It is a sad but undeniable truth that some of the world’s most profitable products are terrible. That lightbulb realisation dawned on me when I worked on the FT’s Lex column and learnt that the most successful pharmaceutical drugs — for manufacturers if not patients — were those that alleviated symptoms but did not cure the complaint. Eliminate the problem and you kill demand. Where is the financial incentive in that?

Lightbulbs, curiously enough, are another example of the same phenomenon. Why develop everlasting lightbulbs (the Centennial Bulb has been in continuous operation in a Californian firehouse since 1901) when you can sell ones that blow periodically? Economic theory suggests that these inefficiencies should be competed away. Real life does not always work that way.

Take cyber security, perhaps the ultimate industry for providing highly lucrative, but ineffective, solutions. Companies and governments are spending an escalating fortune on IT security but cyber crime is only growing worse. According to Gartner, global spending on information security will increase by some 13 per cent to $184bn this year creating a bonanza for many cyber security companies. Even so, the “speed and ferocity” of cyber attacks continues to accelerate, according to the latest global threat report from CrowdStrike. The security company itself botched a software update in July that temporarily shut down 8.5mn Windows devices.