GDP Estimate Was of Unusually Poor Quality.  This morning’s "advance" estimate of annualized 3.17% real (inflation-adjusted) GDP growth was nonsensical, even though it was somewhat shy of consensus.  Most of the reporting was based on guesses; hard data simply are not available this early.  Consider that more than the total reported fourth-quarter growth was accounted for by a narrowing of the trade deficit.  The Bureau of Economic Analysis (BEA) indicated that 3.44 percentage points of growth was generated by an improved net export account.  That estimate, however, was based on just the two months of available data (October and November) for the quarter.  December’s data will not be available until February.

As noted in Commentary No. 345, the relative improvement suggested in the trade deficit for the fourth-quarter (based on the October and November reporting) could have added 1.3 annualized (0.3 quarterly) percentage points to fourth-quarter real GDP growth, but not 3.44 percentage points.  That differential required extremely optimistic assumptions on the part of the BEA as to the December trade results.  Accordingly, the upcoming trade release will be particularly interesting in terms of its implications for GDP revisions.

Separately, after quarters of a significant inventory build-up, a reduced pace of relative inventory increase reduced the reported real fourth-quarter GDP growth rate by 3.70 percentage points.  Inventories at this point in time are even less reliable than the trade data.  Nonetheless, inventory build-up still accounted for half the annual average GDP growth in 2010.  

Also, despite the 30% annualized (8% quarterly) quarter-to-quarter contraction in housing starts, residential investment rose at a 3.4% annualized pace. 

The point here is that reported 3.17% annualized growth, with the regular +/- 3.0% 95% confidence interval, along with such unusually large swings in unreliable components, should not be taken as a serious or meaningful measure of quarterly economic growth.  I believe that realistic growth would have been flat-to-minus and eventually that should prove out in long-range revisions.   

Where early GDP reporting generally is of extremely poor quality, some catch-up should be seen in the annual benchmark revisions due for release on July 29th.  At that time — as will be seen with the payroll employment reporting due for revision a week from now — the revisions to prior economic growth generally will be to the downside, showing a more-protracted and deeper economic contraction in place than officially is recognized at present.

With quarterly weakness in the housing starts and in new orders for durable goods, the indications remain in place for a re-intensifying economic downturn, as discussed in Special Commentary No. 342.

 

"Advance" Guesstimate on Fourth-Quarter 2010 GDP Was Unusually Flimsy.  The opening comments covered several unusual issues with the current GDP report.  A more traditional problem lies in how inflation was handled.  On a one-to-one basis, the lower the inflation rate used to deflate the GDP, the higher will be the real or inflation-adjusted GDP growth rate.  Annualized GDP inflation — the GDP Implicit Price Deflator — was reported showing annualized inflation of 0.3% in the fourth-quarter, down from 2.0% in the third, while annualized CPI inflation rose to 2.6% in the fourth-quarter, up from 1.5% in the third. 

GDP — The Detail.  Published this morning, January 28th, by the Bureau of Economic Analysis (BEA), the "advance" or first estimate of fourth-quarter 2010 Gross Domestic Product (GDP) was 3.17% +/- 3% (95% confidence interval), up from an estimated 2.56% in the third-quarter.  For the year, average 2010 GDP versus 2009 was up by 2.86%, versus a 2.63% decline in 2009 against 2008, while year-to-year change in fourth-quarter 2010 over fourth-quarter 2009 was up by 2.79%, higher than the 1.95% year-to-year gain in fourth-quarter 2009, but lower than the 3.25% seen in third-quarter 2010.

The fourth-quarter GDP implicit price deflator was estimated at an annualized pace of 0.26%, versus 2.03% annualized quarterly inflation for third-quarter 2010.  In contrast, annualized seasonally-adjusted quarterly inflation for the CPI-U in the fourth-quarter spiked to 2.57%, following 1.47% annualized inflation in the third-quarter.  The lower the inflation rate used in deflating the GDP, the stronger will be the inflation-adjusted number and vice versa.

The SGS Alternate-GDP estimate for fourth-quarter 2010 is an approximate annual contraction of 2.1% versus the official estimate of a 2.8% gain. Such is more negative than the alternate 1.4% annual contraction (3.2% official gain) in the third-quarter (see the Alternate Data tab).  While annualized real quarterly growth is not estimated formally on an alternative basis, a flat-to-minus quarter-to-quarter change still appears to have been realistic for the fourth-quarter in what generally has been a protracted period of bottom-bouncing.

The following graphs show, first, year-to-year percent change in the official inflation-adjusted quarterly GDP for the history of the series.  The record annual contraction for the series was 4.11%, seen in second-quarter 2009.  The second graph shows the average annual real growth by year for the history of the series.  The record annual contraction there, following the World War II production shutdown, was 2.63% in 2009.

 

 

 

 

 

GNP and GDI.  The initial estimates for fourth-quarter GNP (Gross National Product) and GDI (Gross Domestic Income) likely will not be published until the third release of fourth-quarter GDP on March 25th, since the actual data underlying these national income aggregate numbers are of so little substance — sometimes educated and sometimes hopeful guesses — that the BEA recognizes these data would not be meaningful at present.  Where a one-month reporting delay is typical, two months appears to be standard when annual data are involved.  What the markets and the media do not realize is that today’s aggregated GDP report is about as worthless as the aggregated GDI and GNP measures would have been had they been reported also.

