What Is US Federal Debt Worth?

Dean Baker raises a fun point albeit to disparage the much deserving Reinhart-Rogoff 90% analysis. His point is that there is a difference between the publicized value of the US debt and its current worth. This difference, as every finance major can tell you, is because US debt is issued and redeemed at par and has a fixed interest rate during its lifetime. At any stage between issue and redemption the actual, or market, value of a bond is calculated as the worth of the stream of interest payments its owner receives adjusted to reflect current interest rates. So a bond issued with a 2% interest rate is worth less when interest rates are 5% – it is producing less income than an investor could receive from a more current bond. Conversely a bond issued with a 5% interest rate goes up in value if rates fall to 2%. The general rule is that bond values move inversely to the interest rate.

This prompts us to ask what the value of the US debt is currently. In other words how much would it cost to redeem the debt in today’s rate environment? Obviously this is an academic question, no one is suggesting the debt is redeemed. But as Baker points out the value of the debt is a somewhat misleading figure because it is constantly moving. Thus calculating simple debt to GDP ratios is just that – simple. It doesn’t tell the whole story, and it ignores a highly relevant piece of information, which is what the market would pay for all that debt today. Helpfully, the Federal Reserve Bank of Dallas does the calculation for us. If you examine their charts closely you can see that the current value of US debt held by the public – a figure that ignores US debt held by the government itself – is slightly higher than its par value. This is because there is a pile of bonds issued some time ago when rates were higher making their market value greater than their par value. Indeed if you look at the charts you can identify periods of high interest rates, in the late 1970’s for instance, because the total debt’s market value was below par. Whereas in the more recent period, say from the late 1980’s, the debt’s market value has tended to be above par because some of it still carries higher rates.

Now, this is a largely academic exercise because the US government doesn’t really care about its current balance sheet value, it cares more about the interest cost burden on its budget. So ought we. From our point of view the most important statistic is the amount of interest the government has to pay each year on the debt we have borrowed along the way. As that amount rises, because the government borrows more or because rates go up on the new debt, or a combination of both, its interest cost eats into the amount it has available to pay for other things – tanks, roads, health care and so on. And, just as important, barring default the interest cost is an unavoidable obligation. It is not discretionary. At least not in any non-apocalyticla sense of that word. So the only reason we worry about US government debt is that its cost can put a squeeze on other things the we may ant the government to do.

This is all very straightforward. Except you wouldn’t get that sense by listening in on the current debate about the government’s debt and budget. That debate focuses on the amount debt, not its cost. The debt is always being positioned as being a burden on future generations, with the burden being the total outstanding, not its cost, nor any of the things the debt bought or produced. And without the context of its cost or purpose any discussion of the debt is meaningless. For instance were we to borrow enough to build a world class railway system we would be bequeathing not just the debt, but a cleaner environment and faster inter-city travel, bot of which are major benefits. And were we able to borrow that money at 1%, thus increasing future cost by a negligible amount, future generations might look back at us as being both wise and thrifty. After all when the bonds come due they always have the option of rolling them over rather than redeeming them outright.

It gets to be more problematic doing this calculation when the purpose of the borrowing is to finance current spending not long term investment in a Keynesian style stimulus of the economy. In these circumstances the value is found in the speed with which the economy can be shifted from underperforming back to operating at its potential. The return on investment from a stimulus is the elimination of a long period of underperformance. Each month that passes with the economy growing below its potential is a month with income lower than it might have been.

These two uses for debt speak to different aspects of the economy. One is expanding the supply of services – like building a road. The other is expanding demand which feeds back as income to those selling things to meet that demand. Either way the value to society is measured by the increase in income creating activity less the cost required to get that increase. The market value of the debt is irrelevant. Only the par value and its interest cost is.

Which brings us all the way back to Baker’s point.

Debt to GDP ratios have very little meaning. They are more an effect of underlying activity, not a cause of that activity. Economists and politicians who hyperventilate about high debt to GDP ratios usually are not talking about the debt, but about the role of government and, specifically, about the size of government. They are saying, for instance, that the government ought not to borrow at 1% to build roads, and that we should leave road building to someone else. Even if that someone else has to borrow at 3%. They are concerned that the government’s borrowing needs are so large that it sucks up all the available cash from the market and leaves less space for private borrowing – which they axiomatically view as being more productive.

So, if the amount of debt is less of a concern than its cost, why does anyone look at its ratio to GDP? Because that ratio is used as a proxy for the future ability of the government to pay the interest cost, not as a proxy to determine future economic growth. Take Greece as an example. Greek debt ratios were troublesome because they indicated future interest costs were going to swamp the budget. Investors began to fear they would not be paid. They began to ask for higher rates to compensate them for this possibility. And this set in motion a panic. Rates rose because of the fear of non-payment. And rising rates made future non-payment more likely. The debt to GDP ratio was simply a rough indicator of future insolvency problems. It said nothing about the Greek economy’s likely rate of growth other than that rate of growth was insufficient to be able to pay the cost of debt, which was rising more rapidly.

Phew.

Now revert back to the US.

Its rising debt to GDP ratio has not triggered any Greek style reaction.

There are two reasons for this.

One is that there is a zero chance of a US default on its debt because US debt is denominated in its own currency. Greek debt is not denominated in its own currency, it uses Euros, not the old Drachma. So the US cannot go bankrupt  the way a bank can, it merely has to print enough money to pay off its debt. [Parenthetically: an American private bank’s debt is not denominated in its own currency either, it uses US Dollars, which is why private banks do face a risk of bankruptcy. Oh, to have your own printing press!]

The second is that US borrowing costs are so low that the future cost of debt is still manageable. The markets clearly think that US growth will be sufficient – at some point – to generate the necessary tax revenues to pay the debt cost. So the US is not perceived as being insolvent, at least with its current debt level and taxation capacity.

OK, one last point.

When we talk of a government “going bankrupt” what we really mean is that the cost of its debt is greater than its ability to raise taxes to cover that cost. It faces cash flow insolvency. In the private sector, as in the case of a bank, there is another form of insolvency: balance sheet insolvency. This is when the entity’s assets are worth less than its liabilities. Banks are particularly prone to tho balance sheet insolvency because they rely on their ability to sell off assets to pay off their liabilities. Obviously if their assets aren’t worth enough they have a problem. This is why regulators and creditors like to know what a bank’s assets and liabilities are worth in current market terms rather than in par terms. The question is simple: if we were to wind the bank up today, could it pay off its creditors? And since its assets and liabilities fluctuate in value depending interest rates knowing their market value is crucial to that determination.

The question of balance sheet insolvency doesn’t occur in the case of a government because creditors cannot seize its assets in bankruptcy. The primary asset a government possesses is the value of its future tax revenues which is a function both of its tax policy and its economic growth. Government’s are cash flow operations and creditors can only ever renegotiate their claim on that cash flow, they cannot close the government down and carve up its assets to repay themselves.

So.

All this means – once again – that the only thing we all need to worry about when we discuss the level of US government debt is whether it can pay the interest cost on that debt. The amount is only relevant as an indicator of that cost. And debt to GDP ratios are a very crude, at best, indicator of potential problems.

With US borrowing costs at near all time lows we should ignore debt to GDP ratios even more than usual. Our questions ought to be: how much to borrow? and What do we do with all this cheap money?

After all, debt is only worth what you do with it.