Public choice economics: where is there consensus?

Whaples, Robert ; Heckelman, Jac C.

Introduction

Economists love to invade other social scientists' turf. The archetypical example may be the subfield called public choice economics, which has been defined as "the economic study of nonmarket decision making, or simply the application of economics to political science" (Mueller, 2003, 1). Since its founding in the late 1950s, public choice economics has become a burgeoning, increasingly important part of mainstream economics, while challenging the approaches and conclusions of many traditional political scientists.

In this paper we ask about the outcome of this invasion. Are public choice scholars' conclusions accepted by rank-and-file economists and political scientists? If not, why not?

To answer these questions we use survey results to compare the conclusions of self-identified public choice scholars with those in the adjacent disciplines of economics and political science. Doing so allows us to observe the final product of the intellectual production process--the judgments reached by the large, sometimes silent, body of scholars after the evidence in support of competing views has been weighed. We begin with scholars' basic assumptions about how individuals act and how democracy should work. Then we compare their conclusions about public choice outcomes on a range of propositions. We conclude that, although there is consensus on many of the issues, there is substantial remaining disagreement on many questions that appears to be tied to the competing presuppositions of scholars in economics and political science.

The survey we use was mailed to 193 randomly selected members of the American Economic Association (AEA) and 202 randomly selected members of the American Political Science Association (APSA), as well as to almost everyone on the Public Choice Society (PCS) mailing list, including 191 economists and 116 political scientists. (1) Our response rate was highest from members of the AEA. Among them 36.3% returned completed surveys (70 surveys were returned), which is very similar to the Alston, Kearl and Vaughan (1992) survey response rate of 34.4% among economists. The response rate for APSA members was only slightly lower at 33.2% (67 surveys), while the response rate for public choice scholars was modestly lower at 29.8% (57 surveys) among economists and 29.3% (34 surveys) among political scientists. The differences in these response rates are not statistically significantly different at the 90 percent confidence level?

In this paper, we refer to the AEA sample as "economists," the APSA sample as "political scientists," the PCS-economist sample as "public choice economists," and the PCS-political scientist sample as "public choice political scientists." When we don't distinguish between the latter two groups, we refer to them collectively as "public choice scholars."

Below we examine thirty-five propositions in seven areas: (1) assumptions about political actors; (2) normative beliefs about government and voting; (3) elections and economic performance; (4) parties, platforms, voting and preferences; (5) government's purposes and growth; (6) individual behavior--voting-with-feet and free riding; and (7) government and the market. Our list of propositions was compiled following years of teaching and researching in the public choice field. Much of it draws on the themes covered in Mueller (1989), the field's leading textbook. (3) Finally, the list and wordings were honed after consultation with colleagues in economics, political science and public choice. (4)

For each of the thirty-five statements, respondents were asked to note how strongly they agreed with the given proposition, on a scale from 1 (strongly disagree) to 5 (strongly agree). Tables below report the percentage responses to survey questions, along with the percent who did not answer (N/A) and the average numerical response for those who answered. The wording of each proposition is in hold. Note that we generally leave out those who are neutral on a proposition when discussing the ratio of those agreeing to those disagreeing. The tables also report whether or not the mean responses of the other three groups are statistically significantly different from the group under consideration at the 95% confidence level. In all of the tables that follow, a = significantly different than the responses for AEA members; b = significantly different than the responses for APSA members; c = significantly different than the responses for PCS-E members; d = significantly different than the responses for PCS-P members. Finally, a * indicates that the group mean is different from 3 (neutral) at the 95% confidence level, using a two-tailed t-test.

I. Assumptions about Political Actors

Perhaps the most fundamental proposition on the survey states that "Individuals are rational utility-maximizers." In his survey of the field, Mueller (2003, 1-2) argues that "The basic behavioral postulate of public choice, as for economics, is that man is an egoistic, rational, utility maximizer." Our findings bear out his conclusion and show that this assumption sets public choice scholars--both those affiliated with economics departments and those affiliated with political science departments--off from typical political scientists. More political scientists reject this proposition than accept it (43% vs. 33%), while economists and public choice political scientists who aren't neutral accept it by a margin of over 5 to 1. Public choice economists accept it by the largest margin (Table 1).

