In a recent television interview, a well-known economist and an equally well-known hedge fund manager debated two things: whether the recent hysteria over Greece was just that, temporary hysteria, or whether the Greek problem was problem a long time in the making that was recognized long before. This post talks about how to remove the drama from sovereign risk.
This video posted on www.zerohedge.com shows hedge fund manager Hugh Hendry, economist Jeff Sachs, and journalist Gillian Tett debating whether the Greek crisis was a surprise and whether or not Greece should “just get it over with” by defaulting sooner rather than later:
From the point of view of the Minister of Finance, looking at his own nation’s finances, should a rising level of risk ever be a surprise? In this blog, we looked at the same question from the perspective of the manager of a university endowment fund:
van Deventer, Donald R. “Optimal Balance Sheet Strategy: An Appreciation of Robert C. Merton’s “Optimal Investment Strategies for University Endowment Funds,” Kamakura blog, www.kamakuraco.com, May 4, 2009. Redistributed on www.riskcenter.com, May 6, 2009.
This blog is available at this link:
In our blog on endowment fund management, we agreed with Robert Merton’s 1993 observation that most universities made a very basic and very serious mistake in managing their own risk-they tended to look at endowment fund returns versus market indices of securities returns, instead of management the assets of the endowment fund versus their true liabilities, the fund’s obligation to provide funds to the university in the future.
Is the same kind of mistake being made by Ministers of Finance in Greece and the many other sovereigns whose credit is currently being called in to question? In order to answer that question, let’s look at the financial management of a sovereign as if the national government were a financial institution or major corporation.
We’ve addressed the many differences in predicting the default of a sovereign from an “outside in” perspective, that is, using publicly available information accessible to outsiders. We did that in this series of blog entries
van Deventer, Donald R. “Key Issues in Modeling Sovereign Defaults,” Kamakura blog, www.kamakuraco.com, August 3, 2009. Redistributed on www.riskcenter.com on August 4, 2009.
van Deventer, Donald R. “Case Studies in Sovereign Defaults,” Kamakura blog, www.kamakuraco.com, August 4, 2009. Redistributed on www.riskcenter.com on August 5, 2009.
van Deventer, Donald R. “Comparing Sovereign and Corporate Default Models: Facts and Figures.” Kamakura blog, www.kamakuraco.com, August 6, 2009. Redistributed on www.riskcenter.com on August 7, 2009.
Keeping that “outside in” perspective, these blog entries point out that the average annualized default rate on sovereigns is nearly 4 times higher than it is for public firms. They make the point that it is easier in the long run to predict sovereign defaults than it is for corporates. The opposite is true in the short run, however, because as my colleague Professor Jens Hilscher is fond of pointing out, “corporates default because they have to, sovereigns default because they want to.”
What if we take a different perspective, the “inside out” perspective where we have complete access to financial information on the sovereign? There are a number of key elements in analyzing the risk of default of the sovereign from this “inside out” perspective:
- “Cash in” stems primarily from corporate and individual taxes whose volume rises and falls with the business cycle and with macro factors relevant to that country: copper in Chile, oil in Bahrain, palm oil in Malaysia.
- Other assets of the country, which are rarely properly accounted for, include every fixed asset, land, and other property which the national government owns. This is not the same as fixed assets, land and other property “in country” but not owned by the government.
- Bonds and other borrowings of the country are the most obvious liability of the country. By number of transactions, because of the lumpiness of financings, for most countries the sheer number of these liabilities will be much smaller than the number of liabilities that a major retail bank would have after summarizing the completely homogeneous securities issuance to retail investors in government debt.
- Other liabilities include the on-going salary and other expenses that the government must pay in order to meet the services it pledges to provide its citizens.
- A liability that is potentially very large but which is often ignored is the obligation to meet the pension requirements of government employees. The current crisis with California pension funds is a classic case in this regard.
