In “Do Leaders Matter? National Leadership and Growth since World War II,” the authors make the correct point that much of the literature attempting to explain long-term trends in growth do not take into account the role of national leadership.
However, there are several potential weaknesses to the authors’ argument. First, it is not entirely clear that changes in levels of national growth are direct determinants of the actual deaths of the leaders of the ability of specific types of institutional arrangements to deal with these changes. As the authors note, the death of autocrats tends to have a more significant impact on economic growth than the deaths of democratic leaders. Second, some of the variables that the authors use to proxy for changes in monetary and security policy may be mis-specified. For example, changes in overall levels of inflation are not totally reflective of a regime’s intentional economic policy.
Observing changes in inflation is indicative of changes in the outcome and not necessarily of the policies that caused the outcomes. Lastly, the variable for security policy may be mis-specified as well. Whether or not a country is in a state of war is not totally reflective of security policy. Variables better suited to measuring changes in states’ security policies may be levels of military spending or the number of international security treaties to which the state is a party.
While the results of this analysis are quite intriguing, it is not entirely clear that the variables that the authors have selected match with the theoretical logic that they have proposed. As a result, the significance of the results are compromised.
Keefer and Knack are tackling an important issue: the link between income distribution, institutions and economic outcomes. In their account, property rights are endogenous to the level of “social polarization”: more polarization in the socioeconomic structure increases the probability of policy instability. In turn, this uncertainty makes capital flight a dominant strategy for investors: as a result, growth suffers.
Let me point out two problems with this paper. The first problem is one of “concept stretching” and measurement: Keefer and Knack interchangeably refer to inequality and polarization as the same thing. However, at the conceptual level, these two are in fact different dimensions of the income distribution of a society. While inequality refers to differences in income levels across the population, polarisation is related to groups that are internally homogeneous and that antagonize each other. The difference can be illustrated by considering the following situation: Suppose a country with six persons labelled as A, B, C, D, E, F with incomes equal to $ 1, 2, 3, 4, 5 and 6, respectively. Consider now two transfers of one dollar: the first one from C to A, and the second one from F to D. The two transfers are equalizing (from richer to poorer persons), so all inequality indices will fall, or at least not increase. The inequality analysis assesses the new situation as “better” than the initial one. In particular, Keefer and Knack would expect that the new distribution leads to a more stable society. However, note that in this example the new income distribution has three persons with $ 2 (A, B and C), and three persons with $5 (D, E and F). The population in this country is divided into two clearly differentiated groups that are internally perfectly homogeneous. Although less unequal, this society has become more polarized! One should not overemphasize this example too much though: it is likely that in most cases polarisation and inequality go in the same direction. However, given that there exist readily available measures of economic polarization (Esteben, Ray, Duclos 2004), why not use these instead of the blunter inequality indices?
The second problem is more substantial and theoretical. The authors simply assume that socioeconomic polarization translates into political polarization to affect economic outcomes. But this link, rather than assumed away, needs to be empirically justified. Political polarization can be measured in a number of ways. Some authors have proxied it by the percentage of seats that one party holds in the assembly when the opposite party is in hold of the cabinet (Frye 2002). It is also possible to consider the objective characteristics of the institutional environment that would make the policy enviorment more or lees secure from policy switches or instability. For example, institutions that create “winner-take all” politics are arguably more likely to generate the lack of government credibility that the authors are worried about. Some examples include majoritarian electoral formulas, single seat districts, presidentialism and concurrent elections. Similarly, the level of judicial independence is a key variable to consider as well. All these political variables shape the party systems in place and the ways polities are organized, and it at this political level (not society) that policy decisions are taken that either induce or prevent secure property rights from being established.
In sum, an interesting line of further research is to explore in depth the connection between socioeconomic and political polarization, studying how both variables shape together the quality of institutions, the security of property rights and affect economic outcomes.
I was pleased to read this week Coases seminal and widely cited work on the “Problems of Social Cost” (Coase 1960). Having now read the article however I can only concur with Glaeser, Johnson and Shleifer (2001) that its findings have been broadly misapplied. Coases theorem that emerges from this article states that in conditions where property rights are well defined and transaction costs are zero, market participants will maximize production value in the absence of government intervention. This theorem has been employed by laissez faire advocates to oppose regulation of financial (and other) markets. Yet in so doing they overlook Coases explicit recognition of the role of both the firm and government in overcoming coordination dilemmas in most conceivable situations – i.e. where property rights are not precisely defined and transactions involve costs (Coase 1960, 16-17). They also disregard what I view is a far more important contribution of this article - namely Coase’s call for an alternate approach to regulation. Coase advocates for methods that maximize net product yielded rather than narrowly focus on assigning private versus social costs and benefits (41-44). The merit of this argument however is hindered by Coase’s singular focus on the value of production as a measure for maximization. Whilst he recognizes the ‘desirability’ of broadening the analysis so that “the total effect of these arrangements in all spheres of life should be taken into account” (43) he fails to convincingly demonstrate how attributes such as ‘aesthetics’ and ‘morals’ should be appropriately quantified and incorporated into this rational framework.
David Stasavage (2002) adds valuable conditions to North and Weingast's (1989) finding that increased numbers of veto players raise an institution's ability to make credible commitments. To what extent can we apply this framework to informal politics, or to interactions between non-state advocacy organizations or coalitions with public or other non-state actors?
