$chools in Crisis:

Making Ends Meet

The Opportunity Cost of Smaller Classes:
A State-By-State Spending Analysis
Marguerite Roza and Monica Ouijdani

December 2012

Two seemingly different threads are in play on the issue of class size. The first is manifested in media
reports that tell us that class sizes are rising to concerning levels. The second thread appears in the
Rapid Response

work of some researchers1 and education leaders2 and suggests that repurposing class-size reduction
funds to pay for other reforms may not be such a bad idea.

Both threads are evidence of current budget tensions. Constrained revenues3 are forcing some
districts to cut staff, and any tinkering with staffing which can drive up enrollment in some classes.
While smaller classes are on many levels desirable, it is also true that smaller classes come with
a hefty price tag. And so, in an environment of scarce resources, those seeking better outcomes in
education have begun rethinking previous decisions to lock up their funds in small classes.

For those who support smaller classes, the argument has been that smaller classes yield better
outcomes for students, all else being equal. The other side generally doesn’t dispute some level of
benefit to smaller classes, but suggests that it is not enough to consider only the benefits. Rather,
leaders must weigh the relative benefits of smaller classes against the benefits of other possible uses
of funds in order to seek the best possible effect for a given expenditure level.4

Practically speaking, class sizes already vary from state to state. Any state’s decision about class size
changes might depend on the magnitude of current class sizes, as well as how much money can be
repurposed if class sizes are increased. This brief provides a state-by-state context to the class size
discussion by showing how class sizes differ among the states, and by estimating how much money
could be freed up by modest changes in each state’s average class size.

1. Matthew Chingos and Grover J. “Russ” Whitehurst, “Class Size: What Research Says and What It Means for State Policy” (Brown
Center on Education Policy at Brookings, 2011), http://www.brookings.edu/papers/2011/0511_class_size_whitehurst_chingos.aspx.
2. Eva Moskowitz, “The Cost of Small Class Size,” The Washington Post, March 27, 2011.
3. Phil Oliff and Michael Leachmen, “New School Year Brings Steep Cuts in State Funding for Schools” (Center on Budget and Policy
Priorities), http://www.cbpp.org/files/9-1-11sfp.pdf.
4. Chingos and Whitehurst, “Class Size: What Research Says.”
Rapid Response The Opportunity Cost of Smaller Classes:
A State-By-State Spending Analysis

Just how big are classes these days?

In following news and blog coverage of class-size increases, one might think class sizes have now jumped
to levels unprecedented in recent decades. Most reports begin with a statement about how class sizes are
“swelling,”5 “ballooning”6 or in some cases, “skyrocketing.”7 We then hear about a single case example, such
as the 9th grade Spanish classroom in an Oregon school with 40 students8 or a Las Vegas elementary class
with 41 students.9 By contrast, when reports involve district averages, the numbers are often more muted,
showing increases of just one to three students.10 Thus, the current class size debate may not be informed
by the facts. For example, a survey of Washington, D.C., insiders working at the federal level reports that 92
percent of respondents assumed average elementary class sizes are substantially larger than national aver-
ages indicate.11

Also notably missing from coverage of class sizes are statewide trends, largely because states simply do not
report year-to-year changes. Instead, statewide class-size data are reported every four years by the National
Center for Education Statistics (NCES) via a large-scale staff survey, with the most recent data reported for
2007–08.

This analysis attempts to fill that gap in current class-size information by modifying the 2008 figures accord-
ing to subsequent changes in student-to-teacher ratios by state. Student-to-teacher ratios data are available
through 2011–12 via the National Education Association (NEA). While student-to-teacher ratios are slightly
different than average class-size figures, both track student enrollment and staffing, so the rate of change in
one measurement is likely to mirror the change in the other measurement.12 For each state, we compute the
rate of change in student-to-teacher ratios using the NEA data and is used to adjust the 2008 NCES class-size
data between 2007–08 and 2011–12. The results nationally are captured in Figure 1.

As Figure 1 shows, the estimated national average class sizes for 2011–12 is actually slightly smaller than it
was in 1999–2000. Over the last twelve years, elementary classes dipped and then began rising again (likely
due to some states implementing class-size reduction in early elementary years, and then relaxing those
requirements). Secondary classes, in contrast, rose by an average of one student before falling again to an
estimated 22.7 students per class.

