Book Reviews / 535

Laurence E. Lynn, Jr.EditorBook Reviews

Journal of Policy Analysis and Management, Vol. 17, No. 3, 535–568 (1998)© 1998 by the Association for Public Policy Analysis and ManagementPublished by John Wiley & Sons, Inc. CCC 0276-8739/98/03535-34

Note to Book Publishers: Please send all books for review directly to the BookReview Editor, Professor Laurence E. Lynn, Jr., Harris Graduate School ofPublic Policy Studies, University of Chicago, 1155 E. 60th Street, Chicago,IL 60637.

Eric A. Hanushek

What Money Can’t Buy: Family Income and Children’s Life Chances by Susan E.Mayer. Cambridge, MA: Harvard University Press, 1997, 256 pp., $35.00 cloth.

It surprises nobody to say that parents are important in guiding children’slives. Having said that, most people also go on to conclude that there is asystematic relationship such that children of parents with higher socioeconomicstatus do better in many ways than children from lower socioeconomic statusfamilies. But what are the implications of that, particularly if one thinks ingovernment policy terms?

Susan Mayer takes this problem head-on. There is an extraordinarily simplelogic to her book. Beginning with the common observation that parental incomeis associated with a variety of child outcomes, she questions whether theobserved relationship is a causal one. In other words, if we provided extra moneyto poor families, would we expect to see the performance of children in thesefamilies improve? If yes, there might be a strong case for a variety of transferprograms to the poor. If no, well, what do we do? The answer from her researchis simple: No, the observed correlation does not appear to be causal.

The analytical attack on this question is sophisticated. Mayer recognizes afact that has been oft-repeated but perhaps not taken seriously enough:Correlation does not imply causality. This old adage has been part of theintroduction to basic statistics courses throughout the social sciences, even ifempirical studies too often ignore it. Particularly within the last decade,increasing attention has gone to the possibility that the underlying causalstructure is more complicated than the simple correlations and regressionsmight suggest. Mayer questions whether the income-outcome relationship socommonly found in the data reflects a true causal relationship.

536 / Book Reviews

The heart of Mayer’s work is a series of new empirical investigations thatpursue alternative solutions to this causation problem. The fact is that the datafor looking at family effects are frequently far from ideal. Her own empiricalwork is subject to this. In addition, the solutions to the issues of causality typicallyrely upon a series of strongly maintained hypotheses about the structure ofperformance relationships and about the way the data were generated. Thesemaintained hypotheses simply cannot themselves be tested within the analysis,and any particular conclusions from the empirical work will typically hold onlyif these hypotheses are true, but not necessarily otherwise. The uncertaintythat these two problems introduce leads Mayer to marshal evidence from avariety of analytical approaches.

The contribution is not, however, so much the novelty of any of the individualempirical forays. Instead, it is the unified approach to the study. The analyticalwork would make good textbook material for students who are beginning theanalysis of natural data. By traveling in a clear and organized manner across arange of underlying theoretical constructs and data analysis strategies, Mayerprovides an easily motivated tableau of approaches that could serve as a primerfor students being introduced to regression analysis. In the course ofinvestigating a range of outcome variables—standardized achievement testscores, school completion, and teenage childbearing, among others—and avariety of estimation approaches, she demonstrates not only how to handleand manipulate data to focus on specific issues but also how to interpretstatistical estimates. Her underlying statistical analyses, while individually opento question, collectively add up to a set of forceful conclusions. The plain fact isthat there is considerable consistency across analytical approaches that are notnecessarily correlated with each other.

The results of her empirical work are straightforward. There are significantoutcome differences between children in poor and richer families. But thecombined results of the statistical analysis is that this relationship does notappear to be a causal one.

The implications of her analysis, however, are not straightforward. Almostcertainly her results will enter into the “battle of the sound bites.” One of theever-present arguments for expanding various welfare programs has been the“clear and well-documented” adverse effects of poverty on children. For example,AFDC was motivated in part as a way of providing for the minimal materialneeds of children but in larger part as a way of supporting children’s developmentso that they would have different life outcomes. Those who wish to reducewelfare programs will undoubtedly find Mayer’s results to be the kind of researchconclusions for which they have been looking. At the same time, Mayer doesshow that there are important and systematic differences in outcomes acrossfamilies with different incomes. In other words, just because simple incomedifferences may not be the cause, a substantial problem remains.

The objection that people will have with the analysis is that the true causalfactors are not well identified. The evidence points to a set of systematic butunmeasured factors that are correlated with both income and child outcomes,but these are not convincingly defined or measured. This shortcoming leads toinherent difficulties when thinking about how to address the very real problemsof child development.

I suspect that it will be a while before the discussion turns more seriously toaddressing the problems—as opposed to using the results as convenient debatingpoints. At least a portion of this is natural, because the policy problems are

Book Reviews / 537

truly hard. The policy implications are also the missing chapter of this book.This is not really a criticism of the book, because it has already done considerablework to clarify the issues and to provide serious research. But when one turnsfrom the empirical work in this book to view the policy implications, one is onhis own.

My guess is that it is going to take a long time to digest this overall body ofresearch and to develop sensible policy implications. If I had to speculate, theoverall empirical conclusions are likely to be supported and accepted in therelatively near future, but the policy discussion is unlikely to become eitherfocused or helpful for a much longer period. Two key issues are raised by aconsideration of the policy issues. First, there is the legitimate problem of adversechild outcomes, but society differs in opinions about the appropriate role ofgovernment in addressing this issue. Second, even if there is consensus thatgovernment should intervene, there are serious questions about whatinstruments to use.

Maybe because of my own interests in schools and education policy, I seeparallels to the school resource case. In addition, I see the problems of policydevelopment with family as more severe than those with schools.

The school resource case revolves around the same issues as the familyresource case. Since the Coleman Report [Coleman et al., 1966], questions havebeen raised about whether school spending directly influences student outcomes.My own version of “what money can’t buy,” presented in the inaugural issue ofthis journal [Hanushek, 1981], had much the same character as Mayer’s work.It set out some data and interpretations that suggested little causal impact ofschool spending, even if there were observed correlations between spendingand outcomes. That article also stopped well short of developing the policyimplications.

The reactions to conclusions about school resources are similar to those Imight expect for conclusions about the effects of family income. First, the schoolresource conclusions conflicted with the conventional wisdom of the day, sothat many people were unconvinced (and many still are). Second, and moreimportant, there was a policy disconnect. Legislatures and, in the case of schoolspending, courts find spending to be an ideal policy instrument. The policymakerneeds only to specify how much to put into the activity, and then local schools(families) can make the detailed allocation decisions. But if schools (families)cannot be trusted to turn money into the child outcomes that we want, whatpolicies should we pursue to deal with the problems we see?

Where, on the other hand, do these policy areas differ? For schools, there isan extensive and well-organized lobbying group that is directly affected by thespending itself, that is, all current school personnel. Moreover, given the broadparticipation in public schools, a large segment of the population wants tosupport school quality, which is frequently translated into spending. Althoughthere is a conservative lobbying group that rallies around limiting schoolspending, its strength and its support by the population face natural limitationsthat differ noticeably from that of groups lobbying to limit the amount of welfarefunding and transfers, spending which directly affects much smaller segmentsof the population. Even though there is a “welfare lobby,” the differences inpolitics are still likely to make it easier to get agreement on the basic factualconclusions in the family research than in the school research. A second areawhere the policies differ is that there is much more general support for the ideaof public involvement in the schools. With some 90 percent of the population in

538 / Book Reviews

public elementary and secondary schools, there is little question about a rolefor government policy and intervention. Again, while some groups would liketo lessen the role of government, say, through the use of vouchers, this is not ageneral argument for keeping government out of schooling. With families,however, there is much more debate about when and how the government shouldintervene. Although transfers to families are well established, other policiesinvolving more active intervention are subject to much less consensus. Finally,there currently are a number of ideas about how to improve schools by meansother than pure resource strategies. Specifically, a variety of policies rangingfrom merit pay to private contracting to vouchers involve ways of changing theincentive structure in schools so that they might provide better studentperformance [Hanushek, with others, 1994]. Such a range of policies, even atthe conceptual level, is much less well developed when one talks about families.In addition, the possibilities clearly interact with the willingness to intervene inhow families conduct their normal activities.

My assessment is that improving the functioning of families—at least givencurrent knowledge and the present form of our underlying social contract—is amuch more difficult and uncertain activity than developing policies to improveour schools. Indeed, while the original Coleman Report made the case for theimportance of families in school achievement, few have taken this as a seriouscall for developing family improvement policies rather than school policies.Essentially, there is no agreement from researchers, politicians, or the publicon how to improve families from the perspective of improved child outcomes.Some use “bad families” as an excuse for bad student achievement, in the senseof “how can we hope to improve achievement if we don’t first improve families?”This in turn frequently becomes a call for larger income redistribution policiesto deal with student achievement. I have always interpreted this as either peoplewith a different agenda or people who want to take the heat off of schools.Mayer reinforces the view that improving school performance through simpleredistributive policies is neither easy nor obviously effective.

A broader way to interpret Mayer’s book is that it provides more evidenceabout how little we know about some key social issues. This becomes particularlyrelevant whenever we think about governmental policies that intervene incomplicated social relations.

Mayer’s book is one that can be read from a number of vantage points. It isalso one that will provoke a lot of new thinking about governmental policies,even if it takes some time before this thinking settles down to serious discussion.

ERIC A. HANUSHEK is Professor of Economics and Public Policy at the Universityof Rochester.

REFERENCES

Coleman, James S., Ernest Q. Campbell, Carol J. Hobson, James McPartland, AlexanderM. Mood, Frederic D. Weinfeld, and Robert L. York (1966), Equality of EducationalOpportunity (Washington, DC: U.S. Government Printing Office).

Hanushek, Eric A. (1981), “Throwing Money at Schools,” Journal of Policy Analysis andManagement 1(1), pp. 19–41.

Hanushek, Eric A., with others (1994), Making Schools Work: Improving Performanceand Controlling Costs (Washington, DC: Brookings Institution).

Book Reviews / 539

LaDonna A. Pavetti

It Takes a Nation: A New Agenda for Fighting Poverty, by Rebecca M. Blank. NewYork Russell Sage Foundation, 1997, 340 pp., NPA.

It Takes a Nation is one of those rare books that meets all the standards ofacademic rigor but is attractive to a broad range of audiences. For policyresearchers, it provides a great compilation of useful facts on poverty, theeconomy, and the effectiveness of antipoverty programs, all organized in anaccessible manner. I found myself turning to the book for information evenbefore I had time to read it cover to cover. For professors, especially in publicpolicy schools, it provides sufficient information and a broad enough scope toform the foundation of a course on poverty policy. For public policy students, itprovides an example of policy analysis at its best, using the distillation of acomplex set of facts as the basis for policy recommendations that take intoaccount fiscal and political constraints and organizational challenges. Forprogram administrators, the audience to which this book is targeted, it providesa concise review of the evidence on whether various antipoverty programs haveachieved their intended outcomes, identifies management practices commonlyfound in successful programs, and presents new ideas on how to think aboutthe design and structure of antipoverty programs. In short, this book offerssomething for just about everyone who has an interest in understanding povertyand using the knowledge that we have amassed over the years about what worksand what does not to rethink how we might reduce poverty in this country.

