PRIVATE AND PUBLIC EDUCATION IN WISCONSIN

Transcription

1 PRIVATE AND PUBLIC EDUCATION IN WISCONSIN Implications for the Choice Debate John F. Witte Christopher A. Thorn Kim A. Pritchard The Robert M. La Follette Institute of Public Affairs Created in 1984, the Robert M. La Follette Institute of Public Affairs is a department of the University of Wisconsin-Madison, which carries out a multifaceted mission of instruction, research, and outreach. The Institute takes no stand on policy issues and publishes papers that present many points of view. Opinions in these pages reflect the views of the authors. John Witte is professor at the Robert M. La Follette Institute of Public Affairs and in the Department of Political Science at the University of Wisconsin-Madison. Christopher Thorn is associate researcher at the La Follette Institute and the University of Bielefeld, Germany. Kim Pritchard is research assistant at the La Follette Institute Board of Regents of the University of Wisconsin System All rights reserved Additional copies may be requested from: Robert M. La Follette Institute of Public Affairs The University of Wisconsin-Madison 1225 Observatory Drive Madison, Wisconsin PRIVATE AND PUBLIC EDUCATION IN WISCONSIN IMPLICATIONS FOR THE CHOICE DEBATE This paper describes the location and demographic characteristics of families with children in kindergarten through twelfth grade in both private and public schools in Wisconsin. Wisconsin has pioneered public funding for private schools through the Milwaukee Parental Choice Program (MPCP) enacted in The program is targeted toward low-income families and students who were not in private schools in the prior year. It is limited to nonreligious private schools in the City of Milwaukee. If more students apply to a school than there are places for the students, the schools must select students randomly. Enrollment in the program is limited to approximately 1,500 students in any given year and is limited within each school to 65 percent of the student enrollment. Private schools receive the state aid that would have gone to the Milwaukee Public School System (MPS) in lieu of tuition ($3,209 in ). Governor Thompson has recently proposed expanding the program to include any private school in Milwaukee.

2 He would also increase the enrollment limit to 3,500 in , 5,000 in , with no limit after that. All other current program conditions would continue to apply. Four years of data support the conclusion that the current program has primarily benefited poor families (annual income approximately $12,000) that are often headed by single parents (75%). Approximately 58 percent of parents in the program were on public assistance, and only 36 percent were employed full time. Students applying to choice schools are primarily racial minorities (95%). Choice families are headed by parents (primarily mothers) who on average are more educated than Milwaukee Public School parents.1 Because this program may eventually expand beyond Milwaukee, and may relax income restrictions and the random selection stipulations, we feel it is important to include in the debate descriptions of private school students and their schools in Wisconsin. The general picture of the private school population, in almost all respects, contrasts sharply with the portrait of families currently enrolled in the Choice Program. Data Sources The data in this paper came from two sources. The location of students by school district and the characteristics of those districts (shown in tables 1-4) are drawn from public records collected by the Wisconsin Department of Public Instruction for the school year. Demographic characteristics of students and households presented in tables 5 and 6 are from the 1990 U.S. Census Public Use Micro Sample (PUMS).2 Most census data are released in aggregate form (cities, counties, etc.), but the Census also releases a 5 percent stratified random sample of individual household data. For this study we used the Wisconsin sample of PUMS. We included all households in which a child between the ages of 5 and 19 was present and enrolled in some grade between kindergarten and twelfth grade. Unfortunately, neither data source allows us to distinguish between private parochial schools and private nonsectarian schools. In the United States, approximately 90 percent of private school enrollment is in religious schools. We have no way of judging whether Wisconsin deviates significantly from this average. It is probably safe to conclude, however, that a considerable majority of private school students in Wisconsin are enrolled in religiously affiliated private schools.3 Finally, as indicated in the footnote for table 6, the definition of "Milwaukee Suburbs" is not precisely the same in the two data sources because PUMS data do not allow breakdowns by school district. In tables 1 through 4, Milwaukee suburbs are those suburban districts participating in the Milwaukee Chapter 220 program; in table 6, the suburbs include Milwaukee County (minus the city), and Waukesha and Ozaukee counties. Where Do Students Attend Private and Public Schools? Tables 1 through 4 present geographic data based on the enrollment by school district in private and public schools in Wisconsin. We focus on the five largest cities, the Milwaukee and Madison suburbs, and the rest of the state. In tables 1 through 3, enrollments are broken down by kindergarten through eighth grade and high school.4 Table 1 provides percentage breakdowns; table 2 a set of ratios of private school to public school percentages; and table 3 the exact enrollment figures. Table 4 provides average household income and wealth characteristics of districts aggregated by the relevant categories. Figures in table 1 should be read as the percentage of either public or private school enrollment in the relevant category. For example, the figure in the upper left-hand cell-for Milwaukee, K-8, percent in public school-means that percent of the public school K-8 students in were enrolled in MPS. This can be contrasted

