Last week I wrote about Machine, Platform, Crowd: Harnessing Our Digital Future, a recently published book by MIT’s Andy McAfee and Erik Brynjolfsson. The book is organized into three main sections: Mind and Machine, Product and Platform, and Core and Crowd, each based on a business trend that’s significantly reshaping the business world. I based my discussion on the Mind and Machine section, which I found to be an excellent explanation of recent advances in artificial intelligence.
I now want to focus my attention on the provocative questions raised in the book’s latter chapters. “In this era of powerful new technologies, do we still need companies?” How are firms likely to evolve given the continuing advances of Internet-based technologies, as well as emerging blockchain-based technologies such as distributed ledgers and smart contracts? “Are companies passé” in an increasingly decentralized economy?
Following the Great Depression and WW2, the country welcomed the stability promised by corporate capitalism. Big, multinational companies dominated most industries, - GM, Ford and Chrysler in cars, Citibank, American Express and Morgan Stanley in finance, and Esso/Exxon, Mobil and Texaco in oil and gas. It was an era characterized by bureaucratic corporate cultures, - not unlike military hierarchies - focused on organizational power and orderly prosperity.
This all started to change a few decades later with the advent of a more innovative, fast moving entrepreneurial economy. The 1980s saw the rise of young, high-tech companies, - e.g., Microsoft, Apple, Oracle, Sun Microsystems, - and Silicon Valley became the global hub for innovation, emulated by regions around the world. Turtlenecks and jeans replaced blue suits, white shirts, and ties.
The advent of the Internet and e-commerce pushed all these trends into hyperdrive in the 1990s. The Internet made it much easier for companies to transact with each other around the world. Vertically integrated firms evolved into virtual enterprises, increasingly relying on supply chain partners for many of the functions once done in-house. Management experts noted that large firms were no longer necessary and would in fact be at a disadvantage in the Internet era when competing against agile, innovative smaller companies.
But, not only are large companies alive and well, but as The Economist noted in a 2016 article, “the most striking feature of business today is not the overturning of the established order. It is the entrenchment of a group of superstar companies at the heart of the global economy.”
The growth of large companies has been most evident in the US. The Economist analyzed almost 900 industries across a number of sectors, and found that the weighted average share of the top four firms in each sector rose from 26% to 32% between 1997 and 2012. In addition the share of GDP generated by the 100 biggest companies rose from 33% to 46% between 1994 and 2013, and the five largest banks now account for 45% of all banking assets, up from 25% in 2000.
While the Internet, blockchain, and other technology advances would seem to support the decentralization of the economy away from big companies, the data indicates that, at least so far, their demise is not evident. In fact, the economy has been moving in the opposite direction. To understand this seeming paradox, let’s turn to The Nature of the Firm, Ronald Coase’s 1937 seminal paper, which along with other achievements earned him the 1991 Nobel Prize in economics.
Why do firms exist? Professor Coase provided a simple, elegant answer to this question. In principle, a firm should be able to find the cheapest, most productive goods and services by contracting them out in an efficient, open marketplace. However, markets are not perfectly fluid. Transaction costs are incurred in obtaining goods and services outside the firm, such as searching for the right partners and prices; coordinating an ecosystem of partners; negotiating, monitoring and enforcing contracts; and so on. Firms came into being to make it easier and less costly to get work done.
A firm will keep expanding, adding people and layers of management and staff as long as doing so is less expensive than securing the additional services in the marketplace. But, there are limits to how big a firm can get and still remain competitive against faster moving companies. The additional layers of management and staff can cause the organization to become bureaucratic, significantly impacting its ability to quickly embrace new ideas and technologies when market conditions change. A well managed company strives to achieve an optimal balance between what work gets done within and outside its boundaries.
Advances in IT systems and networks have had a big impact on the nature of firms over the past few decades. The Internet has significantly reduced external transaction costs, giving rise to global supply-chain ecosystems. Nevertheless, big companies keep growing instead of shrinking as the economy becomes more digitized.
Blockchain technologies, - e.g., distributed ledgers for replicated, shared, non-revocable data; self-executing, self-enforcing smart contracts; mathematical models to enable cooperation among non-trusted parties; - could further redefine the balance between companies and markets over time. “It’s pretty clear that a large movement is under way to take much of the work that used to be done within the single hierarchy of the firm and move it to the market,” write McAfee and Brynjolfsson. “It’s also clear, however, that firms are still going strong, and that in many ways, their economic influence is growing, not shrinking.”
The reason, they argue, is based on transaction cost economics (TCE), built on Coase’s path-breaking work as well as that of other prominent economists. Let me attempt to summarize their argument.
An important TCE concept is that “if every possible contingency for use of a building, machine, or patent were spelled out in contracts, then labeling one party the owner of the asset would confer no additional rights. However, when contracts are incomplete, owners have the residual rights of control, meaning they can do whatever they want with the asset except for what’s in the contract.”
If one could write a complete contract, embed it in software as a smart contract, and store it in a blockchain-based distributed ledger, transactions could be handled with no need to worry who owns what assets because all future decision are being adjudicated by a comprehensive, tamper-proof smart contract. In such a case, all decisions can be made in a purely decentralized market with no need for firms, bankers, lawyers or other intermediaries. However, when a decision about an asset has to be made that’s not specified in the contract, only the firm that owns the asset, - that is, that holds the residual control rights to the asset in TCE terms, - can make such decisions.
Virtually every economist believes that, in practice, complete contract aren’s possible. No matter how smart contracts get, they’ll still be incomplete. The reasons are fairly straightforward. The world is very complicated and the future is largely unknowledgeable. “These and other factors combine to make it prohibitively difficult, and most likely actually impossible, to write a complete contract - one that truly does away with the need for ownership - for any realistic business situation…”
“One of the fundamental reasons that firms exist, then, is that it’s just not possible for market participants to get together and write complete contracts - ones that specify who does what, and who gets what, in all possible contingencies: all the ways the real world could unfold in the future.”
What about the future? Could advances in computers, AI, blockchain and other technologies make it possible to anticipate future possible contingencies and outcomes? McAfee and Brynjolfsson noted that like the Red Queen in Alice in Wonderland, machines would have to run ever faster to keep up with an increasingly complex and unpredictable world. “In the end, contracts would still be incomplete,” they concluded.