There are undoubtedly specific states, districts, and schools that buck the national trend. In fact, uneven class
sizes are not a new phenomenon, and selective coverage of class size “swelling” regularly hits the headlines
as budget gaps come and go.13

5. Sam Dillon, “Tight Budgets Mean Squeeze in Classrooms,” New York Times, March 6, 2011.
6. http://www.spotlightnews.net/features/story.php?story_id=131664094638217300
7. “Will Education Become a Causality of the Recession?” Randi Weingarten interviewed on The Ed Show, MSBNC, May 25, 2010.
8. Nicholas D. Kristof, “Our Broken Escalator,” New York Times, July 16, 2011.
9. “Class Size Fight: Debate Looms During a Year of Overcrowding,” Huffington Post, October 5, 2011.
10. Anna Phillips, “Class Sizes Grew Again, New Figures Show,” School Book, WNYC, November 15, 2011.
11. National Journal Education Poll: Overview of Results (National Journal Group and Gates Foundation, 2010), http://www.gatesfoundation.org/
college-ready-education/Documents/national-education-journal-survey-results.pdf.
12. Rankings and Estimates: Rankings of the States 2011 and Estimates of School Statistics 2012 (National Education Association, 2011).
13. See for example, “Class Sizes Grow as States Confront Budget Woes,” Los Angeles Times, May 21, 2002.

Center on Reinventing Public Education, University of Washington 2

425 Pontius Ave N, Seattle, WA 98109 • 206-685-2214 • www.crpe.org
Rapid Response The Opportunity Cost of Smaller Classes:
A State-By-State Spending Analysis

Figure 1. National average class sizes have shifted only slightly over the last decade

Class sizes vary substantially from state to state

Table 1 displays each state’s average class size for the last three survey years, as well as estimated class sizes
for 2011–12. Of interest is the fact that average class sizes vary considerably between states and among
elementary and secondary levels. In 2011–12, Oregon had the largest estimated elementary classes at 24.6
students, whereas New York had the smallest at 15.5. At the secondary level, California’s classes average the
largest at 32.3 students, in contrast to North Dakota’s classes, which are the smallest at 17.0.

Center on Reinventing Public Education, University of Washington 3

425 Pontius Ave N, Seattle, WA 98109 • 206-685-2214 • www.crpe.org
Rapid Response The Opportunity Cost of Smaller Classes:
A State-By-State Spending Analysis

Table 1. In most states, class sizes are smaller now than they were in 1999–200014

*
Both NEA and NCES reported pupil-teacher ratios through 2011, but only NEA produced them for
2012, so NEA data were used here. Where NEA data for 2009–11 diverged from NCES pupil-teacher
data for the same time period, the analysis was deemed unreliable and not included.

14. Sources: Authors’ calculations based on class sizes from NCES Schools and Staffing Survey (1999–2000, 2003–04, and 2007–08), and applica-
tion of the rate of change in pupil-teacher ratios from NEA Rankings of the States 2011 and Estimates of School Statistics 2012.

Center on Reinventing Public Education, University of Washington 4

425 Pontius Ave N, Seattle, WA 98109 • 206-685-2214 • www.crpe.org
Rapid Response The Opportunity Cost of Smaller Classes:
A State-By-State Spending Analysis

State-by-state analysis identifies current spending on marginal class sizes at $15.7 billion

Class-size reduction efforts are found at the district, state, and even federal levels. Determining the price tag
for a specific policy initiative requires some determination as to where the costs are born (e.g., federal or
state level versus locally),15 how the policy is structured (e.g., a hard cap on class size versus targets for aver-
age class size),16 and to which classes the policy applies.17

Yet at a more practical level, we can understand the level of public funds tied up in smaller classes by examin-
ing the cost implications of having classes with fewer students. At the very least, smaller classes require more
teachers, yielding higher salary and benefits costs. There can also be additional facilities and administrative
costs associated with class size reduction. However, these costs might be difficult to eliminate even if class
sizes are increased.

Using average teacher salary18 and average class-size data19 by state, this brief estimates how much was
expended on teacher salaries (without benefits) to enable 2008 class sizes and compares them to salary
costs for classes with two additional students.20 This study is a departure from other recent studies in that it
computes cost implications by modeling a change to class-size data in contrast to other analyses that model
changes in pupil-teacher ratios, which do not translate directly to class sizes. For a detailed explanation of
the methodology, see Appendix A.

As is summarized in Table 2, raising class sizes nationally by two students could free up a total of $15.7 billion
in public funds. Applying a national average benefit load of 33%, the sum raises up over $20 billion.

