COMMENTARY NUMBER 416
Payrolls, Unemployment and Revisions, M3, PCE Deflator

February 3, 2012

 

__________

 

Basic Economic Outlook Unaltered by Stronger Labor Data

 January Jobs Reading Still at Levels of 11 Years Ago

January Unemployment: 8.3% (U.3), 15.1% (U.6), 22.5% (SGS)

Money Supply M3 Growth Is Picking Up

 

__________

 

PLEASE NOTE: The next regular Commentary is scheduled for Friday, February 10th, covering the December and annual 2011 trade balance.

Best wishes to all — John Williams

 

Opening Comments and Executive Summary.  Welcome to Election-Year 2012!  Last week’s comments on today’s (February 3rd) January 2012 labor market reporting included, “Given pending benchmark revisions, monthly reporting results are unusually vulnerable to any reporting needs of the political establishment.”  The political environment remains worthy of consideration here as a factor that could impact reporting.  Such issues are beyond the regular monthly cautions I offer on the payroll and household data as to the worthlessness of the current labor numbers—irrespective of whether the consensus outlook has been exceeded or disappointed—tied to the irregular and unreliable nature of the seasonal-adjustment process that has become so distorted in the last couple of years (see Hyperinflation 2012).

The hard data (not-seasonally-adjusted) on payrolls always tank in January (January 2012 was down by 2.68 million jobs month-to-month), while the unemployment rate always rises (January’s unadjusted unemployment rate rose to 8.8% from 8.3% in December).  The only difference between those numbers and the headline 243,000 January jobs gain and 8.3% unemployment rate, is how the seasonal adjustments were applied.  There are serious issues with the current quality of those adjustments, and extremely small distortions in those seasonals can make big differences in the resulting headline data.

Whenever there is a major systemic change to an economic series, and the data overtly are massaged so as to prevent distortions, unusually large changes in the resulting near-term reporting automatically are suspect as to accuracy.  Today’s reporting included the annual benchmark and seasonal-factor revisions to the payroll survey, as well as to the annual population controls and revised population assumptions for the household survey.  Where both of those revision processes involve making assumptions (i.e., Is the economy going to be growing, or not?), it is not surprising that assumptions yielding stronger results might be selected over those that did otherwise.

In any event, beyond the revisions, the headline numbers for January 2012 generated by the revamped systems simply were not believable.  New online help-wanted advertising fell sharply in January, indications from the January purchasing managers survey were mixed, and anecdotal evidence still is running to the contrary of happier numbers.  Accordingly, I would expect reporting in the months ahead to revise and weaken with the payrolls, and would expect deterioration in the headline unemployment rate ahead, assuming some catch up in seasonal factors, if that is an issue.

As an aside, there is precedent for direct political manipulation (the first kind) of headline economic data, from a number of administrations—both Democrat and Republican—from the early 1960s and from the onset of modern economic reporting, into the 2000s.  A down economy is extremely difficult for an incumbent party to overcome politically in a presidential-election (see Hyperinflation 2012).  During the first Bush Administration, with George Bush up for re-election, the economy was in recession.  An administration official approached an individual in the computer industry about boosting reporting of computer sales to the Bureau of Economic Analysis (BEA), which reported the GDP.  The sales reporting was boosted, the reported GDP improved, but the public viewed the administration’s improving economic claims as being out of touch with reality.

Most political manipulations, however, are of the second kind, involving methodological changes and in choosing assumptions, where the changes or assumptions tend to give some relative boost to economic reporting.

Payroll Employment and Benchmark Revision.  Headline payrolls rose by 243,000 in January 2012, following a revised 203,000 December gain.  Year-to-year gain in the series rose to 1.51% in January, up from 1.35% in December.  These results were in the context of the 2011 annual benchmark revision, where unadjusted March 2011 payrolls were revised upward by 162,000 (as opposed to an initial estimate of 192,000).  That upside revision was expanded to 231,000 as of December 2011, based on Bureau of Labor Statistics (BLS) imputations, with a related increase to the upside birth-death bias factors, which look like they will total about 580,000 in the current year.

The benchmark revision was relatively small, and despite the upside revisions to recent payroll numbers, and the stronger-than-usual monthly growth in January 2012, January’s payroll still is shy of the peak employment level seen going into the 2001 recession, eleven years ago.  As shown in the following graph, though, today’s level is closing in, just 120,000 jobs shy of the February 2001 level.

