US corrosion cost study and Congress funding

The 1976 → 1982 → 1991 → ~2014 series of cost-of-corrosion studies (NIST and NACE). Tom uses the escalating numbers to teach skepticism about aggregate-statistic methodology ("lies, damn lies, and statistics") while granting the underlying point that corrosion is genuinely expensive.

I just happened to get this book — it arrived after class yesterday — on cathodic protection. I'm sure you all are busy purchasing books like this. It says that the first study on the cost of corrosion was done in 1976, and they found it cost the U.S. seventy billion dollars in 1976, somewhere between three and a half and four and a half percent of the country's gross national product. They redid the study in 1982 and it was a hundred twenty-six billion. By 1991 it was two hundred seventy-six billion. If you calculate that, it's higher than the inflation rate. What it is — lies, damn lies, and statistics — they were including other things. For example, they assume that every automobile that's junked is junked because of corrosion. That's not necessarily true. A lot of them are junked because they're totaled in a crash. That's not corrosion.