Laura Sabani

Agency and Communication in IMF Conditional Lending: Theory and Empirical Evidence

We focus on the role that the transmission of information between a multilateral (the IMF) and a country has for the optimal design of conditional reforms. Our model predicts that when agency problems are especially severe, and/or IMF information is valuable, a centralized control is indeed optimal. To the contrary, when local knowledge is more important than the agency bias we expect delegation to dominate. Controlling for economic and political factors, our empirical tests show that the number of IMF conditions is lower in countries with a greater social complexity, while it increases with the bias of the countries’ authorities, openness, and transparency, consistently with the theory. ∗A preliminary verison of this paper was presented at the Workshop on Sovereign and Public Debt and Default (University of Warwick, 2008), ASSET meeting (EUI, 2008), Workshop on Political Economy (Silvaplana, 2008), Bank of Finland monthly research seminar (Helsinky, 2008), the Annual Meetings of the Political Economy of International Organizations (Ascona, 2008 and Geneva, 2009). We would like to thank Pascal Courty, Jakob de Haan, Francesco de Sinopoli, Federico Etro, Raquel Fernandez, Patrizio Tirelli, Julien Reynaud, Stephanie Rickard, Michele Santoni, Randall Stone, Tuomas Takalo, Otto Toivanen, Mark L.J. Wright for useful comments. Corresponding author: Silvia Marchesi, Dipartimento di Economia Politica, Piazza dell’Ateneo Nuovo 1, I-20126 Milano. Tel. +39 02 6448 3057; Fax. +39 02 6448 3085; E-mail: silvia.marchesi@unimib.it.

Prolonged Use and Conditionality Failure: Investigating IMF Responsibility

Prolonged use of Fund resources has consistently expanded since the 1970s among both lowincome and middle-income countries. Overall this phenomenon suggests a lack of effectiveness of Fund supported programs. Such conditionality failure has been explained by the literature by looking both at the characteristics of the borrowing countries and at the lack of credibility of the IMF threat of interrupting financial assistance in case of non compliance with conditionality. In this paper we suggest that such lack of credibility might be attributed to the dual role played by the IMF, which acts at the same time as a creditor and as a monitor (or as an advisor) of economic reforms. We show that the Fund desire to hide its surveillance failures, in order to preserve its reputation of being a good monitor/advisor, may actually distort its lending decisions towards greater laxity in punishing non-compliance with economic reforms. Such laxity may be exacerbated by the length of the relationship between a country and the Fund. Thus we claim that prolonged use of IMF resources is not only a consequence of a lack of effectiveness of adjustment lending but it might itself be a determinant of conditionality failure.

Sovereign Debt Capacity and the Distribution of Domestic Wealth: A Common Agency Model

This paper proposes a stylized two-period, two-country model illustrating the role of distribution of domestic wealth in determining a countrys level of access to international lending. We model sovereign debt redemption policy in a common agency framework. Within this framework, policy is the outcome of the interaction between government and local and foreign interest groups with conflicting preferences on debt repayment. Our main result is that in full lobby competition, when all interests are represented, the only equilibrium solution is repudiation and the consequent inability of government to access international capital markets. Conversely, when the ability to lobby depends on wealth, governments can access international credit up to a given maximum external debt capacity, determined by the skew in the distribution of domestic wealth.

Does it Take Two to Tango? Improving Cooperation between the IMF and the World Bank: Theory and Empirical Evidence

In this paper we present a theoretical model which, focusing on the quality of information transmission between the IMF and the WB, analyzes the sources of the expected loss in the overall performance of the two institutions relative to the first best outcome, which is characterized by centralized decision and perfect information. In particular, given the Bank-Fund strong complementarities, we show that strategic communication is indeed the primary source of loss for the two institutions. A testable implication of the model is to relate Bank-Funds performance to their willingness (or ability) to communicate. We find evidence that a Bank-Fund simultaneous loan is beneficial to growth and, consistently with the theory, such beneficial effect is reduced by factors preventing full communication, such as the degree of Bank-Fund competition and the salience of their private information.

