crisis? I understand that large corporate penalties can hurt society or innocent employees, but we can still deter bad behavior if we hold individuals accountable. Of course the rich will always be more likely to evade justice, but I feel like 2008 was more blatant than usual and sent a strong message that if you're rich you can act with impunity.

So am I, but look at the vitriolic 'class warfare' narrative that opponents of the administration have been carrying on for the last 6 years. Every time the government makes a bank pay out money to settle accusations of financial asset misrepresentation (eg some of the shenanigans around mortgage-backed securities), the Wall Street Journal and its ilk wax wroth about 'government shakedowns' and so forth. So imagine the response if the DoJ launched criminal prosecutions of big bank CEOs in addition to fraudsters like Bernie Madoff - there would be cries about a communist takeover and what-all else.

It's also true that the regulatory environment was sufficiently lax, and the chain of responsibility sufficiently convoluted, that it would have been hard to make any such prosecutions stick. Financial crime is generally slow to come to trial anyway because it's so hard to prove intent beyond a reasonable doubt, and so hard to pin it on any individual in a sufficiently large organization. As well as (possibly) leading to political upheaval and (definitely) causing a major chilling effect on business activity in the middle of the worst financial crisis in almost everyone's living memory, it would have been ethically, legally, and strategically problematic to bring charges that would not be easily provable at trial. If the CEO of (imaginary) MegaBank had been put on trial and the case collapsed, it would have poised every other prosecution for 20 years. That's not rational, but at the national level prosecutions have a big political dimension, there's just no getting away from it.

My read is that the administration opted to go the route of administrative penalties with the twin goals of maintaining relative stability in the financial industry and getting Dodd-Frank passed on congress to provide much stronger regulatory options for the longer term. It's not an emotionally satisfying strategy, but a pragmatic one taken in the interests of continuity. I would have preferred a top-to-bottom shakeout of the financial industry, and maybe a change in the economic and social order that would result, but the problem with revolutionary changes is that once begun it's hard to stop them turning. Look at the French revolution; half the instigators ended up on the guillotine themselves and 10 years after France had got rid of a king they ended up with an emperor.

> "64,000" people were not effected. Those numbers are just the work forces of the companies involved in the lawsuit.

But the class is defined as just the employees of these companies from 2005-09. This has two implications: 1) the proof issue is much easier--you don't have to get into a calculation of economic damages for the sector as a whole; 2) the settlement release theoretically doesn't preclude people who weren't employees of these companies from bringing a different lawsuit based on the indirect injury.

Don't underestimate the significance of (1). You're talking about a battle of economic expert witnesses, where your burden will be to prove that the defendants' conduct actually suppressed salaries during a time period when engineering salaries were skyrocketing. Whether or not that theory of the case is true, I don't know if its winnable.

Also, using 64,000 employees as the denominator is a bit misleading. Some people might have gotten a lot more money over the time period by taking an offer that was prohibited by a no-poaching agreement. But other members of the class may not have gotten those calls anyway, and so weren't harmed.

EDIT: 'gojomo and 'harryh hit the nail on the head. The theory is certainly plausible, proving it is very ambitious.

I agree with everything you said, and it makes sense for a trial. But people shouldn't think the only people affected were those that had a strong case in court. This salary fixing scandal is black spot on how American companies treat workers.

(It's not "hire Pinkertons to shoot strikers" bad [1] or "pay employees in our own private currency that only we accept" bad [2], but its pretty damn bad.)

[1] http://en.wikipedia.org/wiki/Homestead_Strike [2] http://en.wikipedia.org/wiki/Company_scrip

> How many new technology companies did not get created because people's depressed salaries reduced how much they could save to bootstrap something new?

I agree with everything else you're saying except this. Higher wages could also cause less people to consider founding a startup.

True, higher pay would allow people to save more and thus making it financially easier - but it also would mean a higher opportunity cost [for the same risk], making it psychologically more difficult.

Good point. There also is the case that a lower salary might make the person more likely to try side products to supplement income. Perhaps my original thought here was a reach.

We're talking about physical barriers to entry. Without the money, you can't start the business. If you have the finance, but with a higher opportunity cost, you are still able.

Another great contribution to society by Steve Jobs.

I know this is kind of a throw away poke at Steve Jobs, but its very believable as well.

In fact, the selfishness that Jobs displayed repeatedly in his life helps you understand how this whole thing happened. I can totally see Steve Jobs says "damn it, I'm trying to make cool stuff and these 2-3 other companies get stealing my guys. Stop doing that so we can go back to making awesome stuff!"

But that's the problem and my broader point. These companies acted to reduce churn of top talent and possibly secondarily to save some cash. I can see this seeming like a good idea. But the broader ramifications of their actions are insane.

They could have equally reduced churn by offering better compensation packages... and letting the market forces do what market forces do. Particularly from such a libertarian, "the market solves everything" set of people. (not a position on libertarianism itself, just pointing out the irony in big name capitalists acting against the market).

Technically, nobody was "effected". They were, however, "affected"

http://grammar.yourdictionary.com/style-and-usage/affect-eff...

