An Explanation of Negative Swap Spreads: Demand for Duration from Underfunded Pension Plans

• Author: SURESH SUNDARESAN,SVEN KLINGLER
• DOI: http://doi.org/10.1111/jofi.12750
• Published date: 01 April 2019
• Date: 01 April 2019

THE JOURNAL OF FINANCE •VOL. LXXIV, NO. 2 •APRIL 2019

An Explanation of Negative Swap Spreads:

Demand for Duration from Underfunded Pension

Plans

SVEN KLINGLER and SURESH SUNDARESAN∗

ABSTRACT

The 30-year U.S. swap spreads have been negative since September 2008. We offer a

novel explanation for this persistent anomaly.Through an illustrative model, we show

that underfunded pension plans optimally use swaps for duration hedging. Combined

with dealer banks’ balance sheet constraints, this demand can drive swap spreads to

become negative. Empirically,we construct a measure of the aggregate funding status

of defined benefit pension plans and show that this measure helps explain 30-year

swap spreads. We find a similar link between pension funds’ underfunding and swap

spreads for two other regions.

INSEPTEMBER 2008, SHORTLY AFTER THE default of Lehman Brothers, the differ-

ence between the swap rate (which is the fixed rate in the swap) of a 30-year in-

terest rate swap (IRS) and the yield of a Treasury bond with the same maturity,

commonly referred to as swap spread, dropped sharply and became negative.

As we explain in more detail later, this is a theoretical arbitrage opportunity

and an asset pricing anomaly. In contrast to other crisis phenomena, the 30-

year negative swap spread has been highly persistent, still at around −40 basis

points as of December 2015. In this paper, we examine the persistent negative

30-year swap spread and offer a new perspective on the possible reasons behind

this anomaly. Our hypothesis is that demand for duration hedging by under-

funded pension plans coupled with the balance sheet constraints faced by swap

dealers put pressure on long-term swap-fixed rates and ultimately pushed the

30-year swap spread to become negative.

The negative swap spread pricing anomaly presents a challenge to the view

widely held prior to the financial crisis that swap spreads are indicators of

∗Klingler (corresponding author) is from the Department of Finance, BI Norwegian Business

School. Sundaresan is from Columbia Business School. Weare grateful to Stefan Nagel (the Editor),

the Associate Editor, two anonymous referees, Darrell Duffie, Robin Greenwood, WeiJiang, Tomas

Kokholm, David Lando, Harry Mamaysky, Scott McDermott, Stephen Schaefer, Pedro Serrano,

Morten Sørensen, Hyun Shin, Savitar Sundaresan, and Dimitri Vayanos for helpful comments.

Klingler acknowledges support from the Center for Financial Frictions (FRIC), grant no. DNRF102.

Sundaresan acknowledges with thanks the productive sabbatical spent at BIS, where the paper

received useful suggestions and comments. We also thank participants in the seminars at the

Swiss Finance Institute in Gerzensee and at the Indian School of Business for their comments.

The authors have no conflicts of interest, as identified in the Disclosure Policy.

DOI: 10.1111/jofi.12750

675

676 The Journal of Finance R

market uncertainty that increase in times of financial distress. This is because

the fixed payment in an IRS is exchanged against a floating payment, which

is typically based on Libor, and entails credit risk. Hence, while IRSs are

collateralized and viewed as free of counterparty credit risk, the swap rate

should be above the (theoretical) risk-free rate because of the credit risk that is

implicit in Libor. Swap spreads should therefore increase in times of elevated

bank credit risk (see Collin-Dufresne and Solnik (2001) for a treatment of this

and related issues). In addition, Treasuries (which are the benchmarks against

which swap spreads are computed) are regarded as a “safe haven,” that is

assets that investors value for their safety and liquidity. In times of financial

distress, investors value the convenience of holding safe and liquid assets even

more, which decreases the Treasury yield and makes them trade at a liquidity

premium or convenience yield (see, for instance, Longstaff (2004), Feldh¨

utter

and Lando (2008), or Krishnamurthy and Vissing-Jorgensen (2012)). Taken

together, these arguments suggest that the 30-year swap spread should have

increased around the default of Lehman Brothers.

