The Oversold Society
Democracy is a method of finding proximate solutions to insoluble problems.
Leave your theory, as Joseph his coat in the hands of the harlot, and flee.
—Ralph Waldo Emerson
Today we take it as a given that the state of the economy affects presidential elections. Ronald Reagan won the 1980 election after he stared into the camera during a debate with Jimmy Carter and asked the voters whether they were better off than four years ago. Bill Clinton won the 1992 election by keeping his campaign focused on “the economy, stupid.” And most recently, Barack Obama’s November 2008 victory over George Bush was certainly aided by the long recession that began in December of 2007.
Since 1960, the American National Election Studies have surveyed voters about the most important problem facing the nation. From 1960 to 2000, the most common answer was a concern about the economy or business or consumer issues, easily outpacing foreign policy or national defense issues (Johnston and Williamson 2005). Among academics, a small industry has formed to assess the effect of economic conditions on voting; by 2006, the number of scholarly articles and books stood at 400 (Lewis-Beck and Stegmaier 2007). The articles have found that economic conditions have strong and statistically significant impacts on how people vote. The best known of these pieces are by economist Ray Fair of Yale University; in a 1996 article, Fair was able to explain 96 percent of the variation in the votes for president from 1916 to 1992 by looking only at growth rates, inflation rates, whether an incumbent was running again, and whether the country was at war.
For Barack Obama, the same recession that swept him into power came to dominate every other issue in his administration. Ambitious plans on health care and energy played second fiddle to fiscal stimulus bills and decisions about bailing out investment firms. In an interview shortly after his inauguration Obama admitted, “Look, the only measure of my success as president when people look back five years from now or nine years from now is going to be, did I get this economy fixed? … I’m not going to be judged on whether or not I got a pet project here or there, I’m going to be judged on, have we pulled ourselves out of recession?”1
But are the voters’ expectations rational in this regard? As a practical matter, what can the president really do about the economy? A review of the evidence shows that there is an enormous gap between the public’s perception of the president as an economic catalyst and the reality of the president’s power in this area. Indeed, the evidence suggests that voters pondering their decisions ought to ignore President Reagan’s question about their economic circumstances. Among prominent macroeconomists, many believe that fiscal policy cannot have real effects on the economy because of something called the Ricardian equivalence theorem. The consensus among the remaining macroeconomists is that, although fiscal policy can theoretically boost the economy, the time needed to recognize business cycles and the long lags involved in implementing policy make it extremely difficult for presidential administrations to have positive impacts on the economy (see chapter 3). An analysis by William Keech (1995, 162) of postwar economic stimulus packages before 1995 showed that all of them took effect too late to help aid economic recovery: “There is no example of a program that was passed before the final month of the recession it was designed to correct.” In a review of fiscal policy over the last 40 years, Bradford DeLong (1996, 47) concluded, “Looking back at the budget since World War II, it is difficult to argue that on balance ‘discretionary’ fiscal policy has played any stabilizing role” (italics from the original).
As discussed in chapter 3, even the New Deal programs of the 1930s had little effect on recovery from an economic downturn. The Great Depression was so long that it gave policymakers plenty of time to recognize the problem and craft innovative responses. And the sweeping mandate given to Franklin Roosevelt in the 1932 and 1936 elections gave him free reign to run unprecedented peacetime deficits. Yet among economic historians, there is widespread consensus that FDR’s fiscal policies were not a major factor in bringing us out of the Depression.
It seems likely, then, that we elect presidents to perform a task over which they have little control, and we pay insufficient attention to the things they can do, such as conducting foreign policy and ensuring that the federal government efficiently provides public services. Voting your wallet for a presidential election is a bit like choosing a news broadcast depending upon whether the news was good or bad the last time you tuned in to that station. The news would have been just as good (or just as bad) regardless of who was behind the anchor desk.