On the face of it, it seems that the informal sector yields an unlimited number of veto players: political action by Muslim minorities in Europe, for example, seem always to be threatened by potential attacks or moves by extremists within (or perceived to be within) their own communities. By coalitioning themselves in wide but centrist-oriented groups, they might be seen as accomplishing two goals, in relation to the above frameworks: they incoporate a broad number of players from the community to give credibility to their commitements. On the other hand, they disassociate themselves, in public perception from extremists who may no longer act as veto players, but rather as outside spoilers to the process.
This is one preliminary application of these studies to non-state processes. Would such an interpretation and use be reasonable, or is it assuming too much similarity between public and private actors?
I really enjoyed reading David Stasavage's piece "Credible Commitment in Early Modern Europe", particularly as a complement to North as Weingast's piece. They enrich North and weingast's analyses by arguing that the institutions created in Europe to allow for credible commitment in the part of the government with regards to its debts were important, yet not sufficient. Equally or more important was the political composition of these institutions and how they reflected land and capital interests. While I found their arguments very compelling, a few points deserved better clarification in my opinion. First, a big part of their claim is that the Whigs, despite being landowners, consistently supported the 'monied interests'. The authors claim that this can be explained by 'cross-issue' coalitions/trading. While this is entirely plausible, evidence that monied interests reciprocated the support for the whigs in pushing through policy in other areas (foreign policy, religious affairs) is entirely missing and would have strengthened their case. Second, they quickly dismiss the role of what they call the 'dubious' South Sea company. As interest rates grew under the tories (due to the lack of a credible commitment), the South Sea Company was launched to help provide funding that was made difficult through the bank of england. The authors dismiss the role the South Sea Company (which was to provide funds through tariffs charged on products coming from the americas) could have played as a guarantee on loans by providing a source of income. The last point that I believe could have been better clarified also relates to the whigs voting against their own landed interest. The authors state that the whig majority in the house of commons voted between 1708 and 1710 to increase the rate of taxation of land INCOME, not land per se. While they make clear that the majority of whigs were land owners, it is unclear how much of their income came from the land. Had they been moving capital to the financial interests in london and away from that of the land perhaps the impact of the tax would not be as dramatic as is claimed (and thus the importance of the coalition strategy attenuated)
Glaeser et al. 2001 provide a thorough and thought provoking discussion of how one might encourage particularly desirable normative goals of justice and economically efficient outcomes through institutional manipulation. While their analysis sheds some theoretical light onto why regulators might be preferable to judges in particular cases, the analysis does not shed light on a potential problem with regulators in transitional environments- credible commitment. Although they discuss how changes in the value of the a parameter (incentives to turn in guilty verdicts), it is unclear how or why such payoffs may be altered because such circumstances are completely left out of the model. This in turn implies, erroneously, that if regulators are initially given the proper institutional incentives (proper levels of a and b), they provide more normatively desired outcomes than judges. As Stasavage indicates, however, this is all predicated on the broader political incentives of office holders, whose policy preferences and veto point configuration can easily shift in such a way as to change policy. I believe that Stasavage’s point is particularly apt given that inexperience with markets and a history of state intervention require governments in transitional economies, such as those examined by Glaeser et al., to make credible commitments to encourage investors to invest. Although the efficiency of the judicial system (e.g. avoiding the leniency or abuse equilibria) is important, so to is the ability of investors to believe such equilibria are self-enforcing.
The cases that Glaeser et al. bring up to provide empirical support provide excellent intuition as to what environmental caveats should be made to their theory, however. In the two main cases, there was a large amount of investment regardless of how regulation was structured, although the Czech Republic’s weaker regulation caused it to grow more slowly than Poland. In both cases, initial investment took off because the total discrediting of pro-state intervention forces at the foundational moments of the market meant that business leaders feared no appropriation because none of the major parties could support such intervention after the demise of Communism. As such, the markets- and regulations meant to protect investors- were credible even if the government prioritized efficiency and used politically vulnerable regulators rather than judges.. One can imagine counter factual cases, however, such as the Latin American transitions, where investors invest less in transitional moments because populist forces on the left have not been discredited. In such cases, shifts in the political configuration can credibly lead to potentially expropriatory measures. In such settings, independent judges might be preferable, even if they are on balance less efficient than regulators, because they are insulated from changes in political power. Thus, we see that in transitional economies, the choice over judges or regulators not only depends on the incentives offered in the system, but is also predicated on the potential changes to that system.
Weingat develops two game-theory models to demonstrate that self-enforcing limits on the government result when society groups resolve their coordination problems about the appropriate limits on the government. He also suggests that it is elites that construct solutions to the coordination problem of citizens, often through pacts. I think Weingast’s arguments could be improved if he takes the following problems into account.
First, his models assume that there are only two social groups. What if there are more than two social groups? In this case, the configuration of social groups may make difference too. As Keeper and Knack argue that, if the social groups are polarized, the coordination problem among them to form consensus regarding limits on the government become more difficult. In addition, when there are more than two social groups interact in the models, free-riding problems among those groups might arise besides the increasing difficulties of coordination.
Second, his models seem to work in an institutional vacuum: there is no specified institutional context in which the social groups, elites interact with each other and make choices. Stasavage’s article suggests that some institutional arrangements such as partisan structure, cross-issue coalitions and government delegation can affect the credibility of the limited government. Likewise, the coordination problem among citizenry groups to limit the government power may also be affected by existing institutions. Institutional context such as whether citizens have the freedoms of associations, whether the society has the freedom of press, whether there exist representative political parties and organizations to articulate and coordinate individual citizens’ preferences etc. might change the consensus formation within society regarding the limited government.