3 with the fact that only 8.93% of the high school students in public schools are in Milwaukee. Figures in the bottom panel of table 1 combine the five largest cities, the suburban areas of the two largest districts, and the remainder of the state. Adding across rows and allowing for rounding errors, these figures total 100 percent. A comparison of public and private enrollment allows us to ascertain the areas in which private schools are overor underrepresented relative to public school enrollments. The private to public school ratios computed from table 1 are presented in table 2. For example, for K-8 in Milwaukee, the private/public ratio of 1.09 is computed by taking the Milwaukee private school K-8 population in table 1 (13.42%) and dividing it by the K-8 public school percentage for the district (12.27%). The ratio of 1.09 indicates that private school elementary students are slightly overrepresented in Milwaukee relative to public school enrollment. Multiplying these ratios by 100 gives an over- or underrepresentation percentage. The findings reported in these tables support a number of conclusions. Overall private school enrollment is spread throughout the state, with only percent of private school students in the five largest cities. Public schools are less concentrated, with only percent of public school students in large cities. If we look at grade level, however, the large cities account for close to half (47.54%) of all the private school enrollment in grades nine to twelve. Milwaukee alone accounts for percent of the private high school enrollment. This is in sharp contrast to the 8.93 percent of the public high school enrollment accounted for by MPS students. The ratio of private school overrepresentation for Milwaukee in grades nine to twelve is 3.84 (table 2). This means that private high school enrollment in Milwaukee is 384 percent higher than would be expected if public and private enrollments were proportionate in geographic regions. Two factors may explain this pattern. First, private high schools must respond to economies of scale and so are most likely to exist in concentrated population areas. Thus in Milwaukee there are two large parochial high schools (Pius and Marquette), which presumably draw from the larger metropolitan area. Green Bay, which has almost 200 percent overrepresentation also supports two large parochial schools (one Luthern, the other Catholic). Second, the overrepresentation at the high school level in Milwaukee may be associated with parental concerns about Milwaukee public high schools. The elementary school pattern is considerably different. Although the five largest cities also have some overrepresentation of private school students, the ratio is only That percentage is brought down by Madison, which has lower private school enrollment in all categories. The elementary school number that stands out is the considerable overrepresentation of private schools in the Milwaukee suburbs. They have 177 percent of the expected private school enrollment in comparison to an assumption of even distribution between public and private schools. That enrollment does not carry over into high schools, but some of these students undoubtedly enroll in private high schools in the City of Milwaukee. Table 4 provides income and wealth statistics for these geographic areas. The results are not unexpected. The last two columns, which provide ratios of average income and property wealth relative to state averages, show that cities have both lower income and lower wealth than the rest of the state. The two suburban areas have much higher income and wealth than the state averages. The most extreme contrast is between MPS and the surrounding suburban districts. In terms of average income per pupil, the suburbs have almost twice as much as the city ($45,010 compared with $23,627). In terms of property wealth per pupil, suburbs have 2.45 times more than the Milwaukee average ($359,735 to $146,623). Who Goes to Public and Private School in Wisconsin?