However, this contrast between the claims of public debate and the reality of government power is not unique to the presidency or to national economic issues. While debates on education tend to center on testing, class size, and teacher qualifications, the disappointing conclusion reached by education studies is that the effects of public schooling are dwarfed by the influences of family and socioeconomic background. And while gubernatorial campaigns center on attracting new jobs and stimulating the economy, the literature suggests that state tax and spending policies have little effect. Similarly, while welfare policies focus on altering the educational, marriage, divorce, and childbearing decisions of the poor, there is surprisingly little evidence that we can do so.
The purpose of this book is to explore this paradox between the exaggerated claims made for public policy interventions and the reality of their limited effectiveness. Because policy impacts are often overstated, our national debate is misguided, concentrating on illusory “fixes” rather than on the true choices before us. I provide detailed examples of such overstatements in five areas: short-term fiscal policy, long-term growth policy, education, state and local economic development, and welfare. Each of my arguments is grounded in the best available evidence—literature reviews of the quantitative empirical work in each field. The book analyzes the reasons for this paradox, and makes recommendations for changes in the way we teach economics, statistics, and law in order to narrow the gap between rhetoric and reality. The following chapters emphasize three themes, which are briefly summarized in the sections below.
Government policy is consistently oversold, to citizens, to politicians, and even to academics. It is oversold by both conservatives and liberals, in different but curiously similar ways. Over and over, we elect officials in the naïve belief that they can pull some magic lever to fix our social and economic problems. When that doesn’t work, we “throw the bums out” and elect someone else to pull a different lever or to pull the same lever in the opposite direction. But what many Americans don’t understand, but empirical studies bear out, is that in many circumstances the governmental levers are simply disconnected from the problems they are supposed to address.
The Three Views of Government Policy
The first view of policy, the liberal view, is that government policy can change the world, that it can fundamentally alter the social, economic, and demographic fabric of our society. After a decade of teaching courses on the application of statistics to public policy and having reviewed countless empirical articles on the effects of government programs, my conclusion is that it typically does not. The record shows remarkably little evidence that government programs can alter these fundamental factors. The growth of the service economy, the decline of traditional family structures, the acceleration of globalization, the appearance of new technologies that both threaten established industries and change how we interact—all of these inexorable changes are, I would argue, largely beyond our control.
Since Ronald Reagan, conservatives have done an excellent job of pointing out holes in this first view. But what is less well recognized is that conservatives suffer from a parallel fallacy. The second view of the world, the conservative view, is that government policy is not the remedy for social problems but is actually the cause of those problems. As Reagan put it in his first inaugural address in 1981, “Government is not the solution, it’s the problem.” Permissive welfare policies, so the theory goes, encourage families to break up, encourage parents to have more children, and encourage workers to quit their jobs and take government assistance (see chapter 9). High taxes, conservatives argue, drive industries out of the city, the state, and the country, leading to unemployment and slow growth (see chapters 4 and 7). The irony here is that conservatives see government as powerful, just as liberals do, but with opposite effects. Many of these fundamental social problems could be solved, according to conservatives, if we just got the government out of the way.
Again, the empirical literature on the effects of government programs and taxes indicates that, in most cases, conservatives greatly oversell the power of government actions. A look at the data suggests that our economy and our society are being shaped by powerful historical forces that are largely independent of government policy. Welfare programs probably have little to do with the increase in out-of-wedlock births, and tax policies have little to do with the distribution of economic growth or the ebb and flow of economic activity over time.
Throughout this book, I argue for a third view of government policy, based on the empirical literature—one that acknowledges the limitations of government policy for good and for ill. This third view concentrates on what we can control about policy problems, rather than on what we cannot control. If government cannot solve the fundamental problem of poverty in a capitalist economy, it can have a profound effect in relieving suffering and ameliorating hardships among the poor. If public policy cannot alter the fundamental forces of globalization and post-industrialization, it can nevertheless help people adapt to these changes and reduce the pain and suffering they cause. Rather than chasing after imaginary solutions to social problems, the third view focuses on helping people cope with them.
