TY - JOUR
T1 - Pooling risk among countries
AU - Callen, Michael
AU - Imbs, Jean
AU - Mauro, Paolo
N1 - Funding Information:
We are grateful to two anonymous referees, and to Eric van Wincoop, Tamim Bayoumi, Mick Devereux, Michael Kremer, Raghuram Rajan, Jaume Ventura, and seminar participants at U.C. Berkeley, Cambridge U., the International Monetary Fund, U. Pompeu Fabra, the San Francisco Fed, and Tufts U. for insightful suggestions; to Nicolas Metzger, and José Romero for superb research assistance; and to Jean Salvati and especially Huigang Chen and Alin Mirastean for invaluable help in programming and computational support. Financial support from the Banque de France Chair at the Paris School of Economics is gratefully acknowledged. All errors are our own.
Publisher Copyright:
© 2015 Elsevier B.V.
PY - 2015/5/1
Y1 - 2015/5/1
N2 - Suppose that international sharing risk-worldwide or with large numbers of countries-were costly. How much risk-sharing could be gained in small sets (or "pools") of countries? To answer this question, we compute the means and variances of poolwide gross domestic product growth, for all possible pools of any size drawn from a sample of 74 countries, and compare them with the means and variances of consumption growth in each country individually. From the difference, we infer potential diversification and welfare gains. As much as two-thirds of the first best, full worldwide welfare gains can be obtained in groupings of as few as seven countries. The largest potential gains arise from pools consisting of countries in different regions and including countries with weak institutions. We argue that international risk-sharing fails to emerge because the largest potential gains are among countries that do not trust each other's willingness and ability to abide by international contractual obligations.
AB - Suppose that international sharing risk-worldwide or with large numbers of countries-were costly. How much risk-sharing could be gained in small sets (or "pools") of countries? To answer this question, we compute the means and variances of poolwide gross domestic product growth, for all possible pools of any size drawn from a sample of 74 countries, and compare them with the means and variances of consumption growth in each country individually. From the difference, we infer potential diversification and welfare gains. As much as two-thirds of the first best, full worldwide welfare gains can be obtained in groupings of as few as seven countries. The largest potential gains arise from pools consisting of countries in different regions and including countries with weak institutions. We argue that international risk-sharing fails to emerge because the largest potential gains are among countries that do not trust each other's willingness and ability to abide by international contractual obligations.
KW - Diversification
KW - GDP-indexed instruments
KW - Growth-indexation
KW - Risk sharing
UR - http://www.scopus.com/inward/record.url?scp=84929129835&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84929129835&partnerID=8YFLogxK
U2 - 10.1016/j.jinteco.2015.01.006
DO - 10.1016/j.jinteco.2015.01.006
M3 - Article
AN - SCOPUS:84929129835
SN - 0022-1996
VL - 96
SP - 88
EP - 99
JO - Journal of International Economics
JF - Journal of International Economics
IS - 1
ER -