A Wandering May


“Rough winds do shake the darling buds of May.” – William Shakespeare, Sonnets

The radio news reports from Friday and over the weekend would have you believe that the stock market rose that day in response to the April jobs report. Too bad that wasn’t the case.

The stock market reaction wasn’t in complete isolation from the jobs release, of course – had the number been a shocker instead of a mild shortfall (160,000 vs. an expected 200,000), trading would surely have proceeded down a different path. But the number wasn’t much of a surprise after the private-sector ADP payroll report had checked in with its own estimate of 156,000 two days earlier. While the ADP report is never taken as a substitute for the real thing, it does serve to modify expectations and has been on something of a hot streak the last half-year or so, with predictions quite close to the Bureau of Labor Services (BLS) number that it tries to anticipate.

It was a lackluster report for a lackluster week that allowed some lackluster trading volume to push things around. Equity futures did rise from a loss Friday morning to a gain in the pre-market hour after the report’s release, but prices began falling almost immediately after the open. So traders did what they do – stepped back and allowed the S&P 500 to bounce off of its 50-day moving average (2040), which it did at 11 AM, or two and a half hours after the number and ninety minutes after the open. Without some serious bad news, a technical bounce was a virtual lock and so the index duly rebounded.

At 1:45 PM, heading into the home stretch of the weekend, the indices were flat on the day. Then someone or some group started throwing big SPY (S&P 500 ETF) buy orders at the tape, and in thirty minutes the index rallied another eight points. Here’s a hint – it wasn’t because somebody suddenly realized that wage growth was 0.3% or some other spurious data point. Stocks went sideways from there, fading in the last hour until the usual Friday fillip of the last ten-fifteen minutes that pushes the tape a little higher. So markets did indeed close higher on the day, but for reasons that were unrelated to the news about the jobs report – apart from the fact that it wasn’t a stunner. The actual rise was all about tape-positioning and nothing more.

It’s nothing to call in the authorities about. Stocks are an auction market and within some broad limitations, buyers are free to buy what and when they want. Consider such behavior as akin to a poker game – the other players can’t control the cards that you get or the amounts that you want to put down, but they can bluff you and some can bet more than you can afford to lose.

Earnings for the S&P as a whole in the first quarter now appear set to come in with a decline of between six and seven percent from the year-ago quarter. The second quarter estimate is currently creeping towards a decline of about five percent, but that will probably grow to between eight and ten percent by the end of the current quarter.

The economy and earnings have been weakening, but employment is still hanging on and until a recession hits, the lure of more juice to be squeezed from central banks should help markets keep from sinking too deep. Sentiment has gotten fairly negative of late, so the market is starting to set up for another bout of buying, but the impetus from the February rebound is spent and rally traction will be difficult to come by. First I think we are more likely to test the 150-day average at 2020, as there isn’t much to trade on during the coming week that is likely to be good. Only experienced short-term traders should be thinking about putting any new money to work in the current market.

The Economic Beat

The report of the week was the jobs report, naturally, with its below-estimate report of 160,000 new jobs (seasonally adjusted, or SA) simultaneously fanning fears that the economy isn’t doing so great after all, and hopes that perhaps another number like it a month from now will keep the Fed on hold at its June meeting. That left the markets in a wandering frame of mind.

I should add that the report probably didn’t do much to change minds – the fears that it fanned were mostly existing ones, and ditto for the hopes that it raised. You get a lot of conflicting sentiment near the ends of business cycles (or if you prefer, the “mature leg” of the cycle). The report itself was mixed in many ways, so it’s hardly a surprise that markets were uncertain – what’s more, the number closely tracked the ADP estimate from two days prior, muting the surprise effect.

In fact, there was mixed news all across the report. The April release included downward revisions to March and February that reduced their combined total by 19,000 (19K). Downward revisions are bad in trader lore, with many considering revision direction more meaningful than initial estimates. But the revisions were entirely seasonal adjustment-based, as the unadjusted total estimate for the two months actually rose by a little over 10K.

One item that I would point to as being of some concern is the job replacement rate over the first four months. What I mean by that is that every January, about 2% of the establishment payroll count (unadjusted) disappears through a combination of layoffs, job changing, retirement and so on. Seasonal adjustments smooth out the annual process, and data geeks like me track how many months it takes for the new year to surpass the end-of-year total. If the total isn’t positive by May, it’s recession time.

The good news is that it would take a really terrible May number not to get to positive territory on the year, and there isn’t any current sign of that happening. The bad news is that the data are weak through April – not terribly weak, but the weakest since 2010, when annual job growth finally resumed after the recession. Not as weak as April 2007 either, when recession was clearly on the horizon (though Fed-bettors outnumbered the recession bettors), and not great either.

CNBC hastened to put lipstick on the pig, blaming the shortfall on the psyches of businessmen frightened by the bust of selling that began the year (all misleading, you understand) and roiled markets. Confidently hoping that growth will resume shortly, by next week they will be parading a long line of dinosaurs and reports that will be blaming it all on the government again and assuring us that there is nothing wrong that can’t be fixed by cutting taxes and doing away with regulations.

On the positive side, hourly wages grew by 0.3% and the average workweek ticked back up a tenth to 34.5, though that’s also where it was this time a year ago. The unemployment rate remained steady at 5.0%. Private-sector employment increased by 171K.

The negatives were more numerous. Average weekly earnings are up only 2.5% year-on-year. The unemployment rate owed its success to another big jump (562K) in the “not-in-labor-force” (NILF) category. The employment total in the household survey, used to calculate the unemployment rate, actually fell by 362K. The labor force fell also and the participation rate dropped two-tenths back to 62.8%, compared to a year-ago 62.7%. In sum, employment is still growing, but at a slower and slower rate. That is also starting to show up in weekly claims, which had a big jump last week. One week is nothing to hang your hat on, but a look at the year-to-date comparisons indicate the party for claims is nearly over (it’s been good for an unusually long time, so we shouldn’t complain).

News from the manufacturing sector was tepid. The ISM survey had a flat reading of 50.8 (50 is neutral), down slightly from the April result of 51.5. The growth-contraction score was good, at 11-4 (3 unchanged) and the responder comment section, not exactly overflowing with enthusiasm, showed more optimism than worry. The April factory orders report came in with a positive read of 1.1%, but it was all due to defense spending and March was revised down further to a loss of 1.9%. The service sector survey, by contrast, was fine at 55.7 and generally better than March across the board. Spring cometh, I suppose, with the growth-contraction score a healthy 13-4. Construction spending appears to be up about 8% year-on-year, or at least somewhere in that neighborhood. The revisions to the monthly data tend to be so large (February was revised from a loss of 0.5% to a gain of 1%!) that caution is always recommended.

The full international trade report was a little better than expected, but mostly because of falling import demand, not a good sign. Chain-store sales reports point to another weak monthly report for retail sales in April, with the full report due next Friday. Truck sales that are probably tied to the better-than-usual start to the residential construction season (weather-related) held up the monthly vehicle sales data in April, but car sales are weakening.

Next week features retail sales on Friday, with the other report of interest being the wholesale trade report on Tuesday, though the market doesn’t give it much heed. The otherwise quiet week will feature import-export price data on Thursday and producer inflation on Friday, along with business inventories.

share this blog post
Facebook Twitter Plusone Linkedin Digg Delicious Reddit Stumbleupon Tumblr Posterous