The Elements of Government Programs
Domestic government programs contain various levels of five fundamental activities: redistributional, investment, service, regulatory, and behavioral. A redistributional activity is one that simply moves money from one group of citizens to another. Social Security remains our most popular federal program precisely because it doesn’t try to do anything beyond giving money to old and disabled citizens. It doesn’t try to convince senior citizens to work more, or stay married, or go back to school. Since it is not measured by such yardsticks, Social Security is considered successful by the only yardstick left: whether it has raised the incomes of senior citizens and disabled people, bringing them out of poverty.
An investment activity is one that directs money into human or physical capital in the expectation of greater productivity in the future. The federal highway system or federally sponsored basic medical research, for example, are highly investment-oriented programs.
A service activity is one that provides citizens with goods or services. For example, a lighthouse provides an invaluable service to mariners at sea, and an airport control tower provides essential services for airplane pilots and passengers.
A regulatory activity is one that establishes rules for social and economic activity. The federal courts enforce property rights, for example. Antitrust regulations prohibit monopolies, antipollution legislation restricts individuals and companies from damaging the environment, and workplace safety requirements ensure that firms do not abuse their workforces.
My focus in this book is on the fifth element of government programs—the behavioral component. Behavioral activities try to change targeted groups’ behavior, such as reducing teenage pregnancy, or convincing a company to change the location of a plant, or getting children to read more. Although the behavioral element is only one of five elements, it receives a disproportionate share of our attention in public debates. Typically, political discussion of the national economy is all about getting consumers to spend more or getting firms to invest more. Talk of local economic development centers on getting firms to move to a particular state or locality. Debates on welfare always seem to return to the question of getting welfare recipients to work more, marry more, and have fewer babies.
Notice that many controversial programs are actually a combination of these five components. What we call welfare is a combination of a redistributive element (cash and services) and a behavioral element (provisions to get the poor to stay married, have fewer children, and have children later). And what we call education is a combination of an investment element (increasing human capital), a service element (providing learning experiences for children), and a behavioral element (getting children to learn more and pursue higher education). The lines between redistributive, investment, service, regulatory, and behavioral elements of programs are often blurry. Nevertheless, my thesis is that, despite the claims of conservatives and liberals, the behavioral impact of domestic public policy in the
United States is usually small in relation to that of other social forces at work.2 This means that we really ought to be evaluating government programs by focusing on the other four elements of policy, rather than concentrating on the behavioral element so often discussed by politicians and the news media. Not all redistributional, investment, service, or regulatory activities are successful, of course—money can be wasted in administration, or wrongly invested, or used to provide services people don’t want. My point is simply that we ought to judge our leaders based on their demonstrated performance in those four areas, rather than on an illusory ability to change behavior.
An Illustration of the Pattern
The theoretical overestimation of public policies is vividly illustrated in the issue of public financing for sports stadiums. Given its finite resources, a community faces a choice between sports services and other things it might do with its money. Suppose that the community is approached by a sports team or a group of investors about building a sports stadium. Under the first view of government policy3, the public discussion about the stadium will be all about theoretical economic benefits: more construction jobs, increased tourism, new jobs at the ballpark, and growth in the restaurant and hotel businesses. The argument will be that the new park will be an expenditure that pays for itself in faster economic growth and new tax revenues.
The second view of government policy is skeptical of the government’s ability to recognize profitable growth opportunities and its ability to seize on them. The construction jobs are temporary, and much of the spending at the ballpark will simply divert expenditures from other local entertainment. The jobs at the ballpark are often part-time, low-wage service positions, and much of the profit goes to owners and highly paid athletes who live elsewhere. Under the second view, government subsidization of sports stadiums actually makes the local economic situation worse by wasting precious local capital. If the local government raises taxes to support the new ballpark, the new taxes will weaken the incentives to invest in the city, reducing jobs and economic growth. As in the first view, the policy debate centers strictly on the economic factors of job growth, income increases, and multiplier effects.
