Download Political Selection and more Lecture notes Political Economy in PDF only on Docsity!
Political Selection
Timothy Besley
J
ames Madison (1788 [1961]), the primary author of the U.S. Constitution, wrote in the Federalist Papers (#57): “The aim of every political Constitution, is or ought to be, first to obtain for rulers men who possess most wisdom to discern, and most virtue to pursue, the common good of society; and in the next place, to take the most effectual precautions for keeping them virtuous whilst they continue to hold their public trust.” The second issue that Madison identifies— that is, the need to get institutional incentives right so that politicians act in the public interest— has been studied at length. But the first issue—the need to form a political class that is competent and honest enough to discharge its duties— has received far less attention. Yet many of the problems of government, especially those in emerging democ- racies, arise because the political class uses the state for self-dealing. Any institu- tional improvements that try to rein in their proclivities can have only a limited impact as long as policymakers are intent on using public office for self-enrichment. Democratic theorists have long recognized that the state functions in the context of a nexus of values that restrain narrow self-interest. One of the key roles of elections is to uphold those values, both by disciplining politicians and providing a means of finding those who are fit to serve. As the great American political scientist V.O. Key (1956, p. 10) remarks: “The nature of the workings of government depends ultimately on the men who run it. The men we elect to office and the circumstances we create that affect their work determine the nature of popular government. Let there be emphasis on those we elect to office.” The past 200 years of political history justify an emphasis on selection. Almost
y Timothy Besley is Professor of Economics and Political Science and Director of the Suntory
and Toyota International Centres for Economics and Related Disciplines (STICERD), both at the London School of Economics, London, United Kingdom.
Journal of Economic Perspectives—Volume 19, Number 3—Summer 2005—Pages 43– 60
every major episode of economic change over this period has been associated with key personalities coming to power with a commitment to these changes. In Ger- many of the 1870s and 1880s, for example, Otto von Bismarck was the architect of early moves toward the creation of a welfare state. No account of the landmark economic reforms of the Progressive Era in the United States fails to give a central role played by Theodore Roosevelt, and no account of the New Deal fails to give a central role to Franklin Delano Roosevelt. The history of China’s Great Leap Forward is written in terms of Mao’s vision and how he imposed it. In the United Kingdom, Margaret Thatcher’s economic legacy of privatization and deregulation was in significant measure due to her single-minded pursuit of these goals. But if such dynamic leaders are so important, then we need to understand how they come to hold the reins of power. This outcome could be viewed as largely the product of random events colored by idiosyncratic personalities and chance en- counters. However, at least some role must be given to the underlying institutional structure, which has a more systematic influence on who rises to the top. Thus, it is essential to understand how political selection works. Much of the modern literature on political economy literature has not only neglected the problem of political selection, it has been positively hostile to the topic. For example, James Buchanan (1989, p. 18) argues: “To improve politics, it is necessary to improve or reform rules, the framework within which the game of politics is played. There is no suggestion that improvement lies in the selection of morally superior agents who will use their powers in some public interest.” A similarly narrow view has also been encouraged in political science, too, through influential commentators such as Mayhew (1974, p. 5) whose analysis casts U.S. congressman as “single-minded seekers of re-election.” But it seems implausible to argue that the quality of government can always be improved with such a narrow focus. Institutions that shape incentives are often fragile and provide only limited restraints. At one extreme is the case of Adolf Hitler who, after his election as Chancellor of Germany, showed contempt for the constraints placed through the constitution. A number of modern day autocrats, some of whom were initially elected, show similar contempt for institutions. The reluctance of academic writers on political economy to embrace the importance of selection is somewhat strange, since applying these ideas in our own backyard is second nature. For example, in the preamble to its “Principles Govern- ing Research at Harvard,” the Faculty of Arts and Sciences at Harvard (http:// www.fas.harvard.edu/research/greybook/principles.html) states: “The primary means for controlling the quality of the scholarly activities of this Faculty is through the rigorous academic standards applied in selecting its members.” Recent accounts of corporate performance similarly place weight on charis- matic chief executive officers (CEOs) and the way in which they shape corporate strategies. Indeed, Bertrand and Schoar (2003) argue that the data can be de- scribed in terms of CEO “fixed effects” indicative of management styles. Corporate governance is an area where there is an apparent disconnect between the obsession
