Walter J. “John” Williams is the founder and CEO of Shadow Government Statistics, a web site and consulting practice specializing in government economic reporting.  His clients range from individuals to Fortune 500 corporations.  Williams publishes an electronic newsletter that exposes and analyzes the flaws in current U.S. government economic data and reporting, as well as in certain private-sector numbers, and provides an assessment of underlying economic conditions, net of financial-market hype.

We spoke with Mr. Williams on May 2, 2008.

We’d like to begin with the GDP numbers.  The government’s initial GDP report came out, showing 0.6% annualized growth in Q1 of 2008, or 2.5% year-to-year, both the same as in Q4 of 2008.  Your analysis shows a very different number – -2.7% year-to-year.   Can you explain the major adjustments that you make to arrive at this number? 

Over time we have developed a database and an analytical practice specializing in measuring the quality of government statistics, including how they are reported and manipulated.  To get meaningful forecasts for our corporate clients, we need to know very precisely what to believe and what not to believe in the government’s reported numbers.

There are two types of manipulation For a related discussion, see the article by John Mauldin in Advisor Market Commentaries.we encounter. The first are very long term methodological shifts.  These changes have not had much of an economic basis, and their effects over time have been fairly consistent and very deliberate in building biases into the reporting. As a result of these methodological shifts, GDP and employment are overstated (showing stronger growth) whereas, with reported inflation, the bias is on the downside (showing lower inflation numbers).  As a result, the CPI is significantly understated as compared to how people actually experience inflation.

You can define these numbers any way you like.  But the government’s numbers, over time, have increasingly diverged from common experience.  Most people don’t believe what the government reports.  Main Street is completely right in this respect.

The GDP for Q1 was reported at 0.6%.  This is utter nonsense.

The second type of manipulation does not happen all the time, but it has been happening increasingly often recently, and it appears that it just happened with the reported GDP data.  In this type of circumstance, short-term data are manipulated for political and/or financial market reasons, to show stability in the economy.

Those components of the Q1 GDP which are based on hard data were down.  Retail consumption in Q1, as reported after adjusting for inflation, contracted 4.2% on a quarter-over-quarter basis.  The corresponding number was 1.3% in Q4 of 2007.  Housing sales declined by more than 30% on an annualized basis.  Consumer spending, including housing, is 75% of the GDP.  Industrial production and new orders for durable goods were both down on a quarter-over-quarter basis.  Trade data, after adjusting for inflation, deteriorated, but somehow they improved in the GDP reporting.

The combined effect of this hard data should have been a 1Q GDP contraction.

The rest of the reported data are estimates, and not based on hard numbers.  These include contributions from the service side of the economy, government spending and an inventory buildup.  These guesstimated data showed improvement, enough to produce a reported GDP expansion.

I believe the estimated reported data were manipulated to overcome the poor results from the hard data, in order to report the GDP gain.

We place very little value on the advance report of GDP data, since it goes though two revisions before the final number is reported.  The margin of error on advance GDP data is +/- 3% (as compared to the latest number after years of revisions).  Therefore, the advance GDP of 0.6% could end up eventually anywhere between -2.4% and +3.6%. 

We also do not try to estimate quarter-over-quarter GDP numbers.  Instead, we look at year-to-year changes, which are much more stable.  Quarterly numbers are made to look good for the public.

Based on your analysis, when did the current recession begin, and what is your forecast for when it will end?

Our numbers show GDP is -2.7% on a year-to-year basis, and that we have been in a recession since late 2006.  We are in a structural recession, one which is particularly deep and protracted.  At least in the last decade -- really since the 1990 recession -- official estimates of recession timing have been late in terms of the starts of recessions and early in terms of their ends.

On a more realistic basis, in terms of what businesses are feeling, we are entering the second dip of a double dip recession that began back in late 1999.

Are there characteristics of the current recession that make it different and unique, as compared to previous recessions?

The economy is in a severe inflationary recession, and the rate of inflation is accelerating.  Our economy is also undergoing a structural change.  The inflation is commodity driven, and reflects problems from a weak dollar and a rapidly growing money supply.  There is very little the Fed can do to tame inflation or to stimulate the economy.  It is not uncommon to have inflationary recessions.  The Fed’s primary goal is to keep our banking system stable and, in this respect, they have been successful so far.  In 2002, Bernanke delivered a speech outlining how he could handle these circumstances, and he has followed this path.

The Fed cannot do much about inflation or the economy.  If inflation is driven by a strong economy and consumer demand, it can be tamed by raising interest rates.  But there is no obvious relief for today’s commodity-driven inflation, at least from the perspective of the Fed.

The big problem in the economy facing government today is it cannot stimulate income growth through lower interest rates or otherwise.  Lower rates will just encourage people to borrow more; it will do nothing to address the structural problems inhibiting consumer income.

I fault Greenspan for stimulating growth with financial bubbles, when we would not have had growth otherwise.  His policies prevented the self-correcting actions the regular business cycle usually imparts, and he actually exacerbated long term consumer liquidity problems.

Since the 1970s, much of our manufacturing base has moved offshore, taking with it high paying jobs.  Average real weekly earnings are lower now by 20%, using the government’s numbers.  The effect is the average household has not been able and still is not able to keep up with the cost of living.  This is despite a shift in family dynamics during the last several decades where families with one bread winner typically have become families with two more people having to work.

Display article as PDF for printing.

Would you like to send this article to a friend?