Third, in his models social groups are identical: they don’t differentiate much in resource, power and status. Thus in his models, if I understand correctly, the sovereign has the same probabilities to transgress against either group, either groups, or neither. I don’t think it is the case in reality. For example, the English monarchs were more likely to transgress against peasants rather than capitalists in the seventeenth-century when they badly needed funds to finance wars. If the sovereign’s probabilities of transgression against some social groups but not others are not equal, then the strategies of the social groups whose interests are expected to be damaged by the sovereign and the ones whose interests are to be enhanced might be different from what the models predict.
The innovation of Coase's theorem is often understood to be that private parties will negotiate a socially optimal solution to externalities, given clearly assigned property rights and assuming costless transactions. Since Coase himself acknowledges this assumption of costless transactions (and implores economists to explore it further), it would be a red herring to fault him for that simplifying assumption. He also acknowledges the importance of the initial allocation of property rights. This week's readings address these assumptions by exploring the initial allocation of rights in various institutional contexts.
Keefer and Knack show that social and economic polarization reduce the security of property rights by making the policy area more volatile. Because their goal is to map the relationship between polarization and growth, they do not explore the way that political institutions might affect the degree of polarization or might temper its effect on the security of property rights. In fact, they unconvincingly argue that social polarization is reflected as political polarization. In contrast, North and Weingast explore political dynamics by demonstrating the way that institutional "checks-and-balances" secured property rights in England in the 17th century. Stasavage extends the North and Weingast analysis by showing how institutional veto points and political divisions produced different degrees of property rights in England and France. Furthermore, Weingast's rule of law game leads him to conclude that the Pareto optimal equilibrium (where the government does not infringe on the rights of its citizens, and the citizens do not rebel against the government) is only feasible if there is a self-enforcing coordination device that prevents actors from changing their strategy. This coordination device would come in the form of an institutional framework that prompts otherwise polarized groups to resist transgressions by a sovereign. From these three articles, we develop a picture of the allocation of property rights that is a function of the rational behavior of various interests within an institutional context.
The obvious conclusion to draw from these readings is that initial allocation of property rights is subject to coordination constraints on various levels--a type of transaction cost that can be prohibitively high. In a sense, this problem reduces the Coase theorem to a somewhat uninteresting description of how rational actors engage in economic exchange in a stylized world. Yet, it seems to me that it would be unwise to dismiss Coase on the grounds that his theorem is somewhat self-evident. By bringing us "back to basics," he poses a very powerful challenge to the dominant thinking on social cost, which itself essentially ignores the costs of transactions (e.g., the deadweight loss of taxes) while unknowingly allocating property rights by arbitrary criteria. This week's readings make a valiant effort to unpack the assumptions underlying the Coase theorem.
In "Constitutions and Commitment: The Evolution of Institutional Governing Public Choice in Seventeenth-Century England" North and Weingast (1989) present a fascinating and somewhat counterintuitive argument that higher economic growth can be achieved when sovereigns are bound by certain institutions and rules. Such sovereigns can make more credible commitments to respecting property rights, which in turn reduces risk and interest rates thereby increasing investment.
North and Weingast claim that it is not enough that sovereigns be bound by rules; they need to be bound by self-enforcing rules. That is, parties must have an incentive to abide by the rules of the constitution. Fundamentally, North and Weingast never convincingly explain what it is that makes these rules self-enforcing. Was the English sovereign's commitment self-enforcing because Parliament was given more power, because of the credible threat of revolution, because Parliament had an interest in keeping the government on sounds financial footing, because England did not have a standing army which could be used by the sovereign for repression, or some combination of these factors? There appears to be a fair amount of contingency in what makes these rules self-enforcing. Indeed, military coups and expropriation have been witnesses in states not all that different from England. As Weingast notes elsewhere, there must also exist some consensus in society regarding the legitimate role of the state in order to overcome the coordination and the collective action problem (Weingast, 1997).
Despite its drawbacks, North and Weingast's argument could be extended beyond comparative politics. For example, in international relations, Ikenberry (2000) argues that the US was able to leverage more influence and greater cooperation from the international community following the Second World War by binding itself to a number of international institutions such as NATO, the IMF, the World Bank, and the UN. States were more likely to cooperate with the US (or at a bare minimum not counterbalance), because they had a say in these institutions, which dampened the fear that the US would abuse its position of strength.
References:
John G. Ikenberry, After Victory (Princeton, NJ: Princeton University Press, 2000).
Douglass North and Barry Weingast, "Constitutions and Commitment: The Evolution of Institutional Governing Public Choice in Seventeenth-Century England," Journal of Economic History 49, no. 4 (December 1989): 803-832.
Barry Weingast, "The Political Foundations of Democracy and the Rule of Law," American Political Science Review 91, no. 2 (June 1997): 245-263
Keefer and Knack propose that 1) social polarization decreases the security of property and contract rights, and 2) social polarization decreases economic growth, primarily through the reduced security of property rights. I appreciate how this article draws out and adds to a full causal story about cross-country variation in economic growth: others have shown relationships between inequality and growth and between property rights and growth, but Keefer and Knack make a strong argument for the relationship between inequality and economic growth through property rights. I did have a few concerns about the measurement of social polarization, however. Keefer and Knack use ethnic tensions and income and land inequality as proxies for social polarization, due to data constraints. I was surprised that they only use two measures of ethnic divisions: a variable ranking "ethnic tensions" and one which measures the percent of the population belonging to the largest ethnic group. The "ethnic tensions" variable comes from a guide marketed to overseas investors, which makes me wonder if this measure is correlated with the measure of property rights (which also comes from the same guide). The second variable defines ethnicity by "language, race, and/or religion, depending on which of these is viewed by Sullivan as the most relevant source of divisions within the society" (Keefer and Knack 2002, 134). Ethnicity is undoubtedly a difficult concept to define, but this definition simplifies the fact that often several or all of these characteristics (language, race, religion) contribute to societal divisions. I would find this less troubling if there was a variable that explicitly measured ethnic divisions as they pertain to policy preferences.
