Monday, July 18, 2011

Beyond Hyperbole

. Monday, July 18, 2011

First a little history.
April 3, 1979

The House of Representatives today approved an increase in the national debt's ceiling to $830 billion...ending a threat that the nation would default on its bills for the first time in history.

There have been previous crises over the debt ceiling...This time, the situation was more serious because the technical elements of debt management became more embroiled in the political controversy over a balanced budget...

The bill approved today did not go as far as Republicans had wanted in legislating a balanced budget...Representative Barber B. Conable of upstate New York...said, "All we want is to create a little presumption in favor of a balanced budget."

" U.S. Debt Ceiling Raised; Threats of Default Ended," By CLYDE H. FARNSWORTH New York Times Apr 3, 1979; pg. A12

March 29, 1996

After the Republican leadership threatened for six months to prevent the United States from borrowing billions of dollars more to keep the Government running, today's votes to raise the Federal debt ceiling were a strategic victory for Treasury Secretary Rubin. Mr. Rubin had warned for months that the Republicans were threatening the Government's credit rating. Several of the...leading credit-rating agencies agreed with him, warning investors that the political gridlock could interfere with the Government's ability to repay them.

Initially, Republican leaders dismissed Mr. Rubin's warnings as scare tactics. Some freshman Republican members even talked about impeaching the Treasury Secretary for taking a number of extraordinary keep the government afloat while Republicans tried to force President Clinton to make concessions.

"Brinksmanship to Victory," By DAVID E. SANGER New York Times Mar 29, 1996; pg. B11
It seems, therefore, that the contemporary standoff over the debt ceiling isn't novel. Bargaining theory suggest that both sides in these discussions have incentive to commit to extreme positions. The Republicans need to convince the administration that someone is willing to push the U.S. into default. The Republican leadership thus has strong incentive to push people with extreme views to center stage. The administration needs to convince the Republican leadership that defaulting would be far, far worse for them than a tax increase. The administration thus has incentive to paint the consequences of a technical default as financial apocalypse. Pegging yourself to the extreme is the only way to avoid being driven to your least preferred outcome.

In the past, both sides have found a way to walk back from the extremes and find agreement somewhere in between. Typically this agreement occurs at the zero hour. Why would it come earlier? Of course, past performance in no way guarantees future results, as they say, so things may turn out different this time around. But even if agreement proves elusive, financial apocalypse need not be the result. Indeed, I am puzzled by how people can be so certain about the consequences of an event that has never occurred. Moreover, these doomsday scenarios are hard to reconcile with the little bit of evidence that we do have.

The two observation we have to draw on suggest a more moderate conclusion. In late March and early April 1979, technical problems caused the Treasury to default on a few T-bills. According to the only published paper (gated) to have studied this episode, interest rates on T-bills rose by 60 basis points. The authors assert that this increase was permanent (or at least persistent). This sounds big; in fact it constituted a six percent increase in borrowing costs--rates on T-bills rose from 9.1 to 9.7 percent. Were this to occur today, a six percent increase of the current 3-month T-bill rate would raise borrowing costs by... nothing (.06 x 0 = 0). This seems anti-apocalyptic.

A similar analysis of the 1995-96 standoff (ungated) reached a very similar conclusion. This study explored whether bond markets anticipated default in the spring of 1996 and thus imposed a risk premium. They find evidence of such a risk premium, but find also that the premium was largest for 3-month bills, smaller for 6-month bills, and non-existent for 12-month bills. The authors concluded that these findings might suggest that although "the market is concerned about the budget crisis in the near future, it also believes the standoff will not stretch out into the distant future" (page 262). In other words, markets might distinguish between a sovereign default caused by massive over-borrowing and collapsing export revenues (where the likelihood of being made whole is zero) and a technical default by the United States (where the likelihood of not being made whole is zero).

What seems to have happened in the past, then, is that political deadlock in debt ceiling negotiations leads markets to charge slightly higher interest rates when they lend to the US government. This seems reasonable. It also seems quite far from a doomsday scenario. Does this very slender evidence speak directly to current affairs? One can't know. However, markets don't seem to be spooked, at least not by the American situation. One would think that we would see some sign of impending doom. Higher interest rates at auctions? Sharply falling bond prices in secondary markets? No, instead we see negative yields on US T-bills of less than 3 months. Markets are so concerned about US default that they are willing to pay the US government for the privilege of holding US debt. Perhaps we are still too far from the event to expect a market reaction. Perhaps as Treasury's cash runs down and agreement remains elusive, markets will begin to stir and the world will collapse. Only time will tell.