December Durable Goods Orders Fell for Fourth Straight Month and Contracted Quarter-to- Quarter.  The Census Bureau reported yesterday (January 27th) that the regularly-volatile, seasonally-adjusted new orders for durable goods declined by 2.5% (down by 1.4% before prior-period revisions) month-to-month in December 2010, after a revised 0.1% (previously 1.3%) decline in November orders.  Unadjusted, year-to-year change in December new orders was a gain of 6.2%, softer than the revised 11.6% (previously 10.3%) annual increase reported for November.  Adjusted for seasonal factors, but not for inflation, orders fell by 1.2% (4.8% annualized) in the fourth-quarter 2010 versus the third-quarter, which, in turn, was up by 2.0% (8.2% annualized) from the second-quarter.

The widely followed nondefense capital goods orders fell by 6.3% (down 7.2% before prior-period revisions) in December, versus a revised monthly decline of 8.1% (previously down by 6.8%) in November.  For December, the unadjusted year-to-year growth in the series was 6.2%, down from a revised 20.4% (previously 19.0%) gain in November.

Ongoing Housing Problems.  December housing starts were covered in the prior Commentary No. 346, and I received a couple of comments, which are addressed here.  First was a question as to why building permits were so strong, despite the collapsing housing starts.  While I look at permits, reporting distortions in recent years have impaired the value of the series as a leading indicator, and December’s reporting reflected one of those circumstances.  The reporting of a 16.7% monthly increase (all numbers are significant unless otherwise indicated) in December building permits was accompanied by the following caution:  "Increases in the numbers of housing units authorized by building permits in the Northeast and West in December 2010 may be related to changes to state building codes for new residential construction effective in January 2011 in California, Pennsylvania, and New York [builders were acting in order to beat the changing building codes]."  Indeed, permits were reported up by 80.6% in the Northeast, up by 43.9% in the West, but up a statistically-insignificant 3.3% in the Midwest, and down by 7.6% in the South.

The second was a suggestion that I not give in to annualizing growth rates the way the government does.  I understand the sentiment and shall do my best also to report growth on a simple quarter-to-quarter basis when appropriate.  Where the data are reported on an annual basis (such as the housing numbers), the changes often are viewed on an annualized basis as used in the GDP reporting and for comparison or contrast with the GDP numbers.

December Home Sales Reflected Increasing Numbers of Foreclosures.  Both existing and new home sales continued a pattern of bottom-bouncing.  Wednesday’s (January 26th) reporting of December new-home sales (counted based on contract signings, Census Bureau) showed a statistically-insignificant monthly gain of 17.5% (up 13.4% before prior period revisions) +/- 20.7% (95% confidence interval) from November.  November’s revised level was unchanged from October, where it previously had been a 5.5% monthly decline.  The year-to-year decline in the new home sales series in December 2010 was a statistically-insignificant 7.6% +/- 19.9% (95% confidence interval), versus a revised 23.9% (previously 21.2%) decline in November.

Last week’s (January 20th) release of December existing-home sales (counted based on actual closings, National Association of Realtors [NAR]) showed a 12.3% increase in monthly activity versus November’s revised 6.1% (previously 5.6%) increase, thanks to surging foreclosure sales.  The increase included a 22.6% jump in foreclosures, with standard sales up by 7.3%.  On a year-to-year basis, December sales were down by 2.9% from the year before, which was an annual decline of 8.7% net of foreclosures.

Foreclosure activity remained a major distorting factor for home sales, with "distressed" activity accounting for an estimated 36% of existing sales in the NAR’s December reporting, up sharply from 33% in November.  Short-lived foreclosure moratoriums still likely are having some impact here.

Separately, the Census Bureau acknowledges that a portion of new home sales also is from foreclosure activity but offers no estimates as to the scope of the issue.  Some in the construction trade have difficulty competing with the pricing of foreclosed properties.  Until the foreclosure problems get worked out in the system, monthly changes in these home sales numbers cannot be taken as meaningful positive indicators (when the numbers are positive) of underlying activity in homeowner real estate, as it relates to general economic activity. 

 

 

 

 

The preceding graphs reflect different measures of home sales activity since February 2009.  The updated numbers, through December 2010, reflect the seasonally-adjusted level of monthly sales, rather than the annual rates usually published by the industry.

Week Ahead.  Given the unfolding reality of an intensifying double-dip recession and more-serious inflation problems than generally are anticipated 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 should show up in prior-period revisions.

Payroll Employment and Unemployment (January 2011).  The January 2011 report on labor conditions is due for release on Friday, February 4th.  It will include the annual benchmark revisions to payrolls, which should show major downside revisions to payroll levels, along with revamped seasonal adjustments, as well as imposition of new population estimates on unemployment reporting, which will leave the January 2011 household survey data inconsistent with and not comparable to the December 2010 data.

The most important details from the upcoming report will be tied to the payroll revisions, which should show a deeper and longer economic contraction than previously recognized.  Eventually, GDP revisions will move in tandem.  With a catch-up in seasonal factors, as well, there is a fair shot of disappointing what likely will be a positive consensus estimate, with weaker numbers, even declining monthly numbers for payrolls, and with some upturn likely in the headline unemployment rate.