Blank uses the first three chapters of her book to examine the changing natureof poverty, the economy, and America’s efforts to provide a social safety net. Inthe fourth chapter, she provides a review of the evidence regarding the outcomesof antipoverty programs. There is little information that is new in these chapters;what makes them stand out is the clarity with which the information is presentedand the extremely balanced presentation of a complex set of facts. Blank usesfacts to dispel numerous myths about poverty and antipoverty programs, butin doing so she does not dismiss the very real problems that contribute to thosemyths. For example, Blank presents evidence to show that contrary to publicperception, the majority of the poor do not live in inner-city underclass areasbesieged by crime. At the same time, she acknowledges that these neighborhoodsare “often desperate places, where children are at risk from dying from randombullets . . . where too many residents live with fear.”

As is true in her previous work, one of the major contributions Blank makesto the discussion of poverty in this country is an extremely clear discussion ofthe relationship between the economy and poverty. If one reads this book anddoes not come away convinced that the fate of the poor is intricately tied to thestructure of the economy, one is simply not willing to be convinced. Blankpresents clear evidence that the current economy holds limited opportunitiesfor workers with minimal skills, making employment only a partial solution toreducing poverty.

What makes Blank’s summary of the effectiveness of antipoverty programsunique is that she examines the evidence on whether programs have achievedtheir intended outcomes. Thus, she asks whether cash transfers reduce poverty;food programs improve nutrition; medical programs improve health; orneighborhood redevelopment programs revitalize neighborhoods. Because sheframes her questions so succinctly, she is able to review research findings from

540 / Book Reviews

a broad range of studies and present them in a way that is easy to absorb. Thepicture she presents is not all rosy—she acknowledges that while some socialprograms have achieved exactly what they set out to achieve, others have fallenfar short of expectations.

In the final three chapters of It Takes a Nation, Blank moves away from thepresentation of facts and begins to grapple with who should help the poor andhow we can redesign antipoverty programs, taking into account changingdemographic, economic, political, and fiscal realities. Blank acknowledges thatthere is a big and important role for the private sector to play in addressingpoverty in America. She also concedes that there are times when we have keptpower and authority over some social programs in the hands of the federalgovernment when better outcomes might have been realized if more powerand authority had been delegated to states and localities. But in the end Blankargues that the federal government must ultimately bear the responsibility forguaranteeing the existence of a safety net for the poor. The cyclical nature ofthe economy combined with inevitable state fiscal constraints resulting frombalanced budget requirements and the limited resources available through theprivate sector would make it difficult, if not impossible, for state governmentsand the charitable sector to sustain antipoverty programs when they are neededthe most.

Blank’s discussion of the trend toward targeted programs is one of the mostinteresting parts of the book. She provides a thoughtful interpretation of whywe have started to see movement away from broad-based cash assistanceprograms and toward programs that are aimed at particular groups among thepoor and that are focused on specific behaviors. Her discussion highlights twotypes of targeted programs: behaviorally linked support programs (like foodstamps, Medicaid, and the Earned Income Tax Credit) that tie assistance toparticular behaviors and targeted service programs (like job training or economicdevelopment programs) that provide specific services with the intent of changingthe behavior or environment of the poor. Her discussion of the advantages anddisadvantages of shifting to a safety net that relies on more focused antipovertyefforts provides an excellent framework for beginning to think about the kindsof targeted programs that might replace current broad-based programs andhow such programs could be sold to policymakers who are increasinglyambivalent about spending money on programs to help the poor. Blank viewsthe shift to more focused programs as an opportunity to design and implementmore effective programs. She offers specific criteria for judging whether targetedprograms are likely to be effective, but is careful not to oversell what mightreasonably be expected from such programs.

Building on her discussion of targeted programs, Blank recommends thatthe current patchwork of antipoverty programs be replaced with a three-tiersystem of family assistance. The first tier would provide short-term aid to familiesin crisis. The second tier would provide job search and training assistance tothe potentially employable, and the third tier would provide cash support forgroups who need ongoing cash assistance either because their wages areextremely low or because they cannot find work. Ideally, the three-tier systemwould be supported with a number of initiatives to supplement income andmake work pay, including the Earned Income Tax Credit, Child Care Tax Credits,and Assured Child Support.

Although Blank provides some details on how such a system would work, sheleaves some of the important elements for her target audience, program

Book Reviews / 541

administrators, to work out on their own. The one area in particular where Iwould have liked to have seen more discussion is on the difficulties ofdifferentiating who would be in what tier. Blank acknowledges that there willnever be a foolproof system of determining who needs what kind of assistanceand therefore who belongs in what tier, but she does not go the next step tooffer suggestions on what the options might be for carrying out this extremelycritical task or what criteria a program administrator might use to judge whetheror not a particular strategy might work.

It Takes a Nation provides a pragmatic approach to thinking about andredesigning programs for the poor and is a welcome addition to the literatureon poverty and antipoverty programs. Blank’s ability to present and interpret abroad set of facts and then use those facts to construct practical policyalternatives makes her work stand out from others in the field. She puts herselfvery much at the center of today’s debate on how to best help the poor, makingit hard to ignore both her interpretation of the issues that confront us as anation and her plea that we all bear some responsibility for addressing theproblems that create poverty in our country.

LADONNA A. PAVETTI is Senior Researcher at Mathematica Policy Research, Inc.

Brian K. Gran and Judith A. Levine

Faces of Poverty, Jill Duerr Berrick. New York: Oxford University Press, 1995,214 pp., $25.00 cloth.

Children in Courts: Public Policymaking and Federal Court Decisions, by SusanGluck Mezey. Albany, NY: State University of New York Press, 1996, 206 pp.,$20.00 paper.

Whose Welfare? AFDC and Elite Politics, by Steven M. Teles. Lawrence, KS:University Press of Kansas, 1996, 226 pp., $30.00 cloth.

The most recent round of welfare reform raises several important questions.Why did the Aid to Families with Dependent Children (AFDC) program provideonly meager benefits throughout its history and ultimately become susceptibleto dismantlement? How well does the public image of welfare recipients as lazywomen with thwarted values (the very image that eased passage of time-limited,work-based, welfare legislation) reflect the reality of recipients’ attitudes andbehavior? How is it possible that a program created in the interest of childrencould be dismantled with little attention to the effects on children, whocomprised two-thirds of its beneficiaries? The books by Steven M. Teles, JillDuerr Berrick, and Susan Gluck Mezey each address, directly or indirectly, oneof these questions. Taken together, they provide clues to understanding whyAFDC was vulnerable to dismantlement, whether images of “welfare culture”are valid, and why children’s interests are often discounted in policymaking.

Teles weaves an intricate story of the influence of elites on the Aid to DependentChildren (ADC) and AFDC programs, focusing on the conflicts between elitesand how their actions contributed to the dismantlement of the AFDC programafter a 60-year history. Berrick’s portraits of poor single mothers and theirchildren reveal the often overlooked realities of life in poverty and themisunderstood motives of public aid recipients. Mezey further highlights the

542 / Book Reviews

plight of children. She argues that the U.S. Supreme Court’s hesitation tointerfere with the authority of states, combined with its analytical frameworkof constitutional rights and protections, have left children vulnerable to witheringsocial policy benefits.

Through Whose Welfare? AFDC and Elite Politics, Teles seeks to demonstratehow cultural and intellectual elites prevented AFDC from changing to conformto public opinion. A key contribution made by Teles is the concept of “elitedissensus.” By dissensus, Teles has in mind contentious politics in whichopposing sides struggle over fundamental ethical issues. In contrast to elitedissensus, he suggests that “normal U.S. political forms” are “incrementalismand dominance by interest groups.” Teles contends that comprehensive changeto U.S. social policy is most likely to occur through “consensus politics,” whenelite policy efforts align with public opinion to overcome interest grouppressures.

Although Teles’s primary concern is AFDC’s problems over the last 30 years,he provides an interesting discussion of the history of AFDC leading up to the1960s. Teles describes the background and development of the AFDC programthrough the use of a variety of data sources, including census data, AFDC benefitlevels and recipient numbers, and historical descriptions of programs existingbefore ADC. After providing this synopsis of AFDC’s history, he presents ananalysis of public opinion on welfare. Based on survey data, Teles argues thatthe U.S. public strongly values “work,” believing it enables people to join theU.S. mainstream. The public perceives contemporary U.S. “welfare” and“poverty” programs as failing to support the value of work because the programshave discouraged recipients from joining the paid labor force. Following hisstudy of public opinion, Teles examines the actors and institutions that wereimportant components of elite dissensus for the 1960 to 1995 period. Telesevaluates government and media reports, academic publications, and the publicacts of political leaders to form his conclusions on elite dissensus and AFDCpolitics.

Teles reviews the efforts of the National Welfare Rights Organization (NWRO)in the 1960s. According to Teles, the NWRO used litigation and other strategiesto promote changes in the laws and administration of the AFDC program.Working within the extant policy and institutional structure, the NWRO soughtto ensure that all potential AFDC recipients received benefits. This strategy wasintended to overwhelm the AFDC system and its administrators with benefitapplications, in the expectation that AFDC administrators could not afford thetime and resources needed to deny applications on a discretionary basis.Consequently, in effect AFDC would shift from a discretionary benefit to anentitlement. The NWRO’s efforts came into conflict with President Nixon’sproposed Family Assistance Plan (FAP), which was based on a negative incometax. Teles suggests that the NWRO sought to weaken the political support forFAP. The NWRO opposed FAP for a variety of reasons, but Teles focuses onFAP’s combination of modest benefit increases for NWRO members and theelimination of the AFDC program framework, which the NWRO had learned tomanipulate to its members’ advantage. Nixon’s FAP ultimately failed becausecoalitions could not form to support it, although Teles indicates that parts ofthe program satisfied both conservatives and liberals.

Teles contends that elite dissensus inhibited policy change at the federal levelin the 1980s and early 1990s. As a result, decisions affecting AFDC were madeat the state and local government levels. During a climate of popular demand

Book Reviews / 543

for change to welfare programs, an open and consistent waiver policy from theexecutive branch encouraged governors to experiment with AFDC programs.By the late 1980s, however, the federal government was entering an era of “newwelfare consensus.” This new consensus, which was tenuous, was based on theagreement between political parties that the poor should abide by middle-classnorms and that the federal government should provide resources to aid themin their efforts. Teles suggests that the Clinton administration was poised tochange the AFDC program in a substantial and lasting way, but failed becausethis consensus supporting its efforts was frail.

Two related questions a reader may have from reading Teles’s book are: Whatlevel of elite consensus is needed to permit progress on social policy, and whatlevel of elite dissensus inhibits social policy formation or allows social policydismantlement? Few social policies have enjoyed unanimous support from elitesor the public. One approach to evaluate the impact of elite dissensus is tocompare two or more social policy programs that were both subject to elitedissensus, one which was dismantled and one which was not, for example,AFDC and Old Age and Survivors’ Insurance (OASI), commonly referred to asSocial Security retirement benefits. Prior to the implementation of SocialSecurity, academics, lawyers, and policymakers struggled to overcome U.S.Supreme Court precedent that opposed governmental provision of income[Sterett, 1990, p. 234]. Even after the 1935 Social Security legislation was passed,some experts expressed concern that the U.S. Supreme Court would hold thelaw unconstitutional [Douglas, 1936, pp. 306–307, 320–321]. Today, the SocialSecurity retirement program is considered resistant to welfare stateretrenchment, despite concerns about its financial well-being. Awareness ofthe boundaries of dissensus and consensus will allow creativity in social policydesign without the loss of agreement in such endeavors.

