Michael Rushton

Department of Economics

University of Regina

Regina, Saskatchewan S4S 0A2

Canada

Michael.Rushton@uregina.ca

Paper prepared for presentation at the

Cultural Policy Workshop, University of Chicago

January 2002

The field of cultural economics has devoted substantial attention to the positive and normative analysis of the willingness of citizens to fund art through the public sector. But there has been little economic analysis of the efficiency properties of the various possible methods of organizing public funding of the arts. Recent work on transaction cost models of organizations and politics can be fruitfully applied to the study of publicly funded arts agencies. In particular, the brief history of the National Endowment for the Arts (NEA) provides a vivid illustration of the problem of optimal degrees of political monitoring and statutory control of agencies. The special features of the art world, in particular the status and knowledge of art experts and the lack of a clear sense of mission as to the purpose of publicly funded art, and a rapid change over the past four decades in what artists actually do, have led to changes in the statutory control of the NEA and in the way it is monitored. Examples include the changes in the composition of the National Council on the Arts, the end of grants to individual artists (with the exception of writers), and proportionately large changes in the NEA’s budget allocation. This paper uses a transaction cost approach to link the changes in the art world and the changes in the oversight and procedures of the NEA. Framing the issue as a public administration problem, the paper highlights the challenges in using expert advisory panels when the goals of the panels are uncodifiable.

Keywords: National Endowment for the Arts; cultural policy; transaction cost politics

Analysts working in the field of cultural policy often use the techniques of economics: welfare economics to investigate the role of government where commercial markets and the voluntary sector alone fail to generate acceptable outcomes for the production and distribution of cultural goods; and macroeconomics to calculate the impacts of cultural production on the aggregate income of a region. Each of these methods of analysis has an extensive literature, although there remains a high variance in estimates of the relevant magnitudes. There has been much less systematic attention devoted to the institutional aspects of state support of culture. Yet to a large extent institutional design – the organizational structure and the formal and informal delegation of authority – has been at the core of the most heated debates in cultural policy: What is the role of peer review committees? What should be the statutory restrictions on the types of artistic activity funded by public arts councils? How should the mandates of federal and state arts councils be defined?

The goal of this paper is to apply the tools that have been developed by economists and political scientists to study institutional design to the organization of public funding of the arts. The focus will be on the National Endowment for the Arts (NEA), and the changes it has undergone in its brief history. However, the method is applicable to state and local arts agencies, and indeed there are lessons for all public granting agencies that must rely on professional expertise regarding which projects warrant funding. The analysis is necessarily qualitative rather than quantitative, but there is a theme that binds the various events involving the NEA that are discussed below: changes in the institutional arrangements governing the NEA have been attempts to reduce the transaction costs that arise in the relationships between legislators, the NEA, and the recipients of its funding.

Transaction costs are the costs of conducting political or economic exchanges, including the costs of negotiating arrangements, monitoring the performance of the contracting parties, and enforcement. The various exchanges that arise in the public funding of art are particularly prone to transaction costs. The paper will begin with a review of the more traditional methods that have been used to justify and assess public funding for the arts, and why these methods will sometimes fall short. This is followed by a brief review of transaction cost politics, and a discussion of why it might be a more fruitful way to study the institutions that arise in systems of public funding. This method is then applied to the evolution of the NEA.

Cultural economists working in the welfare economics tradition have found roles for state subsidies to the arts on both distributional and allocative grounds. First, it is important that members of society have access to those goods that build independence and the capacity to be an active participant in the life of their communities. Cultural education requires access to cultural exhibitions and performances. William Baumol, countering claims that public support of the arts benefits primarily the well-off, wrote that "individuals who are unable to use the language well and who are unfamiliar with society’s cultural heritage face marked handicaps in getting good jobs and advancing up the economic ladder. …[C]ultural illiteracy has much in common with linguistic illiteracy" [Baumol, 1997, p. 10]. However, there is more to culture than making us better and more productive individuals (Hughes [1992] claims this view of culture as uplifting is distinctively American). In 1942 John Maynard Keynes, at the time the world’s preeminent economist, accepted the task of becoming chair of the Committee for the Encouragement of Music and the Arts, the forerunner of the British Arts Council. According to his biographer, Keynes "had long been of the opinion that in the modern world art required a new kind of support to take the place of the affluent classes of earlier times. …In the time to come the mass of people should be able to enjoy the delights of fine art which in the past had been reserved for the favoured few" [Harrod, 1951, p. 518]. So we should also consider that there is nothing wrong with the notion that the poor should have access to culture simply because they would enjoy it.

