Is there a government bond bubble?

by on August 23, 2010 at 4:25 pm in Economics | Permalink

Here is a symposium over at The Economist:

It is better labeled a bubble in government spending (Viral Acharya)
No, there is a shortage of safe assets (Ricardo Caballero)
Bond yields are probably appropriately low (Stephen King)
Probably not, but approach low-probability risks carefully (Tyler Cowen)
Yes, and it's huge (Laurence Kotlikoff)
It's possible, but there is not enough evidence to be sure (Paul Seabright)
No, growth and deflation concerns have grown sharply (John Makin)
Current adjustments will reduce the risk of a repricing shock (Harold James)

Andrew August 23, 2010 at 4:38 pm

Isn’t “No, there is a shortage of safe assets” just a different way of saying “Yes.”?

anon August 23, 2010 at 4:55 pm

Andrew, a “bubble” is a pattern where the price of an asset rises following an exponential curve until it collapses at some unpredictable point. Bubbles in this sense can be detected fairly easily, since the expectation of “exponential growth and sudden collapse” will be reflected in derivative markets (such as options) on the original security.

A shortage in safe assets would create a stable premium for “safe” assets relative to “unsafe” ones, not a bubble pattern.

Michael F. Martin August 23, 2010 at 5:09 pm

Wow. Even Tyler basically hedged his positions. I guess uncertainty is quite infectious.

Here’s food for thought. 10-year notes are at their lowest since the 1950s right now:

http://research.stlouisfed.org/fred2/data/GS10.txt

The late great Ken Fisher used to consider the inverse P/E ratio in comparison to 10-year notes to decide whether equity was over/undervalued. At current 10-year rates, the average stock is undervalued by almost half.

I can’t understand why money continues to flood into bonds except that I guess people are sick of seeing unrealized depreciation in real estate and stocks.

Yosh August 23, 2010 at 5:10 pm

A number of these answers seem to be focusing on different time frames rather than disagreeing. Those focusing on the short term are saying there’s no bubble because there are no other safe places to put assets and therefore the pattern is rational.

Those focusing on a longer window are say that these investors are ignoring the instability that government fiscal policy will inevitably create, ultimately having a strongly negative impact on bonds.

Neither of these answers contradicts the other. They are just different time frames.

Bill August 23, 2010 at 5:32 pm

It is interesting that the commentators looked at this solely as a bubble in US government bonds, but then ignored that it must also be a bubble in corporate bonds and non-US government bonds as well given the small differentials between these other bond instruments. If there is a bubble in corporate bonds, and there is a bubble in non-US government bonds (Germany), this might be telling you that other forms of investment that throw off residual income (stocks) are more risky. Could this be a signal there is a bubble in stocks, which are mispriced relative to the risk in bonds.

Bill August 23, 2010 at 8:26 pm

@anon, I don’t disagree that bubbles can arise in markets. In fact, my comment was a point to the contrary–that they do occur–and, even in markets where participants should know better. And, markets that are efficient.

Charlie Plott of Caltech–in my mind, one of THE best economists of our time, who should have gotten a Nobel along with one of his collaborators–has lecture which discusses bubbles and shows some of the metrics and experimental work on how bubbles develop, even in markets where there is common knowledge of values. This is from the Cardiff lectures in New Zealand. I believe this lecture includes one of his bubble experiments involving economists. Look up Plott’s guest lecture in New Zealand. Just google Charles Plott Caltech bubbles and go the UCTV lecture. The lecture has several demonstrations on how bubbles develop even in markets with common knowledge of asset values.

What I’m interested in is the comment by one of the commentators–King–that there is not a bubble in bonds, but rather a bubble in the dollar. Now, what is interesting here is the actions of trading partners purchasing the dollar to devalue their currency relative to the dollar. This could lead to a bubble in US bonds–for investors that were rational if they were pursuing a goal of devaluing their own currency–but not for others. So, to the extent we talk about a bubble in bonds, maybe we are doing a good thing in a game theoretic sense: we are communicating to the Chinese that it would be wise to discontinue purchases of US bonds solely for the purpose of devaluing their currency. Game of chicken anyone.

anon August 23, 2010 at 10:15 pm

Bill, I agree that bubbles could develop even in “efficient” markets with common knowledge of asset fundamentals. But many commentators seem to honestly think that efficient markets cannot possibly develop identifiable bubbles, or (conversely) that the existence of bubbles somehow implies that markets are inefficient in other ways as well.

