Objective, rigorous,
and useful research to
understand the post-Katrina
school reforms.

POLICY BRIEF
January 17, 2017

EducationResearchAllianceNola.org

This paper will remain under media embargo until 5 a.m. CST on Tuesday, January 17, 2017.

DOES SCHOOL REFORM = SPENDING REFORM?
THE EFFECT OF THE NEW ORLEANS SCHOOL REFORMS ON
THE USE AND LEVEL OF SCHOOL EXPENDITURES
By Christian Buerger and Douglas N. Harris, Tulane University

Overview
Over the past quarter century, few ideas in education reform have had more consistent support than giving schools more autonomy. One way to
do that is through charter schools. In particular, these public-private partnership schools, freed from district rules and union contracts, have the
flexibility to spend their funding as they think best.
This study estimates the effects of the post-Katrina school reforms on the operating expenditures of New Orleans’ publicly funded schools. We
compare all the city’s publicly funded schools to a group of Louisiana school districts that had similar spending patterns before the New Orleans
reforms shifted the city to a charter-driven system. We also show changes for both groups before and after the post-Katrina school reforms,
focusing especially on the most recent year of data available (2014). We draw the following conclusions:
• 

New Orleans publicly funded schools spent 13% ($1,358 per student) more per pupil on
operating expenditures than the comparison group after the reforms, even though the
comparison group had nearly identical spending before the reforms.

• 

Spending on administration in New Orleans’ publicly funded schools increased by
66% ($699 per student) relative to the comparison group, far more than the overall
spending increase. Of this increase, 52% ($363 per student) is due to a rise in total

Sample: New Orleans charter and traditional
public schools, plus 17 similar school districts
in Louisiana from 2000 to 2014
Data: District-level expenditures, including

administrative salaries. Roughly one-third of the increase in administrative salaries is

operational and instructional expenses

due to hiring more administrators, and the remainder is due to higher average salaries

Source: Louisiana Department of Education

per administrator.
• 

Sample and Data Sources

Instructional expenditures in New Orleans’ publicly funded schools actually declined by 10% ($706 per student) relative to the comparison
group. This decline is driven by a drop in spending for instructional staff benefits ($353 per student) and in instructional staff’s salaries ($233
per student). Almost all of the decrease in total instructional salaries is due to lower average salaries per instructor, though new teachers still
earn more today than teachers pre-Katrina who had the same years of experience.

DOES SCHOOL REFORM = SPENDING REFORM?

PAGE 1

 Transportation spending and other expenditures, which

scale. In larger traditional districts, schools can share a single system

typically include contracts to outside firms, each increased by

for accounting, busing, and food service. As an additional example,

33%. However, student support expenditures and maintenance

districts can have a single lawyer on retainer rather than having each

were largely unchanged.

separate CMO hire its own. Individual charter schools also tend to

The fact that instructional expenditures have decreased despite a
large increase in operating expenditure is striking. The increase
in administrative spending also suggests either that the lack of
economies of scale has posed a real challenge in this decentralized
system or that the educational models of charter schools involve
higher management costs and perhaps a more top-heavy approach.
There is no one right way to use educational resources, and it is worth
noting that these changes in spending levels and patterns came
alongside a large improvement in education outcomes for students.

have fewer students than traditional public schools, creating the same
economies of scale problem with extracurricular activities and other
specialized services.

“

Still, these results are somewhat surprising given the common concern
that traditional school districts spend too much on large bureaucracies.
We find that charter schools spend even more in that area.
Whatever the reasons, it is clear that the post-Katrina reforms led
to more spending in total and different spending patterns in New

... CMOs tend to be smaller
than school districts,
representing a loss of
economies of scale. In
larger traditional districts,
schools can share a single
system for accounting,
busing, and food service.

Orleans’ publicly funded schools.