On allocative grounds, there is a familiar list of reasons why the marketplace alone will not provide an optimal amount of cultural goods [Blaug, 2001; Frey, 2000; Heilbrun and Gray, 1993; O’Hagan, 1998; Throsby, 2001]. The essential allocative problem is that there may be externalities associated with the production and consumption of artistic works. This is for a number of possible reasons:

• people may feel national or local pride from artistic activities even if as individuals they do not directly take part;
• they may benefit from the innovation in non-artistic fields that results from artistic creation – this may include changes in social life more generally that are partly spurred by art with political dimensions;
• individuals may feel an increased sense of well-being simply because a vibrant arts community provides them with the option of one day enjoying its output; and
• there may be a desire to leave an artistic heritage for future generations.

In addition, there may be public good aspects to artistic activities, such as public sculpture. Finally, there are important educational aspects in artistic production. Individuals lead better lives when they have cultivated an appreciation and understanding of art, and this requires exposure to art. Smolensky [1986] claims this was the primary rationale for the municipal financing of art museums in the US since the 19^th century, and Scitovsky goes so far as to claim that "the only valid argument for government aid to the arts is that it is a means of educating the public’s tastes, and that the public would benefit from a more educated taste" [Scitovsky, 1972, p. 68].

As useful as the welfare economics approach is in clarifying why public funding of the arts is a good thing, it has some significant limitations. First, it is going to be very difficult to determine how much public support is optimal. This is for a number of reasons. The values of the externalities are not obvious, very difficult to measure, and will vary across art forms and genres. Contingent valuation methods have been applied to the preservation of heritage sites, but that is for cultural goods that already exist, not for the funding of new works. Even regarding arts education for children, over which there seems to be less controversy than for other public funding for the arts, it will be challenging to determine what is the best amount of spending when it is set beside all of the other valuable areas of education from which children benefit.

Second, the welfare economics approach is silent regarding the mechanism by which a desire by taxpayers for public support for some kind of art is translated into the actual works of art that ultimately arise from the funding process. While to some degree this is true of all publicly provided goods, taxpayer preferences regarding which highways should first be upgraded are more likely to be closely satisfied than preferences regarding the various art forms and genres that are publicly funded.

Like the welfare economics approach, public choice economists also assume self-interested, knowledgeable individuals. The point of departure is that individuals will advance their interests by political means, through strategic voting, attempting to influence other voters, and direct lobbying of politicians. There is no doubt that public choice economics can tell us something interesting about the public funding of art. Grampp [1989] claims that the arts are particularly ripe for rent seeking behavior; with a demand for arts that is income elastic, increased prosperity means that less rent seeking effort is needed to obtain the desired transfers. Furthermore, given the general weakness of the arguments for state support of the arts generated by the welfare economics tradition, it must be the case that the funding we observe is simply the result of rent seeking [Banfield, 1984].

But the public choice approach to the study of arts funding has some significant gaps in the analysis. First, it is relatively silent on the process by which lobbying activity results in arts funding and policy. Why should politicians listen to artists or arts administrators? Is the structure of arts councils completely explained by the preferences of artists?

Second, there is a significant amount of evidence (although Grampp disputes the point) that the general public supports state funding of the arts – this is not simply the case of a well-organized lobby taking funds from a mass of taxpayers unable to organize themselves to stop the transfer. Although there are many pitfalls involved in conducting surveys on public support of state funding of the arts, it is generally found that there is substantial support. Hansen [1997] found significant public support for state funding of the very specific good of the Royal Theatre in Copenhagen, even though only a small minority of those surveyed had actually attended the Theatre recently. Schulze and Rose [1998] find strong voter support for the state funding of orchestras in Germany. Throsby and Withers [1979] (in Australia) and Morrison and West [1986] (in Canada) find through their surveys that the level of support desired by the median voter is close to what is actually observed in those countries. Pettit and DiMaggio [1998] assess thirteen surveys of American attitudes to the arts and found that most Americans support a federal role in arts funding, although support is "a yard wide and an inch deep" (p. 50), and only a very few would want to expand it. Although there are large confidence intervals around the point estimates of the desired level of state support, the evidence seems to be quite strong that this is not simply a case of art producers obtaining transfers that voters don’t want them to have. This is not to say that there is not rent seeking, or that there are divergences between what the public wants to fund and what is actually funded. But there is certainly a desire for some sort of public funding.

