Alert
JOHN WILLIAMS' SHADOW GOVERNMENT STATISTICS
A L E R T
June 12, 2006
_______
Trade Manipulation Uncovered
Gimmicked Trade Data Boosted GDP and Soothed Markets
_____
Benchmark revisions released along with the monthly April trade data show that the sharp "improvement" in the March trade deficit -- reported at a time of high U.S. dollar and political stress -- was rigged. Further, the same reporting inconsistencies that allowed the gimmicked trade numbers likely also boosted first-quarter GDP growth by 0.5%. If not, then a further upward revision of that amount looms for first-quarter GDP growth in the pending "final" revision on June 29th or in the annual revision due for release on July 28th.
The Bureau of Economic Analysis (BEA) is more politicized than the Census Bureau. The BEA now "participates" in the trade release with Census, which once handled the monthly number exclusively. Violating common reporting principles with the trade data, the BEA likely repeated the process in the GDP reporting. If further, significant upward revisions to first-quarter growth fail to materialize, then the conclusion can be drawn that the 0.5% GDP goosing had taken place by last month's "preliminary" GDP estimate and revision.
Here is what happened. With the March 2006 reporting, the seasonally-adjusted March trade deficit came in at $62.0 billion, down an unusually sharp $3.6 billion from February's $65.6 billion. It appears that the March deficit was reported at what would be the benchmark revision level from the major revisions due for release coincident with the upcoming April report on June 9th.
While it is standard practice by the statistical agencies to adjust pre-benchmark revision reporting to coincide with the benchmarks, such adjustments are made to month-to-month or quarter-to-quarter changes, not to the absolute level. To my knowledge, pre-adjusting the level of a series such as the trade deficit is unprecedented. As a result, the reporting of the February and March deficits were $65.6 and $62.0 billion, respectively, pre-benchmark, and $62.7 and $61.9 billion post-benchmark. The reporting of March's sharp trade improvement appears to have been a deliberate fabrication, aimed at salving the troubled financial markets of the time.
Even the April trade deterioration to $63.4 billion from March's $61.9 billion was misrepresented in the financial media, touted heavily as coming in below consensus forecasts of $65.0 billion. To the contrary, without the gimmicking of the March data and benchmark revision, the April deficit would have widened to $66.6 billion, topping expectations.
The monthly trade report consists of two primary components, goods and services. The trade in goods is as hard a number as the government publishes, with trade flows backed 100% by underlying paperwork. The annual revisions correct for timing of flows for delayed paperwork and corrections tied to reporting from U.S. trading partners, back to 2003. Import data usually are more carefully tracked by a given country, since those numbers typically are involved in the collection of tariffs.
The services sector is as a soft a number as the government reports, backed primarily by guesstimates of activity by the BEA. The guesstimates have been revised back to 1997. The BEA is well practiced with such gimmicking, which is fundamental to early GDP reporting.
In general, the merchandise trade has been in a rapidly deteriorating deficit for years, while the services sector generally has been in a limited offsetting surplus that until recently had been declining.
As shown in the accompanying graphs, the recent trade deficit has narrowed in revision, with downside revisions to the hard merchandise trade balance more than offset by a new accelerating upside pattern in the services surplus.
The problem with the services surplus is that it should be declining, thanks increasingly to the hard-to-guesstimate outsourcing of services to offshore facilities. Accordingly, the services surplus of $6.0 billion -- just reported for April -- should be at least $3.0 billion lower. Going forward, SGS will comment regularly on the separate merchandise and services components, along with the aggregate monthly deficit.
As to the GDP, revisions in the trade deficit flow through to the net exports account, which is subtracted dollar-for-dollar from GDP growth. Based on the trade revisions as adjusted by the BEA for inflation, the first-quarter 2006 already has picked up or will pick up annualized bonus real growth of 0.5%. Otherwise, current economic weakness has been redistributed to the past with implied growth reductions of 0.3%, 0.4% and 0.6% for the second quarters in each of 2003, 2004 and 2005. Such will be reflected in the upcoming GDP benchmark, along with other revisions. That pattern, though, suggests there is some games-playing afoot for second-quarter 2006, and any resulting growth misstatement surely will not be to the downside, coming into the mid-term election.
Further detail will follow in the July newsletter.
July's Shadow Government Statistics is scheduled for release on Wednesday, July 12, 2006. The monthly newsletter regularly is scheduled for posting on the Wednesday following the Friday release of the employment statistics. The posting of the next SGS on the website, as well as any supplements or further interim alerts, will be advised immediately by e-mail