Although elite dissensus provides an intriguing explanation for the demise ofAFDC, a reader may wonder about the importance of other factors behind AFDCretrenchment. For example, what role has the targeted, nonuniversal nature ofAFDC played in its lack of durability, particularly given that its beneficiarieshave been publicly represented as marginal to mainstream society? Therecipients of AFDC were frequently disparaged in the U.S. media and by others,whose depictions sometimes stepped close to sexism and racism [Amott, 1990;Quadagno, 1994]. Middle-class families did not enjoy AFDC’s meager benefits.Some analysts suggest that the provision of a children’s allowance to all families,not only the poor, bolsters the political durability of family allowance programsin other western countries [Goodin and LeGrand, 1987]. Social programs inthe United States that have a broad group of recipients seem to hold greaterappeal than residual programs. Future research on the demise of AFDC shouldexplore whether its dismantlement resulted from a combination of elite dissensusand other factors.

Teles argues that AFDC would have been stronger if its receipt had been basedon participation in the paid labor force. Paid work, according to Teles, has aspecial resonance for the U.S. public because it is deemed honorable and themeans by which individuals assimilate. Whether AFDC receipt based in workwould have improved the durability of AFDC is not as clear as Teles suggests.Another income-tested program that has a strong connection to paid labor forceparticipation is Unemployment Insurance (UI). Although a person becomeseligible for UI on the basis of paid employment lasting a year, the UI programwas subject to retrenchment attacks in the 1980s [Pierson, 1994, p. 115]. Pierson

544 / Book Reviews

describes the impact of the Omnibus Budget Reconciliation Act on UI, notingcutbacks to its extended-benefit program, a tax on some recipients’ bene-fits, and restrictions on federal loans to states’ UI trust funds (pp. 118–120).Other features shared by AFDC and UI perhaps made them vulnerable toretrenchment. The numbers of AFDC and UI recipients were small. Bothprograms were the responsibilities of both the federal and state governments,which presented the opportunity of shifting federal responsibilities to the states[Pierson, 1994].

Another aspect of paid work raises questions about its importance for thedebate over the AFDC program. Child- and elder-care responsibilities, whichare most often borne by women, demand flexibility in the paid labor market[Pearce, 1990]. At the same time, nonpaid workers are not entitled to SocialSecurity retirement benefits or disability insurance. The reliance of paid workerson spouses, children, grandparents, and their paid substitutes to participate inthe paid labor force, and the failure to grant social insurance coverage fornonpaid labor, suggest that the U.S. public does not value all forms of work.

Teles’s argument that the public’s dissatisfaction with AFDC was partiallyrelated to the fact that AFDC was not a work-based program raises questionsabout the public’s perception of AFDC recipient behavior. How accurate arepublic images of welfare recipients as isolated from the paid labor market? Weknow from the literature on welfare dynamics that a majority of those whobegin a spell of AFDC leave within two years, often through employment [Harris,1993] and that some portion of AFDC recipients perform paid labor and receivewelfare benefits simultaneously [Edin and Lein, 1997; Spalter-Roth, Hartmann,and Andrews, 1992]. In her book Faces of Poverty, Jill Duerr Berrick furtherillustrates that perceptions of welfare recipients as homogeneous, irretrievablydependent on AFDC, and completely out of the labor market are misguided.Berrick sets out to debunk stereotypes about welfare recipients by illustratingheterogeneity within the welfare caseload population. To do so, she conductedethnographic work, examining the lives of five individual AFDC recipients eachof whom, according to Berrick, represents a different type of recipient.

Berrick selected four of the subjects from the set of respondents to a largerquantitative study of participant experiences in California’s welfare-to-workprogram, Greater Avenues for Independence (GAIN). Because she wanted herstudy to include both short-term and long-term welfare recipients and becauseshort-term recipients are unlikely to be in the GAIN program, Berrick identifieda fifth respondent by placing ads in newspapers and asking welfare office stafffor appropriate potential respondents. In selecting the fifth respondent, Berrickinterviewed 54 women, conducting in-depth multiple interviews with 11 of them.We mention these details of the sample selection methodology because theyrelate to an opportunity on which Berrick did not capitalize, namely the abilityto use the larger sample to comment on how common each of the five types ofrecipients are.

This comment is not meant as a criticism of the methods used, only as anobservation that the access to additional data had additional potential. Themethods used are exactly what is needed in the welfare literature. Althoughquantitative work on the dynamics of welfare usage has grown into a richliterature, we need to know more about how public aid recipients think abouttheir futures and make decisions about program participation. In-depthqualitative interviews are invaluable for such an understanding and Berrick’sefforts in this direction are commendable.

Book Reviews / 545

It is not entirely clear, however, what audience Berrick had in mind for thebook. Berrick uses the qualitative material to demonstrate mainly that somerecipients are short-term users of public aid, some work while receiving aid,some are investing in human capital and are likely to make successful transitionsto work, and some remain on the rolls for a long time. But these are categoriesone could identify with quantitative data and of which, for the most part,researchers in the field are aware. Consequently, the book is more likely gearedfor a general, popular audience who might contribute to the uninformedstereotyping about which Berrick rightly frets and who might not wade throughquantitative evidence in order to have such stereotypes dispelled. If the bookwere geared to a research audience in the welfare field, the qualitative materialwould instead need to be used to provide a more in-depth understanding of oneor more of the important issues on which Berrick touches, but does not elaborate.

One of the book’s strengths is its descriptive detail derived from the life historiesBerrick tirelessly collected. But somewhere in the thick description, a certainclarity of argument is lost. For example, it is not clear that several of Berrick’srespondents are really all that different from each other on the variables ofinterest, that is, attitude toward work and likelihood of ending a spell of welfare.Berrick attributes a lot of the failure to work to the fact that work does not pay,but then describes “indolence” as at least partially responsible for the long-term public aid receipt of one of her respondents. What does Berrick think isthe nature of interaction between these causes? If good paying jobs wereavailable, would this respondent overcome her indolence? Because the “indolent”respondent attended a basic skills training program five days a week for overfour years and acted on an entrepreneurial spirit by making and selling tacosout of her home, it would also be useful here to distinguish between a cultureof poverty and participation in the informal economy.

Berrick concludes with a summary of 10 types of reforms that have beenproposed or passed in relation to welfare and discusses their likely effects onthe welfare population. We would have preferred it if she instead used theopportunity to discuss the implications of the endeavor of the book—what shethinks the diversity she found means, what specific ways the five women differand what characteristics they share, and what policies would address theirindividual needs. Another detail worth mentioning is that Berrick often assertsfacts without citations and some “facts” (both those with and without citations)are actually somewhat contentious in the literature. For example, while Berrickis likely correct that adolescent parenthood has a negative effect on women’sfinancial well-being, the exact nature of that effect remains elusive andcontroversial. Some acknowledgment of such debates in the literature wouldbetter serve readers, although we sympathize with the hesitation to get into theintricacies of causal analysis for a lay audience.

Berrick’s book highlights the double-edged sword of the facts we have learnedabout welfare dynamics. Most of those who begin a spell of welfare will end itwithin two years. Those who opposed the passage of time limits on public aidreceipt pointed to this fact as a reason for why time limits are unnecessary. Butthose who reluctantly supported the welfare reform law likely used this fact asa reason to rest assured that time limits would not have dire effects. The problemwith this thinking, as Berrick wisely points out, is that while the majority ofthose who start a spell of welfare leave the rolls quickly, the majority of thosereceiving welfare at any given point in time remain, absent time limits, on therolls for many years. In addition, even those who leave welfare quickly are likely

546 / Book Reviews

to return [Pavetti, 1993]. Thus, lifetime time limits affect both short- and long-term recipients. By allowing us a close view of how some recipients use welfareas a means of investing in human capital in order to create above-povertyincomes in the long run, Berrick’s work suggests that one tragic effect of timelimits will be to deny that segment of the welfare population that could usepublic aid receipt as a route to self-sufficiency the opportunity to do so. Mostimportantly, Berrick’s investigation of the diversity within the welfare populationdemonstrates the foolishness of one-size-fits-all welfare policies. Recipients aredifferent from each other and have different needs. In order to create sustainabletransitions to work, some would best benefit from getting a taste of work, seeingwhat skills they lack to do the work they want, and then returning for training[see Herr, Halpern, and Conrad, 1991]. Others would do best by getting theirtraining first and then seeking employment. Still others may not be able tosucceed in work or school until problems such as domestic violence, depression,drug abuse, or other work obstacles are addressed.

The AFDC replacement program, Temporary Assistance to Needy Families(TANF), allows states the freedom to create flexible programs tailored to a diverseset of needs. This flexibility is in fact what some states argued they lacked underthe old federal system. The great test of the welfare reform law will come nowthat states are implementing their new programs—Will states use their newcontrol to better serve recipients or will they use it primarily to cut needy familiesoff the rolls?

Berrick’s poignant descriptions of her respondents’ children’s needs, some ofwhom suffer severe physical ailments, underscore the irony of recent welfarereform debates. As Teles writes, mothers’ pensions began in response to children’sneeds. Policymakers argued that children’s interests were best served when theycould remain living with their mothers. But those who argued for dismantlingAFDC did so based on claims about maternal behavior, paying scant attentionto the effects that removing cash benefits to poor families might have on children.Even the Clinton administration’s assertion that one version of the welfare reformbill would throw an additional 1 million children into poverty did not radicallychange the legislative outcome.

Susan Gluck Mezey’s book, Children in Courts: Public Policymaking and FederalCourt Decisions, offers the opportunity to understand how the social policyprocess can disregard children’s welfare. Mezey considers the influence of acommonly overlooked institution that affects social policy: the federal courtsystem, particularly the U.S. Supreme Court. She examines the federal courtsystem because it is a place where people without sufficient political resourcesto compete in the U.S. democratic system can enforce their rights and seekprotection. Children are included within the category of politically disadvantagedbecause they do not possess legal rights comparable to most U.S. residents.

Mezey focuses on outcomes of lawsuits affecting the status of children,primarily those of the U.S. Supreme Court, for the 1953 to 1993 period. Overthese 40 years, the Supreme Court was led by three different chief justices:Warren, Burger, and Rehnquist. In addition to studying court opinions, Mezeyinterviewed child advocates about their perspectives on the federal courts’changing reception to lawsuits brought on behalf of children. These childadvocates are pessimistic about the role the federal courts will take in improvingthe lives of children.

In her effort to assess the degree to which the U.S. Supreme Court hasinfluenced public policies affecting children, Mezey studies five policy areas:

Book Reviews / 547

AFDC, Head Start, nutrition for young children, protection from child abuse,and financial support from parents. Mezey draws several conclusions from herstudy. Although the Supreme Court has given more independence to childrenfrom the state and parents, it has recently reverted to an aversion towardinterfering with state policymaking. Over the last 10 years, the Supreme Courthas refused to impose an affirmative obligation on the states to protect childrenfrom abuse and to offer children services and income benefits.

The attention given by Mezey to the Supreme Court is needed. Her focusdemonstrates the changes in approach that the Supreme Court has taken andthe ramifications for policies affecting children. The Supreme Court forms policyin both positive and negative ways. The Court can facilitate social policy byallowing parties the opportunity to demonstrate the constitutionality oflegislation, thereby enforcing the legislation’s legitimacy. The Supreme Courthas on different occasions played an active role in policymaking, for example,opening the door to public education for some minorities. On the other hand,as suggested earlier, the Supreme Court can discourage policymaking.

Mezey suggests that a legal claim affecting a child is unique in two ways.First, the claim typically raises questions about the proper degree of authorityover the child by the state. Second, the claim probably forces a court to intervenein a family’s situation. Mezey contends that U.S. Supreme Court decisions reflectthe tendency of the courts to assign a lower value to the interest of children,relative to the family and state. She concludes that the reason for this treatmentis because the Court tries to protect children, rather than giving them rights. Ifa child is given rights, then the Court must balance the rights of the child againstthose of her family and the state. Protections, on the other hand, are moresubject to interpretation and in some ways do not fit into the rights-basedframework used by the Court. As Mezey points out, while it is difficult to supporta complete reliance on either rights or protections, the decisions of the U.S.Supreme Court in recent years suggest that the Court gives more weight tostate authority than the best interests of children.