 

 

In the first graph following, the magnitude of the benchmark revision is evident in the difference between the plots of the new series (blue line) and the payroll levels as reported last month (red line).

The second graph shows the same detail in terms of year-to-year change.

The scope of the 2011 benchmark revision, which has been known for six months, indeed was reasonably shallow.  There is no change in the general economic outlook.  Revisions of the last several years have tended to be negative and more meaningful, as is shown in the third graph, which shows the relative patterns of the 2008, 2009 and 2010 benchmark revisions.

January 2012’s headline 8.3% unemployment rate was in the context of the annual population revisions.  As such, it is not directly comparable with Decembers 8.5% unemployment rate, but it will be used as a comparable number going forward.  The new population data suggested a lower labor force participation rate, with implications for a growing discouraged-worker population.  The U.6 unemployment rate was at 15.1%, the SGS-Alternate was at 22.5%.

 

Chart from Last Year’s 2010 Benchmark Revision

 

PCE Deflator.  The Fed’s newly-targeted inflation number showed 2.4% annual inflation as of December 2011.  That was enough above the 2.0% target that the Fed might be making noises about tightening, if it really were concerned about containing consumer inflation.

M3.  Year-to-year change in the SGS-Ongoing M3 Estimate appears to be on track to hit 3.9% in January 2012, up from 3.1% in December.  With monthly change also now positive, the broad money supply number may be worth keeping an eye on in the next several months.

 

Hyperinflation Watch.   Beyond the issues covered in today’s Commentary, the hyperinflation outlook just was updated fully in Hyperinflation 2012.

Money Supply M3 (January 2012).  Based on roughly three weeks of data, the preliminary estimate of annual growth for the January 2012 SGS Ongoing-M3 Estimate—to be published tomorrow (February 4th) in the Alternate Data section—is on track to hit 3.9%.  That would be the strongest monthly year-to-year growth in 30 months, since August 2009, and it would be up from the revised annual 3.1% (previously 3.2%) growth estimated for December (revisions include the Fed’s benchmark revisions to underlying components).  The increase in annual growth reflects increasingly positive month-to-month change in the seasonally-adjusted monthly average levels of December 2011 and January 2012, versus flat-to-minus changes in the same periods the year before.

The seasonally-adjusted, month-to-month change estimated for January 2012 M3 likely will be around 0.8%, against a 0.3% monthly gain in December.  The estimated month-to-month M3 changes, however, remain less reliable than the estimates of annual growth.

Rising monthly estimates of annual gain, and in month-to-month change, appear to have been the circumstance in January for the narrower M1 and M2 measures (M2 includes M1, M3 includes M2).  M2 for January is on track to show year-to-year growth of about 10.2%, versus 9.6% in December, with month-to-month growth estimated at roughly 1.3% in January, versus 0.5% in December.  The early estimate on M1 for January shows year-to-year growth of roughly 19.2%, up from 18.1% in December, with month-to-month change showing a likely gain of about 2.4% in January, versus 0.7% in December.  The relatively stronger annual growth rates in M1 and M2 still reflect an earlier shifting of funds out of M3 accounts into M1 and M2 accounts.

 

 

__________

 

 

REPORTING DETAIL

EMPLOYMENT AND UNEMPLOYMENT (January 2012)

The Economy Still Is Not Recovering.  As discussed in the Opening Comments and Executive Summary, irrespective of the nature of the factors that helped to boost today’s labor data, the general outlook has not changed.  Even after the 2011 upside benchmark revisions, the January 2012 payroll employment level remains below the level that preceded the 2001 recession, more than a decade ago.

PAYROLL SURVEY DETAIL.  The BLS reported today (February 3rd) a statistically-significant, seasonally-adjusted January 2012 month-to-month payroll employment gain of 243,000 (a gain of 509,000 jobs before the benchmark revisions) +/- 129,000 (95% confidence interval).  December payrolls showed a deceptively close-to-unrevised 203,000 gain (previously a gain of 200,000), all in the context of the annual benchmark revision and revised seasonal factors that shifted relative monthly growth into December, masking other changes.  The seasonal-adjustment error that overstated December’s gain last month, however, was corrected, so that gain could be counted for a second time in the revamped January data.  See the Opening Comments and Executive Summary for a discussion of the revisions and ongoing seasonal-adjustment problems.

