No. 588: November Trade, Construction Spending, Revised Unemployment
COMMENTARY NUMBER 588
November Trade, Construction Spending, Revised Unemployment
January 8, 2014
__________
Plunge in Oil Imports Narrowed the Trade Deficit;
Data Were Positive for Fourth-Quarter GDP
November Construction Gain Was Statistically Insignificant,
Yet, Earlier Numbers Were Revised Higher
Revised Headline Unemployment Due on January 10th
__________
PLEASE NOTE: The next regular Commentary is scheduled for Friday, January 10th, covering December employment and unemployment.
Best wishes to all — John Williams
OPENING COMMENTS AND EXECUTIVE SUMMARY
Instabilities Appear to Have Continued in Late Reporting for November. As discussed in Commentary No. 585, the October government shutdown had disruptive effects on data gathering and reporting for a variety of economic releases, in the last month or so. These issues may have been factors in the headline data just published for the last of the major November economic releases, on trade and construction. Any data-distortion issues should be worked out in the next several months of reporting.
The coverage of today’s (January 8th) Commentary basically is limited to trade and construction numbers, and the upcoming December labor report on Friday (January 10th). A broader review of current economic activity will be included in that Friday missive.
Revamped Unemployment Rates. Discussed in greater detail in the Week Ahead section, unemployment rates will be revised on a seasonally-adjusted basis for recent history, along with the January 10th release of the December 2013 labor data. The revisions will show revised, relative month-to-month changes in the headline unemployment, going back five years. Once per year, the seasonally-adjusted historical household-survey data, including the headline unemployment rates, are revised so that the month-to-month numbers are on a consistent, seasonally-adjusted basis. The revisions do not mean that the numbers are any more accurate—the unadjusted data are not changed—it is just that the latest, monthly concurrent-seasonal-factor adjustments will have been applied consistently to recent history, making the headline month-to-month data comparable for the only time in the last year.
U.S. Trade Balance—November 2013, Deficit Narrowed Sharply As Oil Imports Plunged. The $5.1 billion monthly narrowing in the headline November trade deficit to $34.3 billion, reflected a $3.4 billion decline imports plus a $1.7 billion gain in exports. The unusually large swing in the deficit was driven by a plunge in oil imports and gain in oil exports, but irregularities in general reporting also may have been a factor. Any distortions to data gathering or reporting from the October federal-government shutdown should be worked out of the numbers in the months ahead. The two months of fourth-quarter trade data will be used in the advance reporting of fourth-quarter 2013 GDP, and they suggest that a narrowing of the net-export account deficit will be a positive contributor to the initial GDP estimate on January 30th.
Regardless of short-term monthly volatility, underlying economic reality continues to suggest an ongoing long-range deterioration in the U.S. trade deficit.
Nominal (Not-Adjusted-for-Inflation) Trade Deficit. The seasonally-adjusted monthly trade deficit in goods and services for November 2013, on a balance-of-payments basis, narrowed to $34.252 billion, from a revised $39.328 (previously $40.641) billion in October. The monthly trade deficit shrinkage reflected a $3.4 billion decline in imports and a $1.7 billion gain in exports. Trade in energy-related products) resulted in the bulk of the decline in imports as well as a fair potion of the gain in exports. On the import side, the decline was seen both in lower physical oil-import volume and lower oil prices.
Real (Inflation-Adjusted) Trade Deficit. Adjusted for seasonal factors, and net of oil-price swings and other inflation (2009 chain-weighted dollars, used for GDP deflation), the November 2013 merchandise trade deficit (no services) narrowed to $44.6 billion, from a revised $47.0 billion in October, and versus $52.6 billion in November 2012.
The two months of real deficit now reported for fourth-quarter 2013 annualize out to $549.8 billion, significantly narrower that the current deficit estimate of $584.9 billion for the third-quarter. Where the November trade reporting is the last, prior to the fourth-quarter 2013 advance GDP estimate on January 30th, the implied improvement in the quarterly trade balance should make a meaningful positive contribution to the initial growth estimate of the GDP
Construction Spending—November 2013. Despite an overall boost in activity level from heavy upside revisions, the headline November construction spending gain was not statistically-significant. Construction spending and other recent residential-structure data likely reflect unusual reporting patterns tied to data-gathering disruptions around the October government shutdown. Despite the insignificant 1.0% headline gain in November 2013, October’s level was revised 2.6% higher from its initial reporting.
For November, all the headline monthly gain was in private construction, split between residential and nonresidential spending. The large upside revisions to prior data, though, largely were in residential construction, where severe data distortions have been evident in other, related reporting. Public-sector construction fell in November. The general trend here has been minimally to the upside since 2011, and monthly changes rarely have been statistically-significant. The broad pattern of construction activity in troubled economy has been one of plunge-and-stagnation, not the otherwise-hyped GDP pattern of plunge-and-ever-after-recovery, with a recovery starting in mid-2009.