BACKGROUND
Charter schools have become a growing percentage of publicly funded

“

• 

At first glance, it is hard to predict how charter schools might spend
money differently, and there are only few studies analyzing charter
school spending patterns. Research consistently shows that charter

schools in the U.S. According to their advocates, the goals of charter

schools tend to hire younger and less experienced teachers. Two other

schools include providing schools with more autonomy and holding

studies have considered spending patterns, but this appears to be the

schools accountable for performance, as well as facilitating school

first to look at the effect of switching from traditional public schools to

choice for families. Freed from district rules and union contracts,

charter schools, which helps isolate the effect of the charter approach.

charter management organizations (CMOs) and school leaders can
decide how to use resources in ways consistent with their beliefs
about best practices, parental demands, and the requirements of
performance-based contracts.
Critics point out, however, that district rules and union contracts serve
useful purposes, freeing up school leaders to focus on instruction,
preventing problems, and creating good working conditions and

We try to get beyond the speculation by studying the post-Katrina
school reforms. New Orleans’ publicly funded schools are now almost
all state-authorized charter schools. Because of the increased autonomy
schools now have with regard to personnel decisions, curriculum, and
spending, New Orleans offers a unique opportunity to study how a
charter-driven city might change patterns in school spending.

compensation for teachers. There are also concerns about transparency

We designed our study to analyze changes in citywide operating

in how charter schools use funding, especially in the case of for-profit

expenditure from 2000 to 2014. Specifically, we address two main

charters that might be more likely to use funds for private gain over

questions:

student benefit. While New Orleans does not have for-profit charters,
some of the same issues may arise with non-profits, which can use

1. 

and after the reforms?

increases in revenue to pay higher salaries to their leaders.
One challenge, less widely recognized, is that CMOs tend to be
smaller than school districts, representing a loss of economies of

DOES SCHOOL REFORM = SPENDING REFORM?

Did overall district spending levels differ in New Orleans before

2. 

How have funds been used differently since the reforms—on
administration, instruction, transportation, and other areas?

PAGE 2

 We answer these questions by comparing New Orleans with a group of

likely to be indirect as families demand particular kinds and quality

districts that had very similar spending patterns prior to the reforms so

of services.

that we can be confident that the reforms caused any observed changes.

Other evidence suggests that the intense test-based accountability
created by the state plays a strong role, alongside school choice,

POLICY CONTEXT

in affecting school decisions. Given the accountability focus on

The key aspect of the post-Katrina school reforms, for purposes here, is
that charter schools are granted greater autonomy over their spending

academic outcomes, we might also expect an increase in instructional
expenditures.

relative to traditional public schools. Charter schools are still required

The state is also in the process of returning control of the charter

to follow state accounting rules and to have annual audits, but they are

schools under its authority to the OPSB, which may eventually create

generally not subject to union contracts or district policies.

different pressures on the city’s schools. However, the present analysis

While we focus our discussion on the differences between charter and

“

“

The key aspect of the postKatrina school reforms...
is that charter schools are
granted greater autonomy
over their spending
relative to traditional
public schools.

completely precedes this policy change and focuses on the period when
the state RSD controlled the vast majority of schools.

HOW DID WE CARRY OUT THE ANALYSIS?
We report district-level expenditures from 2000 to 2014 for
New Orleans publicly funded schools, provided by the Louisiana
Department of Education (LDOE). We checked the validity of the
data with charter school and authorizer audits in the years they are
available. The various data sources yield very similar patterns, and we
therefore report only the LDOE data here.
We omit the 2005-06 school year because this was the year of
Hurricane Katrina, and these figures are not informative. The related
technical report for this study also uses data back to 1990, though we

traditional public schools, it is important to note that the control of the

do not report the earlier years for simplicity. All figures are adjusted

vast majority of schools also shifted from the Orleans Parish School

for inflation and reported in 2014 dollars.

