A few months ago, economist Robert Gordon made waves when he published a controversial paper titled “Is U.S. Economic Growth Over? Faltering Innovation Confronts the Six Headwinds.” It argues that most of humanity's economic growth has come from three industrial revolutions:
- 1750 - 1830: Steam power, railroads
- 1870 - 1900: Electricity, petroleum, internal combustion engine, running water, communications, chemicals
- 1960 - Present: Computers, internet, mobile phones
The first two industrial revolutions each took about 100 years to percolate all the way through the American economy, before diminishing returns set in significantly. Gordon believes the third industrial revolution (the digital revolution) is already sputtering out after 40 years, seeing how productivity growth has been decelerating since 2004.
Nevertheless, I think the digital revolution is still in its infancy. Imagine the worst case scenario: starting tomorrow, Moore's law halts. Even amidst such stagnation, I believe society has decades of innovation and investment left, just by using today's technology in new ways.
Today you can buy a computer for $25, and embedded chips for even less. Yet most things in our lives are not yet networked or sensing or intelligent. Why? I think this is not an issue of technology, but time. It does not cost much to manufacture networking, sensing, and intelligence capability into everyday devices. But it doesn't happen right away because there are large fixed development costs and large barriers to market acceptance. However, with enough time (and needing no new fundamental technology) these obstacles will be eventually overcome. The potential is too tremendous to be ignored.
One surprise that the digital revolution may have in store for us is the decoupling of price and cost. Digital information goods like games, movies, TV, sports broadcasts, and e-books all have high fixed costs and very low marginal costs. Very few goods like this have ever existed in history, and never alongside computer software that will allow companies to bundle goods, charge variable prices, or price discriminate at will. Companies that sell digital goods like they sell physical goods (i.e., with a fixed pay-per-item model) are doomed to fail in the long run. This is why services like Apple's iTunes will eventually lose against services like Spotify and Netflix. It is an economic inevitability (one that I will write more about later).
If I were a venture capitalist, where would I invest? What bets would I place on the future? Well, I'd look for opportunities where digital technology is enabling new ways of pricing, selling, and delivering goods. Broadly, these categories would include:
- Companies that increase capital efficiency
- Companies that replace labor with capital
- Companies the increase labor efficiency
- Companies that exploit digital delivery and its economies of scale
- Companies that price or bundle goods with high fixed cost
- Miscellaneous speculative technologies
Over the coming week, I will write a post for each category and the compelling businesses it contains.
Read Part 2 here.