How do individuals maximize in the political arena? Beneath the broad assumption of rationality (or lack thereof) lies a set of assumptions about the process of public choice and the behavior of groups.

Most economists and public choice economists assume that "Most politicians are solely office-seeking vote maximizers." Public choice political scientists are more likely to hold than to reject this axiom but by a slimmer margin. Again political scientists reject economists' preferred assumption about this maximizing behavior, disagreeing with this proposition by about 5 to 3. However, the gap on this proposition is considerably smaller than on the rational utility-maximizer question (Table 2).

The vast majority of economists and public choice scholars accept the assumption that "Bureaucrats are budget-maximizers." On this question, however, political scientists agree with them, with responses that are virtually identical to those of the other groups (Table 3).

Businessmen act in the political arena, as well. All four groups agree strongly that "Firms often seek regulation to protect them from new competition." This is one of the strongest consensuses in the entire survey and virtually no one disagrees with it (Table 4).

Why and how do voters vote? This has proven to be a thorny issue. Economists, political scientists and public choice economists have failed to collectively reach a conclusion (or is this an assumption?) about whether or not "The act of voting in general elections is rational," despite differing substantially on whether or not individuals are best viewed as rational utility maximizers. These three groups are only a little more likely than not to accept this proposition. However, public choice political scientists do accept the rationality of voting by more than 2 to 1 (Table 5).

One problem is that it is difficult to conclude from a cost-benefit analysis that any individual voter should ever vote in a general election. The marginal costs are obviously greater than zero and the material marginal benefits seem to many observers to be zero (or very close to zero). Many economists agree, therefore, that such voting isn't rational--keeping the standard definition of rationality but holding this as an atypical exception to the rule that individuals are rational utility maximizers. Yet, slightly more of them see the act of voting as rational--unwilling to discard this central assumption, perhaps noting that changes in marginal benefits and costs often matter in determining turnout. Interestingly, public choice political scientists are even more willing than economists to view this behavior as rational, even though they aren't as convinced of the general rational utility maximizing behavior of people.

Why would one vote despite the likelihood that the opportunity costs exceed the material benefits? One way to explain this is that there are immaterial benefits, for example that "Voters vote out of a sense of civic duty." A substantial majority of all four groups agrees with this proposition, with very little differentiation (Table 6).

II. Normative Beliefs about Government and Voting

In addition to these varying assumptions about the behavior of political actors, scholars hold differing normative beliefs about the legitimate roles of government and how democracy and voting should work.

Economists, political scientists, and public choice political scientists generally believe that "The redistribution of income is a legitimate role for governments," accepting this legitimacy by wide majorities of about 6 or 7 to 1. However, public choice economists are less enthusiastic, with a substantial minority rejecting this proposition (Table 7).

Only economists (who may not have considered this matter as deeply as the other three groups), are more likely to accept than reject the proposition that "Simple majority rule is the preferred way to make group decisions." Political scientists and public choice political scientists are slightly disinclined toward simple majority voting--wary of the "tyranny of the majority" and other problems--while public choice economists reject it by a ratio of nearly 3 to 1 (Table 8).

Finally, majorities among economists, public choice economists, and public choice political scientists agree that, "Voting rules should limit the incentive for strategic voting." However, political scientists dismiss this injunction by more than 2 to 1 (Table 9).

The remainder of our survey propositions are probably better classified as conclusions about political and economic outcomes that have been widely discussed in the public choice literature. The results indicate that, despite the differing behavioral assumptions and normative beliefs of these scholars, consensus has been reached on several public choice questions. However, there is a lack of consensus on many issues, which appears to be at least partly related to the different first principles of these scholars.

III. Elections and Economic Performance

All four groups overwhelmingly agree that, "National election outcomes are dependent on current economic conditions" (Table 10).

The majority of all four groups reject the proposition that, "Incumbent politicians generate political business cycles." Although the average responses are similar across the four groups, public choice political scientists reject the proposition most strongly, while public choice economists reject it least strongly (Table 11).

The four groups' responses are very similar on a third election-economy proposition. In all four cases, a sizeable majority of those who are not neutral (and many are neutral) have concluded that "Election timing in parliamentary systems is primarily dependent on economic conditions." However, few strongly agree with this proposition (Table 12).

