Dept. of Misleading Stats

One of my pet peeves is misleading statistics. It’s one thing when facts are simply wrong, but it’s another when the numbers are accurate but they’re presented in such a way as to tell a different story.

This morning, Brett Arends, one of my favorite columnists at the WSJ tweeted this out:

Is that right that inflation has eaten up 90% of stock gains? Yes, but no. The chart is accurate, or at least it looks similar to my data. The problem is the 90%. The reason why is that it uses a linear measure against compound growth. I’m throwing out my red flag. When you mix those two measures, the 90% figure is meaningless. All it really tells us is that inflation averaged about 2.8% per year (growth of 2.78% for 84 years equals tenfold).

Put it this way: If you have any data series with positive inflation, after a long enough period, inflation will reach “90%.” The nominal figure will by definition grow faster, and those gains will compound themselves.

If the stock market returned 10% per year in nominal terms and real returns averaged 9.9%, after 2,300 years, by the methodology described, inflation will have taken 90% of the gains. The numbers are correct, but interpretation is misleading.

Here’s another example of misleading stats. Please bear in mind that I’m not saying the numbers are incorrect, just how we read them.

You’ll often hear that from the March 2000 peak to the March 2009 low, the stock market lost 46% (the Wilshire 5000). That’s correct. But if you take a starting point 18 months before the peak, and an ending point one year after the low, then the market made 78%. It’s true that I’m being selective with my data, but what’s so special about measuring from the peak to trough?

Posted by on April 30th, 2014 at 10:04 am

The information in this blog post represents my own opinions and does not contain a recommendation for any particular security or investment. I or my affiliates may hold positions or other interests in securities mentioned in the Blog, please see my Disclaimer page for my full disclaimer.