It’s not “Early” Exit Disease, just Exit Disease, that’s killing innovation outside of (and inside) Silicon Valley

This is in direct rebuttal to a TechCrunch article by Adrian Fortino that asserts that “Early Exit Disease”, meaning a willingness to sell one’s company at a low price due to risk-aversion and a lack of ambition, is what prevents the Midwest from becoming a center of technical innovation. He’s mostly wrong. It’s not Early Exit Disease that’s killing the Midwest (and New York, and even the Valley). It’s just Exit Disease. In the VC-funded world, these companies are not built to last on their own for ten years. They’re built to be sold, either to dysfunctional large corporations forced to “acqui-hire” talent at a panic price, or to the public markets.

When a company is built to be sold as quickly and risklessly as possible, innovation in a technical or cultural dimension is generally unaffordable. Technology companies should strive for the ideal of open allocation, which enables a large number of employees (instead of a protected, and often resented, few) to participate in innovation, but if anything but a sale in 3 years is to be regarded as failure, you will instead see the aggressive closed allocation of “Agile Scrum“. This is exactly what you’d expect in a world where CEOs plan for sales rather than building toward a long-term vision: a CEO who expects to eject within the next 3 years is not going to give a damn about company culture.

The current startup world, in truth, combines the worst of both worlds between the short-term and long-term orientations. Highly-compensated employees are expected to take most of their pay in the form of illiquid equity that may or may not pay off at some unclear time in the future, but with the aggressive race-to-sale attitude that the VCs require the company to have, they’re all forced to chase after short-term targets and to mind unreasonable deadlines. The result is the proliferation of boring companies (because they can’t afford technical risk) with uninspiring, command-and-control cultures. At one time, the startup world may have been populated mostly by highly-compensated experts at the top of their game, but now it runs on “Agile” commodity labor: inexperienced, cheap, untalented and sloppy programmers hired en masse and steered into open-plan bullpens for reasons that have more to do with appearance (investors like open-plan offices) than productivity (at which, these offices are disasters).

What does this have to do with the Midwest, noting the geographic aspect of Fortino’s claim. What’s wrong with the Midwest? Is there something wrong with the Midwest? That’s a big topic, and I’ll have to take it on some other day. Fortino says, “There will never be a true Silicon Valley clone in the Midwest”. I agree. That’s a good thing. Silicon Valley was at its best when it wasn’t trying to be something else. Now, at the upper levels, it’s a second-chance purgatory for people who couldn’t hack it on Wall Street, and the results are pretty awful. There’s no value in trying to clone Silicon Valley, a process that is likely to transport the negatives as well as any positives of that culture. The current Silicon Valley’s way of doing business is used up, played out, and now in mediocrity mode. There’s still money to be made in that game for the well-connected, but trying to replicate it in Ann Arbor or St. Louis is a waste of time. Beating Silicon Valley requires doing something fundamentally different, and the first step is realizing that (as California entrepreneurs did in the 1980s) top technical talent is, once again, undervalued… and figuring out a way to find, deploy, and compensate such talent in a way that existing players do not.

As far as I can tell, Fortino’s central thesis is that the Midwest lacks the massive exits that “create” millionaires in large numbers. He credits Facebook for turning 1,000 employees into millionaires, and argues that their conversion into angel investing is what keeps the Valley going. There are several problems with this claim. First, which is better for the world: one company making 1,000 people millionaires, or a hundred smaller companies making 10 people millionaires? The former is news-making, the latter is more ordinary, but adds more benefit to society. Second, let’s be honest about who makes most of the money in those big exits, aside from investors (the house always wins): executives, not engineers or technologists. These aren’t garage startups making new millionaires because the core ideas happened to just be that good, but bureaucratic companies that have tens or hundreds of non-technical VPs by the time of liquidity. Now, I’m not saying that these people are necessarily less deserving (that would be another topic) of payoffs, but they’re not exactly the sorts of people who are known for vision. Are VPs of BizDev going to build the future with their million-dollar payouts, or are they going to fund their B-school buddies? Thirdly, it’s not angel capital that makes Silicon Valley what it is (for good and bad) but the local venture capital industry. Angel investors participate because the venture capitalists are there, and not vice versa, so the driver is Sand Hill Road, not the network of well-unemployed new millionaires that liquidity events leave behind.

The Midwest is not, and cannot be, Silicon Valley. It probably has more talent in total, but it’s spread out, and I don’t see that changing. It’s a 7-hour drive between Chicago and Minneapolis. While I think that the long-term economic prospects for this region of the country are strong, I think that it will start to shine when technology moves into a post-locational phase, and not before, because there isn’t an obvious winner (like San Francisco) among the much larger set of high-quality locations. Cities like Chicago and St. Louis are going to need interconnection and alliance on a global scale, and not the insular arrogance that we’ve seen in the Bay Area.

More to the point, though, I don’t think that the build-junk-and-sell-it game has much gas left in it, and I certainly don’t think that there’s enough opportunity for those who enter it, this late in its cycle, to justify the cultural sacrifice. We need to think differently: to get back to creating companies worth caring about, in a technical and cultural sense. This isn’t going to be achieved by chasing “unicorns”. It’s going to require tackling the under-capitalized mid-risk/mid-growth (that is, targeting 20-50 percent annual growth, but without the chronic existential risks of the VC-backed unicorn, and with a focus toward becoming profitable) space; that is, the companies too risky for bank loans but not risky enough for headline-obsessed VCs. It’s going to require transmuting the coming startup downturn into something else: a Flight to Substance. Should it happen, it may or may not occur in the Midwest, but there’s no reason that it can’t.

Y Combinator and Paul Graham are bad for the world, Part 2: Fixing founder quality

Part 1 is here

It’s easy to recognize that the current state of Silicon Valley is unacceptable. Innovation has nearly ceased, and what we now have is a resource-extraction culture with no sense of ethics or even a coherent vision. So can we fix it? Or, can we replace it with something better? There’s good news and bad here. The good news is that the problem is simple, obvious, and demonstrably responsible for many of Silicon Valley’s failures. The bad news is that, though simple in etiology and nature, it may be complex and difficult to fix. What’s wrong with Silicon Valley: founder quality.

Almost all of the cultural illnesses of Silicon Valley can be traced to the fact that most startup founders are shitty people. Why are open-plan offices, despite their unacceptable health load, in vogue? Low founder quality. Why is there so much sexism and ageism in the VC-funded world? Low founder quality. Why are there hundreds of IUsedThisToilet apps but few companies doing fundamental research? Because the people being selected as founders aren’t capable of solving real problems, which tend to be way outside of their intellectual frequency bands. Silicon Valley is shitty because it’s creating shitty companies, and it creates shitty companies because VCs are selecting shitty founders.

The irony is that many venture capitalists love to say, “We don’t invest in business plans; we invest in people.” Y Combinator even invited founders to apply without an idea. That canard about “we invest in people” is about the worst thing for that set of individuals to say, because they do such a terrible job of it. I honestly can’t evaluate whether they invest in good or bad business ideas. I don’t think that they know what they’re doing, on that front, but I don’t either, so it’s a wash. When it comes to picking people, however? They couldn’t be doing a worse job of that. Here’s the disturbing part: I think that it’s intentional.

Obviously, I don’t believe that venture capitalists wake up in the morning and say, “I’m going to fund some horrible human beings today!” When it comes to the personal qualities of those they fund, they don’t care. And perhaps they shouldn’t, since their only interest is selling a company for a short-term profit, and the effects of the personal qualities of a company’s leadership tend to unroll over the long term. However, there’s something in their selection process that benefits the worst human beings, and we see it in corporate management as well. Why, despite the supposedly higher level of personal and ethical scrutiny at such levels, are there so many horrible people in managerial positions? Ultimately, most executives who are hiring middle managers are biased toward rewarding those who’ll favor the interests of the executives over the interests of those being managed. The thing about “those being managed” is: they’re not stupid. If it’s obvious that their managers don’t really care about their own growth and advancement, then they’ll become minimum-effort players and the team will be mediocre. Ultimately, this creates an environment in which the people most primed to succeed are the duplicitous: those who will consistently favor the interests of executives over those of the workers, while misleading the workers enough that they actually work hard, and keeping the ruse for long enough to get themselves a promotion.

Over time, this dynamic leads to a moral corrosion of the organization, as well as pervasive distrust. Companies get to a point where the workers don’t trust the managers to look out for their career interests, and the managers don’t trust the workers to get anything done. This leads either to bureaucracy or, worse yet, to its mean-spirited and borderline-psychotic cousin: the micromanagement death spiral, in which aggressive management leads to diminished performance, leading to even more abusive management practices. Micromanagement death spirals can surround individuals, teams, projects, or even whole companies, and they tend to be “fatal” insofar as the only thing that reverses them is for people to leave the company.