Clearly, in most districts, a shift to larger classes wouldn’t immediately free up the funds listed here, as such
shifts are complicated. Part of the challenge is the “integer problem”—the reality that schools are unable
to divide teachers, and so within the constraints of a school’s current enrollment it may not be possible to
raise class size by two students, especially in a smaller school. Cost savings would more likely take place over
several years as districts work to manage their school enrollments around the larger class-size targets (much
in the manner that they manage school enrollments today).

15. http://www2.ed.gov/offices/OESE/ClassSize/index.html.
16. http://www.rand.org/pubs/rgs_dissertations/RGSD156/RGSD156.ch3.pdf.
17. “A Descriptive Evaluation of the Federal Class-Size Reduction Program” (United States Department of Education), http://www2.ed.gov/rsch-
stat/eval/other/class-size/index.html (accessed December, 2012).
18. The data rely on average teacher salaries by state. Although teacher compensation varies widely within states, the use of average salaries
provides a good starting point for analysis. For a more accurate projection, states could use regional salary averages or ranges.
19. We distinguish class size from pupil-teacher ratio in this study; the latter includes auxiliary teaching staff (i.e., librarians, music teachers, etc.)
and does not reflect the actual number of students in a classroom. For a detailed explanation of the difference between these two reporting meth-
ods, see Appendix A.
20. The analysis does not include fringe benefits and reduced facilities usage, which would only increase projected savings.

Center on Reinventing Public Education, University of Washington 5

425 Pontius Ave N, Seattle, WA 98109 • 206-685-2214 • www.crpe.org
Rapid Response The Opportunity Cost of Smaller Classes:
A State-By-State Spending Analysis

Table 2. Current cost of marginally smaller class sizes

Center on Reinventing Public Education, University of Washington 6

425 Pontius Ave N, Seattle, WA 98109 • 206-685-2214 • www.crpe.org
Rapid Response The Opportunity Cost of Smaller Classes:
A State-By-State Spending Analysis

As would be expected, the amount of funds available for other uses would vary by state due to differences in
average teacher salaries and current class sizes.21 Yet in all states, the amount remaining to be reallocated due
to small increases in class size would add up to 2-4% of total K-12 expenditures.

Class-size discussions should be informed by current class-size levels and real opportunity costs

Because of constrained budgets, class size is a high-profile topic among policymakers, education leaders,
parents, and journalists. But the conversations appear woefully ill informed, as perceptions of current class
sizes are out of sync with actual averages. Journalism that reports on class sizes as “skyrocketing” contradicts
current trends that have today’s class sizes at levels below their 1999–2000 levels in the majority of states. The
debate on whether current class sizes are too high or too low should, at a minimum, capture real time levels.

But recognizing the current class-size level is not enough. Consideration of whether smaller classes are
preferable to larger ones requires some recognition of the opportunity costs involved. This brief starts that
conversation by computing the dollars at stake if states accept marginally larger class sizes. To illustrate the
magnitude of public funds involved, the brief shows that the fifty states currently spend $15.7 billion, or an
average of $319 per pupil, per year to maintain their class sizes at current levels as compared with letting
them increase by two. Adding benefits costs to the mix drives up the estimates to over $20 billion nationally.

More recently, new tools (including Education Resource Strategies’ “Budget Hold ’Em” game) are enabling
more informed discussion of spending tradeoffs.22 States might decide to prioritize class size reductions, but
policymakers should understand that such decisions can rule out other potentially desirable uses of funds.
For instance, for those who hope to raise teacher or principal salaries, a marginal increase in class size of two
would enable an average salary raise of over $5000 per teacher, or a $20,000 increase in salary if all the funds
were directed to top quartile teachers. Similarly, the funds are more than sufficient to ensure a laptop for
each student, or lengthen the school day in the poorest 20% of schools.

Other studies have pointed out that reforming schooling by reducing class size often creates the least benefit
for the cost. Alternatives—for example, increasing class size while letting go the lowest-performing teach-
ers and paying high-quality teachers more—could have an even more positive effect on student learning and
cost reduction.23 While this brief doesn’t assess the relative value of each of these investments, consideration
of any investment makes most sense when set against thoughtful alternatives with the projected costs and
savings of each.

