What went wrong? Why did seemingly rational bond investors continue to purchase Puerto Rican debt with only a modest risk premium, even though the macroeconomic fundamentals were dismal? Given weak fundamentals, either investors were stunningly myopic or Puerto Rican debt was implicitly insured by the U.S. Treasury. The rational investor model rules out the former hypothesis. 

 This project examines the latter hypothesis, which we label the “Treasury put.” The expectation of a federal bailout was perfectly reasonable given past behavior. In 1975, the city of New York faced bankruptcy, and it was eventually bailed-out by the federal government. Federal financial assistance to beleaguered institutions has been the norm: Lockheed (1974), Chrysler (1979), the Savings and Loan Crisis (1980's), Brady Bonds (1989), the Mexican Peso Crisis (1995), as well as the massive bailouts during the 2007-2008 Global Financial Crisis.

 Evaluating the “Treasury Put” hypothesis with a minimal set of assumptions is possible given two fortuitous features – the unique characteristic of Puerto Rican bonds and a “seismic shock.” Puerto Rico issued both uninsured and insured general obligation bonds. These bonds were issued on the same day and, in many cases, with the exact same maturity. These features allow us to compute accurately the risk premia on Puerto Rican bonds. For the period 2000 to 2012, the risk premia on Puerto Rican bonds did not reflect macroeconomic fundamentals and was bracketed by the risk premia on Corporate Aaa and Baa bonds. The first set of results suggests that compensation for default risk on Puerto Rican bonds was exceptionally low and that a “Treasury Put” was assumed by investors.

 That assumption was overturned by the 2013 Detroit bankruptcy and the federal government's truancy regarding a rescue package for bond holders. No Puerto Rican bonds have been issued since the Detroit default. To assess the impact of the removal of the “Treasury Put,” we track the trading of matched uninsured/insured bonds and compute the yield-to-maturity (YTM) on a monthly basis. Preliminary calculations indicate that the Puerto Rican YTM's were stable before the Detroit bankruptcy, but widened dramatically thereafter, thus supporting the existence of a “Treasury Put” and a misallocation of capital to Puerto Rico.

 Future work will compare the value of the “Treasury Put” to the values of the “Greenspan Put” (Miller, Weller, and Zhang, 2002) for the stock market and “Paulson's Gift” to banks (a cash infusion and a three-year guarantee on unsecured bank debt in October 2008; Veronesi and Zingales, 2010). Apart from adding to our understanding of the Puerto Rican debt crisis and why rational investors misallocated their capital, our evidence contributes to ongoing discussions about the role of government guarantees in financial markets (Bornstein and Lorenzoni, forthcoming; Diamond and Rajan, 2012). A quantitatively important “Treasury Put” would also raise questions about the proper specification of bond pricing formula, which usually ignores the potentially important channel of implicit government guarantees.