Flash Update
FLASH UPDATE - Mar. 4, 2007
JOHN WILLIAMS' SHADOW GOVERNMENT STATISTICS
FLASH UPDATE
March 4, 2007
__________
Recession Continues to Barrel Along
Market Disquiet Mounts As "Nesting Season" Nears
__________
The SGS Alternate Data pages for the CPI, GDP and U.S. Dollar have been updated. The CPI tables now include monthly index levels for the SGS-Alternate Consumer Price Measure (please see comments in the Inflation section below). -- Best wishes to all, John Williams
Deepening Recession Helps Trigger Greenspan Waffle and Market
Wobble
When consensus economic forecasters start to talk of recession, usually a downturn has become a certainty, with economic activity already having contracted for at least six-months to a year. Wall Street economists, and Administration and Federal Reserve officials, typically are the last to talk of the politically unthinkable, for fear of negative reactions they might trigger in the financial markets. The game is afoot.
Prior to his extended stint as head of the U.S. central bank, Alan Greenspan was noted more for his political skills than his ability as an economic forecaster. Nonetheless, in retirement, he has moved back into the consensus economic forecaster field. In an unusually bold move last Monday, the former Fed Chairman opined that a U.S. recession was possible by year-end 2007. After some ensuing stock-market turmoil, Greenspan hedged that he did not think a recession was "probable."
That said, the panic that hit the U.S. stock market on Tuesday afternoon has not been explained satisfactorily. Granted that the Dow Jones Industrial Average had suffered early Tuesday from the heavy selling in Asia and perhaps from Greenspan's comments or the weak durable goods orders, but that all was old news by the time of the market disruption. With no proximal news accompanying the heavy, panicked selling, the late-afternoon incident gave the impression of someone having major liquidity problems.
On Wednesday, Federal Reserve Chairman Ben Bernanke assured Congress that the President's Working Group on Financial Markets (a.k.a. the "Plunge Protection Team") was on the job. Accordingly, some of the positive market pressures seen at various times from late Tuesday through Friday most likely were at least partially orchestrated by those protectors of financial-market stability.
With the DJIA down 533 points for the week, Treasury-note yields eased slightly, as did the U.S. dollar, particularly against the Japanese yen. Oil prices generally firmed, while the price of gold was down.
In contrast, massive stock selling -- driven by dollar dumping and accompanied by soaring gold prices -- would not be as easy for the Plunge Protection Team to contain. Last week's events, however, appear to have rattled stock-market confidence, and with fundamentals favoring an eventual dollar-driven sell-off, risks of disorderly markets have increased markedly.
Stock market crashes rarely, if ever, happen at historic highs. Instead, the major panics -- such as seen in 1929 and 1987 -- followed market peaks by a month or two with mounting selling pressures. As noted in earlier newsletters, some years back, I retained the services of a mass psychologist in an effort to explain why October/November seemed to be the favored season for stock market crashes. His explanation was that humans have a vestigial squirreling instinct that kicks in as winter nears. He also noted that a vestigial nesting instinct kicked in at a six-month offset to the squirreling season -- in the April/May timeframe -- when the markets also would be more vulnerable to panics.
Unusual Weather Patterns Still Playing Out in Data. Economic reports of the last week or so generally have shown a deepening recession, but there have been some indicators still fluttering from the impact of unusual weather patterns and related poor-quality seasonal adjustments.
For example, on the heels of the devastating January housing starts data, seasonally-adjusted January new home sales plunged by 16.6% for the month and by 20.1% year-to-year. Seasonally-adjusted existing home sales in January, however, rose by 3.0% for the month, but still were down 4.3% for the year. Since new home sales data are based on contract signings, where existing home sales data are based on closings, the latter series still is working on activity from December 2006.
Also mixed were consumer confidence measures. The Conference Board's February consumer confidence gained 2.1% month-to-month, while the University of Michigan's consumer sentiment fell by 5.8%. The Conference Board measure is seasonally adjusted, which does not make much sense and is of suspect purpose, given that the Conference Board does not release the unadjusted number. The Michigan survey is unadjusted. How does one seasonally-adjust peoples' attitudes?