Board (OPSB) to the state and its Recovery School District (RSD),
which introduced some of its own rules. In particular, the RSD requires
all of its New Orleans schools to provide transportation. The OPSB and
RSD also both essentially eliminated attendance zones so that families
could choose schools anywhere in the city, which increased the average
distance from home to school attended by about two miles per student.

Our analyses focus on the per-pupil operating expenditure of all New
Orleans publicly funded schools, including charter and traditional
public schools. We also include expenditures by the CMOs, as well as
OPSB and the RSD, the city’s two school districts, to ensure that we
cover the full scope of costs in the city.

This, combined with the decentralized nature of spending decisions

Our main spending categories, such as administration and

and limited leverage of economies of scale, almost guarantees that the

instruction, are defined by the Louisiana Accounting & Uniform

reforms increased transportation spending.

Governmental Handbook. Instructional expenditures include only

The fact that families have more choice also means that they might

spending for activities dealing directly with interactions between
students and staff. The Handbook also splits administrative spending

place pressure on schools to change how schools work in ways that alter

into several smaller categories. School administration refers to

their spending patterns. However, it is unclear in this case exactly how

spending on the administrative responsibilities at the school level,

this would change school spending patterns. Even the families most

such as the principal’s salary. Central administration includes

involved in their children’s schools are generally unaware of school

planning, research, development, evaluation, information, staff, and

finances; if anything, the spending pressures from school choice are

information technology activities, which usually arise at the CMO or

DOES SCHOOL REFORM = SPENDING REFORM?

PAGE 3

 district. General administration includes other expenses involved in
carrying out CMO and district policies that are not included in the
other categories. Note that all of the funding categories we report
throughout the study are mutually exclusive. Each dollar of spending
is only counted once.
Salaries and benefits represent by far the largest share of operating
expenditures. Some personnel have their positions split between
management and instruction. In these cases, we apportion the salaries

and the change in overall expenditure.
Finally, note that we do not report the usual confidence intervals in
this analysis, but all reported changes are statistically significant by
the usual scientific standards.

DID THE TOTAL OPERATING SPENDING LEVELS CHANGE
AFTER THE REFORMS?

and benefits based on the share of their total salaries in each category.

We find that New Orleans’ operating expenditures after the reforms

Understanding the effects of the reforms on spending requires more

before the reforms, the spending trends for New Orleans and the

than just comparing spending patterns before and after the reforms.

comparison group were nearly identical and gradually rising.

increased more than the comparison group’s spending. In Figure 1,

There might be general trends in spending across schools that have
nothing to do with the reforms, such as economic recessions and changes

Figure 1. Changes in Operating Expenditures

in funding formulas. To address this issue, LDOE also provided data
on the rest of the state, which we used to create a comparison group.
Specifically, we used a technique called synthetic control analysis to
create a weighted average of other districts that almost exactly mirrors
the pre-reform patterns in New Orleans school spending. This gives
us a very good prediction of how New Orleans schools would have
spent their funds in the absence of the reforms. The results are also
very similar when we used alternative comparison groups, such as
limiting the group to districts directly affected by Hurricane Katrina.
The differences in spending allocations between New Orleans and the
comparison can then be interpreted as an effect of the reforms.
We take two steps to ensure that we are identifying the effects of
the reforms on long-run school spending, as opposed to temporary

The vertical line in the middle of Figure 1, and in all later figures,

changes related to rebuilding the school system. First, we focus

indicates the year of the hurricane. After that, the reforms started,

on operating expenditures and completely omit the large capital

and we see sharp and persistent changes in spending levels. By

expenditure required to rebuild schools after the storm. We also focus

2014, New Orleans’ operating expenditures were 13% ($1,358 per

on the most recent year (2014), as the vast majority of the rebuilding

“

brief, when we refer to a single year, it is the spring of the academic
year; in this case, 2014 references the 2013-14 academic year).
To be clear, the changes in total operating expenditures per pupil are
probably not driven by the reforms per se. Since the vast majority of
funding comes from the government, it could be that the changes in
spending levels were driven, for example, by voters’ desire to help