According to Keefer and Knack, social polarization has negative effects on economic growth because of the uncertain security of property rights. Polarization is measured by ethnic tension, ethnic fragmentation, income inequality, and land inequality. I am not sure how social polarization leads to ill protection of the security of property rights. The above variables might each have a different causal story to tell. The authors’ (unconvincing) explanation is that polarization will prevent the incumbent to invest in legal infrastructure that will benefit the enemy, which does not make much sense to me. It assumes that politics is a zero-sum game, in which one’s loss is the other’s gain. This might be true regarding electoral politics, but not necessarily in legal arena. Bureaucratic delegation to an independent agent, such as the central bank or court, is an example. But before we even enter the polarized politics, how do we first transform social polarization into political polarization? Not all political systems represent socioeconomic conditions of their societies. I would like to add a little note regarding the variable of land inequality. This variable is representative only in countries where lands are mostly marketized. For example, the Chinese peasants are entitled to land distribution but they are not allowed to switch the usage of the land (e.g. from agricultural to industrial or commercial. Taiwan also has strict rules regarding transferring farm land usage.). They are “asset-holders” but still socially and economically subordinate; whereas the urban citizens do not have land ownership but are overall better-off. I don’t know how the statistics work, but in cases like this, the correlation might appear odd.
In Coase’s analysis two issues in particular continue to resonate. First, while both are categorized as transaction costs, should we conceive of a qualitative difference between the challenge posed by the incremental increase of participants to an economic transaction and that inherent in public goods problems? It seems that socially and economically efficient outcomes might still be attained by the internalization of externalities even with transaction costs if they do not exceed the benefits from internalizing externalities and as long as all participants – irrespective of their number - contribute to attaining that outcome. As soon as free riding becomes an incentive though, a qualitatively different challenge is faced by society to achieve efficient outcomes given that as opposed to incentives for active participation, the incentive now arises for everyone not to contribute at all.
Second, while the internalization of externalities through side payments may result in a most efficient outcome in which the production of one good ceases completely, to what extent is it in fact still a socially optimal outcome when no alternatives exist, such as for a train that may stop operations on economic efficiency grounds?
In “Constitutions and Commitment: The Evolution of Institutions Governing Public Choice in 17th Century England,” Douglass North and Barry Weingast stress how institutional change in England after the Glorious Revolution allowed the government to “credibly commit” to uphold property rights. In a critique of North and Weingast, David Stasavage notes the role that partisan interests can play in whether the government can or cannot “credibly commit” in light of institutional change. Stasavage’s critique raises a question for debate: how deterministic is North and Weingast’s theory? Are there other important variables other than partisan interests (which Stasavage notes) that may affect how institutions of limited government may credibly commit? North and Weingast and Stasavage use European case studies as the focal point of their analysis but would the story play out differently in other parts of the world? For instance, as Weingast notes in his article “The Political Foundations of Democracy and the Rule of Law,” compared to the United States, the citizenry in Latin America often supports extraconstitutional means. How then can be reconsider North and Weingast’s argument in light of the broader societal context and the way in which society interacts with the limited government? How might the factors that affect “credible commitment” vary in time and place? How can we read Weingast’s piece in light of the North and Weingast’s piece? Further, what exactly does it mean to make a “credible commitment?” Does it mean that states should affirmatively take action to foster property rights? Does it mean that the state should not take arbitrary action to take away rights? Does it mean that the state should just not interfere with property rights in any way? Moreover, how do different definitions of governmental action with respect to property rights interact with economic growth? The Glaeser et at piece reflects one answer, but how widely applicable is that argument to all times and places?
Keefer and Knach argue that income inequality—a form of polarization—affects the sovereign’s credit worthiness, by feeding into an uncertain environment, causing delays and large shifts in the decision-making process. In particular, polarization may lead to shifts in property rights policies, affecting contract enforcement and security, thereby affecting growth, as firms become less willing to invest in scenarios with uncertain property rights. When considering not only relations between firms, but also relations between firms and the sovereign, this line of thought is very interesting. Under high degrees of income inequality, shifts in political leadership have led to drastic changes in policies, such that private goods—usually related to the production or exploration of natural resources—have been confiscated by the sovereign, or even times when the sovereign has been unwilling to pay its debt due to political pressures. However, I thought that there was one aspect that Keefer and Knach could have elaborated more extensively. It is plausible to imagine the scenario where, under high income inequality, the economic elite is concomitantly, the political elite. If they have also have a monopoly on political power, it becomes increasingly difficult for the marginalized poor to voice their preferences, mobilize and eventually cause policy shifts. As such, Keefer and Knach’s analysis would benefit from further insight of the effects of income inequality on property rights’ related policy by incorporating electoral mechanisms and political decision-making institutions. It could be the case that under parliamentary systems, even with high income inequality we observe less uncertainty and less of an effect of polarization on property rights, while under presidential systems the effect may be the opposite, as the division of power is more concentrated on the executive.
Leanne Tyler
ReplyDeleteComparing Institutions—Week Ten—Comment
In “Do Leaders Matter? National Leadership and Growth since World War II,” the authors make the correct point that much of the literature attempting to explain long-term trends in growth do not take into account the role of national leadership.