44 Journal of Economic Perspectives
pivotal role played by Paul Volcker in establishing monetary policy credibility in the 1980s is now legendary. Osborne and Slivinsky (1996) and Besley and Coate (1997) suggest an alter- native to the Downsian paradigm in which selection is fundamental to achieving policy credibility.^1 Competition is modeled between candidates who cannot com- mit to policies in advance. Election promises become credible precisely because a suitable set of candidates can be found to carry them through after they are elected. The approach is manifestly relevant for directly elected chief executives, such as presidents and mayors. However, it is also useful in understanding party com- petition in parliamentary systems. In such cases, it is arguable that parties rather than candidates are selected by voters. In such cases, the party leadership creates the public image of the party. Moreover, electoral fortunes often turn on battles for party leadership, given the importance of leaders in subsequent policy formation. A good example is the leadership battle in the ruling Liberal Democratic Party (LDP) in Japan in 2001 over who would replace Yoshiro Mori. The merits of the prospective leaders were debated in good measure in terms of their competence to lead economic change. The importance of having a credible party leadership explains why it may be difficult for parties to converge to the median of public opinion, even if converging would improve a party’s electoral chances. Support for this approach in the U.S. Congress comes from Lee, Moretti and Butler (2004). They employ an ingenious method that exploits the “regression discontinuity” in elections where one candidate wins by a fine margin. They find that the degree of electoral strength within a district has no impact on subsequent representatives’ roll call voting records. Thus, a large exogenous shift in electoral strength for the Democrats in a district does not lead both parties’ nominees to shift to the left as the Downsian paradigm would predict. The main focus of standard models of policy competition is broad ideological policy conflicts, but the set of specific policy cleavages to which these issues applies is much wider. Candidates can have many “identities” in the sense of Akerlof and Kranton (2005), each of which can affect their policy preferences. Changing the identities of policymakers can serve as a means for making policy change credible. Attempts to engineer the political process to change the complexion of policymaking bodies are frequently greeted with derision. But if identity matters and credibility is important, such steps cannot be dismissed as empty rhetoric. Consider, for example, debates in the United States in the aftermath of the Voting Rights Act of 1965. One result of the act was the creation of majority-minority districts, in which minorities have a majority of voters in the district, and thus a member of minority group has a strong chance of being elected. This practice has led to controversy, with a concern that when minorities are grouped into a few districts, they may end up being marginalized in the political system as a whole. The state of Georgia recently decided to redistrict away from majority-minority districts
(^1) See Persson and Tabellini (1994) for an early contribution recognizing how elections establish credibility through picking the right people.
46 Journal of Economic Perspectives
and instead to move toward a greater number of minority-influence districts, where minorities would be found in a larger share than in the general population but not as an outright majority in the district. The U.S. Supreme Court upheld this “minority influence” redistricting in Georgia v. Ashcroft (539 U.S. 02-182 [2003]). But whether this entire dispute matters ultimately rests on whether selecting minority groups as legislators makes a difference to the conduct of policy. Emerging evidence on the importance of identity in political representation suggests that selection can matter. India, for example, has reserved legislative seats for traditionally disadvantaged caste groups. Pande (2003) finds that reservation for scheduled castes and scheduled tribes at the state level in India affected policies targeted toward these groups. A further case where identity matters is the case of gender representation. In standard models, men and women are equally good at representing policy prefer- ences of women voters. Thence, the historical under-representation of women in elected office should be irrelevant, provided that women have the right to vote. But evidence suggests otherwise. The 73rd amendment to the Indian constitution, enacted in 1992, mandated that a certain fraction of seats in village governments be reserved for women. Chattopadhyay and Duflo (2004) study the outcomes in two Indian states—Rajasthan and West Bengal—and find that the issues favored by women get more attention when women politicians are selected. For example, significantly more attention is paid to water projects, with more than a 25 percent increase in water provision as a consequence of reservation. Such evidence puts the spotlight on efforts to increase the representation of women in political office. As another example, the Labour party in the United Kingdom has introduced a controversial policy of women-only short lists of candi- dates. Such policies follow from a recognition that selection matters.
