Chicken and Egg issues are at the heart of what economists are supposed to be doing. Do you have a clever instrument? Is "clever" a nice word? Andrew Rose and Mark Spiegel ask an interesting question in their new NBER paper titled: "NON-ECONOMIC ENGAGEMENT AND INTERNATIONAL EXCHANGE: THE CASE OF ENVIRONMENTAL TREATIES". They ask; "when nations work together on international environmental issues does this facilitate economic interaction?". To my naive eye, this is a hard question. A Princeton PHD would demand that they come up with a randomization device that forces pairs of nations to work together one environmental issues. The Princeton researcher would then do a before/after comparison to see whether the two "randomly paired" nations increase their economic interactions relative to control groups (pairs of nations that were not randomly paired to work together). Can this "ideal" be approximated using instruments that can be constructed? In the intro, the authors do a good job sketching reverse causality stories related to; "if two nations trade together a lot for exogenous reasons, they may work together to mitigate tragedy of the commons issues."
This paper has clear policy implications --- The Bush Administration will always have to wonder whether it would have been able to form an International Iraq coalition had it been a better global citizen on issues that interested Western European nations and the U.N.
NON-ECONOMIC ENGAGEMENT AND INTERNATIONAL EXCHANGE: THE CASE OF ENVIRONMENTAL TREATIES
Andrew K. Rose
Mark M. Spiegel
Working Paper 13988
NATIONAL BUREAU OF ECONOMIC RESEARCH
1050 Massachusetts Avenue
Cambridge, MA 02138
Countries, like people, interact with each other on a number of different dimensions. Some interactions are strictly economic; for instance, countries engage in international trade of goods, services, capital, and labor. But many are not economic, at least not in any narrow sense. For instance, the United States seeks to promote human rights and democracy, deter nuclear proliferation, stop the spread of narcotics, and so forth. Accordingly America, like other countries, participates in a number of international institutions to further its foreign policy objectives; it has joined security alliances like NATO, and international organizations such as the International Atomic Energy Agency. In this paper, we concentrate on the interesting and under-studied case of international environmental arrangements (IEAs). We ask whether participation in such non-economic partnerships tends to enhance international economic relations. The answer, in both theory and practice, is positive.
Memberships in IEAs yield costs and benefits. A country can gain directly from such interactions; its air might be cleaner, or there might be more fish in the sea. However, some gains can be indirect. For instance, countries with long horizons and low discount rates might be more willing both to protect the environment and to maintain a reputation as a good credit risk. If they can signal their discount rate through IEA activity, participation in IEAs may indirectly yield gains from improvements in credit terms. Alternatively, countries that are tightly tied into a web of international relationships may find that withdrawing from one domain (such as environmental cooperation), may adversely affect activities in an unrelated area (such as finance). The fear of these spillovers may then encourage good behavior in the first area.
Our theoretical analysis begins with an extension of the “reputation spillover” concept introduced by Cole and Kehoe (1997). In our model, countries – or rather, their policymakers –
differ in their attractiveness as borrowers. Formally, we model these differences as differing discount rates among borrowing country governments, but one could envision alternative differences, such as disparities across countries in the perceived cost of default on external debt obligations. We concentrate on the example of discount rate differences in our theoretical model for analytical tractability, but we do not intend to suggest that this is the only source of heterogeneity in creditworthiness across countries.
With differing discount rates, more patient governments choose to join a greater number of environmental treaties; this sends a credible signal concerning a country’s debt capacity. Creditors respond by lending the country more capital. The predictions of this model are multilateral, since membership in IEAs is easily-accessible common knowledge. A country that joins more IEAs enhances its reputation with all nations.
This multilateral model is an intuitive start. Still, it misses the possibility that membership in an IEA may confer special advantages to its members; if Argentina defaults on its Brazilian debt, Brazil can retaliate by failing to support Argentine environmental initiatives. We thus extend the model to accommodate bilateral spillovers. We allow a creditor to respond to default by reducing the debtor’s gains from involvement in mutual IEAs.1 This extended model demonstrates that cross-country economic interaction can be a function of solo and/or joint participation in environmental treaties. Succinctly: the more international environmental commitments that countries make individually and in common, the easier is economic exchange between the countries.2
We then take these ideas to the data. Using a cross-sectional gravity model to control for other phenomena, we find that participation in IEAs is indeed positively associated with the international exchange of assets. This confirms the notion of positive spillovers between
environmental cooperation and economic exchange. Moreover, we find that multilateral IEA participation is not a sufficient statistic to explain bilateral economic exchange; joint IEA participation is also related to asset cross-holdings. We corroborate these findings using a panel of data that includes bank loans and FDI. We use a number of instrumental variables to show that our empirical results do not depend on the assumption that causality flows simply from environmental engagement to economic exchange. Our empirics thus support our extended model with both a reputation effect and some sort of bilateral punishment mechanism.
A brief survey of the literature section is provided in section 2, while our theoretical framework is developed in the following section. The empirical work is presented in section 4. The paper ends with a brief conclusion.
2. Literature Survey
The concept of reputation spillovers arose as a response to the Bulow and Rogoff (1989b) challenge to the sovereign debt literature. In their seminal paper, Bulow and Rogoff cast doubt on the possibility of sustainable sovereign lending based solely on the desire of borrowers to maintain their reputations. They demonstrated that such relationships would not be sustainable, because a borrower would eventually prefer to default on its debt and “self-finance” its consumption-smoothing.
This challenge was addressed in a series of papers by Cole and Kehoe (1995, 1997, 1998). They show that the problem with reputation-based borrowing stems from the fact that a borrower able to replicate interactions with other creditors receives only transient benefits from such relationships. At some point, the benefits of maintaining a reputation fall sufficiently that default and subsequent self-finance is the rational response. However, Cole and Kehoe (1995)
show that the desire to maintain other interactions with creditor nations may support debt, provided that these other relationships are not transient but enduring. Cole and Kehoe (1998) demonstrate that the desire to maintain reputations in enduring relationships can support a debt relationship with transient benefits. Cole and Kehoe (1997) show that the desire to maintain an enduring relationship can support a transient debt relationship in a simple trigger-strategy model, where a creditor responds to default by breaking off debtor relationships with enduring benefits.3 We borrow this modeling strategy below in our theoretical work.
Other signals of creditworthiness have also been examined in the literature. Milde and Riley (1988) argue that borrowers can signal the quality of underlying investments through the magnitude of borrowing. Alternatively, borrowers may resume payments subsequent to default, either to affect bargaining negotiations (e.g. Gale and Hellwig, 1989) or to signal government stability (e.g., Cole, Dow, and English, 1995). Similarly, creditworthy borrowers may signal their types by pursuing fiscal contractions (e.g. Drudi and Prati, 2000), enduring costly recessions prior to defaulting, as in Alfaro and Kanczuk (2005), or by accepting rescheduling packages at tough terms as in Spiegel (2005).
On the other hand, default decisions can themselves be signals of other types of information. Sandleris (2006) develops a model where government default decisions provide information about economic fundamentals that affect private sector behavior. He shows that the desire to signal to the domestic private sector that fundamentals are good can be sufficient to generate lending in an environment without default penalties.
A different literature of relevance concerns the formation and characteristics of IEAs; references here include Barrett (1994), Carraro and Siniscalco (1998), and Finus et al (2005). Most of the literature is skeptical about the ability of voluntary self-enforcing IEAs to improve