The NEA was established through the National Foundation on the Arts and he Humanities Act, P.L. 89-209 signed by President Johnson on September 29, 1965. This followed the establishment in the preceding year of the National Council on the Arts (NCA), an advisory body on means of increasing cultural activity and enjoyment. Previously, the major federal initiative in arts funding was through the Works Progress Administration in the 1930’s [see Barber, 1999]. However, the appropriate role of the federal government in support of the arts remained a subject of debate from the 1930’s through to the establishment of the NEA [Larson, 1983]. A theme during this debate was the resistance to the establishment of any sort of cultural policy; this was seen as an inappropriate activity for the federal government, although many state governments have subsequently tried to articulate a policy direction [Kammen, 1996].

Why did the NEA come into being when it did?

Speeches by Members of Congress and by President Johnson at the time also make frequent reference to America’s place in the world and in history: how would its civilization be judged if there were no great artistic legacy?

The NEA is an independent government agency, with the Chair reporting directly to the President. When the NEA was founded the NCA was retained as an advisory body to the Chairman of the NEA, although there were concerns that a weak NCA would give the Chairman too much power, allowing him to become a "cultural czar". The NCA was made up of 26 citizens, serving terms of six years, with eight or nine appointed every two years. The Chairman of the NEA would make recommendations to the President for appointments to the NCA.

The goals of the NEA included the encouragement of excellence, support for state and local agencies, access and education in the arts, freedom of expression and cultural diversity, although a later report on the NEA noted that "Congress and the President provided only a few administrative signposts for the Chairman in seeking to implement these goals" [Independent Commission, 1990, p. 16]. In its early years the NEA was, in the words of one commentator, "bold, self-confident, innovative and opinionated. They intended to stamp their own imprint upon the arts rather than to support actions taken by others" [Straight, 1988, p. 97]. Grants were made for projects that raised some eyebrows, including $500 to Aram Saroyan for his poem, which read, in its entirety, "LIGHGHT", and funding for artists’ housing in New York City (which Straight regarded as foolish, but which Livingstone Biddle, who later served as Chair of the NEA from 1977 – 1981, defended on the grounds that the small print of the NEA’s mandate permitted "other relevant projects" [Biddle, 1988, p. 239]).

While the first Chairman of the NEA, Roger Stevens, had a great desire to be innovative, his successor Nancy Hanks, who served as Chair from 1969 to 1977, thought the central purpose was to use the NEA to generate more private support for the arts. This was consistent with the goals of Presidents Nixon and Ford, as well as explicit provisions in the Act [Straight, 1979, pp. 75-6]. Figure 1 shows the tremendous increase in funding levels for the NEA during Hanks’ tenure as Chair. "When Hanks took office … the NEA was a small, young agency teetering on the brink of its ‘initial survival threshold.’ By the time she resigned … the NEA had become a mature, institutionalized, successful federal agency" [Wyszomirski, 1987, p. 207]. Furthermore, it was during Hanks’ time as Chair that the amount of arts activity in America boomed, consistent with Hanks’ goals. Wyszomirski credits Hanks’ success to her ability to create a climate of trust between artists and legislators, to use the endowment to foster an influential political constituency, and to institutionalize the agency, creating defined routines for its procedures and policies.

In 1981 President Reagan created a Task Force on the Arts and Humanities, directed to consider ways of stimulating private support for cultural activities, especially during a period of high inflation. Although the NEA might have felt threatened by such a commission, its recommendations were very supportive of the NEA:

During the 1980’s the NEA’s budget was reduced as a part of government-wide deficit-reduction policies, and also in real terms through inflation.

