Funny Chart
Andrew Biggs compares spending on veterinary services and healthcare spending in the US. The case he makes is that the issue with healthcare is not the rate of growth of spending but the absolute amount of spending. He presents this chart as evidence of both. But it is a totally inappropriate way to represent the data.
The first thing that the chart hits you with is that the ’slope’ of both lines is roughly the same. And the fact that Biggs’s conclusion is that the growth rate over the period is roughly the same for both might lead you to think that they are connected. But they aren’t. The lines are on different scales and so a comparison of slopes is meaningless. If you interpolate the data points, you get growth rates over the entire period of 267% for healthcare and 261% for vetcare. Close enough that the conclusions don’t change. But that’s not the point.
A chart of this kind (Y1/Y2) is the wrong choice to show similar growth rates. I could take any two time series and design the Y1, Y2 scales in such a way that they appear to be growing at the same rate. The right way of doing this would be an indexed chart such as is used to compare the performance of two stocks or a stock against an index.
No related posts.