Quality
Selection also matters in affecting the quality of politicians in office. At the risk of oversimplification, I will suppose that quality has two principal dimensions: honesty and competence. Of course, other qualities are sometimes prized, but their merits are more difficult to assess. A good example is the ability of politicians to manipulate and/or persuade others. Many landmark reforms, such as the abolition of slavery in the 1860s and civil rights legislation in the 1960s required such skills and a willingness to defy convention (even constitutional rules) in a way that could easily be used against the public interest in other contexts. But Abraham Lincoln and Lyndon Johnson are in each case widely praised for their decisive contributions. If honesty and competence are at stake, we should expect politician quality to be what political scientists call a valence issue— every citizen wants more of it
Political Selection 47
“the public trust.” This fidiciary model of good political motivation is consistent with the idea that some people hold public office due to public service motivation. It is also related to the much older notion of civic virtue, whereby individuals have qualities that make them trustworthy in their dealings with others. This kind of virtue can be thought of as hard-wired into preferences rather being dependent on external reinforcement, and being inclined to perform a fiduciary duty in this sense can be thought of as an intrinsic motivation , pursued for its own sake rather than because of some well-defined external reward. This idea, which originated among psychologists, has only recently been brought into economic thinking.^3 Thus, a politician who upholds his duty to pursue the public interest will do so even when it means foregoing an increase in his wealth or income by doing so.^4 On this view, finding a trustworthy politician is a matter of selection, not a matter of incentives, and political institutions should be designed to select politicians with a greater degree of public service motivation (Cooter, 2002). Voters seem to believe that selection of politicians matters. Since 1962, the U.S. National Election Survey (NES) has asked Americans a number of questions about their attitudes toward government, including whether they believe that government is run by crooks. Figure 1 gives the proportion of those who answer this question by saying that there are “quite a few crooks” and those who answer by saying that there are “hardly any crooks”—and shows a best-fit line for each set of answers. Of course, these public survey results don’t measure the actual number of crooks in political office. Recent decades have seen an increase in the laws and rules to enforce greater honesty among politicians, rising public cynicism about politicians, and also rising public expectations about how honest politicians should be. The 1960s appear to be a decade when public opinion about the honesty of politicians fell sharply, while such opinions rebounded toward a perception of greater honesty in the 1990s. However, the findings do strongly suggest that people believe that politicians are differentiated along a spectrum of honesty and that the intrinsic honesty of politicians matters to voters.
Methods of Selection
There are two ways of judging whether selection of policymakers is healthy. First, such judgments could be made on the basis of the interests that policymakers serve when they formulate policy. While there is inevitable disagreement about the
(^3) Frey (1997) provides an excellent introduction with economic applications. For a recent discussion of identity and motivation in this journal, see the article by Akerlof and Kranton in the Winter 2005 issue. (^4) To assume that politicians who take their fiduciary duties seriously are completely immune to influence by incentives is perhaps too strong. The extent of incentives may depend on whether they agree wholly with the ends that they are pursuing. Besley and Ghatak (2005) develop a model of agent motivation in which the extent to which agents take care depends on the extent to which they agree with the mission being pursued by an organization. Thus, a politician could be much more motivated when he/she agrees with the cause.