In 1990 a second major report was commissioned, in response to long-standing complaints about the peer-review panel system (discussed in detail below). The Report stressed the public nature of the NEA:

Coincident with the investigation into NEA procedures were the widely debated exhibitions of art by Robert Mapplethorpe and Andres Serrano, and proposals that there should be prohibitions against federal funding for offensive art. The Independent Commission was opposed to any requirement that grant recipients swear under oath that their work would not be "obscene". However, the Act governing the NEA was amended requiring the Chairperson to ensure that judges of grant applications should take into consideration "general standards of decency and respect for the diverse beliefs and values of the American public."

The watershed year was 1996, with a cut to the NEA budget from $162 million in fiscal 1995 to $99.5 million in fiscal 1996. Further, Congress placed the restriction on the NEA that all individual artists’ grants, except in literature, were to end. As Figure 1 shows, the budget allocation never recovered from this cut. In response to the cuts the NEA restructured its grant-making system from 17 discipline-based programs to four categories: (1) creation and presentation; (2) heritage and preservation; (3) education and access; and (4) planning and stabilization.

In 1998 Congress, in addition to mandating that NEA grants to state arts agencies increase from 35% to 40% of program funds, changed the composition of the NCA from 26 to 20 persons, with six being Members of Congress who serve in a non-voting, ex-officio capacity.

The obvious question to address is what caused the cuts to the NEA’s budget, in real terms now less than one-third its level twenty years ago, and the increased restrictions on its activities.

Welfare economics provides a rationale for government support of the arts, in terms of various market failures, yet there is no evidence that the value placed by the public on art changed so drastically during the period in question. There is not evidence that the American public values any less education in the arts, or making the arts more accessible to all, or leaving a rich cultural heritage for future generations. Public choice similarly leaves the question unanswered; once established, the arts lobby should have had no problems securing its entitlement.

Other reasons have been put forward. Brenson [2001] placed blame on those who believe the only value of the arts is what is measured in the marketplace; the NEA cannot be defended on market grounds, and so the battle to preserve the NEA’s budget and grants to individual artists was lost. But no evidence is provided for a shift in the public’s values and attitudes to artists. Jane Alexander [2000], Chair of the NEA from 1993 to 1997, blames the strength of the Christian right in the 1990s in spreading the word against the NEA. But there were conservative politicians who acted as "frequent and formidable opponents" of the NEA as far back as its founding [Kammen, 1996]. Why did the cuts and restrictions to the NEA not occur until the 1990s? Furthermore, why was the Endowment not simply killed? If the opponents of the NEA were that powerful, surely they could have found a Congressional coalition to eliminate the agency. But instead, there was a sharp cut and a specific direction on the kinds of activities the NEA could fund. Finally, it could be put forward that some particularly controversial grants led Congress to demand changes. But the NEA has always had to deal with controversy in its grants, from LIGHGHT in the 1960’s to support for Erica Jong’s saucy novel Fear of Flying in the 1970s.

The answer lies in an analysis of the delegation of decision-making authority from Congress to the NEA. In 1965 there was a willingness to delegate policy and procedures regarding grants to experts in the artworld. In the early 1990s Congress placed restrictions on the granting procedures, and in 1995 individual grants were eliminated altogether. Recent research into the delegation decision provides away of explaining changes at the NEA that are not easily explained by other means.

Coase [1937, 1960] wrote the seminal articles on the importance of transaction costs in economic organization; in the earlier paper the boundaries of the firm are chosen according to the way transaction costs are present in within-firm organization and in market transactions with other agencies (so transaction costs are crucial to the make-or-buy decision), and in the later paper the externality problem is reconceived as the result of prohibitively high transaction costs preventing the contracts between affected parties that could conceivably internalize the spillover effects. Williamson [1975, 1996] adds structure to the analysis by explaining transaction costs as a result of four features of our economic and political world. First, we live in a world of inescapable uncertainty. Second, people have bounded rationality: there are limits to the time and resources we can spend in acquiring information, even if we use what information we have rationally. Further, there are limits on our language: we cannot communicate perfectly with each other even when we wish to. Third, individuals will exhibit opportunism: we often enter transactions with a lack of candor or honesty, especially when we know something that the individuals with whom we are interacting do not. Fourth, as we become engaged in economic or political relationships, investments are made that are irreversible and that have low value in alternative uses, thus leaving the investor vulnerable to tough negotiations with her trading partner, i.e. the "hold up" problem; we refer to this as small-numbers exchange relations.

