Warren Buffet’s Textile Woes

There's a story related by legendary investors Warren Buffett and Charlie Munger. At a time when they owned a textile company, an inventor approached them with a loom that could do twice as much work as the old one. Buffett replied that he hoped it didn't work. Because if it did, he'd have to close the mill.

Sounds strange, but it's an important lesson in microeconomics. The problem with Buffett's textile business was that it was a commodity business. Buyers of textiles really don't care from which mill they get it. If your textile mill buys machinery that cuts your costs in half, every competitor will do the same, and all the alleged savings will get competed away and passed on to the customers. All that heavy capital investment for zero improvement on the margins. Great for the customers, terrible for the business owners.

If you are in a structurally bad business, i.e., one where you're selling an undifferentiated product into a price-competitive market with low barriers to entry, innovation is bad for you. You have to buy it just to stay in the game, but you can never use it to get ahead.

Given that innovation is all around us, what does that mean for your business? It means all the promised efficiency gains will evaporate and get competed away if they are not serving a differentiated business, where you sell something people want from you, and there are at least some barriers to entry. Sound strategy is becoming more important than ever: What is the unique value your company brings into the world? If you have an answer to that, you can profitably bring all sorts of innovation (including AI) to bear.

Innovation without differentiation is just a more expensive way to break even.