Mark Cuban and the Dallas Mavericks vs. BS Jobs
Why are there so many bullshit jobs? Also, annual letters vs. Substack.
The late anthropologist David Graeber begins his explanation of his theory of “bullshit jobs” by recounting a famous failed prediction: The economist John Maynard Keynes said in 1930 that technology would allow his grandchildren to achieve a fifteen hour workweek, but despite massive technological advances, we now work more than ever.
Instead, Graeber says, we got a proliferation of administrative white collar jobs that provide no benefit to society, and in fact wage “spiritual violence” against the unfortunate souls that must go to work every day pretending that what they do matters, against all available evidence to the contrary. He contends that we could eliminate all of the PowerPoint decks, pointless meetings, and paper pushing, and society would not miss a beat.
His definition of bullshit jobs is narrow; he specifically excludes retail service jobs such as waiters and professionals such as teachers and doctors, whose employment has stayed steady over the last century, as well as jobs in manufacturing and agriculture, which have declined precipitously because of technology. He believes these are all tangible, worthy jobs.
His target is white collar information workers whose ranks have swelled in recent decades, particularly in the private sector: Bankers, lawyers, managers, public relations and marketing roles, and administrators of any kind. (In short, anyone who might be reading this essay.) He cites the 1970s as a turning point, stating that “there seems to be an intrinsic connection between the financialization of the economy, the blossoming of information industries, and the proliferation of bullshit jobs”.
Whether one agrees with his thesis or not, it is difficult to reconcile the technological advancements and productivity growth of the last half century with the explosion of seemingly pointless jobs in the private sector. Information technology has made white collar workers more productive, so why hasn’t technology and competition reduced white collar employment in the same manner that it reduced employment in manufacturing and agriculture?
Mark Cuban sold Broadcast.com (Radio! On! Internet!) to Yahoo! In 1999 for $5.7 billion. On January 4, 2000, he bought the struggling Dallas Mavericks from Ross Perot for $285 million. This was a lot for an NBA franchise at the time; the San Antonio Spurs traded hands for only $75 million less than seven years earlier.
Technology has been a huge tailwind for the NBA and the other major professional sports leagues in the last forty years. Television expanded the size of the arena, but cable television made it possible for teams to broadcast all of their games all over the world in high definition and charge fans (and everyone else) for the privilege of being able to access them. As cable TV expanded in the 1990s and 2000s, so did the size of the league’s rights deals, and with it, player salaries. In 1985, the salary cap—roughly, the amount a single team can spend on all player salaries—totaled only $3.6 million. By 2000, the time of Cuban’s purchase, the salary cap had grown to $34 million. The salary cap would reach $109 million by 2019.
It is against this backdrop that Cuban purchased the Dallas Mavericks, with the intent of turning them around. Looking to make a splash and attract better players, he upgraded the locker room. He bought a private plane. He hired statisticians, an unusual step in the pre-Moneyball era. But his most interesting move was to hire several player development coaches, eventually achieving a coach to player ratio of 1:1.Said Cuban later:
I remember having to explain to the sports media that we spent more on support of our personal computers and software than we did developing our most important employees: our players. Now every team has a staff of player-development coaches.
The intuition behind his insight is simple. In 1985, members of the starting lineup for the Mavs were averaging $400,000 a year. By 2000, the starting lineup for the Mavs was making closer to $5 million each. Hiring a $100,000 coach for a single player would have to improve that player by 25% in 1985 to pay off, but by 2000, a coach making $100,000 would only need to improve the productivity of a single player by 2% to be worth the investment.
The Mavericks would make the NBA Finals in 2006 and win a championship in 2011, and teams throughout the league would follow their lead by investing heavily in new white collar positions that could give them an edge, however slight. Teams now employ analytics experts, psychologists, scouts, trainers, and legions of coaches of all stripes. Even though the rules of the game still only permit five players on the court at a time, rosters have expanded to allow for specialist players and reserves and a developmental league has opened up, to train aspiring players. The trend of hiring data analysts and specialist coaches has extended to other leagues; at the end of 2020, the NFL’s Houston Texans were embroiled in a controversy around a disruptive employee that was originally hired as a “character coach”.
The numbers are actually quite staggering. One NBA team website lists 42 non-player employees in basketball operations alone; coupled with the 17 players now on an NBA roster, that implies an NBA team now has a ratio of 50 “bullshit jobs” to the nine or so players that can actually see meaningful time on a basketball court in games. This does not even account for the dozens of other administrative white collar jobs around the team—marketing, PR, finance, HR, IT, and all of the other functions that one would commonly see in a $200 million a year business.
The NBA illustrates Graeber’s point, in a way. It is likely that an NBA team would be almost as successful if it got rid of the bullshit jobs and just had nine players and a coach drawing up plays, closer to what it was in the early days of the league. But it is also likely that a team’s fan base and players would revolt if an owner got rid of all of the bullshit jobs, because of what it would signal about their willingness to compete.
It is interesting to examine the economic logic at work here. An advance in technology that makes the inputs of your business more productive should induce you to invest more in those inputs (both labor and capital), all things being equal, which is exactly what happens in this example. When computers got more powerful, businesses bought more computers and hired people to use them. Mark Cuban responded to cable TV by hiring coaches and buying planes.
What makes the intuition so difficult is that usually, all things are not equal. In the case of agriculture, for example, the number of calories consumed in the world is relatively stable at any price, so in a simple model, any advance in technology that makes farmworkers more productive has to be offset by a reduction in the number of farmworkers.This is apparently less true in the NBA, where fans around the world seem to have a bottomless desire to watch their team. More industries look like agriculture than professional basketball, but industries like professional basketball are now employing a growing share of the workforce.
