The nine-day teacher strike in West Virginia may soon be followed by walk-outs in Oklahoma and Arizona over low wages and potential increases in the cost of benefits.

We shouldn’t be surprised. For more than 20 years, teacher salaries have not even been keeping up with the cost of inflation.

West Virginia, which has become a national flashpoint, makes for a good case study, and one available in my recent report for EdChoice, “Back to the Staffing Surge.” Like virtually all states, West Virginia has significantly increased its public school expenditures in recent decades. Using publicly available data, since 1992, and adjusting for inflation, West Virginia public schools have increased spending by 39 percent per student to $12,512 per student by 2014, while average teacher salaries fell by 3 percent during that time period.

So while school districts had cash to spend, teachers were seeing money actually leave their wallets. So where did that additional money go?

Like public schools in virtually all states, West Virginia public schools increased its staffing of non-teachers far in excess of what was needed to accommodate changes in enrollment, which fell by 12 percent between 1992 and 2015. Despite the decline in student population, West Virginia public schools increased employment of non-teaching staff by 10 percent during this 23-year period, from 17,533 to 20,029—an increase of almost 2,500 personnel as the West Virginia public schools saw their student population decline by almost 40,000 students.  These increases include new assistant principals, district officials, curriculum specialists and teacher aides.

What could West Virginia public schools have done with $232 million in annual savings? For starters, they could have given all of their teachers an increase in compensation of $11,620 per year—much more than the teachers recently received.

Succinctly put, taxpayers spent a lot more per student, but teachers did not get a raise relative to the cost of living. Instead, schools hired more non-teachers—while student enrollment was falling.  As of 2015, West Virginia public schools had significantly more staffing—of both teachers and non-teachers—relative to the national average.

In West Virginia alone, a cautious estimate of the cost of this increase in non-teaching staff is more than $232 million annually, out of a total K-12 education budget of $3.5 billion, of which about one-third goes to teacher pay. What could West Virginia public schools have done with $232 million in annual savings? For starters, they could have given all of their teachers an increase in compensation of $11,620 per year—much more than the teachers recently received.

In my research, I have found that the surge of non-teaching staff in public schools has been occurring across America since at least 1950 and, at least since 1992, to no measurable positive relationship between this increased staffing and student outcomes. National Assessment of Education Progress (NAEP) Long Term Trend scores for 17-year old public school students fell by three points (Reading) and were literally flat (Mathematics) during the post-1992 staffing surge.  Studies have even found that children today are more advantaged than students decades ago, negating any argument that schools need more staff because students are “harder” to educate now.

When the staffing surge is pointed out, policymakers place blame at different levels of government. In reality, all three levels of government have been responsible for this staffing surge. Levels and rates of change in staffing vary widely across (and within) states—indicating the staffing surge was not entirely due to federal or state mandates.

Shoring up short-term salary increases for our educators is a great first step, but one that doesn’t get to the root of the problem.

Now is the time to begin a serious policy conversation not just about how much they make, but about how we got here in the first place. Whether it’s West Virginia, Oklahoma, Arizona or elsewhere, it’s time for public school advocates to tell the truth about real increases in taxpayer resources and where they’ve been spent.

Part of that conversation is looking at outcomes for students – there is a lack of a relationship between these staffing increases and student achievement – and identifying some of the fixes that could address the underlying problems. States should look to expand school choice and create incentives for schools to compete for teacher talent and students in a more open marketplace.

Empowering parents to choose the best educational fit for their kids—and allowing the funding that goes with those students to follow them—puts those families, not bureaucrats, in the driver’s seat in deciding where and how school dollars are spent. And if schools aren’t paying teachers what they believe they deserve, those teachers should have the right to teach elsewhere—and be paid for their hard work.

The West Virginia teachers have been languishing in a system that for so long has been rigged against them. It’s time to fix the system to help them, and the students they teach.