Startups were supposed to be an antidote to these “big company” woes. In the 1990s, it was because of the bureaucracy, misaligned incentives, and self-dealing management that 20-person startups could compete with giant companies. We no longer see that. In 2015, VC-funded companies either become giant corporations themselves (quickly and sloppily, with exactly the kind of culture that you’d expect) or they are shut down. What used to be about a legitimate “10x” (or more) factor is now just about king-making and throwing resources around. These VC-funded startups aren’t efficient or elite. Mostly, they’re in existence because the mainstream corporate world still hasn’t recovered from the catastrophe of 2008, and for many people, they’re the only game in town.

Why has founder quality tumbled in the past 20 years? More importantly, is there an intrinsic reason why the VC-funded ecosystem might require low founder quality? The first question is easy to answer; the second, not as much. Founder quality is low because VCs fund people whom they trust to put the VCs’ interests over those of employees. It’s the same dynamic as what afflicts middle management. In the inevitable trade-off between performance and control, they’re favoring control. Is this atypical? Is it irrational? I don’t think that it is. No one cares about performance, of a company or an ecosystem like the VC-funded world; people care, much more, about their own careers and social status. This isn’t an indictment, because I’m the same way. Managers favor control over performance insofar as they want to remain managers, and same with founders and venture capitalists.

Where does Y Combinator fit into all of this? Y Combinator, by design, selects for the young, naive, and reverent. It has defined that sort of semi-precocious but fundamentally inexperienced person as the ideal startup founder. Paul Graham explicitly put an upper age bound of 38 on startup creation. Why so, and what does this really mean? Graham claims that it’s about the ability to work long hours. First of all, sustained long hours aren’t productive for anyone, young or old. Second and more importantly, I’ve met plenty of people over 50 who are just as capable of working hard (and, usually, more efficiently) as their younger peers. What’s different, in the older workers, is that they don’t make sacrifices for pointless reasons, such as working 12-hour days to improve a manager’s optics. In other words, they’re harder to control.

It’s not by coincidence that the incubator with the most prestige is one of the most ageist ones. It’s not the ageism per se that attracts venture capitalism to Y Combinator, but the YC Vision of the ideal founder: a starry-eyed, inexperienced, overly eager person who’s easily swayed by charismatic personalities (especially charismatic people whose achievements occurred in a different century) and who’s easy to control. It’s not that these people are picked for their business acumen. They aren’t. They’re picked because they’ll follow orders without complaint or question.

The swirling conclusion… and what it means

Over the past ten years, I’ve seen Silicon Valley become a lot more ageist, as well as more aggressive in micromanagement. Open-plan offices dominate, Scrum boards and “story time” meetings are the norm, and engineer time now has to be justified, in many places, down to a two-week increment. So what’s happening here? It’s our good friend, that micromanagement death spiral, on a massive scale. The larger part of a whole industry is trapped in this undesirable (and often unfixable) pattern. The erosion of trust isn’t limited to one team or even a company. It pervades the whole industry.

How will it end? Badly. We can’t fix the Valley’s founder quality problem without a catastrophe that’s big enough to make venture capitalists care about it. Firing founders is hard, so it’s hard to imagine the needed flush-out happening except in the context of a downturn that closes many of these startups for good. This won’t happen without innocents being affected. We could be in for a repeat of the early 2000s.

Moreover, a micromanagement death spiral is symptomatic. It signifies weakness at higher levels, and perhaps in the business as a whole. People don’t micromanage, in general, when things are going well. When it comes to the performance-versus-control trade-off, executives tend to favor performance during periods of opportunity and expansion, and control when things aren’t as good. On the dog’s-eye level of the worker, if one has a micromanager, this means that one’s manager doesn’t have enough real work to do. Up a level, a company defined by micromanagement (e.g. “a Scrum shop”) is one whose executives are floundering. So, if the VC-funded ecosystem is increasingly defined by micromanagement, then what does that suggest about it? That it’s reaching its limits, and that there isn’t a lot of useful work to be done in it. The picture that comes from that is… imminent decline. Will it take three months or five years? That, I don’t know. Will its decline be a “crash” or more of a sputtering out? That, I can’t predict either. It does seem obvious that this particular incarnation of Silicon Valley, with this particular leadership class, is running out of gas. That’s why we see all the damn micromanagement, down to the open-plan offices and “Agile” methodologies.

We know why Silicon Valley is toxic: founder quality that has reached levels that can only be described as excremental. The two-class Valley wasn’t great, but the three-class one, with founders and executives competing against employees, leaving a raw deal for almost all of those doing the actual work, is far worse. Founder quality is a simple problem that is easy to spot, and indisputably present. Unfortunately, to solve it may be difficult, and reasonable solutions may be complex. That is where we’ll go next.

Y Combinator and Paul Graham are bad for the world (Part 1)

“Give a man a gun and he can rob a bank. Give a man a bank and he can rob the world.”

In early November, 1988, one of the first major computer viruses, “the Morris Worm“, hit the Internet. While the damage caused by the worm was probably unintentional, it’s estimated to have been somewhere around a few hundred thousand dollars, and possibly a few million. To be fair, Morris paid his debt to society, and that’s all I consider worth saying about that. I don’t know the first thing about Robert Morris or his moral character, aside from the questionable company he keeps (he’s a partner in Y Combinator).

Paul Graham was a protege of Robert Morris at the time, and while I won’t comment either way on the speculation of whether Graham was involved in the debacle– honestly, I have no idea, I was five at the time– I will point out that the two men were among the first partners in Y Combinator. So they went from black-hat programmers (“crackers”) to black-hat businessmen. What’s that second part, again, about a man with a bank?

First, a confession

I must confess that there is a time in my life when I admired Paul Graham’s writing, and allowed myself to be influenced by his worldview. I didn’t always hold the opinions that I am sharing now.

In 2006, I was in a PhD program (pure math) and I saw the writing on the wall. I passed my quals in my first year, but the academic future didn’t appeal to me. When I saw some of the smartest people I knew in “commuter marriages”, suffering weekly cross-country air travel (on a professor’s budget) because the terrible job market made it nearly impossible for both to land jobs in the same city, I realized that it was time to get out. I encountered this essay, “How To Make Wealth”. To my 22-year-old idiot brain, it resonated. Blame the last vestiges of adolescence, but Paul Graham’s brand of blithering optimism made sense to me under certain phases of the moon. It inspired me to think that, maybe, this startup game might be worth playing.

Let’s pull out the opening paragraph:

If you wanted to get rich, how would you do it? I think your best bet would be to start or join a startup. That’s been a reliable way to get rich for hundreds of years. The word “startup” dates from the 1960s, but what happens in one is very similar to the venture-backed trading voyages of the Middle Ages.

Off the bat, that’s fucking terrible advice. Let’s be realistic about wealth. It’s usually attained slowly, not in some get-rich-quick gambit that has a better-than-even chance of costing you money and disrupting your career. The lesson of the 19th-century gold rushes, that much of the money was made by people selling tools and water, is well-understood. So why didn’t more of the prospectors go into the tool-selling business? Because that was the boring, slow way to get rich. Discovering a gold vein was the fast, flashy way that had more visceral appeal, although it was far more unlikely to work out.

Technology startups have, compared to other get-rich-quick schemes, far better odds of producing monstrous wealth for founders in a short amount of time. If you have the contacts necessary to become a VC-backed founder, then startup away. Graham takes it further and says “start or join a startup”. Employees of startups get a lot of their downsides, in terms of job insecurity and division-of-labor issues, but their payoffs are, almost always, mediocre. The people who get rich off the Googles are, in general, not the early entrants, but the people (early entrant or not) who manage to become executives in the middle phases. Of course, being an early entrant can sometimes give an inside track to an executive position as the company grows, but not as much of one as a decade in finance or management consulting, or an elite MBA. If you want to get rich in San Francisco and don’t have the contacts to become a founder, find a way to jump from one executive position to another in funded companies. You won’t get as rich as the founders who beat the odds and end up in the “three comma club”, but your chances of attaining reasonable wealth are better, as you’ll still get an order of magnitude or two more than the software engineers joining at the same time.

In other words, the get-rich possibility for non-founding employees is no different in the startup game than in the mainstream corporate one. If you climb the ladder and reach a highly-compensated executive position, you can get wealthy. Usually, it takes time and luck. At the end of the wringer, some people end up not making it, meaning that their time was wasted on a promise that was never delivered.

Graham continues:

Startups usually involve technology, so much so that the phrase “high-tech startup” is almost redundant. A startup is a small company that takes on a hard technical problem.