Although Mezey should be congratulated for her focus on the U.S. SupremeCourt, further research on the judiciary is needed. Researchers will benefit fromgoing beyond the Supreme Court to examine the influence of lower courts onpolicymaking that affects children. It is true that the Supreme Court is a power-ful player in U.S. social policymaking. Yet, because lower courts, both fed-eral and state, are greater in number and because most cases do not proceedbeyond their courtrooms, lower court decisions probably have a substantialimpact on policy affecting children. Mezey looks at court decisions, but theprocess by which cases get to the Supreme Court and the factors that candiscourage or obstruct appeals are an important part of the policy story. Similarto decisions that do not go beyond lower federal courts, expectations of howlower courts will rule on issues and the anticipation of whether a case willmove beyond a lower federal court on appeal can inhibit policy change andsupport the status quo.

The bottom line for Mezey is that children regularly fall outside the U.S.political system. Children have weak political representation in both thelegislature and the court system. They are not allowed to vote, but must rely onlegislative advocates. Mezey predicts that the Supreme Court will not be able todo much to improve the situation of children in the United States.

Teles, Berrick, and Mezey have presented new avenues for scholarly researchand admonitions to policymakers. Their books contribute to the debates on

548 / Book Reviews

how the United States forms social policy, the factors we entertain and fail toconsider, and what we as a society know about our impoverished families andchildren and our obligations to them. The books raise important questions abouthow we should proceed in devising policy to protect the poor. Teles argues thatchanges can only occur when there is consensus both between competing elitesand between elites and the public. Although his argument has merit, it posescertain challenges to the formation of policy that protects the poor fromhardship. First, while big change may most likely occur during periods ofconsensus, creative, big-scale policies upon which everyone can agree are rare.Elites who agree with Teles’s analysis may instead opt for an incrementalistapproach, making small changes at the margins in a long-term effort to patchtogether a package of programs that combine to protect the poor. Incrementalismmay indeed be the most prudent and effective means of securing protections inthe long run, but it does risk the possibility that the rare, large-scale programwith the chance of passing would not be pursued.

Berrick’s work illustrates a second challenge. The public’s opinions are oftenbased not on the facts of the poor’s behavior, attitudes, and needs but on mythsand folklore about how the poor live. Although policies might need to beconsonant with public opinion in order to pass or be strong, opinions based onmyths can lead to policies that are, at best, unable to help and, at worst, harmfulto the poor. The challenge is to devise simultaneously policy in line withcommunity values and educate the community about the realities of poverty.This is a tall order indeed. Attempts to cast AFDC as representative of the public’svalue of protecting children and to educate the public about the effects of cuttingcash aid to the poor failed to save the program. Mezey’s discussion of children’svulnerability and Berrick’s ethnographic evidence of the hardships poor childrenand parents face demonstrate the dangers of such a failure.

BRAIN K. GRAN is a Robert Wood Johnson Foundation Scholar in Health Policyat the Institution for Social and Policy Studies, Yale University.JUDITH A. LEVINE is a Robert Wood Johnson Foundation Scholar in HealthPolicy at the School of Public Health, and Research Fellow at the Poverty Researchand Training Center, University of Michigan.

REFERENCES

Amott, Teresa L. (1990), “Black Women and AFDC: Making Entitlement out of Neces-sity,” in Linda Gordon (ed.) Women, the State, and Welfare (Madison, WI: University ofWisconsin Press).

Douglas, Paul H. (1936), Social Security in the United States (New York: Whittlesey House).

Edin, Kathryn and Laura Lein (1997), Making Ends Meet (New York: Russell Sage Foun-dation).

Goodin, Robert and Julian LeGrand (1987), Not Only the Poor (Boston: Allen and Unwin).

Harris, Kathleen Mullan (1993), “Work and Welfare among Single Mothers in Poverty,”American Journal of Sociology; 99(2)(September), pp. 317–352.

Herr, Toby, Robert Halpern, and Aimee Conrad (1991), “Changing What Counts: Re-Thinking the Journey out of Welfare,” Research and Policy Reports, NorthwesternUniversity Institute for Policy Research, Evanston, IL.

Pavetti, LaDonna (1993), “The Dynamics of Welfare and Work: Exploring the Process by

Book Reviews / 549

Which Women Work Their Way off Welfare,” Ph.D. dissertation, Harvard University.

Pearce, Diana (1990), “Welfare Is not for Women: Why the War on Poverty Cannot Con-quer the Feminization of Poverty,” in Linda Gordon (ed.), Women, the State, and Wel-fare (Madison, WI: University of Wisconsin Press).

Pierson, Paul (1994), Dismantling the Welfare State (New York: Cambridge UniversityPress).

Quadagno, Jill (1994), The Color of Welfare (New York: Oxford University Press).

Spalter-Roth, Roberta M., Heidi I. Hartmann, and Linda Andrews (1992), “CombiningWork and Welfare: An Alternative Anti-Poverty Strategy,” report to the Ford Founda-tion, Institute for Women’s Policy Research, Washington, DC.

Sterett, Susan (1990), “Constitutionalism and Social Spending: Pennsylvania’s Old AgePensions in the 1920s,” Studies in American Political Development 4, pp. 231–247.

Howard Margolis

Visual Explanations: Images and Quantities, Evidence and Narrative by EdwardTufte. Cheshire, CT: Graphics Press, 1997, 160pp., $45.00 cloth.

This is the third of Edward Tufte’s self-published books on how to use graphicsto present information. Here “self-published” most definitely does not imply“not actually worth publishing.” It just means that Tufte’s standards of how topublish a book that is intended to display outstanding charts and graphs exceedthose of any otherwise reasonable commercial publisher. Tufte is a politicalscientist at Yale. He is what is called in political science a “methodologist,” theanalogue of an econometrician in economics. But it is evident that his greatlove is the use of graphics to clearly and elegantly convey information.

Introducing this latest effort, he says the first [Tufte, 1983] was about how toshow numbers with pictures. The second [Tufte, 1990] was about how to showthings (nouns) with pictures, but things that ordinarily would not fit into a picture,like countries. Now, he says, he provides a book about showing verbs with pictures.

This claim to a systematic sequence is not to be taken very seriously. Tufteplainly just loves this stuff, hence he keeps finding he wants to write anotherbook about it. All the books are full of shrewd insight into his topic, and all area pleasure to read. I am not prepared to recommend that every policy analystread all three, but surely every policy analyst ought to read at least one. For ourcraft warrants, far more than it gets, careful attention to finding ways to presentcomplicated information in a manner that is fair, accurate, and effective. Thecharacteristic mode of presentation for policy work is the verbal presentationaccompanied by slides or handouts. (Tufte, correctly I think, says there is toomuch use of slides, which soon disappear from view, relative to handouts, whichdo not.) But whether the information is on slides or handouts, anyone presentingsuch work who does not take a very serious interest in what makes for effectivegraphical aids in presentation of arguments and information is not taking hisor her own work seriously enough.

This latest installment of Tufte’s reports on adventures in the visuals trade isvery good, but the books in his series are all very good. Indeed, if you are toread only one, you should probably make it the first [Tufte, 1983]. On the otherhand, the latest happens to include a discussion of two cases of special relevanceto policy analysis. Tufte pairs one extremely successful use of graphics withanother which proved to be a tragic failure.

550 / Book Reviews

The favorable case is that of John Snow, a London physician who in 1854investigated an outbreak of cholera that depopulated a district of central London.With the help of a splendid graphic, Snow figured out what caused cholera(thus making a permanent contribution to public health) and with that graphicalso convinced others to act on what he had discovered. The unfavorable casewas the Challenger launch decision that killed six astronauts. Although theChallenger project engineers correctly saw where danger lay, they did not clearlypin down the basis of this unfortunately correct intuition. And then their correctand urgent appeal for postponement of the launch went unsupported by a displayof information that clearly linked malfunctions of the O-ring (their concern) toits cause. So although there was a clear link between decreasing temperatureand increasing problems with the O-ring, which separated the shuttle from theflames of its booster engines, Challenger was sent on its horribly brief flight ona morning so cold that icicles had formed on the launch tower in Florida.

Snow produced a detailed map of central London, on which he displayed wherepeople who died of cholera lived, together with the location of drinking waterpumps. Overwhelmingly, deaths were in the vicinity of a particular pump, whichturned out to be near a leaking sewer. Further, Snow could account for mostanomalies. Several deaths were remote from the fatal pump but involved a victimwho nevertheless used water from that pump; several areas in the immediatevicinity were not affected, notably a brewery, but the brewery did not use the localwater, and neither did its workers, who had access to the beer. So a carefullyconstructed plot of data guided Snow in pursuing his inquiry and also let himquickly convince others of the significance of what he had learned.

But for Challenger, the data were never clearly organized. The project engineershad a correct seat-of-the-pants sense that the cold weather predicted for thelaunch would be bad for the O-rings, but they did not organize their data wellenough to make the basis of their concern transparent to themselves orpersuasive to National Aeronautics and Space Administration (NASA) officialswho had strong incentives to go ahead unless there really was a serious problem.Tufte, of course, is keen to show how strikingly the very data that the projectengineers had in the charts they submitted could have been displayed in a waythat made the risk of the launch as transparent as Dr. Snow’s link between thecholera deaths and a particular water pump. In this, obviously, he has theadvantage over the project engineers of both hindsight and plenty of time towork things over. Nevertheless, he provides a very striking lesson in how muchdifference clear organization and presentation of data can make.

Even Tufte is not perfect on visual displays. At least twice, displays in thebook fail to include (p. 44) or present in only a marginally visible way (p. 75)features described in his text. But this is a book good enough to overwhelmsuch slips.

HOWARD MARGOLIS is Professor of Public Policy at the Harris School, Universityof Chicago.

REFERENCES

Tufte, Edward R. (1983), The Visual Display of Quantitative Information (Cheshire, CT:Graphics Press).

Tufte, Edward R. (1990), Envisioning Information (Cheshire, CT: Graphics Press).

Book Reviews / 551

Thomas Sobol

Inclusion and School Reform: Transforming America’s Classrooms, by DorothyKerzner Lipsky and Alan Gartner. Baltimore: Paul H. Brookes, 1997, 414 pp.,$36.95 paper.

Ethics and Decision Making in Local Schools: Inclusion, Policy, and Reform, byJames L. Paul, Neal H. Berger, Pamela G. Osnes, Yolanda G. Martinez, andWilliam C. Morse. Baltimore: Paul H. Brookes, 1997, 336 pp., $27.00 paper.

“Two Cheers for Democracy,” wrote E. M. Forster [1951]. “Democracy is nota Beloved Republic really, and never will be. But it is less hateful than othercontemporary forms of government, and to that extent it deserves our support.It does start from the assumption that the individual is important, and that alltypes are needed to make a civilisation . . . It allows criticism, and if there is notpublic criticism there are bound to be hushed-up scandals . . . Two cheers arequite enough; there is no occasion to give three” (p. ##)

Forster’s measured approbation seems the appropriate stance from which toappraise these two new books on school-based decisionmaking and the policyof “inclusion.” Surely these currently fashionable efforts at education reformare unlikely by themselves to create the “Beloved Republic.” But they start fromthe assumption that the individual is important, they permit criticism, and theyare less hateful than other contemporary forms of education. They deserve oursupport. Two Cheers!