However, there are several potential weaknesses to the authors’ argument. First, it is not entirely clear that changes in levels of national growth are direct determinants of the actual deaths of the leaders of the ability of specific types of institutional arrangements to deal with these changes. As the authors note, the death of autocrats tends to have a more significant impact on economic growth than the deaths of democratic leaders.
Second, some of the variables that the authors use to proxy for changes in monetary and security policy may be mis-specified. For example, changes in overall levels of inflation are not totally reflective of a regime’s intentional economic policy.
Observing changes in inflation is indicative of changes in the outcome and not necessarily of the policies that caused the outcomes. Lastly, the variable for security policy may be mis-specified as well. Whether or not a country is in a state of war is not totally reflective of security policy. Variables better suited to measuring changes in states’ security policies may be levels of military spending or the number of international security treaties to which the state is a party.
While the results of this analysis are quite intriguing, it is not entirely clear that the variables that the authors have selected match with the theoretical logic that they have proposed. As a result, the significance of the results are compromised.
Keefer and Knack are tackling an important issue: the link between income distribution, institutions and economic outcomes. In their account, property rights are endogenous to the level of “social polarization”: more polarization in the socioeconomic structure increases the probability of policy instability. In turn, this uncertainty makes capital flight a dominant strategy for investors: as a result, growth suffers.
ReplyDeleteLet me point out two problems with this paper. The first problem is one of “concept stretching” and measurement: Keefer and Knack interchangeably refer to inequality and polarization as the same thing. However, at the conceptual level, these two are in fact different dimensions of the income distribution of a society. While inequality refers to differences in income levels across the population, polarisation is related to groups that are internally homogeneous and that antagonize each other. The difference can be illustrated by considering the following situation: Suppose a country with six persons labelled as A, B, C, D, E, F with incomes equal to $ 1, 2, 3, 4, 5 and 6, respectively. Consider now two transfers of one dollar: the first one from C to A, and the second one from F to D. The two transfers are equalizing (from richer to poorer persons), so all inequality indices will fall, or at least not increase. The inequality analysis assesses the new situation as “better” than the initial one. In particular, Keefer and Knack would expect that the new distribution leads to a more stable society. However, note that in this example the new income distribution has three persons with $ 2 (A, B and C), and three persons with $5 (D, E and F). The population in this country is divided into two clearly differentiated groups that are internally perfectly homogeneous. Although less unequal, this society has become more polarized! One should not overemphasize this example too much though: it is likely that in most cases polarisation and inequality go in the same direction. However, given that there exist readily available measures of economic polarization (Esteben, Ray, Duclos 2004), why not use these instead of the blunter inequality indices?
The second problem is more substantial and theoretical. The authors simply assume that socioeconomic polarization translates into political polarization to affect economic outcomes. But this link, rather than assumed away, needs to be empirically justified. Political polarization can be measured in a number of ways. Some authors have proxied it by the percentage of seats that one party holds in the assembly when the opposite party is in hold of the cabinet (Frye 2002). It is also possible to consider the objective characteristics of the institutional environment that would make the policy enviorment more or lees secure from policy switches or instability. For example, institutions that create “winner-take all” politics are arguably more likely to generate the lack of government credibility that the authors are worried about. Some examples include majoritarian electoral formulas, single seat districts, presidentialism and concurrent elections. Similarly, the level of judicial independence is a key variable to consider as well. All these political variables shape the party systems in place and the ways polities are organized, and it at this political level (not society) that policy decisions are taken that either induce or prevent secure property rights from being established.
In sum, an interesting line of further research is to explore in depth the connection between socioeconomic and political polarization, studying how both variables shape together the quality of institutions, the security of property rights and affect economic outcomes.
I was pleased to read this week Coases seminal and widely cited work on the “Problems of Social Cost” (Coase 1960). Having now read the article however I can only concur with Glaeser, Johnson and Shleifer (2001) that its findings have been broadly misapplied. Coases theorem that emerges from this article states that in conditions where property rights are well defined and transaction costs are zero, market participants will maximize production value in the absence of government intervention. This theorem has been employed by laissez faire advocates to oppose regulation of financial (and other) markets. Yet in so doing they overlook Coases explicit recognition of the role of both the firm and government in overcoming coordination dilemmas in most conceivable situations – i.e. where property rights are not precisely defined and transactions involve costs (Coase 1960, 16-17). They also disregard what I view is a far more important contribution of this article - namely Coase’s call for an alternate approach to regulation. Coase advocates for methods that maximize net product yielded rather than narrowly focus on assigning private versus social costs and benefits (41-44). The merit of this argument however is hindered by Coase’s singular focus on the value of production as a measure for maximization. Whilst he recognizes the ‘desirability’ of broadening the analysis so that “the total effect of these arrangements in all spheres of life should be taken into account” (43) he fails to convincingly demonstrate how attributes such as ‘aesthetics’ and ‘morals’ should be appropriately quantified and incorporated into this rational framework.
ReplyDeleteDavid Stasavage (2002) adds valuable conditions to North and Weingast's (1989) finding that increased numbers of veto players raise an institution's ability to make credible commitments. To what extent can we apply this framework to informal politics, or to interactions between non-state advocacy organizations or coalitions with public or other non-state actors?