That is also demonstrably false. Is Uber a tech company or a transportation company that uses technology? What about YC-backed Airbnb or Reddit? You don’t need to be at the forefront of technology to found one of these companies. You need to have a friend who knows how to build a website. These companies are marketing experiments using technology. And you know what? That’s fine. I’m not knocking these startups for that. However, the idea that nerds are going to hold all the power because these companies really need “10x” lambda-wielding, machine-learning-literate geniuses is one that doesn’t hold water. The claptrap about technical excellence is marketing, not a reflection of these companies’ actual needs or internal definitions of merit. The 1990s are over, and those who insist on technical excellence are again tribeless in Silicon Valley, hence my turn toward Talent Nationalism.

Where Graham gets dishonest is when he makes this promise:

Economically, you can think of a startup as a way to compress your whole working life into a few years.

Graham acknowledges that startups can fail, but he has argued in many of his essays that a startup’s failure won’t consume that much of one’s time anyway, so it shouldn’t be treated as a concern. In Paul Graham’s world, startups either fail quickly (and your investors will let you walk away and immediately start another one; in fact, they’ll fund you!) or they make you rich inside of 3-5 years. So, even though failure is the median outcome of a single startup, it shouldn’t take more than 10 years to get flamingly rich, right?

Let’s say that VC-backed startups have a 80 percent failure rate (this is about accurate) and in the other 20 percent, make the founders wealthy enough never to have to work again (false assumption). Let’s say that startups that fail do so within no more than 18 months (false assumption) and that founders are allowed to walk away (false assumption) and have someone else do the dismal shutdown work, and that investors will, rather than holding past failure against them, enthusiastically fund their next ventures (false assumption). Then startups might make sense as a career, because a person who spends 20 years in that game has well over a 90% chance of becoming independently wealthy. None of that, however is true.

In 2006, I fully bought what Paul Graham was selling. In 2007, I attended the infamous Startup School presentation where Mark Zuckerberg said that “Young people are just smarter”. I was there, and even though I was 23, my sense about that speech was, “You’re full of shit, dude”. I better liked Mitch Kapor’s talk, which advised against becoming an ageist, racist, or sexist “mirrortocracy” and calling it a meritocracy. At any rate, in 2008, I left a lucrative career in finance to try out the tech startup world. It wasn’t the best decision that I ever made. It’s 2015, and let’s just say that I’m behind where I should be, for someone at my age and talent level, in income and net worth.

I don’t blame Paul Graham for selling a lifestyle that worked for him in a time (late 1990s) of radical wealth creation and when the owning class was less attentive to details (read: not capable of capturing 100% of the value generated) than it is now. I certainly don’t blame him for my decision to try out a startup (more than one, actually). I blame myself for all of that. And I’m not afraid to talk about the mistakes that I’ve made, because I’d like to prevent other people from making the same ones.

Paul Graham doesn’t much like me. On Hacker News, others’ posts get deleted and accounts get banned for the crime of linking to my blog. Why? In 2013, Paul Graham started getting warnings from real venture capitalists (the people to whom Y Combinator must sell its companies) that they’d have trouble placing if something wasn’t done about certain “troublesome” posters on Hacker News. This blog post, in July 2012, may have led Graham to consider me one of the “troublesome” ones. Increasingly irritating adverse measures were taken against my account: slow performance of the site (“slowban”) and dishonest representation of a post’s popularity with other users (“rankban”). This culminated in my being banned from Hacker News last summer, Paul Buchheit talking shit about me on Quora, and (unless Quora is deliberately misdirecting blame toward YC) someone from Y Combinator calling Quora and demanding a ban on my account. Quora, which took YC investment in 2014 due to continuing difficulty with monetization, caved to this demand.

Where did this enmity start? What is about my name that can send Paul Graham into an apoplectic rage like no one else’s? Some have argued that it’s some kind of envious resentment that exists because I’m younger and smarter, and arguably more articulate. I doubt this explanation. He’s a centa-millionaire or a billionaire, and I’m not. It’s inconceivable to me that he envies me. He worships material wealth, which he has and I don’t. He envies me about as much as a 1913 Archduke Ferdinand would envy (then impoverished) Nikola Tesla. He is, however, afraid of me. There’s a huge difference. What does he fear? I’m exactly the sort of person that he says prospers in this VC-backed meritocracy, and I actually played the startup game, and got a 50th-percentile career outcome, which is demonstrably… not that great. There’s nothing to be ashamed of, but not much to be proud of, and I’m several years behind where I’d be if I had stuck with a more conventional career. I’m articulate, which Graham recognizes, because his writing skill is a big part of what got him noticed and generated the reputation that he was able to monetize in the form of Y Combinator, and so he understands that my ability to write well and tell a story is, to the illusion he’s trying to create, a credible threat.

Paul Graham can’t have a 32-year-old, visibly smarter than almost everyone in the startup elite, saying “I’ve worked in a few startups, and I’m not rich, and chances are, you’ll end up like me.” His racket demands that people like me slink away in shame, which I refuse to do. Why are VC-backed startups so ageist? It’s not that they believe older programmers are stupid, but that they want to push, out of view, all of the people who’ve been in that racket and had average outcomes. Those kids throwing down 90-hour weeks (as if they were owners, not workers) for startups of which they own 0.02% might be “poisoned” if they’re exposed to people who’ve played that game (and lost) a time or two. If you sift through the older crowd and throw out the 95-98 percent of who’ve had typical outcomes, then you can convince the young that outlier success is inevitable, and get them to work far harder than they should, for minimal reward.

There was a time in my life when I bought most of what Paul Graham was selling. He was quite a competent programmer, back in his day, and his message of hope in a rather dismal time (mid-late 2000s) for startups, the country, and the young, was an appealing one. Paul Graham justly earned a certain reputation by speaking up for technology startups in a time when the mainstream consensus was that such companies were a one-time, dreamy thing that happened in the late 1990s and weren’t coming back. He monetized that reputation brilliantly, by founding Y Combinator.

Unfortunately, while Y Combinator has been fantastically good for Paul Graham’s bank account, it’s been bad for the world.

The three-class society

The VC-funded world was never a utopia. People complained about investors, the difficulty of raising funding, and shenanigans like multiple liquidation preferences (never mind that founders often agreed to them, knowing what they were doing, to inflate their valuations and attract employees, then as now). Before the Y Combinator era, it was a two-class system. Investors were on one side, with founders and their employees on the other, and the social chasm between the two classes was severe: venture capitalists earned $500,000 salaries in addition to upside on the OPM (other people’s money) they managed, while founders were lucky if their VCs allowed them to take out food money. It was typical (and may still be) that investors’ legal fees, in drafting the term sheet, were paid by the company accepting funding. This sum of money, tiny by investor standards but substantive for a shoestring startup, was extracted not because investors needed it, but to send a message: “Shine my boots, prole.”

Being a founder, for a long time, was a rather undesirable gig, excluding the exceptional cases. You didn’t get an “arranged outcome” (e.g. an acqui-hire) if you failed; you went out of business. Rather than being allowed (and expected) to hire immediately for the grunt work of starting a business, founders worked 90-hour weeks and spent much of that time actually creating the thing. (These days, they work 90-hour weeks raising funding and building up their public reputations, which seems like not much of a change, but they get a lot more for the effort.) It wasn’t an enviable life, but there was, at least, a sense of shared suffering between founders and their employees. The investors continued to make high-6- and 7-figure salaries while playing games with OPM, while founders and employees did all the work. This sense of “We’re all in this together” was reflected in cap tables, because employees who joined early were compensated more like founders than employees are now. Investors still took the lion’s share, and founders obviously got more than later employees, but it was still possible for average employees to make substantial money from their options. The order-of-magnitude (or two) drop between the last founder and the first employee wasn’t there yet. Moreover, early employees often had transparency into the capitalization table, allowing them to assess for themselves whether they were fairly compensated. That consideration is pretty much nonexistent in 2015.

What changed? In the mid-2000s, serial founders and venture capitalists’ friends (whose MBA-school grades weren’t quite good enough to make the investor-side cut, but who could hop from VC-funded executive position to another) discovered that there was job security to be won by sticking together. The two-class society, with founders and employees on the same side, became a three-class society, as founders (plus serial startup executives) and employees separated into two separate clusters. A few social changes happened at the same time. For one, investors and startups began to separate people into “founder-grade” and “employee-grade” categories. Founder-grade employees got executive positions (which didn’t even exist in the old-style startups, until after they were no longer startups) and would be introduced to investors, so they could start their own companies, after one to two years of loyal service. Employee-grade workers were given 0.05% equity slices and would typically see people hired above them as the company grew, but were expected to work hard as if they were founder-grade. Fundamentally, the resource-extraction culture that Silicon Valley has turned itself into, is one built on young people who can be treated as employee-grade, but who will overexert themselves to push out a founder-grade effort.