For the uninitiated, school-based decisionmaking (SBDM, as it is called byPaul et al.) is an effort to get more of the decisions that affect children made bythe people who know them best, the teachers, principals, and parents in localschools. It takes a variety of forms in various jurisdictions, and thus far hasbeen, as we shall see, of unimpeachable intent but dubious efficacy. Inclusion,although the term has now become reified (as in the question, “Does your schooldo inclusion?”), refers to the practice of “including” students with special needsin regular classes as a matter of right and of course. Inclusion and SBDM canbe pursued separately or together. The works under review raise the questions:(1) What do we know about these two initiatives? (2) What should we thinkabout them? and (3) What should we do about them?

Ethics and Decision Making in Local Schools

Ethics and Decision Making in Local Schools is part of an ongoing,comprehensive, and responsible effort at the University of South Florida tocombine ethics and social science in the service of improved school governanceand better circumstances for children with disabilities. As the authors write,“SAC’s (school advisory committees) . . . have extraordinary responsibilitiesand opportunities. Policies may be guided by data, but they are based uponvalues that policy makers consider important and worthy. The decisions thatinterpret those values in practical ways to structure the allocation of resourcesreflect an ethic or set of ethics. Deliberately and consciously or casually andunreflectively, an ethic or set of ethics guides the deliberative process one schoolat a time. It is the purpose of this volume . . . to help define ethical perspectivesand their role in the deliberative process of SACs.”

The structure of the volume reflects the nature of the writers’ work. An opening

552 / Book Reviews

section limns the historical, professional, and legal contexts of school-baseddecisionmaking and the inclusion of children with disabilities. A second section,on the ethical foundations for school policy, lays a base of ethical theory, movingfrom traditional liberal democratic principles and rules to relational ethics basedon “stories” to contemporary feminist and “womanist” perspectives. Third, asection on the social psychology and culture of schools addresses issues of power,school culture, and group dynamics. Finally, a section on ethical decisionmakingin local schools suggests guidelines for developing a curriculum for school-based policymakers. The individual chapters reflect not only the views of theirrespective authors, but much interdisciplinary discussion and the results ofconferences and a three-day “Collegium” with school practitioners andpolicymakers. It is, as noted, a comprehensive and responsible effort.

What do we learn from it? The authors offer an earnest argument and someevidence that school-based decisionmakers can improve the quality of theirdecisions by attending to their ethical implications, especially if the ethics arenot found in rules but in stories that define the local culture and that emphasize“hospitality to the stranger.” The results of their Collegium suggest that the realstrength of school-based teams may be their “ability to construct a moralcommunity” from the diverse people who come together in the school system.(If so, perhaps we had better stop judging the efficacy of school-basedcommittees solely by the rise and fall of test scores.) They express appropriateconcern that children with exceptional needs may fall prey to a utilitarian ethicof the greatest good for the greatest number. They point the way to morereflective, inclusive, and empathetic school communities.

The reviewer wants to believe. Indeed, he does not disbelieve. But naggingdoubts and yet unanswered questions preclude the requisite leap of faith.

For openers, it is hard for this troglodytic white male to swallow whole theconcept of “womanism.” Womanism, apparently, addresses “the interlockingand systemic nature of domination and the ways in which such dominationhas been used to perpetuate compulsory ignorance.” Womanism is not the sameas feminism. “The terms womanist and feminist of color are fairly synonymous,with some similar political, social, and cultural implications. However,historically, African American women have mistrusted feminist ideas and havenot used the label feminist as a descriptor of their liberatory work in AfricanAmerican communities. One of the recurring and painful problems withfeminism and African American women is that feminist analyses and politicsare generally associated with Caucasian women, who, along with Caucasianmen, are the benefactors of systemic racism” (emphasis in original). It is hardfor me to appreciate the full meaning of womanism, but then, I am probablynot expected to do so. All that can be done is to respect as best one can theexperience of African American women and the value of their specialcontributions to society, and move on.

More to the point, the authors themselves are ambivalent in their views ofschool-based decisionmaking. On the one hand, they claim that “one of thechanges with profound implications for school policy is the shift to a philosophyof site-based management,” and they hitch their work and their book to thismovement. On the other, they report that “although SBDMC’s [school-baseddecisionmaking committees] are, for the most part, performing some policy-related functions, we found little by way of deep reform and serious restructuringinitiatives.” They also note that although many educators believe thatdecisionmaking responsibilities should be shared at the local level, “the rhetoric

Book Reviews / 553

associated with the implementation of those responsibilities is, in our experience,far in advance of the reality.” One is left to wonder how real is the movementthey are describing. Whether or not there are formalized arrangements forshared decisionmaking, so long as there are schools there will be teachers andparents and principals who ought to have something to say to one anotherabout what and how the students are doing. Given how important it is thatthese discussions be ethically informed, one wonders if tying ethics to SBDMmay be more trendy than wise.

There are, of course, many reasons why shared decisionmaking at the locallevel has not flourished more quickly and abundantly. There is the inherentdifficulty of the task, the problem of finding time, the need of participants todevelop deliberative skills, and the reluctance of some policymakers andbureaucrats to surrender authority. But most of all, in the reviewer’s experience,there are the conflicts of power among major players. Teachers’ unions,administrator groups, school boards, and state officials vie with one anotherfor their share of the pie. But few seem ready to share authority or influencewith the group arguably most excluded from the table at present—the poor,black, Hispanic, and otherwise ethnically and culturally “different” parents whocomprise the majority in our larger cities. One hopes that the authors will addressthese power relationships more directly as they continue their work.

Finally, one regrets the absence of the stories themselves. The writers arguethat “plugging in formulas based on our values and then calculating the rightconclusion to the individual moral dilemmas that we face” will be of little use—“that ethics is about story, community, and character.” But their volume is devoidof the telling examples that might have demonstrated their point. For the vast,diverse army of inexperienced local decisionmakers, which would be morehelpful: this academically proper set of expository chapters, or a clean, sparselyannotated volume of six stories that helped to create community from diversity?“Show, don’t tell,” said the teacher.

But lamenting the absence of what we miss is not to criticize the gift that wehave received. Ethics and Decision Making in Local Schools remains a usefuladdition to the current literature. If nothing else—and this is much—it remindsus that school talk should be about ethics as well as test scores.

Inclusion and School Reform

If the authors of Ethics and Decision Making are at pains to find the right ethicalstance to inform their deliberations, the authors of Inclusion and School Reformare untroubled by moral uncertainty. They know what is true and good. What istrue and good is inclusion—that “the education of students with disabilitiesshould occur with their age peers in the general education classroom, with thenecessary supplemental aids and support services. Anything other than fullinclusion denies the students’ rights to an appropriate education.” Having thusannounced their position, the authors devote the substance of their book to arecitation of whatever argument and evidence will support it.

And an interesting position it is. Inclusive practices produce better academicand social results for both students with disabilities and those nondisabledstudents in whose classes the disabled students are placed. Teachers and parentslike inclusive practices better than segregated practices. “Data on the fiscalconsequences over time are limited, but most school districts report that theoverall cost of inclusive education is no greater than that of educating students

554 / Book Reviews

in two separate systems and may produce cost savings as the program isimplemented over time.”

Wait—there is more. Not only are inclusive programs more effective, popular,and efficient, anything else we once did or may still be doing is wrong, ignorant,and motivated by prejudice. It turns out that PL 94–142 and the entire systemof due process rights and protections we elaborated over the past quarter centurywere based on a “deficit model (i.e., a model premised on the perceived ‘deficits’of people with disabilities).” The division we have made between general andspecial education—in an effort to provide services to students with disabilitieswho are previously neglected—“is an example of placing blame on the childrather than demanding responsibility from the system.” In fact, in this viewpersons with disabilities have no deficits at all; they unjustly experience thecallous failure of society to make appropriate accommodations to theircondition. “Disability is no longer a personal defect or deficiency; instead it isprimarily the product of a disabling environment.”

Fortunately, according to the authors, enlightened persons have recognized theunjust folly of past practice. More specifically, they have made the transition froma “functional limitations” paradigm to a “minority group” model, in which peoplewith disabilities join the ranks of blacks, women, and homosexuals in seeking toovercome injustice by securing their civil rights. As the text notes, “a growingnumber of disabled people perceive disability as another manifestation of humandifferences rather than as a lack of functional capabilities. In their view, theprinciples of the Brown decision can be applied to disabled students withoutsignificant modification. Separate educational facilities are inherently unequal.”

The authors do a good job of fleshing out these arguments. They provideuseful demographic and financial data, analyze the leading court cases, reviewthe relevant legislation and learning theory, and provide descriptions of effectiveclassroom practice. Unfortunately, all the data they report supports their view.Are there no places where inclusion has not worked well? Are there no teacherswhose lack of compassion or capacity have led them to resist? Is there nobacklash by parents of nondisabled students? Are there, for that matter, no realhandicaps other than those society imposes?

But enough. So long as one remembers that one is reading a tract in academicform, one will find much useful information here. I have no doubt that thisvolume will (and should) be widely read by students of the field so long as thecurrent orthodoxy prevails.

And those of us too old or too cynical to be swept off our feet must still cometo terms with a worldview so passionately argued. What, at the end of the day,are we to think? Having already revealed his distaste for the aroma of politicalcorrectness, this reviewer nonetheless tends to side with Lipsky and Gartner.They present their case in one-sided fashion, but that does not make them wrong.The dual system of “special” and “regular” education may not have been designedto punish the student, but it does not serve us well. Surely our children andsociety would be better-off with schools that respect and provide for the multipletalents of all their diverse students, schools in which children of all backgrounds,abilities, and dispositions are helped to succeed. In an enterprise so right andso important, one can forgive an excess of zeal.

Coda

Most drinks leave an aftertaste. Having sipped from these two volumes, whatlingers on the palate? The flavors that suggest themselves are gratitude and

Book Reviews / 555

sadness: gratitude for two books that skillfully and responsibly amassinformation and arguments in support of good works. But sadness in knowinghow little, in the larger scheme of things, they are likely to affect the schools.These books are carefully made, and useful. But most practitioners andpolicymakers will not read them, nor will all who do so be swayed by theirrational arguments and moral fervor. In most cases, decisions will continue tobe made on uninformed premises and political grounds. The schools will notbe reformed one by one, by thoughtful people thinking ethically; they will change,when they do, in response to the larger social, political, and technologicalimperatives of the times.

Still, the darkness of the void is no reason not to light a candle. Thanks folks,for writing these books—I bid you Two Cheers!

THOMAS SOBOL is Christian A. Johnson Professor of Outstanding EducationalPractice at Teachers College, Columbia University. He is the former Commissionerof Education of New York State.

REFERENCES

Forster, E. M. (1951), Two Cheers for Democracy (New York: Harcourt Brace).

Amy Stuart Wells

Choosing Schools: Vouchers and American Education, by Jerome F. Hanus andPeter W. Cookson Jr. Washington, DC. American University Press, 1996, 160pp., $39.50.

Financing Education: The Struggle between Governmental Monopoly and ParentalControl, by Quentin L. Quade. New Brunswick, NJ: Transaction Publishers,1996, 166 pp., NPA.

Few things are more frustrating than reading books by zealots who paintbroad, unicolored brush strokes across a highly complex and multifaceted terrainof public policy. This explains my frustration in reading two books, one byHanus and Cookson titled Choosing Schools: Vouchers and American Education,and the second by Quade, Financing Education: The Struggle betweenGovernmental Monopoly and Parental Control.

Both books address one of the most significant public policy issues of ourtime—whether parents should be able to spend public money on private schooltuition. Yet each does so in a fairly simplistic and ideologically charged mannerthat does little justice to a contested debate.

