No. 401: Underlying Economic Reality, October Housing Starts
COMMENTARY NUMBER 401
Underlying Economic Reality, October Housing Starts
November 17, 2011
__________
Housing Starts Continued Three-Year Pattern of Bottom-Bouncing
Economic Recovery Never Happened;
Hyped Gains Were Based on Gimmicked Inflation Adjustments
Current Downturn Already Exceeds Duration of Great Depression’s First Down-Leg
__________
PLEASE NOTE: The next regular Commentary is scheduled for Wednesday, November 23rd, covering October existing home sales and new orders for durable goods, as well as the first revision to third-quarter GDP.
—Best wishes to all, John Williams
SPECIAL COMMENTARY – UNDERLYING ECONOMIC REALITY
In line with common experience, I contend that the U.S. economy has been in trouble since at least 2000, when it entered a recession that dragged into 2003. Business activity then began collapsing in 2006, hit a bottom in 2009 and has been bottom-bouncing since. The outlook for the U.S. economy remains bleak, with continued and deteriorating bottom-bouncing ahead. I figure 57 months of recession so far in the current downturn, which tops the 43 months officially in the first down-leg of the Great Depression.
The happier, official version of U.S. business activity is that the economy went through a shallow recession between March 2001 and November 2001, with a much deeper downturn beginning in December 2007 and ending in June 2009. Since then, U.S. economic activity has recovered fully and now exceeds pre-2007 recession levels.
Where some widely followed economic series such as retail sales and industrial production show increasing activity, only GDP reporting shows a full recovery. Those series also all are adjusted by inflation with gimmicked underlying assumptions. Other series such as payrolls, household income, consumer confidence and housing starts support the more dire circumstance. Those series, except for household income, are not subject to inflation adjustments.
Plots of these indicators are shown and discussed, and, where appropriate, viewed with different inflation assumptions. All the graphs plot monthly or quarterly levels of activity, indexed to January 2001 or first-quarter 2001 equal to 100. The shaded areas represent officially pronounced periods of recession.
Consider payroll employment, shown above. Where employment traditionally has been considered a coincident indicator of economic activity, payroll levels suggest the 2001 recession ended in 2003, not 2001, and that the 2007 recession is not over; at least there has been no meaningful rebound in economic activity. The payroll level in October 2011 remained below payroll levels going into the 2001 recession.
GDP and Inflation. The next two graphs are based on GDP reporting. The first graph reflects official levels of real (inflation-adjusted) GDP activity (indexed to first-quarter 2000 equals 100). It shows no clear 2001 recession, with growth generally in an uptrend until the official December 2007 to June 2009 recession, and with third-quarter 2011 business activity fully recovered from the recession.
A significant issue with official GDP reporting is the nature of the inflation rate used to deflate the series. The lower the inflation rate used in the GDP’s implicit price deflator (IPD), the stronger will be the inflation-adjusted level and growth reported for the real GDP. Back in the 1980s, the Bureau of Economic Analysis (BEA) introduced the concept of hedonic adjustments in determining the IPD. Hedonic adjustments altered (usually reduced) inflation estimates, based on nebulous quality concepts that had no relationship to real-world common experience. The effect was to reduce the IPD inflation artificially. Other major countries initially avoided the concept in their GDP calculations, with a number of papers discussing how the U.S. hedonic methodologies gave an artificial boost to reported U.S. economic performance, productivity, etc. relative to the rest of the world. I estimate the hedonics currently reduce the annual IPD by about two percentage points.
There is no easy way to reconcile the official GDP activity with payroll employment activity, without considering the inflation issue. The payroll employment numbers are surveyed and eventually benchmarked. Despite all the issues I have with the employment series, the numbers eventually become fairly solid. In contrast, the GDP estimates are heavily guessed at and modeled, including the IPD.
Backing out the two-percentage point IPD understatement generates the “Inflation-Corrected Real GDP” graph, which is more consistent with the payroll numbers than is the “Headline GDP.” With corrected inflation, the GDP shows the “2001” recession beginning in 2000 and extending into 2003. The “2007” recession begins in 2006, hits a bottom in 2009 and the bottom-bounces thereafter. A small upside bounce is turning down again.
I am attempting to make a simple point here as to how differing assumptions on inflation can throw off official reporting, meaningfully. The inflation-corrected number in the graph is not the same measure as the SGS-Alternate GDP Measure, which also adjusts for hedonics. There are a number of other factors involved in that measure, which complicate the discussion but show an even more-negative “corrected” GDP (see the Alternate Data and Primer tabs on www.shadowstats.com for more detail).