At some point, probably in the late 2000s if not earlier, software engineering also transitioned from being a founder-grade profession to an employee-grade one. If it isn’t obvious, the reason why VC-backed startups love open-plan offices and “Agile”/Scrum is because those devices exist to turn massive teams of “employee-grade”, young and mediocre, programmers into something halfway productive and controllable. Much of this was and is done because acquisitions are so often priced according to headcount, making it sometimes profitable to hire large teams of unqualified engineers.

In my belief, the three-class startup ecosystem is worse than the two-class system. In the former system, the people doing the work were separated from the investors, and a certain share of the power comes just from being on the spot. Investors could take their share of the surplus wealth, but there was enough power and wealth on the other side that you didn’t have to be a founder by 30 in order to get wealthy in the startup game. In the three-class system, founders play against employees. You get formalized performance reviews, aggressive micromanagement in the name of “process” or “Agile”, and commoditization of the important work. This doesn’t make the VC-funded startups worse, in these aspects, than large corporations; it just makes them not better. That said, big corporations offer many things that VC-backed startups don’t: stability, diversity of available work, career planning, and established HR practices that will prevent, say, the extremes of harassment for which the “brogrammer” startup scene is known.

Y Combinator, I believe, shares some moral responsibility for creating the three-class society. It was never designed to be “just another” incubator, but to formalize the existence of the founder/employee separation. Partners and founders in YC-backed companies who joined after YC acceptance are called “half-bloods”; they weren’t selected by YC itself as founder-grade, and the term carries an implicit judgment that they don’t really belong.

Paul Graham and Y Combinator have certainly done a lot to help founders negotiate against investors, and that’s admirable, because even though it’s not a fair fight, it was probably less of one, ten or twenty years ago. They also seem, however, to have shared a lot of information on how to win against employees. Most of the people who got rich in the pre-YC Silicon Valley did so because their employers “made mistakes” (which weren’t mistakes, because their high compensation meant that these employees were legitimately engaged with their work) and offered equity and responsibility based on what seemed fair to all parties, and not based on what their backers told them they could get away with. Let’s say that in 1995, you have two programmers, a person with contacts to investors, and an older person with the organizational skill to evolve into the CEO role. They might agree to an even (25% each) split and dilute equally when the investors come in. In 2015, the investor might meet with the proto-CEO and the guy-with-contacts and says, “You can totally give those programmers 2 percent a piece. Tell them that it’ll either be a billion-dollar company, meaning $20 million, or failed, in which case equity slices don’t matter, inside of 12 months.” One view of this is that the market is becoming more efficient and fewer employees are getting lucky by being “overcompensated” in equity. I doubt, however, that the new arrangement is more efficient, because the shitty cap tables also drive down talent levels. The darker view is that the investors and founders have locked-in the value capturing process and there’s just nothing left for regular employees, who don’t have the contacts to become founders and are unlikely ever to get them.

When it was investors against everyone else, in a two-class society, founders and employees could share some of the bounty that would naturally accrue to them on account of the leverage they’d have, just by being on the spot. There’s power that accrues, in a business, to the people who run the thing. The three-class society, on the other hand, has founders and executives acting as company cops, with investors and founders working together to make sure that employees get as little, while being expected to give as much, as possible.

We don’t need more “cool kids”. We need substance.

Y Combinator has created a “cool kids’ table” in Silicon Valley. While real venture capitalists don’t much care for Paul Graham himself (for reasons I won’t get into here, since I’d risk compromising sources) and his waves in the kiddie pool, many VCs will use his signal, simply because it’s expedient. There are firms now that won’t talk to a non-YC company. They’ve outsourced their talent vetting to a man who routinely makes foot-in-mouth statements about foreign accents or about those uncanny people with vaginas who, according to legend, make up half the species. This is unsurprising (VC groupthink is a notorious plague) but sad. Y Combinator now provides a service that is harmful to the world: it, increasingly, means that founders only get one chance. If YC judges you as employee-grade and passes, then the rest of Sand Hill Road will conclude the same.

Frankly, the best thing that the United States can do if it wishes to preserve private-sector innovation is to break up Sand Hill Road. I recognize that, formally speaking, it’s a collection of companies that are theoretically competitors, but their sharing of information means (even regardless of what YC does) that a founder gets only one real chance to make his case, because the VCs decide as a group who they like and who they don’t, like middle-school girls passing around “cutest guy” lists. The culture of note-sharing and co-funding isn’t new to finance; it’s a subcase of what would be called market manipulation (and highly illegal) if applied to publicly-traded stocks instead of unregulated private equity. The result of it is that the market no longer reflects aggregate preferences (however they may fluctuate) but the game-theoretic concerns of the most powerful players. This is bad for innovation and, while it’s good for the well-connected “serial entrepreneurs”, it’s bad for people who are trying to make the employee-to-founder leap.

As for YC, I’ve met about 30 YC founders in my life. I will admit that, a few of them, I ended up liking. They’re not all bad. Another 10 or so, I’m neutral on, or just don’t know much about them. And then, there are a substantial nunber who are dreadful. I don’t mean just that I dislike them; I mean that their personality and ethical flaws are so severe that I have no faith in a talent-scouting organization that would admit them.

I wouldn’t interview with, or work for, a YC-backed company today, and I can proudly say that I’ve never worked for a YC company, but I’ve interviewed with a few in the past. One time, I remember listening to a 25-year-old CEO gloating about firing people in their first 30 days because (you hear this justification a lot) “we’re a startup”. I heard another one, at one point, talk about how he’d be able to exploit his employees because, in his model, they’d so desperately want to be introduced to Y Combinator that they’d do anything, just on that promise, which he could rescind “for performance”. (I don’t know how his company turned out.) Obviously, I didn’t work for either of these companies.

It seems clear that YC people conceive of themselves as superior, founder-grade, relative to the employee-grade masses pounding on the doors of startups (not because startups are great, but because the mainstream corporate world is, due to technological unemployment, depopulating). That really irks me. Some organizations are founded for high-minded purposes but become exclusive and snooty. Such a criticism has been made about some of the most selective schools, for example. However, these institutions weren’t designed to be that way. Harvard doesn’t claim that it admits and gets every decent student, because it knows that it doesn’t have room for all of the “leader-grade” people who apply. The top universities are selective because they’re good universities, and because there are physical limitations regarding their ability to admit students. There’s good and bad to “the Ivy League”. These universities do a lot of good for people who attend them, and I don’t think that the snobbery actually hurts many people. (It makes some people massively insecure, which is part of why corporate executives in their 50s are often caught falsifying educational credentials, despite gaining no benefit, by that age, in doing so; but I don’t think that, in the U.S., there’s a serious job-market effect after age 27.) These universities make earnest efforts to have their positive effects on society outweigh the negatives. I cannot say the same of Y Combinator, which seems to be replicating the negative traits of such institutions, without any of the positive ones (such as noblesse oblige or a value on classical erudition and the preservation of culture). Y Combinator, unlike a university founded as a seminary in the 17th century, was actually created to be exclusive and snobbish, but its elitism is entirely about access to money.

Of course, I can’t blame Y Combinator entirely for shitty cap tables or for the emergence of a 3-class startup world. It’s quite possible that those changes would have occurred, no matter what happened. Similarly, I want to talk about ethics, and I’ll admit that I can’t blame YC alone for teaching founders to be unethical.

However, I’ve been around the startup game for long enough to know that, when it comes to dishonest and unethical behavior, founders and executives share tricks. They learn from each other, when it comes to how to best compete against employees and the public, and I suspect that Y Combinator has played a role. Of course, I don’t think that Paul Graham ever gets up and says to his shiny young proteges, “You should be unethical.” However, I think that his lack-of-statement is a statement. Founder quality is at a dismal low, and Paul Graham doesn’t care. He’s rolling in dough. His lack of concern for founder quality and ethics, however, has encouraged a value system to emerge in which such concerns are willfully trampled.

This has to be addressed. The same unethical tricks– disguising layoffs as performance-based firings to avoid negative press, bait-and-switch hiring, misleading prospective employees about rank and position in the company, placing clawbacks on relocation packages– show up in startups, over and over again, and founders are clearly learning them from somewhere. Y Combinator is clearly not the only source of this information, because you see these behaviors in non-YC companies as well, but it stands to reason that an organization that exists to be an exclusive club within the Valley would also be delivering access to knowledge on how to compete against employees.

