Tuesday, October 27, 2015

Those Fragile, Resilient Financial Markets

by Neil H. Buchanan

As I write this post, the latest news is that there have been secret negotiations between the White House and congressional Republicans that might simultaneously avert a November 3 debt ceiling-related default while also preventing a government shutdown on December 12.  All of this could break down, of course.  [Update: As of 9am Tuesday, 10/27, the deal is reported to be moving forward.]  No matter what happens, Professor Dorf and I will surely weigh in as the story develops.  Here, I want to offer some musings on an issue that is related to the debt ceiling, but that is interesting no matter what the outcome of the current politically manufactured crisis.

In one of the few interviews that I have given recently about the debt ceiling -- where the paucity of interviews is, as I noted in my post last Thursday, caused by an eerie lack of interest by the press in the debt ceiling, even as potential catastrophe looms -- the reporter asked a series of perceptive questions about how financial markets would respond to a full-on debt ceiling-related default.  Alternatively, he also wanted to know how I thought the financial markets would react if President Obama were to agree with the Buchanan-Dorf analysis and issue bonds in excess of the debt ceiling.  For awhile, it appeared that my answers to those two lines of inquiry betrayed an underlying contradiction in the way I think about financial markets.  Seeing why there might have been a contradiction, and then seeing why that contradiction is illusory, might be usefully clarifying.

The first question is what would happen if the government were to default on its obligations, if both sides were to refuse to blink in advance of November 3.  My answer would be familiar to most readers of this blog: the markets would be a bit more upset if the government failed to pay interest on Treasury securities, but defaulting on any government obligation would be a huge deal.  No one knows how bad the damage might be, but it would be very bad, because this is something that has never happened.  Short answer: The financial markets are fragile, and they might break if the government were ever to be untrue to its promisees.

The second question is what would happen if the government, in order to avoid a default on November 3, were to issue "illegal bonds" --  what Professor Dorf and I called "presidential bonds" in our first law review article on this topic, to make clear that the bonds would represent debt that had not have been authorized by Congress -- in advance of the drop-dead date.  My answer was that the financial markets would be upset, but the reason for their being upset would be more political than economic.  A constitutional crisis is always bad for markets, and surely there would be disruption.  But I then argued, again echoing points made in the various Buchanan-Dorf writings and in my book, that the financial markets are very good at dealing with and putting a price on risk.  If the president were to begin to issue illegal bonds, that would be a reason to add a default premium to the returns that Treasuries normally pay (which would be meaningful, because Treasuries have always been thought to be free of default risk), but there is no reason to think that this would be any different from pricing any other risky asset.  A market that can price junk bonds can surely price presidential bonds.  Short answer: The financial markets are resilient, and they can stand up to what the political world throws at them.

As my discussion with the reporter continued, I noted that the pricing of the presidential bonds would become quite interesting, because there would be no way to know which new bonds are presidential bonds and which are fully legitimate bonds.  Why?  On any given day, old Treasuries are maturing, and it is fully legal to roll those over by issuing new Treasuries even when the government is against the debt ceiling, because doing so does not increase the total amount of debt in existence and thus does not exceed the debt ceiling.  So, if there are $100 billion in maturing Treasuries on a given day, and there is also a need for $50 billion in new borrowing to cover the difference between obligations coming due and tax revenues rolling in, the Treasury would float $150 billion in new securities.  Even if one wanted to put an "illegal" stamp on one-third of them, how would one decide which were the good ones and which were the bad ones?  (I cannot find the link, but I am fairly certain that Professor Ted Seto of Loyola-LA is the first person to have described this peculiarity.)  I concluded that those clever financial markets would then paradoxically view this as good news, because if it is impossible ex post or ex ante to separate the good from the bad, that would increase the likelihood that Congress would feel obligated to validate all of it, after the crisis ended.  Resilience!

The reporter then pointed out that, earlier in his career, he had reported from countries like Russia, Argentina, and similar places that have defaulted on their sovereign debts at various times in the recent past.  The amazing thing is that, after such short-term crises, international lenders turn out to have very short memories, and they come right back to those same countries carrying piles of cash to lend.  This reminded me of a lecture that I delivered in Vienna in 2013, where a discussant pointed out that Austria has defaulted on its debt six times since WWII, yet the world kept turning.

My answer to that discussant, in turn, holds the key to figuring out the contours of the fragility/resilience tension in the arguments above.  The fact is that some defaults are truly minor events, where the people who are directly affected are of course unhappy, but the rest of the world barely notices.  On the other hand, some of the other debt crises, such as Argentina's, can be market-shaking events with very real consequences not just for the citizens of the affected countries, but for investors (including pension funds in the U.S. whose clients are middle-class retirees).  As bad as they are, however, these crises have never (so far) caused a global meltdown.

