Q. is 6-sigma good for economic growth or bad? In case you aren't aware, 6-sigma refers to a quality control measure. If you buy a jet engine from General Electric (GE) you'll be comforted to know there is a very high probability the engine conforms exactly to their design intention. Many companies jumped on the 6-sigma bandwagon so they wouldn't be left behind, and so they would qualify as vendor to a 6-sigma firm, in what was another management 'fad'. These fads have shown up as upward bumps of a few percentage points on stock markets, showing they are generally good for business.
At least in the short run, and for certain types of products and production methods. The problem I see is one of bureaucracy. To get the high conformance, workers must do their production exactly as specified. That means more things need to be specified. The equivalent of 'thick binders' of exacting specifications get written. To understand the binders, new workers must go through training sessions, likely with multimedia course ware, and testing at each stage. So by the time a production process has been 6-sigma-ized a bureaucracy has gelled around a very specific process.
And that's bad for innovation in the long run. If a worker comes forward with a suggestion for incrementally improving a process, they are more likely turned down or put on a future list. That's because it's costly to rework the training materials, retrain all the workers, besides the normal changes to production process and testing methods. So in the long run, 6-sigma would tend to slow incremental innovation, and slow economic growth.
But 6-sigma might help the speed of innovation in this way:: if you want to invent some new use for jet engines, it's nice to get an engine that won't have problems, so you can focus on your new innovation.The firms at the end of 6-sigma supply chains don't have to be 6-sigma. They can innovate rapidly in their own domain. But if they need a part design changed, and it's a few steps back up the 6-sigma supply chain, it's going to cost them.
On the other hand, some types of products are easier to quality control. Software enjoys automated testing. A developer can experiment offline with a new algorithm, put it into a copy of the production software. Then pass it to a testing guru to develop automated tests. After they go back and forth with quality touch-ups, the new code can be merged semi-automatically, including the tests, into the final production version. After that, it's easy to do a straight copy of something digital, with no workers fouling up the production process.
To the extent some of that software-like process can be adopted in physical product production processes, the easier it will be to do incremental innovations.
But even in software, it's common to do 'releases' - that is, although the product and process is changing with incremental innovations, that's in the back room, and the end users keep getting the old version until the new one is released. Firms can do the following: always have 2 or 3 assembly lines. One is your current production, one is for your next release in so called beta, and one is for new ideas just getting started. When ready for a release, you change over to the new assembly line. That allows firms to capture and try out incremental innovations. Then they can have it both ways: 6-sigma and incremental innovation.
