NOte: formatted version here.

25 years ago, I thought that the Internet would erode the advantages of large corporations, creating a business playing field geared more toward ordinary individuals and small businesses. Economic inequality would be reduced.
It seems that this was incorrect. Why was I mistaken?
In assessing the factors that determine economic success in the Internet era, I was steered wrong by neoclassical economic thinking. I saw that the Internet and Moore’s Law were making capital relatively abundant. That, I thought, would strengthen the power of the typical individual and weaken the relative power of large corporations. I pictured an economic democracy of small Internet-based businesses, not a world of hegemons like Google or Facebook or Amazon.
If we still lived in a primarily industrial economy, dominated by tangible factors, then capital abundance would indeed benefit the mass of workers. Income and wealth would become less concentrated.
But the 21st-century economy is dominated by intangible factors, including technical skills and managerial talent. The rise of computers and the Internet dramatically increased the economic power of elites in various fields, not all of which are technical. As a result, income and wealth have become more concentrated. Most people are actually at a relative economic disadvantage compared with where they would have been 25 years ago. (They may be better off in some absolute sense, but relative to the economic elites, the masses have lost ground.)
The intangible economy includes several talent effects:
— small teams are now more effective relative to large masses of workers;
— talent tends to become more concentrated, because the most talented individuals get better
— talented individuals gravitate toward teams with other talented individuals
The Mythical Man-Month
In the industrial era, managers increased output by deploying more equipment and hiring more people. They might calculate the required labor input to build 5000 automobiles as a certain number of man-months.
In a classic book of essays that appeared in 1975, Frederick P. Brooks pointed out that such “man-month” calculations do not apply in software development.
when schedule slippage is recognized, the natural (and traditional) response is to add manpower. Like dousing a fire with gasoline, this makes matters worse, much worse. More fire requires more gasoline, and thus begins a regenerative cycle which ends in disaster.
— Frederick P. Brooks, “The Mythical Man-Month”
In a 1970’s-era automobile plant, introducing a new worker to the assembly line was easy. Getting the worker to be productive required very little managerial oversight and only brief communication with other workers. But Brooks pointed out that computer programming is different. Each new person on the project adds significantly to the burden of management and communication.
If you are trying to get a lot accomplished, are you better off with a small, elite team or with a large collection of average workers? Well, if you are trying to build a lot of cars using 1970’s technology, the large team of average workers will out-produce any small team of elite automobile assemblers. But in software, the small elite team is more likely to win. Hence, Amazon is known for its “two-pizza rule” of trying to keep software teams limited to the number of developers who can be fed with two pizzas.
The Best Get Better
Another characteristic of the post-industrial economy is that the most talented individuals tend to get better. Think of the virtuous cycle of a movie star. Because of her talent and reputation, she gets offered the best roles. This in turn allows her to improve her skills and enhance her reputation further.
The virtuous cycle of the best getting better is true in many parts of the business ecosystem. The best software developers are recruited by the leading-edge companies. The best marketing managers get offered the most interesting marketing jobs. The best PR firms are sought after by the most exciting firms. The best venture capitalists get pitched by the most promising start-ups.
This virtuous cycle generates enormous barriers to entry. It creates an imperative to “work with the best and forget the rest.” The best PR firm will have the contacts and the credibility with journalists to achieve your goals; other PR firms, no matter how hard they try, will not serve you as well. As a start-up raising funds, if you go with a top-tier venture capitalist, with their connections and experience they will also help you solve other problems. If you go with a lower-tier firm, chances are that firm will wind up creating more problems than it solves for you.
Elites Marry Endogamously
A tribe in which members are only allowed to mate with other members is said to marry endogamously. Members of the tribe of high-skilled executives tend to gravitate toward teams that include other high-skilled executives. A top marketing executive is not going to join a software company with a mediocre chief technology officer, and conversely. Elites marry endogamously.
This can be seen particularly clearly in the setting of high-growth companies, as described by Elad Gil in his recent book, The High-Growth Handbook. The book is about a small but important segment of the business ecosystem, consisting of companies that have successfully navigated the start-up phase but which are not yet mature. They are in the process of growing a proven experiment into a large-scale business.
Companies in this high-growth phase must very quickly evolve from small, simple, informal outfits to large, complex, and more systematic organizations. The pace of change is mind-boggling. As Gil puts it, a high-growth company can become a completely new organization every six months until it reaches maturity. This is a unique environment, which is disorienting both to someone who is used to small start-ups and to someone who is used to working at large, mature corporations.
As a result, high-growth firms tend to need executives who have been successful at other high-growth firms. This makes for endogamous marriage within a very small tribe.
In fact, we can think of firms that achieve high growth as firms that have created successful endogamous marriages. If it were not for the synergy that arises when elite executives with different specialties work well together, firms would not be able to survive as they took on the complex challenges of high growth. Instead, we would be more likely to see the business ecosystem looking like what I expected 25 years ago.
In Brief
Talent effects are driving inequality in incomes and wealth. In many industries nowadays, small teams of talented individuals can out-compete larger collections of mass workers. Elite skills, reputations, and connections can create barriers to entry that produce high returns. In some important fields, the stars get the best jobs, which in turn enables them to enhance their know-how and their reputations. And the most talented people in one field are likely to work in firms with the most talented people in other fields, creating synergies that increase their rewards even further.