Calling montary nerds

[oldfart]

I may be nuts, but check me out on this.

Treasury securities with an issue maturity of 52 weeks are less are called "T-bills", are issued at a discount, and redeemed at face value. Lately the shot side of the yield curve has tipped up to reflect a market risk premium for anticipated default. But how can a T-bill default? When it matures, Treasury refinances with new T-bills at the next auction. There is no separate interest payment to fail to make. As long as the maturing T-bills are replaced with newly issued T-bills, there is no default and the public debt does not increase.

To carry it further, I suppose that refinancing maturing longer term Treasuries with T-bills might be treated the same way.

US government securities dealers value their inventory at market using present value formulas, but I believe the Bureau of the Public Debt can only treat each instrument at it face or redemption value, regardless of the maturity. I'm ot sure how the Fed values its Treasury holdings.

So my point is that refinancing maturing Treasury issues with new T-bills would not increase the public debt nor constitute a default. The new T-bills should be about as liquid as before.

The only glitch I can see is if there were to be a legal opinion that the auction mechanism constitutes a payment of interest which is not authorized (which would be a direct repudiation of Section 2 of the Fourteenth Amendment). Suppose you hold a 4 week T-bill with a face amount of $10,000 in your Treasury Direct account. When it comes due, you have authorized the Treasury to automatically take the redemption proceeds and buy a new 4 week T-bill. Since the new T-bill sells at a discount, there is a bit of money left over (treated for tax purposes as interest) which is credited to the cash portion of your Treasury Direct account. I suppose some one might argue that those accounts are a liability that should be considered a part of the public debt, but I don't think they treated that way now. They, along with other government accounts payable are not formally part of the public debt. Maybe the Treasury would suspend withdrawals from these accounts, but maybe not.

So folks, how do you think T-bills will be treated in ten days?

 

[expat_panama]

...refinancing maturing Treasury issues with new T-bills would not increase the public debt nor constitute a default. The new T-bills should be about as liquid as before...

That's my take, and it's what I've been seeing since last May when the debt hit the $16.8T limit.

...a payment of interest which is not authorized...

Funds were authorized for interest payments back when the T-bills were sold. What's happening is that interest rates are going up. Seven years ago T-bill rates were over 5% and interest payments were about $250B/yearly. In spite of the soaring debt interest payments have not been going up, in fact last year interest payments were still "just" the same $250B/yearly. The problem is that T-bill interest was only 0.15%. With rates going back up the payments could easily jump to several $T per year. The payments have been authorized; diverting those payments for unauthorized use would be embezzlement.

...how do you think T-bills will be treated in ten days?

Please let me know what happens. I'm hiding under my bed.

 

[Toro]

Prices of assets are affected by perceived risk. All securities are priced with a risk premium. The risk premium is a function of certainty. The market is still saying that it is still extremely likely the bills will be paid in full, but the probability of that certainty has declined slightly. So the price falls slightly and interest rates rise.

 

[oldfart]

Please let me know what happens. I'm hiding under my bed.

If you're like me, you will be in the fetal position sucking your thumb and clutching your blankie.

I just had this nagging fear that I had missed something major about accounting for T-bill interest. You are correct that since the interest is the discount, it has been authorized and paid at issue. Normally we don't even think about how the government accounts compute public debt and how that interfaces with government accounts payable.

 

[oldfart]

Prices of assets are affected by perceived risk. All securities are priced with a risk premium. The risk premium is a function of certainty. The market is still saying that it is still extremely likely the bills will be paid in full, but the probability of that certainty has declined slightly. So the price falls slightly and interest rates rise.

I agree, and this is what we are seeing in the market. Tuesday's auction reflected this. It leaves me with a couple of questions though.

1. Since the risk premium is traditionally computed as the excess over the return on a similar maturity Treasury issue, what base rate do we use to compute an increase in risk premium on T-bills? We can clearly see the effect in the time series data, but in a year or so how would we know if T-bill rates had returned to "normal" or if there was a "permanent" shift reflecting a risk premium on Treasuries? The easy answer is to compare the spread over time, but that is really a circular argument. How do we know that other securities are simply being considered safer? And are not the two explanations identical?

2. Suppose that the market for longer term Treasuries becomes thinner and the Treasury changes the mix of offerings to include a higher percentage of T-bills (God knows after Operation Twist the Fed is flush with long term Treasuries!)? It seems to me that this lowers the burden of debt service further by replacing maturing bonds at 8 or 9% with T-bills at less than 1%. There are dangers to going this short, but I need a couple more envelopes to work it out.

Any thoughts?

 

[expat_panama]

...Normally we don't even think about how the government accounts compute public debt...

--and that may because for hundreds of years both private debt and business debt have both been bigger than government debt. Now, for the first time in American history our national government has grown while the national economy has evaporated, to the point that national debt is now larger than private debt and bigger than total business debt.

Considering the fact that America's private economy is what supports the government, this is something that simply can't go on.

 

[expat_panama]

...lowers the burden of debt service further by replacing maturing bonds at 8 or 9% with T-bills at less than 1%...

The gov't would love that but bond traders are no longer willing to go along. In fact, 30-year T-bills were at their lowest a year ago at 2.5% but now the rate's at 3.7% and climbing back up toward the 5+ %'s we had in '07.

Like I was saying above, I do believe we're looking at $T annual outlays for interest.

 

[Toro]

Prices of assets are affected by perceived risk. All securities are priced with a risk premium. The risk premium is a function of certainty. The market is still saying that it is still extremely likely the bills will be paid in full, but the probability of that certainty has declined slightly. So the price falls slightly and interest rates rise.

I agree, and this is what we are seeing in the market. Tuesday's auction reflected this. It leaves me with a couple of questions though.

1. Since the risk premium is traditionally computed as the excess over the return on a similar maturity Treasury issue, what base rate do we use to compute an increase in risk premium on T-bills? We can clearly see the effect in the time series data, but in a year or so how would we know if T-bill rates had returned to "normal" or if there was a "permanent" shift reflecting a risk premium on Treasuries? The easy answer is to compare the spread over time, but that is really a circular argument. How do we know that other securities are simply being considered safer? And are not the two explanations identical?

2. Suppose that the market for longer term Treasuries becomes thinner and the Treasury changes the mix of offerings to include a higher percentage of T-bills (God knows after Operation Twist the Fed is flush with long term Treasuries!)? It seems to me that this lowers the burden of debt service further by replacing maturing bonds at 8 or 9% with T-bills at less than 1%. There are dangers to going this short, but I need a couple more envelopes to work it out.

Any thoughts?

1. That's a great question and I don't know the answer. My guess would be to compare the price movements of the near term maturity to further out maturities, with the difference being the "credit risk" of holding the near term maturity.

2. That's correct. However, if you are a corporate Treasurer, you are extending the duration of your debt at such low rates, not shortening them. Perhaps the Treasury should be doing the same because rates will go up eventually. The real interest rate is 1.5% to 2.5% below the long term average, so we can expect it to normalize within a few years. Because the duration of the debt has been shortened, we will find our interest costs rising at a fair clip when we roll our debt.

 

[Freemason9]

When it matures, Treasury refinances with new T-bills at the next auction.

I think that's part of your answer . . . if the federal government comes up against a virtual "debt ceiling," it will be unable to issue new T-bills at auction.

 

[expat_panama]

... if the federal government comes up against a virtual "debt ceiling," it will be unable to issue new T-bills at auction.

Wait a sec, the national debt maxed out last April at $16.8T and has stayed there ever since. What's happened is that every time someone cashes in a T-bill, the treasury sells another.