At first, transaction cost analysis was applied principally to understanding the institutions we observe in commercial markets – patterns of vertical integration, hierarchies in the personnel structure of firms, the structure of contracts, and so on. The techniques are brilliantly applied to the commercial art world in the recent book by Richard Caves [2000], which will undoubtedly encourage much future research into contractual relations in the arts.

But the method of analysis is also ideally suited to understanding political institutions. It can help us understand the structure of government bureaucracies as well as the employment relations in private firms. It can be a very useful tool in understanding the legislation we observe, and the degree of political oversight of bureaucracies. In commercial or political exchanges we can ask how institutions have evolved to economize on transaction costs.

"A transaction cost theory of politics is built on the assumptions of costly information, of subjective models on the part of actors to explain their environment, and of imperfect enforcement of agreements" [North, 1990a, p. 355]. These are the same kind of features that lead to transaction costs in economic exchange. Epstein and O’Halloran [1999] use transaction costs to analyze the political equivalent of Coase’s firm’s make-or-buy decision: when should Congress delegate (i.e. "buy") policy making to a bureaucratic agency, instead of devoting its own resources to developing detailed policy (i.e. "make"). They note that the specification of delegated authority is a contract, albeit incomplete, enforceable by the courts. There is always a choice of governance arrangements, and boundedly rational legislators must decide on a degree of delegation that will minimize the political transaction costs, including the costs of developing the agency’s guidelines, monitoring its behavior, and enforcing the statutory provisions. Epstein and O’Halloran find that the optimal degree of delegation rises as the level of expertise held by the agency is more important, and where constant bargaining through logrolling in Congress is likely to be slow and cost-inflating. Optimal delegation will fall when there are significant concerns about policy drift by the agency.

Although the method of transaction cost politics has not been widely applied to cultural policy, there are some examples. Alan Peacock [2000] draws on insights from welfare economics and public choice, as well as his experience as past Chairman of the Scottish Arts Council, to raise questions about how well the current structure of arts councils solves the problem of improving the welfare of citizens. In particular he raises questions about the lack of accountability of arts councils, and of their capture by artists and arts administrators at the expense of arts consumers. Bruno Frey [2000] studies museum behavior from the perspective of the incentives and constraints facing museum directors, asks how state support for the arts will change the kind of art that is produced, and looks at the effects of having voters make direct choices about public art through referenda, among many other considerations. The focus is on institutional design, and how well alternative organizational structures serve the goals that justify public funding in the first place.

The development of the transaction cost approaches to economics and politics was driven by the realization that many of the institutions we observe that govern exchange and contract can be explained by the ways real human beings differ from the hypothetical rational agents of theoretical models. But the features of the world noted by Coase, North, and Williamson are particularly obvious in the art world.

First consider the presence of uncertainty. While it is true for many goods that there is uncertainty surrounding how a product in development will be received by the market, for art it is especially true. Caves [2000] is fond of citing William Goldman’s observation about films: "Nobody knows anything." This explains much of what we observe in the design of contracts in the commercial art world. But it applies equally well to publicly funded artistic creation. Although there is uncertainty in all public projects, policymakers will be able to project much more accurately the use of an upgraded highway, or extra places in public universities, than the response to a publicly funded work of art.

Second, there is opportunism. People have the capacity to be dishonest, to be less than forthcoming with their true feelings and intentions, and to "spin" their communications to others. This is not more or less the case in the art world, but an aspect of human nature found throughout economic and political institutions. However, given bounded rationality in the arts, the possibilities for using opportunism are increased. It is more difficult for individuals not well versed in the arts to evaluate the recommendations of arts boards or peer review panels, which increases the ability of those who are presented as experts to mislead.

Third, and for this paper most important, there is bounded rationality. The ability to critically evaluate proposals for, or finished creations of, art requires specialized knowledge that results from many years of study. It is not knowledge that is easily obtained by those who do not work on a daily basis with artists and their works. It is conceded that there are other areas of public funding that also involve great expertise: physicians’ services in publicly funded heath care, for example, or state funding of scientific research. However, in these examples while there is certainly specialized expertise required regarding the means employed by the specialists, at least most members of the public have enough understanding of the ends of the work: to cure the sick, or to increase our understanding of the physical properties of the world. But the average citizen, if not the art critic, would have a difficult time articulating what art is for. This is not to demean the intelligence or aesthetic sense of ordinary people. Rather, it is a comment on the nature of art: that it defies easy understanding, and especially defies easy communication, of its quality or its purpose. In addition to the difficulties of evaluating art itself, the welfare economics underlying the rationale for public finance is itself complex. Externalities are difficult to quantify at the best of times, but for the arts it seems to be particularly hard, to the point that some would deny their existence at all.

