This is a guest post by Michael Lilley, Founder and President of the Sunlight Policy Center of New Jersey, which seeks to educate residents on the way NJEA “uses their tax dollars against their own interests, secures benefits for the few over the many, and blocks needed reforms.” This was originally published on SPCNJ’s website.
New Jersey public schools are among the best in the nation and our teachers are the primary reason for that. They should be compensated fairly. But teachers are ill-served by flawed research.
Mark Weber, Ph.D., of New Jersey Policy Perspective (NJPP) has come out with a report detailing a “wage gap” between New Jersey teachers and private-sector workers. Unfortunately for New Jersey’s teachers, Dr. Weber does not make the case well. He relies on faulty methodologies that do not properly account for significant factors in his calculations. He cites studies that are irrelevant to his thesis. He assumes facts that cannot be assumed. And he cherry-picks data, prominently highlighting data that support his case and burying or ignoring data that doesn’t. There may be a case to make that there is a wage gap, but Weber does not make it.
Faulty Methodologies
Weber calculates that there is an hourly wage gap of 14.5 percent between New Jersey teachers and non-teachers with bachelor’s degrees.
But wage gaps do not measure total compensation. Non-wage compensation like health and retirement benefits have considerable value. Weber acknowledges that “teachers tend to have more generous benefit packages” than non-teachers, and the Economic Policy Institute (EPI) researchers he cites all quantify these benefits and take them into account to determine a “compensation gap.” Compensation gaps are the more accurate measure.
Weber attempts to account for these superior benefits to arrive at a compensation gap but he does not quantify their value. Instead, his sole source of relevant data is EPI. But other researchers believe that EPI’s methodology greatly understates their value.
Andrew Biggs and Jason Richwine point out that the data the EPI uses comes from the Bureau of Labor Statistics’ Employer Costs for Employee Compensation (ECEC) survey. The ECEC values pensions based on the cost to the employer (in New Jersey’s case, the state government). But an employer’s pension contribution in a given year “bears no legal or mathematical relation to the pension benefits a teacher accrues in that year … Whether the government contributes nothing or an infinite amount, the benefits accruing to employees do not change.”
New Jersey provides the perfect case-in-point. It is widely known that New Jersey has woefully underfunded its pensions. The New Jersey Pension and Health Benefits Study Commission determined that through 2015, the state only made 18.5 percent of the full amount required to fund the state’s public pensions. By EPI’s methodology, the value of a pension to a teacher would be only 18.5 percent of what the teacher is owed. But by law New Jersey actually owes 100 percent and every teacher rightly expects nothing less than 100 percent. In New Jersey’s case, EPI severely understates the value of pensions to teachers, and so does Weber.
In addition, New Jersey teachers are enrolled in defined benefit pensions plans in which the state guarantees a fixed payment every year of retirement. Such defined benefit pension plans are very rare in the private sector because they are so risky and expensive for employers. Private-sector workers’ retirement funds will be determined by the performance of their 401ks. Nothing is guaranteed. Clearly, a state-guaranteed annuity is worth more than a 401k. Biggs and Richwine found that if the true value of state-guaranteed defined benefit pension obligations is taken into account, the cost would rise from 11 percent of wages to 32 percent of wages. That 21 percent jump erases any overall compensation gap presented by Weber.
- Although Weber acknowledges the existence of Biggs and Richwine’s research, he does not address the substance of it. A proper analysis would address and presumably refute it.
Irrelevant Sources
As for his other sources for valuing benefits, Weber quotes extensively from a 2014 NJSpotlight report by Mark Magyar. Magyar makes two comparisons between what New Jersey public employees pay for their benefits and what workers for other state governments pay. But these comparisons are irrelevant to Weber’s study of the compensation gap between New Jersey teachers and private-sector workers.
When actually making comparisons to private-sector workers, Magyar notes that New Jersey public employees pay more for individual health insurance than private-sector workers but also pay LESS for family coverage (the plan was deliberately designed this way to help families and their larger healthcare bills). For the purposes of Weber’s research, these findings are ambiguous and not quantified, and in any event are limited to what employees pay for their health benefits. They do not address the overall value of the benefits, which would account for employee contributions but would also account for economic value of the health benefits provided to the employees. The quantified values of the overall benefits would be the relevant data points in Weber’s discussion of compensation gaps. But Weber does not provide these.
Finally, Weber cites an NJPP analysis that found that New Jersey’s public pensions rank 95th in “generosity” among the country’s 100 largest public plans. This comparison is made with other state pension plans and is irrelevant to a comparison with New Jersey private-sector workers. As is Weber’s final point about how much New Jersey spends on instruction-related staff benefits compared to other state governments.
- Citing irrelevant research obfuscates rather than illuminates.