Consumer Reality. The two graphs following the GDP plots are indicators of consumer willingness and ability to consume. The first graph on real median household income is courtesy of www.SentierResearch.com. This series was discussed in Commentary No. 394. As published, the numbers suggest a 2000 start to the “2001” recession, a late-2006 start to the “2007” recession. The numbers suggest a continuing plunge in activity, with no recovery whatsoever.
The next graph shows the Conference Board’s consumer confidence index (rebased to January 2000 equals 100). Here, the pattern shows a 2000 to 2003 recession, with a downturn in late-2007, a 2009 trough, subsequent bottom-bouncing, and a renewed downturn in the most recent numbers.
These charts provide a backdrop for the next graph, which is a plot of “Inflation-Corrected Real Retail Sales.” The standard real retail sales graph (CPI-U) was published in yesterday’s Commentary No. 400.
As with official real GDP reporting, reporting of real retail sales (deflated with the CPI-U by the St. Louis Federal Reserve) uses understated inflation, with the result of overstated levels of real growth and activity. Instead of the CPI-U, I have used the SGS-Alternate Consumer Inflation Measure (1990) to deflate the nominal retail sales number, adding about three percentage points back into annual inflation on the recent CPI-U. As has been discussed frequently in monthly Commentaries and elsewhere, the Bureau of Labor Statistics (BLS), has changed methodologies in recent decades so that the CPI no longer reflects the cost of living for maintaining a constant standard of living, and it no longer reflects what most people consider full out-of-pocket expenses, as a result of using hedonic adjustments.
The inflation-corrected real retail sales shows a 2000 to 2003 recession, a plateau of activity into the next downturn, which begins in 2006, a trough in in 2009 and bottom-bouncing thereafter. As a result of the new deflation of retail sales, the series more closely reflects that patterns seen in consumer confidence and household income, particularly in terms of the 2003 to 2006 plateau in activity.
The housing starts graph does not rely on an inflation adjustment, and accordingly shows plunging activity from 2006 into 2008, with bottoming-bouncing since. Today’s release of the October housing starts data showed a continuation of the bottom-bouncing pattern, which is within one month of the three-year mark.
An inflation-corrected industrial production plot is not included here, but one will be generated for the revision to the Hyperinflation Report. The production index includes components, such as computers, where the volume is calculated from dollar amounts that are deflated by hedonically-dominated inflation estimates. As a result, an adjusted production series should show a somewhat deeper downturn since 2007 and a shallower upturn since the 2009 trough in activity.
There is no question that the government has taken actions in recent decades to depress inflation reporting artificially. As a result, inflation-adjusted economic growth has been overstated, and that helps to account for the discrepancies seen between GDP estimates and employment patterns. Accordingly, the employment estimates are closer to common experience.
These general patterns are in place irrespective of the precise level of inflation understatement by the government. I believe the estimates used here are quite conservative.
Hyperinflation Watch—Systemic-Solvency and Economic Crises Continue. Yesterday’s Special Commentary briefly touched upon the long-term solvency issues facing the United States. An annual GAAP-based deficit running greater than five-trillion dollars is beyond containment. It promises eventual hyperinflation if there is inadequate political will in Washington to address the issue, as currently is the circumstance. Minimal deficit reduction efforts under consideration by the Congressional Super Committee are likely to do nothing meaningful in restoring long-term U.S. solvency, at best, and could devolve into circumstances that bring U.S. solvency issues back into focus for the global financial markets. A new dollar selling panic and crisis reactions actions by the Fed easily could bring the pending hyperinflation threat rapidly to a head.
Today’s Special Commentary took a different approach towards making the case that the economic collapse that began in 2006 or 2007 is ongoing and is worse than popularly is recognized. With no relief in sight for the structural income and credit problems facing consumers, there is no near-term prospect for broad economic recovery in the United States. An ongoing economic downturn has severely negative implications for the projected U.S. federal budget deficit, for projected U.S. Treasury funding needs and for banking-system stress tests and systemic stability. It also promises a volatile political environment coming into the 2012 election, where pocket-book issues historically have dominated national election results more than any other single issue.
There remain no happy solutions available here to remedy the crises, only tools—devil’s choices—for the Fed and the U.S. government to buy a little extra time. Domestic systemic instabilities, and possibly instabilities outside the United States, make substantial, expanded “easing” actions of some form likely by the Federal Reserve, sooner rather than later. From the Fed’s standpoint, keeping the banking system afloat remains its primary concern, not expanding the economy or containing inflation. The ultimate cost in propping the system, however, remains inflation.