The next three propositions concern how the political situation affects economic growth.

All four groups overwhelmingly agree that "Political rights and civil liberties promote economic growth" (Table 13).

All four groups, by margins of about 2 or 3 to 1, reject the proposition that "Economies grow faster after coups and revolutions due to the destruction of existing special-interest organizations" (Table 14).

Finally, none of the four groups accepts the proposition that "Economic growth is higher when parties of the left are in office." Public choice economists soundly reject this conclusion (with only 5 percent accepting it), while economists and pubic choice political scientists are about three times as likely to reject as to accept. Political scientists are only slightly more likely to reject than accept the conclusion. These differences may reflect differences in conclusions drawn from the research literature, but they hint at ideological predilections with political scientists being more open to the idea that leftist parties do more to enhance economic growth and public choice economists more open to the opposing argument. A one-point increase in the response about individuals as rational utility maximizers is associated with a 0.25-point decrease (significant at the 99 percent confidence level) in the left-brings-higher-economic-growth response, which accords with the popular perception that those accepting the rational maximizing postulate are more to the right and those rejecting it are more to the left politically (Table 15).

IV. Parties, Platforms, Voting and Preferences

All four groups are nearly unanimous, strongly agreeing with the proposition that "In a two-candidate election, both candidates will moderate their positions to attract votes." This is one of the two strongest consensuses in the survey (Table 16).

Likewise, all four groups have concluded that "Simple majority rule prevents effective third-party competition" (Table 17).

All four groups reject the proposition that "Voters seek to balance representation across the parties," with both types of public choice scholars most strongly denying the conclusion by ratios of 5 or 6 to 1 (Table 18).

Political scientists, public choice economists and public choice political scientists generally disagree with the conclusion that "Legislative voting cycles occur frequently." Political scientists are especially likely to reject this conclusion by a margin of almost 4 to 1. Economists accept the conclusion, but almost half are neutral on the issue and nearly a third declined to answer this question, perhaps reflecting unfamiliarity with the issue (Table 19).

Finally (ironically), there is little consensus among any of the groups on the issue of whether or not, "Group preferences are typically intransitive" (Table 20).

V. Government's Purposes and Growth

The next series of related questions deals with the issues of why government exists and why it has grown. The questions are purposely broad, not specifying place and time, although it is probably the case that (since all the respondents live in the U.S.) the reference is the U.S. and its experience with government growth during the twentieth century.

The first question asks if "Governments exist primarily to produce public goods." Given the manifold roles and functions, both economic and non-economic, of government, it is not surprising that few strongly support this idea. However, a sizeable majority of political scientists agree that the production of public goods (nonrival, nonexcludable goods that range from defense to environmental protections) is the top reason governments exist. Public choice economists and economists disagree. More of them reject than accept the idea that the production of public goods is the main role of government. Public choice economists are least supportive of this explanation, with rejecters exceeding accepters by 2 to 1, but public choice political scientists are about evenly split on the issue (Table 21).

One alternative is that "Governments exist primarily to redistribute wealth." Political scientists reject this proposition by a 3 to 1 margin--strongly preferring the alternative that governments exist primarily to produce public goods. Economists and public choice political scientists also generally reject the primacy of wealth redistribution. However, a slight major of the public choice economists who aren't neutral accept the proposition (Table 22).

A closely related proposition is that "Most government programs are driven by rent seeking"--i.e, driven by efforts to use the political system to transfer resources to a favored group. The responses are strikingly similar to the previous question, with average responses all about half a point higher. Political scientists lean against the proposition (though their high non-response rate may indicate an unfamiliarity with the term "rent seeking"). Public choice political scientists are almost evenly split, economists slightly favor this interpretation, and public choice economists clearly endorse it. In addition, a one-point increase in agreement with the rational utility maximizer proposition is associated with a 0.27-point increase (significant at 99%) in this response, while a one-point increase in agreement that governmental redistribution is legitimate is associated with a 0.26-point decrease (significant at 99%) in the response to the rent-seeking question. Thus, those who reject the legitimacy of redistribution/rent-seeking are more likely to believe that it is ubiquitous (Table 23).