We offer a demand-driven explanation for negative swap spreads. In par-

ticular, we develop a model in which underfunded pension plans’ demand for

duration hedging leads them to optimally receive the fixed rate in IRSs with

long maturities. Pension funds have long-term liabilities in the form of un-

funded pension claims and invest in a portfolio of assets of various duration,

such as stocks and government bonds. They can balance their asset-liability

duration by investing in long-term bonds or by receiving fixed in an IRS with

long maturity. Our theory predicts that if pension funds are underfunded, they

prefer to hedge their duration risk by entering into IRSs rather than by buying

Treasuries, which may not be feasible given their funding status. The pref-

erence for using IRSs to hedge duration risk arises because a swap requires

only modest investment to cover margins, whereas buying a government bond

to match duration requires outright investment.1This demand, when coupled

with dealer balance sheet constraints, results in negative swap spreads.

Greenwood and Vayanos (2010) show that pension funds’ demand for dura-

tion hedging in the United Kingdom can affect the term structure of British

gilts by lowering long-term rates. In this sense, our paper bears a close relation-

ship to their work. Our approach differs from theirs, however, as we focus on

underfunded pension funds’ optimal preference for the use of IRSs in duration

hedging. The model that we develop shows that the demand for IRSs increases

as the fund becomes more underfunded.

We provide nonparametric evidence suggesting that the swap spreads tend

to be negative in periods when defined benefit (DB) plans are underfunded.

We thus illustrate a new channel that may be at work in driving long-term

swap spreads down. Using data from the Financial Accounts of the United

States (formerly named Flow of Funds Accounts) from the Federal Reserve,

we construct a measure of the aggregate funding status of DB plans (both

private and public) in the United States. We then use this measure to test the

1The leverage implications of IRSs on pension sponsors are not investigated in this paper.

An Explanation of Negative Swap Spreads 677

relationship between the underfunded ratio (UFR) of DB plans and long-term

swap spreads in a regression setting. Even after controlling for other common

drivers of swap spreads recognized in the literature, such as the spread

between Libor and repo rates, the debt-to-GDP ratio, dealer-banks’ financial

constraints, market volatility, and the level as well as slope of the yield

curve, we find that UFR helps explain 30-year swap spreads. In line with

our narrative, we also show that swap spreads of shorter maturities are not

affected by changes in UFR.

We conduct a number of robustness tests. One potential concern with using

UFR to explain swap spreads is that the same factors known to affect swap

spreads can also affect pension funds. For example, a change in the level of the

yield curve can affect not only swap spreads but also the level of pension funds’

underfunding. Toaddress this concern, we use stock returns as an instrumental

variable in a two-stage least squares setting. The idea here is that while stock

returns directly affect pension funds’ funding status (through the asset side),

there is no obvious economic reason as to why they should be related to swap

spreads. Our results are robust to this test. Next, we include additional control

variables, such as U.S. corporate bond issuance, and we examine the effect

of pension funds’ underfunding on swap spreads in different time periods. We

find that the link between UFR and swap spreads is most pronounced in the

immediate aftermath of Lehman Brothers’ default, when derivatives dealers

faced stringent balance sheet constraints. Finally, we examine the relation

between a modified version of UFR and swap spreads and find that our results

are robust to using this alternative measure.

We conclude our paper by testing the effect of pension funds’ underfunding on

swap spreads in two additional countries with significant pension plans: Japan

and the Netherlands. We find that the funding status of Japanese pension

funds contributes significantly to Japanese swap spreads with 10 years and 30

years to maturity while Dutch pension funds’ funding status is a significant

factor behind 30-year swap spreads in Europe. Next, we review the related

literature.

As mentioned above, Greenwood and Vayanos(2010) show that demand pres-

sure by pension funds lowers long-term yields of British gilts. In addition,

Greenwood and Vayanos (2010) mention that pension funds also fulfill their

demand for long-dated assets by using derivatives to swap fixed for floating

payments. They note that pension funds have “swapped as much as £50 billion

of interest rate exposure in 2005 and 2006 to increase the duration of their as-

sets” but do not investigate the effect of such demand on swap spreads further.

Moreover, their focus is on U.K. gilt markets. Our paper complements their

analysis by showing that underfunded pension funds’ demand for long-dated

assets can have a strong impact on swap rates, globally.

More generally, swap rates and Treasury yields have been extensively

studied in previous literature. A stream of literature calibrates dynamic term-

structure models to understand the dynamics of swap spreads (see Duffie and

Singleton (1997), Lang, Litzenberger, and Liu (1998), Collin-Dufresne and Sol-

nik (2001), Grinblatt (2001), Liu, Longstaff, and Mandell (2006), Johannes and