ReplyDeleteOn the face of it, it seems that the informal sector yields an unlimited number of veto players: political action by Muslim minorities in Europe, for example, seem always to be threatened by potential attacks or moves by extremists within (or perceived to be within) their own communities. By coalitioning themselves in wide but centrist-oriented groups, they might be seen as accomplishing two goals, in relation to the above frameworks: they incoporate a broad number of players from the community to give credibility to their commitements. On the other hand, they disassociate themselves, in public perception from extremists who may no longer act as veto players, but rather as outside spoilers to the process.
This is one preliminary application of these studies to non-state processes. Would such an interpretation and use be reasonable, or is it assuming too much similarity between public and private actors?
I really enjoyed reading David Stasavage's piece "Credible Commitment in Early Modern Europe", particularly as a complement to North as Weingast's piece. They enrich North and weingast's analyses by arguing that the institutions created in Europe to allow for credible commitment in the part of the government with regards to its debts were important, yet not sufficient. Equally or more important was the political composition of these institutions and how they reflected land and capital interests. While I found their arguments very compelling, a few points deserved better clarification in my opinion. First, a big part of their claim is that the Whigs, despite being landowners, consistently supported the 'monied interests'. The authors claim that this can be explained by 'cross-issue' coalitions/trading. While this is entirely plausible, evidence that monied interests reciprocated the support for the whigs in pushing through policy in other areas (foreign policy, religious affairs) is entirely missing and would have strengthened their case.
ReplyDeleteSecond, they quickly dismiss the role of what they call the 'dubious' South Sea company. As interest rates grew under the tories (due to the lack of a credible commitment), the South Sea Company was launched to help provide funding that was made difficult through the bank of england. The authors dismiss the role the South Sea Company (which was to provide funds through tariffs charged on products coming from the americas) could have played as a guarantee on loans by providing a source of income. The last point that I believe could have been better clarified also relates to the whigs voting against their own landed interest. The authors state that the whig majority in the house of commons voted between 1708 and 1710 to increase the rate of taxation of land INCOME, not land per se. While they make clear that the majority of whigs were land owners, it is unclear how much of their income came from the land. Had they been moving capital to the financial interests in london and away from that of the land perhaps the impact of the tax would not be as dramatic as is claimed (and thus the importance of the coalition strategy attenuated)
Glaeser et al. 2001 provide a thorough and thought provoking discussion of how one might encourage particularly desirable normative goals of justice and economically efficient outcomes through institutional manipulation. While their analysis sheds some theoretical light onto why regulators might be preferable to judges in particular cases, the analysis does not shed light on a potential problem with regulators in transitional environments- credible commitment. Although they discuss how changes in the value of the a parameter (incentives to turn in guilty verdicts), it is unclear how or why such payoffs may be altered because such circumstances are completely left out of the model. This in turn implies, erroneously, that if regulators are initially given the proper institutional incentives (proper levels of a and b), they provide more normatively desired outcomes than judges. As Stasavage indicates, however, this is all predicated on the broader political incentives of office holders, whose policy preferences and veto point configuration can easily shift in such a way as to change policy. I believe that Stasavage’s point is particularly apt given that inexperience with markets and a history of state intervention require governments in transitional economies, such as those examined by Glaeser et al., to make credible commitments to encourage investors to invest. Although the efficiency of the judicial system (e.g. avoiding the leniency or abuse equilibria) is important, so to is the ability of investors to believe such equilibria are self-enforcing.
ReplyDeleteThe cases that Glaeser et al. bring up to provide empirical support provide excellent intuition as to what environmental caveats should be made to their theory, however. In the two main cases, there was a large amount of investment regardless of how regulation was structured, although the Czech Republic’s weaker regulation caused it to grow more slowly than Poland. In both cases, initial investment took off because the total discrediting of pro-state intervention forces at the foundational moments of the market meant that business leaders feared no appropriation because none of the major parties could support such intervention after the demise of Communism. As such, the markets- and regulations meant to protect investors- were credible even if the government prioritized efficiency and used politically vulnerable regulators rather than judges.. One can imagine counter factual cases, however, such as the Latin American transitions, where investors invest less in transitional moments because populist forces on the left have not been discredited. In such cases, shifts in the political configuration can credibly lead to potentially expropriatory measures. In such settings, independent judges might be preferable, even if they are on balance less efficient than regulators, because they are insulated from changes in political power. Thus, we see that in transitional economies, the choice over judges or regulators not only depends on the incentives offered in the system, but is also predicated on the potential changes to that system.
Weingat develops two game-theory models to demonstrate that self-enforcing limits on the government result when society groups resolve their coordination problems about the appropriate limits on the government. He also suggests that it is elites that construct solutions to the coordination problem of citizens, often through pacts. I think Weingast’s arguments could be improved if he takes the following problems into account.
ReplyDeleteFirst, his models assume that there are only two social groups. What if there are more than two social groups? In this case, the configuration of social groups may make difference too. As Keeper and Knack argue that, if the social groups are polarized, the coordination problem among them to form consensus regarding limits on the government become more difficult. In addition, when there are more than two social groups interact in the models, free-riding problems among those groups might arise besides the increasing difficulties of coordination.
Second, his models seem to work in an institutional vacuum: there is no specified institutional context in which the social groups, elites interact with each other and make choices. Stasavage’s article suggests that some institutional arrangements such as partisan structure, cross-issue coalitions and government delegation can affect the credibility of the limited government. Likewise, the coordination problem among citizenry groups to limit the government power may also be affected by existing institutions. Institutional context such as whether citizens have the freedoms of associations, whether the society has the freedom of press, whether there exist representative political parties and organizations to articulate and coordinate individual citizens’ preferences etc. might change the consensus formation within society regarding the limited government.