There are five key delegation relationships when art is financed through a public arts council. In each of these relationships the rules will be set to maximize the difference between the benefits of delegation and the transaction costs of the implicit contract.

• Voter « Legislator: The contracts between voters and the legislators they elect are very vague, and not easily enforced. Elections do not allow for voters the chance to pass judgment on how well their elected legislator kept each of her many promises - the vote is a very blunt instrument. Furthermore, if a legislator had promised to seek increased funding for the public arts council, and the increased funding never occurred, voters have great difficulty in knowing whether this is because the legislator reneged on her promise, or if unforeseen circumstances made it good public policy not to seek increased funding. Like all transactional relationships involved in public funding of the arts, there is a great asymmetry of relevant information between the principal (the voter) and the agent (the legislator). Voters have little incentive to investigate the matter further, because it is such a small-scale policy issue, and a salient issue only to a very small minority whose well-being is directly tied to levels of funding. Yet it is important for sound policy that legislators have some idea of what voters want. Note that this applies not only to levels of funding, but to what art is funded. Rushton [2000] notes that whatever is the justification for state support for the arts in the first place, whether it be to solve the externality problem, to ensure that poor young people have a chance to acquire cultural literacy, or to strengthen community or national bonds, the preferences of voters on the amount of funding and on the nature of the art they wish to fund must be included.

• Legislator « Executive Branch: The general problem is to design an oversight system that ensures effective public service, but that follows the direction set by the legislature, and does not engage in "bureaucratic drift" towards the goals desired by either the bureaucrats themselves or by the interest groups with whom they must deal. Particularly relevant is the very limited time and resources available to the member of the legislature, who must deal with a very large and diverse range of issues. McCubbins and Schwartz [1984] provide a description of the choice of institutional framework that seems quite relevant to arts councils. Police-Patrol Oversight is "comparatively centralized, active and direct: at its own initiative, [the legislature] examines a sample of executive-agency activities, with the aim of detecting and remedying any violations of legislative goals and, by its surveillance, discouraging such violations" (p. 166). This may involve holding hearings, commissioning studies, and the like. Fire-Alarm Oversight is less centralized, active or direct. Instead, the legislature "establishes a system of rules, procedures, and informal practices that enable individual citizens and organized interest groups to examine administrative decisions (sometimes in prospect), to charge administrative agencies with violating [legislative] goals, and to seek remedies" (p. 166). These procedures may take the form of ensuring that citizens have access to information and the ability to challenge bureaucratic discretion, and generally entails more detailed statutory constraint on government agencies. On the assumption that members of the legislature want to take credit for benefits gained by their constituents, and avoid blame for their losses, and that there are high opportunity costs for the legislator's time and effort, increasing the problem of limited knowledge, fire-alarm oversight comes to be preferred. Politicians will get little credit from voters for attending "police-patrol"-type meetings, and violations of legislative direction will still occur and often go undetected even with regular committee oversight, but with fire-alarm oversight only those issues that matter to constituents will be brought to the attention of the elected member, who can then take credit for fixing a problem in a visible way. For arts councils the interesting questions surround how the discretionary behaviour of the councils is limited by statute. To what degree are arts councils constrained by the possibility of citizens pulling a "fire-alarm", and to what degree do arts councils attempt to deflect the possibility of citizens raising concerns by providing on their own initiative information about their activities?

• Executive Branch « Arts Council: At this level of delegation we encounter the practicalities of defining, in the context of public arts funding agencies, "arm's length." The governing legislation of arts councils may specify the criteria by which applications for grants are to be assessed, whether there should be geographic balance in the awards, whether outreach and education are important, and whether there should be limits set on controversial content in the works of artists receiving funding. It is generally the case that the Executive Branch appoints the Chair of the arts agency and the Board.