4 District breakdowns and aggregates do not allow us to make accurate comparisons of families and households sending their children to public schools and those sending them to private schools. Individual level data are required for those comparisons. Such data are presented in tables 5 and 6. Because the Milwaukee Parental Choice Program is in Milwaukee, and because the Milwaukee area has more than its share of private school students, we focus on the Milwaukee metropolitan area and the state as a whole. Tables 5 and 6 differ in that table 5 is based on individual student statistics, table 6 on household characteristics. In most regards, both tables present a picture of private school students and households that is very different from the portrait of families and students participating in the current Milwaukee Parental Choice Program. For the state as a whole, in terms of race, as indicated in the top panel of table 5, private school students were more likely to be white than public school students.5 The differences are larger in elementary school than in high schools. For the latter, the percentage of nonwhite students is quite close to the percentage of whites (9 percent of private school students are nonwhite; 11.5 percent of public school students are nonwhite). The racial differences between public and private schools in Milwaukee were considerable (see panel b of table 5). Between 84 and 85 percent of the private school students in the City of Milwaukee were white, but only 33 percent of the public elementary and 37.9 percent of the public high school students were white. For African Americans the differences were particularly sharp, with 55 percent of elementary students and 51 percent of the public high school students being African American. In contrast, only 12.8 percent and 9.8 percent of the private school students were African Americans. Racial differences in the suburbs were trivial, with almost all students in both types of schools being white (see panel c of table 5). Family demographics indicate sharp contrasts between public and private school users. Parental employment status is indicated in the three panels of table 5. Across the state private school parents were more likely to be employed full time and less likely to be unemployed. Again, the contrasts are most pronounced in the City of Milwaukee. From 28 to 29 percent of the public school families reported neither parent being employed. This is in contrast to 4 to 5 percent private school parents being unemployed. Again, in the suburbs the two groups look very similar, with exceptionally high employment levels for both public and private school families. Statistics in table 6 are for households and are presented in three categories: households with students only in public or only in private schools, and households with students in both. The statistics provide a uniform picture of higher socioeconomic status for households with at least one child in private school. Statewide, private school households were more likely to headed by a married couple; household income was approximately 23 percent higher; the likelihood of being on public assistance was considerably less; parent education was higher (by.9 years); and property values and property taxes were more for the private school households. As with other data, the contrasts are considerably more extreme in the City of Milwaukee. This is especially true when comparing the largest groups-public school households-with households having students only in private schools. For example, although household income of both groups in Milwaukee was below the state average, private school household incomes in the City of Milwaukee were 69 percent higher than public school incomes ($42,583 compared with $25,203). Thirty-five percent more private school households were headed by two parents, and the householder had, on average, 1.7 more years of education. Finally, Milwaukee public school households were over eight times more likely to be on public assistance than Milwaukee households with children only in private schools (25.8% to 3.1%). The contrasts between the city and suburbs were stark for both public and private school users. As with earlier results, the differences between public and private school suburban households were less pronounced because

