I feel there needs to be something to clarify today's employment report. Also about how the unemployment rate could rise while employment also rose (it isn't because of some meteoric rise in the labor force). As everybody has already heard non farm payroll rose 308,000 in March, and the unemployment rate ticked up to 5.7% from 5.6%. February was also revised to a 46,000 job gain and January was pushed up to 159,000 (giving the first quarter 513,000 jobs).
The numbers were not cooked by the people at the BLS. The people there do good work, and I don't know of a case of that ever happening, but I'm not old enough to remember too many years ago, so I could be wrong.
The 308,00 number is the seasonally adjusted number, meaning that month to month changes that usually occur are factored out. On a not seasonally adjusted basis, meaning the literal job growth, was 1,007,000. Yes, over a million jobs. However, the seasonal adjustments are done by looking at the previous years, so it removes the average seasonal factors (it's done though auto-regression). Seasonal adjustments are not perfect. However, don't discount jobs because they may have a seasonal factor. A heavy Christmas season might have required more workers. Or good weather might have allowed more work to happen. Those both have a seasonal component, but are just as real as any other job. (But if that good weather is just pushing work into this month and will subtract from summer hiring, that isn't really netting jobs.)
Hiring has been across a wide range of sectors. The BLS's diffusion index of employment measures the breadth of labor change. Is jumped to 61.0 from February's 51.4. It hasn't been that high since July of 2000.
The payroll employment number comes from the establishment survey and the unemployment comes from the household survey. They both count slightly things and both have different methodologies. They are both "correct" in terms of what they count. However, in the last year the two have been diverged and talked about a lot. This month, even with the 308,000 job growth in the payroll numbers, the household survey showed a drop in employment of 3,000 people (might as well call it flat since this only says that there is a 90% chance employment growth was between -293,000 and +287,000). However, the labor force grew by about 179,000 people (once against that is a 95% confidence +/-290,000). With 179,000 more people looking for work and 3,000 fewer people employed according to the household numbers, that means 182,000 more unemployed people. 8,352,000 total unemployed in a labor force of 146,650,000 gives an unemployment rate of 5.6951%. That gets rounded up to 5.7%, so it "essentially unchanged" from last month. Last month the employment participation rate (the number people either employed or looking for work divided by the number of potential workers) was 65.87%. This month the number of potential workers grew by 193,000. To keep the same level of participation 127,133 people needed to join the labor force. So the 179,000 is slightly more, causing the unemployment to rise, but it doesn't account for all of it, and it isn't that huge. So the unemployment rate went up because of rounding, a drop in employment, and slightly more people entering the workforce than the population grew.
74,000 jobs were not from the end of the supermarket strike. Nobody I know of has made any prediction on how much of an increase can be attributed to that. While 74,000 supermarket workers did return to work, tens of thousands of people employed to fill in were also released. The number of new jobs from this could be quite low and difficult to tell.
And all the jobs added were not of poor quality. There was 277,000 private sectory jobs. Of the 31,000 government jobs, federal employment fell by a 1,000, while state and local education jobs rose by 25,000. In the private sector some of the biggest gains were 71,000 in construction; 42,000 in professional and business services; and 39,000 in education and health services. While some of the construction jobs might be because of almost an early summer in March, the largest subgroup gain was 50,000 in specialty contractors. That would include some like roofing and house painting that are weather related, but it would also include tile, electrical, plumbing, and a number of other jobs that would be less influeced by sunny skies.
Pres. Bush is still 1.8 million jobs in the red. That means a little over 250,000 a month before November to get back in the black or 200,000 a month to avoid being placed in the same category of Herbert Hoover.
Bonds took an absolute beating today. The 10-year fell, moving the yield up almost 30 basis points last time I checked (0.3 percentage points). The 10-year TIPS tanked similary rising 20 basis points.
It is just one month (well, given the revisions call it 1½). So don't get too happy (or upset for some of you).