Reframing the Collective Action Problem in Sovereign Bond Restructuring
Lazard—the leading sovereign debt advisory firm in the world right now—makes an interesting point in their recent white paper on debt issues, namely that in practice, emerging market bond holders aren’t all that dispersed and atomized.
Rather than thinking of thousands of individual accounts, there are some really big players, such as Blackrock, Fidelity and Vanguard—or Ashmore, Franklin Templeton, WAMCO, PIMCO, and T. Rowe Price and no doubt others. They trade with each other and some smaller players with the help of a few large intermediaries. But at the end of the day, these big players have the heft, in Lazard’s view, to play an active role in coordinating sovereign restructurings. They could play the same role that Citigroup and a few other global banks helped played in the sovereign workouts of the 1980s and 1990s. They just have chosen not too. Lazard:
“Latin American debt restructurings in the 1980s were thought to be simplified by the limited number of actors—the large U.S. banks notably—and the role of the U.S. Treasury. As a result, the view, a decade ago, was that the replacement of bank debt by bond debt would make restructurings much more complicated as a result of the atomization of creditors. In that respect, the case of Ukraine in 2015, where one single institutional investor could hold a large share of total debt and play a systemic role in the resolution of the crisis looked somewhat like an anomaly. In fact, the fund management industry has gone through its own concentration trend and, combined with the increased popularity of emerging market investing, large fund managers now dominate the field. In the most recent restructurings of sovereign debt, the largest fund managers have all been at the negotiation table. As a result, a more institutionalized group of the largest fund managers could be a decisive stakeholder in the near future—if fund managers wish to collectively engage in such a way.” [Emphasis added.]*
It is an interesting point, especially considering the absolute failure of bond holders to step up and “voluntarily” contribute to the World Bank’s Debt Service Suspension Initiative even though many of the world's poorest countries have tapped global bond markets over the last ten years and as a result debt service on these bonds is no longer de minimis.
The basic issue there has been simple—there are not many bonds maturing this year, but the coupon on the existing stock of bonds is high (data). Paying those high coupons are more of a burden for countries in a world with limited and costly new inflows, falling export, and shrinking tax receipts.
But the transaction costs associated with restructuring a bond to push back a coupon are high. All outstanding bonds would need to be restructured, and the process of restructuring the bonds would almost certainly result in a temporary credit downgrade. As Lazard notes in their paper, most countries reasonably concluded that the costs were higher than the benefits if the restructuring did nothing more than push out six to eight months of coupon payments by a year or two. The IMF reached a similar conclusion in its recent paper on the sovereign debt restructuring architecture: "the potential benefits of [a limited debt service suspension] have been viewed by many sovereigns as smaller than the potential costs."
As a result, no bonds have been voluntarily restructured to provide flow relief to match the flow relief offered by traditional bilateral creditors (the Group of Seven [G7] countries together with the other creditors who gather at the Paris Club). Zambia is seeking a temporary standstill while it pursues a more comprehensive restructuring to address what sure seems like a clear cut case of insolvency. But that is it. The G7 recently noted: "voluntary private sector participation has been absent, which has limited the potential benefits for several countries."
Now, financially speaking, there is a simple solution to the “downgrade” problem: creditors could commit to recycling coupon payments into a new low-rate bond (say a five-year bond with a coupon linked to LIBOR).
Payments on existing bonds would be made, so there would be no default or cause for a downgrade. Yet by recycling coupons payments into a new bond creditors would not be taking cash out of low income countries amid a devastating pandemic, and recycling coupon payments into a new bond even on concessional terms would not cost creditors that much in the grand scheme of things—the new low coupon bonds would not trade at par, but the amount of cash flow at stake is something like $5 billion, and the new bonds would have substantial upside as countries recover (bonds of solvent countries converge to par as they approach maturity).
And by setting the issue of “comparability” and private participation in the Debt Service Suspension Initiative aside, bond holders would clear the air, which might lead spreads on existing bonds to fall and for the market to rally. Big investors could market their social chops, at a fairly modest cash flow cost.
So why didn't this happen?
The usual argument is that such an approach could never work because it is impossible to coordinate dispersed bond holders. Lazard though suggests that bond holdings are in fact often fairly concentrated, and I tend to agree.
The collective action problem then becomes something slightly different.
Any major fund manager that participated in a “voluntary” effort to provide interest relief needed this year and next amid the pandemic would under-perform relative to peers who took the cash payment and invested back in high coupon existing bonds. Accepting as payment in kind a bond with a concessional interest rate isn't as good as getting cash, and their relative performance would suffer.
That is a collective action problem no doubt.
But it is a bit different than the problem posed by thousands of dispersed investors.
If all the big investors agreed to participate, their relative performance would not be determined by participation... everyone would be pretty much left in same position (better than if they would have been had been widespread defaults and downgrades, worse off than if they had gotten cash in a narrow sense)
This should have been a solvable problem.
And it may well need to be solved in at least a few forward looking cases if the debt service suspension initiative is extended—or if in some cases the IMF concludes broader and deeper restructurings are needed—and the IMF and the Paris Club club condition their concessions on real participation by bond holders and Chinese policy banks.**
* I have been reliably informed by Ted Truman that the Federal Reserve did the bulk of arm-twisting back in the 1980s, not the Treasury.
** China has indicated that the China Development Bank should be viewed as a commercial lender, not a bilateral (official) creditor. China Exim's concessional loans, by contrast, are generally considered as official bilateral exposure and China Exim will participate in the Group of 20 debt service suspension initiative. But in some cases, China Exim has lent on fairly commercial terms and even as part of a consortium that includes Chinese state commercial banks, and it isn't clear if such debts will be covered by the debt service suspension initiative (the debt initiative was limited to direct government exposure and many of these loans will be to special purpose vehicles set up to finance the construction of a hydroelectric dam or a power plant). The IMF's paper on the "International Architecture for Resolving Sovereign Debt Involving Private-Sector Creditors" thus was silent on the thorniest coordination problem of all—how to coordinate the restructuring of China's (state-owned) "private" creditors with the restructuring of truly private claims.