Overall these responses seem to imply that quite a few public choice economists have a different, darker view of government than those in the other three groups. Many of them believe that economic redistribution is not a legitimate role of government, but see redistribution as its primary function and rent-seeking behavior as rampant.

Why has government grown? We contrast five competing, but not mutually exclusive, explanations. All four groups show support for all five of the explanations, although they differ on which receives the strongest support.

All four groups agree that "The size of government has grown due to the proliferation of special-interest groups," with public choice economists most likely to support this claim and public choice political scientists least likely to do so (Table 24).

All four groups are more likely to agree than disagree that "The size of government has grown due to expansion of the franchise," with public choice economists very strongly agreeing and economists showing signs of unfamiliarity with the issue (Table 25).

All four groups are more likely to agree than disagree that "The size of government has grown due to bureaucratic self interest." Sizeable majorities of both economists and public choice economists accept this thesis (Table 26).

Economists, public choice political scientists and especially political scientists generally agree that "The size of government has grown due to an increase in the size of government desired by voters." Public choice economists, however, are on average neutral on this proposition (Table 27).

Finally, all four groups lean toward the proposition that "The size of government has grown due to the failure of government to return to its initial size after a crisis." Public choice economists are the strongest supporters of this conclusion (Table 28).

Despite many similarities, public choice economists (and to a lesser extent economists) stand out as being most likely to think that government growth does not reflect the overall desire of voters, but has occurred by "accident" (due to the lingering affects of crises such as world wars and the Great Depression, for example), aided and abetted by self-interested bureaucrats and rent-seeking special-interest groups. Likewise, political scientists are most likely to conclude that voters got the government growth that they wanted, downplaying (relative to the other groups) the roles of crisis, bureaucrats and special-interests.

VI. Individual Behavior: Voting-with-Feet and Free Riding

Two questions about "voting-with-feet" behavior show substantial disagreement. Political scientists reject (by a ratio of more than 3 to 1) the conclusion that "Individuals often move in response to local government taxes and expenditures." Public choice political scientists reject the idea by almost 2 to 1. Economists are almost evenly split on the proposition, while public choice economists strongly back it, with those agreeing outnumbering those disagreeing by over 4 to 1. The response to this question is more closely associated with the rationality proposition than almost any other. A one-point increase in agreement with the rational-utility-maximizer response is associated with a 0.35-point increase (significant at 99%) in the numerical response to this question (Table 29).

Nor is there consensus on the proposition that "Voting-with-the-feet migration sorts people into groups with homogeneous tastes." Political scientists are almost evenly split on this question. Economists and public choice political scientists lean in favor of it, but responses tend to indicate weak approval or disapproval. However, public choice economists clearly agree with the conclusion--although their responses are mostly 4's rather than 5's. Again, these two questions show a significant divide, with public choice economists likely to conclude that voting-with-feet behavior is frequent and important, but political scientists much less likely to believe this is the case (Table 30).

There is accord on a third question about individual behavior. Only a handful reject the conclusion that "In large-group settings, individuals typically free-ride on others' contributions" (Table 31).

VII. Government and the Market

All four groups favor the conclusion that "Government does more to protect and create monopoly power than it does to prevent it." However, public choice economists are convinced most strongly and political scientists least strongly (Table 32).

The next question asked whether or not, "When transaction costs are low, the market achieves the efficient allocation of resources regardless of how property rights are assigned." This proposition is also known popularly as the Coase Theorem. Political scientists (who are probably much less familiar with the argument) reject this conclusion by more than 2 to 1. The other groups generally favor it. Among public choice economists agreement is more than four times the level of disagreement (Table 33).

Finally, the survey also included two propositions that are not on public choice issues per se, but which deal with the economic effects of government policies. The following two have been used previously in surveys of economists and verify that our sample of economists is similar to previous surveys (for example, Alston, Kearl and Vaughan, 1992). These questions also help pinpoint important differences among the four groups.

Does "A minimum wage increase unemployment among young and unskilled workers"? Political scientists generally answer this question negatively. Their conclusion is sure to raise eyebrows among many economists (who have probably studied the issue in greater detail), as they generally answer affirmatively (by about 2 to 1). Nearly two-thirds of public choice economists strongly agree that minimum wages increase unemployment, with only a handful rejecting this conclusion. This gap between the political scientists and the public choice economists is the largest numerical gap in the survey (Table 34).

