(Editor’s note: This story was authored by Jon Greenberg of nonprofit, nonpartisan news organization PolitiFact and was published April 4 on Oklahoma Watch, also a non-profit journalism organization. The story appears here in accordance with Oklahoma Watch’s republishing terms.)
Teachers in about 70 Oklahoma school districts staged a week of walkouts over low pay and slender school budgets. A common talking point is that they are paid far less than teachers nationwide.
Lawmakers and the governor agreed that education spending should rise after many years of austerity, passing a tax increase to help cover a salary hike of about $6,100 a year. Many teachers, however, say they’re looking for a long-term commitment to wages and school support.
Our recent fact-check confirmed that Oklahoma teacher salaries rank near the bottom, but one point deserves a closer look: What’s the best way to compare teacher pay from state to state?
The Oklahoma Council of Public Affairs, a free-market, small-government oriented think tank, argued that, in reality, Oklahoma’s teachers are doing about average.
“Every recent analysis shows that Oklahoma teacher pay jumps to between 30th and 35th in the nation once cost of living is considered,” wrote executive vice president Trent England.
England cited three analyses, two of which explained how they reached their results, and their methods were similar. (We did not hear back about the third.)
The 1889 Institute and EdSurge used salary estimates from the National Center for Education Statistics, part of the U.S. Education Department, and a cost of living index from the Missouri Department of Economic Development. From these, they produced adjusted average salaries and put them in order.
The 1889 Institute analysis ranked Oklahoma 30th, and the EdSurge one put Oklahoma at 35th. (Differences as small as $20 or $30 can shift the rankings slightly.)
But labor and education economists say it’s not that simple.
The people who study education labor markets raise big issues with using cost of living adjustments this way.
But before we go there, it’s worth noting that there’s an urban tilt to the cost of living measure used by the 1889 Institute and EdSurge. And they aren’t the only ones who rely on that index: An education think tank called EdBuild, as cited by NPR, followed the same path.
Behind all three salary estimates is the work of the Council for Community and Economic Research, a subscriber-based research group.
The Council for Community and Economic Research said on its website that it aims to answer, “How do urban areas compare in the cost of maintaining a standard of living appropriate for moderately affluent professional and managerial households?”
So, the index is not only about urban dwellers but more specifically the costs faced by those who are more financially comfortable.
The council goes on to warn:
There’s no doubt that small rural places offer an alternative to an urban professional or managerial standard of living that many people find attractive, but such places are qualitatively different from urban areas, and they simply don’t support the kind of urban lifestyle embodied in the Cost of Living Index.
With about 33 percent of its population living in rural areas, Oklahoma is more rural than two-thirds of the rest of the country. So a cost of living index based on professionals living in cities is not a good fit, because the labor market is different.
What it takes to hire a good teacher
A statewide cost of living adjustment misses the fundamental challenge faced by school districts across the state, economists told us.
The question, according to economist Christiana Stoddard at Montana State University, is one of attracting and holding talent.
“If school districts want to have the same ability to compete for teachers, how much do they need to pay?” Stoddard said.
For labor markets, location matters — and generally speaking, cities have an advantage. Whether it’s nightlife or proximity to great skiing, amenities can have an impact on salaries. Cities are more expensive, but in fact, several studies show that people are willing to give up a bit of pay to live there. (This can be true even if urban pay is higher once experience and other factors are taken into account.)
In the real world of hiring, rural schools might have a low cost of living, but that hardly makes them competitive.
“They are remote, far from hospitals and shopping centers,” Stoddard said. “They have a much harder time recruiting even though it doesn’t cost that much to buy a house there.”
To get the same quality teacher as an urban school, a rural school district would actually need to pay a little extra or pay the same for a less competitive teacher. Stoddard said looking at the cost of living pushes attention away from that reality.
“If you apply statewide cost of living to a rural area in Oklahoma, you might conclude that teachers should be paid less,” she said.
Economist Lori Taylor, professor and director of the Robert A. Mosbacher Institute for Trade, Economics and Public Policy at Texas A&M University, also noted the amenity gap issue, and pointed to another fundamental problem with the underlying data.
“It is not appropriate to compare average teacher salaries, because average salaries reflect not only differences in the wage level, but also differences in the mix of teacher credentials,” Taylor said.
For instance, Taylor warned that not all states have the same percentage of teachers with master’s degrees, but the national data miss that.
“I would be very careful in drawing conclusions from such data,” she said.
Another data problem is inconsistent accounting for health and retirement benefits in the state-level statistics.
A more comprehensive approach
In 2015, economists Dan Rickman, John Winters and Hongbo Wang at Oklahoma State University brought together many of the key factors into one big computer analysis. They controlled for teacher characteristics, regional price differences, pay, the amenities of different places, tax rates and metropolitan vs. non-metropolitan settings.
Using data from 2009 to 2011, they looked at details from the lower-48 states.
“Where we adjust for everything, Oklahoma ranks 44th out of 48 states,” said Rickman.
The state’s rank did bounce around. It started at 47, then moved to 45 based on regional price adjustments from the Bureau of Economic Analysis. After accounting for teacher characteristics and the salaries of other professionals, the state rose to No. 40. But when the metropolitan/non-metropolitan factor was added, the state’s rank fell to 44.
“The gist of our paper is there is no way to compare Oklahoma teachers salaries to those of other states without concluding that the teacher salaries in Oklahoma are low or near the bottom,” Rickman said. “We debunk all the popular myths for the differences.”
This analysis is a bit out of date and excludes Hawaii and Alaska. (Hawaii ranks very low in simple cost of living-adjusted pay, which invites the question about what would happen when amenities were factored in). But Rickman said, and the 1889 Institute analysis concurred, that Oklahoma teacher pay has lost more ground since 2010 than for teachers in most other states.
So, using current numbers would change the specifics but not the overall findings.
Temptation of cost of living adjustments
The calculations that take average teacher salary and divide it by a correction factor are easy to do and understand. For sure, if a teacher were weighing competing offers from Kansas City and Oklahoma City, the standard cost of living indexes would be helpful.
Applying that formula to statewide rankings, however, misses many factors. Cost of living overlooks the fact that part of the compensation package for working in a city is the city itself.
States with larger rural populations face different issues than more urban states, and that plays out in teacher pay. It takes more than a simple formula to sort that out.
(PolitiFact is a nonprofit, Pulitzer Prize-winning news organization that specializes in fact-checking; it is owned by the Poynter Institute of Media Studies, based in St. Petersburg, Florida. Reach the reporter at firstname.lastname@example.org. Email suggestions to email@example.com.)