I still think that one of the best responses that I have heard to that question in Vienna was offered by one of my then-RA's, which I reported in the blog post linked above: "Well, isn't it interesting that your country defaulted without destroying the world economy.  Good for you!  Your country is just adorable."

What makes the possibility of a U.S. default on its obligations so scary, then, is that it is not just a default, but a U.S. default.  Global financial markets are uniquely dependent on U.S. Treasuries to grease the skids of commerce, so anything that undermines what amounts to the lynch-pin of global finance is a much bigger deal than other types of defaults.  As Professor Dorf and I have argued all along, the further in advance the president were to announce a plan to view the debt ceiling as non-binding, the easier it would be for markets to treat the ultimate issuance of presidential bonds as a non-event.  (Happily, it would also reduce the likelihood of ever needing to issue the bonds at all, because the hostage-takers would have no negotiating power.)

That is not, of course, to say that there would be no negative consequences at all in such a situation.  That is why it is so important for Congress to repeal or effectively neuter the debt ceiling, or at least to increase it in a timely way as needed.  But the difference between the two possible ways to deal with a crisis -- defaulting, or issuing unauthorized bonds in order not to default -- ultimately hangs on what the financial markets do well versus what they do poorly.  They price risk well, but they are very poorly equipped to deal with the huge unknown of a U.S. government default.

Of course, the craziest members of Congress have their own view of the resilience of financial markets.  As I noted last year, one of those guys has actually argued that the global financial markets would rally in response to a default, because it would prove that Congress had gotten serious about reducing U.S. debt.  Right.  First, there is no evidence that markets are worried about U.S. debt at all, with 10-year Treasuries now yielding just a touch more than 2% (effectively borrowing for free in an economy where the Fed's inflation target is 2%), and even 30-year Treasuries are paying less than 3%.  Second, even if the markets did think that the U.S. debt situation would get out of hand sometime in the next few decades, the markets would want to see evidence of changes in taxes (up) and spending (down).  Defaulting on government obligations is not evidence that sanity has at long last returned to fiscal decision making.

I cannot believe that I felt the need to write that last sentence.  But here we are.


Shag from Brookline said...

So, government bonds, legal and illegal, would be fungible [my second favorite word]?

Neil H. Buchanan said...

Not just fungible. Indistinguishable.

t jones said...

Would there be any reason/incentive for the Treasury to issue series PR ("Paul Ryan") bonds to cover the unauthorized $50 billion - thus making clear which bonds carried the higher risk?

Joe said...

The pressures of the sanity caucus again appear to be enough to make these discussions somewhat academic.

Shag from Brookline said...

Joe, your "sanity caucus" is an infringement of the Marx Bros.' "sanity clause." Fox may pick up on this with its upcoming seasonal "War on Christmas" schtick.

Elon said...

Is there any heightened market risk associated with the presidential bonds you describe other than a risk of a judicial intervention (or perhaps an action by a subsequent administration) that causes the government to stop honoring the presidential bonds in a fashion tantamount to heightened risk of default? (For example by a constitutional court that disagrees with your analysis of the constitutionality of the presidential bonds). If not, couldn't a judge fairly easily fashion a remedy that deals with the basket of "good" bonds and "bad" bonds that you describe? For instance if on Default -1 the government reissued $50 billion dollars of existing debt and $100 billion of new presidential debt, couldn't a judge order the treasury to pay only a third of the coupon on all debt with maturities/rates pegged to D-1? As a result, would an efficient market really have problems pricing in some risk premium associated with the issuance of presidential bonds simply because those bonds were quasi-anonymous?

Hank Morgan said...

I'm pretty skeptical of the president unilaterally issuing bonds under the executive power, and I think you're underestimating the nature of the market reaction to that decision. Ultimately, I think there would be a huge legal cloud over all of that debt, and it would not stave off a financial crisis.

The one idea I've heard that seems much more well-grounded would be for the Treasury to change its rules to allow for the issuance of high-coupon bonds, which would raise more money without increasing the face value of the public debt. As long as the Treasury properly changes its rules to allow this, it would seem to be a legally impeccable approach, and far less risky than your suggestion of "presidential bonds." More detailed explanation here:

I'd be curious to hear the Dorf/Buchanan take on this proposal, which seems like the most sane thought I've heard to date, should Congress ever be so insane as to fail to raise the debt ceiling.

David Ricardo said...

You can't fool me. There ain't no sanity caucus.