In terms of the actual benchmark revision, the benchmark March 2011 month was revised higher by 162,000 (not the 192,000 estimated initially), on a not-seasonally-adjusted basis.  That change was imputed back in time on a monthly basis by the BLS back to the last benchmark of March 2010, and imputed forward in time so that the unadjusted December 2011 data were revised high by 231,000.  The birth-death model annual upside bias, accordingly, was revised higher by about 70,000 jobs.  Current data next will be revamped, based on the March 2012 benchmark.

In terms of year-to-year change, the unadjusted January 2012 annual growth rate was 1.51%, up versus the benchmark-revised December 2011 growth rate of 1.35% (previously 1.28%).

The following graphs of year-to-year unadjusted payroll change include the benchmark revision (see the Opening Comments and Executive Summary section for revision-specific graphs).  Having shown a rising trend in annual growth through 2010, which primarily reflected the still-protracted bottom-bouncing in the payroll series, that pattern of growth flattened out in 2011, as shown in the first graph of the near-term detail in year-to-year change.  These numbers recently have reflected short-lived year-to-year distortions as a result of the year-ago hiring surge and then full layoffs of temporary census workers.  The January number has shown somewhat higher relative growth, but that circumstance likely will be short-lived.

As shown in the longer-term graph (historical detail back to World War II), with the bottom-bouncing of recent years, current annual growth has recovered from the post-World War II record 5.06% decline in August 2009, which was the most severe annual contraction seen since the production shutdown at the end of World War II (a trough of a 7.59% annual contraction in September 1945).  Disallowing the post-war shutdown as a normal business cycle, the August 2009 annual decline remains the worst since the Great Depression, yet the current level of employment is far from any recovery.

 

 

 

The regular graph of seasonally-adjusted payroll levels and a graph of revision detail, which still show that the current employment level is at pre-2001recession levels, are located in the Opening Comments and Executive Summary section.

Concurrent Seasonal Factor Distortions.  As discussed in prior writings (see Hyperinflation 2012, for example), seasonal-factor estimation for most economic series has been distorted severely by the extreme depth and duration of the economic contraction.  These distortions are exacerbated for payroll employment data based on the BLS’s monthly seasonal-factor re-estimations and lack of full reporting.

Where the BLS recalculates the monthly seasonal factors each month for payroll employment, going back a number of years, it usually only publishes revised data for the last two months of reporting.  This month’s benchmark revision, in theory, includes a full update of the revised concurrent seasonally-adjusted data (actually it is off by a month or two).  Accordingly, the regular graph here is skipped this month, with the plot of new monthly distortions due to begin with next month’s reporting.

Nonetheless, a further big issue remains that the month-to-month seasonally-adjusted payroll data have become increasingly worthless, with reporting errors likely now well beyond the official 95% confidence interval of +/- 129,000 jobs in the reported monthly payroll change.  Yet the media and the markets tout the data as meaningful, usually without question or qualification.

Birth-Death/Bias Factor Adjustment.  Despite the ongoing and regular overstatement of monthly payroll employment—as evidenced usually by regular and massive, annual downward benchmark revisions—the BLS generally upped its monthly biases in post-benchmark reportings of recent years.  Based on the just-published benchmark revision, the BLS appears to have upped what now is a 490,000 jobs annual upside fudge factor for the twelve months ended February 2011, to about 580,000 as of revamped reporting.  The January 2012 bias was a negative 367,000, versus a minus 339,000 in the revamped January 2011, and versus a revamped December 2011 negative bias of 1,000 (previous an 11,000 monthly subtraction).

The aggregated upside annual bias still reflects an ongoing assumption of a net positive jobs creation by new companies versus those going out business.  Such becomes a self-fulfilling system, as the upside biases boost reporting for financial-market and political needs, with relatively good headline data, while often also setting downside benchmark revisions for the next year, which traditionally are ignored by the media and the politicians.  Where the BLS cannot measure the impact of jobs loss and jobs creation from employers starting up or going out of business, on a timely basis (within at least five years, if ever), such information is estimated by the addition of a bias-factor generated by the Birth-Death Model (a model of the effects of new business creation and old business bankruptcies).  The fundamental defects of the Birth-Death Model are discussed as usual in the ensuing paragraphs.

Positive assumptions—commonly built into government statistical reporting and modeling—can become self-fulfilling prophesies, with “stronger” economic data being reported as a result of happy guesstimates, or underlying assumptions of ongoing economic recovery.  Indeed, historically, the Birth-Death Model biases have tended to overstate payroll employment levels—to understate employment declines—during recessions.   There is a faulty underlying premise here that jobs created by start-up companies in this downturn have more than offset jobs lost by companies going out of business.  So, if a company fails to report its payrolls because it has gone out of business, the BLS assumes it still has its previously-reported employees and adjusts those numbers for the trend in the company’s industry. 