Does it take two to tango? How to improve cooperation between the IMF and the World Bank

Despite a series of official agreements aimed at strengthening Bank-Fund cooperation, it is widely believed that coordination between the two organizations often falls short of what should be rationally expected. A greater pressure for conformity, required by the need of coordination, may contrast with the objective of enhancing policy adaptation to the specific conditions revealed by each institution’s specialized expertise. In this paper we present a theoretical model which, focusing on the quality of information transmission between the IMF and the WB, analyzes the impact of different governance structures on the degree of Bank-Fund cooperation. We compare the performance of a “delegation-scheme” with that of a “centralization-scheme.” A centralized structure better addresses the necessity of coordinating policy actions but greater consistency between the Bank and the Fund policy actions will be achieved at the expenses of a less satisfactory adaptation to “local conditions.” When the need for coordination is relevant, as it is generally the case for the Bank and the Fund operations, we find that a centralized governance structure allows to achieve a level of overall payoff greater than those of a decentralized one.

The Role of Trade and Offshoring in the Determination of Relative Wages and Child Labour

ABSTRACT Incorporating family decisions in a two-period model of the world economy, we predict that trade liberalization raises the skill premium and reduces child labour in developing countries where the adult labour force is sufficiently well educated to attract production activities from abroad that will increase the demand for skilled relative to unskilled labour. Elsewhere, liberalization will reduce the skill premium, but it will not necessarily raise child labour. Our prediction is not rejected by the data, and it explains why child labour is negatively associated with trade openness in those developing countries where the labour force was relatively well educated when the liberalization took place, but not elsewhere.

Agency and communicaton in IMF conditional lending: Theory and empirical evidence: theory and empirical evidence

We focus on the role that the transmission of information between a multilateral (the IMF) and a country has for the optimal design of conditional reforms. Our model predicts that when agency problems are especially severe, and/or IMF information is valuable, a centralized control is indeed optimal. To the contrary, when local knowledge is more important than the agency bias we expect delegation to dominate. Controlling for economic and political factors, our empirical tests show that the number of IMF conditions is lower in countries with a greater social complexity, while it increases with the bias of the countries’ authorities, openness, and transparency, consistently with the theory. ∗A preliminary verison of this paper was presented at the Workshop on Sovereign and Public Debt and Default (University of Warwick, 2008), ASSET meeting (EUI, 2008), Workshop on Political Economy (Silvaplana, 2008), Bank of Finland monthly research seminar (Helsinky, 2008), the Annual Meetings of the Political Economy of International Organizations (Ascona, 2008 and Geneva, 2009). We would like to thank Pascal Courty, Jakob de Haan, Francesco de Sinopoli, Federico Etro, Raquel Fernandez, Patrizio Tirelli, Julien Reynaud, Stephanie Rickard, Michele Santoni, Randall Stone, Tuomas Takalo, Otto Toivanen, Mark L.J. Wright for useful comments. Corresponding author: Silvia Marchesi, Dipartimento di Economia Politica, Piazza dell’Ateneo Nuovo 1, I-20126 Milano. Tel. +39 02 6448 3057; Fax. +39 02 6448 3085; E-mail: silvia.marchesi@unimib.it.

Agency and communicaton in IMF conditional lending

We focus on the role that the transmission of information between a multilateral (the IMF) and a country has for the optimal design of conditional reforms. Our model predicts that when agency problems are especially severe, and/or IMF information is valuable, a centralized control is indeed optimal. To the contrary, when local knowledge is more important than the agency bias we expect delegation to dominate. Controlling for economic and political factors, our empirical tests show that the number of IMF conditions is lower in countries with a greater social complexity, while it increases with the bias of the countries’ authorities, openness, and transparency, consistently with the theory. ∗A preliminary verison of this paper was presented at the Workshop on Sovereign and Public Debt and Default (University of Warwick, 2008), ASSET meeting (EUI, 2008), Workshop on Political Economy (Silvaplana, 2008), Bank of Finland monthly research seminar (Helsinky, 2008), the Annual Meetings of the Political Economy of International Organizations (Ascona, 2008 and Geneva, 2009). We would like to thank Pascal Courty, Jakob de Haan, Francesco de Sinopoli, Federico Etro, Raquel Fernandez, Patrizio Tirelli, Julien Reynaud, Stephanie Rickard, Michele Santoni, Randall Stone, Tuomas Takalo, Otto Toivanen, Mark L.J. Wright for useful comments. Corresponding author: Silvia Marchesi, Dipartimento di Economia Politica, Piazza dell’Ateneo Nuovo 1, I-20126 Milano. Tel. +39 02 6448 3057; Fax. +39 02 6448 3085; E-mail: silvia.marchesi@unimib.it.