5 both sets of households were relatively affluent, well educated, and living in expensive homes. On all of these measures, the differences between public school households in the city and the suburbs were greater than the differences between private school households in the respective locations. The widest differences of all, however, can be obtained by comparing City of Milwaukee public school families with suburban Milwaukee families who have at least one child in private school. Policy Implications Private school students and their families in the Milwaukee Parental Choice Program are very different from the private school students and families in the rest of Wisconsin and from other students and families in the City of Milwaukee. The average student in the Choice Program is not white, is from a single-parent family with a very low annual income and a high probability of being on public assistance. In contrast the average private school student in both the state and in the City of Milwaukee is white and comes from a two-parent family with a considerably higher than average income and very low possibility of being on public assistance. Private school families throughout the state and in Milwaukee are considerably higher on every socioeconomic measure than the average public school user. Students in private schools tend to come from households with higher incomes, more education, more two-parent families, higher employment rates, and much lower incidence of being on public assistance. Finally, the contrasts between public and private school students and households are most extreme where private school use is heaviest-in the city and suburbs of Milwaukee. Changes in participant demographics in an expanded Choice Program. Given the sharp differences between public and private school households in Milwaukee, it is likely that if the Choice Program expands according to the governor's current proposal, the socioeconomic level of families and the percentage of white students in the program are likely to increase. Several safeguards exist, however, against too great a change in the short term. First, the income limit of 175 percent of the poverty line will remain. Second, schools will still be required to select students randomly if they are oversubscribed. And, third, students already in private schools would not be eligible. Changes are likely, however, because there remains considerable difference between the actual income of current Choice Program applicants and the income limit of the program. For the first four years of the program, the average applicant had family income just under $12,000, while the program limit for a family of three was approximately $21,000. Thus we must anticipate that some families with higher incomes who would or are currently using private schools will enter the program, causing the average income level to rise. This conclusion is based not only on our estimates of the income of the average private school student in Milwaukee, but also on data on a privately funded, low-income scholarship program (Partners Advancing Values in Education, or PAVE) which parallels the Choice Program. That program has income limits similar to the Choice Program's, but allows students to attend religious and independent schools while paying half the tuition for the students (up to $1,000). It enrolled 2,370 students in (compared with 742 in the Choice Program) and gave 95 percent of its scholarships to students in parochial schools.6 One common feature of both programs is that the mothers of enrolled students tend to have more education than the average MPS parent. The PAVE Program differs from the Choice Program in several important respects, however. The Choice Program has very few white students (5%). In contrast, of the parents of students in the PAVE program, "roughly half" (46% of the females and 52% of the males.) are white.7 In addition, while only 25 percent of the families in the Choice Program have two parents, 43 percent of the PAVE participants come from two-parent households.8 These higher incomes and more prevalent two-parent families are undoubtedly due in part to the tuition-matching

6 requirement of the PAVE program, but some of the higher socioeconomic status is also due to the inclusion of families attending religious schools. The increased voucher program proposed by the governor, with much higher levels of school subsidy ($3,200), could be expected to displace many of the PAVE scholarships. The result would be a more integrated Choice Program serving families with somewhat higher socioeconomic status. Implications for expanded vouchers. If the Choice Program conditions are relaxed (such as dropping geographic and income limits, eliminating random selection, and dropping the ban on current private school students), we can anticipate a major shift in beneficiaries of the program. If most current private school families would continue to apply to private schools, and most private schools would continue to admit students similar to the ones they admit now, an open-ended voucher program would clearly benefit households that are more affluent than the average household in Wisconsin. If religion is a primary motivation for private school enrollment, and a voucher program is in effect, we would expect that motivation to remain with a voucher program in effect. Thus the current characteristics of public and private school users are a useful guide for what would occur under a general voucher system. One could argue, however, that with vouchers available to everyone, private schools would change considerably-essentially opening them up to the poor. But the opposite argument seems equally plausible. With more money available, private schools that now cannot afford to be selective (such as inner-city private schools) could become more selective. And without regulations preventing tuition increases, already highly selective schools could maintain that status by requiring add-on payments in addition to vouchers. New private schools undoubtedly would develop to serve families who have vouchers to spend. But again it is unclear how many new schools would spring up, where they would be located, and what type of clients they would seek. It is impossible to predict the answers to these questions. There are no dynamic equilibrium models to simulate even estimates of new school creation. The existing private school market provides mixed signals. Clearly, private schools cater to families with higher socioeconomic status, although the range is considerable. Nonreligious private schools are in general not the inner-city schools in the Choice Program. They are highpriced schools that provide "elite" education. Approximately half the private schools in the United States are Catholic. Catholic, and undoubtedly other religious schools, consider aiding the poor part of their mission. For example, Milwaukee Archbishop Weakland, in commenting on the prospects for an extended voucher system, recently said, "Others fear that the private schools would then take only elite children and leave those with special difficulties behind. I can say without equivocation that Catholic schools would be eager to educate all if they had the funds to do so. Such challenges are not foreign to our tradition."9 This may be true, but the Archbishop does not make decisions for individual parish schools or for independently run Catholic high schools. And repeated studies confirm that those currently attending Catholic schools remain above the average in socioeconomic status.10 Whether that would change with an influx of public money is simply not clear. Other issues. Several issues in the voucher debate are beyond the scope of this paper. One finding that proponents of broad-based vouchers can use to support their view is that current private school families pay considerably more property tax than public school families (22% more statewide). And because their children do not attend public schools, they get much less in return. Other arguments involve issues of church and state separation and the potential regulation of private schools which might result if this separation is relaxed to allow vouchers. Finally, there is the matter of funding such a program. Is it good public policy to use public monies to fund what until now have been private choices? And can the state afford such a venture?

