When we first saw the 38,000 print, we thought it was an error with our news feed. But no, the U.S. economy added only 38,000 non-farm jobs in May, the lowest since September 2010, and the biggest miss on record. It's really hard to overstate how morbidly bad May's payrolls were. It is, however, commensurately easy to kiss a June rate hike by the Fed a sweet goodbye.
Everyone was shocked. We were definitely shell shocked, because we couldn't believe just how bad the numbers were. Neither could the markets.
Flash crashes and surges sprouted the instant the first HFT algos recognized that it was really an epic disaster. Yes, here we are 5 days late to reporting on this monumental event; but you see, we have more to bring to the table, now that 3 trading days have already passed for the dust to settle.
Before we even go into the details of what is a very shocking development to the entire Fed tightening game, we want readers to put this entire edifice in proper context. To appreciate just how wildly the Fed has swayed the market's rate hike expectations, try to follow this series of chronological events:
- First it was the April FOMC statement that left the door open for a June hike;
- Then the markets braced for the April NFP report which was released in early May;
- April's payrolls missed big, busting June rate hike expectations;
- But the April FOMC meeting minutes, which were released later in May, seemed to indicate willingness to hike in June, once again raising expectations for a June hike.
If you're still not tired of all this ping pong action, we shall overtly state that we are. While we cannot speak for the rest of the markets, the folks we talk to indeed share the same sentiments — enough with this nonsense, the market will go wherever it wants to go, and the Fed will do whatever it wants to do. Hike or no hike; in June, July, September, or when the cows come home.
Now that we've gotten that out of the way, let's dive down and get up close and personal with the details.
According to the BLS, May saw a total of 38,000 non-farm jobs added to the economy, against expectations of 164,000. This is largely down from April's 124,000 (revised lower from 160,000), and March's 186,000 (revised lower from 208,000).
- 38,000 non-farm jobs added in May vs. 164,000 expected
- U3 unemployment rate down from April's 5% at 4.7% vs. 4.9% expected
- April's 160,000 figure was revised sharply lower to 124,000, and March's 208,000 figure was revised lower to 186,000; totaling net revision of -59,000
- 25,000 private jobs added in May vs. 152,000 expected, down from April's 130,000 (revised sharply lower from 171,000)
- Average hourly earnings rose +0.2% MoM vs. +0.2% expected, down from +0.3% in April
- Average weekly hours at 34.4 vs. 34.5 expected, down from April's 34.5
Our take & market reactions
Sometimes, when payroll reports come out bad, but not mega bad (such as April's, which was actually terrible by itself, but nothing compared to May's), they leave some wriggle room for market interpretations.
However, this report was so ugly that we feel it leaves almost no room for an alternative interpretation. Specifically, one of forgetting about a June rate hike. The Fed is almost certainly not going to hike later this month.
Everyone (including the BLS) talked about how the massive strike by Verizon workers contributed to May's depressing print. However, do understand that consensus forecasts had already factored in the impact of that strike which was widely documented, and was likely not undermined. We therefore do not believe that May's report was a complete outlier.
The 0.3% decline in the U3 unemployment rate should not be read positively. The rate fell due to an exodus of Americans from the labor force, causing the participation rate to fall once again to some 40-year lows, sending the number of people not in the labor force to another record high.
The modest level of wage inflation doesn't even iron out the pages in this badly frayed book. The fact that Fed Chair Janet Yellen had to take to the podium on Monday urging calm in the markets, is a testimony in and of itself to the magnitude of this disaster.
As you will see in the charts below, the markets went into the second half of May with heightened hopes for a June or July hike. We have talked a lot about such dynamics and will therefore not be repetitious here. Nothing much has changed as these dynamics are still the same.
Rate hike odds have crashed to almost nothing for June — 4% currently (was 2% at the lows of last Friday). July odds have fallen from highs of 48% to 36% last Friday. This places more weight on September, but it's going to be much tricker and precarious to hike then, mostly because of the U.S. Presidential Elections.
The Fed mentioned that June's meeting would be "live"; that means taking into account May's payrolls disaster. Only a mindless FOMC member would vote to raise rates in such a cataclysm. We now officially do not see a June hike at all. We feel that a July hike is unlikely too, but aren't fully convicted yet.
Translating rate hike odds into moves in various markets, the initial (acute) knee jerk was massive dollar weakness (which persisted till the close of Friday), risk aversion (stocks, carry, credit all down), and anti dollar assets outperform majorly (precious metals, emerging currencies, and commodities especially crude).
As we've shown in the above chart, the follow through (as of Tuesday) has mainly been concentrated to a weak dollar, and strong anti-dollar assets. WTI is pushing $50, the highest since July 2015, while gold and silver are attacking key technical resistances. Risk off flows mostly dissipated by Monday, and most equity indices are up strongly from Friday's lows.
In the immediate aftermath of April's payrolls, we said the following:
"For now, we are still sticking to our bearish dollar bias but do so with extra caution...
We are ready to flip to being bullish the dollar for the medium term contingent on how the market trades the following week. Do not however be mistaken, macro and fundamentals point to a weaker and not a stronger dollar."
It turns out that we were pretty accurate on our take.
With the addition of May's payrolls, we are outright bearish the U.S. dollar and will be looking for good opportunities to gain some short exposure. We are of course fully ready to stop dead in our tracks and turn on a dime, as must be expected when dealing with such troll-worthy elements.
Post analysis note from Goldman Sachs:
Weak May Payroll Report Likely to Stay Fed’s Hand in June
- Nonfarm payroll employment increased by just 38k in May, far below consensus expectations.
- The unemployment rate fell to a new cyclical low due to a drop in labor force participation.
- In light of the weaker-than-expected employment report, we have revised our subjective odds of the timing of the next FOMC rate increase.
- We now see probabilities of 0% for June, 40% for July, and 30% for September.
- Nonfarm payroll employment increased by just 38k in May, well-below consensus expectations for a 160k increase. Earlier months were also revised down by a net 59k, making the result much weaker than expected overall. Employment in telecommunications related jobs declined by 37k, reflecting the strike at Verizon Communications. However, the weakness extended much beyond that category. Goods-producing employment fell by 36k—the largest decline of the recovery—with declines in each of its major components (mining, construction and manufacturing). Excluding the Verizon strike, employment in service-providing industries increased by 98k, down from 144k in April. Temporary help services jobs contracted by 21k, and growth in trade/transportation/utilities was unchanged. A small number of categories improved: education and health employment gained a solid 67k and government employment rebounded, with a 13k following a 7k drop in April.
- Average hourly earnings increased by 0.2% (mom), or 2.5% from a year earlier. The 12-month gain was above our expectations due to upward revisions to prior months. The average workweek was unchanged at 34.4 hours.
- The household survey measure of employment gained just 26k after a decline of 316k in April. However, because of another decline in labor force participation, the unemployment rate fell to a cycle low of 4.7% (4.692% unrounded). At this level the unemployment rate is now in line with Fed officials median projection for end-2016, and below their longer-run estimate (from the March FOMC meeting).
- In light of the weaker-than-expected employment report, we have revised our subjective odds of the timing of the next FOMC rate increase. We now need see probabilities of 0% for June, 40% for July, and 30% for September. Although the report lowers the odds of near-term action, in our view, it also arguably raises the range of possible outcomes. If employment growth rebounds next month but the unemployment rate remains low, the case for hiking after June would become quite strong. Alternatively, if sluggish employment growth were to persist, the FOMC could remain on hold for longer than we currently expect.