21. In the United States, the average teacher salary for the 2011–12 school year was $56,643; however the range is between $39,850 for South
Dakota and $74,449 for New York. See Summary Table G, “Estimated Average Annual Salaries of Total Instructional Staff and of Classroom Teachers,”
2010–11 (revised) and 2011–12.
22. “School Budget Hold 'Em Tradeoff Exercise” (Watertown, MA: Education Resource Strategies).
23. See for example, Chingos and Whitehurst, “Class Size: What Research Says;” S. Dynarski, J. Hyman, and D.W. Schanzenbach, “Experimental
Evidence on the Effect of Childhood Investments on Postsecondary Attainment and Degree Completion,” NBER Working Paper No. 17533
(Cambridge, MA: National Bureau of Economic Research, 2011); and D.N. Harris, “Toward Policy-Relevant Benchmarks for Interpreting Effect Sizes:
Combining Effects with Costs,” Educational Evaluation and Policy Analysis, 31, 3–29 (2009).

Center on Reinventing Public Education, University of Washington 7

425 Pontius Ave N, Seattle, WA 98109 • 206-685-2214 • www.crpe.org
Rapid Response The Opportunity Cost of Smaller Classes:
A State-By-State Spending Analysis

Appendix
Measuring Class Size

This analysis differs from other attempts at modeling the incremental spending of class-size changes because
it uses changes in class size and not commonly used pupil-teacher ratios (while pupil-teacher ratios are by
no means the equivalent of class sizes, they are, however, readily available and easy to manipulate). This
analysis is derived by dividing the total number of students by the total number of teachers. In this case,
teachers include all certificated staff (excluding administrators) whether they are primarily responsible for a
classroom or teach pull-out classes such as gym, art, or music. Because pull-out staff are included in the total
number of teachers, the pupil-teacher ratio is often lower than average class size.

Class size is determined by either direct observations of classrooms or surveys of the number of students
in a classroom. Then, averages are tallied for the school, district, and state. Due to the fact that surveys and
observations are more time-consuming and costly, average class-size data lags behind pupil-teacher data,
which is another reason the latter is more widely used.

Methodology: Calculating Projected Funds Available

This analysis used data from two sources. The National Education Association’s 2010–11 Rankings & Estimates
provided data on the number of elementary and secondary teachers, the average salary for classroom teach-
ers, and the fall enrollment of students.24 The 2007–08 Schools and Staffing Survey from the National Center
for Education Statistics provided state-by-state figures for average class size.25 This analysis used the data to
compute the number of teachers needed when increasing the average class size by two additional students.
The difference in salary costs due to the decreased workforce was then found. As noted earlier, these calcula-
tions do not include fringe benefits and changes in facilities.

The following four-part calculation was used to determine the available funds for each state:

1. Average Class Size (ACS) x Number of Teachers (NT)/New Class Size = New Number of Teachers
(NNT)

2. NNT x Average Teacher Salary (ATS) = New Salary Costs (NSC)

3. Old Salary Costs (OSC) – New Salary Costs (NSC) = Projected Funds Available (PFA)

24. Summary Table F: Estimated number of instructional staff members in public elementary and secondary schools by type of position, 2010–11,
Rankings and Estimates: Rankings of the States 2010 and Estimates of School Statistics 2011 (National Education Association, 2011) and Summary
Table G: Estimated average annual salaries of total instructional staff and classroom teachers, 2009–10 (revised) and 2010–11, Rankings and
Estimates: Rankings of the States 2010 and Estimates of School Statistics 2011 (National Education Association, 2011).
25. Table 71. Highest degree earned, years of full-time teaching experience, and average class size for teachers in public elementary and second-
ary schools by state: 2007–08, "Public Teacher Questionnaire" Schools and Staffing Survey (SASS) 2007–08 (Department of Education, National
Center for Education Statistics, 2009).

Revised February 2013

Center on Reinventing Public Education, University of Washington 8

425 Pontius Ave N, Seattle, WA 98109 • 206-685-2214 • www.crpe.org
Rapid Response The Opportunity Cost of Smaller Classes:
A State-By-State Spending Analysis

The $CHOOLS IN CRISIS: MAKING ENDS MEET series of Rapid Response

briefs is designed to bring relevant fiscal analyses to policymakers amidst the
current economic crisis. For more information, please visit www.crpe.org

Dr. Marguerite Roza is Director of the Fiscal

Analytics Unit at Georgetown University and
Senior Research Affiliate at the Center on
Reinventing Public Education.

Monica Ouijdani is a Research Analyst at the

Center on Reinventing Public Education.

Funding for this work was provided by the Bill & Melinda Gates Foundation. We thank
the foundation for its support, but acknowledge that the findings and conclusions
contained here are those of the authors alone and do not necessarily reflect the opinions
of the foundation.

Center on Reinventing Public Education, University of Washington 9

425 Pontius Ave N, Seattle, WA 98109 • 206-685-2214 • www.crpe.org