Issue #34: What’s the Deal with Nominally Negative-Yielding Bonds?
In this issue of Monetary Mechanics, I am going to explain why people would willingly buy nominally negative-yielding bonds. On the surface, this seems like a ridiculous proposal. What nominally negative yields translate to, in simple terms, is that someone is paying you money for the privilege of lending you money – a proposition that seems ridiculous. Yet somehow, there are trillions of dollars of negative-yielding bonds in the modern financial system today, which include not only negative-yielding sovereign bonds but also negative-yielding investment-grade and junk bonds.
Why are investors willing to pay trillions of dollars for bonds for which they will earn negative returns in nominal terms with 100% certainty? When I am presented with a market phenomenon that appears paradoxical, I find that it is often an invitation to investigate it deeply and in detail to attempt to find the logic that lies underneath the apparent illogicalness. I believe that this is the only method to broaden one’s own perspective and to understand the fundamental structural forces that drive market behavior.
I believe that an understanding of this phenomenon necessitates an understanding of the various methods to profit from investments in fixed income instruments. There are two standard explanations that try to rationalize the existence of the trillions of dollars of negative-yielding bonds in the modern financial system today:
Deflation – investors believe that inflation rates will be low or perhaps even negative in the future, so nominally negative yields will be positive in real terms in the present.
Profit through capital gains rather than through interest income – this is essentially a variant of the “greater fool theory,” whereby one can buy a bond with a yield of -1% and still profit from it if they can sell that bond to someone else for an implied yield of -2% (essentially profiting from the price appreciation rather than from the interest income).
I do not find either of these two standard explanations particularly satisfying.
First, even though I am not an “inflationista” by any means, I cannot argue against the fact that so far inflation has been higher than the previous decade’s average inflation levels by all measures. Furthermore, even before the recent post-COVID-19 pandemic supply chain driven bump in inflation, inflation levels were never low enough to justify nominally negative yields. Hence, deflation does not seem like a reasonable explanation, because it is not reasonable to think that investors bet more than $10 trillion dollars of capital on inflation rates being negative from 2016 onwards.
Second, it is true that betting on the price of bonds to increase during periods of economic and financial turmoil has been a good investment strategy in the past, as demonstrated by the strength of the classic 60/40 portfolio model and the inverse correlation between bonds and stocks for the past 40 years. However, I do not believe that it is reasonable to think that large institutional investors are betting on nominally negative-yielding bonds exclusively for the purpose of speculation on capital gains, given that there is a certain loss on interest income regardless of the end result.
Large institutional investors have increasingly allocated capital away from traditional fixed income instruments and towards equities and alternative investments due to declining bond yields.1 2 However, while this capital reallocation is happening in the background, large institutional investors are still buying nominally negative-yielding bonds despite the fact that doing so no longer provides diversification benefits.3
So, we must ask, once again, why there are trillions of dollars of negative-yielding bonds floating around in the modern financial system today. One final explanation is the proliferation of bond index exchange-traded funds (ETFs) from asset managers like Vanguard that offer diversified exposure to fixed income instruments around the world, some of which may be negative-yielding. A prime example is the Vanguard Total International Bond ETF, one of the largest exchange-traded bond funds, whose top allocations include Japan, France, and Germany, which are all countries with negative-yielding sovereign bonds.4 Consequently, while the investors invested in these ETFs are not making an active decision that the negative-yielding bonds are a good investment, passive flows into these ETFs are creating an outcome that is similar to the relentless upwards march of the equities markets.
While I believe that there is merit to this explanation, I also believe that a few billion dollars of flows into diversified bond ETFs cannot completely explain the existence of the trillions of dollars of negative-yielding bonds. In my mind, there are only two explanations that can fully account for such a burgeoning of negative-yielding bonds: post-GFC legislations and regulations (particularly Dodd-Frank and the liquidity coverage ratio (LCR) requirement) and securities lending income.
I have already talked about various post-GFC legislations and regulations in Issue #26 of Monetary Mechanics, as well as their collective effect on the demand for high-quality liquid assets (HQLA) in Issue #32 of Monetary Mechanics. Thus, I am going to talk primarily about the existence of securities lending in relation to the burgeoning of negative-yielding bonds in this issue of Monetary Mechanics.
I believe that the demand for collateral stemming from collateral transformation trades is driving securities lending income above and beyond any loss that one would have to suffer by holding negative-yielding bonds. As of December 31, 2020, approximately $29.6 trillion of assets were available for lending globally.5 As of the first half of 2021, the total lendable amount of assets had grown to over $34 trillion.6
So now there are a lot of passive holders of US government bonds, foreign sovereign bonds, and other fixed income instruments, who hold them even though they do not have yield, precisely because these securities can be lent out. Consequently, the more aggressive these investors are able and willing to be in the class of collateral that they are able and willing to take when they lend them out (i.e. the riskier the collateral that they are able and willing to take when they lend them out), the more money these investors get paid.
The higher securities lending income that these funds offer to these investors incentivizes increased capital inflows, which causes adverse selection in the sense that the highest performing funds are highest performing precisely because they are able and willing to participate in the riskiest types of securities lending contracts.
Let’s consider a hypothetical example to hammer this point home. Currently, 10-year US treasuries yield approximately 1.5%, which is significantly less than the inflation rate. However, if the securities lending income from lending these US treasuries is 3% or more, suddenly these US treasuries do not seem like such an unattractive investment.
This is exactly what has been happening ever since the implementation of post-GFC legislations and regulations simultaneously decreased circulation of C1 collateral and manufactured demand for C1 collateral. Consequently, the decrease in collateral velocity brought about this new obscure cottage industry of collateral transformation and securities lending, enabling financial institutions to earn positive returns on nominally negative-yielding bonds.
https://crr.bc.edu/wp-content/uploads/2017/06/slp_55.pdf
https://www.nomurafoundation.or.jp/en/wordpress/wp-content/uploads/2014/09/20101015_Betsy_Palmer.pdf
https://www.blackstone.com/docs/default-source/black-papers/seeking-an-alternative_standard.pdf
https://www.investmentnews.com/negative-yields-find-their-way-into-u-s-investor-portfolios-169723
https://www.blackrock.com/us/individual/literature/brochure/us-retail-securities-lending-brochure.pdf
https://www.institutionalinvestor.com/article/b1v73mkpcd7scr/New-Opportunities-in-Securities-Lending