Timothy Besley 49
validity of distributional goals, most people would agree that the pursuit of broad- based social objectives rather than narrow interests is better. The experience of ethnically factionalized politicians in parts of Africa is a vivid illustration of the problems that arise when the narrow politics of identity is dominant. Selecting leaders who can encompass broader interests and overcome the strictures of tribe or clan then has many advantages. However, the difficulty of establishing credibility in such contexts is clear. Second, the health of selection could be judged in terms of the honesty and competence of those who are picked for office. History suggests that four main methods of selection to political office are available: drawing lots, heredity, the use of force and voting. The modern world is converging toward the last of these. To understand why, it is useful to discuss the pros and cons of each. For a period of time, ancient Athens filled seats on its legislative council by drawing lots from among its citizens, as Manin (1997) explains in an excellent discussion. Each citizen served for one year, and there was a restriction to two terms in a lifetime. The Greeks understood the downside of this method in terms of ensuring good politicians. They did impose safeguards in the form of a kind of confirmation hearing in which the character and competence of the selected candidate was scrutinized. However, the basic premise behind selection by lot is that civic virtue was widely distributed in the population, so that random selection made it relatively unlikely that anyone picked by the lottery would be a bad politician. Selection by lot was deemed preferable to elections for three main reasons. First, it guaranteed rotation in office, so that politicians were guaranteed to experience both political and everyday life. Second, selection by lot guaranteed the widest possible access to public office and hence was viewed as egalitarian. Third, lots seemed more likely to maintain a unity of purpose in the community, while elections increased the chance that citizens would group into factions.
Figure 1 Perception about Crooks in Government, 1958 –
10
20
30
40
50
1960 1970
quite a few crooks hardly any crooks
Fitted values Fitted values
1980
Response score
1990 2000
50 Journal of Economic Perspectives
Enlightenment thinkers, such as John Locke, who championed the idea that government could only gain its legitimacy from consent given freely by the citizens as expressed through the ballot box. This claim about perceived legitimacy has nothing directly to do with whether elections provide incentives to discipline rulers; instead, it’s a claim about the importance of selection to a well-functioning political system. In the modern world, most autocrats rule by force rather than inheritance. Autocracy’s selection rule is to pick those leaders who can muster the greatest loyalty from coercive institutions such as the army and/or police. While autocracies display a fair amount of heterogeneity in their performance, it seems fair to say that the most brutal regimes in human history have typically been of this kind. Selection by force does not appear to be an effective method either for selecting people who will offer wide-ranging political representation or those who will be the most competent to hold public office. Jones and Olken (2004) offer some evidence that the low-quality government under autocracy is rooted in selection problems. They look at economic growth before and after the death in office of world leaders to provide evidence on whether leader quality matters. Their main finding is that the death of an autocrat leads to a change in growth. Democratically elected leaders show no corresponding pattern. This finding is consistent with selection by force having a disadvantage in selecting good leaders compared to democratic selection. Elections are now the most commonly used institution for selecting a political class, but enormous diversity exists across electoral institutions. There are differ- ences in the wages paid to politicians, the kind of public duties that are required, and the kind of policy discretion that the politician enjoys. There are diverse rules about who can stand for office, such as age, wealth or literacy requirements. Rules about the conduct of elections also differ across democratic systems, including whether there are primaries for multiple candidates from the same party or runoffs between the top two candidates. Rules also differ with regard to the conduct of campaigns and the kind of financial support that candidates may receive. This institutional diversity makes it difficult to draw general lessons about selection in democracies. The details of electoral rules are likely to matter. In the next section, we will discuss how elections work in selecting policymakers.