7 Notes 1. See John F. Witte, Christopher A. Thorn, Kim M. Pritchard, and Michele Claibourn, "Fourth-Year Report: Milwaukee Parental Choice Program" (Madison, Wis.: Department of Public Instruction, December 1994). 2. The subset of data from the U. S. Census used in this study was purchased from the Wisconsin Applied Population Lab at the University of Wisconsin-Madison with private funds of John and Mary Witte. Time devoted to this project by the authors was volunteered and strictly separate from their other duties as faculty, staff, and students at the University of Wisconsin-Madison. 3. For example, of the 109 private schools in Milwaukee, only 22 are not religious. 4. For the sake of brevity, we will refer to the kindergarten through eighth grade population as "elementary" students. 5. One longstanding "problem" with census data is that Hispanic origin is not considered a racial category. Thus those of Hispanic origin must select one of the racial groups. There is little reason to expect, however, that public and private school Hispanics would select differently. Thus the relative comparisons are probably quite accurate. 6. Janet Beales and Maureen Wahl, "Given the Choice: A Study of the PAVE Program and School Choice in Milwaukee," The Reason Foundation (Los Angeles, January, 1995), p Ibid., p Ibid., p Catholic Herald, February 16, 1995, p See, for example, Anthony Bryk, Valerie Lee, and Peter Holland, Catholic Schools and the Common Good (Cambridge, Mass.: Harvard University Press, 1993), especially chapter 7. For an update on this literature see John F. Witte, "School Choice and Student Performance," Paper presented at The Brookings Institution Conference on Performance-Based Approaches to School Reform, Washington, D.C., April 6, Table 1. Where Do Wisconsin Students Go to School ( )? % in Public School Milwaukee Madison Racine Green Bay Kenosha PK Total % in Private School PK

8 Total % in Public School Five Largest Suburbs of Suburbs of Rest of Cites Milwaukee Madison Wisconsin PK Total % in Private School PK Total The first six rows of percentages describe what percent of students in Wisconsin were enrolled in public and private schools in , broken down by grade level, for each of the five urban areas listed. For example, the first row of Table 1 indicates that for all Wisconsin students in grades PK through 8 attending public schools, 12.27% of those students are in Milwaukee, 3.01 % of those students are in Madison, and so forth. For Wisconsin students in private schools in all grades (the sixth row), Table 1 indicates that 16.90% of those students are in private schools in Milwaukee, and 2.50% are in private schools in Madison. The last four columns are read the same way. The "Five Largest Cites" column is simply the sum of the percents in the five urban areas. These last four columns sum to 100%. Table 2. What is the Ratio of Private to Public School Enrollment in Wisconsin ( )? Milwaukee Madison Racine Green Bay Kenosha Public/Private PK Total

9 Five Largest Suburbs of Suburbs of Rest of Cites Milwaukee Madison Wisconsin Public/Private PK Total The rows describe the ratio of private school students to public school students in the given areas. Thus, the 13.42% of students in private schools in grades PK through 8 in Milwaukee is divided by the 12.27% of students in public schools in grades PK through 8 in Milwaukee to get a ratio of 1.09, indicating a slightly higher percent of PK through 8 students in private schools in Milwaukee. The same ratio is run for grades 9 through 12 and for all grades (Total) for each of the given areas. Table 3. What are the Actual Enrollment Numbers ( )? Public Schools Private Schools PK PK Total Wisconsin Milwaukee Madison Racine Green Bay Kenosha Five Largest Cities Suburbs of Milwaukee Suburbs of

