Adjust As Needed

Does anyone actually read the Census Bureau’s report on economic progress, I wonder? They say, for instance, that “retail and food services sales for February, adjusted for seasonal variation and holiday and trading-day differences, but not for price changes, were $427.2 billion, an increase of 0.3 percent (±0.5%)* from the previous month, and 1.5 percent (±0.9%) above February 2016.” If your head didn't explode, you might conclude that retail sales are increasing, but are they really?

I think they forgot to adjust for the phases of the moon, and perhaps the adiabatic effect of the solar wind, so we’ll just have to live with those inaccuracies, but let’s parse that statement for actual meaning.

First of all, what does “an increase of 1.5 percent (±0.9%)” mean? Well, it means the increase may have been as little as 0.6% or as much as 2.4%. How informative is that, really? They are giving us information which has a 60% margin of error. Your average sports betting bookie can do better than that; much better.

And, when comparing between February of 2016 and February of 2017, how much adjusting for “seasonal variation and holiday and trading-day differences” do we need to do? Between February of one year and February of the next year? Really?

Finally, what, precisely, are those “price changes” for which they are not adjusting? They are, simply enough, due to something called “inflation.”

Looking at retail sales increase of 1.5 percent which is “not adjusted for price increases,” and noting that inflation in the same period was 2.1%, one might actually draw the conclusion that people are buying less and paying more money for it, and few sane people would report that as “sales are increasing” as the Census Bureau does.

How often does the government report that the budget, or taxes, or some such thing “adjusted for inflation” is stable? But inflation is almost the very definition of instability, so they are saying that “the economy, adjusted for instability of the economy, is stable.”

I’m not trying to suggest that the Census Bureau has any sort of political axe to grind with their reporting, because I don’t think that they do. But government bureaucracies do have a sort of nonpolitical bias against reporting that creates discomfort, that makes things look bad, and so they tend to shape their reporting in a manner that makes the public comfortable.

It creates an uninformed public, actually a misinformed public, which is the opposite of the purpose of the report, but...