How does the three-class Valley encourage unethical business practices? The social distance between founders and employees enables the former to behave as if the people in the latter category are peasants who don’t matter. Moreover, the short-term focus of these companies enables them to pull moves that you’d be unlikely to see from long-lived corporations. For example, it’s probably well-known at this point that technology companies disguise business-based layoffs as performance-based firings. Investment banks are willing to come forward and say, “Times are tough for us, and we’ve had to let good people go.” They pay severance, and move on. Startups, on the other hand, need to keep up the mythology of monotonic expansion, and are too stingy to pay severance, so they dress layoffs up as performance issues, preserving their reputations at the expense of departing employees. Why is there a difference in approach? Banks are old and large enough to know that other peoples’ good will isn’t an inexhaustible resource. The person let go might, a decade later, be a decision-maker at an important client. VC-funded startups, on the other hand, are designed to be dead or sold within five years, and therefore don’t harbor any concern for the long-term implications of their founders’ actions.

The performance/control tradeoff

Before I close out Part 1 of this essay, let me point out one final thing. VC-funded startups have, rather unceremoniously, moved themselves to the wrong side of economic history. With commodity investors funding commodity founders to work on commodity ideas and hire commodity engineers, we’ve replaced what an engine of innovation with a value-capture tap, exploiting the reputation for innovation that a certain geographic region earned decades ago.

The “magic” of the 1990s was observed when small startups out-competed large corporations, in a manner that we’re not really seeing today. The 2015 analogue is a VC-funded giant (rather than a small, elite team) overtaking a corporate giant. The disproportionate high performance, the “10x” and “100x” factor, that we saw in the 1990s, has mostly gone away. Why is this? The Valley is, in general, no longer funding startups that are capable of “10x” performance.

I doubt that it is controversial to say that, in work, there is a trade-off between control demanded and performance rendered. What made the earlier waves of startups so much more productive than large companies was their absence of the control and permissions systems that impede performance. These earlier crops of startups were founded by engineers who wanted to avoid the negatives of the corporate world and who were disinclined to micromanage because, first of all, they typically had several years of work experience on the other side of the managerial relationship and, secondly, because they identified as lifelong technologists. For a contrast, the current crop of founders are people who want to be corporate titans: they just want to get there faster than their parents did. What used to be a secessionist colony for corporate misfits has become a glass elevator for connected, entitled Millennials. Consequently, we have a startup world in which toxic micromanagement is the norm, clept in the Orwellian name of “agility” and “openness”.

I’ve seen enough startup failures to have observed the Micromanagement Death Spiral. Long-term micromanagement leads to disengagement and underperformance in those being managed, and this tends to bring on even more micromanagement. Hostility and distrust mount until the micromanaged employee(s) leave or the team (or company) collapses. When you have a constitutional micromanager (“control freak”) in charge, the damage can be immense. Now, there are at least two plausible etiologies of Micromanagement Death Spirals, and I consider both to have some truth in them. One is that they cause organizational decline directly. While the damage is downward-directed and therefore limited if the control freak is a middle manager, the eventual effect of such a person is a “cone of disengagement” within the company. The other is that micromanagers are symptomatic of decline. If a manager is demanding frequent, detailed status reports, then he clearly doesn’t have enough to do. When things are going well and the business is expanding, micromanagement is rare, but when executives are panicked and middle managers need to define and protect turf, people start breaking out the TPS reports. So, while micromanagement often drains the business, it’s also true that weakness in the business causes micromanagement. This is worth pointing out, because a culture of cargo-cult micromanagement in the VC-backed software business just might signify structural weakness in that whole ecosystem.

In the three-class Valley, mounting micromanagement is to be expected. The founder class has to maintain enough control over the employees to protect their positions. At the same time, they need continually to justify their superior status to the investor class that sits above them. Founders are now expected to hire rapidly and reach enormous valuations, regardless of whether the business is ready to handle that scale, and that insistence on unsustainable growth (Y Combinator, for example, demands 5-7 percent per week) creates an environment of frequent crisis, which begets micromanagement that ripples throughout the whole company.

In the 1990s, “startup” meant a small, elite, high-performance company that could take on a large, established megalith. In 2015, it means a newly-created, VC-backed corporate giant taking on an old corporate giant. What we’re learning is that dysfunction isn’t necessarily a function of chronological age, because these new huge companies are often just as dysfunctional, ethically challenged, and inefficient as the old ones, if not moreso. This obliterates Silicon Valley’s original reason for existing: once innovation, now merely shuffling and “disruption”.


So, could it be better? If Y Combinator weren’t made by Paul Graham, would the emergence of something similar, some other toxic, overconnected, and ethically challenged meta-organization, be inevitable? Or is it possible to do better, and how? Moreover, if one had recognized the ethical paucity of Y Combinator’s black-hat founders, could someone have interceded and built something better, sturdier, and less bad for the world?

These are some of the topics I’ll attack in the second half [edit: third?].

Part 2 is here.

Rapid headcount growth kills a company’s culture, but not in the way people think

Even though Silicon Valley fetishizes rapid growth and “unicorns”, it’s well-understood that rapid headcount growth is bad for company culture. The firm gets big, “politics” start happening, and the company becomes a less interesting place to work, or so the complaint goes. I’m not fully bought in to this analysis, because politics is always there; the question is whether it interferes with getting work done or with a person’s life outside of work. Still, I’ve seen it happen first-hand that rapid headcount growth damage a company’s culture. The conventional explanation for this is that new people are brought in and not properly acculturated. Having seen this dynamic play out multiple times, that’s not how it actually works.

The newcomers, for the most part, aren’t to blame. Most new employees are going to come into an organization near the bottom and, therefore, not have much influence over the corporate culture in any case. Even when there are more of “them” than there are old hands, the threat posed by new hires en masse is pretty much nonexistent. Of course, sometimes a company will hire externally for an executive position, and that new executive hurts the culture. That’s not necessarily a growth problem, though. Hiring a rotten executive will hurt the company, regardless of whether its headcount is growing quickly, slowly, or not at all. That’s a different problem.

So if rapid headcount growth hurts the company culture, but the newcomers are not to blame, then what exactly happens? In general, when a company hires over-quickly and loses its culture, it’s the fault of the people who are already there. This style of hiring is typically done to build a bottom. The workers are burning out, and management sees this, and the shared solution is to promise the worn-out workers that they’ll have a bunch of people under them within a year. Concern with sustainable practices, business focus, work/life integration, and technical solvency (as opposed to “technical debt”, whose interest rate is usurious and unpredictable) can go out the window, because “you’re/we’re all going to be VPs and Directors by next Christmas!” I was once in a conversation where a programmer was directed not to automate certain reliability-oriented tasks because, as his manager argued, “We can always hire a state-school grad to carry a pager.” I don’t know if I’m more offended by the snobbery or amused by the incorrectness of it. In what world do competent people allow themselves to be poorly managed solely because they didn’t go to “the right schools”. Only an idiot would presume that another person’s reasoning might be, “This place is a shitshow, but I’m an Ohio State graduate so this is my lot in life.”

Hiring one’s way out of problems doesn’t work for very long, if ever. Recruiting is hard, and recruiting good people is extremely difficult. I grapple with abstruse programming languages, advanced linear algebra, and complicated production systems on a daily basis. Recruiting, especially under limited resources, is harder. When business circumstances and the press are favorable, there’s a palpable “talent pressure” that leads a company to believe in an inexhaustible pool of potential workers. Inexhaustible, it ain’t. For all the negative things that are said about large corporations and investment banks, such companies often treat their best people quite well, knowing just how difficult and expensive it is to find talent. Only hubristic, inexperienced companies can convince themselves that a limitless pool exists.

VC-funded startups often lose their cultures not by intention but through negligence as they grow. The cultural decline is blamed on regression to the mean as the company is forced to compromise and hire people “not like us”. That story, while favorable to the originators, is completely false. Culture is de-prioritized by the founders and investors, who demand rapid and unsustainable growth or they will lose interest. For example, Y Combinator– an institution that is bad for the world in more ways than I have time or space to get into– demands 5 percent growth per week, and writes off a company that doesn’t achieve this target. A company that started with 5 people and grew at this rate would have 63 people after one year, 800 after two years, and over 10,000 after three years. That’s an absurd demand! To throw an idea out, simply because it is not on track to build a 10,000-person company inside of 36 months, is ludicrous. Many important problems don’t require 10,000-person companies (or even 100-person companies) to solve them. As founders struggle to meet this unreasonable growth demands, cultural integrity can’t be afforded. Worse yet, the unreasonable expectations of the investor community create an adverse selection in the moral quality of founders; or, to put it more bluntly, the system rewards those most willing to make dishonest promises.