By 2014, New Orleans’
operating expenditures were
13% ($1,358 per pupil) higher
than the comparison group.

the city rebuild, or the government may have been more willing to

“

efforts had been completed at that point (here and throughout this

pupil) higher than the comparison group.

support New Orleans’ schools with the reforms in place and with

We report only expenditures here, though the revenue trends closely

greater confidence that the money would be used well. Whatever the

match Figure 1. This just reflects the fact that schools spend the vast

reason, we view the changes in spending patterns (administration

majority of the funds they receive. The technical report associated with

versus instruction, etc.) as effects of the combination of the reforms

this brief provides additional detail on revenue sources.

DOES SCHOOL REFORM = SPENDING REFORM?

PAGE 4

 The key point here is that if schools increased all their spending

rise in the number of administrators per pupil. In other words, there are

categories equally, then we would expect all the individual categories

more administrators, and they typically earn higher salaries than they

to increase by the same 13% reflected in Figure 1. As we show below,

would have without the reforms.

this is far from the case. Some categories increased far more than
this, and others actually decreased.

Figure 3. Reform Effects on Administrative Spending, by Category

HOW DID SPENDING ON ADMINISTRATION CHANGE
BECAUSE OF THE REFORMS?
One of the most striking conclusions of our analysis is the steep
increase in administrative spending. Figure 2 shows this clearly. Total
administrative expenditures, which again were nearly identical prior
to the reforms, were 66% ($699 per pupil) higher than the comparison
group in 2014.

Figure 2. Reform Effects on Overall Administrative Spending

HOW DID SPENDING ON INSTRUCTION CHANGE BECAUSE
OF THE REFORMS?
Despite the large increase in total expenditures per pupil, instructional
spending actually declined because of the reforms. As of 2014, New
Orleans publicly funded schools spent 10% less ($706 per student) on
This increase is five times larger than we would have expected if the

instructional expenditures than the comparison group. The decline in

increase in total expenditure had been spread evenly across spending

instructional spending is even larger than the increase in administrative

categories. School administration is the largest contributor to the

spending.

increase in administrative spending ($503 per student) followed by
central administration and general administration, which only grew

Figure 4. Reform Effects on Overall Instructional Spending

marginally. Note also that the scale of y-axis differs between Figures
1 and 2 and others below so that the changes in spending in smaller
categories are easier to see.
To better understand the sources of these trends, we break the categories
down further into administrative salaries, fringe benefits (health care
and pensions), and other expenses (Figure 3). Roughly half of the increase
is due to salaries, and the other half is due to these other categories.
We break down the salary figure even further. The right side of Figure 3
shows that 62% of the increase in salaries is caused by an increase in the
average administrative salary, while the remaining 38% is caused by a

DOES SCHOOL REFORM = SPENDING REFORM?

PAGE 5

 To better understand what happened, we again break the results

schools in New Orleans participated in TRSL. With the charter-based

down into smaller categories. In this case, we see large declines in

school reforms, schools were not required to participate in TRSL

salaries and even larger declines in fringe benefits (Figure 5). The

and could alternatively offer 403(b) programs which require only

decline in benefits for instructional staff explains 50% of the decrease

small contributions by the schools. In 2014, only 42% of schools were

in instructional spending. The vast majority of the decrease in

enrolled in the state pension plan. The TRSL rates—that is, the amount

instructional staff salary spending (83%) is explained by a reduction

the state charged schools for TRSL participation—also doubled after

in the salaries themselves, while a small portion (17%) is due to a

the reforms, making it even more expensive for participating schools.

decrease in the number of teachers per pupil.

Since a large share of these rising rates went to pay legacy costs of
already-retired teachers, current teachers would not receive all the

Figure 5. Reform Effects on Instructional Spending, by Category

schools’ pension contributions back even in the long run.