Third, in his models social groups are identical: they don’t differentiate much in resource, power and status. Thus in his models, if I understand correctly, the sovereign has the same probabilities to transgress against either group, either groups, or neither. I don’t think it is the case in reality. For example, the English monarchs were more likely to transgress against peasants rather than capitalists in the seventeenth-century when they badly needed funds to finance wars. If the sovereign’s probabilities of transgression against some social groups but not others are not equal, then the strategies of the social groups whose interests are expected to be damaged by the sovereign and the ones whose interests are to be enhanced might be different from what the models predict.
The innovation of Coase's theorem is often understood to be that private parties will negotiate a socially optimal solution to externalities, given clearly assigned property rights and assuming costless transactions. Since Coase himself acknowledges this assumption of costless transactions (and implores economists to explore it further), it would be a red herring to fault him for that simplifying assumption. He also acknowledges the importance of the initial allocation of property rights. This week's readings address these assumptions by exploring the initial allocation of rights in various institutional contexts.
ReplyDeleteKeefer and Knack show that social and economic polarization reduce the security of property rights by making the policy area more volatile. Because their goal is to map the relationship between polarization and growth, they do not explore the way that political institutions might affect the degree of polarization or might temper its effect on the security of property rights. In fact, they unconvincingly argue that social polarization is reflected as political polarization. In contrast, North and Weingast explore political dynamics by demonstrating the way that institutional "checks-and-balances" secured property rights in England in the 17th century. Stasavage extends the North and Weingast analysis by showing how institutional veto points and political divisions produced different degrees of property rights in England and France. Furthermore, Weingast's rule of law game leads him to conclude that the Pareto optimal equilibrium (where the government does not infringe on the rights of its citizens, and the citizens do not rebel against the government) is only feasible if there is a self-enforcing coordination device that prevents actors from changing their strategy. This coordination device would come in the form of an institutional framework that prompts otherwise polarized groups to resist transgressions by a sovereign. From these three articles, we develop a picture of the allocation of property rights that is a function of the rational behavior of various interests within an institutional context.
The obvious conclusion to draw from these readings is that initial allocation of property rights is subject to coordination constraints on various levels--a type of transaction cost that can be prohibitively high. In a sense, this problem reduces the Coase theorem to a somewhat uninteresting description of how rational actors engage in economic exchange in a stylized world. Yet, it seems to me that it would be unwise to dismiss Coase on the grounds that his theorem is somewhat self-evident. By bringing us "back to basics," he poses a very powerful challenge to the dominant thinking on social cost, which itself essentially ignores the costs of transactions (e.g., the deadweight loss of taxes) while unknowingly allocating property rights by arbitrary criteria. This week's readings make a valiant effort to unpack the assumptions underlying the Coase theorem.
In "Constitutions and Commitment: The Evolution of Institutional Governing Public Choice in Seventeenth-Century England" North and Weingast (1989) present a fascinating and somewhat counterintuitive argument that higher economic growth can be achieved when sovereigns are bound by certain institutions and rules. Such sovereigns can make more credible commitments to respecting property rights, which in turn reduces risk and interest rates thereby increasing investment.
ReplyDeleteNorth and Weingast claim that it is not enough that sovereigns be bound by rules; they need to be bound by self-enforcing rules. That is, parties must have an incentive to abide by the rules of the constitution. Fundamentally, North and Weingast never convincingly explain what it is that makes these rules self-enforcing. Was the English sovereign's commitment self-enforcing because Parliament was given more power, because of the credible threat of revolution, because Parliament had an interest in keeping the government on sounds financial footing, because England did not have a standing army which could be used by the sovereign for repression, or some combination of these factors? There appears to be a fair amount of contingency in what makes these rules self-enforcing. Indeed, military coups and expropriation have been witnesses in states not all that different from England. As Weingast notes elsewhere, there must also exist some consensus in society regarding the legitimate role of the state in order to overcome the coordination and the collective action problem (Weingast, 1997).
Despite its drawbacks, North and Weingast's argument could be extended beyond comparative politics. For example, in international relations, Ikenberry (2000) argues that the US was able to leverage more influence and greater cooperation from the international community following the Second World War by binding itself to a number of international institutions such as NATO, the IMF, the World Bank, and the UN. States were more likely to cooperate with the US (or at a bare minimum not counterbalance), because they had a say in these institutions, which dampened the fear that the US would abuse its position of strength.
References:
John G. Ikenberry, After Victory (Princeton, NJ: Princeton University Press, 2000).
Douglass North and Barry Weingast, "Constitutions and Commitment: The Evolution of Institutional Governing Public Choice in Seventeenth-Century England," Journal of Economic History 49, no. 4 (December 1989): 803-832.
Barry Weingast, "The Political Foundations of Democracy and the Rule of Law," American Political Science Review 91, no. 2 (June 1997): 245-263
Keefer and Knack propose that 1) social polarization decreases the security of property and contract rights, and 2) social polarization decreases economic growth, primarily through the reduced security of property rights. I appreciate how this article draws out and adds to a full causal story about cross-country variation in economic growth: others have shown relationships between inequality and growth and between property rights and growth, but Keefer and Knack make a strong argument for the relationship between inequality and economic growth through property rights. I did have a few concerns about the measurement of social polarization, however. Keefer and Knack use ethnic tensions and income and land inequality as proxies for social polarization, due to data constraints. I was surprised that they only use two measures of ethnic divisions: a variable ranking "ethnic tensions" and one which measures the percent of the population belonging to the largest ethnic group. The "ethnic tensions" variable comes from a guide marketed to overseas investors, which makes me wonder if this measure is correlated with the measure of property rights (which also comes from the same guide). The second variable defines ethnicity by "language, race, and/or religion, depending on which of these is viewed by Sullivan as the most relevant source of divisions within the society" (Keefer and Knack 2002, 134). Ethnicity is undoubtedly a difficult concept to define, but this definition simplifies the fact that often several or all of these characteristics (language, race, religion) contribute to societal divisions. I would find this less troubling if there was a variable that explicitly measured ethnic divisions as they pertain to policy preferences.
