Goldman's Advance Look At The NFP Number, And Why The Firm Continues To Be More Bearish Than Consensus

With just 20 minutes left until the real deal, Goldman highlights its case for ongoing economic weakness, in this case manifesting in an NFP number due at 8:30 am that is about 35K below consensus, at 60K, and Private Payrolls at 75K, also below exp of 80K. The three reasons for Goldman's bearishness: 1. State and local cutbacks, particularly in education; 2. The remaining wind-down of temporary Census employment; and 3. Little change in federal non-Census employment. Will they be right? Check back in 20 minutes to find out.

From Goldman's Andrew Tilton

Still Waiting for Signs of Stronger Hiring (Tilton)

We expect the October employment report to feature a gain of 25,000 in nonfarm payrolls (75,000 in the private sector), a steady unemployment rate at 9.6%, and a meager 0.1% increase in average hourly earnings.

The main culprit in the weak labor market recovery has been slow hiring; layoffs are actually fairly low for this stage in the economic cycle.  At this point we see no strong indications that hiring picked up in October, though the employment report itself will provide the definitive judgment. 

We expect the October employment report to feature a gain of 25,000 in nonfarm payrolls (75,000 in the private sector), a steady unemployment rate at 9.6%, and a meager 0.1% increase in average hourly earnings. This a slightly more cautious forecast than the “consensus” view on payrolls, with the median forecast currently at 60,000 for nonfarm payrolls and 80,000 for the private sector.

The main difference between our view and the consensus view is what we expect to be a continued decline in government employment.  This breaks down into three main components:

1. State and local cutbacks, particularly in education.  The fiscal constraints on many state and local governments are well known, and state and local hiring has slid from an average of 18,000 per month in 2006-2008 to -30,000 per month so far in 2010.  In September, state and local employment fell by 83,000 jobs in September, the largest monthly loss in more than a quarter-century.  The bulk of that decline was in educational employment: although more than a million teachers were hired in September on a non-seasonally adjusted basis, the increase fell short of seasonal norms, and so was reported as a seasonally adjusted decline.  This issue should recur in October – “normal” hiring is more than half a million, and insofar as budget constraints cause the actual increase to be smaller, seasonally adjusted employment should decline.  One factor that could go the other way is the late timing of Labor Day; in years when this occurs some teacher hiring appears to be pushed into October.  But we think the structural issues probably dominate, and so expect a loss on the order of 40,000 state and local jobs.

2. The remaining wind-down of temporary Census employment.  Approximately 6,000 temporary Census workers were still on the government’s payroll as of the September payroll survey week. Essentially all of these should have finished their work by the time of the October payroll survey.

3. Little change in federal non-Census employment.  The average monthly change in federal government employment, excluding the Census, has been almost zero in recent months, and we see no obvious reason for this to change in October.

As for the private sector, the key issue remains the rate of hiring.  The share of workers who are losing jobs seems to be relatively low for this stage of the business cycle (see “Jobless Claims—No Underlying Improvement,” US Daily, October 22, for more details.) 

But hiring is anemic: the latest Job Openings and Labor Turnover Survey (JOLTS) from the Department of Labor showed a “hires rate” (number of new hires as a percentage of total employment) of 3.2% in August, the latest data available.  This is barely above the 3.0% rate at the depths of the recession in late 2008 and early 2009, and well below anything seen in the 2001 recession or its aftermath. Similarly, 46% of all US households view jobs as “hard to get” according to the latest Conference Board survey, with only 3.5% characterizing them as “plentiful”; these figures are not far from their worst levels in the nearly half-century that the survey has been conducted.

Some analysts have suggested that the low rate of hiring may reflect structurally higher unemployment, or a mismatch between job seekers and jobs available.  Indicators of job openings do look somewhat more encouraging than actual hiring.  For example, the JOLTS job openings rate has increased from a low of 1.8% in 2009 to 2.4% currently—back to the trough levels in the aftermath of the 2001 recession, a somewhat bigger improvement than the hiring rate.  But the ratio of the hiring rate to the job opening rate is not unusually low if we compare the history of the two series.  And the various indicators of job openings have not improved materially over the past few months.  For example, the Conference Board online job advertising survey, despite benefiting from a secular shift of advertising from print to online media, has shown very little improvement since the summer, and a similar survey from Monster.com peaked (on a seasonally adjusted basis) back in June.   Rather than mismatch, we think the bigger problem is simply weak economic growth and consequently limited demand for labor.  An acceleration in real GDP growth will ultimately be necessary to drive a substantial increase in hiring. We therefore expect private-sector hiring to be around 75,000 in October, fairly similar to September.

Despite our expectations for mediocre private sector hiring and a decline in government employment, we are not forecasting an increase in the unemployment rate (currently 9.6%).  There are two main reasons for this:

1. Recent outperformance of household employment.  The unemployment rate is based on a survey of households rather than business payrolls. So far in 2010, employment growth as measured in the household survey (about 200,000 jobs per month, after adjusting for population growth and the payroll employment concept) has consistently outperformed the estimate of the payroll survey (68,000 per month).  Possible reasons for this discrepancy include: a) payback for 2009, when the household survey underperformed (though not by as wide a margin), b) the “birth-death model,” which estimates the number of payroll jobs created by new firms and tends to lag behind turns in the economy, c) growth in contract work, which could result in households’ describing themselves as “employed” in the household survey without any new payroll jobs. Insofar as the recent outperformance of the household survey continues—admittedly far from a sure thing—it would reduce the chance of an increase in the unemployment rate.

2. Subdued labor force participation.  The share of the working age population who are employed or looking for work is only 64.7%, just off a 25-year low.  In part this is due to the aging of the population (participation rates peak in the 35-44 age group and fall off substantially beginning in the 55-59 age bracket) and in part it is due to the weakness in the labor market.  Our estimates suggest that after adjusting for demographic trends, the drop in labor force participation is roughly in line with the cyclical decline following prior recessions.  With the usual caveat that anything can happen in a given month, we do not see a strong reason for the participation rate to rebound soon and push the unemployment rate higher.

Comments

[Sudden Debt]

Maybe GS is fired all of their traders which could trash the employment numbers?

[obamaphobe]

Even when gs is wrong, they can frontrun their way to being right.  Today could be ugly.   I don't see any buyers stepping up in front of the report. 

[Robslob]

there are always buyers...even when the masses are selling...China?

[Cojones]

That would mean that BLS lets down Ben B. Ain't gonna happen. I expect better than expected figures with downward revisions next month.

[Cojones]

Quod erat demonstrandum, bitchez.

[scratch_and_sniff]

Typical Goldman...fucking wrong again.

[sodbuster]

LOL! Maybe- maybe not. What really is the point of trying to guess what the numbers will be, from an incompetent and unreliable agency, that uses flawed methodology to begin with? F'n B(L)S.

[Grand Supercycle]

DOW/S&P500/FTSE daily and weekly overbought charts are now at an extreme level. Similar extreme conditions were detected before the correction started in mid 2007.

http://stockmarket618.wordpress.com

[cheap uggs for sale]

Welcome to our website-- http://www.ugghots.com , we are making the promotion for many uggs now.Here is a chance for you that you want to buy good ugg classic boots by cheap price.At present our hot sale snow boots has :
| UGG Delaine Boots
| UGG Gaviota Boots
| UGG Gissella Boots
| UGG Evera Shoes
| UGG Upside Boots