Beware of Distorted Economic Reporting.  In this holiday-shortened week, July’s monthly trade-balance report is the only major release available for comment.  The large monthly trade deficit volatility reported in the last two months likely was due more to poor-quality seasonal adjustments than to actual violent swings in trade activity.  The nature of reporting distortions triggered by the severity and extreme duration of this economic contraction has been discussed previously, and will be discussed again in some detail in the upcoming Special Commentary.  At work in the trade data, and as seen in some of the recent unusual volatility in weekly new claims for unemployment insurance, likely are seasonal distortions created by unusual and irregular patterns of activity in the financially-stressed and largely reorganized U.S. automotive industry, and from short-lived year-ago stimulus activity.  Related issues could impact next week’s reporting of August retail sales, industrial production and CPI inflation (see Week Ahead below).  As a separate issue, today’s market-moving large drop in new claims likely resulted from the Department of Labor’s inability to adjust the weekly data meaningfully for regular seasonal variations in activity, particularly around holiday periods such as Labor Day, and not from a sudden shift in weekly economic activity.

July Trade Deficit Reporting Suggests Minor GDP Revisions and Minimal Third-Quarter Impact, So Far.  With a minor downside revision to the size of the previously reported June 2010 trade deficit, the July trade release suggested a possible small, upside revision to estimated real second-quarter GDP growth, due for re-estimation at the end of the month.  Despite the sharp month-to-month reduction reported in the July monthly deficit, though, the inflation-adjusted monthly shortfall still was in line with the second-quarter average, suggesting, so far, no major swing in third-quarter 2010 trade data that would have meaningful impact on the "advance" third-quarter GDP estimate, due at the end of October.

Nominal (Not Adjusted for Inflation) Trade Deficit.  For July 2010, the Bureau of Economic Analysis (BEA) and the Census Bureau reported today (September 9th) that the nominal seasonally-adjusted monthly trade deficit in goods and services narrowed to a better-than-expected $42.8 billion, versus a revised $49.8 (previously $49.9) billion in June, but widened sharply from the $31.1 billion monthly deficit of July 2009.  Monthly nominal data were revised back to January 2010.

Against June 2010, the July trade balance showed a gain in exports and drop in imports, reversing the trends of June.  The lower imports were impacted minimally by trade activity in oil, although unadjusted oil imports reflected both lower physical volume and oil prices.  Specifically, for the month of July 2010, the not-seasonally-adjusted average price of imported oil was $72.09 per barrel, versus $72.44 in June 2010 and $62.52 in July 2009.  In terms of not-seasonally-adjusted physical oil imports, July 2010 volume averaged 10.056 million barrels per day, versus 10.398 million in June 2010 and 9.504 million in July 2009.

Real (Inflation-Adjusted) Trade Deficit.  A narrowing trade deficit directly boosts GDP growth and vice versa.  The July revisions in the trade data set up a possible minor revision to the goods element of the net-export component of the GDP, which would be a small positive for the upcoming, second revision to second-quarter GDP (due September 30th).  While earlier monthly nominal data were revised, the real monthly numbers were not.  The real July number, itself, largely was neutral (slightly positive) in terms of its implications, so far, on the pending "advance" estimate of third-quarter GDP, due October 29th.

As reported by the BEA, adjusted for seasonal factors and inflation (2005 chain-weighted dollars as used in reporting real GDP), the second-quarter deficit was ran at an annualized pace of $575.1 billion, revised lower from the $577.1 billion reported last month.  The annualized real July deficit, as reported, was $572.3 billion, but that will be supplemented by the August estimate before the third-quarter GDP estimate is put together. 

Due to underlying assumptions used by the BEA in its GDP estimate, something shy of the full differential in the revised second-quarter number is likely to flow through to the GDP revision.  Important here is that this particular series is just the merchandise sector, which eventually has a paper trail behind it, from reporting by U.S. Customs.  It does not include the services area, which largely is a guesstimate by the BEA, and which traditionally runs a smaller offsetting surplus to the merchandise deficit.  

Week Ahead.  Given the unfolding reality of a weaker economy (or re-intensifying downturn) and more-serious inflation problems than generally are expected by the financial markets, risks to reporting will tend towards higher-than-expected inflation and weaker-than-expected economic reporting in the months ahead.  Increasingly, previously unreported economic weakness will be evident in prior-period revisions.

Retail Sales (August 2010).  Due for release on Tuesday, September 14th, August’s monthly retail sales number is expected to show a seasonally-adjusted 0.2% monthly gain (Briefing.com), versus a 0.4% gain initially reported for July.  Anecdotal evidence of continuing weakness in consumption and possible seasonal-factor distortions from last year’s "cash-for-clunkers" program suggest a fair shot at a downside surprise in reporting.  The monthly change also is a fair bet to be negative, after inflation adjustment.  Broad economic activity, such as reported in the GDP, usually is measured net of inflation, so that changes in sales levels reflect changes in physical volume as opposed to changes in prices.

Industrial Production (August 2010).  Due for release on Wednesday, September 15th, August industrial production should be due for some reporting catch-up, where some recent strength was due partially to irregular auto production patterns (and resulting seasonal adjustment distortions), as mentioned in the opening comments.  Briefing.com reports a consensus expectation for seasonally-adjusted monthly gain of 0.4% versus an initial 1.0% gain for July.  Again, reporting risk favors a downside reporting surprise.

Producer Price Index — PPI (August 2010).  Due for release on Thursday, September 16th, the August PPI has a fair shot at topping a consensus estimate of a 0.2% seasonally-adjusted monthly increase (Briefing.com), the same level as in July.  Reversal of seasonal factor biases that depressed adjust PPI reporting earlier this year likely will add upside pressure to this somewhat randomly volatile series.

Consumer Price Index — CPI (August 2010).  Due for release on Friday, September 17th, the August CPI-U and related series also will receive seasonal-factor boosts to gasoline price inflation.  Separately, in August 2009, the Bureau of Labor Statistics (BLS) did not include the government’s cash-for-clunker payments as part of auto pricing at the time.  Such artificially knocked off about 0.3% from the CPI-U in August 2009, although it was recovered in later, post-clunker reporting.  Nonetheless, those events may affect the seasonal-adjustment of August 2010 prices.  In combination, the gasoline and auto circumstances offer a good bet for an upside surprise to the 0.1% consensus seasonally-adjusted monthly gain for the CPI-U (Briefing.com).  July reporting showed a 0.3% increase.

Year-to-year inflation would increase or decrease in August 2010’s CPI-U reporting, dependent on the seasonally-adjusted monthly change, versus the 0.37% adjusted monthly gain seen in August 2009.  I use the adjusted change here, since that is how consensus expectations are expressed.  To approximate the annual inflation rate for August 2010, the difference in August’s headline monthly change (or forecast of same) versus the year-ago monthly change should be added to or subtracted directly from July 2010’s annual inflation rate of 1.24%.  A consensus result of 0.1% would leave annual CPI-U at about 1.0%.