« The exogeneity-plausibility trade-off | Main | The macroeconomics of double pole dancing »


Feed You can follow this conversation by subscribing to the comment feed for this post.

I have been arguing the "downgrade should be a stimulus" point of view vs. Noah Smith and others on Twitter. But here I'm going to be a devil's advocate and make the argument that I think he should have made. It's all about maturity (and in this case I'm not talking about the maturity, or lack thereof, of the freshman Republicans in Congress). Although the downgrade theoretically applies to all US debt, nobody except a Beckite lunatic can seriously believe (now that the ceiling deal has passed) that there is any significant risk associated with short-term US debt. So if one "believes the downgrade" (or has come to the same conclusion independently), one appropriate way to respond is to shorten ones maturities of US debt. Since the zero constraint is binding on short-term rates but not long-term rates, this action is unambiguously contractionary unless the Treasury or the Fed responds by changing the mix of outstanding maturities.

Regarding China, I do think it is good for the US if China stops buying Treasuries. Indeed, I have been assuming that most economists (at least salt water economists) would agree, and I'm always surprised when economists make statements that seem to disagree. I can see a couple of counterarguments, but I don't find them persuasive. First, a sudden change in Chinese policy could be disruptive, and the chaos produced by the disruption could outweigh the direct stimulus effect. Second, the price level may be disproportionately affected by the exchange rate, and the Fed could react to an exchange rate shock by tightening to an extent that would outweigh the direct stimulus effect.

Andy: Good comment.

On Andy's point: the S&P downgrade only applies to l.t. debt. Also, to Nick's question, the S&P explicitly cites the loss of principle from inflation as a form of default.

The two issues above are related. Nick makes the point that increases in expected inflation should raise RGDP and therefore be seen by S&P as a positive, not a negative. Why would S&P think otherwise? Let's leave aside the possibility that they are simply unsophisticated. A downgrade and the prospect of higher (and more volatile) inflation may either: 1) cause investors to shorten portfolio duration; or 2) force them into risk assets. The two responses have very different implications for AD. I notice that many economists seem to assume the "buy risk assets" response as a given. In high inflation economies, however, the real term premium on local currency sovereign bonds is high and rises on negative economic news. The reason is that investors shorten duration when they expect yet more monetary stimulus, more inflation, and more loss of purchasing power.

So here's a question: what determines whether investors choose 1) or 2)? Obviously if investors perceive that higher inflation drives real growth, then 2) will dominate. If they believe, instead, that the impact on real growth is uncertain while the impact on real purchasing power loss is certain, then they will shorten duration (choose 1)). The question is not what economists believe or what their models indicate; the question is what the majority of investors will decide. How do we know this in advance?

The question I am more interested in is what does the S&P know that the market does not know? When 5-year TIPS at constant maturity are negative, you must have a very good reason for arguing that the USA isn't AAA. You said it yourself, it's your 'crude and very amateur political analysis' that leads you to that view. There are however a lot of people out there with a 'very crude and amateur political analysis' that are willing to pay money to park their funds into TIPS. How much better is your analysis than theirs? Not to mention the people with a more sophisticated political analysis who are willing to pay to hold TIPS.

I mean I do understand that view, however when those views are indistinguishable in quality, it seems a bit odd to change the rating from AAA to anything lower. They have basically the same information as before yet they come to a different conclusion. It looks more like marketing to me than anything else. "Look at the S&P, they're independent enough to downgrade even the US, surely their rating now will be fair".

How is this, "and in this case I'm not talking about the maturity, or lack thereof, of the freshman Republicans in Congress," even a remotely constructive remark? It does not make you look well informed either.

"I do think it is good for the US if China stops buying Treasuries." If buy stop you mean, stop accumulating, that's already happened. They've been net sellers of US govt debt for the past five months. Still this process is going to be a slow unwinding since about ninety percent of their money base is backed by treasury notes.

What I find interesting about 'AAA' ratings is how rare they are. There are now only four US companies rated AAA: Automatic Data Processing, Johnson & Johnson, Microsoft, ExxonMobil. Ten years ago there were eight, and there used to be about fifty.

Berkshire Hathaway lost it in 2010
GE lost it 2009
Pfizer lost in 2007
Bristol-Myers Squibb lost it in 2002

Meanwhile, the default rate on AA+ corporate bonds is zero.

I think the S&P downgrade is partly political. They were criticized heavily for not anticipating the mortgage loan debacle and now have decided to make sure they are on the ball with U.S. debt. They are being extra cautious as well as swinging back at the politicians.

