Comments on: GDP bonds are a really bad idea, part 3 A slice of lime in the soda Sun, 26 Oct 2014 19:05:02 +0000 hourly 1 By: traducator romana daneza Mon, 29 Sep 2014 13:57:35 +0000 If some one needs expert view regarding running a blog then i advise him/her to pay a quick visit this weblog, Keep up the fastidious work.

By: handleym Tue, 28 Feb 2012 07:32:18 +0000 ““As a result, if you’re a risk-averse person who wants a perpetual US government security and your discount rate is say 3%, then the expected value of a singe Trill is actually infinite.”
This statement doesn’t make sense – discount rates should differ based on the riskiness of the cash flow they are discounting, and 3% is way too low for an equity-like instrument like a GDP bond. ”

It’s worse than that. It’s the same as valuing ecological assets — there is uncertainty in the appropriate “interest” rate, in the one case the relevant discount rate, in the other case the relevant growth rate. This uncertainty is extremely important because of the projection to infinity. If your range of possible valid values for this rate includes 0, then you get strange things happening that depend on exactly how you take your limits.

The bottom line is that, in a context of the real uncertainties of the situation, projecting all the way to t=infinity makes no sense — all that makes sense is to project as far as you are confident in projecting and then make a reasonable assumption about what happens after that. There are various reasonable assumptions one might make, but plenty of them do not value such a financial instrument as having a price of infinity.

By: handleym Tue, 28 Feb 2012 05:45:20 +0000 “But Bob Shiller is much more ambitious when it comes to such things”

The value of these things depends on how you want to use them, surely?
For example, I forget the details, but I seem to remember that part of John Geanakoplos’ social security plan involved this sort of bond.

Geanakoplos’ plan was to defuse the Republican claim that “your social security money doesn’t REALLY belong to you and can be taken away any time the government wants” while retaining the security of Social Security (none of the volatility of the stock market, and indexing to the general wealth of the nation). There are various additional bells and whistles to solve various specific problems, but the large scale plan, as I recall, involved “individual accounts” — ie a specific record in a ledger somewhere of what you are owed — but money spent against GDP bonds rather than in the stock market or some other financial market.

By: TFF Thu, 23 Feb 2012 21:55:54 +0000 “Governments issuing GDP bonds or inflation-linked bonds have an incentive to manipulate statistics, ala Argentina.”

You don’t think that happens here?

I thought about TIPS as well. Like GDP bonds, they are fundamentally pro-cyclic in their payout (they pay very little interest in a deflationary environment), yet still increase in value in a recession (due to flight to safety and falling real interest rates).

Could also argue that inflation will dominate real GDP growth over the coming decades in both the US and Europe. Technological innovation will have to fight a strong demographic headwind to make any progress.

Which brings me back to the top — does anybody SERIOUSLY believe that the official CPI represents the on-the-ground reality? Supposedly we saw just an 8.3% *cumulative* increase in the CPI from June 2007 to December 2011. My impression would be something closer to 15%.

Of course, that would make the economy look like it is in pretty bad shape. Much happier to pretend that boom times are here again…

By: revelo Thu, 23 Feb 2012 20:10:11 +0000 30-year TIPS already have some of the characteristics of bonds that pay a percentage of nominal GDP. Commenters above have already addressed some of areas where you are confused about this issue. You did get one thing right, which is that measuring nominal GDP is not an exact science. Same problem as with CPI-inflation which drives TIPS. Governments issuing GDP bonds or inflation-linked bonds have an incentive to manipulate statistics, ala Argentina.

Before we experiment with perpetual GDP bonds, I think we should see how perpetual TIPS work, since that is a more natural evolution from what we already have.

By: engineer27 Thu, 23 Feb 2012 17:59:27 +0000 I suggested something like this during the US debt ceiling showdown. The Treasury was prevented by law from issuing any additional debt, but there was no law against issuing equity, right?

Dividends are certainly attractive to investors, and would get a better offering price and subscription to the IPO. But they certainly aren’t essential. Investors would have to think about what “a claim on the future profits” of country X really means; but they would still come up with a value. The Government of X would get some operating funds, without being tied to a stream of future payments.
Any subsequent issues, of course, would dilute current holders. So countries would not be able to return to investors every year like they can with debt issues.

Also, no one has yet mentioned Gilts — perpetual obligations of the UK government. The coupon on Gilts is periodically raised by parliament.

By: TFF Thu, 23 Feb 2012 14:42:22 +0000 “If the Treasury wants to see the results of a prudential experiment, it can conduct one itself just by doing a small issue.”

Not a bad idea, but you might see different dynamics with a small float than if you have a large, active market. Even with TIPS, I get the impression that the market is noticeably less liquid than with conventional Treasury bonds.

By: Th.M Thu, 23 Feb 2012 14:09:49 +0000 The elasticity of tax revenue (which will be used to actually pay those dividends) to GDP is not one (and can vary quite a bit) so the idea needs refining, even if the principle is interesting.

By: Greycap Thu, 23 Feb 2012 13:05:16 +0000 I am impressed by the quality of most of the comments here. Your post is not completely useless if only because it has drawn such good commentary.

I have only two things to add to them. First, your worries about measurement error are misguided. Are these errors unbiased? In that case they will nearly completely cancel, requiring only a small convexity adjustment to compensate. Are they biased? Then a deterministic adjustment will compensate.

Second, there is no need to cower fearfully, waiting for some bold private company to blaze the trail. If the Treasury wants to see the results of a prudential experiment, it can conduct one itself just by doing a small issue. No matter how bad these things are compared to straight bonds, they just can’t be very bad in absolute terms if there aren’t very many of them.

By: NYC_Economist Thu, 23 Feb 2012 13:03:57 +0000 Felix, it seems odd to write a whole post about GDP bonds without discussing Argentina, which issued GDP warrants as a sweetener during its own restructuring. These warrants have been trading since 2005 and therefore provide an obvious reference point for any discussion on the topic.

The Argentine experience is supportive of your argument that GDP bonds are not an especially promising means of borrowing for most countries. These instruments are necessarily complex, with payouts contingent on both the level and growth rate of GDP as well as inflation (via the GDP deflator). Perhaps because of this complexity, Argentine GDP warrants have persistently traded at very cheap valuations relative to reasonable assumptions about future growth. While this has made them very appealing as an investor, it obviously reduces their utility as a source of funds in the future.