HOW DID OTHER CATEGORIES OF SPENDING CHANGE
BECAUSE OF THE SCHOOL REFORMS?
The vast majority of school spending goes to pay the salaries and
benefits of teachers, administrators, and other staff. However, a few
additional categories are worth noting.
Recall that families were given almost complete choice over any
publicly funded schools in the city and that RSD charter schools were
required to provide transportation. Given these two policies and
the reduced economies of scale, transportation costs were almost
guaranteed to rise.
Figure 6 shows that this is exactly what happened. These results
show an increase in roughly $300 per pupil as of 2014. In this case,
the comparison group was spending $100 more before the reforms,
so the $300 figure reflects sum of the initial difference and the $200
difference in 2014.
The decline in salaries only makes up one-third of the reduction in
instructional spending and is mainly driven by the 12-year decline in

Figure 6. Reform Effects on Transportation Spending

the average experience level in the district. In all professions, including
teaching, people are generally paid more when they have more
experience. Post-reform New Orleans teachers actually earn more than
those from the pre-Katrina period who had the same years of experience.
We do not break down benefits into smaller categories, but these are
naturally similar to what we report for salaries. That is, the decline
in benefits, too, is almost entirely due to the decline in benefits per
instructor. This is mostly because, in the pre-reform period, charter
schools were required to pay into the Teachers Retirement System of
Louisiana (TRSL).
The size of the decline in benefits—50% of the total instructional
decline—is noteworthy, however. Before the reforms, all public

DOES SCHOOL REFORM = SPENDING REFORM?

PAGE 6

 It also turns out that charter schools are much more likely to contract

These changes might seem surprising given the criticism that traditional

out for services with outside organizations. These expenses, along with

public schools spend too much on management, but the patterns are not

some smaller categories, are reflected in the other spending category

difficult to explain. A loss in the economies of scale and a reduction in

shown in Figure 7. Spending in this category increased by 33% ($704

services by districts and authorizers lead us to expect higher spending

per student).

on administration and contracted services by schools and CMOs. The
declining participation in the state pension system and drop in average

Figure 7. Reform Effects on Contracted Services and Other Spending

teacher experience also explain declining spending for instruction.
There are several reasons to be confident in these results: First, prior
research on the subject, though limited, has also found that charter
schools spend a larger share of funds on administration. Second,
most of the changes are driven by salaries, which tend to be very
accurately reported. Third, the results hold up regardless of which
data sources and comparison groups we choose. These patterns
almost certainly reflect real differences in the allocation of resources
and real effects of the reforms.
If there is any uncertainty, it is over whether some categories of
spending that were previously categorized as instructional are now
being placed in the administrative category. However, this is strictly
hypothetical, and we have no evidence to suggest this is the case.

DISCUSSION

Some might also wonder whether charter schools are contracting out

The first important observation here is that the New Orleans school
reforms did not only entail a shift to charter schools and school
choice, but also in spending. Research increasingly shows that money
is a contributing factor in any school improvement effort. While our
calculations suggest that this increase in spending could not plausibly

some forms of instruction, such as special education, which would
shift funds from instruction to the other spending category. However,
if that were driving the results, there would be a substantial reduction
in the number of teachers. What we actually see is a reduction in
average salaries and benefits per teacher.

have explained all the reform effects on student academic outcomes,

One possible reaction to the higher salaries for administrators is that

these funds almost certainly explain some of the effects. Money

charter school educators are using the reforms to enrich themselves.

matters for charter schools and traditional public schools alike.

However, it is worth noting that charter schools are more likely to

Regarding the pattern of spending, the main effects of the reforms
have been the increase in administrative expenses and contracted
services, offset by a decline in instructional expenditures. When we
look more closely, we also see that salaries for administrators and

“

teachers are moving in opposite directions.

hire educators with degrees from elite colleges and school leaders
with business backgrounds, people who typically earn high salaries
outside of education. The higher salaries may partly reflect the other
professional opportunities available to the new educator workforce.
The hiring of additional managers might be related to the reliance
on younger teachers who no doubt need more active leadership and

... the main effects of the
reforms have been the increase
in administrative expenses
and contracted services, offset
by a decline in instructional
expenditures.