I think the S&P downgrade is partly political.

If you put your triangulation hat on for a moment, there has been a lot of active pressure to eliminate the rating agencies. For instance, just a few days ago the following piece appeared http://www.slate.com/id/2300572/

I wonder how that changes in light of the downgrade. Or does it?

If the US removes the regulations that pertain the the views of the ratings agencies what objective measure will they put in their place? The ratings agencies were probably used an objective source because there wasnt much else that was performing as well as them. If the ratings agencies are still relatively accurate (they may have missed the housing collapse but so did everyone else) then removing them from regulations just seems to be the US governments way to remain unaccountable. If we remove too many checks and balances and rely solely on market numbers then we set up the possibility to blow up a T-bill bubble that no one will see coming. Doesnt really sound like a good idea to me.

I can't find a reference, but my recollection was that the 1992 Canadian Govt. debt downgrade was only for foreign-currency (not in CAD) denominated bonds, of where there were few. Anyone know for a fact?

The 1992 downgrade was an attack on the Rae government, led by the federal government who requested the downgrade.

@Ian The ratings agencies are not relatively accurate. Read "The Big Short". Also, anybody who saw the Youtube video of the Case-Shiller index translated into a roller coaster ride knew housing was overpriced. Of course, anyone who said that at the time was not considered a Serious Person.

"But when ratings agencies grade sovereign bonds, for printer governments, are they talking merely about the likelihood of legal default, or do they also consider the likelihood of the real value of those bonds falling due to inflation?"

Standard and Poor's Methodology is easy to find; go to their home page (http://www.standardandpoors.com/home/en/us) and click on "Sovereign Government Rating Methodology and Assumptions" on the right under "Highlights". It states "The sovereign rating methodology ... addresses the factors that affect a sovereign government's willingness and ability to service its debt on time and in full." In other words, it ignores inflation.

Note that bond raters like S&P carefully distinguish between "foreign currency" and "local currency" debt. The latter is never rated lower than the former and the former is usually downgraded before the latter. Note that the Business Week graph that you linked to doesn't indicate whether the downgrades were for the local currency or foreign currency debt.

This, of course, is where America's "exhorbitant priviledge" comes into play; most sovereign governments with lots of debt have lots of foreign currency debt, so they can't print their way out of trouble. (I think Japan is an important exceptions, with lots of JPY debt held domestically.) The US, however, has no trouble convincing foreigners to buy USD debt; in fact, the more uncertainty in world markets, the bigger the premium investors are often willing to pay for USD debt.

Let me adjust that first statement somewhat. The downgrade was part of a campaign to put deficit reduction at the top of the political to-do list. It also was particularly damaging to Ontario at the time. Which was the main course and which was gravy is not obvious.

Nick, you'd have fun comparing your crude political bond-rating analysis to the pros; the S&P methodology document I mentioned above has a nice schematic on p. 4 and a long discussion. Political Analysis is one of the five pillars on which they base their judgement. You can imagine that given (a) the stated willingness of some Tea Party members to bring about a default, and (b) their ability to effectively veto Republican proposals that do not mesh with their views, and (c) Republican control of the House and effective veto-power in the Senate, S&P would not be impressed.

Jim; do you have a source for the statement that the federal government requested the downgrade in 1992? and whether the downgrade was for all debt or just foreign currency?

BTW, "AAA" ratings are not that rare for sovereigns; S&P note that about 15% of all sovereigns that they rate make it. Corporates are a whole different kettle of fish.

"... my recollection was that the 1992 Canadian Govt. debt downgrade was only for foreign-currency ..."

I don't think that's quite right. The overseas issues were the most strongly affected, but they were mostly denominated in CAD. Confusing the question is that not all CAD eurobonds are issued by the Canadian government or even by Canadian corporates; in the early 90's, many international corporates had issued CAD eurobonds for the sake of lower interest rates. Prices of these bonds fell noticeably when Canada was downgraded. Canadian government bonds also fell, but by a much smaller amount.

"... wouldn't a downgrade reduce the demand for government bonds?"

For the government of Lower Slobovia, yes. But for the US government, conventional thinking is that it will just induce a preference for shorter maturities, as other commentators have mentioned. We shall see.

There is a large and somewhat contradictory literature examining whether ratings changes contain information. The answer appears to depend on many factors: whether one is considering upgrades or downgrades, bond or CDS spreads, sovereigns or corporates, liquid or illiquid issues. If you have the patience you can find much of this literature through the references of this paper by Hull, Predescu and White (which is very interesting itself):http://www.rotman.utoronto.ca/~hull/DownloadablePublications/HPWPaperonCDSSpreads.pdf.