Making Elections Work
Most polities impose explicit rules that restrict who can run for elected office. Age and residency restrictions are most common. For example, candidates for the U.S. presidency must be American-born and at least 35 years of age. Such barriers are, however, on the whole less stringent than in the past. For example, Britain and France traditionally used wealth tests for candidates, which were much more stringent than those required to be allowed to vote. Some barriers to entry can be justified on grounds of promoting candidate
52 Journal of Economic Perspectives
quality. However, others, like wealth tests, are of more dubious standing. While wealth could be a guarantor of independence or a mark of success in private life, it could equally well be correlated with a taste for personal enrichment. It is an interesting question, which probably deserves more attention, as to whether there are grounds for restricting candidature beyond the qualifications for voting. Supposing that a substantial pool of potential candidates for public office pass any tests to become a candidate, we need then to understand what makes it more likely that good candidates will put themselves forward for election. We will look at this selection process in four stages: 1) the “attractiveness ratio,” which considers the motivation to run for office; 2) the “success ratio,” which is the likelihood that good candidates succeed in elections; 3) the “opportunity cost ratio,” which cap- tures the relevance of outside options on the decision to run for election; and
- the “accountability ratio,” which captures the relative reelection rates of good and bad politicians.
The Attractiveness Ratio Who will have a high level of motivation to run for office and why? Rewards to holding public office come in three main forms: formal compensation (like wages, benefits, travel allowances), public service motivation and other rents. Good poli- ticians are attracted by public service motivation and wages, while bad politicians care about rents and wages. Then a key ratio for determining who will be attracted to run for office in a political system is the attractiveness ratio, defined as
A
rents wages public service motivation wages.
When the attractiveness ratio A appears large to many people, the pool of bad politicians who would like to run for office will be larger. The attractiveness ratio is higher, for example, when available rents are higher. When A appears small to many people, the pool of good politicians who would like to run for office will be larger. The attractiveness ratio will be lower when society has a strong public service ethic. Any system of selection depends on the qualities of the pool of citizens who are potential candidates. Societies comprise individuals with a range of values that are learned in schools, community organizations and families, which have a bearing on how they behave in positions of trust. If relatively few individuals in the population take such responsibilities of trust seriously, then finding such individuals to fill elective offices is likely to be more difficult. A body of recent work on “social capital” has argued that there has been a decline in civic values that manifest themselves in reduced participation in certain social activities. Putnam (2000) argues that this decline in social capital is linked to other social forces in society. In his earlier work on civic traditions in Italy, Putnam (1993) argued that greater civic
Timothy Besley 53
The Success Ratio The attractiveness ratio shapes who will seek to gain political office. But voters care about finding the best politicians in the available pool. If voters are able to filter out bad politicians completely, then it does not matter how attractive political office is to such politicians. Hence, the attractiveness ratio needs to be multiplied by the success ratio defined by
S
probability of election of bad politicians probability of election of good politicians.
In an ideal world, the success ratio would be equal to zero, and bad politicians would never be elected. But the ratio is not zero, for a number of practical reasons. Voting involves a multidimensional decision, involving both a judgment about political competence and also about policy agreement. In some cases, some voters may prefer a low-quality candidate who shares their own policy preferences to a high-quality candidate from the other group. This factor limits the effectiveness of elections in filtering out bad candidates. This insight could explain why ethnically factionalized societies seem so often to spawn bad politicians, like when the break-up of the former Yugoslavia brought Serbian leader Slobodan Milosˇevic´ into office. Candidates are typically chosen by political parties. This fact raises the question of why a party would ever put a bad candidate up for election. One possibility is that if rents are earned by parties as well as successful candidates, and protection of those rents is dependent on selecting bad politicians with little public service motivation, then the party may have an interest in putting up bad candidates. The problem that parties face in making this choice arises from the risk that voters will choose the other party. A striking example of how the absence of competition reduced politician quality is the recent history of the U.S. south, where the near monopoly on power of southern Democrats gave less incentive to select good candidates. Besley, Persson and Sturm (2005) argue that the pattern of policy
Table 1 Relative Wage of U.S. Senators and Law Firm Partners
Year
Senators’ yearly salary (1)
Average salary for a law firm partner (2) (1)/(2)
1954 12,500 10,258 1. 1976 44,600 55,000 0. 2002 150,000 593,800 0.