“

DOES SCHOOL REFORM = SPENDING REFORM?

professional development. The substantive issue is whether this
alternative model—trading off more managers for lower teacher
experience—is good for students. From what we know so far, the
system is working to improve measureable student outcomes.
Even if the New Orleans reforms are judged positively, however, it
is not clear whether this model is sustainable financially. There are
signs that interest in New Orleans is abating among younger teachers

PAGE 7

 and growing concern about the constant churn of teachers who do
not plan to stay or make teaching a career here. The state government
is also in dire financial straits and may eventually have to either
reduce spending or force charter schools to pay into the state pension
system. Something will have to give.

How Does This Research Relate to Other
ERA-New Orleans Studies?
The research agenda of the Education Research Alliance
for New Orleans is designed to understand the effects of the

There is no right way to spend money in schools, but money clearly

reforms on students and the ways in which those effects arise.

matters and provides important information about how schools

Spending is one mechanism for school improvement.

operate and what they value.

Since our results suggest that charter schools spend a larger
share on administration, we are beginning to examine the
organizational structures of charter schools and CMOs to
see how the allocation of specific management positions has
changed (for example, whether schools have more broadly
distributed leadership).
Given that the changes in instructional spending are driven
by changes in the teacher workforce, readers may also be
interested in our earlier descriptive analysis of teacher
certification, experience levels, and teacher degrees in the
report, Significant Changes in the New Orleans Teacher
Workforce (Harris & Barrett).
In an upcoming paper, we will be examining more closely the
factors that explain educator compensation. An additional
study will examine the decrease in New Orleans public
schools’ participation in the state’s teacher pension system, the
Teachers’ Retirement System of Louisiana (TRSL).
Since the ultimate goal of spending and other aspects of
education is to improve student outcomes, our analyses of
student outcomes are also relevant. In What Effect Did the
Post-Katrina School Reforms Have on Student Outcomes?,
Douglas N. Harris and Matthew Larsen find that the charterbased reforms have had large positive effects on student test
scores.

DOES SCHOOL REFORM = SPENDING REFORM?

PAGE 8

 About the Education Research
Alliance For New Orleans
The mission of the Education Research Alliance for New Orleans
(ERA-New Orleans) is to produce rigorous, objective, and useful
research to understand the post-Katrina school reforms and their
long-term effects on all students. Based at Tulane University, ERANew Orleans is a partnership between university-based researchers
and a broad spectrum of local education groups. Our Advisory

About the Authors
Christian Buerger
Christian Buerger holds a Ph.D. in Public Administration from
the Maxwell School of Citizenship and Public Affairs at Syracuse
University and is a Postdoctoral Fellow at the Education Research
Alliance for New Orleans.

Board includes (in alphabetical order): the Louisiana Association of

Douglas N. Harris

Educators, the Louisiana Association of Public Charter Schools, the

Douglas N. Harris is Professor of Economics, the Schleider

Louisiana Federation of Teachers, the Louisiana Recovery School

Foundation Chair in Public Education, and the founding Director

District, New Orleans Parents’ Guide, New Schools for New Orleans,

of the Education Research Alliance for New Orleans at Tulane

the Orleans Parish School Board, the Orleans Public Education

University. He is also a Non-Resident Senior Fellow at the Brookings

Network, and the Urban League of Greater New Orleans. For more

Institution.

information, please visit the organization’s website.

EducationResearchAllianceNola.com

Contact Information
1555 Poydras Street
7th Floor, Room # 701
New Orleans, LA 70112
(504) 274-3617
ERANewOrleans@gmail.com

An Initiative of

DOES SCHOOL REFORM = SPENDING REFORM?

PAGE 9