In the meantime, we might contemplate the purpose of a reputation. The US can borrow at negative interest rates not because it never faces obstacles to repayment, but because it makes its creditors whole in spite of any obstacles it faces. And it might well be precisely this determination to make its creditors whole that prevents a bargaining stalemate over the debt ceiling from precipitating financial doomsday.


Kindred Winecoff said...

Good post and welcome back. Few things:

1. If our options are between "doomsday" and "pay more for borrowing while we have a $1tn/year primary surplus" I'll still take "raise the damn debt ceiling please". Hence, I don't think any discussion of the doomsday scenarios is hyperbole.

2. "One would think we would see some sign of impending doom". Like what? Spikes in prices of gold and other commodities? Check. Ratings agencies flipping out? Check. Wall Street going nuts on the GOP? Yep. Obviously right now bond markets haven't freaked out on Treasuries, but I think that's because they think a deal will get done in which everyone is made whole (as do I) and because if a deal doesn't get done and an actual default occurs, everything goes to hell so nothing's safe anyway.

3. Comparing a technical difficulty in 1979 (in which the full faith and credit of the United States is still promised) to a measured political decision to intentionally default for no good reason in 2011 (in which the full faith and credit of the United States is deliberately eschewed) is apples to oranges, to put it mildly.

4. On your math (0.06 x 0) you should be adding rather than multiplying, in which case it isn't nothing. A 60 basis point increase is a 60 basis point increase, no matter the starting point. I would guess that the 1979 penalty is a best-case scenario, given #3, but even still this would cost billions of dollars. For no reason at all.

5. "In other words, markets might distinguish between a sovereign default caused by massive over-borrowing and collapsing export revenues (where the likelihood of being made whole is zero) and a technical default by the United States (where the likelihood of not being made whole is zero)." This is a key point, and in every prior event where there's been a gov't shutdown everyone (including workers) have been made whole. OTOH, to my knowledge we've never had an event where the legislature simply refuses to pay. And we've never had an event (again, to my knowledge) where the ratings agencies downgraded our debt, thus really straining the banks. We've earned our reputation as payers of our doubts, which is why it could hurt so much if we tarnished it.

6. I also agree on the bargaining stuff. I actually had a post planned along the same lines. It makes sense for the GOP to toe the line. It makes sense for Obama to call the bluff and push for a bigger bargain than the GOP really wants, and it makes sense for a McConnell-type plan to be the equilibrium outcome. But I worry about trembling hands.

Thomas Oatley said...

Thanks. Simple responses here, maybe another post on the bigger points you raise.

1. I don't disagree that raising the ceiling is better than not. One can share that view without thinking that doomsday is likely if we don't.

2. Gold prices? Seems that we ought to see direct evidence in the relevant markets. If people were selling treasuries to buy gold, gold price spikes should be accompanied by govt debt valleys.

3. Comparing 1979 to current is apples to oranges. This seems tautological: doesn't apply because it doesn't apply. Why doesn't it apply?

4. On your math...My point was that 60 basis points was 6 percent of the current yield. So, a 6 percent increase in the current yield would be zero. Not sure why we would expect rates to increase by 600 percent (as would be implied if we added the 60 basis points to current yield).

5. This is indeed THE key point. And it needs further elaboration. Here I will say only that this isn't a debt repudiation nor even an insolvency-based default. No one questions whether the US has the income to service or the intention to service its sovereign debt. What is in question is whether the US government will have the cash on hand to make about $30 billion of interest payments due in August. And even if it doesn't, then this default has zero implications for what happens in September. Yet, the conversation seems to assume that if we fail make a payment in August everyone concludes that we will refuse to pay all of our obligations. I had a hard time understanding the logic that supports this inference.

jdwill said...

Holy Pilloried Pile, Batman!
You mean the debt ceiling was under 1T in 1979? I think I saw somewhere it was about 4T in 1996.
Now on our way to 16T.

Methinks this aspect of the story is being under-reported.

Beyond Hyperbole
There was an error in this gadget




Add to Technorati Favorites