Official Reporting. The total value of construction put in place in the United States for November 2013 was a headline $934.4 billion, on an annual-rate basis, not adjusted for inflation. That estimate was up for the month by a statistically-insignificant 1.0%, versus a revised 0.9% monthly gain in October and a revised 1.4% gain in September. Adjusted for PPI new construction inflation, aggregate real spending in November 2013 was up by 1.5% for the month, versus a 1.4% October gain and a 1.3% September gain.
On a year-to-year basis, November construction spending was up by a statistically-significant 5.9%, versus 7.2% in October and 5.0% in September. Net of construction costs indicated by the PPI current construction index, year-to-year growth in spending was 4.7% in November, 6.5% in October and 4.2% in September. More-realistic private surveying suggests annual costs to be up by enough to come close to turning those annual construction-spending growth rates flat or into annual contractions.
Reflected in the usual full spectrum of eight graphs in the Reporting Detail section, the 1.0% monthly gain in November total construction was composed of private residential construction up by 1.9%, private nonresidential construction up by 2.7%, and public construction down by 1.8%.
[For further details on November trade and construction spending data,
see the Reporting Detail Section.]
__________
HYPERINFLATION WATCH
Hyperinflation Outlook. With Hyperinflation 2014—The End Game Begins published yesterday, January 7th, a new Hyperinflation Summary for this section will be added in the near future.
__________
REPORTING DETAIL
U.S. TRADE BALANCE (November 2013)
November Trade Deficit Narrowed Sharply, As Oil Imports Plunged. The $5.1 billion monthly narrowing in the headline November trade deficit to $34.3 billion, reflected a $3.4 billion decline imports plus a $1.7 billion gain in exports. The unusually large swing in the deficit was driven by a plunge in oil imports and gain in oil exports, but irregularities in general reporting also may have been a factor. Any distortions to data gathering or reporting from the October federal-government shutdown should be worked out of the numbers in the months ahead. In any event, the two months of fourth-quarter trade data will be used in the advance reporting of fourth-quarter 2013 GDP, and they suggest that a narrowing of the net-export account deficit will be a positive contributor to the initial GDP estimate on January 30th.
Regardless of short-term monthly volatility, underlying economic reality continues to suggest an ongoing long-range deterioration in the U.S. trade deficit.
Nominal (Not-Adjusted-for-Inflation) Trade Deficit. The Bureau of Economic Analysis (BEA) and the Census Bureau reported yesterday, January 7th that the nominal, seasonally-adjusted monthly trade deficit in goods and services for November 2013, on a balance-of-payments basis, narrowed to $34.252 billion, from a revised $39.328 (previously $40.641) billion in October. The monthly trade deficit shrinkage reflected a $3.4 billion decline in imports and a $1.7 billion gain in exports. Trade in energy-related products (including crude oil) resulted in the bulk of the decline in imports and was just shy of accounting for half the gain in exports. On the import side, the decline was seen both in lower physical-import volume and lower oil prices.
Energy-Related Petroleum Products. Lower prices and physical imports of oil led to a 16.9% decline in the not-seasonally-adjusted (down by 11.2% adjusted) the value of imported, energy-related petroleum products in November 2013. For November 2013, the not-seasonally-adjusted average price of imported oil fell to $94.69 per barrel, from $99.96 per barrel in October, and was down from $97.45 per barrel in November 2012. Not-seasonally-adjusted, physical oil import volume in November 2013 averaged 7.091 million barrels per day, down from 7.820 million in October, and down from 8.099 million barrels per day in November 2012.
Ongoing Cautions on Data Quality. As previously discussed, meaningful distortions in the regular monthly, physical flow-of-trade, and related paperwork, likely have been impairing the quality of recent trade reporting. The October revisions may have alleviated some of those issues. Separately, potentially heavy distortions in headline data also continue from seasonal adjustments, much as has been seen in other economic releases, such as retail sales and payrolls, where the headline number reflects month-to-month change. As has been discussed frequently (see Hyperinflation 2012 for example), the extraordinary length and depth of the current business downturn have disrupted regular seasonality patterns. Accordingly, the markets should not rely heavily on the accuracy of the monthly headline data.
Real (Inflation-Adjusted) Trade Deficit. Adjusted for seasonal factors, and net of oil-price swings and other inflation (2009 chain-weighted dollars, used for GDP deflation), the November 2013 merchandise trade deficit (no services) narrowed to $44.6 billion, from a revised $47.0 (previously $48.3) billion in October, and versus $52.6 billion in November 2012.