All four groups conclude that "Tariffs and imports quotas usually reduce general economic welfare," but again there are wide gaps, with political scientists showing less (but still strong) support and public choice economists virtually unanimous in this conclusion (Table 35).

These final four questions indicate that economists--and especially public choice economists--are impressed by the power of the market and fairly wary of government intervention, which they agree brings unemployment, reduced well-being and monopoly in the cases discussed above. Political scientists are much less sanguine about the market and less wary--though still somewhat wary--of government interventions in these areas.

VIII. Major Conclusions

The purposes of this inquiry are to examine differences in the assumptions and conclusions reached by economists, political scientists, and public choice scholars and to discover where consensus has and hasn't been reached on important public choice questions.

Where has consensus been reached? There are several propositions on which for all four groups the percent of the non-neutrals agreeing outweighs the percent disagreeing by at least 60 to 40 percent (an electoral blowout). These are:

* Bureaucrats are budget-maximizers.

* Firms often seek regulation to protect them from new competition.

* Voters vote out of a sense of civic duty.

* The redistribution of income is a legitimate role for governments.

* National election outcomes are dependent on current economic conditions.

* Election timing in parliamentary systems is primarily dependent on economic conditions.

* Political fights and civil liberties promote economic growth.

* In a two-candidate election, both candidates will moderate their positions to attract votes.

* Simple majority rule prevents effective third-party competition.

* The size of government has grown due to the proliferation of special-interest groups.

* The size of government has grown due to expansion of the franchise.

* The size of government has grown due to bureaucratic self interest.

* The size of government has grown due to the failure of government to return to its initial size after a crisis.

* In large-group settings, individuals typically free-ride on others' contributions.

* Government does more to protect and create monopoly power than it does to prevent it.

In addition the proposition that "Economies grow faster after coups and revolutions due to the destruction of existing special-interest organizations" and "Voters seek to balance representation across the parties" are soundly rejected by all four groups.

Obviously some (and in a few cases many) disagree with these "consensus" results. One use of this study is to quantify the field's beliefs so that they are all the more clear and so that scholars can compare themselves to one another--perhaps reassessing their conclusions when they are at odds with the majority, perhaps redoubling their efforts to convince others of their errors. The cycle of scholarship never ends and few of these debates will soon cease. Future studies can use these responses to track the evolution of the field.

The responses also seem to indicate that some of these debates will be timeless because scholars do not see eye to eye in their basic assumptions about human behavior. Several conclusions about debates on public choice issues are associated with fundamental assumptions about rational utility maximization and/or the legitimacy of redistribution by government. Divisions between political scientists and economists on the question of rationality and self-interested behavior seem to play an important role--pushing economists (and to an even greater extent public choice economists) to a greater wariness than political scientists about government. (5)

TABLE 1

Category 1 2 3 4 5 N/A Average

AEAI (b,c) 7 6 17 56 14 0 3.64 *
APSA (a,c,d) 10 33 24 30 3 0 2.82
PCS-E (a,b,d) 2 7 5 49 37 0 4.12 *
PCS-P (b,c) 0 12 24 50 15 0 3.68 *

TABLE 2

Category 1 2 3 4 5 N/A Average

AEA (b) 4 19 19 38 20 1 3.51 *
APSA (a,c) 12 42 16 27 3 0 2.67 *
PCS-E (b) 11 13 23 36 18 2 3.38 *
PCS-P 12 21 21 38 9 0 3.12

TABLE 3

Category 1 2 3 4 5 N/A Average

AEA 3 18 16 47 16 3 3.56 *
APSA 2 20 20 41 17 5 3.52 *
PCS-E 5 11 12 54 18 0 3.68 *
PCS-P 15 15 15 47 9 0 3.21

TABLE 4

Category 1 2 3 4 5 N/A Average

AEA (c) 0 4 9 49 38 1 4.20 *
APSA (c) 0 2 19 46 34 3 4.12 *
PCS-E (a,b) 0 0 5 28 67 0 4.61
PCS-P 0 0 15 29 56 0 4.41 *

