by Dr. Terry Stoops
Former Director of the Center for Effective Education, John Locke Foundation
In 2017, the plaintiffs and defendants in Leandro v. State of North Carolina agreed to allow an independent consultant to advise Judge David Lee on how to proceed in the 25-year-old case. They selected California-based consulting firm WestEd to recommend an action plan. WestEd, Learning Policy Institute, and the Friday Institute for Educational Innovation delivered its main report, Sound Basic Education for All: An Action Plan for North Carolina, and supporting reports to Judge Lee in July 2019. The contents of the report were released five months later.
In his January 21, 2020 consent order, Judge Lee declares that the reports “will inform decision-making and this Court’s approach to the case.” I believe this is a mistake. Judge Lee should ask a diverse group of stakeholders to assess the strengths and weaknesses of the WestEd report before it is used as the basis for court action.
To justify their recommendation to increase spending on North Carolina, WestEd researchers document a decrease in inflation-adjusted per-pupil spending since 2010. WestEd offers two different percentages for the decrease.
“When adjusted to 2018 dollars, per-pupil spending in North Carolina has declined slightly overall, about 6% since 2009–10.” (p. 21)
“Furthermore, when adjusted to 2018 dollars, per-pupil spending in North Carolina has declined slightly overall, about 2% since 2009–10, and year-over-year funding has been relatively flat (see Exhibit A2).” (p. 176-177)
“When adjusted to 2018 dollars, per pupil spending in the state has declined overall, about 6% since 2009-10.” (January 23 presentation to the Commission on Access to Sound Basic Education)
The first and third passages do not reference a data source, but Exhibit A2 shows a 2 percent decrease from 2010 to 2018. According to the graph, year-over-year funding has fluctuated significantly between 2010 and 2018. Between 2010 and 2014, per-pupil spending decreased, mostly due to lingering effects from the Great Recession and Medicaid shortfalls. But per-pupil spending has not been “relatively flat” since 2014. Exhibit A2 shows that inflation-adjusted per-pupil spending increased by 6.2 percent between 2014 and 2018 as revenue rebounded and the General Assembly reigned in Medicaid spending.
WestEd’s decision to begin their analysis in 2010 was unwise and potentially misleading. Ultimately, a timeframe is an arbitrary decision, but it should provide an appropriate context for the broader discussion. I believe that the WestEd report fails to do so.
As mentioned above, the Great Recession ended in June 2009, so the 2009-10 expenditure reflects some of the lingering fiscal effects from the recession. To their credit, WestEd researchers acknowledge the “cutbacks that began during the recession after 2008.” (p. 14) But there is no mention of Medicaid shortfalls or any other contextual information that may be beneficial to readers. Indeed, public school funding is subject to constraints imposed by revenue fluctuations and other budgetary commitments.
A more significant concern is that Leandro v. State of North Carolina did not begin in 2010. The case was filed in 1994 and decided by the N.C. Supreme Court in 1997. WestEd researchers include student performance data starting in the early- and mid-1990s (see exhibits 10 through 15), so why not also document expenditures for the same period?
Despite the decrease in education spending during and after the Great Recession, North Carolina has still been able to increase real per-pupil spending during the entire Leandro period. Inflation-adjusted per-pupil spending increased by 34 percent between 1992-93 to 2018-19. If we prefer the starting point to be the Leandro I decision in 1997, inflation-adjusted per-pupil spending in 1996-97, it increased 27 percent. Documenting expenditures over a longer period accounts for outliers and offers additional context that is lost by limiting the timeframe.
Finally, documenting per-pupil expenditures over a longer period avoids unnecessary partisanship. North Carolina Republicans won legislative majorities in 2010 and sustained those gains for a decade, so the 2010-18 period captures mostly Republican budgets. Yet, per-pupil spending stagnated under Democratic majorities. Inflation-adjusted per-pupil expenditures were relatively flat between 2000-01 and 2005-06 and increased by only 1.5 percent over this period.
WestEd researchers make multiple references to “deeper legislative cuts over the last few years” (p. 14, 214) but fail to identify the nature of these “deep” cuts or specify the timeframe. As mentioned above, WestEd-generated figures show the inflation-adjusted per-pupil expenditures decreased by only 2 percent between 2010 and 2018, and inflation-adjusted per-pupil spending increased by 6.2 percent between 2014 and 2018, arguably a period that qualifies as the “last few years.”
