He Who Begins to Count Begins to Err

“If you spend more than 13 minutes analyzing economic and market forecasts,” the famous investor, Peter Lynch, once remarked, “you’ve wasted 10 minutes.” Oskar Morgenstern (1902-1977), a professional economist, probably would have agreed with Lynch.

I found myself thinking about Morgenstern over breakfast recently as I was skimming financial headlines in the newspaper. He was a “numbers guy” who understood the limitations of numbers.

Today’s papers carry lots of angst over rising prices, commonly called inflation. “Global Price Fears Mount,” reads one Wall Street Journal headline. “New Push at Fed to Set an Official Inflation Goal,” reads another.

Within these articles are a number of “facts.” I want you to go through these and see if you can tell what the unifying fallacy is:

“Annual inflation in China is almost 5%…”

“Last month [in Europe], inflation unexpectedly jumped to 2.2% in the eurozone from 1.9%, the first time in more than two years it has exceeded the ECB’s target of just below 2%. Some economists say it will rise about 2.5% in the next two months”

“Inflation [excluding food and energy] was still weak at 1.1% in December”

“UK inflation is approaching 4%…”

“The Fed informally has said its goal is inflation of around 2%”

“Forecasts cluster around 1.75% and 2%.”

There are more, but this sampling is large enough to make my point. The underlying fallacy here essentially boils down to the idea that government officials can calculate a true inflation rate to within one-tenth of one percent. And that’s not all! Government officials can also dial inflation up or down to within tolerances of one-tenth of one percent.

This is where Oskar Morgenstern comes in. Morgenstern was a German-born economist, educated in Vienna. When Adolf Hitler took over Austria, Morgenstern happened to be in the US. He decided to stay. Good move. Morgenstern is most famous today for his work with John von Neumann in founding game theory.

I was thinking of him because of a famous essay he wrote entitled, “Qui Numerare Incipit Errare Incipit” (“He Who Begins to Count Begins to Err”). He is not the only one to take economists to task for their abuse of statistics, but his thoughtful analysis stands out in my mind.

Essentially, Morgenstern’s point is that we need to be more aware of the errors in such numbers. He criticized the way in which people report and use these numbers. They purport an accuracy that does not exist.

Here is an extended quote from his essay, which gives us a small list of sins:

“Changes in consumers’ total spending power down to the last billion or less (i.e., variation of less than one-half of one percent) are reported and taken seriously. Price indexes for wholesale and retail prices are shown to second decimals, even though there have been so many computing steps that the rounding-off errors alone may preclude such a degree of precision. Unemployment figures of several millions are given down to the last 1,000s (i.e., one-hundredths of one percent ‘accuracy’), when certainly 100,000s, or in some cases perhaps the millions, are in doubt.”

Yet, despite the huge error in such numbers, people treat them very seriously. Wage increases, in some cases, are based on changes in price indexes. People plan and make big decisions based on such poor data.

“Economics is not nearly so much of a science,” Morgenstern writes, “as the free use of allegedly accurate figures would seem to indicate.” He suggested such figures report estimated error rates. For example, inflation might be 2%, plus or minus 2%.

Doing so would make the following statement look entirely hollow and devoid of meaning, as it essentially is: “Inflation unexpectedly jumped to 2.2% in the eurozone from 1.9%.” And you would laugh at a statement such as “Forecasts cluster around 1.75% and 2%.”

As an investor, you ought to look at the whole constellation of economic numbers with doubt and skepticism. They are not accurate. They can never be accurate.

Far worse is the idea that we should target a certain inflation rate. I am simply beside myself that anyone can believe that an “inflation target” is anything other than ridiculous. We can’t even measure it, but the Fed thinks it’s going to control it! And not only control it, but thread it with precision to 2%!

As investors, I think it is a great mistake to sit around thinking about things like unemployment, GDP, price indexes and the whole lot of garbage that gets reported and commented on by nearly everyone. The whole thing is a fraud. I think you could be a very successful investor and never parse a consumer price index in your life. In fact, they may do you harm by making you afraid to invest when the economic picture seems dim… or reluctant to sell because it is so bright.

Official inflation in the US may be reported as 2%, but it may well be four times as high. Directionally, it’s enough to know that prices are rising. You don’t need the consumer price index to tell you that the dollars in your pocket are buying less, just like you don’t need a thermometer to tell you that it’s freezing cold.

When inflation really gets going, the economists will be the last to know.


Chris Mayer
for The Daily Reckoning

The Daily Reckoning