Written by Steven Hansen
ADP reported May 2012 non-farm private jobs growth at 133,000. The market expected 157 to 165K (versus the 133K reported). These numbers are all seasonally adjusted.
There is no question that any jobs growth is positive. However, this is the second month in a row of low jobs growth – yet this 133,000 growth is on the theoretical numbers for the current economic conditions. Econintersect believes ADP is the best real time indicator of jobs growth. Jobs growth has been over potential until recently.
This is the second month in a row of jobs growth under 150,000 (after the previous five months were over 150,000). Jobs growth of 150,000 or more is calculated by Econintersect to the minimum jobs growth to support population growth (see caveats below).
This jobs growth data continues to be anti-recessionary, although not strongly pro-recovery. Employment is a rear view indicator, and looking at this ADP data – the averaged growth rate trend is essentially flat since November 2011.
The above graph shows that employment growth is better than last month. The overall trend for the rate of growth had been slightly improving – and one month of bad data does not establish a new trend.
Small and medium sized business historically create most of the new jobs (analysis here) when using the ADP data. A continuing take from the ADP data is that small and medium size business continue to be the employment driver. However, the BLS totally disagrees (see caveats below).
The relative strength is in the service sector which literally provided almost all the jobs growth in 2011. Manufacturing jobs contracted this month for the second month, while the goods production sector is weak at adding jobs.
In our April 2012 economic forecast released in late April, we estimated non-farm payroll growth at 130,000 – all based on historical pressures from an economy running at near potential. But as explained in Ben Bernanke and the Puzzle of Employment, our fudge factored estimate would have been a growth of 204K.
Caveats on the Use of ADP Employment Data
Historically employment is the confirmation that real economic growth is occurring. As background, many economic factors impact jobs growth. How many jobs businesses create in any one month is not directly dependent on these economic factors, but on individual decisions. The impact of all the economic factors is averaged out over many months.
ADP tends to revise slightly their data one month after issuance, and does an annual revision of employment similar to what is done with the BLS employment data.
On March 7, 2012, Automatic Data Processing, Inc., in conjunction with Macroeconomic Advisers, LLC, will publish annual revisions of the estimates of employment shown in the ADP National Employment Report. These revisions will reflect:
- Updated regression estimates used to adjust for historical differences in the variances of the monthly growth rates of employment reported by the Bureau of Labor Statistics and computed from ADP data;
- Updated estimates of the differences in the historical averages of the growth rates of employment reported by the Bureau of Labor Statistics and computed from ADP data;
- Updated estimates of historical seasonal factors;
- The reference month for establishing the levels of employment shown in the ADP National Employment Report will be advanced from March 2010 to March 2011; and
- There are no revisions to the methodology of computing the estimates of employment shown in the ADP National Employment Report.
This report was developed and maintained by Macroeconomic Advisers, LLC.
It is a measure of employment derived from an anonymous subset of roughly 500,000 U. S. business clients. During 2010, this subset averaged about 337,000 U.S. business clients and represented over 21 million U.S. employees working in all private industrial sectors.
The data ……. is collected for pay periods that can be interpolated to include the week of the 12th of each month, and processed with statistical methodologies similar to those used by the U.S. Bureau of Labor Statistics to compute employment from its monthly survey of establishments. Due to this processing, this subset is modified to make it indicative of national employment levels; therefore, the resulting employment changes computed for the ADP National Employment Report are not representative of changes in ADP’s total base of U.S. business clients.
Basically this employment index is designed to mimic BLS private non-farm employment – which does not include government employment. The headline BLS Employment Report includes government employment.
Econintersect believes the simplistic sampling extrapolation technique of ADP yields a far better picture of the employment situation in real time than Bureau of Labor Statistics (BLS) methodology. Although the BLS employment numbers eventually are correct, their data gathering technique does not support the quick release schedule.
Because of the differences in methodology, many pundits ignore the ADP numbers. Although there can be a low correlation in a particular month, the different methodologies tend to balance out, and the correlations are excellent outside of the data turning points. We are now 22 months past the post recession turning point in employment. Based on current estimates, BLS non-farm private jobs growth in 2011 was 1,920,000 while ADP was 1,925,000. The difference is insignificant.
There is now a known disconnect between BLS and ADP on the breakdown of jobs growth between small, medium and large business – as the graphs below illustrate. Basically the BLS sees large business as the employment driver, while ADP sees small and medium size business as the employment driver. For full details please read: BLS Experiments With New Data Series: Now I Am Confused.
There is the proverbial question on what is minimal jobs growth each month required to allow for new entrants to the market. Depending on mindset, this answer varies. According to Investopdia, the number is between 100,000 and 150,000. The Wall Street Journal is citing 125K. Mark Zandi said 150K. Econintersect is going with Mark Zandi’s number:
- If Econintersect used employment / population ratios to determine the number, the exact number seems to be between 140,000 and 160,000. The graph below uses the historical employment-population ratios to show jobs growth per month if the population was 300 million.
- If Econintersect uses employment – population ratios, the correct number would be the number where this ratio improved. Using the graph below, the ratio began to improve starting a little after mid-year. This corresponds to the period where the 12 month rolling average of job gains hit 150,000.
The following graph (not seasonally adjusted non-farm private payroll) shows that most of the employment growth are in the first half of the year, and there is little real growth of employment in the second half of the year. Therefore seasonal adjustment algorithms understate employment growth in the first half of the year, and overstate growth in the second half of the year.