Also, http://www.gingersoftware.com/english-online/spelling-book/c...

Yeah, I was able to hold out for the first two or three times, but by #5....

What is your evidence that the entire "sector" was depressed? Hand-waving about "Billions and Billions" doesn't count.

During the entire period of the "non-poaching" agreement, many other non-cartel bidders for tech talent were vigorously operating in the same market. If talent outside the cartel was underpriced, others would pay its full value.

Also during this whole period, industry salaries soared, and the number of opportunities at smaller companies (non-cartel bidders) multiplied.

Unless you spent time as a long-tenured employee at one of the specific cartel companies, their non-poaching agreements likely did not affect your compensation level.

And, to the extent cartel employees realized they could make more in the (large and growing!) non-cartel world of startups and smaller firms, they could (and did!) leave quite often. If you've worked with ex-Apple, ex-Google, ex-Intel, ex-Adobe, etc people, your projects' success and resulting total compensation levels may actually have been higher because of the stinginess of the cartel.

If some of the biggest employers in the bay area artificially cap the salaries of their workers, of course it will effect how other companies price their openings.

Three Hypothetical, but perfectly plausible conversations (no one would ever negotiate like this, but ignore that, these are inner motives/dialogs to some extent)

Company: "And what were you thinking for starting salary?"

Interviewer: "I'm at Apple now, I make X. I'd like 1.15X"

Company: "Done! I was going to pay you 1.5X but whatever"

or

Interviewer: "My asking pricing is X"

Company: "Well, I know DevOps people working at Google make .75X, so we will offer you 0.95X"

or even more likely:

Slightly drunk friend #1: "I really hate my job, and I'm only making X. Am I underpaid?"

Slightly drunk friend who works at Adobe: "Nah man, I only make X if we hit our bonus numbers! At least its a stable job with benfits, unlike all these startups that flame out."

Friend #1: "Yeah, I guess you're right."

I see, you have no evidence for sweeping claims of "Billions and Billions". You only have some 'just-so' stories and intuitions. But in each of your cases, a person checking compensation at just a few non-cartel places would do just fine.

So, if the cartel employees and their friends in your conversations stay slightly-drunk/stupid/lazy, they'll keep working at rates lower than are available to them. Meanwhile, anyone shopping around gets paid a competitive rate, unaffected by the non-poaching agreements.

I can believe there's some localized harm there, especially for long-tenured people at the cartel companies, but any damage is self-limiting. It's not massive on an "entire greater Bay area technology job sector" basis.

Also note, these cartel companies were also constantly hiring from non-cartel sources, and there was no coordinated "cap" or "ceiling" on such offers. They had to win such employees in unrestricted competition versus both cartel and non-cartel companies – and they often did!

Since employees talk (as in your examples), even long-tenured in-cartel employees would then get info about fully competitive compensation offers, and up their expectations for the next compensation-review. So again, any damage from non-poaching collusion faces practical checks on its magnitude.

If you're an Apple/Google/etc employee who wishes you'd been cross-recruited with a big compensation offer, you have a right to be angry. (But, you're still probably not that angry, given how well things have been going for top employees at those places in the relevant era!)

But if you've been employed elsewhere, your anger is misplaced. The non-poaching agreements didn't hurt you... and maybe even helped for the reasons I gave above.

Software engineers as a rule are grossly underpaid relative to the actual value they provide, whether they are employed at Google or Web Body Shop X in Podunk Midwest Town Y. Collusion of this sort is deleterious on the whole precisely because of the indirect but very much desired effects of wage depression, even if some individuals benefit and others are more disproportionately harmed.

What's your evidence for "grossly underpaid"? Do all of the tens of thousands of employers meet in secret to hold the line against higher salaries? Are there laws against software engineers starting their own firms where they'd be paid their full value?

Your claim doesn't pass the smell test. Pay levels are set by a giant, uncoordinated, massively competitive process.

If you know any software engineers who are grossly underpaid for the value they can provide, hire them! Then "underpay" them just a teensy bit, instead of the "grossly underpaid" exploitation they suffer under now. Pocket the difference for yourself, as a reward for seeing their true value so much better than every other employer.

Where is your evidence that pay is "set by a giant, uncoordinated, massively competitive process?" Pay is set by a variety of factors that generally favor the wealthy and connected, usually in such a way that workers rarely benefit from expanded wealth (wage stagnation). Additionally there is a coordinated effort (see: any Koch political outfit or its allies) to weaken and destroy unions, which are the principle means other than law that laborers gain in wages and working conditions. Moreover, there is very strong evidence in the form of the subject of this thread that pay is set, at least to some non trivial extent, by acts of collusion.

It seems to me you believe very strongly in a black-and-white world where "the market" functions just fine. It is terribly out of sync with actual evidence, though, which you simply brush aside while scribing arguments containing numerous unsupported assertions and demands that others support their statements. As if we should simply accept your point of view as the truth by default.