It's like the old story of three government statisticians who went deer hunting. The first shoots at a buck and misses him to the left. The second shoots but misses by a like amount to the right. The third statistician works on his calculator for a minute and, then, instead of shooting, jumps up and down in jubilation, shouting, "Hey, seasonally adjusted, we got him!"
Although the manufacturing purchasing managers survey bounced above 50.0 (recession/expansion dividing line) in February, along with its employment measure, seasonal distortions also appear to be at work. The prices paid component, which is not seasonally adjusted, surged from 53.0 to 59.0.
On the employment front, help-wanted advertising fell to 32 in January from an upwardly revised 34 (was 33) in December. The January number was down 15.8% from the year before. Even so, this series seems to have been distorted to the upside by the weather abnormalities. Also, year-to-year, unemployment claims have continued to soar in recent weeks, an economic negative.
February's employment report (due Friday, March 9th) is at risk of showing weaker jobs growth and higher unemployment than suggested by already tepid consensus forecasts. Reflecting a deepening recession, monthly payrolls shortly should begin contracting month-to-month.
Seasonally-adjusted new orders for durable goods fell by 7.8% (8.1% net of revisions) in January, following a 2.8% gain in December. Year-to-year change was 4.0%. Nondefense capital goods contracted by 19.9% for the month, following a 10.5% gain in December, with annual growth at 0.8%. Although these series are highly volatile on a monthly basis, they are nearing an overall recession signal, which will be discussed in the next newsletter.
Finally, the "preliminary" estimate revision to the seasonally-adjusted, annualized quarterly real (inflation-adjusted) growth in fourth-quarter GDP was revised to 2.22% +/- 3% (statistically indistinguishable from a contraction for the third consecutive quarter) from the "advance" estimate of 3.47%. Annual growth, measured fourth quarter to fourth quarter, was 3.07% (revised from 3.38%) and now is lower than the 3.15% seen in fourth-quarter 2005. In terms of annual average GDP growth, 2006 now stands at 3.30% (revised from 3.38%) versus 3.22% in 2005.
Although the downside revision was unusually large, the numbers remain of such poor quality as to be meaningless. The growth patterns, however, remain weak enough to allow for subsequent annual benchmark revisions to show a recession in place as of the second quarter of 2006. The SGS-Alternate GDP now is estimated as showing an annual real contraction of 1.5% in the fourth quarter versus a similar 1.5% in the third quarter.
The official GDP estimates for the fourth quarter still are too crude for the Bureau of Economic Analysis to estimate gross national product (GNP) or gross domestic income (GDI) growth rates for the fourth quarter.
Inflation. The seasonally-adjusted January CPI was reported up by roughly 0.2% (0.3% unadjusted), with annual inflation at roughly 2.1% versus 2.5% in December. While the Bureau of Labor Statistics went to three decimal points with the new index-level reporting, it did not restate any prior periods on the same basis, where index levels were rounded to the nearest decimal point. As a result, annual growth rates will be less precise or consistent until the BLS gets around to reporting January 2008 data. Distressing to the inflation gimmickers, the annual "core" CPI rate rose to 2.7% from 2.6% in December.
Annual inflation for the Chain Weighted CPI-U was 1.9% in January, versus 2.4% in December.
Annual inflation reflected in January's personal consumption expenditure (PCE) deflation measure was 2.0%, versus 2.3% in December, while the "core" PCE was 2.3% in January, versus 2.2% in December. Again, the "core"-number concept has started to backfire.
Adjusted for all the methodological gimmicks since 1980, the SGS-Alternate Consumer Price Measure shows annual inflation at 9.9%, down from 10.0% in December.
A Note on Alternate CPI Index Levels The levels posted for the SGS Alternate Consumer Price Measure on the Alternate Data tab are not seasonally adjusted, and there is no plan to adjust them. Like consumer confidence, prices are what they are. Inflation is viewed most meaningfully on a year-to-year basis, instead of dealing with a heavily-massaged, seasonally-adjusted monthly number. While there are some seasonal variations, such as companies tending to raise prices effective January 1st, or the Bureau of Labor Statistics usually implementing methodological changes in January, most uses of the CPI measure are based on the not-seasonally-adjusted year-to-year change.