TABLE 5

Category 1 2 3 4 5 N/A Average

AEA 16 21 15 37 12 3 3.07
APSA 11 27 20 30 12 2 3.06
PCS-E 16 18 18 30 18 2 3.16
PCS-P 12 15 9 41 24 0 3.50 *

TABLE 6

Category 1 2 3 4 5 N/A Average

AEA 0 6 16 57 21 0 3.94 *
APSA 0 17 12 55 17 2 3.71
PCS-E 2 4 20 54 21 2 3.89 *
PCS-P 0 3 9 70 18 0 4.03 *

TABLE 7

Category 1 2 3 4 5 N/A Average

AEA (c) 7 4 9 30 49 1 4.10 *
APSA (c) 6 6 9 44 35 2 3.95 *
PCS-E (a,b,d) 19 16 14 26 25 0 3.21
PCs-P (c) 6 6 18 38 32 0 3.85 *

TABLE 8

Category 1 2 3 4 5 N/A Average

AEA (c) 8 20 30 34 8 13 3.15
APSA 12 29 29 26 3 3 2.78
PCS-E (a) 22 38 18 18 4 4 2.44 *
PCS-P 15 15 39 24 6 3 2.91

TABLE 9

Category 1 2 3 4 5 N/A Average

AEA (b) 6 18 28 33 16 27 3.35 *
APSA (a,c,d) 24 32 21 21 3 6 2.48 *
PCS-E (b) 2 13 22 50 13 5 3.59 *
PCS-P (b) 6 32 12 29 21 0 3.26

TABLE 10

Category 1 2 3 4 5 N/A Average

AEA (b) 0 3 13 61 23 0 4.04 *
APSA (a,c,d) 1 6 24 60 9 0 3.69 *
PCS-E (b) 0 4 16 60 21 0 3.98 *
PCs-P (b) 0 0 15 59 27 0 4.12 *

TABLE 11

Category 1 2 3 4 5 N/A Average

AEA 16 28 30 21 5 13 2.70 *
APSA 11 42 22 22 5 5 2.68 *
PCS-E 7 35 27 27 4 4 2.85
PCS-P 18 39 12 27 3 3 2.58 *

TABLE 12

Category 1 2 3 4 5 N/A Average

AEA 0 22 31 41 6 27 3.31
APSA 3 20 24 46 7 12 3.32 *
PCS-E 0 8 38 42 12 12 3.58 *
PCS-P 6 12 18 65 0 0 3.41 *

TABLE 13

Category 1 2 3 4 5 N/A Average

AEA (b) 2 3 16 45 34 9 4.08 *
APSA (a,c) 0 16 15 45 24 0 3.76 *
PCS-E (b) 0 9 11 41 39 2 4.11
PCS-P 0 7 14 50 29 3 3.94 *

TABLE 14

Category 1 2 3 4 5 N/A Average

AEA 16 38 26 19 2 17 2.53 *
APSA 23 37 23 14 4 15 2.39 *
PCS-E 15 27 40 10 8 9 2.67 *
PCS-P 23 27 33 13 3 12 2.47 *

TABLE 15

Category 1 2 3 4 5 N/A Average

AEA (b,c) 22 29 35 13 2 10 2.43 *
APSA (a,c) 11 29 35 23 3 2 2.79 *
PCS-E (a,b,d) 40 24 31 5 0 4 2.02 *
PCS-P (c) 18 30 36 12 3 3 2.52 *

TABLE 16

Category 1 2 3 4 5 N/A Average

AEA (b) 0 1 1 40 57 3 4.53 *
APSA (a,c,d) 3 9 11 41 36 2 3.98 *
PCS-E (b) 0 2 2 42 54 0 4.49 *
PCS-P (b) 0 3 6 44 47 0 4.35 *

