The Myth of the Platform Company

Thoughts on GaveKal’s Our Brave New World

Does the decline of U.S. manufacturing and the growing trade deficit portend a collapse of the dollar? Or are we in a brave new world in which countries with developed financial markets produce nothing, have no physical capital, run permanent trade deficits, and pay for it all with perpetually overvalued financial assets? The latter, according to Charles Gave, Louis-Vincent Gave, and Anatole Kaletsky (GaveKal) as told in their book Our Brave New World.

The authors believe that many seemingly problematic features of the global economy can be understood as consequences of the growth of a new breed of companies. These “platform” companies come into being through the restructuring of existing companies. What starts out as an integrated manufacturing firm eventually divides into two specialized firms. The company first relocates its manufacturing operation offshore, typically to China, and then spins off the offshore manufacturing arm entirely. What’s left, ex-manufacturing, is a “platform company,” a firm whose business is to buy finished goods and then provide marketing, distribution, branding, and retailing.

The authors believe the growth of these companies explains the more or less permanent advantage that the developed world has over emerging economies. Their argument can be summarized as follows:

  1. The global economy is divided into two sectors: manufacturing (located in China) and platform company economies (the developed world)
  2. The process of reorganizing a firm into a platform company is a model that can be adopted by firms in the developed world
  3. As more firms make this transition, the developed nations are becoming platform company economies
  4. Platform companies earn a high return on investment, while manufacturing firms earn a low return on investment. This permanent returns differential accounts for a sustainable higher standard of living for those in the platform nations.

It is not clear that the trade deficit as such is a problem. While the arguments offered by the trade deficit gloom and doomers overstate the problem, if there is one, the solution offered in Our Brave New World, while provocative, is even worse. In the next section, I will elaborate the authors’ theory of superior sustainable returns, and then follow with some more detailed arguments against them.  

The Competitive Process and the Platform Model

What reason do the authors give for the sustainability of a higher rate of profit for platform companies? Platform companies earn a high return primarily because they don’t have much capital. Capital serves no particularly useful purpose and ensures that its owner will realize low returns. Getting rid of capital liberates firms to focus on high-margin portions of the value chain. The authors believe that the branding, marketing, retailing, and development of intellectual property are inherently high-return activities because they don’t use much physical capital.

This essay will attempt to criticize GaveKal’s view from a viewpoint of Austrian economics. The article does not assume an extensive background in Austrian economics, but as a preliminary, it is important to clarify the Austrian concept of economic profit. The Austrian school view is that firms on average will earn an accounting profit equal to the normal rate of interest. Austrian economics does not count this normal return as profit, though accountants would do so. Austrian economics uses the terms “profit” and “loss” (or, better, “economic profit” ) to mean any return greater or less. As Murray Rothbard explained:

“Realized profit tends to disappear because of the entrepreneurial actions it generates. The basic rate, then, is the rate of interest, which does not disappear. If we start with a dynamic economy, and if we postulate given value scales and given original factors and technical knowledge throughout, the result will be a wiping out of profits to reach an [equilibrium] with a pure interest rate.”

Austrians see profits and losses as a temporary phenomenon, indicating maladjustment between supply, demand, and price. The reason that profits or losses are temporary is that entrepreneurs are constantly attracted to profit opportunities and averse to losses. As the profits or losses are realized in a sector or industry, entrepreneurs enter or leave the sector; they buy and sell factors of production, and then increase or contract outputs, until price differentials narrow, and, ultimately, profits and losses are eliminated.

The higher return that the authors attribute to platform companies would be an economic profit. Their thesis, put in alternative terminology, is that platform companies earn a permanent economic profit by being platform companies. The main point of my criticism is that the sources of competitive advantage that the authors attribute to platform companies are not inherent in the platform idea and are not invulnerable to competition.

Splitting off the capital structure platform of a firm would expose its retailing business to competition from other retailers. Without a capital structure, the startup costs for a competitor are decreased; there is nothing preventing other similar retailers from simply buying the same manufactured inputs from the same supplier in China and then competing on branding and retailing. For that matter, there is nothing to prevent the independent manufacturing firm in China that owns the capital from expanding down the value chain and competing with the platform company in retailing.

