Previous articleNext article FreeCommentAlbrecht RitschlAlbrecht RitschlLondon School of Economics Search for more articles by this author PDFPDF PLUSFull Text Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinked InRedditEmailQR Code SectionsMore“The End of Gatekeeping: Underwriters and the Quality of Sovereign Bond Markets, 1815–2007” by Marc Flandreau, Juan Flores, Norbert Galliard, and Sebastián Nieto‐Parra presents new, fascinating historical insights into sovereign bond issues. Underwriters until World War II specialized on certain default probability ranges in a case of assortative matching: first‐rate debt would be issued by first‐rate issuing houses, debts of slightly lower quality by second‐rate houses, and so forth. In contrast, underwriters nowadays appear to be rather despecialized with regard to default risk. In the old regime, the underwriting industry was highly concentrated, whereas today it is not. The authors also find that in the olden days, the issuing houses would assume the risk of issue failure, whereas today they do not. The puzzle stated by the authors is that in the old equilibrium, the underwriter’s type would reveal the type of the borrower, whereas today it does not. The interpretation given to this evidence by the paper is that under the old regime, information asymmetries existed between debtors and creditors, which the issuing houses resolved by signaling quality to investors. Truthful revelation of borrower types was ensured by reputation effects. According to the authors, the increasing role of rating agencies since the postwar period has rendered this function of sovereign debt underwriters redundant: owing to the activity of rating agencies, default risk is now common knowledge (or that is at least what we thought until recently). The disappearance of information asymmetries leaves no role for signaling and reputation effects, and hence no role for underwriter specialization.The paper convincingly demonstrates that an equilibrium switch in the underwriting market indeed took place in the postwar period. But is there also a puzzle? One puzzling fact is that rating agencies appeared already before World War I, half a century earlier than the authors’ story would imply. If underwriters were indeed so critical in resolving information asymmetries, their position should have been eroded earlier by competition from the rating agencies. However, up until World War II the rating agencies and the traditional underwriting business coexisted, without undermining the assortative character of underwriter/debtor matching that the authors’ story would imply. Hence, the underwriter puzzle might be worse than the authors think—or is it?The sovereign bond issue process as described by Flandreau et al. is characterized by several stages. In the first stage, a debtor enters the market and reveals its type (i.e., its default probability) to the underwriters. In the second, a match with an underwriter of the same type (i.e., specialized in the appropriate default probability range) is established, and the underwriter buys up the bonds. In the third, the bonds are sold to the market, and the debtor’s type is revealed to the public. Given the information asymmetry between debtors and the market, underwriters will be able to extract a rent to the extent that they have control over their market segment.This reasoning looks fine as long as this game is one‐shot for each debtor and is repeated for the underwriters and the public. One‐shot debtor participation implies that debtor types have to be evaluated every time a bond is issued. Repeated market participation of the underwriting houses gives rise to the desired reputation effects in signaling debtor quality. Under repeated market participation of debtors, however, debtor types would become common knowledge after the first debt issue. This would erode the market power of the underwriters, an effect that would have to be assumed away in order for the authors’ story to hold. Assumptions that guarantee this are not entirely implausible: some debtors may choose not to participate repeatedly, whereas others may exhibit low serial correlation of debt quality. However, the principal problem with the authors’ interpretation remains that in their setup, signals about debtor quality are a public good, and returning borrowers could go with low‐cost issuing houses or use other facilities. Add to this the possible role of rating agencies in revealing debtor types in the first place, and the authors’ story looks somewhat less than robust.A possible alternative interpretation of the equilibrium switch reported by the paper rests on the increasing role of international organizations. Prior to World War II, few such organizations existed, and where they did, their weakness was often notorious. In this environment, issuing houses were able to operate as bondholders’ gatekeepers, helping to coordinate creditor interests in the face of default threats. Beginning with the Bretton Woods era, international organizations have increasingly assumed this role, crowding out underwriter banks from the business of managing sovereign debt crises.In summary, this fascinating paper describes a new and important stylized fact about sovereign bond issues. The authors’ own interpretation of this fact rests on information asymmetries between sovereign debtors and investors. To this commentator, this interpretation seems less than robust to repeated games, which would wash out the role of financial intermediaries in resolving the information asymmetry. Other tasks historically performed by these underwriters, such as their coordination of bondholder interests, might provide an alternative and more plausible interpretation. Previous articleNext article DetailsFiguresReferencesCited by Volume 6, Number 12010 Article DOIhttps://doi.org/10.1086/648698 Views: 112 © 2010 by the National Bureau of Economic Research. All rights reserved.PDF download Crossref reports no articles citing this article.