"... are they talking merely about the likelihood of legal default, or do they also consider the likelihood of the real value of those bonds falling due to inflation?"

In theory, they account for inflation, as David Pearson has mentioned. In practice, though, that seems doubtful: the US retained its rating in the postwar period even as it inflated away its wartime debts. Perhaps only hyperinflation is really accounted for. And note that domestic and foreign currency obligations are rated separately, as Simon van Norden has mentioned. Here is S&P's own explanation of how they arrive at their sovereign ratings: http://www.investinginbondseurope.org/uploadedFiles/Learn_About_Bonds/What_You_Should_Know/Market_and_Economic_Influences_on_Your_Bond_Investments/Sovereign%20Credit%20Ratings%20Primer%20-%20SP.pdf.


I had thought that Moody's/S&P specifically mentioned inflation risk in their June/July "creditwatch" warnings, but now I cannot google the exact news item. If I dig it up, I will post the link.

Simon, there is no reason to speculate on whether the political culture itself drove the downgrade. Remember, S&P's warning letter to the treasury was pretty explicit:

- A "clean" increase in the debt ceiling (without deficit reduction measures) such as Obama initially demanded would result in a downgrade.
- They further stated that they would likely downgrade unless those deficit reduction measures included structural reforms to entitlement programs.

The message from S&P was clearly stated; it's a shame the details of their message weren't more prominent in the media reports, but this outcome was obvious without reaching to the partisan blame game.

For the sake of clarity, private foreign issuers of bonds denominated in CAD and sold in Canada are known as Maple Bonds.

S&P and later Moody's downgraded Government of Canada debt instruments denominated in foreign currency other than the CAD. These consist of Canada Bills for terms of 271 days or less and Canada Notes for terms longer than that. Both series have an issue limit of $10 Billion each. CAD denominated Government of Canada Bonds ("Canadas") remained rated AAA or Aaa by all agencies.

Canada Bills/Notes have been issued in US Dollars and Euros and these were the debt series which suffered the downgrade in 1992.

The fact that Maple Bonds also took a hit is a result of the fact that private debt issues are rated for quality in relation to their underlying sovereign currency issuer.

The most bizarre commentators have been those implicitly arguing that S&P should just pretend the federal government deserves a AAA credit rating for the sake of the world economy. We just came out of a financial crisis that was a result, in large part, of pretending that mortgage-backed securities were AAA when they weren't, and we can all see how well that worked out. People amaze me sometimes.

Determinant: "CAD denominated Government of Canada Bonds ("Canadas") remained rated AAA or Aaa by all agencies."

That's an important bit of info. You sure about that? You got a source? Any idea how far back that is true?

"Maybe we have had separatism as a worry, and the US hasn't, at least since their civil war?"
Pierre Fortin adressed that in his 1996 CEA Presidential address "The Great Canadian Slump"
CJE 1996 vol 29 pp 761-87
His answer: no effect.

Jim Rootham, Simon van Norden
I am on vacation so no personal library but in the early 00's , a guy from the economic section at La Presse (Rudy Lecours IIRC) wrote a book on the Great Canadian Slump. He quoted an anonymous-by-request investment banker in NY who said to him (my appoximate translation from memory):"Canada is the only country I ever saw asking to downgrade its bonds." Given where the budget cuts were directed and the well-planned press campaings, yes it was a political inside job, an internal coup if you will...
If anyone has info on that book, please post.

It's the double secret plan for recovery.

1. Behave like irresponsible morons for a couple of months.
2. Say: Hey everybody! Look! We suck! Stop buying our bonds (and hoard someone else's cash)!
3. Devaluation and ... Profit!

I like James Kwak's suggestion: the rating agencies should just throw in the towel on the big economies. They have nothing to add that the market hasn't already priced in.

The current state of Canada's credit ratings can be found in Table one of this report from the Department of Finance: http://www.fin.gc.ca/dtman/2009-2010/dmr-rgd1001-eng.asp.

That table is unclear, but the upgrades refer to our foreign debt.

The primary sources on Canada's downgrade in the 1990's are a bit stale as they are pre-internet, but here is a good summary http://www.activistpost.com/2011/08/could-us-regain-aaa-rating-if-it-loses.html

You could verify this by a call to the Commerce Faculty or perhaps some friendly contacts at the Bank of Canada.