I’ve written about technology’s being a resource-extraction culture, surprisingly similar to what you’d expect in an oil-rich Middle Eastern country. When a company’s getting good press and raising money easily, it’s in a similar “bonanza” mode. No thought is given to the future, under a shared belief that the company will either have failed or made everyone rich (with some variation in how “rich” is defined) within a few years. And the perception that there will always be more eager grunts allows a company to create a culture of professional hazing and overall disregard for those at the bottom. This is a common pattern and, yes, it kills the culture. When managers no longer see mounting grunt work as a problem that requires automation or complexity-cutting, because they can just triple headcount and give “team lead” positions to those who’ve suffered the most, it’s a really bad transition. But don’t blame the newcomers for “killing the culture”. When this decline occurs, the new people usually have nothing to do with it.

Harem queens in the corporate world

I recently came across a New York Times article about shitty behaviors that our society often associates with the female gender. It’s about the stupid, catty, bitchy behavior that makes for an entertaining evening in a film like Bridesmaids, but that actually wastes an enormous amount of emotional energy and gives the female gender a bad name. The choice paragraph, in my opinion, is this one (emphasis mine):

There are two main theories of why women are competitive in indirectly aggressive ways. Evolutionary psychology, which uses natural selection to explain our modern behaviors, says that women need to protect themselves (read: their wombs) from physical harm, so indirect aggression keeps us safe while lowering the stock of other women. Feminist psychology chalks up this indirect aggression to internalizing the patriarchy. As Noam Shpancer writes in Psychology Today, “As women come to consider being prized by men their ultimate source of strength, worth, achievement and identity, they are compelled to battle other women for the prize.” In short: When our value is tied to the people who can impregnate us, we turn on each other.

“Indirect aggression” is the author’s more-polite term for passive aggression, and here’s what I take out of the article. I’ll withhold personal comment, so please don’t get offended just yet. From the article:

  • women, in a system where their value is attractiveness as assessed by men, work to undermine each other.
  • women generally prefer passive aggression over overt conflict.
  • women exhibit an especial and envious hatred toward highly attractive women, that is divorced from the reality of that person.

Is this true of “women”? Well, it’s true of some women, and not of others.

Now, I want to talk about these horrible patterns of behavior: passive-aggression, cattiness, sabotage, bitchiness. I’ve seen a lot of this stuff in my decade in the software career. And, here’s the thing… most of the worst offenders have been men.

In my personal opinion, malicious social aggressions are not especially gendered. Women may be two to three years ahead in this game during high school and college, because they’re farther along (in general) when it comes to social development, but corporate executives’ cattiness makes high-school cheerleader drama look innocent in comparison.

That humans perceive certain styles of aggression as being gendered is, however, telling. Stereotypically “male” aggression’s virtue is that it is short-lived and honest; its vice is that it’s brutal and often disproportionate (and if it leads to violence, its effects can be permanent). For a contrast, the “female” aggression has the virtue of precision, but the vice of being dishonest, long-lived, and cancerous. In reality, we observe all classes of aggression in all genders of people, so why do we have this need to create a gendered correlation? I think the answer is, and I’m going to have to define this word (lest it be useless) patriarchy.

What is patriarchy? I’m going to use it to mean a society in which women are largely blocked from holding official power, and in which male dominance (often backed by threat of physical harm, whether delivered personally or through a military) is the prevailing force in determining which men get power. A large number of men, for the record, lose in a truly patriarchal society, because the high-status men take the lion’s share of the rewards and sexual attention and the rest get very little. The extreme of patriarchy is a polygynous society in which high-status men are afforded large numbers of “wives”, most of whom experience the status of chattel slaves. Patriarchy is morally reprehensible and it is good that our society is beginning to reject it, but it’s important to acknowledge its probable normalcy in our evolutionary frame. Men are physically stronger than women, and humans are not naturally monogamous, and there is a general observed tendency in humans to desire to “possess” others, at least in terms of their sexuality. That this would lead to a world, speaking of most of human history, in which powerful men owning harems isn’t surprising. In that lens, it seems inevitable.

What is this supposedly “female” pattern of “indirect” social aggression all about? Well, it’s about polygynous patriarchy. In a monogamous society, children usually inherit the social status of the parents. If it’s a monogamous patriarchy, the status of the father will be weighted more heavily. In a polygynous patriarchy, the warlord or chieftain with 20 wives might have 150 children, so having a high-status father isn’t enough to confer social status to the children, because there are just too many alpha-brats to give that status and wealth to everyone. Often, the children with the best chance of inheriting the father’s wealth, status and connections are those born to the most favored wives. If a woman achieves that status, her children will be heirs, and the other womens’ children will be labelled bastards. Obviously, a woman who is in a harem has an evolutionary incentive to want to be that favored wife.

Passive aggression works well for a woman in a harem who is campaigning to become queen of the harem. Let’s consider how this contest works:

  • fitness or “performance” or “merit” will be judged by an owner who is largely distant from the day-to-day operations (e.g. division of labor) within the harem.
  • overt violence against a potentially superior competitor will be judged as destruction of the owner’s property, and punished. It will also fail to make the target appear less fit, from an evolutionary perspective.
  • however, emotional cruelty, abuses of power, and other “indirect” aggressions will cause a targeted competitor to become discouraged, stressed-out, or depressed, and therefore appear less healthy (that is, less fit) to the owner-husband.

What is an aspiring harem queen going to do? She can’t gain her status through physical dominance, as men in such societies often did, because provable harm to other women will cause her to lose favor with the man who ultimately selects the winner. Anyway, physical injury doesn’t make a woman look genetically unhealthy, and is therefore less likely to sway the opinion of the owner/husband (if we assume that he’s also acting under evolutionary imperative); while the long-term effect of emotional violence (and stress) will make the target appear sickly and suggest genetic inferiority.

In the modern world, harems mostly don’t exist. There is some harem-queen behavior in American high schools and colleges, related to the severe desirability differential between the men and the women at that age, and the intense competition (for women trying to attract the few unusually mature men) that this creates. 18-year-old women are physically attractive to men of all ages; 18-year-old men are attractive to… no one they can legally date. So I suspect there is some incentive for harem queen behavior, given how few desirable men there are in those ghettos for adolescents that we call schools. Of course, the men become more desirable and mature as they grow up, and the women seem to grow out of any harem-type behavior, and given that the teenage years are a period in which everyone’s pretty fucked up, I’m going to guess that the overall behavior of women isn’t any worse than that of men. The movies give us adolescent “mean girls” for entertainment’s sake, but I don’t think that such meanness is gendered.

Let’s jump away from adolescent drama (it’s not that interesting) and focus on the corporate world. The parallels are shocking.

  • corporate competitors strive for a “favored subordinate” status, because the odds, for a person not born into it, of joining the ownership class are only slightly better than those of a woman in the neolithic era becoming a man.
  • fitness or “performance” is judged by parties who are distant from the action and oblivious to day-to-day operations and any emotional or political drama that might be going on. They are about as capable of assessing individual merit as neolithic harem-owning men were at assessing genetic health among their wives.
  • like attractive women in pre-monogamous societies’ harems, people who threaten to out-perform the others draw undesirable attention and adversity within the harem. They’re sometimes singled out and set up to fail.
  • obvious interference with a potentially superior competitor’s performance (much less physical harm against that person) will push the perpetrator out of favor, as “The Company” will view that as damage to its own property.
  • non-obvious (from the owner’s perspective) interference or aggression toward a potentially superior competitor can degrade that person’s apparent performance and is therefore the weapon of choice.

Corporate executives can be understood, then, as successful harem queens. Under this lens, the entire corporate world makes sense. The CEO isn’t the alpha male, but the champion bitch.

So who’s the alpha male? Well, it’s arguably “The Company” or “shareholders” or rich people or “capital”. We’ve personified Mammon, and then gone a step further. We’ve created an alpha male more powerful than any actual human can be. As I’ve said before, that’s not a corporate “person” but a god. Worse yet, this alpha male isn’t one corporate “person”/god but the whole professional system, a supreme and supremely intangible meta-god among the myriad corrupt gods that our socioeconomic system has made. One could imagine, in a corrupt and slowly declining religious order, harem-queen behavior among the priesthood, and that’s basically what we have within the secular priesthood of corporatism. This kind of shitty behavior, clearly, was never limited to women. And now we have a society that runs on it.

Tyler Durden, in Fight Club, lamented that “We’re a generation of men raised by women”. Now, Fight Club was on to something. The male gender in the U.S. is failing and has been failing for some time. (It’s even worse in societies that are more thoroughly corporatized. Look up Japan’s salaryman culture and then read up on its hikikomori and “herbivore men”.) There is a toxic something in many advanced societies that has been emasculating men over the past forty years. But it is not that we were “raised by women” (I’ve seen neither evidence of declining paternal involvement, nor that increasing maternal influence is detrimental) and it’s not the fault of feminism. It’s the fault of corporatism. We’ve created a society that requires us to believe that our richest people are “alpha males”, when they are actually champion harem queens– preening private-sector social climbers. This has spawned a generation of men who are far too concerned with others’ opinions of them (and of the schools they went to, and of the jobs they held) and mostly unable to succeed on their own or think for themselves.