- More subtle obligations include either explicit or implicit obligations to bail out failing institutions. These would include not only government-affiliated entities but also those firms for which there was explicit support (i.e. bank deposit insurance) or implicit support (i.e. those institutions that are too big to fail).
None of these assets or liabilities is unusual or difficult, compared to what financial institutions have to deal with, and many of them (like pension obligations and bonds outstanding) are literally identical.
- Given that managing the finances of a sovereign are, if anything, less complex than the finances of a large financial institution, how can difficulties like those of Greece be a surprise? Robert Merton pointed out in his piece on endowment management that in many cases universities simply ignored their obligations to pay, for example, inflation adjusted salaries to faculty members in their analysis of risk and financial strategy. On the sovereign side, the most common mistakes can be listed in short order:
- Ignoring the value of national assets (i.e. a national highway system)
- Ignoring the revenue from national assets (i.e. toll revenue from that asset)
- Ignoring the expenses associated with national assets (i.e. maintenance obligations to maintain the revenue stream)
- Ignoring all cash flows beyond a short run modeling period
- Ignoring the true obligations to meet pension obligations by assuming unrealistic risk-return scenarios for national pension funds, as in the State of California
- Ignoring how movements in macro economic factors affect both “cash in” (i.e. tax revenue) and cash out (unemployment insurance, health care costs, etc.)
- Ignoring costs of implicit insurance, like bank deposit insurance
Remedying these mistakes is tedious and time consuming but not conceptually difficult and not difficult from a risk management systems point of view.
There are a number of other scenarios that are much more difficult that face national governments. Here are a few examples:
- Irrational acts by other governments, like the recent attack on a South Korean naval vessel by North Korea
- Terrorist attacks, like 9/11
- Natural disasters, like the earthquakes in Haiti or Chile
- Man-made disasters, like the oil drilling crisis in the Gulf of Mexico that began April 20, 2010.
The potential costs of national defense are the most difficult to quantify because, from the perspective of the 21st century, collective acts of violence by nations and terrorist groups are so irrational and so likely to result in failure. The third and fourth types of problems, however, are often priced in the catastrophe insurance market. While national governments may not be able to buy insurance in the volumes needed to hedge this risk, pricing for small amounts of insurance can at least be used as a guide for how much “cash in” this year needs to be set aside to “self-insure” against the next big earthquake in Japan, for example. Regrettably, very few governments have shown the collective will to do this self-insurance in the same way that a worker approaching retirement age would do to protect his own standard of living.
In the debate between Prof. Jeffrey Sachs and hedge fund manager Hugh Hendry, who was right? Is the hype about the Greek default just market over-reaction (Prof. Sach’s view on a sovereign that is risky but for whom ‘default’ would be foolish) or a long anticipated problem for which default sooner than later is the right strategy (Hugh Hendry’s position)? What we do know is this:
1. The credit default swap market does not have the depth to be a reliable indicator of sovereign credit quality, as we explained in this blog:
van Deventer, Donald R. “Sovereign Credit Default Swaps and Lessons from Used Car Dealers,” Kamakura blog, www.kamakuraco.com, May 11, 2010. Redistributed on www.riskcenter.com on May 13, 2010.
As we explained, only 20 of 183 sovereigns for which Kamakura provides default probabilities on www.kris-online.com have more than 1 non-dealer credit default swap trade per day. Even Greece, at the height of the market concern over the nation’s credit worthiness, averaged only 7.9 non-dealer CDS trades per day
2. If Greece was performing only “common practice” financial forecasts and risk management simulations, its financial difficulties should have been forecasted long in advance. Greece’s problems are not due to acts of war, other violence or natural disasters. For the same reasons, Greece and the State of California could have only gone beyond the point of no return, at which default is inevitable, by failing to meet minimal finance management standards.
The citizens of governments around the world deserve financial and risk management by the governments that at least meets the standards of what each citizen is required to insure their own well-being.
Donald R. van Deventer
Honolulu, June 16, 2010