TABLE 17

Category 1 2 3 4 5 N/A Average

AEA 4 11 18 50 18 20 3.68 *
APSA 8 12 8 46 26 3 3.71
PCS-E 0 18 24 41 18 11 3.59 *
PCS-P 9 24 3 35 29 0 3.53 *

TABLE 18

Category 1 2 3 4 5 N/A Average

AEA 18 40 21 19 1 4 2.46 *
APSA 27 25 31 16 0 0 2.37 *
PCS-E 31 35 22 13 0 4 2.16 *
PCS-P 35 24 32 9 0 0 2.15 *

TABLE 19

Category 1 2 3 4 5 N/A Average

AEA (a,c,d) 4 10 43 37 6 30 3.31 *
APSA (a) 13 21 36 26 4 21 2.87
PCS-E (a) 8 28 43 20 2 11 2.80
PCS-P (a) 18 36 30 6 9 3 2.52 *

TABLE 20

Category 1 2 3 4 5 N/A Average

AEA 3 24 32 36 5 16 3.15
APSA 4 27 39 25 6 22 3.02
PCS-E 4 23 32 32 9 7 3.21
PCS-P 0 33 30 24 12 3 3.15

TABLE 21

Category 1 2 3 4 5 N/A Average

AEA (b) 17 33 16 26 9 0 2.76
APSA (a,c) 6 19 16 45 13 0 3.40 *
PCS-E (b) 24 30 19 24 4 5 2.54 *
PCS-P 9 29 24 24 15 0 3.06

TABLE 22

Category 1 2 3 4 5 N/A Average

AEA 14 33 19 31 3 0 2.76
APSA (c) 17 35 33 14 2 2 2.48 *
PCS-E (b,d) 13 19 28 28 13 5 3.09
PCs-P (c) 15 41 32 12 0 0 2.41

TABLE 23

Category 1 2 3 4 5 N/A Average

AEA (c) 6 23 23 34 13 9 3.23*
APSA (c) 9 30 33 26 2 31 2.83
PCS-E (a,b,d) 0 23 13 38 27 2 3.68*
PCS-P (c) 3 27 42 21 6 3 3.00

TABLE 24

Category 1 2 3 4 5 N/A Average

AEA 1 12 19 39 29 1 3.83 *
APSA (c) 6 15 16 51 12 0 3.48 *
PCS-E (b,d) 0 16 7 37 39 2 4.00 *
PCs-P (c) 3 27 18 35 18 0 3.38 *

TABLE 25

Category 1 2 3 4 5 N/A Average

AEA (c) 6 20 30 37 7 23 3.20
APSA (c) 6 20 20 48 6 2 3.29 *
PCS-E (a,b,d) 4 9 20 42 25 4 3.76 *
PCs-P (c) 12 18 18 44 9 0 3.21

TABLE 26

Category 1 2 3 4 5 N/A Average

AEA (b) 4 7 19 48 22 1 3.75 *
APSA (a,c) 3 20 30 44 3 2 3.24 *
PCS-E (b) 2 14 19 35 30 0 3.77 *
PCS-P 3 18 24 44 12 0 3.44 *

TABLE 27

Category 1 2 3 4 5 N/A Average

AEA (b) 13 13 22 39 13 1 3.26
APSA (a,c) 6 4 19 45 25 0 3.79 *
PCS-E (b) 11 30 11 47 2 0 3.00
PCS-P 6 18 15 47 15 0 3.47 *

TABLE 28

Category 1 2 3 4 5 N/A Average

AEA 4 19 24 42 10 4 3.34 *
APSA (c) 9 21 18 44 8 2 3.20
PCS-E (b) 2 16 19 49 14 0 3.58 *
PCS-P 3 21 24 41 12 0 3.38 *

TABLE 29

Category 1 2 3 4 5 N/A Average

AEAI (b,c) 7 35 19 35 3 3 2.91
APSA (a,c) 18 45 18 16 3 0 2.42 *
PCS-E (a,b,d) 4 11 23 46 16 2 3.61
PCs-P (c) 9 46 15 24 6 3 2.73

TABLE 30

Category 1 2 3 4 5 N/A Average

AEAI (b,c) 0 23 31 46 0 7 3.23
APSA (a,c) 17 25 15 37 6 3 2.91
PCS-E (a,b,d) 2 4 21 63 11 2 3.77 *
PCs-P (c) 3 35 15 44 3 0 3.09

