Virulent Word of Mouse

April 7, 2004

The paradox of commodities

Filed under: Capturing and creating value,Essays — Shane Greenstein @ 9:52 pm

A commodity is available from many vendors without distinction. Bananas are commodities. So are potatoes, salt, and some types of potato chips.

Can a microchip be a commodity? How about a consumer electronics product, such as a cellular handset? These are not merely academic questions, but standard topics of conversation in executive board rooms, Silicon Valley coffee shops, and Wall Street stock analysts’ conference calls.

What is not a commodity

Simply stated, a commodity is not a high-margin product. For example, Microsoft prices Windows at more than 20 times its unit cost. In contrast, Michael Dell feels lucky when a PC’s unit margin reaches 5 or 10 percent above cost. Get the idea?

Pricing at multiples above unit cost requires something special–valued brand, frontier features, unique service, or lightening-fast distribution. Buyers must be willing to resist the temptation to obtain their product from another supplier that offers it at a lower price but without something special.

In the beginning, computers were not commodities. During its heyday, IBM had the most storied distribution and servicing network in the world. IBM’s engineering was pretty good, too. The combination of advanced technology and tailored service was extremely potent. IBM dominated the market. It was able to charge prices at multiple levels above a machine’s unit cost.

Computing has come a long way since then. Today, very few firms can garner high margins on their products. This fact fosters a myth that all high-tech product markets eventually evolve into commodities. There is an element of truth to this myth, but also plenty of confusion.

Misunderstanding the role of obsolescence contributes to the myth. Rather trivially, every product eventually loses pricing power to obsolescence. There are plenty of examples: mainframes, minicomputers, VCRs, and a host of buggy- whip technologies from the past.

Such examples are misleading because they overlook the fortunes of particular firms. For example, IBM’s mainframes staved off commoditization for decades and so did DEC’s minicomputers, even when the market was shrinking. As another example, consider Computer Associates. For many years this company earned tons of revenue from specializing in software markets with declining demand.

Another part of the myth arises from a careless misreading of the history of the personal computer industry. Stated bluntly, in 1983 IBM priced its hardware at two- times its cost; not anymore.

There are three errors in the way people usually tell this story. First, many conveniently forget the less-glorious era that preceded IBM’s entry into the market. From 1975 until 1982, the only firm that made serious money was Apple–on the Apple II. During the early years, virtually all successes were small, and eventually everybody sucked wind.

The second error stems from failing to recognize that IBM’s experience was the aberration, not the rule. From the outset, IBM used very common parts for its PCs to make them competitive. They also embedded a proprietary ROM BIOS. The company expected to protect this one part from imitators and maintain its margins with its distribution. Neither BIOS nor distribution was unique for long. Compaq reverse engineered the BIOS, and retailers figured out how to stock the product.

In other words, margins were high for a brief time because a particular player in that market had some unique and special assets. These were useful and profitable assets, but could only remain uncommon for a short period.

More generally, high margins do not beget low margins in a predetermined sequence. Rather, low margins are the usual state of the world and occasionally a firm manages to break away. It is rare, but possible. To be sure, it is also eye- catching, but whether it happens early in the history of the product category or later is just a side detail.

This last general observation introduces the third subtle error in using PCs as an example: It simply overlooks the variance across firms in other electronics markets. That is, some firms manage to stave off commodity status year after year. They figure out something that others do not. This last point is a big one and will take some explaining.

Fighting against commodity status

There are four positive and three negative strategies firms use to fight commodification. Let’s start with the positive.

Positive strategies

One positive strategy focuses on a niche with loyal buyers. Many software game firms have tried this approach. The BlackBerry handheld is the latest consumer device to also use this approach. The Palm handheld started this way. The list goes on.

To be sure, this strategy relies on a cool idea, a novel approach, or a fad among fanatics. But this does not guarantee high margins in perpetuity. Companies can keep margins high only with careful updating of the core concept and intelligent marketing to an enthusiastic audience with predictable tastes.

A related and second approach in fighting commodification is to expand beyond a niche into a broad product line. It involves manipulating product replacement cycles for maximal effect.

There are several firms with a history of doing this well. Sony is one of them. Sony has spent decades developing its reputation for making frontier products that do not fall apart after consumers open the box. They extract a premium for that reputation. Visit any electronics retailer and compare Sony’s prices with everyone else’s. Sony’s are higher. Yet many customers will think about it for a minute, then pay the premium and walk out with a Sony.

Another honorable mention goes to Hewlett-Packard with its laser printer product line. HP introduces a new broad line of printers about every six months, charges little for them, and then soaks consumers later with the cartridge replacement price. Everybody knows the game plan and yet HP has stayed atop sales charts for more than 15 years; over half of the market goes to HP. That is an impressive run.