Sources: Senators’ salary is taken from the U.S. Senate website: http://www.senate. gov/artandhistory/history/common/briefing/senate_salaries.htm. Average salary for law firm partners is taken from Liebenberg (1956; 1954 data), American Bar Association Journal (1977; 1976 data), and Internet Legal Research Group: http://www.ilrg.com/employment/salaries/ (2002 data).
Political Selection 55
formation in the U.S. south before and after the Voting Rights Act of 1965 is consistent with a significant improvement in the quality of state governors in the region as a consequence of the increase in political competition. The selection procedures within parties may also influence the likelihood that a bad candidate is chosen. In many party structures, candidate selection is struc- tured to maximize the power of party elites, with candidate selection being a highly secretive procedure where personal connections could play a large role. This process could allow bad candidates, intent on using political office for private ends, to use their influence. The trend in modern democracies has largely been toward more systematic and transparent systems of selection. If party members who don’t share in the victory rents get a say in picking candidates, then the tendency toward low-quality candidates is diminished. The advent of the primary election in the United States is a case in point, with candidates forced to compete for the party nomination. If parties have relatively homogeneous policy preferences, then the main remaining basis for within-party competition should be candidate quality. Information is also a key determinant of the success ratio. Laws that affect media freedom can be important institutional determinants of political selection. The media can assist voters in identifying the quality of candidates and conversely, political selection will often work poorly in countries where the media is repressed. It is no surprise, therefore, that low-quality leaders frequently try to repress the media. This was famously a feature of Peru under Alberto Fujimori. The impor- tance of press freedom is confirmed by studies such as Brunetti and Weder (2003), which finds a negative correlation between corruption and press freedom. Repres- sion of the press is more difficult when there is a plurality of relatively independent media (Besley and Prat, 2004). Similarly, Djankov, McLeish, Nenova and Shleifer (2003) find that countries with state-owned media tend to be more corrupt.^6 The rules governing campaign finance can also affect the success ratio. But exactly how such rules work depends on who is contributing to campaigns. Cam- paign contributions by special interests are likely to do most damage to politician quality (Coate, 2004). However, broad-based contributions may have the opposite effect by leading to greater advertising of the qualities of good candidates. Political reservation of the kind seen in India also affects the success ratio. By restricting the domain over which candidates are picked, it could make the pool of good candidates smaller. On the other hand, political reservation may limit the extent of policy disagreement between the competing candidates, making it more likely that political competence is a deciding factor. In a study of local government in south India, Besley, Pande and Rao (2005) show that political selection in reserved and unreserved seats is very similar—in both cases politicians tend to be from wealthier, politically connected and more educated backgrounds.
(^6) This view of the media may be somewhat rosy-eyed. The intrusiveness of the media could also be viewed as a negative filter. Moreover, the media could be more aggressive in doing this than they used to be. For example, President John F. Kennedy’s personal transgressions were famously overlooked, in contrast to President Bill Clinton’s more recent experience.
56 Journal of Economic Perspectives
incentives emerges in this context. If politicians are better behaved in office, then there is less chance of filtering out the bad ones. Whether this tradeoff is advan- tageous for voters depends on the longer-run consequences of keeping bad poli- ticians in office. If there is costly end-game in which bad politicians are term limited or decide to retire, the voters will eventually pay. But otherwise, it may be better for voters that bad politicians recognize they must be on their best behavior to retain their position. The importance of such factors is borne out in evidence from political life cycles that exploit the fact that some politicians periodically face term limits (for example, Besley and Case, 1995). Democracies offer fairly limited means for entrenchment by incumbents.^7 However, the possibility of re-election does allow bad politicians to strike bargains with special interests in exchange for campaign contributions—a pattern that will tend to favor bad politicians and hence raise the accountability ratio. However, the range of devices for entrenchment in autocratic settings is larger than in electoral systems. Indeed, autocratic politicians may emasculate the electoral process by limiting press freedom and intimidating opposition politicians. For example, Daniel Arap Moi of Kenya created a de jure one party state in 1982. In similar vein, Zimbabwean leader Robert Mugabe has all but eliminated the possibility of his removal from office by electoral means.