Assuming Facts that Cannot Be Assumed
Even taken on its own terms, the most important factor in Weber’s case for a wage gap is the value of academic degrees. Because almost all teachers have bachelor degrees, they are compared to private-sector workers with bachelor’s degrees in Weber’s and EPI’s regressions. But not all bachelor’s degrees have the same economic value.
Biggs and Richwine note that roughly half of all teachers have bachelor’s degrees in education, which have substantially less economic value than engineering and science degrees. A recent Bankrate study found that the top 50 most remunerative college majors ALL came from science, technology, engineering, business and finance. The top 12 – all engineering and science majors – averaged a median income of $87,200 per year. The 12 education-related degrees ranked from 83rdto 135th (out of 162 majors) and averaged a median income of $46,700 – 46 percent less than what the top 12 degrees earned. Clearly, the market differentiates between the economic value of these degrees. Yet Weber and the EPI studies he cites just assume that they are equal and make no effort to adjust for the differing economic values.
- So all the analyses that undergird Weber’s conclusions about a wage gap wrongly assume that all bachelor’s degrees have equal economic value. Can these be regarded as valid conclusions?
Cherry-Picking Data
Weber cites a long-standing series of EPI national studies that echo his conclusions about New Jersey’s wage and compensation gaps. For example, Weber cites a 2019 EPI report by Sylvia Allegretto and Lawrence Mishel which found a 21.4 percent wage gap and a 13.1 percent compensation gap. All these EPI reports use weekly wages rather than hourly wages, not accounting for the fact that teachers work fewer hours a week than private-sector workers.
This is an important difference. Weber cites Keefe’s report that found that teachers had a weekly wage gap of 16.8 percent and a 12.5 percent compensation gap. Yet the same report found that the gap shrinks to 9.4 percent when based on hourly compensation rather than weekly compensation. Weber does not mention this. Furthermore, Keefe finds that the hourly compensation differential between public school teachers and other full-time workers who both live and work in New Jersey (as most teachers do) shrinks further to 6.8 percent. Weber does not mention this either. But the bottom line is that EPI reports that use weekly data are overstating the wage gap.
Remarkably, using the hourly data for New Jersey that Weber ignores, the Keefe report shows that if academic degrees are excluded there is no compensation gap at all. Using data from 2012-2014, Keefe determined that teachers worked 89 percent of the hours per year other full-time employees worked (1,980 versus 2,227). Other full-time employees earned $106,912 in total compensation versus $95,703 for teachers. But 89 percent of $106,912 is $95,090, which is $613 LESS than teachers were actually compensated. So, per Keefe, teachers are actually compensated MORE per hour than other full-time employees.
To his credit, Weber provides both weekly (19.4 percent) and hourly wage gaps (14.5 percent), a 4.9 percent difference. But in both cases, Weber includes private-school teachers (who are paid less) in his teacher data, acknowledging that this likely “makes the gap between teachers and other workers larger.” Only in the Appendix do we learn by how much: 3.6 to 7.8 percent, according to NJDOE. So looking at public school teachers, which is what Weber is focused on, the hourly wage gap shrinks to 6.7 to 10.9 percent. Per Keefe, this would erase the 6.8 percent hourly wage gap between teachers and full-time workers who live and work in New Jersey.
Even more remarkably, in the 2019 EPI report cited by Weber, Allegretto and Mishel indicate that New Jersey teachers likely have a wage ADVANTAGE over private-sector workers. In the main body of his report, Weber cites their headline of a 21.4 percent wage gap nationally but buries the New Jersey-specific data on the very last page of his Appendix, dismissing it as an “outlier.” No wonder he buries it: Allegretto and Mishel determine that New Jersey’s weekly (not hourly) wage gap is a mere 3.8 percent. Per Weber, the difference between weekly and hourly wage gaps in New Jersey is 4.9 percent, so on an hourly basis, New Jersey teachers actually have a 1.1 percent wage ADVANTAGE.
In addition, Keefe’s New Jersey-specific data shows that New Jersey teachers have an even greater total compensation advantage. Keefe finds that the value of benefits gives teachers an additional 4.3 percent in total compensation. So, according to the very reports that Weber cites and using his own calculations, in 2018 New Jersey teachers received total hourly compensation that is 5.4 percent greater than private-sector workers.
- Weber presents conclusory headlines about wage and compensation gaps and cherry-picks the data to support his case. But a closer look at the research he cites thoroughly undermines his case.
Weber’s research on a wage gap in New Jersey is deeply flawed. He relies on faulty methodologies that do not properly account for significant factors in his calculations (pensions). He cites studies that are irrelevant to his thesis (Magyar, NJPP). He assumes facts that cannot be assumed (value of academic degrees). And he cherry-picks data, prominently highlighting data that support his case and burying or ignoring data that doesn’t (Allegretto and Mishel, Keefe).
Perhaps Weber or some other NJPP researcher will undertake the task of actually proving there is a wage gap. But this report does not make that case. New Jersey teachers will have to wait.
1 Comment