But a look at the empirical literature on sports stadiums reveals weaknesses in both the first and second views of public policy. These studies show a strong consensus that sports stadiums are not major engines of economic growth. By themselves, stadiums cannot turn around the fortunes of inner cities, and they cannot pay for themselves in the long run with increased tax revenues (Noll and Zimbalist 1997,15). These findings call into question the first view of ballpark economics. What stadiums do provide, however, is more entertainment and sports services, because they hold the key to the location of sports franchises. New clubs will typically refuse to locate in an area without an attractive new stadium, and older clubs threaten to move unless provided with up-to-date facilities. Sports teams do unite communities: in increased community pride, in a sense of shared experience. Using survey data, scholars have been able to demonstrate the intangible benefits of sports franchises, and these may be large enough to justify public investment (Rappaport and Wilkerson 2001). These findings suggest that the first and second views, with their tight focus on economic factors, may ignore crucial community benefits.
Is building the stadium with public assistance a good idea? Once we abandon liberals’ and conservatives’ theoretical arguments and simply rely on the highest-quality empirical information available, a different perspective emerges. The lesson of the third view of public policy is this: forget the inflated estimates of job growth and dubious spending multipliers. And forget the exaggerated talk about tax disincentives—as chapter 7 will demonstrate, state and local tax rates have small impacts on economic activity. The real question is whether the community is willing to pay the freight for the personal and psychological benefits of having a team in town. The authors of the largest and most comprehensive study of the economics of sports stadiums put it this way: “Properly reckoned, the value of a sports team to a city should not be measured in dollars of new income but should be appreciated as a potential source of entertainment and civic pride that comes with a substantial net cost” (Noll and Zimbalist 1997, 498).
In terms of the five-part division of government activities discussed above, the first and second views of policy have been looking at the wrong element entirely. These have been concentrating on the behavioral element of sports stadiums—how they can change the behavior of fans and firms and improve local economic conditions. But that’s an illusion—the behavioral effects are much smaller than we thought. Instead, we should look at the service element of the question. Sports franchises provide a service to their communities—a place for recreation and a common bond between citizens. The right question to ask here is not “how will this stadium change our town?” but rather “are the benefits of this franchise worth the costs to our citizens?”
Explanations from Political Science and Psychology
Political scientists would probably be unsurprised by our first theme. Of course politicians overstate the effects of their policies; they have every incentive to do so if distorting information will enhance their careers. Voters are “rationally ignorant”—their share of the costs or benefits of any particular program is too small to warrant careful analysis of politicians’ claims. In addition, voters are not always rational deliberators of policy options; as Kathleen Hall Jamieson puts it (1992, 41), “our quirks as consumers of political information” make us particularly susceptible to negative information, cause us to pay disproportionate attention to vivid case studies, diminish our reasoning capacity in fearful situations, and make us believe that repeated messages are more credible. According to political scientists, the problem is not so much a lack of information, but a distorted political system that allows policymakers (or, as political scientists call them, policy elites) to dupe irrational voters into making poor choices.
The trouble with the political science perspective is that it assumes politicians know the truth and deliberately mislead the rest of us. But the evidence suggests otherwise. As detailed in chapter 8, politicians’ educational backgrounds are strong in economic theory and legal procedure, but terribly weak in statistics and empirical policy studies. For example, an examination of the curricula of six of the top law schools in the country revealed that none of them required a course in statistics, and only 3 percent of the students took advantage of optional statistics courses that were offered at three of the schools.4 Chapter 5 details the poor quality of the information used in making policy decisions by contrasting it with the data and analysis used in medical decisions. For example, the evidence underlying the largest tax cut in U.S. history consisted of a weak analogy to a tax cut during the Kennedy administration. Taken together, these chapters demonstrate that elites are not omniscient Machiavellians, but suffer from the same misinformation and confusion as the rest of us. Certainly politics and political gamesmanship play a role in poor decisions, but learning to understand and interpret the evidence is a vital prerequisite to making better choices.