There's no "wage stagnation" in this industry. Also, for the positions under discussion, no unions to "weaken and destroy". I can't tell if you are a software engineer in the bay area, but as one trust me that there are hundreds of high-tech employers who don't coordinate, and they do spend a lot of effort and money to attract and retain employees.

You know who wasn't part of the alleged cartel? Facebook. Microsoft. Yahoo. Amazon. Oracle. Salesforce. Twitter. Yelp. Autodesk. HP. Cisco. VMWare. AMD. Workday. Electronic Arts. Zynga. Netflix. Etc. Etc. Etc.

Also, no seed- or VC-funded or YC-backed startups of the past 10+ years were part of the cartel. Also, no consulting/contracting shops were part of the cartel.

You seem to be applying some sort of undifferentiated-laborer/factory-floor model to a region/industry where it doesn't apply.

There certainly are market failures; when they happen, they can be studied and quantified. There's no evidence of that here. All we have is angry crusaders with grandiose damage claims, who probably weren't even affected by the no-poaching agreements.

maybe less than the US but the average for devs in London UK is less tan a tube driver gets.

An it is statistics 101 that if you change the skewness and kurtosis of a distribution if effects not just those at the top.

I see it the other way around. Collusion in hiring and wage fixing is illegal, and it is considered serious. It can be considered a crime.

It is difficult to establish the exact damages in a situation like this. This is one of those things we have made illegal because we understand it is very harmful, but it's difficult to determine just how harmful. And yes, it is important to provide evidence of harm.

But here's where I start to deeply disagree with you. This isn't a situation where two parties were both operating in presumably good faith, and some damages occurred, and we're trying to figure out who should foot the bill for what damage. This is a situation where one party behaved highly illegally, and based on the emails, there's really no question it was ugly.

So ok, I'm not going to let plaintiffs pick a number out of thin air, there needs to be some justification. But I'd say at this point that the burden should be more on the defense. Why shouldn't we say billions at this point?

I also probably disagree with you deeply on how harmful this sort of thing is.

If "these cartel companies were also constantly hiring from non-cartel sources," then why did they enter an agreement to stop poaching employees from each other? Clearly it was concern enough to potentially violate the law to stop or slow.

Obviously they were "hiring from non-cartel sources" – after all, their staffs grew significantly, at all experience levels, during the no-poach period. Those people had to come from somewhere else, and I've seen no allegation the cartel coordinated on offers to this large pool of external people.

I believe the companies entered the cartel because staff churn is highly-disruptive to ongoing and innovative projects. Churn leaks proprietary info, it deters investment in staff development, it inconveniences or demoralizes those that remain, and it spreads an often-corrosive mercenary attitude.

Also, bidding wars for cross-recruited employees will often mean the "winner" suffers the "winner's curse": they are the bidder that erred in overestimating value. And while that delivers a temporary cash bonanza to the employee – "woohoo, this year I'm significantly overpaid!" – it could mean an unsustainability that later hurts their career and lifetime earnings.

The best projects and workplaces instead have a strong continuity-of-staff, where firm-specific expertise compounds over time, and teams "gel" to become familiar, highly-effective collaborators. As a result, the welfare of the firm and the employee are not in zero-sum opposition in such matters! Many cartel employees, and especially average employees, may have gained when their workplace stability and productivity was protected from constant poaching. You'd need a lot more study to know for sure. But the law isn't often based on economic analysis, or even domain expertise, but more often on lawyer/legislator superstitions.

It's not helpful to think of this as something perpetrated by cartoon villain managers, twirling their mustaches thinking of ways to sneak a few dollars out of their poor, ignorant workers' wallets. It was talented tech managers doing what they thought led to the most sustainable and productive teams, made up of highly-educated and well-informed professionals. Consider: the very middle managers who needed to implement this policy, for their own hires, also knew it applied to themselves as well.

What evidence do you have that "The entire greater Bay area technology job sector" experienced depressed salaries as a result of what these companies did?

AFAIK the companies involved agreed to not solicit each other's employees. While somewhat shady, it's not on the same level as an agreement to cap wages. Basically the damages are covering the crime of not affirmatively using your recruiters to go after people.

The "solution" sounds preposterous- You are forced to attempt to poach another company's employees, without regard to any other aspect of the business, such as maintaining good relationships with your suppliers or partners.

"the companies involved agreed to not solicit each other's employees. While somewhat shady, it's not on the same level as an agreement to cap wages."

It was more than an agreement not to solicit each others employees, it was an agreement not to hire each others employees. [1].

"Alan,

We’d strongly prefer that you not hire these guys.

Steve" [2]

These companies created an agreement which prevented people from getting paid their market rate. John Siracusa does an excellent job discussing this at length on ATP [3]

[1]- http://pando.com/2014/03/22/revealed-apple-and-googles-wage-...

[2]- http://pando.com/2014/03/27/how-steve-jobs-forced-google-to-...

[3]- http://atp.fm/episodes/59-the-little-puck-that-could and http://atp.fm/episodes/76

The solution is not: "you're forced to poach" - the solution is: "you're not allowed to collude."