If capital-lite service, IP, or branding resulted in economic profits, why would not entrepreneurs in the developing world start such companies as well? Japan and Korea, both of which began their industrialization process providing cheap manufactured goods, have since expanded up the value chain. There are many well-educated people in both China and India who can form service-oriented companies. Zacks Investment Research, for example, produces high-quality research reports using equity analysts in Mumbai.

Do the Math

One of the ways that the authors try to show that less capital-intensive processes yield higher returns is with a simple — and incorrect — mathematical argument. Let’s look in detail at how they do this.

The ratio of profits to the cost of investment is a measure of profitability that can be used to compare different firms. This ratio tends to be equalized by the competitive process that Rothbard describes. In other words, on the market, there is a tendency for every unit of invested funds to earn the same return.

In Our Brave New World, we read that a firm in the developed world will increase this ratio by getting rid of its capital. The argument is based on the observation that capital is in the denominator of the ratio, so having less of it reduces the denominator (Page 88). However, shrinking the denominator of a ratio increases the ratio if the numerator is held constant. If the numerator decreases by the same amount as the denominator, the ratio stays the same. The ratio could even fall if the numerator decreased by more than the denominator.

Does getting rid of capital increase or decrease the revenues of a firm? If capital had no relation to a company’s profits, having less of it would be good, and none of it best. But capital is not a blob; it consists of specific goods that each makes a unique contribution to the output of the firm. The problem with GaveKal’s argument is that it ignores the economic function of capital within a firm. Economic historian Sudha Shenoy emphasizes that capital and skilled labor are interdependent — “The capital has to be combined with the appropriate skills…There’s an awful lot of ‘tacit knowledge’ inside firms — passed on as ‘the way things are done.’” It may be impossible to separate the individual functions of a firm from its capital.

The following example illustrates this dependence between capital and output. A financial analyst for a hedge fund might, looking at his personal balance sheet over the course of a week, conclude that working at his job is a high-return activity because that is the only thing generating revenues, while eating, sleeping, and exercise are low-return activities. After consulting with GaveKal’s research firm, the poor fellow decides to spin off his body. The body is shipped to China, where he is able to outsource eating, sleeping, and exercise by hiring Chinese workers to perform those functions for him. Meanwhile, his mind has become a platform company. The mind can work 24 hours a day, analyzing the shares of platform companies.

A friend of mine works for a large integrated firm consisting of a consultancy and several manufacturing divisions. The firm was formed in a merger in which the manufacturing firm purchased what was a pure consultancy. (How could the authors’ model account for “reverse platform” mergers such as this?) In my friend’s view, his firm has a competitive advantage over pure consultant firms because it can single-source an entire solution for its customers and manage supply risk.

The book offers as an example of a capital-lite company, the former Marriott hotel chain (p.7). Marriott has spun off its hotel properties to a REIT and become a hotel servicing company. Should we expect the hotel “platform” company to offer higher returns than a hotel chain because it no longer has the burden of costly capital? Not necessarily. Why would the company that now owns the physical hotels not take competing bids from various management firms, driving down the price of hotel management services? Unless something very special distinguishes the Marriott service company from other hotel service companies, there is no permanent source of profits to be derived from this organizational structure. On the contrary, it could be a disastrous decision.

The New York Times recently covered the emergence of hypoallergenic hotel rooms. According to the article, a clean hotel room is no longer a differentiating factor: “If all you’ve got is good service and ‘Gee, my room was clean’ — well, you kind of expect that today.” The new super-clean rooms require capital expenditures of around $30,000 per room for replacing carpets with hardwood floors, improving ventilation, supplying dust mite-proof pillowcases, and other changes. The rooms also require additional hotel labor for improved cleaning procedures. Hotels have found that they can charge a premium and have higher occupancy for them.

This story illustrates how hotel management learned about the preferences of hotel customers — including what type of capital is necessary to earn profits — by operating hotels. Hotel management got the idea after receiving a lot of requests for special cleaning procedures and materials from certain sensitive guests. The contributions to the returns of management, labor, and fixed capital are interdependent and easily separated. A hotel company that owns its fixed capital is in a better position than a service company to provide super-clean rooms.