Psychologists would probably also be unsurprised that policy levers are disconnected. In 1975, Ellen J. Langer showed that people systematically overestimate their control over circumstances, a result she called “the illusion of control.” In one experiment, Langer’s subjects strongly preferred a lottery ticket they chose over one another person chose, even if both tickets had an equal probability of winning. Some research even indicates that the illusion of control is beneficial in some circumstances because it leads to greater optimism and perseverance. Alloy and Clements (1992) showed that college students with stronger illusions of control were less discouraged by poor test results and negative life experiences than more realistic students. Of course we overestimate our abilities—that is one means of coping with an uncertain and unyielding world.
My point in this book is not to dispute political scientists and psychologists, but to find ways around the problems they raise. Ultimately, our political beliefs are dysfunctional, and given the magnitude of the issues facing us, we simply cannot afford to let them stand unchallenged. For example, real human lives are involved in the welfare debate—mostly children’s. Getting welfare policy wrong means subjecting those children to real suffering, including hunger, isolation, illness, and poor life prospects. Can we really afford to remain ignorant about the problem or react to information irrationally?
Unfortunately, the political science and psychological explanations for disconnected levers aren’t helpful in solving the problem. Knowing that politicians sometimes distort the facts doesn’t help much, because we need to know exactly how the facts are distorted in order to clarify our understanding. Knowing that we are subject to illusions of control isn’t helpful, unless we know which areas are subject to illusion and how far our illusions stray from reality. What we really need are some empirical observations about when we are typically fooled and some benchmarks against which to measure both politicians’ claims and our beliefs about the world.
The next two themes, therefore, deal with identifying the circumstances in which we overstate policy outcomes and with educating us—and our policymakers—to think clearly about policy issues.
Too Much Theory, Not Enough Values
The ideological overestimation of government policy is aided by a parallel theoretical overestimation. Due to the power, precision, and sophistication of economic tools, particularly theoretical ones, policymakers tend to clothe their arguments in economic theory. Since labor supply curves slope up, the argument goes, a tax on wages reduces the net wage and reduces the supply of labor, increasing unemployment. A worker-training program, it seems, increases the marginal productivity of labor, increasing wages and employment. And universal health insurance ought to decrease the welfare rolls, since the current system penalizes mothers who try to give up welfare (which provides health insurance) for low-wage work (which often does not).
All of this makes perfect sense in the two-dimensional, static world of undergraduate economics courses. The trouble is that, although everyone knows that demand curves slope down and supply curves slope up, only a very few have the education and motivation to estimate the slopes of the curves, and these parameters have a profound effect on the practicality of policy prescriptions. Since we approach policy problems from a theoretical, rather than empirical, standpoint, we fail to appreciate that in the complex, multidimensional world in which we live, the individual effects of government policies are often small.
In the case of welfare policy, for instance, economic theory suggests that higher welfare benefits encourage welfare mothers to have more children, because increasing the marginal benefits of any activity ought to get us to do more of that activity. But the childbearing decision is extremely complex, involving availability of birth control, education, class, family structure, and race. When most analysts have looked at the numbers on birth rates and welfare mothers, they have concluded that welfare benefits have almost no effect on birth rates; government policy is a piccolo in a veritable symphony of causes (chapter 9).
Because the language of economics is powerful, we use it to oversell the positive or negative effects of government policies, arguing for increases or reductions in government activity. Once we abandon the inflated expectations of government policy embodied in the first and second views, preferences matter a great deal more than is generally acknowledged. Economists give these preferences an abstract name like “the social welfare function,” but political scientists and philosophers use terms like “values” and “ethical judgments.”
Politicians continually try to skirt the knotty questions of values by resorting to theoretical, economic arguments, but the empirical evidence draws us right back into the normative sphere. The question is not what welfare system will fix the problem of poverty—none currently available will do that (see chapter 9)—but what welfare system is compatible with our obligations toward the poor (whatever those may be). The question is not what economic development policy will revitalize our state’s economy—no such policies currently exist (see chapter 7)—but what economic development policy will help us build a community that matches our preferences. The question is not what fiscal policy will eliminate the swings of the business cycle—at present, we don’t have the tools to do these things (see chapter 3)—but what tax policy represents a fair sharing of the public-service burden.