TABLE 31

Category 1 2 3 4 5 N/A Average

AEA 1 1 15 54 28 4 4.06 *
APSA (c) 1 5 20 52 23 2 3.89 *
PCS-E (b) 0 4 9 54 33 0 4.18 *
PCS-P 0 6 9 44 41 0 4.21

TABLE 32

Category 1 2 3 4 5 N/A Average

AEA (c) 1 27 26 26 20 0 3.36 *
APSA (c) 2 18 40 29 11 3 3.29 *
PCS-E (a,b) 0 14 19 21 46 0 3.98 *
PCS-P 0 21 21 44 15 0 3.53 *

TABLE 33

Category 1 2 3 4 5 N/A Average

AEA (b,c) 11 19 13 43 14 10 3.30
APSA (a,c,d) 18 39 18 18 7 9 2.56 *
PCS-E (a,b,d) 4 14 5 35 42 0 3.98 *
PCs-P (b,c) 15 21 6 47 12 0 3.21

TABLE 34

Category 1 2 3 4 5 N/A Average

AEA (b,c) 7 22 13 20 38 1 3.59 *
APSA (a,c,d) 26 37 15 14 8 3 2.40 *
PCS-E (a,b,d) 4 4 5 25 63 2 4.39 *
PCs-P (b,c) 6 29 18 35 12 0 3.18

TABLE 35

Category 1 2 3 4 5 N/A Average

AEA (b,c) 1 3 9 29 58 1 4.39 *
APSA (a,c,d) 2 17 22 37 22 12 3.61 *
PCS-E (a,b,d) 0 0 4 21 75 0 4.72 *
PCs-P (b,c) 0 3 12 53 32 0 4.15 *

Category AEA APSA PCS-E PCS-P

AEA -- 16 14 1
APSA 16 -- 24 7
PCS-E 14 24 -- 12
PCS-P 1 7 12 --

Notes:

(1.) The survey was mailed in October 2000. The figures on number of surveys sent do not include envelopes that were returned as undeliverable. Among AEA members we sampled only those with a Ph.D., teaching in economics departments in the U.S. Among APSA members we sampled only those with a Ph.D., teaching in departments of political science, politics, or government in the U.S. There is no formal public choice association. Rather, anyone who registers for the Public Choice Society conference is included on the mailing list. We thank the late Carol Roberts for supplying us with this list. Among PCS members we define economists as anyone teaching in an economics department, and political scientists as anyone teaching in a politics department, regardless of their degree specialization. Anyone whose mailing address did not list an economics or politics/political science/government department was not included in the survey.

(2.) The cover letter followed this form: "The enclosed questionnaire is designed to assess the opinion of political scientists on issues concerning public choice. Your name has been randomly selected from the APSA's membership directory. We hope to get a truly representative sample of APSA members and, obviously, a higher response rate will help ensure this. Feel free to skip over any questions about which you do not have an opinion or do not have enough information to respond. Please return the survey in the enclosed stamped envelope to ensure anonymity." Other than the name of the discipline and professional organization, the wording was identical across the groups. Because we were on a tight budget and didn't know the identity of the respondents, we didn't send any follow-up reminders. Because the surveys were anonymous, we do not know whether or not nonresponders are systematically different than responders in any dimension except for sex, where the response rates do not differ significantly.

(3.) The updated and expanded version is Mueller (2003).

(4.) We thank John Dinan, Keith Dougherty, Don Frey, P.J. Hill and Andrew Morriss for their comments.

(5.) The summary table below gives the number of cases across the 35 questions in which the responses of each pair of groups differ significantly at the 95% confidence level. It is obvious that the views of the APSA members and the PCS economists are the most divergent.

References

Alston, R.M., Kearl, J.R. and Vaughan, M.B. (1992) "Is There a Consensus among Economists in the 1990s?" American Economic Association Papers and Proceedings, vol. 82, no. 2, pp. 203-9.

Mueller, D. (1989) Public Choice II, New York: Cambridge University Press.

Mueller, D. (2003) Public Choice III, New York, Cambridge University Press.

Robert Whaples * and Jac C. Heckelman **

* Department of Economics, Wake Forest University, Winston-Salem, NC 27109-7505; 336-758-4916, whaples@wfu.edu

** Department of Economics, Wake Forest University, Winston-Salem, NC 27109-7505; 336-758-5923, heckeljc@wfu.edu