03/18/2010 05:12 am ET Updated May 25, 2011

Lies, Statistics and Economic Statistics

Weathering a blizzard of statistics is the fate of the public minded. Numbers come at us from all points of the compass. Some are raw data, some massaged, some naked and some fitted out for the occasion by their sponsors. In this wintry economic season, they all come with a message and meaning. Making sense of the figures demands a large measure of skepticism and an eye for misrepresentation and forgery.

Here is a quick everyman's guide to economic statistics. First, the "recession." No term has been more abusive of statistics. There is a narrow technical meaning for economists. It is strictly a matter of GDP numbers having nothing to do with our feelings of being "in the money or out of the money." If the economy registers a quarter in which GDP rises - by however small an amount, we are no longer in recession. That's it. That happened in the third quarter of this year when it rose at a 3.4 annualized rate.

If GDP drops 4 percent in the fourth quarter and then rises just 0.4 percent in the first quarter of 2010, the business headlines could rightly proclaim that the "recession" is still over. The public understanding of the term "end of recession" is that we have hard data of a return to prosperity - or, at least, that a growth cycle is imminent. Nothing of the sort is in the numbers per se. The discrepancy between these two meanings are cynically played upon by all those who have a vested interest - political, intellectual, or financial - in creating the impression of an economic morning in America. That covers just about everyone marketing the news.

Then there are GDP numbers. Politicians and markets alike rise and fall on these magic numerals. In fact, they are as soft as statistics get. At times they are fanciful. Consider the financial sector. The "virtual" wealth generated by the razzle-dazzle of recent years may have created nothing of tangible economic value - whether goods or services. Indeed, the
largest part of the value created by the trading of exotic financial instruments is fictitious. President Obama admitted as much in remarking on one occasion that we shouldn't worry that much about the evaporation of trillions in investment accounts since it wasn't real money anyway. Economists routinely talk blithely of the "financial" economy and the "real" economy. Yet GDP - and all other aggregate statistics - make no distinction whatsoever. Nor do economic models.

GDP figures are no more than the sum of all expenditures. Every time a piece of financial paper (actually, electronic dots) with little intrinsic value is transferred from one party to another the national cash register records it as a number in the tally, and does so at the face value of the transaction. This is an absurd methodology based on an absurd measure of value. We all may have been living in a world of statistical make believe. The latest numbers, for example, say that GDP grew at a 3 percent annualized rate. But if that reflects the big Wall Street banks reverting to go-go of virtual assets, then the real number is substantially lower.

Rates of economic growth are further overstated by discounting population increase. If the number of Americans increases by 3 percent and GDP numbers grow by 3 percent, we are no better off in real terms. This relationship is prominent in calculating economic dynamics in poor countries but for some inexplicable reason largely ignored in the U.S.

The implications of all this slips through our mental fingers. Yet the implications are profound for calculating national wealth, the United States' place in international league standings, productivity and even inflation. Inflation as represented in the government's cost of living index is another statistical fiction.

For one thing, the formula was rigged by Bill Clinton to produce lower numbers in order to keep down increases in Social Security payments that are tied to annual inflation. The method was crude: When the price of one item in the index rises sharply, a cheaper item that supposedly is a functional equivalent is substituted. More generally, there are a plethora of distortions in data gathering that bias the index toward the low side. One small example, when all of my personal medical expenditures rose due to changes in my employer provided plan (as happened this year to millions), they never registered in the official inflation numbers.

Productivity statistics are also manipulable. Every time a company "downsizes" - that is to say, fires workers - its productivity figures go up if the same quantity of goods/services is produced. Heavier work burdens for remaining employees do not count. Unpaid overtime does not count. Nor do the costs borne by customers who must wait longer for help in a box store, or on lines at an airline ticket counter, or go through the ordeal of dealing with mechanical telephone programs intended to be as painful as possible so as to force you to spend your time on the Web. Corporate and government statistics register none of this. Economists' models do not either.

The common thread running through this recitation of how economic statistics are abused is that it is the little person - as employee, as customer, as retiree - who gets the short end. Surprise, surprise.