The Hanus and Cookson book is by far the better of the two, in part becauseit attempts to look at the so-called “voucher” or “private school choice” issuefrom two oppositional perspectives—Hanus supports vouchers, while Cooksonargues against them. The problem is that Hanus’s contribution is more thantwice as long as Cookson’s and about two times as bombastic; thus, the book israther lopsided. Quade’s book is little more than a diatribe puffed up to a full160-plus pages of ranting, raving, and repetition by a “good” Catholic manconvinced that the state ought to pay for the religious education of his andother people’s children.

556 / Book Reviews

Although I am not an advocate of vouchers myself, I do believe that the pro-voucher, pro-Christian, pro-“traditional” morals perspective that both Hanusand Quade offer should be heard and must be part of a larger debate aboutpublic funds and private schools. Still, it is exasperating to read Hanus andQuade, university professors of political science and government respectively,spout personal opinions masquerading as serious scholarship. Both authorseasily slip into universal statements about the failure of public schools and thewonders of private (particularly religious) schools with little or nodocumentation to back them up. Quade, in particular, quickly reverts to blamingthe public education “monopoly” and the teachers’ union for “all” that is wrongwith public education.

A prime example of the type of unsubstantiated statements common to bothbooks is found in a chapter by Hanus in Choosing Schools where he writes, “wenow realize that our schools do not seem to be doing anything particularlywell. Both national and international achievement tests indicate that Americanpupils do not have grasp of core subjects” (p. 5).

There is no citation and no specific evidence presented to support thisstatement, as if this were the new common sense. In reality, statements such asthis one are controversial among educational researchers. Authors on each sideoffer “full-proof” evidence that U.S. public schools are either falling furtherbehind or doing a miraculous job, serving a larger percentage of the populationthan schools in other countries and closing the achievement gaps betweendisadvantaged and advantaged students [for a good overview, see Schrag, 1997].

Quade’s book is even worse on this account. In fact, the level of analysis herewould have been far more appropriate for a Wall Street Journal editorial thanan academic book. He includes no bibliography or reference section, only anoccasional mention in the text of another publication that he has read. He opensthe book with a broad attack on the “educational finance monopoly” or EFMand a call for an alternative funding method of school choice “without penalty”for parents who choose to put their children in private schools. Without citationor even a description of the so-called EFM and the history of its regulatorynature, he states:

EFM is fundamentally injurious to children, parents, and the nation . . . maintained bypolitical defenses of financial interests, not for reasons of educational merit . . . schoolchoice without financial penalty would create better educational conditions and out-comes for those same children and parents and for the general welfare . . . we realizethat educationally, there is no downside to school choice. (p. 3)

Yet, what quickly becomes apparent is that neither Hanus or Quade (especiallyQuade) is interested in evidence about the educational issues surrounding thisdebate. These are not well-thought out public policy arguments grounded in areflective review of literature regarding what we do or do not know about therelationship between school choice plans and students’ educationalopportunities. Rather, the central purpose of both books is to provide platformsfor two men who believe, due to their religious and philosophical convictions,in providing public funds to parents enrolling their children in private (especiallyreligious) schools.

(One illustration of how off-the-cuff each of these pro-voucher writers tendsto be: Hanus rejects the philosopher Jean-Jacques Rousseau for his “strongdemocracy” argument, which he claims supports the status quo in education.

Book Reviews / 557

Meanwhile, Quade embraces Rousseau as a great thinker who recognized thepowerful reasons for putting ourselves in an imaginary “state of nature”—thatis, outside the box of the current system of funding and organizing publiceducation.)

Quade states up front, in his introduction, that he disdains much of thestatistical evidence about school choice, calling it the smoke screen thatsurrounds the issue and protects the EFM. Arguing that because there are “notrue” school choice programs in the country,1 there is no “proof” or “guarantee”that such programs work. (Ironically, he later cites come research literaturethat supports his argument.) Quade writes that “general knowledge” or “analyticevidence” as opposed to “statistical evidence” should be brought to bear on the“greatest issues of American education” because “the subject matter cannot besubmitted to arithmetic, laboratory comparison, and testing” (p. viii).

Despite his denunciation of evidence other than “general knowledge,” Quadeis fairly circumspect until chapter 2 about which or whose general knowledgehe deems to be valuable in this debate. Here he presents the heart of hisargument: Namely, the central problem with public education is what he calls“ethical inadequacy” or the inability of public school educators to serve as the“ethical extensions” of parents in teaching morals and implementing discipline.This inability, according to Quade, is the result of legal and political restrictionson moral (read religious) pedagogy in the finance monopoly of public education.Such inadequacy, he writes, may be, from parental and family perspectives,“the most fundamental, compelling and ubiquitous of today’s schoolingproblems” (p. 13). Indeed, Quade notes, too often, the public schools havebecome “centers of counter-family values.”

It is on these issues of values and morality, not student achievement oreducational opportunity, that both Hanus and Quade seem to center theirarguments. Both authors are basically writing about a “moral” as opposed toan “academic” crisis within a public education system that has been stripped ofits ability or desire to teach traditional values and beliefs. As Hanus notes, sincethe 1960s when traditional Judeo-Christian values were attacked, theautonomous individual displaced the family as the object of public policypriorities, and religious presuppositions were rejected. In their place, Hanusargues, the government interjected secular assumptions, which led to the“promotion of worldwide birth control methods, expansion of welfare thatcontributed to the expulsion of fathers from the family, state-endorsed abortionpolicies and court decisions that excluded acknowledgment of God and the TenCommandments from schools and public ceremonies” (p. 7).

Thus, after years of recognition within Western civilizations of the centralityof families and churches in the regulation of family life and education, theseand other so-called Great Society programs ushered the government into newareas of private lives such as health care, family planning, welfare, and education.What made matters worse, according to Hanus, is that new taxes were imposedduring the 1960s to force families that were morally opposed to the content ofthese policies to pay for them. “Such disaffection is even more pronounced in

1 The book was written before the second phase of the Milwaukee voucher plan, which wouldallow religious schools to receive vouchers to serve students from the public schools, was sched-uled to go into effect. It was also written before the implementation of the Cleveland voucher plan,which also allows students to attend religious schools.

558 / Book Reviews

education because schooling is compulsory—parents must pay taxes to supportthe public school system, and they are forced to send their children to schoolsthat emphatically offend their values and way of life” (p. 7).

In other words, Hanus and Quade are making moral and cultural arguments,grounded in their own religious beliefs, about who will control the schools andwhose interests will be served. They are both tapping into anxiety and anger onthe part of parents who do not understand why their Christian values do not drivethe curriculum and practices in public schools. In doing so, they both demonstratean embarrassing naiveté about the “moral” bases of other religions and seculartraditions. (Quade notes, for instance, that we seem to forget that secular lawsprohibiting murder have a Judeo-Christian root in the Ten Commandments—asif Christianity were the only religion to claim killing immoral.)

Quade, especially, presents a highly dogmatic and condescending argumentfor why his values and morals should prevail in public institutions such asschools, and if they do not, why he and other parents should be free to taketheir public money and run. Both authors are intent on assuring that childrenneed not be exposed to belief systems or values that run counter to those oftheir parents. They believe that the way to avoid this is to give parents thepublic funds to send their children to schools that are their moral match.Whether or not these schools teach hatred or disrespect toward people of adifferent race, ethnicity, country of origin, religion, or gender than the studentsin the school does not seem to matter to either author. Nor does the fact that, ifparents are free to spend public funds on any private school they like, there willstill be many taxpayers in this country who feel that they are being forced tosupport educational institutions that teach values and beliefs that theyfundamentally oppose.

What makes matters worse is that neither author believes that thegovernment—the source of much evil, they claim—should be able to regulatethe private schools receiving public funds through a school choice plan beyondbasic issues of safety and nondiscrimination in admissions. (Although they wouldbe allowed, under Quade’s proposal, to maintain their current admissionsstandards, which must, in some of the schools include consideration of students’religious background.) They see no problem with public money going to privateschools with little or no oversight of how that money is spent and little or noaccountability to the public taxpayers for the use of these dollars.

In addition, both of these authors seem oblivious of equity issues inherent insuch school choice plans. For instance, they both argue that private schoolsshould be allowed to charge tuition over and above the amount of the voucheror certificate. They do not question who then would really have access to thebest and most prestigious private schools under this scenario or how publicfunds would quickly become subsidies for wealthy parents to send their childrento very expensive private schools. (No, doubt the most prestigious private schools,some of which charge more than $10,000 a year in tuition, would simply increasetheir tuition to absorb the public subsidy and still charge more than most familiescould afford.) Meanwhile, there would be less money left in the public coffer topay for the education of the poorest students.

Cookson’s comments in the Choosing Schools book address several of theseissues. He argues, for instance, that parental accountability is not the finalaccountability and that society as a whole must be responsible for the treatmentof children. Cookson also brings his background as a researcher of privateschools to the discussion, which allows him to dispel the all-private-schools-

Book Reviews / 559

are wonderful rhetoric. But pound for pound, his more reasoned response cannotmatch the political hyperbole of Hanus or Quade.

Read together, these two books are long on polemics and short on substanceor even consideration of alternative ways of framing or approaching the issueof public funds for private school tuition. They make for good sermons, butlousy policy arguments.

AMY STUART WELLS is Associate Professor of Educational Policy at the Universityof California–Los Angeles.

REFERENCES

Schrag, Peter (1997), “The Near-Myth of Our Failing Schools,” The Atlantic Monthly,October, pp. 72–74,76,78,80.

John S. Robey

The Political Economy of Special-Purpose Government, by Kathryn A. Foster.Washington DC: Georgetown University Press, 1997, 288 pp., $55.00 cloth.

Reconstructing City Politics: Alternative Economic Development and UrbanRegions, by David L. Imbroscio. London: Sage Publications, 1997, 210 pp., $42.00cloth, $18.95 paper.

The Dimensions of Federalism: State Governments and Pollution Control Policies,by William R. Lowry. Washington, DC: Georgetown University Press, 1997, 269pp., NPA.

Federalism’s Continuing Fecundity

These volumes demonstrate the complexities and dynamism of federalism andpublic policy today. Each of the volumes addresses different levels of government,and they are excellent case studies of the dynamic nature of contemporarypolicymaking. The authors examine different policies. Kathryn Foster’s ThePolitical Economy of Special-Purpose Government looks at issues such as publichousing, natural resources, flood control, and soil and water conservation thatspecial-purpose governments address. David Imbroscio’s Reconstructing CityPolitics examines urban politics and the normative and empirical aspects ofeconomic growth policies. William Lowry’s The Dimensions of Federalism looksat the state level of government. He develops a model that examines some ofthe interrelationships that exist between the federal and state governments andthe policies attempting to control pollution.

Professor Foster is associated with the State University of New York-Buffalo.Her work is part of a series of forthcoming volumes from Georgetown UniversityPress on policy formulation and implementation Foster’s analysis of special-purpose governments is unique. We usually think of federalism as the typicalthree-level “layer cake.” The levels were, of course, federal, state, and local.Nobody said anything about a “special-purpose governmental” level. However,if there ever was any question about the importance of including special-purposegovernments into our conceptual schema, Ms. Foster dispels that doubt.

560 / Book Reviews

Special-purpose governments are defined by the author as “. . . autonomouslocal governments that provide a single or limited services.” Between 1952 and1992 the number of special-purpose governments grew by some 156 percent.General-purpose governments increased by only 5 percent during the same timeframe. The growth has been nationwide but is most predominant in themetropolitan areas. Half of local governmental expenditures for public housingare made by independent authorities. Special districts (as of 1992) accountedfor 26 percent of local governmental long-term debt.

