Clayton Christensen responds to Jill Lepore. One passage:

One criticism Lepore makes is that some of the firms you describe as failed incumbents—whether it’s in the disk drive industry or the mechanical excavator industry or the steel industry—the companies that are ostensibly being disrupted, don’t disappear but continue to do very well, in some cases continue to dominate their industry. In 1960 there were 316 department stores in North America—department stores like Macy’s. Then the discount department stores like Korvettes and Kmart and Woolco and Target and Walmart came in, starting in 1962, and they were disruptive because for the department stores to go down-market and compete with discount prices, their profitability would have been decimated, so they had to move upmarket and get out of hard goods where margins were small and get into clothing and cosmetics where margins were higher. (My emphasis.) Now, how long has it been? Fifty-two years, Jill. Just so you understand, disruption doesn’t happen overnight. There are now six or eight traditional department stores in existence in North America. Let’s just call it less than 10. And Walmart is quite a large company. Target is quite a big company. So has disruption been at work in the retailing industry? It’s a question. Macy’s still exists. So—Jill, tell me, what’s the truth? If you could just be Jill’s answer for me.

The stuff in italics is a slightly more elaborate version of the replacement effect from a previous post. So it seems Christensen’s logic is in fact based on the replacement effect. But actually Christensen does not correctly account for the impact of competition.

Suppose an incumbent is making profits from the high-end hard good market. If there is no threat of entry, he has a weak incentive to create discount stores for these self-same goods because of the risk of cannibalizing his own high-end sales. This is the replacement effect. But facing the threat of entry, the incumbent’s logic changes. The entry destroys the incumbent’s profits from his core activity. Since these profits were holding the incumbent back and now there will be no such profits, the incumbent has good incentives to enter the low end. Knowing that the incumbent will enter the low end, the entrant may actually stay out as profits are going to be shared at the low end and price competition will dissipate them anyhow. The bottom line is that “disruption” can rationally increase the incentives of the incumbent to innovate even at the low end.

How does this fit in with the previous post? If the entrant’s product is differentiated from the incumbent’s (eg MOOCs are quite different from HBS classes) and poses little threat to the core business, the replacement effects is the key force and the incumbent innovates less than the entrant.

(All this analysis is quite generic. We have MBA homeworks exercises based on it. Tirole’s textbook has a great exposition of key intuitions and cites papers from the early 1980s. I should not be blogging about old stuff. But if old stuff is having impact on business practice and is only partially understood, why not? After all, Krugman goes over IS-LM repeatedly!)