One might make the counterargument that the macho-subordinate “bro” culture that is observed in some corporate environments (such as most Silicon Valley startups) is, to the contrary of my argument, excessively masculine. Actually, it’s just stupidly masculine. There’s a lot of posturing going on in it, and women are justified to find that posturing threatening, but it’s just another form of the harem-queen game. The difference, in the young macho-subordinate culture, is whom these wannabe harem-queen men are whoring themselves out to. In the adult corporate world, their target is the amorphous nondescript alpha-male-figure (i.e. the invisible ultra-powerful humanoid that evolves into a god) known as “capital”. A seasoned corporate executive knows that his job isn’t to slut it up for a specific manager or CEO but to make himself maximally attractive to the system itself, and that system’s god persona is too aloof for it to be assigned a definite gender, so it does not suffice to appease one presumed aspect of it. Consequently, it demands an appearance that is bland, inoffensive, and neutrally toned. On the other hand, the macho-subordinate brogrammers haven’t figured out yet that they should focus on their attractiveness to a larger system rather than their immediate bosses. They’re still powdering themselves for the visual delight of, say, a 30-year-old and invariably male founder. This drives them not toward the manicured, soft-faced look of the corporate executive, but toward the absurdity of the frat boy who exhibits the worst traits of masculinity in caricature, but whose membership in such a pointless, groupthink-driven organization shows his desperate need for approval. Therefore, the macho-subordinate brogrammer culture of Silicon Valley does not refute my claim that corporate competition is a harem-queen dynamic; rather, it supports it.

Right now, the world is run mostly by cowardly, talentless, champion harem queens angling for approval from that invisible god we call “capital”. That’s unfortunate, and my view is that it has to end. In order to get this far, human society had to overcome many challenges, and it seems that the 21st century is destined to bring us more challenges (e.g. climate change, religious radicalism, economic inequality and petroleum depletion) and I don’t think that we can handle a single one of them with the business leadership that we’ve got. So who should be running the world? To be honest, there are some men whom I consider supremely qualified, and there are some women whom I consider supremely qualified. Gender, to me, is irrelevant. However, we need independent thinkers, and this means that the harem-queen world (for which, many of the worst offenders are men) has got to go.

Why I’m not quitting technology

Every now and then, I encounter a blog post about “Leaving Tech”. Sometimes, it’s a whiny, self-indulgent rant by someone who didn’t expect startups to be so hard. Other times, the person has solid reasons for leaving the software industry. Now, I don’t begrudge people for making the decision to leave technology. A lot of great people are leaving this industry, especially when they get to my age (early 30s, which is middle-aged in Silicon Valley years). That’s a personal decision, and I understand why people make it, and I don’t fault them. This is an industry run by the wrong people and, consequently, it has a toxic, horrible culture full of junk that has nothing to do with computer science or programming. It’s ageist, sexist, racist, classist, and macho-subordinate. It allows horrible ideas like employee stack-ranking, micromanagement in the Orwellian name of “agility“, and open-plan offices designed to test people by seeing who gets the least sick. In software especially, we’ve allowed ourselves to be colonized by people who’ve failed out of the mainstream business culture. People leave this industry every day, often for good reasons, but I’m not going to do so. I’m going to stand my fucking ground.

It’s not really to my economic advantage to stay. People far dumber than I am make a lot more money in private equity, and people at or around my level of intelligence make seven figures, easily, after a decade in the hedge fund world. So why do I spend time and emotional energy on computer science and technology? This industry never loved me much– Google was pretty horrible to me, as is well-documented by this point– so why do I care about it?

I could get on a moral high horse about the importance of technology to economic growth, the noble cause of defending it, and all that, but the reason’s much simpler. I’m not leaving because, well, this is my fucking territory. End of story. No, it’s obviously not mine personally but it belongs to my tribe (Talent) and not the colonizing one. Technology happens to be one of society’s major profit centers, but even if it were irrelevant, I’d still be inclined to defend it. I’m fucking smart and I’ve worked fucking hard to understand this stuff, and this means that I belong here, and these colonizing fucksticks don’t. I’m a Talent Nationalist, and I would have no credibility as such if were to I give up so easily on the tribal birthrights of Talent, which include control of the technology industry. Sure, I could join the other side, tow their stupid line, become a smiling idiot just like them, and make a shit-ton more money, but there’s one problem with that… namely, the fact that would be disgusting.

As one result of technology’s colonized status, founder quality is at an all-time low. Every time I hear an investor say, “We don’t invest in ideas, we invest in people” I cringe, and here’s why. I’d respect those guys much more if they said, “We invest in ideas, and sometimes we fund a horrible founding team if we like the idea”, because that would be honest. Some of the ideas they fund are decent ones, like Uber. The people they fund, 95 percent of the time, are just atrocious. If their job is to invest in people, then I would be better at their jobs than they are by far, because unlike them, I wouldn’t fund so many people who are talentless and unethical.

I’m going to fight because, morally speaking, I have no other option, as a fervent Talent Nationalist. I doubt that I am Talent’s last line of defense, but I am certainly zealous in my defense of it and its rightful territory. If you come in to a scientific or technological company and impose employee stack-ranking, then you are invading my homeland with your shitty corporate culture, and I will defend my homeland past your demise and mine, because it’s my fucking job.

It may be fashionable for self-involved Silicon Valley hotshots to write blog posts about how they’re quitting tech because they’re tired of working hard. Fuck that. For me, it was never about that. The people I care about aren’t the ones writing “startups are hard” blog posts. I care about the highly competent men and women (and a disproportionate share are women) who are pushed out of this industry by its inept management, stupid practices, terrible companies and founders, and its exclusionary and macho-subordinate culture. We can’t have that. We can’t be driving out the true Talent while letting ourselves be overrun by bros and corporate yes-men. I may be a masochist, and I may be tied to an idea– that this industry can still be saved– that is absolutely insane. Even if it’s a losing war, I am going to stand my ground and fight. Someone must.

Really, what else can one expect me to do?

The “I.T.” stigma as the source of VC-funded technology’s rotten culture.

People outside of “Silicon Valley”, meaning VC-funded technology startups, have an often hilarious and rosy view of it. To hear them tell it, a tech job means free massages whenever you want, access to futuristic computing technology, interesting problems to solve, and enough money to retire by age 40. That is not what the technology industry really looks like, even in the hot-shit “unicorns” and the West Coast megaliths. The pay is sub-adequate given the level of talent involved and the extreme job insecurity, early retirement is usually involuntary, and boring work tends to pile up not because it can’t be automated, but due to mismanagement and political intrigue. Of course, it’s not very useful for me to go on about how much the VC-funded software industry sucks. People who’ve been in it, for more than two years or so, will agree; and those who haven’t reached that conclusion, at some point, will.

Why does the software industry have such a terrible culture? It’s easy enough, and certainly comforting, to blame “the VCs”. It’s always appealing, when a class of people is that highly paid, seems not to share one’s values, and appears to be in one’s way, to argue that one could do better without them. Could we? Not unless we understand what ruined this particular crop of venture capitalists– if, even, they’re the ones who deserve the blame.

I’ve written about the Damaso Effect. It is, I would argue, the most important sociological insight about Silicon Valley that has been made in the 21st century. According to the Damaso Effect, the technology industry’s woes exist because technologists (programmers, engineers and scientists) are a colonized tribe, and because the colonizing culture sends its worst people over: those who failed in the mainland, for some reason. I’ve argued that technology has been colonized by “the mainstream business culture”, but what is that culture? Mostly, for good and bad, it’s finance culture: Wall Street.

There’s good and bad to be said about Wall Street. For one thing, the ethics on Wall Street are actually way better than in Silicon Valley. Wall Street ethics, by the standards of the corporate world in general, are average at least. This surprises people, but the ethical issues surrounding the banks usually come from their sheer power and importance. If the bankers do something unethical, it ends up in the New York Times and there’s often jail time involved. If VCs or founders do something unethical, it silently ruins a few careers. VC-funded technology just isn’t important enough to attract the kind of scrutiny that is put on the banks. Trust me, though, when I say that in every ethical measure– diversity, meritocracy, fair treatment of third parties– the VC-funded startups are far worse. It’s not even close. It’s not just about the quality of people, but also the incentive structures. Bankers actually care enough about their careers and reputations not to do anything stupid, while a techie will trash a person’s career and life over the smallest grudge. VC-backed founders don’t care about doing the right thing, because they expect their companies to be dead, or sold for billions, inside of 5 years. Bankers know that a reputation for honesty is an important asset in a long career. Techies lie about their companies, their products, and the job positions into which they’re hiring prospective talent, all the time. To their credit, most of the technologists themselves— the programmers, engineers and scientists– are typically honest to a fault. Their aversion to dishonesty is a consequence of this group’s general desire to avoid “being political”, even when it’s necessary. The problem is that, in the colonial dynamic that characterizes the VC-funded world, this set of people has no power.