ReplyDeleteAccording to Keefer and Knack, social polarization has negative effects on economic growth because of the uncertain security of property rights. Polarization is measured by ethnic tension, ethnic fragmentation, income inequality, and land inequality. I am not sure how social polarization leads to ill protection of the security of property rights. The above variables might each have a different causal story to tell. The authors’ (unconvincing) explanation is that polarization will prevent the incumbent to invest in legal infrastructure that will benefit the enemy, which does not make much sense to me. It assumes that politics is a zero-sum game, in which one’s loss is the other’s gain. This might be true regarding electoral politics, but not necessarily in legal arena. Bureaucratic delegation to an independent agent, such as the central bank or court, is an example. But before we even enter the polarized politics, how do we first transform social polarization into political polarization? Not all political systems represent socioeconomic conditions of their societies.
ReplyDeleteI would like to add a little note regarding the variable of land inequality. This variable is representative only in countries where lands are mostly marketized. For example, the Chinese peasants are entitled to land distribution but they are not allowed to switch the usage of the land (e.g. from agricultural to industrial or commercial. Taiwan also has strict rules regarding transferring farm land usage.). They are “asset-holders” but still socially and economically subordinate; whereas the urban citizens do not have land ownership but are overall better-off. I don’t know how the statistics work, but in cases like this, the correlation might appear odd.
In Coase’s analysis two issues in particular continue to resonate. First, while both are categorized as transaction costs, should we conceive of a qualitative difference between the challenge posed by the incremental increase of participants to an economic transaction and that inherent in public goods problems? It seems that socially and economically efficient outcomes might still be attained by the internalization of externalities even with transaction costs if they do not exceed the benefits from internalizing externalities and as long as all participants – irrespective of their number - contribute to attaining that outcome. As soon as free riding becomes an incentive though, a qualitatively different challenge is faced by society to achieve efficient outcomes given that as opposed to incentives for active participation, the incentive now arises for everyone not to contribute at all.
ReplyDeleteSecond, while the internalization of externalities through side payments may result in a most efficient outcome in which the production of one good ceases completely, to what extent is it in fact still a socially optimal outcome when no alternatives exist, such as for a train that may stop operations on economic efficiency grounds?
In “Constitutions and Commitment: The Evolution of Institutions Governing Public Choice in 17th Century England,” Douglass North and Barry Weingast stress how institutional change in England after the Glorious Revolution allowed the government to “credibly commit” to uphold property rights. In a critique of North and Weingast, David Stasavage notes the role that partisan interests can play in whether the government can or cannot “credibly commit” in light of institutional change. Stasavage’s critique raises a question for debate: how deterministic is North and Weingast’s theory? Are there other important variables other than partisan interests (which Stasavage notes) that may affect how institutions of limited government may credibly commit? North and Weingast and Stasavage use European case studies as the focal point of their analysis but would the story play out differently in other parts of the world? For instance, as Weingast notes in his article “The Political Foundations of Democracy and the Rule of Law,” compared to the United States, the citizenry in Latin America often supports extraconstitutional means. How then can be reconsider North and Weingast’s argument in light of the broader societal context and the way in which society interacts with the limited government? How might the factors that affect “credible commitment” vary in time and place? How can we read Weingast’s piece in light of the North and Weingast’s piece? Further, what exactly does it mean to make a “credible commitment?” Does it mean that states should affirmatively take action to foster property rights? Does it mean that the state should not take arbitrary action to take away rights? Does it mean that the state should just not interfere with property rights in any way? Moreover, how do different definitions of governmental action with respect to property rights interact with economic growth? The Glaeser et at piece reflects one answer, but how widely applicable is that argument to all times and places?
ReplyDeleteKeefer and Knach argue that income inequality—a form of polarization—affects the sovereign’s credit worthiness, by feeding into an uncertain environment, causing delays and large shifts in the decision-making process. In particular, polarization may lead to shifts in property rights policies, affecting contract enforcement and security, thereby affecting growth, as firms become less willing to invest in scenarios with uncertain property rights. When considering not only relations between firms, but also relations between firms and the sovereign, this line of thought is very interesting. Under high degrees of income inequality, shifts in political leadership have led to drastic changes in policies, such that private goods—usually related to the production or exploration of natural resources—have been confiscated by the sovereign, or even times when the sovereign has been unwilling to pay its debt due to political pressures. However, I thought that there was one aspect that Keefer and Knach could have elaborated more extensively. It is plausible to imagine the scenario where, under high income inequality, the economic elite is concomitantly, the political elite. If they have also have a monopoly on political power, it becomes increasingly difficult for the marginalized poor to voice their preferences, mobilize and eventually cause policy shifts. As such, Keefer and Knach’s analysis would benefit from further insight of the effects of income inequality on property rights’ related policy by incorporating electoral mechanisms and political decision-making institutions. It could be the case that under parliamentary systems, even with high income inequality we observe less uncertainty and less of an effect of polarization on property rights, while under presidential systems the effect may be the opposite, as the division of power is more concentrated on the executive.
ReplyDelete