The book’s final theme concerns the widespread inability of citizens and policymakers to understand basic statistical concepts. In chapter 5, I attempt to illustrate the empirical poverty of our policy debates by contrasting them with debates in health care. In 1981, two landmark decisions were made within weeks of each other. In July, the medical community approved a controversial vaccine for hepatitis B; in August, the Reagan administration engineered the largest tax cut in the history of the country. The claims for the hepatitis vaccine were backed by one of the most successful experimental trials in
U.S. history; the claims for the tax cut had almost no empirical support.If politicians “sell” us distorted information, why are we willing to “buy” it? Why are our standards for evidence in policy disputes so low? Part of the answer may lie in the public’s lack of statistical knowledge. Because we don’t know how to weigh empirical evidence, we fall victim to poorly supported but intuitively plausible ideas. This lack of sophistication is fed by unhelpful teaching approaches in three disciplines.
In economics, the primary emphasis is on learning the theoretical mechanics of demand, supply, and markets, not on the empirical magnitudes of economic phenomena. In fact, nonmajors who take an economics class rarely receive any instruction at all in statistics or econometrics. This leaves them with a head full of theoretical pathways by which government policy could conceivably affect market outcomes, but with no means to discern which actions are most likely to work.
In statistics, our current approach emphasizes theoretical concepts (such as probability theory) rather than the application of those concepts to real problems. And when statistical theories are applied to real problems, they are most often focused on experimental statistics in health care, not on the nonexperimental statistics common to public policy.
In law, the background favored by the vast majority of politicians, almost no time is devoted to the study of statistics. While students are extremely well trained in the legal nuances of particular cases and in learning to distinguish the fine details between cases, they have almost no training in making valid inferences from a large body of such cases. District attorneys, for example, are well prepared to prosecute and convict as many defendants as possible; but if elected to the town council or the mayor’s office, they are not prepared to decide how to allocate scarce resources to best control crime.
The examples in this book do not show that government policy never affects behavior—only that these effects are typically overstated. For example, in the cases of education and state economic development discussed in chapters 6 and 7, policy has a positive effect but one that is much smaller than commonly assumed. To make sensible policy, decisionmakers and their advisers must have a realistic appreciation of the relevant parameters and the inevitable policy trade-offs. Such an appreciation requires the ability to draw valid conclusions from masses of data, which is precisely what inferential statistics teaches us.
To see the policy world clearly, we do not have to turn students into printout-carrying statistical geeks. They simply must learn the skills to be knowledgeable consumers of statistical information. Paying attention to the appropriate evidence, I will argue, typically leads us toward the third view of public policy.
Organization of the Book
These three themes—disconnected levers, overreliance on economic theory, and statistical innumeracy—are woven together in the remaining chapters of this book. Chapter 2 explains the empirical methodology we can use to sort out effective from ineffective policies. For those tempted to skip this chapter, let me add a word of encouragement here. I promise: no Greek letters, no equations, and a minimum of multisyllabic technical terms. My goal is to explain the differences in quality of statistical evidence in straightforward, nonmathematical terms. I do so early in the book, so that you can appreciate the strength (and often, weakness) of the evidence debated in the rest of the book. Think of our political discourse as a flea market for policy ideas—it helps to know some indicators of quality beforehand. Chapter 2 provides some things to look for before you “buy” a particular policy prescription.
Chapters 3, 4, 6, 7, and 9 explore five policy areas that illustrate the oversold policy pattern. These chapters address the first two themes of the book. The cases were not randomly selected but do represent voters’ most frequently cited domestic concerns. As discussed above, the American National Election Studies5 show that the most frequently cited domestic problem was a concern about economic conditions (including business and consumer issues). The next most commonly mentioned problem was social welfare (including public welfare and education).6 Accordingly, the book contains multiple cases from each category.
Chapter 3 examines federal fiscal policy, particularly the role of the president and Congress in the short-term behavior of the economy. Chapter 4 explores the connections between government policy and economic growth. And chapter 7 takes up the issue of state economic development, specifically the tax incentives used in bidding wars between states for jobs. Chapter 6 addresses education policy, focusing on the effect of increased educational resources on student performance. Finally, chapter 9 discusses welfare policy, which has been transformed by the 1996 welfare reform act.
