In the day-to-day flood of news reports about debt problems of Greece, Ireland, Portugal, Italy, and even the U.S., it's easy to get lost in a sea of opinions about the effects of a given bailout measure. The best tool to cut through this confusion is simple arithmetic.
As my friend John Mauldin wrote last week, "No country save Britain at the height of its empire has ever recovered from a debt-to-GDP ratio of over 150% without a default. None. And the reason is simple arithmetic. Even a nominal interest rate of 6% means that it takes 10% of your national income just to pay the interest. Not 10% of tax revenues, mind you; 10% of your total domestic production."
For most countries in Europe, government revenues typically run between near 40% of GDP, while government spending presently runs several percent ahead of that. In Greece, government debt now represents about 150% of GDP at interest rates between about 10% for very short and very long-maturity debt, to about 25% annually on 2-year debt (reflecting a high expectation of default). The overall average yield on Greek debt is close to 15%. The problem is that 15% interest on 150% of GDP works out to 22.5% of GDP in interest costs if the debt actually has to be rolled-over without restructuring it. That would be more than half of the government revenues of Greece. The only way Greece can avoid default with that math is if investors quickly become willing to roll over the existing debt at an interest rate in the low single-digits.
While last week's extension of further bailout provisions for Greece certainly gave the markets a rousing "risk on" day on Thursday, the main effect on Greek debt was to extend the expected date of certain default (estimated based on interest rate spreads and credit default swaps) from about 1.5 years away, where the expectation was on Wednesday, to closer to 2 years away, which is where the expectation was by Friday.
Unlike countries with an independent monetary policy, Greece can't print its own currency, and can't devalue its obligations on the foreign exchange markets. So it is stuck with impossible math, save for the willingness of other European countries - mainly France and Germany - to pay the tab. It's still possible for Europe as a whole to summon the political will to do that, given that the GDP of Greece is only about 10% that of Germany. Still, Italy has a debt-to-GDP ratio of about 120%, and a GDP about two-thirds that of Germany, so the math becomes fairly daunting if contagion spreads past small countries such as Ireland and Portugal (which also have debt-to-GDP ratios near 100%).
As I've noted several times in recent months, bond market spreads imply very low near-term (3-6 month) probability of default in any Euro-area country. A sovereign default is much more likely to occur near the end of the next bear market, whenever it occurs, than at the start. As Ken Rogoff and Carmen Reinhart noted in their book This Time It's Different , "Overt domestic default tends to occur only in times of severe macroeconomic distress." The most likely window for a Greek (or other Euro-nation) default will be at a point when France and Germany are experiencing economic downturns sufficient to douse the political will to bail out their neighbors at a cost to their own citizens.
Here in the U.S., total Federal debt to GDP is also approaching 100%, but the debt held by the public (outside of that held by Social Security and the Federal Reserve) amounts to about 60% of GDP and rising, due to recent budget deficits of about 10% of GDP annually. This is presently manageable since so much of that debt is of short-maturity and is being financed at very low interest rates. And though U.S. Federal tax revenues have historically run near 19% of GDP (they're presently only about 16% due to the sluggish economy), those depressed interest rates mean that debt service doesn't consume a huge chunk of revenues just yet.
Still, it's precisely that short average maturity that makes the debt problematic from a long-run perspective, because it can't be inflated away easily. In the event of sustained inflation, the debt would have to be constantly refinanced at higher and higher yields. Contrary to the assertion that the U.S. can easily inflate its debts away, it is clear that sustained inflation would create enormous risks to our long-run fiscal condition by driving interest costs to an intolerable share of revenues. At that point, any shortfall in GDP growth or government revenues would result in a rapid spike in debt-to-GDP (as Greece and other peripheral European nations are experiencing now). Prior to embarking on an inflationary course, the first thing a government would want to do is dramatically lengthen the maturity of its debts.
For Greece, and increasingly for Portugal and Italy, our view continues to be "certain default, but not yet." For the U.S., our view is that, barring significant restructuring of mortgage obligations, our debt problems are more likely to take the form of sluggish economic growth for an extended period of time. I continue to believe that the main window of inflation risk will begin in the back-half of this decade - not yet. Even so, we are already observing a sustained shift away from fiat currencies toward alternatives like gold (though there will certainly be fits and starts to that trend). Meanwhile, bond yields continue to offer very low yields-to-maturity, while credit spreads on corporate debt (even the riskiest types) have been squeezed as thin as they were in 2007. In stocks, on the basis of a wide variety of fundamentals, we expect the total return on the S&P 500 to average about 3.6% annually over the coming decade.
In short, the present menu of investment options provides little basis to expect significant nominal or real returns from long-run, passively-held investments purchased at present levels. I strongly doubt that investors will be deprived of opportunities to accept risk at higher expected returns and lower prices in the coming years. With few exceptions, the markets presently offer all of the risk with weak prospects for long-term return.
