The total value of the companies we’ve funded is around 10 billion, give or take a few. But just two companies, Dropbox and Airbnb, account for about three quarters of it.
In startups, the big winners are big to a degree that violates our expectations about variation. I don’t know whether these expectations are innate or learned, but whatever the cause, we are just not prepared for the 1000x variation in outcomes that one finds in startup investing.
-Paul Graham, Black Swan Farming
The freelance writer has to hustle every day for gigs, and some months are better than others. The staff editor is always well fed; the freelance writer is hungry on some days. Then the day comes when print finally dies, the magazine industry collapses, and the staff editor gets laid off. Having built up no resilience, he will starve. He’s less equipped to bounce to the next thing, whereas the freelance writer has been bouncing around her whole life— she’ll be fine. So which type of career is riskier in the long run, in the age of the unthinkable?
-Reid Hoffman and Ben Casnocha, The Start-up of You
The Industrial Revolution was characterized by the rise of well-defined, specialized, routinized jobs. Adam Smith made his observations about pin factory workers more than a hundred years before Henry Ford’s assembly line became an icon of modernity and efficient industry.
With routine work came routine jobs, and routine paychecks. Modern industrial era employment, while taken for granted today, is something of a historical novelty. Before this Bourgeois Era we live in, an overwhelming supermajority of humanity lived on farms, and the rest were aristocracy or warlords of one stripe or another.
Farm life was lumpy–every year had the high point of the harvest, sometimes even with a subsequent festival in the nearby town. Then every year had its long, hard winters. Then there were particularly lumpy years; a bad harvest could wipe out a whole village while a very good one would be the subject of conversation for years afterwards, and might result in a temporary growth in the population.
Lumpiness in Modern Life
This is not to say that modernity has been all smooth trend lines and uninterrupted flow. Nassim Taleb would certainly protest such a claim. Even if we are speaking in strictly economic terms, there have been big, dramatic events of the negative and positive sort. The Great Depression comes to mind. The hyperinflation of Weimar Germany. On the flipside, the German and Japanese post-war Miracles. The sudden gentrification of American cities that had been in decline for decades.
And on a company by company and individual by individual basis, there has been a lot of lumpiness. Google went from a Stanford computer science project to a multibillion dollar company within a handful of years. Apple rose and fall and then rose far more spectacularly than ever.
Taleb has argued that the more informational economic activity is, the lumpier it will be. Thus, the content industries, and finance, have always been lumpy. The scalability of informational goods makes it possible for a book, such as Harry Potter, to be a best seller across the entire planet, raking in enormous amounts of money. Meanwhile hundreds of thousands of books that come out every year won’t sell more than a handful of copies; for we are a groupish species and we like to focus on a small subset of things that can create a common experience.
This latter piece is preferential attachment; if one person’s consumption of an informational good increases the odds that someone else will consume it “by even a fractional amount“, it will create extremely skewed distributions. And there are well understood reasons why the book business has always been skewed, and why globalization and digitization will only skew it further.
Also skewed, though not quite so dramatically, is income in a human lifespan–into your “peak earning years”. Then there is the well documented phenomenon of extremely skewed healthcare spending–dramatically backloaded into the last handful of years and handful of months of your life.
So we are no strangers to lumpiness. But it seems to me that we are blind to it. As Paul Graham notes, it “violates our expectations”. We expect life to be more like the smooth streams of compensation that the industrial revolution has provided us.
We are going to have to adjust, though, because there is good reason to think that those smooth streams are going away for good. Things are about to get a lot lumpier.
The Robots Are Coming
The paradox is this. A job seeker is looking for something for a well-defined job. But the trend seems to be that if a job can be defined, it can be automated or outsourced.
Our capacity to automate seems, at times, to be limitless. One thing is for certain, however, and that is that if it is repetitive and has clearly defined parameters, we can automate it. The Kling quote above actually understates the extent of the circumstances by bringing outsourcing into it. The fact is that even in China, where labor is much, much cheaper at a far higher scale than any developed nation, they are moving towards automation. Does this sound familiar:
China’s manufacturing output was over 70% greater in 2008 than it was in 1996. Over the same period, manufacturing employment in the country declined by more than 25%.
This is the exact same trend that we have been seeing in the United States for half a century, only, as with everything else, China is playing catch up and so the trend has accelerated there. While politicians and pundits in America blame outsourcing for the loss of manufacturing jobs, the fact of the matter is that our manufacturing output never stopped growing; it was only manufacturing employment that declined.
This trend, explored at length in Race Against the Machine, is not without historical precedent. Remember, we were an agricultural nation before we were an industrial one.
A century ago, 40 percent of Americans worked on farms. Today, the farm sector employs about 3 percent of our workforce. But our agriculture economy still outproduces all but two countries.
Some believe that the pattern will play out in a similar way all over again–manufacturing and anything else that can be automated will shrink down to single-digit percentages of our employment. But entrepreneurs will think up new ways to put people to work en masse.
A more pessimistic story, believed by Robin Hanson for example, is that there’s no going back. Automation has grown so good that the majority of people simply will never gain the skillset to be able to provide comparable value, in any sector. Anyone who has a successful company will be able to use automation to produce on an unimaginable scale and thus become unimaginably rich even by today’s standards, but a large segment of the population will not be able to find any way to contribute value whatsoever.
I am proposing a different story: we will all learn to live with ultra-lumpy incomes.
A World of Black Swan Farmers
Join me for a minute in our automated future. It only takes a few tens of thousands of people to produce agricultural and manufacturing output per capita on a scale we would consider absurdly large today. Delivery and postal workers have been put out of work by tacocopters. Maids, fast food workers and cooks have all been replaced with robots. What are we to do?
Well the first upside is that everything is extremely cheap. We can produce so much food, and so much stuff, and provide so many services, that our huge supply will drive prices straight down. So you don’t need a lot of money to maintain a standard of living that would be considered affluent by historic standards.
OK, but where does even that little bit of money come from?
We will all have to adjust to the lack of routinized and easily definable jobs by becoming a little like venture capitalists. We will put out blog posts, and Kindle books, and apps, and any other sort of informational good that we can, in the hopes that one blockbuster will support us for a while.
Since these black swans are, by necessity, very rare on a case by case basis, we will probably combine our efforts and share the spoils. The most obvious way would be for the member of the family who manages to get a hit to take care of the rest until the next hit comes along. But perhaps we will explore many more kinds of partnerships and legally binding revenue-sharing arrangements in order to cope with this radically different labor market.
And again, because of expanding supply and falling prices, you need not have a big hit in order to support yourself. Maybe 100,000 blog views will pay enough, through AdSense revenue, to feed you for a few months.
I can imagine a world where people have only periodic income and they have a higher standard of living than we currently do. I can imagine things would seem psychologically more tenuous in such a world, but it’s not as though anything was ever guaranteed under the old way. And maybe we will adapt, psychologically.
Do you think that you could live happily in a high volatility gig economy?