• Arts Council « Peer Review Panel: This is the transaction where the problems posed by specialized information are largest. Peer review panels are implemented precisely because they have knowledge of their fields that is not generally held. But this lends itself to opportunism: what factors ensure the panel will make its recommendations honestly? One potential measure is to have permanent arts council staff on hand at jury meetings, to ensure fair debate. There is also reliance on the professionalism of the jurors, an ingrained sense that there is a duty to adjudicate fairly. Although arts council boards must ratify the recommendations of the jury panels, there exists a trade-off: panel jurors will only put a high degree of effort into their task if they have sense that they are unlikely to be overruled. Baker, Gibbons and Murphy [1999] show that the recommendations of the peer review panel might always be followed even though in the statutes the formal authority to make a grant rests with the arts council board. They note it might be the case that the board would desist from overruling a jury recommendation even if the board knew that a particular recommendation was a bad one, in order to maintain a high degree of effort in future peer review panels.

• Peer Review Panel « Artist: There are two transactional issues that have some importance at this stage. First, what kind of art activity will be proposed by those seeking grants? Second, how will opportunism manifest itself? Applicants are likely to be well-informed about what kind of projects have received funding in the past, and about the kind of person who is typically selected to be on the jury panel. It is hard to imagine that this knowledge would have no effect on the kind of artistic activities that will be proposed. Young [1997] makes the case that those in the art world, such as panel jurors, who confer status on works should use their position "responsibly" and so encourage excellence. Opportunism may arise as arts institutions seeking funding inflate their budget requests. Peacock writes that in his experience, "the problems of moral hazard and uncertainty are particularly prevalent in the direct subsidization of arts organizations and the incentives to solve these problems are blunted because of the lack of rewards to committees and officials if they take tough decisions" (1994, p. 179).

A crucial input into Congress’s decision about the delegation of policymaking authority is the gap in expertise between legislators and the staff of the executive agency. The evidence suggests that a major change in the artworlds of 1965 and 1995 was the decline in the credibility of experts.

The panel system was formalized during Nancy Hanks’ tenure as Chair. As the budget for the endowment grew very rapidly, the NCA could not handle the volume of applications and grants. And so discipline-based peer review panels were formed, each of which would work with permanent members of the NEA staff to prepare recommendations to the NCA, which would subsequently advise the Chair. Ideally, the panels would provide expert advice, prevent the grant-making process from becoming yet another pork-barrel fund, and provide cover for the NEA when a controversial grant was awarded [Mulcahy, 1985]. However, as Galligan [1993] notes, there were tensions in the panels right from the start. The mandate of the panels was not precise: they were to promote artistic excellence, but also to promote access to the arts. Indeed Hanks’ success at achieving such growth in the NEA budget has been attributed to her ability to persuade Presidents Nixon and Ford of the importance of bringing art to all the American people. So it was important to increase the diversity of membership on the panels. In addition, there were fears that the panels would be dominated by mainstream views of artistic excellence, rewarding those who were already well known. And so the pressure was on to include representatives from relatively marginalized genres or approaches.

The panel system is crucial in the decision of Congress to be willing to delegate policy making in the arts. A loss of faith in the value of the panels should lead to a change in the willingness to delegate.

Three events in the 1990’s illustrate the change in the perception of expertise in the arts. These episodes are often included in accounts of the "culture wars", but the focus in this paper is how art is evaluated in the public sphere.

Mapplethorpe: From December 1988 through January 1989 the Institute of Contemporary Art in Philadelphia, with the support of an NEA grant, held a retrospective of the photography of Robert Mapplethorpe. All galleries which were to host the exhibition were obliged to include the X Portfolio, which contained 13 photographs containing images of nudity, including children, and sadomasochism. The exhibition was scheduled to open at Washington’s Corcoran Gallery of Art on July 1, 1989, but was cancelled by the gallery director amidst fears that it would bring the wrath of some Members of Congress upon the NEA and the Corcoran. This turned out to be too little, too late, as on July 14 Representative Dick Armey and 107 other congressmen sent a complaint to the NEA that it had used federal money to fund indecent art [Steiner, 1995].