10 Madison Rest of Wisconsin Table 4. What Are The Income and Wealth Ratios of These Areas (1990)? Average Average Per Ratio of Avg. Ratio of PPP Household Pupil Property Income/ State Value/ State Income in 1990 Value in 1990 Avg. Avg. PPP ($) ($) Value Milwaukee Madison Racine Green Bay Kenosha Five Largest Cities Suburbs of Milwaukee Suburbs of Madison Rest of Wisconsin Table 5. The Race and Parental Employment Status of Students in Wisconsin (1990) 5a. State Public Public HS Private Private HS of Elem. Elem. Wisconsin Race: White African American Hispanic Asian Native American

11 Other Minority Parental 1 or more Employment full-time Status: 1 or more part-time No parent employed (N) (515183) (269489) (117725) (25182) 5b. City of Public Public HS Private Private HS Milwaukee Elem. Elem. Race: White African American Hispanic Asian Native American Other Minority Parental 1 or more Employment full-time Status: 1 or more part-time No parent employed (N) (54714) (26200) (16325) (5856) 5c. Public Public HS Private Private HS Milwaukee Elem. Elem. Suburbs Race: White African American Hispanic Asian Native American Other Minority Parental 1 or more Employment full-time Status: 1 or more part-time No parent employed (N) (67472) (42892) (17138) (9206)

12 Table 6. Household Family Structure, Income, Housing Characteristics, and Education in Wisconsin (1990) 6a. State of Wisconsin Public School Public & Private Private School Family Structure: 75.8% 86.9% 86.2% Married 4.4% 2.1% 2.4% Male Householder 19.7% 11.0% 11.4% Female Householder Income: $39,625 $48,640 $49,202 Mean HH income Receiving Public 7.9% 4.8% 2.4% Assistance: 2.4% 1.1% 0.9% Householder Spouse Housing: $61,800 $68,600 $67,850 Mean property value $1,529 $1,893 $1,870 Mean property tax $444 $496 $471 Mean rent 28.2% 27.7% 23.9% Rent as a % of income Mean Education: (Years) Householder Spouse (N) (418491) (30160) (59,894) 6b City of Milwaukee. Public School Public & Private Private School Family Structure: 45.4% 61.1% 80.0% Married 4.8% 5.3% 3.7% Male Householder 49.8% 33.6% 16.3% Female Householder Income: $25203 $35201 $42583 Mean HH income Receiving Public 25.8% 16.3% 3.1% Assistance: 6.5% 1.7% 1.3% Householder Spouse Housing: $49060 $51122 $57425 Mean property value $1428 $1695 $1883 Mean property tax $443 $473 $473 Mean rent 35.8% 31.8% 25.3% Rent as a % of income

13 Mean Education: (Years) Householder Spouse (N) (46660) (3323) (11183) 6c. Milwaukee suburbs* Public School Public & Private Private School Family Structure: 83.2% 90.0% 89.9% Married 3.5% 3.0% 1.5% Male Householder 13.3% 7.0% 8.6% Female Householder Income: $54385 $67980 $62689 Mean HH income Receiving Public 2.3% 1.9% 1.3% Assistance: 2.2% 3.2% 0.0% Householder Spouse Housing: $83383 $92171 $91153 Mean property value $2413 $2731 $2623 Mean property tax $568 $616 $595 Mean rent 28.9% 36.8% 29.0% Rent as a % of income Mean Education: (Years) Householder Spouse (N) (64163) (3638) (11061) *The Milwaukee suburbs in the census data are composed of the remainder of Milwaukee County, excluding Milwaukee, plus Waukasha County and Ozaukee County. The Milwaukee suburbs in tables 1 and 2 are roughly equivalent to the Chapter 220 Program districts. They include only about half as much area, so the (N) of the two categories will not match.