Ultralong bonds?

[wiggin]

There have been rumblings for a while that Mnuchin might try to inaugurate offerings of ultralong bonds (~50 year maturity, likely) in order to help pay for infrastructure spending. Bloomberg has a nice piece summarizing it today (linked below), but the gist is that the government can get very long maturities without necessarily sacrificing much interest, which will help pay for big spending on capital projects. Detractors argue it can reduce liquidity for both the ultralong bonds and the 30 years it will likely partially replace, especially since such long maturities will likely be held by buy-and-hold investors like insurers and pensions. It also will cost more, even if the increase is largely minor. More broadly, they argue that the US Treasury market has been so deep and so liquid because of predictable and regular issuance, and throwing in highly granular auctions of these ultralong bonds would damage that reputation.

I'm curious what you folks think. On the one hand, it seems like a tempest in a teapot to me - will an illiquid ultralong market really affect the rest of the Treasury market? Will it be that bad to have a corner of the market that caters to the needs of long-term liability investors while simultaneously locking in low interest rates for infrastructure investment? On the other hand, I've already seen concerning signs about liquidity at much shorter maturities - there's a massive amount of demand for safe assets nowadays, especially for capital buffers and collateral, and it's only going to get worse as the population ages. As more and more of the available bond supply is grabbed by buy-and-hold investors along with central bank balance sheets swollen by QE and corporate holdings skyrocketing, liquidity of certain critical products may indeed become a problem. That in turn would threaten the entire edifice of US debt markets.

How worried should we be about liquidity in these markets? It seems silly at first given just how deep and liquid the market currently is, but these precise strengths have dramatically increased their attraction to the market and exacerbated these concerns. A liquidity crunch in Treasuries would be absolutely horrific to imagine.

I know other countries have experimented with ultralong bonds in recent years, but the Treasury market is unique so it's hard to determine how generalizable their experiences might be. I also think it's interesting that sometimes the US has taken a pretty big financial hit in order to achieve a specific monetary goal - the article mentions TIPS as an example, and there are others. Structuring bond issuance to make healthier and more stable markets isn't the main mission of the Treasury, but it might be worth it even if it costs a bit more. I'm inclined to be cautious about ultralong issuance.

https://www.bloomberg.com/news/artic...tra-long-bonds

[RandBlade]

The one thing there is not a shortage of is Treasuries. Yes there is a lot of demand but due to the persistent failure to manage the budget there is also a lot of supply.

I believe and may be mistaken that the UK has one of the highest proportion of long and ultra long bonds and has done for quite a long time, it is a long standing policy and not a modern experiment. This, along with having our own central bank of course, was a major advantage to our economy in recent years. The UK ought to have been one of the worst affected nations by the credit crunch: highly exposed to the financial sector, our deficit ballooned to worse than Greece's, our debt ratio shot up etc but one problem we never faced was difficulty borrowing at very low rates despite all that. One reason, other than QE, was that due to our long portfolio of long and ultra long bonds only a relatively small proportion of bonds had to be reissued and rolled over at any one stage. This helped us absorb the credit crunch in a way other comparable nations struggled to do so.

Now of course there is a world of difference between Treasuries and UK bonds. But the precautionary principle seems to me to be the same. A credit crunch that affected US Treasuries would be cataclysmic and having ultra long bonds to ensure there is a regular but not overwhelming supply seems a good idea to me. Nobody I think is proposing that all bonds should be ultra long but in a mix these should provide Stability and smooth our markets. Sensible planning should be able to make sure that the bonds are timed in a way that assists with stable supply rather than stunting it.

[wiggin]

The one thing there is not a shortage of is Treasuries. Yes there is a lot of demand but due to the persistent failure to manage the budget there is also a lot of supply.

I believe and may be mistaken that the UK has one of the highest proportion of long and ultra long bonds and has done for quite a long time, it is a long standing policy and not a modern experiment. This, along with having our own central bank of course, was a major advantage to our economy in recent years. The UK ought to have been one of the worst affected nations by the credit crunch: highly exposed to the financial sector, our deficit ballooned to worse than Greece's, our debt ratio shot up etc but one problem we never faced was difficulty borrowing at very low rates despite all that. One reason, other than QE, was that due to our long portfolio of long and ultra long bonds only a relatively small proportion of bonds had to be reissued and rolled over at any one stage. This helped us absorb the credit crunch in a way other comparable nations struggled to do so.

Now of course there is a world of difference between Treasuries and UK bonds. But the precautionary principle seems to me to be the same. A credit crunch that affected US Treasuries would be cataclysmic and having ultra long bonds to ensure there is a regular but not overwhelming supply seems a good idea to me. Nobody I think is proposing that all bonds should be ultra long but in a mix these should provide Stability and smooth our markets. Sensible planning should be able to make sure that the bonds are timed in a way that assists with stable supply rather than stunting it.

Re: shortage of Treasuries, the issue isn't that there isn't enough US debt out there (ha!) but that certain maturities are needed for certain collateral/capital buffers, and those may have limited liquidity under current market conditions. Remember that a huge chunk of US debt is currently owned by the Fed, and that rising capital requirements in the financial industry (a la Basel 3) have also dramatically increased the appeal of US debt. The US is also unique (compared to, say, UK markets) in that much of the demand is not only domestic but also international - since most trades, especially including forex, involve dollars, Treasuries are often needed as collateral even if neither party is actually in the US.

I'm aware of the relatively long maturity of UK debt (though I don't know how long ultralong is), which can indeed insulate the country from transient shock to sovereign debt markets. But that's only really an advantage if you expect a lot of volatility in said markets due to concerns over default/etc. (such as with the eurozone crisis a few years back). Otherwise, long maturities are not specifically valuable unless you want to lock in unusually low rates (such as the current time). If rates are just average or above-average (and you are insulated from concerns over default or currency risk), it makes sense to have more shorter-dated paper to get lower interest rates. Essentially, the UK may have overpaid for their debt to get security they didn't need.

I'm not sure I can parse your last paragraph. I'm not at all concerned about a credit crunch in US markets (meaning a sudden complete lack of demand, causing yields to spike). Far more likely are constrictions on supply, especially at specific maturities. I guess part of this concern is a bit silly; people are worried that demand for ultralong debt will be high enough that you won't have market liquidity, which could in turn reduce demand. So it sorta seems like a self-correcting problem, right? I think the bigger issue, though, is the reputational risk involved. If US Treasuries are ever seen as an unreliable store of easily-accessed value (due to deep and liquid markets), that could indeed cause a credit crunch of massive proportions. It doesn't seem very likely here in particular, but I'm aware that the US Treasury spends a lot of time thinking carefully about how they structure offerings in order to keep the market stable and happy. Ignoring these concerns just because we'd get slightly longer-dated debt doesn't seem like an obvious move for me. After all, they could just shift to offering more 30-year bonds, which are pretty long maturity already compared to current average maturity of US debt:

https://seekingalpha.com/article/402...level-35-years

[Hazir]

One has to wonder what's the difference if your liabilities are all in the currency you control AND hugely to foreign entities.