Further, the presumed net additional “surplus” jobs created by start-up firms, get added on to the payroll estimates each month as a special add-factor.  These add-factors are set now to add an average of 48,000 jobs per month in the current year.  With the economy continuing to falter, however, I expect a significant downside benchmark revision for next year (March 2012), given current details of the BLS’s overly positive estimates.

HOUSEHOLD SURVEY DETAILS.  The once statistically-sounder household survey, which counts the number of people with jobs, as opposed to the payroll survey that counts the number of jobs (including multiple job holders more than once) just went through its annual discontinuity manipulations, with the BLS revamping population estimates.  Accordingly, comparative monthly changes in key factors such as employment are not meaningful, based on the reported hard data.  The changed data, which reflect both 2010 census readings and all-important government assumptions, are highly suspect both as to quality as well as to touted effects. 

One always has to be suspect of unusually large changes in data, which show up with systemic revisions and which theoretically have been neutered statistically so as to not be distorted by the revisions.  Consider that the revamped January universe showed an incredible seasonally-adjusted 847,000 monthly jump in household-survey employment.  Net of population control effects, the suggested 631,000 employment gain was equally unbelievable.

More importantly, the new population basis suggested an additional 1,252,000 people not in the labor force, with a resulting 0.3% reduction in the labor force participation rate.  While the numbers have been juggled to neuter the impact on the headline unemployment rate, the implications are that the long-term discouraged worker problem is larger than it had appeared previously.  The issues here suggest that the headline 8.3% unemployment for January has moved well outside the realm of common experience and credibility, into the arena of election-year political shenanigans.

Separately, severe issues with monthly seasonal factors continue and still meaningfully cloud the significance of the reported monthly levels in the adjusted headline U.3 unemployment rate and other adjusted household-survey numbers.  The data have been shifted by highly unstable seasonal factors that are artifacts of the severe and extraordinarily protracted downturn in U.S. economic activity (as well as distortions created by last year’s census hiring and firing effects), not by the regular and stable seasonal patterns that were in place before the current economic crisis.  Unlike the payroll or establishment series, the household survey does not use the concurrent seasonal factor adjustment series.  The household series, however, does go through other revisions and distortions, as partially seen above.

Unemployment Rates.  The January 2011 seasonally-adjusted headline (U.3) unemployment rate was reported at 8.26% (purportedly 8.25% net of population control effects).  The December headline unemployment rate was published at 8.51%.  The numbers are highly suspect, politically, and are subject to massive seasonal-factor distortions beyond the January population control effects.  The January and December numbers likely are neither comparable nor accurate. See the Opening Comments and Executive Summary. 

As has been the recent trend, the indicated reduction the January headline unemployment rate would reflect an ongoing flow of unemployed workers moving to discouraged-worker status and moving out of the headline labor force.

On an unadjusted basis, January’s U.3 unemployment was 8.8%, versus a non-comparable 8.3% in December.

Heavily skewed by bad seasonal factors and the population controls, the January U.6 unemployment rate notched lower to a seasonally-adjusted 15.1% from 15.2% in December.  The unadjusted U.6 rate jumped to 16.2% from 15.2% in December.

The broadest unemployment rate published by the BLS, U.6 includes accounting for those marginally attached to the labor force (including short-term discouraged workers) and those who are employed part-time for economic reasons (they cannot find a full-time job).

In 1994, during the Clinton Administration, “discouraged workers”—those who had given up looking for a job because there were no jobs to be had—were redefined so as to be counted only if they had been “discouraged” for less than a year.  This time qualification defined away the long-term discouraged workers.  The remaining short-term discouraged workers (less than one year) are included in U.6. 

 

 

Adding the SGS estimate of excluded long-term discouraged workers back into the total unemployed and labor force, unemployment—more in line with common experience as estimated by the SGS-Alternate Unemployment Measure—notched higher to 22.5% in January from 22.4% in December.  The SGS estimate generally is built on top of the official U.6 reporting, and tends to follow its relative monthly movements and annual seasonal-factor revisions. Accordingly, it will suffer some of the current seasonal-adjustment woes afflicting the base series, such as the November and December distortions in part-time employment for economic reasons.  There still should be catch-up reporting the months ahead, since the series and underlying do not use the concurrent-seasonal-factor methodology that is used with the payroll employment series. 

