In 1981, IBM introduced its personal computer. The computer industry immediately blew up, killing all Route 128 minicomputer manufacturers such as Digital Equipment Corporation and Prime Computer, Inc. Silicon Valley, Austin, and Seattle are now the new centers for computing technology, filled with dozens of new startups trying to cash in on the PC explosion. Among those startups was Compaq, a company started by Texas Instruments engineer Rod Canion alongside venture capitalist Ben Rosen. Canion’s plan was to produce a computer with 100% compatibility with the IBM standard, but with higher quality components, superior technology, and premium prices.
To achieve higher quality than the rest of the pack, Compaq used vertical integration. Instead of buying off-the-shelf components like everyone else, they designed many of their components in-house. Power supplies, electronic controllers, and keyboards made by Compaq were superior to the competition. Compaq became popular among corporate buyers, who cared more for reliability and quality than the lowest price tag, and the company made decent profits for a little under a decade.
However in 1991, Compaq’s sales started to decline. The sales force reported that corporate customers no longer wanted to pay for the premium prices of Compaq because the quality differential of the competition had diminished. Rosen and a group of others went to Comdex, where they found that many of the components made in-house by Compaq were now produced by specialists. Independent manufacturers were now making the same power supplies, controller chips, and keyboards as Compaq at the same or even better quality for a much lower price. Compaq’s vertical integration is now a serious burden, as buying the parts from outsiders has become much cheaper than building them in-house.
What happened here? Many people believe that firms who vertically integrate are guaranteed to lower their costs, gain finer control over their end product, and ultimately raise profits. While there are a number of examples where vertical integration has helped firms, here is a clear example where vertical integration became a hinderance to Compaq as the computer industry matured. When the industry became large, specialists emerged and could take advantage of economies of scale. Because they sold their components to all firms, they could spread R&D costs over many more units sold. Compaq’s components are used only by Compaq, and therefore they cannot spread their fixed costs to a as much as the competition. Per component, they would always have higher costs than the specialists.
Ford Motor Company came across the same issue around 60 years earlier. In their relentless pursuit of vertical integration, Ford’s River Rouge plant had all the necessary components to make its own steel straight from iron ore. It had blast furnaces to smelt the iron ore, open hearth mills to turn the crude iron into steel, and steel rolling mill to create sheet steel, which would finally be ready for vehicle production. The River Rouge plant also made its own glass and supplied its own power. However, eventually Ford fell victim to the same problem as Compaq, as the automobile industry matured and external suppliers could produce better components for less due to economies of scale.
Intel has forward vertical integration, doing the lithography for their chips in-house. Intel has been enjoying high profits for a long time now, being the dominant chip maker for the past 20 years or so. However, with the advent of smartphones, the market for processors has grown, Taiwan Semiconductor Manufacturing Company’s (TSMC) lithography outperforms Intel’s. Now suddenly Intel is in deep trouble, as their chips are performing far worse than rival Advanced Micro Devices (AMD), who is outsourcing their lithography to TSMC. Apple is also dancing around Intel with their ARM-based mobile processors, which are much faster and run cooler than even even Intel’s notebook chips. Just recently Apple announced that they are moving the Mac off of Intel, as using in-house chips has proved to be both faster and cheaper. Like AMD, Apple also uses TSMC to produce their chips. Intel’s vertical integration, which has given them tight control of their chip quality and yields over all these years, is now causing them to lose valuable marketshare. This allows their rival’s economies of scale to grow relative to Intel’s, making each step easier than the last to beat Intel at their own game.
Vertical integration is no magic pill for a firm become profitable. Integration will only be profitable if you can make your components in-house better or cheaper than purchasing them from outsiders. Otherwise, you are wasting valuable management time and R&D resources over reinventing the wheel.
Sources: Competition Demystified by Bruce Greenwald and Judd Khan, National Register of Historic Places documentation on the Ford River Rouge Complex
PS: In my high school classes always touted that vertical integration is a magic pill to success, and companies that owned their suppliers were very profitable. It was also implied that these companies exhibited monopolistic tendencies. I am not sure how popular these ideas are in today’s world, but this article is directed to anyone who has the same idealism over vertical integration as my teachers did in high school. Producing things in-house doesn’t guarantee higher profits, it only can be utilized if outside specialists aren’t good enough for the job. Since vertical integration is just a manner of efficiency, it really has nothing to do with monopoly profit.