Further details follow in the next newsletter.
___________________________________________
March's "Shadow Government Statistics" monthly newsletter is targeted for the week of March 19th.
FLASH UPDATE
March 4, 2007
__________
Recession Continues to Barrel Along
Market Disquiet Mounts As "Nesting Season" Nears
__________
The SGS Alternate Data pages for the CPI, GDP and U.S. Dollar have been updated. The CPI tables now include monthly index levels for the SGS-Alternate Consumer Price Measure (please see comments in the Inflation section below). -- Best wishes to all, John Williams
When consensus economic forecasters start to talk of recession, usually a downturn has become a certainty, with economic activity already having contracted for at least six-months to a year. Wall Street economists, and Administration and Federal Reserve officials, typically are the last to talk of the politically unthinkable, for fear of negative reactions they might trigger in the financial markets. The game is afoot.
Prior to his extended stint as head of the U.S. central bank, Alan Greenspan was noted more for his political skills than his ability as an economic forecaster. Nonetheless, in retirement, he has moved back into the consensus economic forecaster field. In an unusually bold move last Monday, the former Fed Chairman opined that a U.S. recession was possible by year-end 2007. After some ensuing stock-market turmoil, Greenspan hedged that he did not think a recession was "probable."
That said, the panic that hit the U.S. stock market on Tuesday afternoon has not been explained satisfactorily. Granted that the Dow Jones Industrial Average had suffered early Tuesday from the heavy selling in Asia and perhaps from Greenspan's comments or the weak durable goods orders, but that all was old news by the time of the market disruption. With no proximal news accompanying the heavy, panicked selling, the late-afternoon incident gave the impression of someone having major liquidity problems.
On Wednesday, Federal Reserve Chairman Ben Bernanke assured Congress that the President's Working Group on Financial Markets (a.k.a. the "Plunge Protection Team") was on the job. Accordingly, some of the positive market pressures seen at various times from late Tuesday through Friday most likely were at least partially orchestrated by those protectors of financial-market stability.
With the DJIA down 533 points for the week, Treasury-note yields eased slightly, as did the U.S. dollar, particularly against the Japanese yen. Oil prices generally firmed, while the price of gold was down.
In contrast, massive stock selling -- driven by dollar dumping and accompanied by soaring gold prices -- would not be as easy for the Plunge Protection Team to contain. Last week's events, however, appear to have rattled stock-market confidence, and with fundamentals favoring an eventual dollar-driven sell-off, risks of disorderly markets have increased markedly.
Stock market crashes rarely, if ever, happen at historic highs. Instead, the major panics -- such as seen in 1929 and 1987 -- followed market peaks by a month or two with mounting selling pressures. As noted in earlier newsletters, some years back, I retained the services of a mass psychologist in an effort to explain why October/November seemed to be the favored season for stock market crashes. His explanation was that humans have a vestigial squirreling instinct that kicks in as winter nears. He also noted that a vestigial nesting instinct kicked in at a six-month offset to the squirreling season -- in the April/May timeframe -- when the markets also would be more vulnerable to panics.
Unusual Weather Patterns Still Playing Out in Data. Economic reports of the last week or so generally have shown a deepening recession, but there have been some indicators still fluttering from the impact of unusual weather patterns and related poor-quality seasonal adjustments.
For example, on the heels of the devastating January housing starts data, seasonally-adjusted January new home sales plunged by 16.6% for the month and by 20.1% year-to-year. Seasonally-adjusted existing home sales in January, however, rose by 3.0% for the month, but still were down 4.3% for the year. Since new home sales data are based on contract signings, where existing home sales data are based on closings, the latter series still is working on activity from December 2006.
Also mixed were consumer confidence measures. The Conference Board's February consumer confidence gained 2.1% month-to-month, while the University of Michigan's consumer sentiment fell by 5.8%. The Conference Board measure is seasonally adjusted, which does not make much sense and is of suspect purpose, given that the Conference Board does not release the unadjusted number. The Michigan survey is unadjusted. How does one seasonally-adjust peoples' attitudes?
It's like the old story of three government statisticians who went deer hunting. The first shoots at a buck and misses him to the left. The second shoots but misses by a like amount to the right. The third statistician works on his calculator for a minute and, then, instead of shooting, jumps up and down in jubilation, shouting, "Hey, seasonally adjusted, we got him!"
