U.S. May non-farm payrolls may be a calmer affair after April shock
The May U.S. non-farm payroll report on Friday may be a much less volatile affair than last month, when shock news of 288,000 new jobs topped even the most optimistic views.
This time, there is more certainty around a less spectacular but still solid outcome, based on an analysis of forecasts from the most accurate economists in Reuters polls on recent U.S. data.
The range of views among economists who were closest to reality in forecasting recent key releases on jobs, manufacturing activity, and the magnitude of the contraction in first quarter U.S. GDP is significantly narrower than the range in the wider Reuters poll.
Broadly, this supports expectations for U.S. economic growth to accelerate.
Those top forecasters based on recent data ranged from 200,00 to 230,000, with the Reuters median from 105 economists at 218,000, almost right in the middle. The wider sample is 110,000-325,000.
The latest consensus is the highest in all Reuters polls since January 2008, before the financial crisis.
Last month, the consensus called for 210,000, well below the surprise 288,000 outcome, which marked the strongest net monthly job creation in well over two years.
Almost no economist got even close to that figure, except for Joel Naroff of Naroff Economic Advisors, who said 279,000. He is expecting 219,000 for May, just 1,000 above consensus.
While Naroff missed on first quarter GDP, he got his forecast for the May Institute for Supply Management manufacturing barometer almost exactly right, calling for 55.2, just 0.2 percentage point below the actual number, at 55.4.
I think 240-250 is the trend all the way into the fall and that could even accelerate as we get towards the end of this year. So what I’m essentially looking at is if you start with 280 let’s just say, and we add another 220-230, what you are looking at is about 250 or so. And it’s pretty much where I see the trend at this point.
I’m getting that from a variety of factors – layoffs, most importantly the unemployment claims numbers, which I think are accurately reflecting fewer layoffs and therefore somewhat better growth in job gains.
We are going to see some volatility – a little slower job gains. But still if we come out 220-230 or even 210, and people are disappointed in that, it just shows how far we have come.
I wouldn’t be surprised if the April number is revised to above 300, and if that is the case – and even if we get 200 in May – then it still will be great two months, or I should say a fantastic two months.
For Friday’s NFP release, 89 of 105 economists predict job gains of over 200,000. Only two economists expect May job creation to top the 288,000 added in April.
While markets tend to focus on NFP misses, because that creates volatility where money can be made, a surprising number of consensus forecasts have been extremely close to the actual outcome over the past year.
Two-thirds of the Reuters consensus forecasts published over the last year have been accurate to within at least 32,000 jobs. In March, the consensus was out by only 8,000. Last August it was off by 11,000. And last May, it was off by only 5,000.
That is all the more remarkable given that the margin of error on the survey itself is plus or minus 90,000, according to the Bureau of Labor Statistics.
To be sure, revisions to the April number could change the overall picture, as Naroff says.
But on the basis of the May figure alone, this could be one of those months where it comes in closer to consensus.
— reporting: Rahul Karunakar, Swati Chaturvedi, Deepti Govind, Sumanta Dey, Sarmista Sen