There is a third approach for overcoming commodification. Instead of beating it, the firm revels in it. Just as McDonald’s makes tons of money selling French fries, so too can an electronics firm from selling tons of low-margin products. After all, a penny times a very big volume results in a very big number.

This strategy requires the lowest cost and highest volume. It requires extraordinary execution. Also, it is not for the fainthearted. Unexpected downturns in demand can be devastating.

Best Buy and Circuit City operate this way. So did Egghead and many other retailers who are now long gone. That said, for sheer guts, the top dog is Dell over the last decade. Dell doesn’t focus much on margin, but it sure sells a lot of product.

A fourth positive approach combines all the preceding three, but with the additional twist that the managers think like marathon runners. The managers continually discuss how to remain strategically consistent, that is, how to stick to the game plan, accumulate learning, and wait for others to misstep. It requires a special sort of organizational discipline.

Nokia has thrived through this strategy. For years its products were not the best nor the worst. It managed to ship a new release about the same time as its biggest rivals, Motorola and Erikson, among others. Then eventually those rivals tripped, erred in some way, and angered their customers. It left Nokia reaping the rewards for not screwing up.

Negative strategies

There are also three negative strategies for delaying commodification. All are controversial.

One strategy is to commodify everyone else in a value chain so that it leaves only one brand with something special. Intel and Microsoft have tried this from time to time in the PC value chain and have met with varying degrees of success. That said, the importance of this strategy is exaggerated.

First of all, the key PC parts are common, so most commodification would have occurred whether or not Intel and Microsoft helped it along. Second, Microsoft and Intel are both pretty good at the four positive strategies I’ve outlined and would not be where they are without them. Third, many seemingly negative strategies are reactive, not proactive. For example, Intel saw the demand for Wi-Fi explode and reacted strategically by inventing Centrino. Sure, this branding commodified a number of Wi-Fi card producers as a result. But I find it hard to criticize Intel for that. Intel would fire CEO Craig Barrett if he did not try to market something like Centrino, so is he really to blame for doing his job?

A second negative approach is to try to kick out any potential rivals by fair or foul means. After all, nothing staves off commoditification like un threatened monopoly status. On this score, many firms deserve awards. Yet, nobody holds a candle to Bill Gates for this behavior, so I focus on him.

Perhaps Gate’s most imaginative strategy was the per-processor licensing contract with assemblers at the end of the DOS era. This meant that assemblers paid for a license whether or not they included it in the product that shipped out the door.

Assemblers certainly did not like paying for something they did not ask for. Gates just had his lawyers insert it into the contract. He told the assemblers to live with it or go out of business. They relented and kept quiet in public, lest they anger their important supplier.

Here’s the rub: The folks at DR-DOS thought that it closed the market to them, which it did. DR-DOS also thought it had a chance to do well. Alas, you and I never found out who was right, because Microsoft’s contracts closed the distribution channel to everyone but itself.

By the way, that strategy is now a standard example in MBA courses. You know something is clever if it makes the MBA canon.

The third negative strategy is to combine the two preceding ones, commodify others, and solidify monopoly at the same time. Again, nobody does this better than Gates.

The browser wars brought many such actions to the public’s and court’s attention. Of all the stuff Gates authorized, the most audacious were the first- screen restrictions on all contracts with assemblers. This forbade firms such as HP from putting help screens on their systems, which HP wanted to do to compete with Compaq. At the heart of it, Microsoft did not want business partners to profit from peddling the Internet, and they did not want HP to help their users navigate the Internet with a Netscape browser.

Again, nobody wanted these restrictions. Gates just put it in the contract and made sure everyone knew he was watching for compliance.

Did it matter? Arguments about the importance of this clause can become heated in a Silicon Valley coffee shop. My own view is that without this illegal stuff, the browser wars would have been much more expensive for Redmond and more protracted than the fight everyone witnessed. It might have become a marathon. But, alas, you and I will never find out who would have won in that situation.

Even the most conservative appeals court in the country singled out this clause for removal. Once again, you know something is clever when seven appellate court judges agree it has to go.

The paradox of commodities

It is simply wrong to argue that firms cannot make money in a commodity technology market.

Yet, that leads to the central paradox of commodity markets: Staving off commodification requires investing in something special. But if that were easy to do, then everybody would make such investments, which would make the “something special” common. If they were common, then they no longer would be useful for staving off commodification.

Published PDF version of this entry

Leave a Comment »

No comments yet.

RSS feed for comments on this post.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

Blog at

%d bloggers like this: