Financial Restructuring: Which Debt, How Much, and the Range of Choice

by William G. Shipman
Chairman, CarriageOaks Partners LLC

The RosenBerg Institute of Global Finance

Harvard Law School
Program on International Financial Systems

Symposium on Building the Financial System of the 21st Century: An Agenda for Japan and the United States
September 21-23, 2001

The Symposium on Building the Financial System of the 21st Century: An Agenda for Japan and the United States will soon convene at the Keidanren Guest House near Mount Fuji, Japan. Participants will explore many issues including debt restructuring by governments, banks and other debtors. As we discuss this topic we should be mindful that conventionally defined debt is just the tip of the iceberg. Whatever challenges the world faces in restructuring government debt, they are significantly more daunting when considering how to deal with a much larger obligation: the unfunded liability of government mandated pension systems.

Time is often a friend in the restructuring of debt. In pension reform, because of demographic realities, time is the enemy. As the demographic clock ticks, the range of choice narrows. As we gather for three days near Mt. Fuji perhaps we will find a way to speed up the process of reform as we think through and discuss its many tradeoffs.

Debt and the Range of Choice

Prohibitively high debt reduces choices. Argentina, much in the news lately, faces the possibility of defaulting on some or all of its $130 billion of government debt. In order to secure outside financing the government may be forced into a "zero deficit" position, limiting its spending priorities. The government is considering changing the reserves that back the peso from just U.S. dollars to a combination of dollars and euros. This has been interpreted by some as a prelude to a peso devaluation. The world has a rich history of governments paying their debts with devalued currencies.

Japan faces high and rising unemployment. At the same time banks harbor an unsustainable level of bad loans, and government debt-both central and local-hovers around 120 percent of GDP. Reform proposals such as requiring banks to write off their non-performing loans face opposition because such action would add to job losses. Here, too, debt limits choices.

In Europe, a maximum level of government debt was a criterion of being eligible to join the single currency union, the euro. Under the Maastricht Treaty cumulative government debt was limited to 60 percent of GDP. Although other financial criteria were required to be met under the Treaty, the debt to GDP hurdle was considered the most difficult to meet. Perhaps this is why, by design, the Maastricht definition of debt was limited. Had it been more inclusive, the march toward a single currency most likely would have halted.

And of course the willingness to lend to individuals is predicated on their ability to pay principal and interest when due. This in turn is related to other debts one must service so a high level of existing debt limits the choice of taking on additional debt.

Whether for the individual, a lending institution or a sovereign government, debt matters. The perceived ability to honor one's obligations affects one's range of choice.

Lesser Known Debts: Big and Small

Government debt is generally considered to be the value of its bonds outstanding, often referred to as explicit debt. There are other debts or liabilities, however, such as uncancelable long term leases, insurance commitments and unadjudicated claims against the government to list but just a few. These are real liabilities but as a practical matter they are relatively small and not well known. They are the equivalent of a balance sheet footnote.

There is another debt that most countries face, however, one that is also not well known; it is the unfunded liability of government mandated pensions or Social Security systems. In most countries this liability is greater than explicit government debt. If debt matters, if debt narrows the range of choice, then this liability must be addressed.

Government Pensions: a Brief History

Social Security started in Germany in 1889 and spread quite rapidly throughout the world in response to many and varied socioeconomic problems. In the case of the United States Social Security legislation passed 66 years ago in response to the Great Depression of the 1930s.

During the first half of the 1930s the U.S. economy contracted about 25 percent, unemployment reached 22 percent and the stock market virtually imploded, plunging by 70 percent. Our nation's economy was on its knees and President Roosevelt had to do something. Much like world leaders before him he chose, as part of his social reform, to transfer resources from the wealthy to the needy-generally the elderly-through a tax on payroll; for at that time those that just had a job were considered society's wealthy.

Pay-As-You-Go Financing: Intentions and Problems

Providing benefits for the elderly through a payroll tax is called pay-as-you-go (PAYG) financing. Under such an arrangement benefits paid to today's elderly are financed by payroll taxes from today's young. Benefits to be paid to today's young will be financed by payroll taxes from tomorrow's young. Thus, benefit payments are an intergenerational wealth transfer from younger workers to older retirees. There is no saving, investing or accumulation of wealth in a pay-as-you-go system. More than 100 countries presently employ PAYG financing.

As compassionate in intent these systems may be, their financial structure suffers from two long-term problems. The first is a low embedded rate of return. The second is that they rely on unrealistic demographic assumptions.

Benefit payments can be no greater than the payroll subject to tax times the tax rate applied to that payroll. Holding the tax rate constant, benefits can increase no more than the increase in payroll. For most countries the real increase in payroll is less than the return on equity and debt capital, an alternative for financing retiree benefits. In the United States, for example, taxable payroll increased annually by about 1 percent in real terms over the last 4 decades. In contrast, a portfolio of 70/30 percent stocks and bonds, respectively, earned an annual real rate of return of about 6 percent during the same period. The five percentage point difference requires greater resources from a pay-as-you-go system to achieve a specified retirement income than is necessary from a market-based structure because of the latter's higher return.

