Is A Bad NFP Print Days Away - Goldman Says Warm Weather Added 70,000-100,000 Jobs; Now It's Payback Time

Three months ago, this site was the first to discuss the impact of abnormally high temperatures on "better than expected" economic data, which the mainstream media in its perpetual permabullish bias attributed to economic "growth", and not even to $1.3 trillion in ECB liquidity, which today even the ECB's Constancio admitted was nothing but QE: "The purpose of the European Central Bank's two three-year longer-term refinancing operations was to address banks' short-term funding issues and "nothing else." "The sole aim of the LTRO was to cater to the funding stress of euro area banks in general," Constancio said at a colloquium on macro-prudential regulation here. "It never crossed our minds that we were solving the sovereign debt crisis" with these measures. Hence QE, albeit masked by worthless collateral exchange to make the naive Germans believe the ECB was not outright printing money. It was. Now that the 'economy', and by that we mean the stock market of course, is finally turning over, the topic of the weather will start being far more prominently featured, as there will have to be a validation to unleash QE at either the April or the June FOMC meeting (something which the Chairman hinted at on Monday, and which Bill Gross has been saying for months). Why blame it on the weather of course. It is in this context that we show the latest Goldman Sachs economic outlook piece from Zach Pandl who now states that "unseasonably warm temperatures have lifted the level of nonfarm payrolls by 70,000-100,000 as of February." Call it erroneous seasonal adjustments (as we have for the past two months), call it a trigger happy BLS, or just call it people leaving their home more than if there was 6 feet of snow outside, the point is that now up to 100,000 jobs will have to be "given back." Which in turn means that next Friday's NFP forecast of +213K may just end up being as low as 113K, with the print coming just in time for the Chairman to commence warming up the printers, and soon enough to where more QE will give the president the sufficient bounce in stocks he needs to mask the debt ceiling breach in September.

How does Goldman get to its estimate? Why by using something called "state-level panel regressions." Obviously.

The advantage of the panel approach is that by using the cross sectional information as well as variation over time, we will have a much richer dataset and hopefully more precise coefficient estimates. We created two versions of the model: one in which we controlled only for state-specific factors (e.g. faster or slower trend growth) and another in which we controlled for both state- and time-specific effects (e.g. the state of the business cycle and average weather across the nation).


Encouragingly, the coefficient estimates are very similar to the national results: a one standard deviation shock to the HDD variable implies around 30,000 additional payrolls. The main difference with the national regression is that the statistical significance is much stronger (with t-statistics for the contemporaneous HDD variable around seven instead of three). The estimated equations imply a weather impact on the level of payroll employment of about 70,000-100,000 as of February.

In other words, everyone got confused by the weather and assumed it was a symptom of a healthy economy. It wasn't. And now... it's payback time.

The “payback” from this weather related boost could be a drag on payroll employment growth in March and April.

What else does Goldman find?

As a crosscheck to these results, we also tested whether the states in which warm temperatures matter most have also been the states with accelerating job growth in recent months. First, we re-estimated the state-level panel regressions with the states grouped into the nine official Census Divisions (with divisional and period fixed effects). Second, we compared the estimated coefficients of the weather effect by Census Division to changes in employment growth in recent months.


As expected, changes in temperatures have much different effects across regions. Weather has the biggest impact on the “West North Central” region, which includes Minnesota, North and South Dakota, and Kansas. Changes in the weather, at least as measured by the HDD index, have a minimal impact on the Pacific region (Alaska, California, Oregon and Washington). As shown in the chart below, the regions with the largest (most negative) weather effect coefficients have also generally seen faster payroll employment growth recently (the acceleration in payroll growth is measured as the monthly average change in December and January compared to the average over the previous three months).


The state level evidence therefore seems reasonably clear that unseasonably warm weather lifted employment growth over the past few months. However, changes in temperatures cannot account for all (or even most) of the acceleration in job growth. Using our highest estimate of the weather effect, average monthly payroll growth over the last three months ex-weather would have been about 210,000 instead of the reported 245,000. This lower figure still represents a significant acceleration from a three-month average growth rate of 157,000 in November.

What does this mean? Well for one thing, that the market is now about 200 points higher than where it should be if global warming was not a factor. Because as a reminder here is what January and February NFP prints were:

  • February: 227K, Expectations 210K (source) + 17K Beat
  • January: 243K, Expectations 225K (source) + 18K Beat
  • December: 200K, Expectations 150,000 (source) + 50K Beat

In other words, combined the NFP beats of the winter, excluding the favorable impact of the weather, would have been a series of meets, if not outright misses. Whether or not that would have resulted in a lower market is unknown: after all it may have simply made the market soar even more on greater QE demands and expectations. Frankly who cares. It is all centrally planned now.

However, now the time has come for the payback. Especially since Goldman says so. And remember that Goldman, along with all the other banks, demands QE. However for that to happen, not only will the economy have to slow down drastically, but stocks to drop materially.

Next Friday's NFP miss will be just the first part of the plan. As for the second one, why we are sure that Brian Sack has something up his sleeve as usual.