Although the manufacturing purchasing managers survey bounced above 50.0 (recession/expansion dividing line) in February, along with its employment measure, seasonal distortions also appear to be at work. The prices paid component, which is not seasonally adjusted, surged from 53.0 to 59.0.
On the employment front, help-wanted advertising fell to 32 in January from an upwardly revised 34 (was 33) in December. The January number was down 15.8% from the year before. Even so, this series seems to have been distorted to the upside by the weather abnormalities. Also, year-to-year, unemployment claims have continued to soar in recent weeks, an economic negative.
February's employment report (due Friday, March 9th) is at risk of showing weaker jobs growth and higher unemployment than suggested by already tepid consensus forecasts. Reflecting a deepening recession, monthly payrolls shortly should begin contracting month-to-month.
Seasonally-adjusted new orders for durable goods fell by 7.8% (8.1% net of revisions) in January, following a 2.8% gain in December. Year-to-year change was 4.0%. Nondefense capital goods contracted by 19.9% for the month, following a 10.5% gain in December, with annual growth at 0.8%. Although these series are highly volatile on a monthly basis, they are nearing an overall recession signal, which will be discussed in the next newsletter.
Finally, the "preliminary" estimate revision to the seasonally-adjusted, annualized quarterly real (inflation-adjusted) growth in fourth-quarter GDP was revised to 2.22% +/- 3% (statistically indistinguishable from a contraction for the third consecutive quarter) from the "advance" estimate of 3.47%. Annual growth, measured fourth quarter to fourth quarter, was 3.07% (revised from 3.38%) and now is lower than the 3.15% seen in fourth-quarter 2005. In terms of annual average GDP growth, 2006 now stands at 3.30% (revised from 3.38%) versus 3.22% in 2005.
Although the downside revision was unusually large, the numbers remain of such poor quality as to be meaningless. The growth patterns, however, remain weak enough to allow for subsequent annual benchmark revisions to show a recession in place as of the second quarter of 2006. The SGS-Alternate GDP now is estimated as showing an annual real contraction of 1.5% in the fourth quarter versus a similar 1.5% in the third quarter.
The official GDP estimates for the fourth quarter still are too crude for the Bureau of Economic Analysis to estimate gross national product (GNP) or gross domestic income (GDI) growth rates for the fourth quarter.
Inflation. The seasonally-adjusted January CPI was reported up by roughly 0.2% (0.3% unadjusted), with annual inflation at roughly 2.1% versus 2.5% in December. While the Bureau of Labor Statistics went to three decimal points with the new index-level reporting, it did not restate any prior periods on the same basis, where index levels were rounded to the nearest decimal point. As a result, annual growth rates will be less precise or consistent until the BLS gets around to reporting January 2008 data. Distressing to the inflation gimmickers, the annual "core" CPI rate rose to 2.7% from 2.6% in December.
Annual inflation for the Chain Weighted CPI-U was 1.9% in January, versus 2.4% in December.
Annual inflation reflected in January's personal consumption expenditure (PCE) deflation measure was 2.0%, versus 2.3% in December, while the "core" PCE was 2.3% in January, versus 2.2% in December. Again, the "core"-number concept has started to backfire.
Adjusted for all the methodological gimmicks since 1980, the SGS-Alternate Consumer Price Measure shows annual inflation at 9.9%, down from 10.0% in December.
A Note on Alternate CPI Index Levels The levels posted for the SGS Alternate Consumer Price Measure on the Alternate Data tab are not seasonally adjusted, and there is no plan to adjust them. Like consumer confidence, prices are what they are. Inflation is viewed most meaningfully on a year-to-year basis, instead of dealing with a heavily-massaged, seasonally-adjusted monthly number. While there are some seasonal variations, such as companies tending to raise prices effective January 1st, or the Bureau of Labor Statistics usually implementing methodological changes in January, most uses of the CPI measure are based on the not-seasonally-adjusted year-to-year change.
Further details follow in the next newsletter.
March's "Shadow Government Statistics" monthly newsletter is targeted for the week of March 19th.