Thursday, April 6, 2017

School Choice: An Economic Winner

There are many reasons why parents should have the last say in their children's education. For one, it gives them the status they have the right to as free, independent people; having grown up in a country (Sweden) where the government had the right to educate your child, and parents were arrested and charged with a crime if they tried to homeschool their kids, I deeply value this aspect of school choice.

In addition, there are many economic reasons why it is better to have a market for K-12 education:
it encourages youth entrepreneurship, it keeps costs down without jeopardizing student performance (in fact, the exact opposite is often true) and it is a private-sector job creator. Ultimately, the winners are our children, but on the way to harvesting those profits from educational choice, we also benefit as taxpayers. 

As a first indicator of how school choice can help reduce the cost of education, I have compared per-student spending in all 50 states with school choice opportunities. The result is interesting: even though the method I used is experimental, it is fair to say that school choice, done right, may cut the per-student cost of public education by almost 15 percent. 

To get to the exact numbers from that comparison, let us first review per-student spending in all the 50 states. Table 1 reports U.S. Department of Education data on statewide, total operational expenditures - in other words it does not include capital spending on school facilities - per enrolled student:

Total current expenditures per student
New York  19,901 Louisiana  10,774
District of Columbia  19,558 Iowa  10,645
Alaska  18,359 Washington  10,194
New Jersey  17,845 Missouri  9,856
Vermont  17,288 Oregon  9,760
Connecticut  17,206 New Mexico  9,706
Wyoming  15,885 South Carolina  9,671
Massachusetts  15,138 California  9,599
Rhode Island  15,071 Arkansas  9,568
New Hampshire  14,362 Indiana  9,481
Maryland  13,997 Kentucky  9,443
Pennsylvania  13,861 Kansas  9,414
Delaware  13,701 Georgia  9,150
Illinois  13,091 Alabama  9,025
Maine  13,005 Colorado  8,953
Hawaii  12,458 South Dakota  8,873
North Dakota  12,339 Florida  8,714
Nebraska  11,715 Tennessee  8,622
Minnesota  11,407 Texas  8,572
Ohio  11,255 North Carolina  8,460
West Virginia  11,233 Nevada  8,358
Wisconsin  11,067 Mississippi  8,268
Montana  11,010 Oklahoma  7,693
Virginia  10,968 Arizona  7,412
Michigan  10,912 Idaho  6,527
Utah  6,459
Source: U.S. Department of Education

As we can see, when D.C. is included we are the seventh costliest jurisdiction. As we know from last week's article about capital school construction, Wyoming has the second highest capital-to-current spending ratio of all states and the DC; in fact, only the nation's capital beats us in terms of capital construction as share of total spending on education. 

Since capital construction is not included in this part of the U.S. Department of Education's database, we have to leave it aside. There is, however, a good side to that: if we slowed down capital construction to levels that are more in tune with the rest of the country, we would see a rapid decline in school spending per student. 

Before we get to the school-choice comparison with educational outlays, let me also report on the share of total current school spending that is defined as instructional by the Department of Education:


Instructional share of total spending
New York 71.8% Missouri 63.7%
Minnesota 69.4% Montana 63.6%
Nebraska 68.9% Nevada 63.3%
Tennessee 68.1% Hawaii 63.3%
Florida 67.5% Delaware 63.2%
Massachusetts 67.3% Colorado 63.2%
Utah 67.1% Maine 63.1%
Maryland 67.0% Alaska 62.8%
Virginia 67.0% South Dakota 62.8%
New Hampshire 66.0% North Dakota 62.8%
Iowa 65.9% Kentucky 62.8%
Georgia 65.9% Michigan 62.5%
Kansas 65.2% Rhode Island 62.3%
California 65.1% West Virginia 61.9%
Connecticut 65.1% South Carolina 61.8%
North Carolina 65.0% Ohio 61.8%
Wyoming 64.8% Oregon 61.6%
Arkansas 64.8% Alabama 61.5%
Texas 64.5% Louisiana 61.3%
Idaho 64.4% New Jersey 61.1%
Vermont 64.4% Indiana 60.9%
Washington 64.3% Mississippi 60.7%
Pennsylvania 64.1% Arizona 59.2%
Illinois 63.8% Oklahoma 58.2%
Wisconsin 63.8% District of Columbia 56.3%
New Mexico 56.3%

When it comes to operational costs, we do not stand out: spending just below two thirds of our education dollars on instruction is a somewhat high share compared to most other states. 

In addition to the absence of capital-construction spending in this comparison, there are two questions to keep in mind: first, how does the U.S. Department of Education define "instructional"? According to their database, instructional support functions are included as costs for instruction. This raises some pertinent definitional questions. 

Secondly, how does the rate of centralization of a school system affect the rate of instructional dollars in school spending? There is interesting data on this, that I will have to get back to at a later point in time. For now, let me suggest that if the state or the federal government pick up a disproportionate part of school funding, more of it is going to go into what is defined as instructional spending. That does not mean instruction gets better - suggesting a comparison of per-student outlays to student performance - but it is definitely a question that merits further inquiry.

Now for school choice and educational cost. A comparison of the above-reported total current expenditures for K-12 education with school-choice data is not easily done. School choice is an institutional variable, thus a representation of the "capital stock" in the school system; current expenditures, by definition, represent the flow of an activity. In order to address the methodological problems that can arise from this, I compiled an experimental index to represent access and use of school choice by state. The index, which is based on school-choice data from the Friedman Foundation, takes into account the number of school-choice programs, participation rates and eligibility rates by program. The final index number is imperfect, as the disparity between states with many programs and states with few programs is a bit too high. 

With that in mind, the following numbers, preliminary as they are, indicate that the more accessible school choice is, the lower the cost of public education tends to be. This means, plain and simple, that public schools have to lower their costs in the face of competition from private education options. Table 3 reports the unweighted average per-student cost of education (as defined above) for three groups of states:


Average per-student current expenditures
Strong choice Moderate choice No choice
Wisconsin $10,427 Pennsylvania $10,867 Mississippi $12,184
Ohio Maine Alaska
Minnesota D.C. Arkansas
Iowa Alabama California
Louisiana Georgia Connecticut
Arizona Colorado Delaware
Indiana N. Carolina Hawaii
Illinois Maryland Idaho
Florida S. Dakota Kansas
Rhode Island Kentucky
Virginia Massachusetts
Oklahoma Michigan
S. Carolina Missouri
Utah Montana
Mississippi Nebraska
New Hampshire
New Jersey
New Mexico
New York
North Dakota
West Virginia
U.S. Department of Education and Friedman Foundation

The per-student cost in the "strong choice" group is 85.6 cents to the dollar spent in the "no choice" group. While, again, this is an experimental study, it does suggest that there is a positive correlation between school choice and educational expenditures. 

I would have liked to report on specific numbers for Wyoming under the premise that we would move from "no choice" to either of the "choice" columns. However, that requires quite a bit of fine tuning of this study, and therefore I will have to get back to these numbers at a later point in time. For now, take this for what it is: yet another finger pointing in the direction of school choice as an economic winner for the Cowboy State.

No comments:

Post a Comment