Concluding Comments
Economic models of markets typically bring together individual preferences and the incentives that individuals face. However, modern political economy has tended to focus only on the incentives faced by politicians for good or bad behavior, while neglecting the importance of selection. But no society can run effective public institutions while ignoring the quality of who is recruited to public office and what they stand for. By reclaiming the selection agenda for political economy, we find not only a tool for improving government, but a rich agenda for economic research. Theoretical models of political competition are at last being developed to recognize the importance of selection, and empirical evidence is bearing out the relevance of these ideas in practice. A huge amount remains to be done in assembling more data on the qualities of the political class and then relating these qualities to the institutions that shape political selection. The focus here has been on politicians. However, issues of selection run into all aspects of public service. Similar issues arise, for example, in understanding the performance of public bureaucracies. Here, too, attention must be paid to selection of those who are competent and motivated toward public service as well as to traditional concerns about incentive design. For example, the National
(^7) However, there is the possibility that a low-quality political class can affect future outcomes. For example, Caselli and Morelli (2004) suggest that, by influencing future rewards to political office, this can lead to path dependence in candidate quality.
58 Journal of Economic Perspectives
Commission on the Public Service (2003) discussed how to rekindle the values of public service in America. Commission chair Paul Volcker remarks in the preface (p. 4): “Too few of our talented citizens are seeking careers in government or accepting political or judicial appointments.” One interesting question that seems poorly understood involves when elections rather than some other form of selec- tion mechanism is able to select a better cadre of public servants. Selection may also be important in the conduct of business. A key issue is whether socially responsible business decisions can be achieved without socially responsible managers. A recent op-ed piece by Robert Shiller (2005) observes that “the view of the world that one gets in a modern business curriculum can lead to an ethical disconnect. The courses often encourage a view of human nature that does not inspire high-mindedness.” Finally, economists and other social scientists should pay more attention to the origins of civic virtue and the possibility of increasing the number of trustworthy public servants. While economists often resist the idea that values can or should be taught, their importance is manifest in the operation of public and private institutions—from Watergate to Enron.
y I am grateful to Riccardo Puglisi for excellent research assistance. Andrei Shleifer provided
constructive comments and encouragement throughout the process. I am also grateful to Francesco Caselli, Maitreesh Ghatak, James Hines, Rohini Pande, Ken Shepsle, Jim Snyder, Timothy Taylor and Michael Waldman for very helpful comments and advice on an earlier version.
References
Akerlof, George A. and Rachel E. Kranton.
- “Identity and the Economics of Organiza- tions.” Journal of Economic Perspectives. Winter, 19:1, pp. 9 –32. Alesina, Alberto. 1988. “Credibility and Policy Convergence in a Two-Party System with Ratio- nal Voters.” American Economic Review. 78:4, pp. 796 – 806. American Bar Association Journal. 1977. 63, 171. Barro, Robert. 1973. “The Control of Politi- cians: An Economic Model.” Public Choice. Spring, 14, pp. 19 – 42. Bertrand, Marianne and Antoinette Schoar.
- “Managing with Style: The Effect of Man- agers on Firm Policies.” Quarterly Journal of Eco- nomic. 118:4, pp. 1169 –207. Besley, Timothy and Anne Case. 1995. “Does
Political Accountability Affect Economic Policy Choices? Evidence From Gubernatorial Term Limits.” Quarterly Journal of Economics. 110:3, pp. 769 –98. Besley, Timothy and Stephen Coate. 1997. “An Economic Model of Representative Democ- racy.” Quarterly Journal of Economics. 112:1, pp. 85–114. Besley, Timothy and Maitreesh Ghatak. 2005. “Competition and Incentives with Motivated Agents.” American Economic Review. 95:3, pp. 616 –36. Besley, Timothy and Andrea Prat. 2004. “Handcuffs for the Grabbing Hand? Media Capture and Government Accountability.” Typescript.
Political Selection 59
