Goldman Sachs forecasts a 200k increase in non-farm payrolls for March – in line with consensus – and believe last month’s 175k print supports the ongoing positive trend (in light of the weather effect). Key employment indicators looked mixed-to-better in March, and despite the continued cold temperatures, less extreme weather conditions overall should give an additional boost to job gains this month. Citi suggests the weather could have knocked 172k off payrolls overall from Dec to Jan and are more hopeful, expecting a 240k print. Their biggest fear, a greater than 275k print (which is the high bar that Joe Lavorgna has set) could see asset markets reacting badly (on the basis of quicker Fed tightening).
As Citi’s Stephen Englander notes,
How much weather in NFP? — 172k
Conveniently the BLS has published February State employment data so we can estimate ‘weather corrected’ national data based on how employment is less affected parts of country evolved. We estimate a variety of models to project national NFP from six states – California, Texas, Washington, Oregon, New Mexico and Arizona. We average their projections for December to February to get a composite estimate of national NFP.
Our estimate of the cumulative weather effect based on the average of these models is 172k over the three affected months (Dec.-Feb)., with the biggest impact by far in December. Figure 1 shows the average estimate of our eight models. The models differ in estimation period, lags, whether explanatory variables are each state or the sum of all the states, among other dimensions. We don’t have enough experience with these models to be comfortable picking a ‘best’ model but there is enough consistency in their estimates to suggest that they are capturing something.
Across the eight models, the minimum weather effect was 70k, the maximum 251k. None were negative and only one of the eight showed a cumulative weather effect of less than 100k. If anything we would think these estimates may be downward biased since some of these states themselves had episodes of bad weather. So the risk in our view lies very much on the strong side of estimates.
Our economists estimate 240k, which relative to a baseline of 190k would represent a recovery of about 30% of the weather-related losses. It still looks to us as if investors are lowballing NFP tomorrow, and that the risk is to the topside (acknowledging that the tail of forecasts is a bit more skewed to the upside than to the downside.)
We keep uncovering evidence that FX crosses are becoming more sensitive to rates moves, particularly in the 2-5 years range so we see upside USD risk, especially versus EUR and JPY, if our weather-effect estimates are correct. We still see a 200k or lower outcome as risk positive, a 240k or higher as risk negative and 200-240k as the grey area. It may take investors a few days to decide whether a weather bounce is good, bad or indifferent for asset markets. Where we see asset markets reacting badly is if we get a really strong print – say 275k+ – which would suggest a strong snapback. Investors would likely view this as an indication of a US economy that cannot wait to burst out of the box. We think this would be risk-negative for asset markets since it would point to a quicker Fed tightening of liquidity. This is clearly a minority view with equities at all-time highs.
Goldman Sachs’ David Mericle is less positive (right at consensus 200k)…
We forecast a 200,000 increase in nonfarm payrolls in March, in line with consensus expectations. We view the reasonably solid February gain of 175,000 despite extremely adverse weather conditions as providing some confirmation that the underlying trend growth rate of payrolls remains solid. Key employment indicators looked mixed-to-better in March, and despite the continued cold temperatures, less extreme weather conditions overall should give an additional boost to job gains this month.
We expect that the unemployment rate declined to 6.6% in March (vs. consensus 6.6%). We also expect that hours worked, which tend to show a larger impact from severe weather conditions, will rebound from their February decline. As the flip side of this rebound in hours, we expect a softer +0.1% gain in average hourly earnings (vs. consensus +0.2%) as last month’s unusually large gain–likely driven by weather distortions–partially reverses.
We forecast a 200,000 increase in nonfarm payrolls in March, in line with the consensus estimate of 200,000. We expect private payrolls increased 195,000 (vs. consensus 200,000). While the average payroll gain seen over the last three months now stands at a disappointing 129,000, adverse weather conditions have weighed heavily on the economic data in recent months, and we would instead view the six-month average of 177,000 or the 12-month average of 180,000 as better approximations of the trend rate of payroll growth. We expect March payrolls gains to come in a bit higher than that, reflecting both the improvement in weather conditions and the month’s mixed-to-better employment indicators summarized below.
Arguing for a stronger report
As we noted earlier this week, despite cold temperatures in March as a whole, weather conditions showed a considerable improvement from February in two senses. First, our rule-of-thumb for the effect of temperatures on payrolls–which places 50% weight on the reference week itself and 25% on each of the two prior weeks–points to a moderate improvement from February to March. Second, there were no major snowstorms in March, while there was a major storm from Tuesday-Friday during the reference week in February and another major storm two weeks before the reference week. We expect weather to provide a roughly 25k boost in March, with some risk of a softer contribution in recognition of the seemingly more-modest-than-expected weather impact on the February report. While weather should provide a boost in March, the month’s colder-than-usual temperatures leave room for additional bounce-back in April, if temperatures normalize.
The four-week moving average of initial claims for unemployment benefits fell 7k to 330k from the February to the March reference week. During the reference week itself, claims dropped to 323k.
Announced layoffs were down 30.2% year-over-year in March after falling 24.4% in February, according to Challenger, Gray, and Christmas. The heaviest job cuts in March were seen in the health care and telecommunications sectors. Challenger noted that the heath care job cuts reflected both lower Medicare reimbursements and layoffs of temporary workers at the end of the sign-up period for health insurance under the Affordable Care Act.
Private job gains reported by ADP rose strongly to 191k in March from an initially-reported February gain of 139k. In addition, the ADP report tends to show less weather impact than the official payrolls report. At their current level, ADP job gains are close to the roughly 200k trend seen in the second half of 2013. That said, we attach only limited weight to the ADP report because its initial print has yet to prove itself as a reliable indicator of payroll job growth as measured by the Labor Department.
Arguing for a weaker report
The labor differential?the difference in the percentage of respondents in the Conference Board’s consumer confidence survey describing jobs as plentiful vs. hard to get?worsened slightly by 0.9pt to -19.9 in March, following four months of consistent improvement. The index has shown a fairly steady recovery since late 2011.
The employment component of the ISM nonmanufacturing index–the single best survey measure of employment growth–recovered most of its sharp February drop, rising 6.1pt in March to 53.6, a level indicating a moderate rate of expansion. However, the employment components of the Richmond Fed and New York Fed service sector surveys showed declines in March. In addition, the employment components of the major manufacturing surveys were also weaker this month, with the ISM manufacturing, Chicago PMI, Philly Fed, and Empire all showing declines but remaining in neutral-to-expansionary territory.
Both new and total online job ads fell substantially in March, to roughly the level seen prior to a spike in February. However, this series tends to be quite volatile and is a forward-looking rather than coincident indicator, meaning that the February jump is likely to have some positive impact on the March data that roughly offsets this month’s decline.
Overall, we view the softer job gains seen this winter as a temporary deviation from a still-strong trend. While weather conditions remained far from normal in March, the improvement from last month should provide at least some boost. As growth accelerates later in 2014, we expect the trend rate of payrolls growth to rise to about 225,000 per month.
We expect that the unemployment rate ticked down to 6.6% (vs. consensus 6.6%) in March from an unrounded 6.72% in February. We also expect average weekly hours to reverse last month’s decline, which was probably weather-related. As the flip side of the rebound in hours, we expect average hourly earnings to post a softer +0.1% gain (vs. consensus +0.2%) in March after a stronger-than-usual +0.4% jump in February. While this strong print led some commentators to suspect that wage growth might be picking up, we suspect that it was largely a statistical artifact caused by the severe weather conditions in February, which probably shifted the composition of the workforce toward salaried and away from hourly workers.