The two months of real deficit now reported for fourth-quarter 2013 annualize out to $549.8 billion, significantly narrower that the current estimate of $584.9 billion for the third-quarter. Where the November trade reporting is the last, prior to the fourth-quarter 2013 advance GDP estimate on January 30th, the implied improvement in the quarterly trade balance should make a meaningful positive contribution to the initial growth estimate of the GDP
CONSTRUCTION SPENDING (November 2013)
Following Heavy Upside Revisions, Headline November Construction Spending Growth Was Insignificant. Headline construction spending and other recent residential-structure data likely reflect unusual reporting patterns tied to data-gathering disruptions from the government shutdown in October. Despite the statistically-insignificant 1.0% headline gain in November-2013, October and September data were revised sharply higher, with October’s level now 2.6% higher than the initial reporting, and with the September level revised higher by 1.4%.
For November, all the headline monthly gain was in private construction, split between residential and nonresidential spending. The large upside revisions to the aggregate September and October numbers, however, largely were in residential construction, where severe data distortions have been evident in other, related reporting. Public-sector construction fell in November. Although the general trend—even in nominal terms—has been minimally to the upside since 2011, and monthly changes rarely have been statistically-significant. The general pattern of activity has been one of plunge-and-stagnation, not the hyped GDP pattern of plunge-and-ever-after-recovery, with the recovery starting in mid-2009.
Official Reporting. The Census Bureau reported (December 2nd) the headline, total value of construction put in place in the United States for November 2013 at $934.4 billion, on a seasonally-adjusted—but not-inflation-adjusted—annual-rate basis. That estimate was up month-to-month by a statistically-insignificant 1.0% +/- 1.9% (all confidence intervals are at the 95% level), versus a revised $925.1 (previously $908.4) billion in October, which was up by a revised 0.9% (previously up by 0.8%) from September. In turn, annualized activity in September was $916.5 (previous $901.2) billion, which was up by a revised 1.4% (previously down by 0.3%) versus an unrevised August level.
Adjusted for PPI new construction inflation, aggregate real spending in November 2013 was up by 1.5% month-to-month, versus a 1.4% gain in October and a 1.3% gain in September.
On an annual-growth basis, November 2013 construction spending was up by a statistically-significant 5.9% +/- 2.3%, down from a revised 7.2% (previously 5.3%) in October, and a revised 5.0% (previously 4.4%) in September. Net of construction costs indicated by the PPI current construction index, year-to-year growth in spending was 4.7% in November, versus 6.5% in October and 4.2% in September. More-realistic private surveying suggests annual costs to be up by enough to come close to turning those annual construction-spending growth rates flat or into annual contractions.
The statistically-insignificant 1.0% gain in monthly November 2013 construction spending, versus the 0.9% gain in October, included a 1.8% contraction in November public construction spending, versus a 3.1% decline in October. November private construction was up by 2.2% for the month, having been unchanged in October. The following graphs reflect the 1.0% monthly gain in November total construction, with private residential construction up by 1.9%, private nonresidential construction up by 2.7%, and public construction down by 1.8%. Also reflected is the 0.9% monthly gain in October total construction, with private residential construction unchanged, private nonresidential construction up by 0.1% and public construction up by 3.1%.
The preceding two graphs reflect total construction spending through November 2013, before and after inflation adjustment. The inflation adjusted graph is on an index basis, with January 2000 = 100.0.
There is no perfect inflation measure for deflating construction, but the PPI new construction index is the closest found in publicly available series. Adjusted for the PPI measure, construction spending shows the economy slowing in 2006, plunging into 2011 and then turning minimally higher in an environment of low-level stagnation, through the most-recent reporting. The pattern of inflation-adjusted activity here does not confirm the economic recovery shown in the headline GDP series (see Commentary No. 585). To the contrary, the latest construction reporting, both before and after inflation adjustment, shows a pattern of ongoing stagnation. Please note that the inflation series here will be revised after the publication of the revamped producer price index series, due for publication in February 2014 (see Commentary No. 582).
The next graph reflects the reporting of November 2013 construction employment, which will be updated with the January 10th Commentary and the release of December 2013 labor conditions. In theory, payroll levels should move more closely with the inflation-adjusted aggregate series, where the nominal series reflects the impact of costs and pricing, as well as a measure of the level of physical activity.
The next graph shows total construction spending, broken out by the contributions from total public spending (blue), private-nonresidential spending (yellow) and private residential spending (red).
The next two graphs cover private residential construction, including housing starts data, for November 2013. As detailed in Commentary No. 584, the November surge in housing starts activity likely was an artefact of surveying and reporting disruptions around the October government shutdown. The November surge was not confirmed by November building permits. Keep in mind that the construction spending series is in nominal (not-adjusted-for-inflation) dollars, while housing starts and building permits reflect unit volume.