Professor Foster orients her investigation around four fundamental questions:

• What is the most useful way to think about special-purpose governments?• What factors explain the rise of special-purpose governments?• What are the cost consequences of service provision by special-purpose

governments? [and]• What are the policy implications of specialized governance? (pp. 4–5)

The author employs “institutional choice theory” to address these questions,and she offers a convincing defense of this approach versus “conventionalanalyses.” Conventional analyses are viewed as having overlooked the “specificityand importance of special-purpose governments.” Moreover “institutional choicetheory” understands that “different geographic and financial subtypes of districtsmay respond differently to institutional, economic, political and legal factors....”

The author first explains the different types of special districts. She views thecontinuum of special districts as a mixture of public and private sectorinvolvement. Most special-purpose governments lack police power althoughmany possess the right of eminent domain. What makes special-purposegovernment so potentially influential is its independence from other localgovernments. The districts have “... sovereignty over their administrative andfinancial affairs and enjoy substantial freedom from oversight by their parentgovernment.”

Special-purpose governments are products of enabling legislation from thestate legislature but citizens may also initiate steps that could result in thecreation of such entities. Once created they are governed by a board that isindependent from the authority of the government.

Most special-purpose governments are created to deal with water, housing,community development, fire protection, or flood control. Other purposes thatthese governments serve include irrigation, parks, recreation, soil conservation,health, hospitals, libraries, and highways. Special-purpose governments areinvolved in many policy areas that traditionally have been the responsibility ofcity and county governments. According to the 1992 Census of Governments, therewere 26,783 special-purpose governments in the United States. The majority wereless than one county in size, and about 63 percent had some taxing power.

Foster reviews four theoretical perspectives on metropolitan economies. Shefinds that no one theory offers a satisfactory explanation of the rapid growthand emergence of special-purpose governments. She provides the reader witha new conceptual model that links institutional choice theory with the attributesthat make special-purpose rather than general-purpose governments so popular.One of the products of this model is the identification of six geographic andeconomic system subtypes of special-purpose governments. Each subtype hasdistinct characteristics that influence the importance and use of special districtsin metropolitan areas.

Book Reviews / 561

After developing the six subtypes, the author tests four competing theoriesthat purport to explain the uneven distribution of the district subtypes. ProfessorFoster also tests for the economic consequences of specialized governments.This is accomplished by examining the correlation between per capita servicecosts and the use of special-district governments. By testing for the impact ofthe use of special-district governments and variations in resource allocations,the author is able to analyze policy implications.

Professor Foster concludes that interest groups are more focused on specificgeographic or financial subtypes than on districts as a whole. This is becausespecialized subtypes are more likely to result in service delivery than a specialdistrict. Professor Foster maintains that “. . . the likelihood that a special-purposegovernment will ultimately provide services depends on the government’s . . .legal, institutional, and political environment.” The most important determinantof reliance on special-purpose governments is the legal environment withinwhich decisions are made. Not surprisingly, districts are more numerous wherelegal burdens are modest.

Special districts have not resulted in competition and reduced prices for servicedelivery. In fact, the opposite is the case. Service delivery is found to be moreexpensive in areas that have turned to special-district governance than innondistrict areas. This is due to higher administrative costs and characteristicssuch as financial and political isolation. The most specialized areas of themetropolis (those that commonly rely on the use of special districts) allocatefewer resources for social welfare causes than less reliant areas. The morespecialized areas invest more in economic development. The author’s findingsregarding the relationship of special-purpose governments and economicdevelopment mesh well with the central concern of Professor Imbroscio’sReconstructing City Politics.

David Imbroscio is associated with the political science department at theUniversity of Louisville. His work is one of a series that is being published bySage Publications. Rejecting the traditional liberal and conservative viewpointson urban planning, Professor Imbroscio attempts to merge economic theoryand regime theory. This results in a synthesis that is useful for those who wishto pursue both economic development as well as social welfare investments inchildren, schools, jobs, and assistance for the less fortunate.

This synthesis brings together the “urban political economy” literature withthat of a less developed body of work that stresses the importance of developinga just, compassionate urban political economy. Professor Imbroscio maintainsthat “. . . a consensus exists in the urban political economy field regarding theempirical nature of contemporary central city politics. . . .” The consensus isthe product of two widely held beliefs. First, cities are usually governed by acoalition of public officials and business interests. Second, this alliance producesa pro-economic development agenda for the city. The author argues that thisresults in policies that enhance political inequality and damage liberaldemocracy. The cure for this problem is found in the need to “reconstitute”central city politics.

To understand how such a reconstitution may be achieved, ProfessorImbroscio maintains that we need to understand how urban regimes evolve.Urban regimes are presented as products of “external economic dependence”and “internal resource dependence.” External economic dependence is the needfor urban areas to attract capital, and internal resource dependence is the needfor local community support.

562 / Book Reviews

The author writes that in order for an urban regime to be reconstituted,strategies to reduce both dependencies need to be developed. ProfessorImbroscio points the reader in the direction of urban economic developmentpolicies as the possible sources of such strategies which he refers to as“entrepreneurial mercantilism,” “community-based economic development,”and “municipal enterprise.”

Entrepreneurial mercantilism’s vision sees economic fertility occurring whenthe city is supportive of policies that would be beneficial to small entrepreneurs.Proponents of this school of thought view entrepreneurial mercantilism as oneway to promote innovation and import reduction or replacement. Using the“Homegrown Economy Project” in St. Paul, Minnesota, as a case study, ProfessorImbroscio finds that confronting political barriers is more important in thedevelopment of economic growth policy than are the dangers posed byinstitutional factors. Entrepreneurial mercantilism is essentially individualistic.

The community-based economic development school views economicdevelopment as a product of collective action. The author examines the urbanredevelopment efforts in Pittsburgh, Pennsylvania, and concludes that “. . .despite earlier evidence suggesting the community-based strategy’s possibleeffectiveness, . . . the feasibility of implementing it on a scope necessary to alterthe structural context of urban regimes remains seriously constrained.”

The objective of the municipal-enterprise model is to promote local growthby allowing cities to own profit-making enterprises. This is the least developedof the three strategies, but the author maintains that it has some promise.However, he also acknowledges that it faces serious obstacles. ProfessorImbroscio concludes that the most promising course of action “. . . would be topursue entrepreneurial-mercantilist development goals—stimulating small-business development . . . in ways that nurture and buttress the alternativemeans of capital ownership and control embodied in the community-basedand municipal enterprise strategies.”

With the combination of these two schools of thought, Professor Imbroscio’swork departs from the usual empirical descriptive constraints. It elegantlydescribes the current reality and at the same time allows us to ponder whatcould be. It is reminiscent of “speculative philosophy.”

William Lowry, who is associated with the political science department atWashington University in St. Louis, offers yet another volume relating tofederalism. His work relies on both quantitative and qualitative analyses ofpollution control policies of U.S. state governments. The author argues thatstate policy is shaped by federal influence, state leadership, and competition.He maintains that although his focus is on pollution control his findings areapplicable to a broader spectrum of policies including welfare and health.

Professor Lowry argues that for many years there was justifiable distrust ofthe competence of state governments. However, recently a perspective hasemerged that portrays state governments as the truly innovative component ofthe federal system. State governments are not uniformly innovative and thismakes them ideal units of analysis for comparative examination. One basicpremise of the volume is that if a policy is to be successful it must becommunicated and have organizational support.

The author examines political leadership in states that have developedresponsive policies. He finds that there is a significant relationship betweenstate leadership and the vertical involvement of the federal government, andthe horizontal potential for interstate competition. Professor Lowry maintains

Book Reviews / 563

that his “. . . basic thesis is that state leadership, including the overall matchingof response to need . . . is affected by intergovernmental dimensions of thepolicy involved.”

Using leadership as the study’s dependent variable and relying on variousperspectives of federalism the author builds a model of state behavior. Hemaintains that new responsibilities and respect for the ability of state leadersto successfully implement new policies is due to Baker v. Carr [369 US 186(1962)] and a more equitable representation of women and minorities in statelegislatures. This increased respect is also a result of increased publicparticipation in the policymaking process, fiscal surpluses in some states (forexample, California), enhanced media attention of the record of statepolicymakers, and the increasing professionalization of state politicalinstitutions.

To test for the effect of federalism on state leadership the author relies onquantitative and qualitative analyses. He also selects four states for in-depthanalyses. The selection of Wisconsin, North Carolina, California, and Iowa stemsfrom their high scores on earlier work on policy innovation and their geographicdispersal.

The author’s analysis of stationary source air pollution found that there islittle vertical involvement because of the large degree of autonomy left to thestates by federal legislation. This is most apparent in the problem of acid rain.States like Wisconsin may develop innovative programs, but unless neighboringstates also cooperate, the problem will not be successfully addressed. This is anarea that the author views as unlikely to be resolved by the states alone.

Point source water pollution control is characterized as high in both verticalinvolvement and horizontal interstate competition. The federal government hasspecified the methods to be used to attain clean water standards. There is somecorrelation between the severity of the problem and the state’s response, whichis strongly related to its economic resources. The high vertical involvementallowed for the dissemination of many of the innovative ideas that wereattempted in North Carolina. The author writes that, “[s]tate governments havemade serious progress in the control of point source water pollution since 1972.Much of the innovative behavior in this policy area is taking place at the statelevel.”

Mobile source air pollution is characterized as high in vertical involvementbut low in horizontal competition. The federal government has been veryinvolved in the development of automobile inspection and maintenance (I/M)programs to curb air pollution. States with high levels of mobile source airpollution are found to be more likely to have both I/M programs and morestringent standards. Low levels of interstate competitiveness have allowed stateslike California to develop standards that exceed federal guidelines.

Low vertical involvement and low horizontal competition characterizenonpoint source water pollution. The low federal involvement is attributed totraditions of local control. Some state officials, for example in Iowa, are describedas willing to try innovative ideas that exceed federal guidelines. However, thelow level of federal involvement has made dissemination of the results of stateefforts more difficult.

Professor Lowry concludes that his model of state leadership and federalismdemonstrates that although there is a correlation between the severity of theproblem and the response, the state degree of matching is more apparent whenvertical involvement is high and horizontal competition is low. The less interstate

564 / Book Reviews

competition, the more likely it is that a state may exceed federal guidelines and,the higher the degree of vertical involvement, the more likely that state policiesmay be disseminated and coordinated. In sum, pollution control policies “. . .vary according to the dimensions of federalism.” As the author asserts “. . . it isnot enough to say that institutions matter, or that states matter; rather we mustacknowledge that the institution of federalism itself matters.”

These three volumes are a testimony to the continuing vitality of the federalsystem. Although they each address different levels of government and differentpolicy issues, they nevertheless collectively present the reader with a view ofthe fecundity of the research that is taking place in subnational policymaking.

All three of the volumes address the issue of federalism. Imbroscio places hiswork in the tradition of Alexis de Tocqueville and others who have recognizedthe need for decentralization of authority to protect liberty, and he writes that“. . . decentralization of state power stands as an important buttress for themaintenance of the liberal-democratic political way of life.” Professor Lowrycloses his work quoting Tocqueville’s acclaim for federalism and its powerfulimpact on the policymaking process. Professor Foster’s study of special-purposegovernance found that over half of the districts receive funding from othergovernments. She writes, “[e]specially important is aid from the federalgovernment, which . . . represented 37 percent of all federal aid to localgovernments.”

The state, city, and special-district governments all appear to be growing inimportance as the federal government attempts to divest itself of responsibilitiesand return them to local decisionmakers. These volumes capture this dynamicpoint-in-time in the evolution of American federalism. The authors are to becommended for presenting us with a methodologically rich and diverse set offindings. Their works chronicle the changes that are occurring in therelationships between the various levels of governments in the 1990s. Thesebooks should be of interest to a wide audience, especially those studying publicadministration, political science, economics, and public policy as well aspractitioners interested in federalism, policy analysis, economic development,and pollution control.

