Friday, October 10, 2014

Moody's Zandi: "Have We Underestimated U.S. Wage Growth?"

by Calculated Risk on 10/10/2014 02:47:00 PM

My view is that there is still significant slack in the labor market. Evidence of slack includes the elevated unemployment rate at 5.9%, the elevated level of U-6 at 11.8% (an alternative measure of labor underutilization), the large number of people working part time for economic reasons (included in U-6), and the low level of wage growth.

Some new research from Moody's Mark Zandi suggests wage growth might be picking up: Have We Underestimated U.S. Wage Growth?  A few excerpts:

[W]hat if wage growth is accelerating already, and the BLS wage measures have yet to pick this up? This is the message in new data from ADP, based on payroll processing records for more than 24 million employees, or about one-fifth of all U.S. workers.

Moody’s Analytics helped ADP separate the changes in hourly wages paid to those staying in their jobs, labeled job holders, from wages paid to those who change jobs, new entrants to the workforce, and those leaving it. The data track changes quarterly from the second quarter of 2011 to the third quarter of 2014, long enough allow for seasonal adjustments. The data can also be broken down by industry, region, company size, worker age and gender, tenure on the job, pay scale, and part- vs. full-time.

The hourly wage rate for job holders is the most telling. It is up 4.5% from a year ago in the third quarter, a strong and steady acceleration from its low two years ago. The acceleration in hourly wage growth occurs across the board, although it is up most for younger workers, those with one to five years on the job and at lower pay levels, and those who work at small companies. Wage gains have also picked up most in financial services and construction in the West and South.

Baby boomers who work at big companies in healthcare and in leisure and hospitality in the Northeast and Midwest have experienced the slowest acceleration in wages.

The ADP-based hourly wage data likely overstate the acceleration in wage growth in the broader labor market for several reasons. First, ADP’s client companies tend to perform better than average, particularly among smaller companies. A small firm will not use a payroll processing service unless its prospects are good.

How much this inflates the ADP data has not been estimated, but judging from the impact on employment, which we have quantified in our work estimating payroll employment (the ADP National Employment Report), it is meaningful. This may suggest that the ADP data foreshadow broader trends if conditions continue to improve.

Second, the ADP hourly wage for job holders more accurately measures workers’ base pay. This is unlike other BLS wage measures, which include other forms of compensation. The BLS average hourly earnings gauge includes pay for overtime hours and incentive pay (though not irregular bonuses). In 2011 and early 2012, overtime hours were increasing as was incentive pay. This supported growth in the BLS average hourly earnings metric, but not in the ADP measure. Indeed, ADP hourly wage growth lagged BLS hourly earnings growth during this period.

During the past two years, overtime hours and incentive pay have leveled off, weighing on the BLS hourly earnings growth gauge, but not on ADP’s base hourly wages measure. This effect should wear off soon, however, and stronger growth in base pay, which is evident in the ADP data, should appear as well in the BLS data.

The increase in personal income tax rates at the start of 2013 also affected the timing of some workers’ income. Businesses moved income into 2012 to benefit from the lower tax rate, reducing income afterward. This likely had a bigger impact on the BLS measures of labor compensation than on ADP’s base wage rates for job holders. Yet this impact too should also fade quickly.

Third, the significant increase in the number of job leavers and new entrants since early 2013 may also be weighing more heavily on BLS wage measures. In the third quarter of this year, there were approximately 400,000 more job leavers and 500,000 more new entrants than in the first quarter of 2013. Since new entrants are paid less than those at the ends of their careers (a large share of the leavers), measured wage growth has been depressed. While this also affects growth in ADP’s hourly wages for job holders, it does so later than it affects the BLS wage measure.

Implications

If the acceleration in ADP hourly wages presages an imminent acceleration in broader measures of labor compensation, the implications are substantial. Most encouragingly, it signals that workers will finally participate more equitably in the benefits of the economic recovery.