The root source of current global systemic instabilities primarily has been the financially-dominant United States, and it is against the U.S. dollar that the global markets ultimately should turn, massively. The Fed and the U.S. Treasury likely will do whatever has to be done to prevent crises in the euro-area from triggering a systemic collapse in the United States. That precedent was established in 2008. Accordingly, it is not from a euro-related crisis, but rather from within the U.S. financial system and financial-authority actions that an eventual U.S. systemic failure likely will be triggered, seen initially in a rapidly accelerating pace of domestic inflation—ultimately hyperinflation.
The financial markets remain extremely volatile and unstable. Underlying the various market upheavals fundamentally is the deepening crisis of confidence in the U.S. dollar and in the long-term doubts of U.S. financial, economic, systemic and political stability. For those living in a U.S. dollar-denominated world, regardless of any ongoing near-term extreme volatility in the U.S. dollar—in either direction—versus the stronger major currencies and gold, the stronger currencies and precious metals remain the fundamental, long-range hedges against what lies ahead.
Massive, fundamental dollar dumping and dumping of dollar-denominated assets may start at anytime, with little or no further warning. With a U.S. government unwilling to balance or even to address its uncontainable fiscal condition; with the federal government and Federal Reserve standing ready to prevent a systemic collapse, so long as it is possible to print, spend, loan or guarantee whatever money is needed; with the U.S. dollar at increasing risk of losing its global reserve currency status; much higher inflation lies ahead, in a circumstance that, again, could evolve rapidly into hyperinflation.
The economic and systemic-solvency crises and the broad inflation and economic issues detailed in the Hyperinflation Special Report (2011) and in recent Commentaries, continue to unfold with outlooks that remain unchanged. A fully updated Hyperinflation Report will be published in December, after the publication of the 2011 GAAP-based financial statements for the U.S. government.
__________
REPORTING DETAIL
Housing Starts Continued Bottom-Bouncing in October. A small decline reported in October housing starts followed a large downward revision to the September number. Nonetheless, the regularly volatile housing starts series continued its protracted bottom-bouncing in the latest reporting, a month shy of completing three years of activity at 75% below 2006’s record construction level. There is no relief in sight. The Census Bureau reported this morning (November 17th) a statistically-insignificant monthly contraction in seasonally-adjusted October 2011 housing starts of 0.3% (a drop of 4.6% before prior-period revisions) +/- 12.7% (all confidence intervals are at the 95% level). September starts were revised sharply lower to a 7.6% (previously a 15.0%) monthly increase. The year-to-year change in October starts was a statistically-significant increase of 16.5% +/- 12.5%, following a downwardly revised 5.2% (previously 10.2%) annual gain in September. The recent swings in the direction and magnitude of annual change in the housing starts series reflect comparisons with volatile year-ago monthly activity that was exaggerated in both directions by stimulus and post-stimulus effects. With housing starts well into their third year of post-housing-market-crash bottom-bouncing, there remain no indications of a reprieve for the industry or the otherwise deteriorating broader economy. Indeed, the pattern of housing starts has remained one of stagnation at an historically low-level plateau of activity, over the last 35 months. Since December 2008, housing starts have averaged a seasonally-adjusted annual rate of 576,000. In that period, all monthly readings have been within the normal range of monthly variability for the aggregate series around that average, with the 628,000 October 2011 monthly reading at 9.0% above average. Again, though, such is within the normal reporting range of this extremely volatile series. As shown in the following graphs, current monthly housing starts activity remains near the record monthly low seen for the present series in April 2009. The current number also is well below any level reported in the predecessor nonfarm housing starts series, which was introduced in 1946. The first graph shows recent detail for current housing starts activity, the second graph shows the same data within the historical context of the post-World War II period. An indexed version of the first graph is included in the introductory Underlying Economic Reality section. Existing Home Sales (October 2011). October existing home sales are due for release on Monday, November 21st. The sales data likely will show ongoing bottom-bouncing, impaired activity that is reflective of many of the fundamentals impacting the October housing starts release. Any upside surprise to the numbers likely would be within the normal monthly volatility of the series. The October new home sales release is scheduled for Monday, November 28th. Gross Domestic Product—GDP (Third-Quarter 2011, Second Estimate, First Revision). The first revision to the third-quarter GDP estimate is due for release on Tuesday, November 22nd. While any revision is likely to be little more than statistical noise, underlying economic reality would favor something of a downside revision. Initial estimates of third-quarter gross national product (GNP)—which is the broadest measure of U.S. economic activity (GDP is GNP net of trade in factor-income, or interest and dividend payments); and of third-quarter gross domestic income (GDI), which is the theoretical income-side equivalent to the GDP’s consumption-side; always have the potential of offering a surprise. New Orders for Durable Goods (October 2011). October durable goods orders are scheduled for release on Wednesday, November 23rd. Monthly change in this highly volatile series rarely is meaningful, with odds currently favoring a monthly contraction in October’s headline monthly activity.
__________