Nonetheless, there continues to be a noticeable divergence in the SGS series versus U.6.  The reason for this is that U.6, again, only includes discouraged workers who have been discouraged for less than a year.  As the discouraged-worker status ages, those that go beyond one year, fall off the government counting, and new workers enter “discouraged” status.  Accordingly, with the continual rollover, the headline workers flow into the short-term discouraged workers counted in U.6 continue, and from U.6 into long-term discouraged worker status (SGS Measure) at what appears to be an accelerating pace.  See the Alternate Data tab for more detail.

As discussed in previous writings, an unemployment rate above 22% might raise questions in terms of a comparison with the purported peak unemployment in the Great Depression (1933) of 25%.  The SGS level likely is about as bad as the peak unemployment seen in the 1973 to 1975 recession.  The Great Depression unemployment rate was estimated well after the fact, with 27% of those employed working on farms.  Today, less that 2% work on farms.  Accordingly, for purposes of Great Depression comparison, I would look at the estimated peak nonfarm unemployment rate in 1933 of 34% to 35%.

 

PCE DEFLATOR (December 2011)

Paying lip-service to its “inflation-containment” mission, the Federal Open Market Committee (FOMC) announced on January 25th that it would target year-to-year inflation—as measured by change in the PCE deflator—at 2.0%.  The PCE deflator is the heavily massaged and modeled inflation rate for personal consumption expenditure, published on a monthly basis by the Bureau of Economic Analysis (BEA), and quarterly as part of the GDP release.  The monthly series, which is a surrogate measure of consumer inflation—fully substitution and hedonic-based—currently is yielding the lowest annual consumer inflation rate of the major series (see the following graphs of the PCE deflator versus the CPI-U and the C-CPI-U. 

Unlike the more widely followed CPI-U measure, which never is revised and is published on a seasonally unadjusted-basis, the PCE deflator is heavily revised for at least two years following initial reporting, and it is available only on a heavily-massaged, seasonally-adjusted basis.  At subscriber request, we begin regular coverage of the PCE deflator.

In theory the PCE deflator measure should run close to the chain-weighted C-CPI (C-CPI-U) (see Commentary No. 413 for details of the C-CPI-U and other inflation measures).  Yet, the December 2011 PCE deflator showed 2.4% inflation, with the C-CPI-U at 3.3%, the CPI-U at 3.4%, CPI-W at 3.8% and the SGS-Alternate (1980-Base) at 11.0%. 

As reported January 30th, by the BEA, the month-to-month change in the seasonally-adjusted December 2011 PCE deflator was a gain of 0.07% (up 1.10% net of prior period revisions), versus an unchanged November (previously was a decline of 0.04%).  Year-to-year inflation (the Fed’s presumed target item) was 2.40% in December, down from a revised 2.59% (previously 2.54%) in November.

Instead of suggesting a boost in interest rates, which would be the normal response to inflation being above the targeted 2.0%, the Fed keeps promising low rates into the foreseeable future, along with suggestions of possible renewed purchases of U.S. Treasury debt, actions that run counter to containing inflation.  Again, this “inflation targeting” move primarily is pabulum for those in the markets who think the Fed really would move to contain inflation at the cost of impairing already-fragile banking-system solvency.  The Fed’s primary function remains keeping the banking system afloat, at any cost.

 

 

Week Ahead.  Recognition of an intensifying double-dip recession as well as an escalating inflation problem still is sporadic.  The political system would like to see the issues disappear until after the 2012 election; the media does its best to avoid publicizing unhappy economic news; and the financial markets will do their best to avoid recognition of the problems for as long as possible, problems that have horrendous implications for the markets and for systemic stability. 

Until such time as financial-market expectations move to catch up fully with underlying reality, or underlying reality catches up with the markets, reporting generally will continue to show higher-than-expected inflation and weaker-than-expected economic results in the months and year ahead.  Increasingly, previously unreported economic weakness should show up in prior-period revisions. 

U.S. Trade Balance (December and Annual 2011).  Due for release on Friday, February 10th, the December trade deficit likely will widen beyond consensus estimates.  To the extent that the report is much worse than expected, the first revision to the fourth-quarter 2011 GDP estimate (due February 29th) would be pressured towards the downside, and vice versa.  The annual shortfall in the headline trade number likely will top $560 billion for 2012, versus a deficit of $500 billion in 2011 and $381 billion in 2010.  A widening trade deficit is a direct reduction to broad U.S. economic activity.

 

 

__________