Saturday, October 31, 2009

Some graphics on the US fiscal position

At the end of my last post, ( I mentioned that we were between a rock and a hard place as far as the fiscal situation of the US goes. Below are some graphics to show the history of our government expenditures, receipts, and deficit. All of the underlying data in this post are taken from two pages (GDP, FISCAL sections) of the St. Louis Fed website.

Perhaps the most salient point is that since 1947, expenditures have never represented a higher percentage of GDP while government receipts are the lowest percentage of GDP since 1967! (PLEASE CLICK ON THE PICTURES TO ENLARGE)

Peter Bernholz's work studying hyperinflation has been cited widely in the financial press recently. In his most recent book, Monetary Regimes and Inflation: History, Economic and Political Relationships, Bernholz analyzes the 12 largest episodes of hyperinflations - all of which were caused by financing huge public budget deficits through money creation. His conclusion: the tipping point for hyperinflation occurs when the government’s deficit exceed 40% of its expenditures. This is mathematically equivalent to when expenditures>1.67*receipts. The graph below illustrates this ratio through 2009q2; it will continue to deteriorate for at least 2 more quarters and probably several more after that.

Just as alarming from a creditor perspective is that government receipts are lower now, in real terms, than they were in 2000! The fundamental ability of any entity to service its debt has an upper bound determined by two variables: available revenue, and creditor faith. Of course practically speaking, costs cannot drop to zero, but from a strictly theoretical perspective, revenue provides an upper bound to how fast the debt can be repaid.

The faith of creditors is integral because it determines the interest on the debt. If the interest on the debt was >30% we would already be incapable of servicing our debt even if the government had zero expenditures. It is amazing how quickly a country can go from totally credit worthy to bankrupt just based on the perception of risk from investors. As David Einhorn put it in a recent newsletter: "events can move from the impossible to the inevitable without ever stopping at the probable." a pithy interpretation

To get an idea of what would be necessary to repay the government debt, we look at revenue stream, future expenditures, perception of risk, and nominal GDP growth.

First let's suggest a maximum theoretical revenue stream available to the US. The historical maximum (in % of GDP) for receipts is 31.9%, achieved in the first quarter of 2000. [WWII and the early 1980s both had percentages approaching 30% of GDP]. There is good reason to think that we cannot achieve a rate much greater than 30% of GDP. This is because the incentive to dodge taxes goes up exponentially with the tax rate. There is also an argument that high tax rates disincentivize the activity that is being taxed (particularly true for capital gains taxes.) For the purpose of argument, let's be generous and assume that we can achieve a sustainable rate of taxation of 32% of GDP. With this assumption, we can say that the US must default if the interest payments on our debt exceed 32% of our GDP. We are far off from that point now, as the interest on our debt represents only 5% of our receipts, and a little over 1% of GDP.

At first glance, 1% appears to be a pretty safe number. However, there are several things that must be true to keep it at that level. Most obviously, we must stop increasing the debt:GDP ratio. Most importantly however, market interest rates cannot rise. The current low cost of servicing out debt is merely a function of record low interest rates. These record low interest rates are dependent on two things: the nominal market interest rate must continue to hover around zero, and creditors must not perceive a risk of default. Market interest rates are (essentially) zero now only because those are the current expectations for returns. If the market shifts to the perception that GDP growth is sustainable, then interest rates will increase by an amount corresponding to the shift in perception. It is likely that fiscal and monetary policy will remain accommodative
until this perception changes! Therefore, we will continue to increase debt levels and increase monetary stimulus until perceptions of future growth, and the real interest rate, shift. So it is only a matter of time until the interest on the US debt will rise. If the average interest on US debt rose to a very modest 5%, interest payments would suddenly jump to 10% of our maximum receipts. Keep in mind that these calculations are made on our maximum receipts, not maximum budget surplus. Expenditures in the future will certainly not be zero.

Incorporating a realistic but austere forecast of expenditures, let's say that the country can cut government expenditures to 15% of GDP (which is less than half of our current expenditures of 35.5% of GDP). To be clear, cutting government spending to this level would seem draconian: social safety nets would be cut, defense spending would be decimated, medicare and social security would only be token programs, etc. We would be down essentially to education, a reduced military, bare-bones infrastructure spending, and emergency local and state services. With this austere assumption, the budget surplus could theoretically be 15% of GDP; even in this best-case scenario however, a 5% market interest rate would imply 20% of our budget surplus would go simply to service debt, and only 80% to pay down the principal.

Finally, and most importantly, we come to the risk premium demanded by investors. If investors think (for whatever reason) that there is an "x" percent chance of default, than the interest on government bonds will be approximately x%+real market rate. Let's illustrate this with an example: if the market perceives a 10% risk that the US will default on half its debt, then the risk premium charged on debt would be 5%. So if the nominal market rate of interest was 5%, and investors felt they were risking 5%(10% chance of losing half their principal) of their capital to default, the real rate would be 10%. A budget surplus of 5% of GDP with an interest rate of 10% on our current debt would imply that we were spending 120% of our surplus on servicing the debt. If we could not increase this rate of repayment, then we would be in default. You can see how quickly a loss of faith results in insolvency.

Adding in the element of inflation and GDP growth completes the picture. I talk about this more in the next post, but briefly, the rate of inflation needs to be added to the real interest rate to give a nominal interest rate. If there is a 5% rate of inflation and 3% rate of GDP growth, even with a zero percent risk of default, the nominal interest rate would be about 8%. This would mean even a 5% current account surplus would not be sufficient even to service the debt on the US balance sheet.

No comments:

Post a Comment