The last two graphs, preceding, show the patterns of the monthly level of activity in private nonresidential construction spending and in public construction spending. The spending in private nonresidential construction remains well off its historic peak, but has bounced higher recently off a secondary, near-term in late 2012. Public construction spending, which is 98% nonresidential, continues in a broad downtrend, with intermittent, short-lived bounces.
__________
WEEK AHEAD
Weaker-Economic and Stronger-Inflation Reporting Likely in the Months Ahead. [The only major economic release in the week ahead is the December labor report, covered in the Pending Releases. Otherwise, this Week Ahead section is unchanged from the prior regular Commentary.] The markets generally remain overly optimistic as to the economic outlook, although expectations have softened during the last year. That circumstance, and underlying fundamentals that remain highly suggestive of deteriorating business activity, mean that weaker-than-consensus economic reporting should remain the general trend. Inflation likely will be higher than market expectations. Data distortions resulting from the October government shutdown temporarily increase the risk for unusual reporting and revisions in most federal-government and related series.
In terms of monthly inflation reporting, energy-inflation-related seasonal-adjustment factors will be turning negative in December. They were not positive enough in November to offset declines in unadjusted energy prices fully in the PPI and the CPI. That said, upside pressure on oil-related prices should reflect intensifying impact from a weakening U.S. dollar in the currency markets, and from ongoing political instabilities in the Middle East. The dollar faces pummeling from continuing QE3, and the ongoing U.S. fiscal-crisis debacle (see Hyperinflation 2014—The End Game Begins). Particularly in tandem with the likely weakened dollar, inflation reporting in the year ahead generally should reflect much higher-than-expected inflation.
A Note on Reporting Quality Issues and Systemic Reporting Biases. Significant reporting-quality problems remain with most major economic series. Headline reporting issues are tied largely to systemic distortions of seasonal adjustments. The data instabilities were induced by the still-ongoing economic turmoil of the last six-to-seven years, which has been without precedent in the post-World War II era of modern economic reporting. These impaired reporting methodologies provide particularly unstable headline economic results, where concurrent seasonal adjustments are used (as with retail sales, durable goods orders, employment and unemployment data), and they have thrown into question the statistical-significance of the headline month-to-month reporting for many popular economic series.
PENDING RELEASES:
Employment/Unemployment (December 2013). The release of December 2013 employment and unemployment by the Bureau of Labor Statistics (BLS) is scheduled for Friday, January 10th. With the recent monthly data still recovering from the effects of the October government shutdown, the December numbers still may be haphazard. The extraordinary misreporting of October and distortions in November labor data were covered in Commentaries No. 572 and No. 580.
From the payroll employment perspective, the BLS trend model suggests a 170,000 jobs gain for December. The consensus tends to close in around that trend, yet expectations appear to be running nearly 30,000 jobs higher. Underlying economic reality would suggest a downside surprise with the payrolls.
Expectations appear to be for the headline December U.3 unemployment to hold at November’s level of 7.0%. That is not unreasonable for the U.3 reporting, but watch out for revisions. If U.3 drops, there likely would some further labor-force loss associated with that. The broader U.6 and ShadowStats unemployment measures would tend to hold, or increase anew, at their higher respective levels.
Again, all these numbers are unsettled and could come in well outside general expectations.
Annual Revisions to Seasonally-Adjusted Unemployment Rates and Household Data. Once per year, along with the December release of the unemployment numbers, the seasonally-adjusted historical household-survey data, including the headline unemployment rates, are revised so that the month-to-month numbers are on a consistent, seasonally-adjusted basis, going back for five years. The revisions do not mean that the numbers are any more accurate—the unadjusted data are not changed—it is just that the latest, monthly concurrent-seasonal-factor adjustments will have been applied consistently to recent history, making the headline month-to-month data comparable for the only time in the last year.
Normally, the BLS recalculates the seasonal factors for the current reporting and prior history, each month, but it only publishes the current month’s data on the recalculated basis. The data in every prior month remain as they were reported initially, with their own unique set of seasonal factors. Accordingly, other than with the December revisions, the seasonally-adjusted monthly numbers are not consistent. Month-to-month comparisons, such as the “the unemployment rate dropped by 0.1% in September,” cannot be made legitimately and otherwise are meaningless.
ShadowStats will show the before and after data in its accompanying analysis. In the BLS release of January 2014 data, planned for February 7th, the population estimate will be revised for the household (unemployment) survey, making all the respective December 2013 and January 2014 details not comparable, that is in addition to the resumption of the regular scrambling of the monthly unemployment data, with inconsistent, concurrent seasonal factor adjustments. Also, the annual benchmark revision for the payroll survey will be published at that time.
__________