JOHN S. ROBEY is Associate Professor of Political Science at the University ofTexas-Brownsville.

John B. Horrigan

Universal Service: Competition, Interconnection, and Monopoly in the Making ofthe American Telephone System, by Milton L. Mueller. Cambridge, MA: MIT Press,213 pp., $40.00 cloth.

The story of the American telephone system usually goes something like this:In the years immediately after Alexander Graham Bell uttered “Watson, comequickly,” the Bell System used its patent-conferred monopoly to establishhegemony over the U.S. telephone market. Certainly, there were concerns aboutthe Bell System using its size to muscle out competitors, and under the 1913Kingsbury Commitment, Bell was prohibited from acquiring small independentphone companies. But this prohibition was relaxed by Congress in 1921 andthe Bell System proceeded to consolidate its hold on the nation’s phone system.This made sense because it was a natural monopoly. The Communications Act

Book Reviews / 565

of 1934 established a regime of regulatory oversight that checked potentialmonopoly abuses while providing a regulatory framework to build a telephonesystem “with adequate facilities at reasonable charges.” In this dominant andhagiographic portrayal of the evolution of the American telephone system,government is credited with wise regulation as it promoted a system of internalsubsidies by which the Bell monopoly kept local phone rates low. Private industryis credited with public-spiritedness and technological virtuosity. The happyoutcome has been universal service; just about everybody who wants a phonegets one at affordable rates.

Much of this is wrong, according to Milton Mueller’s book Universal Service:Competition, Interconnection, and Monopoly in the Making of the AmericanTelephone System. Mueller argues that universal service never meant keepingrates low to promote phone subscribership, but really referred to networkinterconnection, that is, ensuring that the networks of competing phonecompanies were technically capable of transmitting messages between them.Moreover, neither the level of phone rates nor the magnitude of subsidiesdesigned to keep them low ever had much to do with promoting telephonesubscribership in the United States.

Mueller engages in careful historical analysis in portraying the dynamics ofcompetition in the U.S. telephone industry. “Dual service” was the mode ofcompetition in the Wild-West early years of the telephone industry (roughly1894 to 1912) whereby different companies built incompatible networks andtried to attract the most subscribers. This mode of competition becamecumbersome for subscribers after awhile; many users (mostly businesses) hadto incur the costs of subscribing to both networks in order to reach everyone inan area in which “dual service” reigned. Eventually, consumer preferences droveconvergence to a single network; as enough consumers found it more convenientto purchase access from one network, the remainder followed to that network.Before the convergence occurred, however, dual service caused competingproviders to build telephone lines in most areas of the country. This copper-wire network formed the backbone of a universally available telephone system.Universal service, a phrase first invoked by AT&T president Theodore Vail in1907, referred not to a policy by which phone rates were kept low so that morepeople could purchase services, but rather to interconnection—a policy by whichcallers from one network could complete calls to another.

By focusing on the dual service nature of competition, Mueller argues that theeventual consolidation of the industry to the Bell monopoly was driven by demand-side efficiencies, not the fact that the telephone network was a natural monopoly.Unit costs of providing service did not decrease with the number of subscribers inthe early telephone network, but increased quite substantially. Therefore, “demand-side economies of scope” prompted the convergence to a single network. Muellerbuilds the notion of a demand-side economy of scope on an observation that eachpossible connection on a telephone network constitutes a separate service. Whena new subscriber pays for a new connection, existing subscribers gain access to aseparate service without having to pay the entire cost of the new connection. Thiscreates efficiencies for consumers when one company provides access; consumershave a wider universe of potential callers than when there are separatenonconnecting networks. In economists’ parlance, the demand-side economy ofscope is essentially the same as a network externality.

If universal service was really about interconnection and the phone systemwas not a natural monopoly, how did we arrive at the ill-conceived notion that

566 / Book Reviews

universal service was about people having telephones with the Bell System aidingthe cause? The answer, according to Mueller, lies in the Bell System’s publicrelations machine. Bell sought to forestall antitrust attacks in the 1970s byarguing that its monopoly supported the low rates that gave the United Statesone of the highest rates of telephone subscribership in the world. In a discussionof only two pages, Mueller traces this public relations offensive to AT&Tchairman John deButts’s speech before the National Association of RegulatoryUtility Commissioners in 1973, which was then extended to legislative initiativesbefore Congress in the mid-1970s. The Bell System was, of course, ultimatelydismantled but not before the perception took hold that the monopoly hadbeen responsible for low rates for all. This, in turn, created an entire field of“universal service policy” founded on the lamentable premise (in Mueller’s view)that the telephone is an entitlement.

Once Mueller begins to rely on the fact that the universal service myth was aresult of AT&T rewriting telephone history in the 1970s, his argument takes adifferent turn: Universal service mythology was conceived as part of a monopoly’srent-seeking sins, and born into an environment hospitable to policy activistslooking to make the telephone an entitlement. For the argument to stick, theremust be no evidence that elements of the universal service myth (that is, lowrates being thought of as important to subscribership, widespread access to thetelephone being thought of as an important social or policy goal) existed priorto the 1970s. But there is such evidence. Writing in the Atlantic Monthly in1913, Theodore Vail said that “with a large population with large potentialities,the experience of all industrial and utility enterprises has been that it adds tothe permanency and undisturbed enjoyment of a business . . . if the prices areput at such a point as will create a maximum consumption at a small percentageof profits.” This was part of Vail’s “policy of reducing rates” so as to expand thebusiness [Stehman, 1925]. In 1914, McClure’s Magazine remarked on Bell’s effortsto “democratize this [the telephone] instrument—to make it part of the dailylife of every man, woman, and child” [cited in Pool, 1977, p. 131]. By 1931, apresidential commission stated that “to be without a telephone or a telephonelisting is to suffer a curious social isolation in a telephonic age,” even thoughabout one-third of all Americans had a telephone in the home [cited in Hadden,1994]. In 1950 Arizona Senator Ernest MacFarland pressured the FederalCommunications Commission (FCC) into making sure Bell’s long-distancedivision—rather than residential subscribers—would bear a greater share ofthe network’s fixed costs [Temin, 1987]. MacFarland worried that FCC policywould result in high rates for the “average housewife” while lowering the costburdens to “big users” such as “large national corporations.”

The point is that there was a good bit of discussion among politicians and inthe popular press about the importance of having a telephone, its socialdimensions, and the role prices might play in fostering subscribership. Indeed,the real price for local telephone service fell by half from 1940 to 1975 asthe share of households with phones grew from 37 percent to over 90 percent[Temin, 1987].1 It is no wonder that policymakers thought prices had somethingto do with telephone subscribership. Universal service “mythology” was notsomething that was created in the 1970s so that AT&T could mount a strategy

1 It is worth noting that Mueller states that average local rates stayed higher than 1952 levels until1965 (p. 160), while Temin’s (1997) chart (p. 60) shows real rates falling in that interval.

Book Reviews / 567

against antitrust attacks, even if AT&T tried to use the concept to its advantage.Rather, the idea that the telephone was a distinctly important service forAmericans to have—and that regulatory policy should pay special attention tothat goal—dates to the early part of the century. That this general policy goalbecame wrapped up in a broadly defined notion of universal service does notmake it a myth.

Although Mueller’s handling of a so-called universal service myth is notconvincing, this does not detract from his point about network competition beingimportant in the build-out of communications networks. The neglect of networkcompetition among historians of the telephone industry, Mueller argues, has ledto ill-considered choices in contemporary telecommunications policy, particularlywith respect to interconnection. In laying out a framework for ushering incompetition, the Telecommunications Act of 1996 requires that incumbenttelephone companies allow entrants to connect to their network atnondiscriminatory rates. This has resulted in a flurry of rulemaking by the FCCdefining the network elements to be “unbundled” to provide interconnection anddetermining what reasonable and nondiscriminatory rates are. This rulemakinghas been doomed to litigation. The Bell Operating Companies collectively havefelt that FCC rules create unjust advantages to entrants and a federal court hasagreed, throwing out the FCC’s interconnection rules. Rather than encouragecompetition, litigation over implementation of the Telecommunications Actseemingly is delaying it. Mueller questions the premise of the act’s entireinterconnection framework, noting that it ignores the possibility that mandatinginterconnection may actually inhibit competition. He recommends thatcompetitors be allowed to build proprietary networks, compete for customers,and let consumer choice settle interconnection and standards issues.

Mueller’s suggestion to unleash competition with little government-administered transition is provocative, though not fully compelling. If the worldcan withstand competition between VHS and Betamax or Microsoft and Apple,then surely it can do the same for communications networks. There may besomething to this, but given the growing social and commercial importance ofcommunications, noninterconnecting networks in the same or different regionswould be hard to tolerate. Moreover, with the almost daily reports of mergersin the telecommunications industry, concentration apparently is the choice ofthe marketplace, not competition—unless policymakers heed Mueller’s analysisand use antitrust tools to prevent mergers. Even if mergers are discouraged,the expense of building new networks might nonetheless constrain competition.Paradoxically, from Mueller’s perspective, requiring that new entrants into localtelephone markets build their own networks has been used in some states (Texas,for instance) as a phone-company-sponsored legislative tactic to preserveincumbents’ monopolies.

Time horizon thus becomes a key issue for telecommunications policymakers.Is it better to rely on interconnection rules and lengthy litigation to promotecompetition? Or is it better to allow unfettered network competition, whichwill require patience and deep pockets from potential entrants as they buildnetworks while, in the meantime, incumbents strengthen their hold on markets?Mueller would probably say that network competition is the best way to hastencompetition, but that outcome is not certain. With the current outbreak ofmergers, the network competition strategy runs the risk of allowing incumbentsto solidify market power. Moreover, if mergers result in combinations betweenproviders of network services and programming content, consumer choice could

568 / Book Reviews

be further limited. Litigation associated with interconnection may be protracted,but the alternative may be no better and create a new set of problems asincumbents use a variety of tactics—aided by long-established avenues forinfluencing policymakers—to protect market shares.

Mueller does not puncture the so-called universal service myth, but hisresurrection of the network competition issue deserves serious considerationby policymakers. Network competition should not supplant a rules-basedapproach to interconnection, but it should be another arrow in policymakers’quivers as they try to foster universal availability of broadband networks.Policymakers could, for instance, use government procurement as a strategictool; state or local governments could contract for the construction of broadbandnetworks for service delivery, lease necessary capacity, and allow the contractorto use remaining capacity to compete with incumbent communicationsproviders. Governments would thus encourage construction of a universallyavailable network, while additional competition would address affordabilityissues for low-income populations.

Communications networks will increasingly become the nervous system forsociety, with applications ranging from education to energy management tohome health care. Equity issues with respect to communications access willhave greater consequence, meaning that discussions about universal servicewill not end. As a policy concept, universal service will benefit from creativethinking by scholars. Universal Service certainly qualifies as that, even if it doesnot score the knockout punch to universal service for which Mueller had hoped.

JOHN B. HORRIGAN is a Consultant to the National Research Council andConsultant to the World Resources Institute.

REFERENCES

Hadden, Susan G. (1994), “Expanded Universal Service for the 21st Century,” New TelecomQuarterly 2(2), pp. 28–34.

Pool, Ithiel de Sola (ed.) (1977), The Social Impacts of the Telephone (Cambridge, MA:MIT Press).

Stehman, J. Warren (1925), The Financial History of the American Telephone and Tele-graph Company. (New York: Houghton Mifflin).

Temin, Peter (1987), The Fall of the Bell System (New York: Cambridge University Press).