Mark Cuban was clever enough to recognize that in a growing business, he should be looking for creative ways to invest more in his team, and he figured out that coaches and statisticians cost about the same as they did before, but because of technology, they were now many times more productive. As the stakes grew, other teams copied him and looked for new ways to gain small edges that might now be economically feasible, leading to the proliferation of specialist jobs described earlier.
Information technology has had a similar impact in other industries. It has rendered some specific roles obsolete–in 2020, the Wall Street Journal reported that 1.6 million secretarial and administrative-assistant jobs have been lost in the U.S. since 2000–but overall, the number of information workers is growing. Again, this is generally what the theory predicts: If information technology makes information workers more productive, we should see more information workers in the world, until we hit a point of diminishing returns, or until there is no more demand for what they produce.
If Graeber’s thesis is right, that people with bullshit jobs are not productive, we should see more successful companies avoid employing them. But in fact, it seems like every memoir by a very successful entrepreneur or executive recounts how they tested this thesis and failed. Google, prior to its IPO, had a well-known failed experiment where they eliminated managers. Frank Slootman recounts in Tape Sucks how he tried to underinvest in finance and HR when he took his first CEO job at Data Domain, and narrowly avoided disaster; the CFO he hired to clean it up recently followed him to Snowflake, where Slootman got him an options package worth $300 million a year (at the current stock price).
The best examples of startups that avoided bloat and achieved great exits are Instagram, sold in 2012 for $1 billion when they had 13 employees, and WhatsApp, sold for $19 billion in 2014 when they had 55 employees. These ended up being fantastic acquisitions for Facebook, so much so that the FTC recently sued to have the deals reversed. Perhaps they would have had a better outcome if they had invested more in the business and held on.
In a world where technology increasingly enables a small number of superstars to satisfy a large portion of the needs of consumers around the globe,we are left with the fact that most of us cannot be all-stars, or even part of the starting lineup. If current trends continue, the rest of us must be content to be backups, benchwarmers, practice squad players, coaches, assistant coaches, junior assistant coaches, video analysts, and so on. (One wonders if this explains the increasing popularity of startups–the ability to be a starter on a smaller team.) Hopefully, Graeber is overstating his case when he claims that bullshit jobs wage spiritual violence, or that modern work arrangements resemble sadomasochism without a safe-word. If technology has truly deprived us of fulfilling jobs, at least we can all go home at the end of the day and watch the NBA on TNT.
Annual Letters: Newsletters before Substack
Just in time for annual report season, Bread Crumbs Research on Twitter shares a curated list of best all-time annual letters. Before the Internet, you could keep up with a company by buying a few shares, at which point the company would be obligated to mail you a glossy annual report each year, which included the CEO’s letter, containing their thoughts on the industry, the business, and any other relevant topic that came to mind. (You still can, of course, but it’s easier to just visit the company’s investor relations website now.) As the author of the thread observes, “Most shareholder letters are either outright garbage or some boilerplate stuff. But there are gems as well, which are rare, but worth looking for.”
Warren Buffett, when asked, would say that he never owned Microsoft stock, but he would caveat that by saying that he did make a token purchase of a hundred shares when he met Bill Gates in 1991, so that he could keep up with the company through its annual reports. As a method of getting interesting reading material about business, buying a bit of stock requires more upfront cash than a Substack subscription, but sometimes there’s an interesting return on the back end. By my accounting, Buffett’s token $7,000 investment in Microsoft is worth over $1.2 million today, not that he needs the extra cash.
There is probably some correlation between the ability to write a good annual letter and the ability to achieve a high long term return on investment. If you (or a ghostwriter at the company) cannot explain the business in a clear manner, you probably will not be able to manage it either. The list includes Jeff Bezos’s famous Year 1 letter in 1998; if you bought a hundred shares when that letter came out, you would have about $3 million today. (Amazon and Microsoft are either a testament to the value of good business writing, or a testament to the value of Seattle-based cloud computing companies.) For the record, Warren Buffett began writing his famous annual letters in 1978; a hundred shares purchased then (a manageable amount, at a bit over $100 per share) would be worth $35 million today.
It is worth noting here that some of the initial logic here is driven by the NBA salary cap, which limits spending on player salaries, but leaves the door wide open for spending elsewhere in the organization. However, the same trends quickly spread to sports like soccer and baseball, which have no salary cap.
I know this is an oversimplification. Also, yes, you invest more in farm equipment.
Economics students will recognize that this essay is about income and substitution effects, which allow you to analyze a technology or policy shift by breaking it into its constituent components: A new technology or subsidy makes people richer (income effects), but it will also make a good relatively more or less attractive (substitution effects). The Internet makes everyone richer, but will cause consumers and firms to substitute what they invest in and consume.
This is the part of the essay where the writer usually cites Sherwin Rosen’s The Economics of Superstars, and bemoans the growth in income inequality that results, but if you are reading this, I will assume you are already familiar with it.
Most of us can't be all-stars if all-stars are billionaires or multi-millionaires, but the real truth is that not even a single person needs to be an all-star if we're going by that. One problem -- and this doesn't necessarily disprove or disagree with anything you've said -- is that a lot of people get into entrepreneurship these days with the goal of becoming a star or sorts, rather than creating a profitable business.
"What makes the intuition so difficult is that usually, all things are not equal. In the case of agriculture, for example, the number of calories consumed in the world is relatively stable at any price, so in a simple model, any advance in technology that makes farmworkers more productive has to be offset by a reduction in the number of farmworkers."
This is not quite true, or not solely. In this case, we can afford to be less efficient and spend said resources to feed animals so we can consume more and cheaper meat.