The second problem is that the amount of retirement income any particular age group can receive depends on how many workers there are to tax relative to the number of eligible retirees at that time. This ratio is primarily determined by two factors: life expectancy and the fertility rate, both of which are moving in the wrong direction from the point of view of pay-as-you-go financing.

Life expectancy is highly correlated with wealth. Citizens of wealthy countries, those with high percapita GDP, live longer than citizens of poor countries. Wealthy countries can more easily afford clean water, clean air, vaccinations and effective medical treatment to list but just a few things that make it easier to survive.

One of the wealthiest countries in the world is Switzerland; life expectancy at birth is 80 years. One of the poorest countries is Mozambique; life expectancy at birth is only 38 years. Most all countries of the world lie between Switzerland and Mozambique in both measures of wealth and life expectancy.

The fertility rate, or births per woman of child bearing age, is also related to wealth. Women of wealthy countries have fewer children than women of poor countries. This is in part because poor countries tend to be agrarian based economies wherein a large family is of economic value. Many other variables such as culture and the availability of contraceptives also play a part, but wealth is a dominant factor.

Using again the example of Switzerland and Mozambique, fertility rates are 1.47 and 4.93, respectively. Again, most countries lie between these two extremes. The fertility rate that stabilizes a population is 2.11. Fertility rates in many European countries are particularly low: for example, France-1.7, Germany-1.3, Italy-1.2, Spain-1.15. There is now no longer any country in Europe where people are having enough children to replace themselves when they die.

As any single country becomes more wealthy, life expectancy rises, fertility rates fall and the ratio of workers to retirees shrinks. The United States is a case in point. In 1950 there were 16 workers per retiree, today there are just 3.4. In 2030 there will be only two.

The Historical Response: Taxes

As these demographic forces progressed over the last half century governments responded by raising payroll taxes to pay benefits. In the U.S. the maximum payroll tax in 1950 was just $90, for the maximum tax rate was just 3 percent on the payroll subject to tax, which was then $3,000. Today, just for the retirement portion of Social Security, the tax is 10.6 percent of $80,400 of payroll, a maximum tax of $8,522.40. That is a tax increase of over 1,200 percent after adjusting for inflation. Other countries have responded in a similar fashion; they tax payroll at a higher rate than the U.S. does, in most cases significantly higher.

The Future: Unfunded Liabilities

But as taxes have gone up pay-as-you-go systems have not become more solvent or secure. Indeed, they have become less secure. They face massive unfunded liabilities, or debt if you will. According to a World Bank Policy Research Report, titled "Averting the Old Age Crisis:"

In OECD countries, the social security debt varies from 90 percent to more than 200 percent of GDP. Although not well known because it is implicit, the implicit social security debt for all countries is much larger than the explicit debt. Adding both components would in most cases triple the total national debt.

Some Questions and Propositions

What are the implications of explicit debt being only a fraction of total government debt? Would a government's creditworthiness remain the same if its true debt were known? Would the market price government bonds differently? Would interest costs rise? Would rating agencies, such as Moody's, change their risk assessment of sovereign debt? Would they change their ratings?

As to the last question, one answer is to compare Moody's rating of sovereign debt for 18 countries-the US, EU, Japan and Australia-that have significantly different levels of implicit pension debt as a percent of GDP. As it turns out, ratings are quite similar across all 18 countries. In all cases the rating category states that "issuers demonstrate very strong creditworthiness."

Why would a rating agency apparently not consider significant unfunded pension liabilities when assessing the creditworthiness of sovereign debt? There are at least two possible answers. The first is that the rating agency is unaware of the pension obligations. The second is that it is aware but assumes that countries will default on those obligations. Should this second answer be correct, although there may be future social costs, such costs may not impact the ability to honor explicit debt obligations, which is the creditworthiness issue.

As to the first proposition, Moody's wrote in a November 2000 article titled: "EU Public Finance: Important Structural Risks Masked by Positive Cyclical Effects"

Expenditures on state PAYG pensions are large. They range from 10%-11% of GDP in Germany and France, to more than 14% in Italy, and will increase substantially until 2030. OECD simulations from 1996 - which are only broadly indicative - already provide some idea of the magnitude of future problems if no further policy action is taken. In 2030, the net present value of unfunded pension liabilities would be more than three times the level of GDP in some EU countries. These unfunded pension liabilities add to outstanding general government debt levels, which in some countries are already very high.

At least for Moody's, the first proposition should be rejected.

As to the second proposition, if it is correct, the social fabric of many countries will face significant stress because for most elderly people government pensions are their sole source of income.

One Additional Proposition

Neither proposition is attractive. More importantly, neither has to occur. There is a third proposition, one that may alter the debt landscape and the social fabric for generations to come. It is the structural reform of pay-as-you-go systems to fully funded or market-based systems. Such reform, if properly designed and implemented, meets multiple goals simultaneously. From the point of view of financial restructuring that we will discuss during the Symposium on Building the Financial System of the 21st Century, this proposition may yield great hope for severely indebted countries.