These claims are inaccurate, although more nuanced claims could be defensible: for instance, it is almost certainly the case that bubbles develop more easily in markets where it is difficult or impossible to sell the asset short, or to exchange derivatives such as options on the asset (see the market for structured finance in the 2005-2008 bubble).

Note that e.g. the options markets in energy commodities in 2005-2008 correctly forecast the possibility that a bubble would pop, but prices stayed high because of concerns that future prices could spike even higher.

Yancey Ward August 24, 2010 at 12:13 am

In other words, bond yields probably tell you nothing about how or if stocks are over-valued.

Andrew August 24, 2010 at 6:47 am

I use the broad definition of ‘irrationally overpriced’ to describe a bubble. Although, that could even be broadened to ‘overpriced.’

But, rather than symantics, specifics may be easier. The NASDAQ lost about 3/5ths of its value reverting to the mean trendline. Just about everyone accepts that as a bubble. I know squat about bonds but if the duration of Treasurys is 8 then a 60% loss would accompany a 7.5% rise in interest rates. Is this inconceivable? I would even say that a bond bubble wouldn’t have to be as dramatic as a stock bubble since people consider bonds safer.

susie casalou August 24, 2010 at 9:38 am

I am an ignorant person regards to stocks and bonds. My sister scalled me this morning and has asked me about government bonds. Is it really safe to invest in our government bonds rather than stocks or any other types of investment?
Please someone kindly provide me with accurate answer not a personal opinion.

libert August 24, 2010 at 10:17 am

It’s very hard to have a bubble in bonds for the simple reason that bond prices have a ceiling. That is, no one is going to pay more than the face value (plus coupons) of the bond. If you did so, you would be getting a negative nominal interest rate, in which case you’re better off holding cash.

There are no similar ceiling on stocks or commodities, since the potential upside is unlimited. It’s possible to get irrationally enthusiastic about the prospects of a company or the demand for a commodity, but what idea could possibly justify thinking that a debtor will pay you more than the contract requires? (Caveat: if interest rates are high, and you expect them to fall, then you could consider it possible for your bond’s price to rally. But rates are pretty low right now, making that ceiling pretty binding).

Also, you can short bonds, so that can’t be an explanation for why a bubble in bonds is theoretically possible.

So that rules out a bubble. The question then becomes whether bonds are somewhat overpriced (not a bubble, just somewhat overpriced). That depends on your expectations about future interest rates (high interest rates coming soon? Then maybe bonds are overpriced), which in turn depends on your expectations about the economy (if you expect a robust recovery, then interest rates will increase. If you expect a slow, dragging recovery, then it’s hard to make that claim).

spencer August 24, 2010 at 11:42 am

Every bull market is not a bubble.
But you would not know that reading the popular press.

Ryan Vann August 24, 2010 at 1:15 pm

“I agree, using the word “bubble” seems incorrect, as opposed to simply “overpriced”.”

Indeed.

The Money Demand Blog August 24, 2010 at 5:27 pm

Many economists think that bubbles are harmful and that the Fed should actively lean against the bubbles. Well, the bond bubble is also very harmful, as bonds are pricing in extended unemployment and below-trend inflation. Bernanke should print more money ASAP in order to burst the bond bubble.
More here (Krugman and the bond bubble): http://themoneydemand.blogspot.com/2010/08/krugman-and-bond-bubble.html

Bill August 27, 2010 at 8:25 pm

This discussion of whether there is a government bond bubble is to hide the fact that the Fed will be raising its inflation target, and institute QE, thereby causing the bond market to drop.

It is easier to say that there was a bond bubble than to say bonds tanked because of new Fed policy.

Blame the victim.

Comments on this entry are closed.

Previous post:

Next post: