by George Taniwaki

Two announcements made yesterday seem to point to a cosmic convergence, or at least a law and economics convergence. First, is the death of Ronald Coase. The other is Microsoft’s plan to acquire the mobile handset business of Nokia.

Ronald Coase 29 December 1910 – 2 September 2013

Why are there firms? Why doesn’t everyone just work for themselves and contract out their labor to each other whenever and wherever it is needed? Conversely, why doesn’t everyone work for a single large employer that tells everyone what to do and produces everything that consumers want to buy?

The answer to this question of firm size is the basis of a branch of economics called the theory of the firm. Theory of the firm sounds like a single idea. But actually there were many competing explanations regarding why firms exist. The most important contribution is by Ronald Coase, a British economist who taught at the University of Chicago who proposed that the ratio of internal transaction costs to external transaction costs drove firm size.

In 1937, Mr. Coase published a paper “The Nature of the Firm” in Economica which outlines his is ideas about how in certain cases transaction costs can be reduced by replacing a market driven  contractual relationship between a buyer and seller with an internal non-market relationship. (Also called insourcing as opposed to the current controversy about firms getting smaller by shedding operations in a practice called outsourcing.)

All transactions involve a contract, either explicit or implicit, written or verbal. Complex contracts often require expensive lawyers who consider possible contingencies and carefully define the outcomes. In addition to legal costs, contracts have to be monitored and enforced to ensure both parties get what they expect in the relationship. The more complex the task, the more complex and expensive the contract is likely to be over its lifetime

Much of the effort in using a contract is to ensure that all risks are explicitly defined and divided between parties. This effort can be eliminated if the parties join into a single firm. Then the risk where one party’s gain is another’s loss can be shared.

We  live in a world with many sized firms, so thinking about two extreme cases posed at the beginning of this blog post may seem silly. But they lead to important questions. What factors determine the size of a firm? What is the “ideal” size of a firm? Given that we see lots of firms of different sizes, is the current distribution of sizes “efficient”? If not, what should we do, if anything, to help owners, managers, and workers “right size” their firms? Finally, can we predict the effect of technological and social changes on firm size?

These are important questions worth investigating. And all of them can be analyzed using the framework first developed by Mr. Coase.

In 1960, Mr. Coase again used transaction costs to develop a framework to study the problem of externalities (the failure of the market to capture all benefits or costs of a transaction, such as pollution). He showed that if the resource being impacted (such as clean air) could be well-defined and protected as a property right, and if transaction costs were low, then rights owners could bargain for use of the property. Further, he showed that it didn’t matter who owned the property rights initially. As long as all parties bargained fairly and rationally, the utility maximizing outcome would prevail.

It is on the basis of his groundbreaking work on transaction costs that Mr. Coase was awarded the Nobel Memorial Prize in Economics in 1991.


Figure 1. Ronald Coase in 2011. Image from


Microsoft and Nokia, better together?

How big should a mobile computing firm be? The biggest firms in the market today are huge. How should a firm profit from the mobile computing market? Should it make its software and hardware available only in combination like Apple with iOS and iPhone? Should it make software and license it to a small set of select vendors, like Microsoft with Windows Phone? Or should it make software available to all comers including itself, like Google does with Android?

Well, it’s pretty obvious that the licensing strategy for Windows Phone has not worked. The only licensee of volume is Nokia, and that is about 3.5% of the market. Handset manufacturers were not willing to pay the estimated $10 per handset fee for it. Instead they licensed Android for free, plus an estimated $5 per handset patent royalty to Microsoft and Apple. (Given that Android has 12 times the market share of Windows Phone, that means Microsoft makes most of its mobile computing revenue from Samsung and other makers of Android phones.)

Now Microsoft has announced it will acquire Nokia for $5 billion plus $2.1 billion to license its patent portfolio. What explains the belief that the performance of a combined Microsoft + Nokia will be superior to Microsoft and Nokia as independent entities? Ronald Coase would probably say that using transactions cost analysis is a good place to start.

In the 1980s, Microsoft made an enormous amount of money realizing that it could sell its operating system software to many hardware manufacturing firms with little or no customization. In the desktop computing era, Microsoft paid for nearly all the research and design for desktop computers and charged it to manufacturers as part of its license fee for Windows. These licensees competed mostly on price and ended up with slim margins. What R&D they did pay for was focused on improving supply chain management and manufacturing efficiency. Almost nothing was spent on new computer designs or user experience.

Apple guessed wrongly that people would pay a premium for computers where the software and hardware worked together better and were a joy to use. They didn’t.

Keeping software and hardware as separate firms made sense for Microsoft. All the profit was in software and there was little innovation or value added in hardware.

Then in the late 2000s, mobile computing took off. Unlike desktop computers where size, shape, weight and design are not particularly important, mobile devices are all about pushing the technical limits of computing while fitting it into a handheld package. Apple, which already had an experienced design staff and a CEO who prized good design, immediately had a hit with the iPod and later the iPhone.

Microsoft revamped its mobile OS (was called Windows Mobile, now called Windows Phone) but had difficulty finding licensees that could create phones with compelling designs. One solution is to include a subsidy to the hardware manufacturer to pay for improved software and hardware integration. But contracts to guarantee innovation are hard to write and nearly impossible to enforce. Thus, Microsoft (and Google) found itself in a difficult position. How can a manager at a software company make hardware innovation a priority at its licensees? One solution is to insource it.

Horace Dediu has an excellent blog post that asks, Who’s buying whom? He states that firms are the sum of three values, resources, processes, and priorities (RPP). He says that, “When one company buys another, it’s the equivalent of one set of RPPs trying to engulf or swallow another set of RPPs.”

It is pretty easy to see that when one company buys another for its resources (people, plants, patents, store locations, customer lists) or processes (workflows, algorithms, supply chains) that it wants to incorporate them into its existing infrastructure.

However, Mr. Dediu states that acquiring another company’s priorities is different. “A company typically only has room for one set. If there are conflicting priorities, they need to be sorted out else the company can end up in a state of internal conflict and dysfunction. So if you’re acquiring a set of Priorities, it’s likely that you’ll have to discard your own… So, in a way, an acquisition of Priorities is almost a reverse acquisition.”

Microsoft has stated its intent to make devices central to the user’s experience. By buying Nokia and making Nokia become the new face of Windows Phone, it would definitely reinforce that priority. Farhad Manjoo on Slate Sept 2013  states it even more forcefully. He says the purchase of Nokia will kill Microsoft’s software licensing strategy, including for Windows.


Figure 2. Ownership share of mobile phone OS for Symbian (Nokia’s old mobile OS), Windows Mobile (Microsoft’s old OS), and Windows Phone (Microsoft’s new OS for which the largest licensee is Nokia). Image from

Disclosures: George Taniwaki is a graduate of University of Chicago’s Booth School of Business. He has been employed at Microsoft both as a full-time employee and as a contractor. The opinions expressed in this blog post are his own and do not reflect those of either organization.