Chapters 5, 8, and 10 illustrate the final theme of statistical innumeracy. As mentioned above, chapter 5 illustrates the current low standards for empirical support currently used by policymakers, by contrasting the evidence medical professionals use to make decisions with the evidence policymakers use. Chapter 8 explores the role of statistics in current economics and legal education. In chapter 10, I make recommendations for changing the way we teach economics, law, and statistics, so that we will be less vulnerable to overzealous claims about the effects of public policies. Because statistical innumeracy is so entwined with the other two themes of the book, chapters 5, 8, and 10 are interspersed with the five chapters examining substantive policy areas.
Too often, our national debates are misguided. We debate phantoms and mirages while leaving the real choices undiscussed. We can only make effective decisions if we are clear about the things that we can, and cannot, change. I hope this book contributes to a better understanding of the true alternatives in front of us.
1. Barack Obama, interview by Anderson Cooper, Anderson Cooper 360, CNN, February 3, 2009. http://www.cnn.com/2009/POLITICS/02/03/obama.qanda/index.html.
2. The evidence used in this book relates to domestic policy initiatives in the United States and cannot be used to evaluate the much larger policy differences between nations. So, for example, chapter 9 examines the effects of the welfare reform act of 1996, but these data should not be used to make arguments about moving to a social welfare state like those in Scandinavia.
3. Throughout this book, I have been careful to not automatically associate the words “liberal” and “conservative” with the first and second views, because the ideological lines are often confusing. For example, in the case of local economic development, many conservatives, especially local industry leaders, oppose larger “public spending” on things like welfare but nevertheless support public sector initiatives to encourage private sector growth. Many liberals, especially those in social service industries, support a larger public budget but oppose publicly assisted private sector initiatives such as sports stadiums, which they feel primarily benefit the middle and upper classes.
4. Author’s calculations, based on examination of the online catalogs of the six schools (see chapter 8).
5. The American National Election Studies are a joint project of Stanford University and the University of Michigan.
6. Author’s calculations, based on Sapiro and Rosenstone (2004).
Alloy, L. B., and C. M. Clements. 1992. “Illusion of Control: Invulnerability to Negative Affect and Depressive Symptoms after Laboratory and Natural Stressors.” Journal of Abnormal Psychology 101(2): 234–45.
Fair, Ray C. 1996. “The Effect of Economic Events on Votes for President: 1992 Update.” Political Behavior 18(2): 119–39.
Jamieson, Kathleen Hall. 1992. Dirty Politics: Deception, Distraction, and Democracy. New York: Oxford University Press.
Johnston, Louis D., and Samuel H. Williamson. 2005. “The Annual Real and Nominal GDP for the United States, 1790–Present.” Economic History Services, October. http://www.eh.net/hmit/gdp/.
Keech, William R. 1995. Economic Politics: The Costs of Democracy. New York: Cambridge University Press.
Langer, Ellen J. 1975. “The Illusion of Control.” Journal of Personality and Social Psychology 32(2): 311–28.
Lewis-Beck, Michael S., and Mary Stegmaier. 2007. “Economic Models of Voting.” In The Oxford Handbook of Political Behavior, edited by Russell Dalton and Hans-Dieter Klingemann (518–37). Oxford, UK: Oxford University Press.
Noll, Roger G., and Andrew Zimbalist. 1997. Sports, Jobs, and Taxes: The Economic Impact of Sports Teams and Stadiums. Washington, DC: Brookings Institution Press.
Rappaport, Jordan, and Chad Wilkerson. 2001. “What Are the Benefits of Hosting a Major League Sports Franchise?” Federal Reserve Bank of Kansas City Economic Review 86: 55–86, 1st quarter.
Sapiro, Virginia, and Stephen J. Rosenstone. 2004. American National Election Studies Cumulative Data File, 1948–2002. Ann Arbor, MI: Inter-University Consortium for Political and Social Research.