Bulow–Rogoff Sovereign Debt Model

The Bulow–Rogoff sovereign debt model is a pair of theoretical results from Jeremy Bulow and Kenneth Rogoff that challenge the reputational explanation of why countries repay foreign debt. In "A Constant Recontracting Model of Sovereign Debt" (Journal of Political Economy, 1989) they prove that a reputation for repayment alone cannot support lending, and in "Sovereign Debt: Is to Forgive to Forget?" (American Economic Review, 1989) they develop a bargaining model in which repayment is enforced by direct sanctions. Together the papers reorient the field of international debt from reputation toward real penalties and renegotiation. The results contrast sharply with the Eaton–Gersovitz sovereign debt model, and they build on tools from game theory and contract theory within business and economics theory.

In the reputational view, a government repays because default would end its access to credit markets and thereby deprive it of the ability to smooth consumption or finance investment. Bulow and Rogoff ask whether this loss can be large enough to deter default if the country has other financial options. Their answer is no, under a simple but general condition: the country can, after defaulting, save abroad and buy insurance-like claims from foreign financial institutions on a pay-as-you-go basis.

The central argument is a cash-in-advance construction. Suppose that a country that has been shut out of borrowing can still enter into contracts in which it pays a premium to a bank in advance and receives a payment in the future states in which it needs funds. Because the country pays first, there is no repayment risk for the bank, and no reputation is needed. Such contracts can replicate the consumption smoothing that lending provided. If the country can reproduce the benefits of access with these transactions, then the loss of access cannot be a punishment, and the incentive to repay disappears. The result holds when the country's alternative contracts yield the same value as the original debt contract: any positive-value debt would be repudiated, so the only equilibrium is no lending.

The intuition is that reputation works in settings where there is a service that the defaulter can obtain only by a reputation for good behavior. When the contracts can be written on a prepaid basis, the borrower is not dependent on trust. The limit of the argument is that the country must be able to find such contracts. Bulow and Rogoff acknowledge that if governments could not access them, reputation could matter, and that the cash-in-advance approach is an extreme case. Nonetheless, they argue it captures the true economics of repeated finance, since sophisticated market participants have strong incentives to offer these contracts.

If reputation cannot explain lending, something else must. In the 1989 American Economic Review paper Bulow and Rogoff model repayment as the outcome of bargaining between a debtor and its creditors after the debtor has stopped paying. The creditor has recourse to direct sanctions: seizure of the country's foreign assets and exports, disruption of its trade credit, and legal action in the creditor's jurisdiction. These sanctions are costly to the debtor, and the debtor and creditors split the surplus from avoiding them.

The framework implies that the amount repaid is determined by the debtor's expected cost of sanctions, not by the face value of the debt. In a Nash bargaining setting the settlement is a fraction of the sanction cost, so the debt can be repaid only up to the amount creditors can credibly threaten to impose. Two consequences follow. First, a debtor with large foreign-asset exposure or heavy reliance on trade finance can borrow more than one with few such exposures, which explains differences in debt capacity across countries. Second, debt trading at a discount in secondary markets reflects the market's estimate of the expected settlement.

Bulow and Rogoff also show that, in the repeated bargaining environment, sanctions are renewed in each period, which is why the first paper is called a constant recontracting model: the terms of existing debt are in effect renegotiated whenever the country benefits from doing so, so only the value of the sanction threat matters at each date.

A second contribution concerns the overhang created by large debts. When a country's debt exceeds the amount creditors can extract, the marginal return of improved economic performance accrues mostly to creditors, because the debt claims would absorb extra output. The debtor then has weak incentives to adopt reforms or invest. Krugman (1988) formalized the same point using the debt Laffer curve, in which a higher face value lowers the expected repayment. Bulow and Rogoff (1988, 1990) argued that creditors, as a group, may therefore gain from forgiving part of the debt, but that individual creditors have an incentive to free-ride, a public-goods style problem in which a Pareto-improving write-down is not implemented. They also argued that buybacks of discounted debt using the country's own resources or donor money transfer wealth to creditors and are an inefficient way to give aid, since the buyback price rises with the amount purchased and reflects the full value of remaining claims.

The two frameworks give different views of what enforces repayment and what limits borrowing.

The modern quantitative literature, for example the models of default with endogenous output costs, often merges both views: it uses the Eaton–Gersovitz structure for pricing, but it adds sanction-like output losses to support positive debt, a nod to Bulow and Rogoff. Empirical work finds that the cost of default includes both temporary market exclusion and trade or output losses, which is consistent with a mixture of the two mechanisms.

The Bulow–Rogoff results informed the debate on the 1980s Latin American debt crisis, the Brady Plan, and the later Heavily Indebted Poor Countries initiative. They support the view that debt relief is a transfer from creditors, that creditor coordination matters, and that official lending should be structured to avoid bailing out private claims. Critics argue that the cash-in-advance contracts central to the reputation result are not readily available to many governments, that the model offers little prediction on the timing of default, and that bargaining outcomes depend on assumptions about sanctions that are hard to measure. The models also interact with questions about information asymmetry between creditors and debtors, and the design of mechanism design solutions such as collective action clauses remains an active research area.

1. Jeremy I. Bulow, Kenneth Rogoff (1986). "A Constant Recontracting Model of Sovereign Debt." Journal of Political Economy. https://doi.org/10.1086/261596
2. Eaton, J., & Gersovitz, M. (1981). Debt with Potential Repudiation: Theoretical and Empirical Analysis. The Review of Economic Studies, 48(2), 289–309. https://doi.org/10.2307/2296886
3. Krugman, P. (1988). Financing vs. Forgiving a Debt Overhang. Journal of Development Economics, 29(3), 253–268. https://doi.org/10.1016/0304-3878(88)90044-290044-2)
4. Arellano, C. (2008). Default Risk and Income Fluctuations in Emerging Economies. American Economic Review, 98(3), 690–712. https://doi.org/10.1257/aer.98.3.690