The most important economic news this week is Friday's employment report from the Bureau of Labor Statistics. This monthly report contains a wealth of data for economists, probably the most significant in the near term being the month-over-month change in Total Nonfarm Employment (the PAYEMS series in the FRED repository).
Yesterday we had the February estimates from ADP (198,000 new jobs) and TrimTabs (100,000 new jobs). This is a rather jaw-dropping spread. How will Friday's BLS number, the one that really matters, compare?
I took a few minutes yesterday morning to create a visualization of the three series over the past twelve months, ending with the ADP and TrimTabs estimates for February.
As the chart above illustrates, neither series has a particularly strong correlation with the government's data, although ADP has a better record. To be specific, over the past 12 months, the TrimTabs:BLS correlation is 0.54 and the ADP:BLS correlation is 0.79.
What's particularly striking in yesterday's ADP and TrimTabs estimates is the astonishing spread. We don't need a calculator to tell us that ADP's 198K is 98% higher than the TrimTabs number. In fact, over the illustrated timeframe, this is the widest spread between the two, second place going to the 92% spread for February 2012. One year ago the BLS number came in about 8% below the ADP estimate but 77% above the TrimTabs' number.
As a forecast for the February BLS nonfarm employment number, the economy and the market are no doubt pulling for ADP.
For a sense of the critical importance of nonfarm employment for the economy, see my Big Four Economic Indicators, which I'll be updating on Friday after the BLS employment report on February data is published.