When the exhibition opened at the Contemporary Arts Center in Cincinnati, the director and the gallery were charged with obscenity. A jury trial found for the defendants. Hughes [1992] and Steiner [1995] provide accounts of the defendants’ use of expert testimony from art critics and curators. Since U.S. law defines obscenity as having no artistic value, if the photographs were demonstrably "art" they could not be obscene. The experts were clearly able to convince the jury. But they did so only by divorcing the formal aspects of the photographs, or their potential utility ("raising awareness," etc.), from what they were actually about, or trying to convey. Hughes called the defenses of the exhibition "an aestheticism that was so solipsistic as to be absurd," while Steiner, who it should be noted finds great value in Mapplethorpe’s work, found that the testimony of the critics and curators at the Cincinnati trial revealed a "vacuum in aesthetic theory in America."

Finley: In 1990 the reauthorizing statute for the NEA stipulated that in its grants the NEA would take into consideration "general standards of decency and respect for the diverse beliefs and values of the American public." Karen Finley and three others were denied NEA grants by Chair John Frohnmayer based on the decency clause, and in turn they sued the NEA. The case ultimately made its way to the Supreme Court, which held that the decency clause neither violated First Amendment rights nor violated constitutional vagueness principles (National Endowment for the Arts v. Karen Finley, 118 S.Ct. 2168 (1998)). In his memoirs, Frohnmeyer [1993] expressed fear that the decency clause would have a "heating effect," as artists would try to be as outrageous as possible in an effort to challenge the rules. There is no evidence that this has come to pass. Indeed, the opinion of Kathleen Sullivan [1991] is that the clause is constitutional only as long as it is vague enough to say "take into consideration" and nothing more.

The case weakened the stature of the NEA and its peer review panels by exposing the fact that any sort of serious content restriction on funding would be subject to court challenge. This puts Congress in a bind: they are then being asked to delegate spending decisions while being unable to restrict the NEA from funding potentially embarrassing (to the legislators) projects. Supporters of unfettered peer review panels would quickly find that fighting battles over the First Amendment is not an effective way to preserve the budget of theagency.

Jones: In 1994 choreographer Bill T. Jones presented "Still/Here" at the Brooklyn Academy. The piece combined dance with video testimony from people dying from AIDS. The piece itself did not stir as much debate as a "review" of the piece by the distinguished dance critic Arlene Croce [1994] in which she wrote that she had not seen the work, since doing so would be pointless: "victim art" is not something that can be subjected to aesthetic critique [see Teachout, 1995, for discussion of the commentary on Croce’s piece]. Croce’s thoughts were echoed by Joseph Epstein [1995], who had also served as a member of the NCA, and who drew attention to the problems that had developed in the peer review system at the NEA:

It needs to be stressed that Epstein, Croce, Hughes, and Steiner are sophisticated commentators on the arts; their views cannot be dismissed as the voices of philistines. But together they reveal that a significant change has happened in thinking about culture since the 1960’s. At the founding of the NEA there was a faith that there were such things as experts in the various fields of art, who could distinguish good art from bad art on the basis of their training and knowledge. Such experts certainly are still with us. But there has been a loss in confidence that these are the people making important funding decisions. It is interesting to note that the only grants to individual artists to survive the 1990’s were for literature. This cannot be because literature is never subversive. But it is worth speculating whether there remains a faith in literary criticism that is no longer present in the visual or performing arts. Munson [2000] notes the peer review panels came to be less and less composed of peers, i.e. artists. Instead, curators and art critics came to take a larger role. This led to claims of an increasing level of attention to the style represented by a particular applicant over the actual quality of the work.

The National Endowment for the Arts receives a far greater share of public and political attention than would be expected from its budget, which even when it was as high as $170 million Epstein described as, "as government spending went, just above the level of walking-around money." A complaint that might be raised about the preceding analysis is that is uses a theoretical sledgehammer to crack a simple peanut: "Conservative politicians were able to get publicity pleasing to their constituents through punishing an agency for spending money on questionable projects. They were able to appear fiscally responsible and on the side of family values."

But this is too easy; it does not explain why the NEA was able to be created and to grow even in the presence of various conservative coalitions in Congress, or why the NEA was reauthorized at all, albeit with a strictly limited mandate. The models typically used by economists and by other commentators are not as able to explain the full sequence of events. A transaction cost model is an effective way of modeling the delegation decision in arts funding, and explains well the fortunes of the NEA as a response to a change in the value awarded to specialized expertise in the field.