It goes “When a measure becomes a target, it ceases to be a good measure.” How does that work in practice?
In: Other
I can give you a good example: Food labeling, specifically protein.
It’s actually pretty difficult to measure components in food. The standard method for measuring protein is to basically blast a sample into atoms and count how much nitrogen is there, because most of the nitrogen present in any living being is in the protein.
Like I said, it’s the standard method. If you’re selling something that’s supposed to have a certain amount of protein in it, you want to hit that target. The problem is that the target is no longer really the amount of protein. It’s the amount of nitrogen. There are much cheaper, nitrogen-rich molecules you could add to the food that will “pump the numbers” and show up an protein in the tests. That’s why Chinese manufacturers started adding melamine to pet food and baby formula.
It was kind of an open secret for years that this was going on. If that sounds bad to you, it gets worse. Melamine is relatively harmless, so at least they weren’t hurting anyone (apart from the risk of protein deficiency). Then some manufacturers realized they could save even more money if they changed from the industrial melamine to using the batches that had been rejected for industrial use due to contamination. It was the contaminants that started killing cats, dogs and babies.
You need to regulate and control things, but you have to be careful about the metrics you use as the target, because if there’s a way that someone could game the system, maliciously comply, or overdue compliance on that one metric to an extreme degree, they probably will.
Latest Answers