I also remember Alan Rock posting a bit on this era and he said specifically it was our foreign debt, not our domestic debt that got the downgrade.

From a 30 July article in the WSJ by TOM LAURICELLA

"Canada, too, suffered a multiple-stage downgrade over a period of several years, and analysts note that its finances at the time were in comparable shape to those in the U.S. today.

In October 1992, when S&P cut Canada's foreign debt rating by one notch from triple-A, there was little impact on Canadian markets. But when Moody's followed in 1994, 10-year note yields rose 0.45% over the next month and stocks fell 6%, according to data compiled by RBC Capital Markets.

Then, in April 1995, Moody's downgraded Canada's key sovereign rating by one notch from Aaa. That day, there was little reaction in the bond, stock or currency markets."

Moody's downgraded Canada's domestic and foreign currency debt in April 1995. You can read Paul Martin's (then Finance Minister) official press release responding to their announcement at http://www.collectionscanada.gc.ca/webarchives/20071122094738/http://www.fin.gc.ca/news95/95-033e.html

Good comments guys. Thanks.

Has anybody heard of Japanese yen being a safe haven? Does anybody know the S&P rating of Japan? Does S&P know the debt/GDP ratio of Japan?

The bond market is not on the "government bonds are risky" bandwagon. I think this is one of those cases where there's a huge disconnect between what sophisticated people privately believe and what "everyone" is saying publicly.

Regarding currency devaluation: this is the "virtuous austerity" option. Hardship but with low unemployment; quite different from the forced-idleness version of austerity. This is the path that the peripheral Euro countries should take (leave the Euro, which would force them to run a trade surplus).

It is quite clear political differences are central to their decision

"The downgrade reflects our view that the effectiveness, stability, and predictability of American policymaking and political institutions have weakened at a time of ongoing fiscal and economic challenges…

The political brinksmanship of recent months highlights what we see as America’s governance and policymaking becoming less stable, less effective, and less predictable than what we previously believed. The statutory debt ceiling and the threat of default have become political bargaining chips in the debate over fiscal policy. Despite this year’s wide-ranging debate, in our view, the differences between political parties have proven to be extraordinarily difficult to bridge."

No, nothing partisan about that ;-)

The one ratings downgrade took place in the early '90s and IIRC at that point the ratings agencies were merely pointing out a structural imbalance between expenditues and revenues was reaching a breaking point. The risk they were citing was Canada tipping into a set of countries where default or deflation was the norm. The solution, as it turned out, was relatively simple, involving tax rises and job cuts and a nice boost from commodity prices if memory serves.

The US situation is not that dissimilar to Canada's two decades ago. They have a structural deficit that is easily solved by adding tax revenue and concentrating expenditures on productive means. Canada had the impetus to take on the solution, the US looks to be more of a wildcard. Bond markets have indicated that the US is nowhere near its tipping point. So S&P may be correct about the underlying situation but its timing is way off.

When confronting the US-China bond buying matched pairs, I would read Michael Pettis's blog posts. To quickly summarise, the US reducing bond issuance can be simply viewed as a direct result of a smaller current account. If the US trades less with China, China's US sov purchases will decrease as well. Foreign bond purchases is the mechanism for trade flows, not the raison d'etre, which we should be looking at.

"...In the early 1990s, according to Canadian investigative journalist Linda McQuaig, Canadian corporate executives encouraged ratings agencies to threaten a downgrade of their nation’s credit as an inducement for cutting social spending and lowering high-end tax rates. It worked..."

Quoted from


StatGuy at Baseline Scenario argues that a downgrade is contractionary because safe assets are Giffen goods (which doesn't seem entirely implausible: for example, take the case where people want to maximize their chance of survival).

I believe his argument would also suggest that higher expected inflation would be contractionary, although QE could still help by reducing duration risk.

Andy: I read that, not thoroughly. Didn't seem convincing to me. A necessary condition for a good being Giffen is that it be an inferior good. Do poor people own a larger stock of safe assets?

...Masquerading as objective think tanks and research organizations, the ratings agencies act as lobbyists for banks and underwriters by endorsing a race to the bottom – into debt, privatization sell-offs and an erosion of consumer rights and control over fraud. “S&P was aggressively killing mortgage servicing regulation and rules to prevent fraudulent or predatory mortgage lending,” Stoller concludes. “Naomi Klein wrote about S&P and Moody’s being used by Canadian bankers in the early 1990s to threaten a downgrade of that country unless unemployment insurance and health care were slashed.”...


The comments to this entry are closed.

Search this site

  • Google

Blog powered by Typepad