Understanding finance culture

I’ve worked in finance as well as technology. So the first thing that I’m going to say is that “Wall Street” is more diverse than Silicon Valley. The VC-funded world has the same juvenile, fratty monoculture whereever you go. It doesn’t matter if you’re in San Francisco, Brooklyn, or Miami. The important decisions are made by the same small set of people. Finance, on the other hand, is very diverse. There are hedge funds where the average age is over 50, and others where it’s under 30. Unlike the Valley, where a woman pretty much can’t be made a founder unless it’s a fashion startup, there are women-led trading desks that deal in extremely technical products. So, I can’t talk about “finance culture” with authority as if it were monolithic, because it’s not. It’s much bigger and much more diverse, and all I can speak about are general patterns.

Techies generally avoid finance, based on a perception that it’s money-driven (which it is, but so is tech) and unethical (see above) and that programmers have low status in financial firms (which is quite variable). Focusing on the last of these points, it’s true that, traditionally speaking, programmers have often been low in banking’s status hierarchy. As it’s often depicted, traders outrank quants, and quants outrank quant developers, and quant developers outrank “I.T.”, meaning regular in-house developers. In some firms, this is true and, in those, programmers often scramble to reinvent themselves as quants as fast as they can. In others, it’s not as true. Most of the best-performing financial firms are those that go out of their way to avoid having that kind of status hierarchy, especially as technical skills become more important with each passing decade. Not all successful financial firms are progressive, nor vice versa, but there seems to be a positive advantage in having a technology-driven culture. If you want the best programmers, you can’t put everyone who writes code at the bottom of a hierarchy.

This essay doesn’t require a focus on the best of finance culture, because the people who do well in finance generally don’t end up in Silicon Valley. The quant funds aren’t where VC-funded founders come from. Those people are making too much money to dance on their hats for VCs. Overwhelmingly, the people moving from finance into technology are coming from the least progressive financial firms, in which the 1980s hierarchy still holds and to be a programmer is still a low-status job. So, when those people become founders, what attitudes should we expect them to have?

Cost center life

Intelligently-managed firms, in finance as well as technology, recognize that the programmers are important. Others don’t, and finance’s negative reputation among programmers comes from them. In the shitty financial firms, programmers (and, to a lesser degree, quants) are viewed as a cost rather than a profit center. This isn’t news, nor is it shocking, and the slow decline of the organizations that think this way is exactly what they deserve. Those less progressive companies are slowly losing air, and this means that they’re cutting jobs. Job cuts are common in finance and often not a big deal, but there is an aura of “musical chairs” around certain categories of non-technical people. Where do they go, if there are no jobs for them on the Street? While they’d like to move to private equity, many can’t make that switch and fall down a couple levels into venture capital. If they’re lucky, they become VCs. If they’re not lucky, they become founders, grab a few million dollars (with many strings attached) and start bossing nerds around. Of course, their attitude toward technical people– that they’re a cost center– is one that they take with them.

This, I think, explains how technologists have been driven out of their own industry. Technologists have always had to answer to finance. A merger or an IPO can’t happen without the banks’ say-so. Technologists and the financiers, at some point, had a deal along these lines: we’ll build the product, you decide our bonuses. This played to the technologists’ middle-class earnestness, and their desire to put their heads down, do great work, and be fairly rewarded. It also gave the upper hand to the financiers: venture capitalists, Wall Street, and corporate acquirers. Over time, the “technology” industry became less technological and more marketing-obsessed.

If it had been competent financial professionals who’d moved into the Valley to manage its money problems, I think that the outcome might have been more positive. A successful hedge fund manager might not understand machine learning or Haskell, but she’d probably be inclined, by experience, to know how to hire smart people and get out of their way. She’d understand that making programmers second-class citizens will lead to shitty programming and that, while some established banks can afford that because their businesses are relationship-driven, technology companies can’t. However, said hedge fund manager wouldn’t dream of moving into the Valley. It’s the failed financiers who make that transition. And where does it lead us?

Ultimately, I believe that the best business people focus on the performance of the operation, while the worst ones focus on control. The talented ones focus on making more winning bets than losing bets, but accept risk and avoid sacrificing quality, performance, and morale just to have more control; while the untalented ones become obsessive micromanagers hell-bent on making sure the peasants know that they’re peasants. I don’t think that this aspect is finance-specific, but management-specific. Effective managers recognize that their teams have come up with ideas that they didn’t have, and that keeping their people creatively enabled is important. Ineffective managers, with painful histories of failure, a willingness to blame others for their problems, and the social status to get away with doing so, turn into distrustful, vengeful bean counters.

When you have someone who failed out of finance or management, has a chip on his shoulder against past and future underlings, and is more likely than not to come from a declining company in which programmers are undervalued… what you get is, well, exactly the sort of person who becomes a founder or VC amid the current Sand Hill Road climate. It shouldn’t be surprising that programmers have been made second-class citizens in their own industry. It should seem inevitable.

Where should one go?

In 2008, I left Wall Street to try out a startup (more than one) and ended up in the VC-funded world. Gigantic fucking mistake. I wouldn’t advise it to anyone. I don’t have a burning love for finance, but my career would be way ahead of where it is, had I stayed on the finance side of things. VC-funded tech (including ex-startups whose cultures were shaped by VCs) is a loser’s game. Sure, the software industry has a great need for intellectual talent, but its leadership is of such low quality (even in beloved companies that show up on peoples’ “Big 4” lists) that such talent is wasted. If I could go back to 2008 and tell my 25-year-old self anything, it’d be to stay in finance and not give this tech startup shit a moment’s thought. It’s called “the dream” because it ends as soon as you wake up.

For an individual, I’d say that a person choosing between a quant role in finance, versus an engineer position in the startup, should take the former. The pay is better and the jobs are better: you’re treated as an adult, rather than having to waste brain space on “user stories” and “scrotum ceremonies”. Finally, you have much better odds of becoming a founder (or a VC, if that’s what you want) after 10 years in finance (and making the right contacts) than after 10 years hopping around the startup world, from one loser engineering position to another. The startup world bends over backward in order to recruit the hordes of (mostly young, untrained, and mediocre) programmers that it needs in order to implement its vision, but the truth is that engineers are second-class citizens. I’ve been in discussions where a CEO sneered, “I’ll quit before I give an engineer 0.5 percent”.

I don’t dislike finance. If I could move back into it on the right terms, I would. I don’t find most of what financial professionals do, on a day to day basis, to be wrong or unethical. Moreover, while there are infrequent high-profile incidents of banks being unethical to customers and the public, the startup people and VCs have a stream of frequent low-profile incidents of screwing over their own people. Banks don’t go out of their way to ruin the reputations of, say, departing employees in the way that techies do. Bankers protect each other, which is something programmers never learned how to do.

All of this said, it’s a bit disheartening to come to the realization that I have, which is that a young person entering the technology world, rather than finance, is making a mistake. I’d love to be able to say something different. In 2007, the unambiguous best career option for a young person coming out of college was finance. It wasn’t even close, and startups weren’t on the map. I knew complete idiots who were 25 and making $500,000 per year in private equity. Then 2008 happened, the financial job market tanked, and it still hasn’t recovered. Many academics and economists decried finance’s mid-2000s strangehold on top talent and hoped that the Crash of 2008 would steer young talent toward Silicon Valley. To its credit, the Crash did move many people from finance into technology, but it didn’t happen in the right way. Quantitative, tech-friendly hedge funds kept their programmers, while non-technical soft players (the sorts of people who actually caused the crash) moved into technology– and in the top ranks. The idiots who were nudged out of real private equity called in some favors and became VC-backed founders. As a result, the Valley got sloppy founders, sloppy management, sloppy HR, sloppy culture, and in the end, sloppy product. And now, here we are: it’s 2015, and the programmers and scientists are second-class citizens– treated as cost centers rather than profit generators– in the world that they built. Finance hasn’t recovered to its pre-2008 levels, but technology has lost its soul as well, and the destination career for a talented young person is probably still Wall Street.

It doesn’t need to be that way. Technology can, and should, be fixed. However, it’s full of people with the wrong attitudes, especially at the top levels, and fixing this industry is going to require driving them out.