Capital is expensive, but capital-intensive firms do not necessarily earn low returns. A large structure of accumulated capital and an integrated supply chain can constitute a significant barrier to entry if the capital is costly or difficult for competitors to reproduce or operate. GMO analyst Ben Inker provides some statistical evidence that over the last six years, more capital-intensive companies have been more profitable than less capital-intensive companies.

Are brand-dependent businesses likely to earn higher returns? Branding is quite costly. Developing a brand requires a considerable investment in marketing and a superior product. Toyota, for example, has a reputation as a good value for a dependable car. But this is only after having invested billions of dollars of capital to create high-quality manufacturing processes.

Survivor Bias

GaveKal believes that the platformization trend is gaining momentum in the developed world. Management will increasingly realize that following this trend will enable a firm to match the high returns earned by those others that have already made the move. If the platform model works, should we expect that more firms will benefit by adopting it? No, for reasons that I will explain.

While the authors can point to the leading firms in various sectors of the U.S. economy that appear to fit the description of a platform company, it is more than likely that the superior returns earned by those firms do not come from this characteristic.

GaveKal is guilty of survivor bias in attributing the success of today’s leading “platform companies” to their organizational form. The total return on investment within an economy must be measured over the sector as whole, which includes the economic losses realized by all of the losses that result either within the same company or among competitors.

The authors’ error is that they take a snapshot of a world in flux and treat it as if it were fixed. The leading firms in any industry are the survivors of a competitive process in which many firms failed, losing some or all of their invested funds. Not all restructurings — platform or otherwise — will be successful, and some investors have lost money on each and every one the firms that no longer exist. As Rothbard observed:

“A grave error is made by a host of writers and economists in considering only profits in the economy. Almost no account is taken of losses. The economy should not be characterized as a ‘profit economy,’ but as a ‘profit and loss economy.’”

Consider a drug company that can manufacture a drug for $1 per dose and sell it for $50. Does this firm earn high returns on capital? That depends on how many prospective drug trials failed in order to find one that worked. Having invested large sums to create their capital (or intellectual property), some firms can be in a position where returns measured from that point forward are higher than average. That is because their capital cannot be reproduced without a similar investment, either the expense of building a large plant or the expense of doing many drug trials. There is not a low-cost way for firms to jump ahead to the point where returns are realized without first making the investment. And if there were, everyone would, and the superior return would get competed away.

The authors commit a fallacy of composition by thinking that a particular firm can increase its returns by transforming into a platform company. This fallacy arises through the confusion between relative and absolute advantages. An absolute advantage is one that benefits anyone who adopts it, regardless of what others do. Tooth brushing regularly is an example. A relative advantage is gained by improving the position of the part within the whole. For another example, everyone can cut in line, but not everyone will end up at the front of the line. What happens instead is that some people cut ahead of other people, but there is still only one person at the front. Radio personality Garrison Keillor made the same point in speaking of the town of Lake Wobegon, Minn., “where all the children are above average.”

What does this have to do with platform companies? Looking at only the survivors makes it appear that becoming a platform company is an absolute advantage, while, in fact, the successful ones have gained a relative advantage by imposing a relative disadvantage on their competitors. Economic profit, as discussed above, is a relative advantage. On average, firms earn zero economic profit; the economic profits of some firms are earned at the expense of losses made by other firms. The profits of a single firm are evidence that it has attained a relative advantage, in relation to its competitors. The more firms that attempt to earn profits by adopting the platform model, the more the competition among platform companies will eliminate those profits.


The main problem of the book is that the authors look at the result of the competitive market process as if it could be achieved through organizational structure alone. But business organization can’t be standardized in this way. Some firms will realize more value by integrating a greater portion of their supply chain within a single firm, while for others, the opposite will be true. The best organizational structure for any firm comes from a trade-off between keeping that part of the production process it does best inside the firm against the advantages of purchasing the same good at lowest cost from a supplier in a competitive external market.

Organizational models can be useful, but any model will work some of the time and not other times. Permanent economic profits cannot be realized by adopting a model. It takes superior entrepreneurial judgment to apply the right model at the right time. There cannot be such a thing as an organizational model providing permanent economic profits, because organizing a firm is an exercise of entrepreneurship. While there can be platform companies, there is no such thing as a platform model, nor a platform company economy.

Robert Blumen

August 29, 2007