WestEd researchers may be referencing other types of cuts, such as personnel or programmatic changes. For example, in describing changes to personnel counts, WestEd again offers conflicting statistics.
“Over the same time period, enrollment in traditional public schools and charters schools increased by 2% (Public Schools of North Carolina, 2009).” (p. 14 of Educator Supply, Demand, and Quality in North Carolina: Current Status and Recommendations)
“As described in a research paper on educator workforce supply, demand, and quality prepared for this project, budget cuts have reduced the total number of teachers employed in North Carolina by 5% from 2009 to 2018, even as student enrollments have increased by 12%.” (p. 18)
“Budget cuts reduced the total number of teachers employed in North Carolina by 5% from 2009 to 2018, even as student enrollments increased by 2% during that span.” (p. 53)
Student enrollment increases have been closer to 2 percent, mostly due to increases in the charter school sector.
But, again, the selected timeframe is problematic. Establishing 2009 as the baseline year is inconsistent with their fiscal analyses, which start in 2010. More importantly, the 2008-09 school year is an outlier; the state employed more teachers that year than ever before. (See p. 14 of Educator Supply, Demand, and Quality in North Carolina and N.C. Department of Public Instruction’s Statistical Profile.)
If 2009-10 and 2017-18 are used as the basis for the teacher count, there is a much different story. According to N.C. Department of Public Instruction statistics, North Carolina employed 95,377 full-time district teachers and 2,512 charter teachers in 2010 for a total of 97,889 teachers. North Carolina employed 94,117 district teachers and 6,358 charter teachers in 2018 for a total of 100,475 teachers. That represents an increase of 2.6 percent, which is in line with enrollment changes during this period. (See note below.)
Of course, WestEd could have compared staffing levels throughout the Leandro period by using teacher-student ratios. Using state staffing and enrollment data, North Carolina’s teacher-student ratios have decreased slightly since the 1996-97 school year. Even the research paper on educator workforce supply, demand, and quality shows that student-teacher ratios have not changed significantly since 2012. North Carolina’s student-teacher ratio remains lower than the average student-teacher ratio of all states and is within half a student of the average of Southeast states.
The authors of the research paper on educator workforce supply, demand, and quality claim that the “reduction in teacher employment and the increase in pupil-teacher ratios is directly related to the concurrent reduction in school funding in the state.” Between 2010 and 2018, state- and locally funded teaching positions both increased (by 5 percent and 38 percent, respectively). Federally funded teaching positions decreased by 3,356 positions or 35 percent as temporary federal funds expired. Because the authors fail to differentiate between teaching positions funded by state, local, and federal dollars, the reader would not know that the “concurrent reduction in school funding in the state” must refer to federally funded positions.
In “A Study of Cost Adequacy, Distribution, and Alignment of Funding for North Carolina’s K–12 Public Education System,” researchers acknowledge the increased stress on the state budget due to employee benefits,
Considering per-pupil spending within North Carolina between 2010 and 2018 (see Exhibit 4), the allocation of resources by category remained fairly stable, with the main exception of salaries and benefits. Although the proportion of spending on labor overall increased 2 percentage points, the share of this spending allocated to salaries went down 3 percentage points while the share of spending on employee benefits increased 5 percentage points. Thus, although a greater proportion of expenditures is going toward labor costs, this does not translate into proportionately higher teacher salaries, due to the increasing cost of benefits — a trend seen nationwide (Aldeman, 2016). Proportion of funds allocated to other categories of spending remained relatively consistent over time. (p. 6)
This is a critical observation and a missed opportunity. In 2019, employee benefits for public school employees were a $3.1 billion expense and consumed 23 percent of the total education operating budget. But WestEd researchers failed to offer recommendations on how to curb health insurance and retirement cost growth. In fact, employee benefits will exert increasing pressure on the state budget for decades to come, undermining efforts to recruit teachers and fund programs.
Obviously, this critique is not exhaustive, but it raises doubts about the analyses that Judge Lee will use to formulate his approach to Leandro. In subsequent essays, I will outline additional concerns with other parts of the report.
Note: The authors of Educator Supply, Demand, and Quality in North Carolina use Occupational Employment Statistics from the Bureau of Labor Statistics, which is a semi-annual mail survey of non-farm establishments. BLS cautions using Occupational Employment Statistics to make comparisons over points in time because of changes to classification systems, data collection, reference period, and methodology. That is why I use N.C. Department of Public Instruction-generated teacher counts based on the Public School Full-Time Personnel Report.)