Or so it seems.
I usually post links to IBO reports as great new information. But here is one that doesn’t do a very good job. To truly communicate the intended information, it should extend back much further in time. The report says the intent was to reduce (eliminate) subsidies as they occurred in 2002, but there is no information about subsides before 2008 in the graph. It suggests possibilities such as choosing a base year that makes everything look bad. A reliable graph would start in a decade before 2002.
This isn’t the kind of performance measurement discussed here; but it is related. I have long been suspicious of the kind of performance rating that this agency is ending.
The practice faced here is called gaming. What do you do to overcome it?
Here we have an argument that performance measurement cheating is not a reason why performance should not be measured. However, and this is important, poor quality measurement may encourage cheating or even create the opportunity for cheating. Just as the linked blog post argues that the allegation of cheating is not proved (accepting it hypothetically for the purpose of completing the argument); so too, the innocence of the measurement system is not proved (it could well be that the system is poorly designed).