KEY DATA: ADP: 169,000/ Help Wanted: -104,500/ Productivity: -1.9%; Labor Costs: +5%

IN A NUTSHELL: “As we get conflicting information on the labor market, the Fed members have to feel very conflicted about when to start raising rates.”

WHAT IT MEANS: I had been hoping that the April data would start showing a clear trend, but alas, that is not yet the case. Decent if not solid Institute for Supply Management reports seemed to point to better economic conditions, including job gains, and that made sense given the low level of unemployment claims. But the ADP estimate of private sector job growth was well below expectations. There were reasonable increases in small business payrolls, moderate gains at mid-sized firms but almost no hiring at large companies. While we all know that the energy sector is cutting back, it is hardly clear that many of the layoffs have already occurred. Notices usually have to be given first. And the ADP report indicated that large service firms didn’t hire that many workers, which doesn’t match the ISM report. That said, the ADP estimate raises serious questions about the strength of Friday’s employment report.

Adding to the uncertainty about the April jobs numbers was the Conference Board’s Help Wanted Online index. There was a decent sized decrease in the number of want ads posted online. These data do bounce around on a monthly basis, so maybe that is all that was going on. Also, it is not clear the reason for the drop. Was it a result of a softer economy? If so, Friday’s number may be disappointing. Or, did firms finally start actually start filling all those open positions? The number of new ads was down much less than the total number of ads, which could mean HR is catching up. That suggests a better than expected jobs reports. I am not certain what Friday’s number will look like, but I am sticking with my view that it will be much, much better than the ADP report indicates.

Meanwhile, labor costs keep rising. Unit labor costs, which adjust compensation for productivity, jumped in the first quarter. GDP growth was weak (and the revisions should show it to have been even weaker), so it was no surprise that with more people working, output per worker would decline. But the jump in labor costs also came from a strong rise in compensation. That is a factor in the Fed’s decision calculus.

MARKETS AND FED POLICY IMPLICATIONS: While it is mostly about jobs, it isn’t all about jobs anymore. Compensation matters. Since the first quarter growth number was likely an anomaly, we can dismiss the productivity weakness. But it looks like compensation is accelerating and that is crucial for the Fed. The rising labor costs will not only put pressure on firms have to raise prices, but growing incomes should increases demand, providing a greater ability to pass those higher costs through. So watch on Friday not only the number of jobs and the unemployment rate, but also the wage data. Indeed, investors may look past a mediocre payroll number if the wage increase is sharp, since that could influence the Fed the most.