There has been a renewed interest in the impact that digital technology is having on the boundaries of the firm. In particular there is a sense that networks operating on top of platforms are competing effectively with traditional firms. Examples include AirBnB as competing with hotels without owning any real estate, Uber with transportation companies without owning and cars, LendingClub with banks without having a balance sheet, Upwork (formerly ELance/ODesk) a systems contractor without employees, etc. (Aside: I am planning to write a future post about the many profound differences between these which get lumped together as if they were somehow the same).
When most people write about the boundaries of the firm they refer to Ronald Coase’s groundbreaking work on “The Nature of the Firm.” This work is well known but it dates back to 1937 and while Coase provided a breakthrough in thinking his grasp of the nature of transaction costs was limited. A lot of work since then has gone into giving much more specificity to the nature of the costs by economists such as Oliver Williamson, Michael Jensen, Bengt Holmstrom, John Roberts, Oliver Hart, Paul Milgrom and many others. If you want a great overview of this work I highly recommend the reader put together by Randall Kroszner and Louis Putterman. If you want a nutshell summary of the most important tradeoff keep reading on.
There are two fundamental problems to organizing economic activity: motivation (getting people to expend effort, companies to allocate resources to an activity) and coordination (arranging activities so that they fit together, such as making sure a part needed for production arrives on time). It turns out that there is a tradeoff between the two and the degree of tradeoff is influenced by the available information technology. The TL;DR version: better, faster, cheaper access to information reduces the degree of the tradeoff and makes the network model possible.
To see the tradeoff just consider what happens inside a firm. The firm pays employees a fixed salary. As a baseline that salary is the same independent of what specific task the employee works on. That allows the firm to highly coordinate the activities of employees. But of course it reduces the employee motivation to work hard (because the salary doesn’t respond to the intensity of effort). Pretty much everything on top of salary that you see in compensation, such as stock options, bonus plans, etc. is designed to bring some of that motivation back.
A competitive market is the opposite extreme. Each party gets to keep the additional gains arising from effort so motivation is maximized. Restaurants are a great example and this explains at least in part the extraordinary work efforts in that segment. But the activities between individuals restaurants are not coordinated. Try ordering an appetizer from one place and a main dish from another with a dessert from a third. In markets then we see devices such as long term contracts as a way to bring back some degree of coordination.
Now the beauty of the advance in information technologies is that they shift out the tradeoff frontier (or envelope): it is now possible to achieve more motivation and more coordination than was possible before. Here is the intuition for it. Without any communication technology you only act on the information available to you locally. At that time in history we had lots of markets and a few small firms (e.g., a farm, a shop, a local manufacturer) in which the employer works directly with the employees. As you add limited communication technology, such as the telegraph, you can relay information to a central office which can figure out the best course of action and send decision back. That stage gave rise to large firms that eventually spanned the globe. Now comes something like the Internet and everyone can have access to all the information, that is each node can evaluate its actions in the global context of any benefits from coordination. This gives rise to networks and other platforms and hybrid organizations that organize the gathering and sharing of information. If you want a more rigorous analysis of this I suggest the second essay from my dissertation.
But it is super important to understand that the motivation-coordination tradeoff has *not* been eliminated. It has just been reduced. And that means we will still see both firms and markets in addition to the network model even though the network model will likely grow a lot. If a market (or strategy within a market) calls for extreme coordination, the firm model will outperform the network. For instance, to create the perfect consumer experience for a smartphone Apple has integrated vertically backwards all the way to making its own chips. Conversely if motivation is of the essence the market model will outperform a network. Athletes in individual sports such as tennis or golf are a great example.
