The importance of market testing

If anyone had tested the market for plastic bottles with non-detachable tops, as now required by EU countries and the UK, they would rapidly have discovered that it was not going to work. The theory was that people would be encouraged to recycle the top as well as the bottle if it remained attached.

Everyone hates it. The top gets in the way if you try to drink from the bottle, and can cause cuts to the lips or fingers. Everyone I know, without exception, rips off the top. Some, like myself, cut away the protruding strip of plastic to prevent damaging fingers when we reseal. Most recycle the top along with the bottle, putting it in the same bin. The tactic has not worked. It causes resentment, irritation, and is ineffective.

Oxo probably did market test the revised shape of the famous cube. It is no longer a cube in the strict sense of the word. The new Oxo cube is shaped like an "X", with notches carved into the sides to make it easier to crumble. It is a great success because it does what it is intended to do. It can be crumbled to a powder in a fraction of a second and have hot water added. In this case, unlike the bottle tops, it makes life better for the consumer, and is appreciated by them, judging by the comments

The Adam Smith Institute has long advocated the market testing of policies to ascertain how successful they might be. We have urged them to be tried out in limited locations to assess how they perform in practice. Most new policies have opponents whose vested interests are tied in with the current way of doing things. There is less resistance to limited trials than there would be to a universal simultaneous rollout. Furthermore, theoretical objections by opponents can be countered by showing that they do not arise in practice. And on occasions where the testing reveals design flaws, the policy can be revised or abandoned.

The ASI originally backed Enterprise Zones and Freeports because it was felt that they could be market testing lighter regulations and taxes within them that could later be rolled out more widely once their success had been demonstrated. It is harder to argue against giving things a limited trial than it is to object to their universal application.

There are boundary problems with limited area tests, in that people might move into the trial areas and invalidate the trial. For example, if a guaranteed income were tried in one town, people might flock there to benefit from it, building in a biased population. These are problems that can usually be resolved by taking care when setting up the trial.

Market testing goes with caution. ‘Let’s see if it works’ is a sensible attitude that is better than plunging directly into the unknown.