July 2021

What Drives Bond

Yields?

Executive Summary
This paper provides an overview of Indeed, ten-year bond yields move
the various factors that influence nearly one-for-one with changes
government bond yields. The current in long-term growth and inflation
short-term interest rate, expected expectations, which explains the
future short-term interest rates, and secular decline in bond yields over
term premia determine the level of the past several decades. Despite
(risk-free) bond yields in the economy. "noise" to the contrary, and despite the
While the current short-term interest exceptionally low yield environments
rate and its near-term expected path we have witnessed, fundamentals
are largely controlled by central banks, continue to drive bond markets.
longer-horizon expectations of future
short-term interest rates, as well as These findings have implications for
term premia, are largely driven by investors. From a strategic allocation
non-policy factors. Different forms standpoint, assuming government
of monetary policy – e.g., traditional bond risk is highly one-sided, with
interest rate policy, forward guidance, low interest rates implying there is no
and quantitative easing – are room for bonds to rally, is misguided.
unified, reacting to the same set of From a tactical perspective, despite low
economic variables and influencing yields and innovations in the conduct
broad economic conditions through of monetary policy since the Global
their impact on longer-maturity Financial Crisis, a fundamentally
yields. Both in theory and in the driven approach to bond market
data, non-monetary policy factors investing should retain its efficacy into
drive significant variation in yields, the future.
particularly at longer maturities.

I thank Cliff Asness, Paolo Bertolotti, Mohsan Bilal, Michael Doros, Tony Gould, Peter Hecht,
Jordan Brooks Bradley Jones, Antti Ilmanen, William Latimer, Zachary Mees, Michael Mendelson, Scott
Principal Richardson, Nick Steinbach, and Zhikai Xu for their valuable comments and suggestions.
Contents
Table of Contents 2

1. Introduction 3

2. Decomposing bond yields 4

3. Understanding interest rates 6

4. Understanding term premia 10

5. Monetary policy drivers of bond yields 12

6. Non-monetary policy drivers of bond yields 17

7. Discussion 20

8. Conclusion 23

References 24

Disclosures 26

Table of Contents
 What Drives Bond Yields? | July 2021 3

1. Introduction

The aim of this paper is to provide a and inflation. Expected future short-
comprehensive, but non-technical, overview term interest rates are also influenced by
of the factors that influence long-maturity monetary policy. “Forward guidance” refers
government bond yields. What determines
1
to the central bank policy of communicating
the level of bond yields in the economy and information about the anticipated path of
what drives their variation over time? future short-term interest rates with the aim
of influencing long-maturity bond yields.
With yield levels recently testing unchartered Long-horizon expectations of short-term
waters, it is hardly surprising there are many interest rates, however, are anchored by
questions, and a lot of “noise,” as to why yields macroeconomic fundamentals: long-term
are low and where they may venture from inflation expectations and the “natural rate
here. I outline a framework for identifying and of interest,” itself closely linked to the trend
understanding the important drivers of bond growth rate of output (“trend growth”).
yields. In this way, I hope to arm the reader
with a structure to thoughtfully produce The determinants of term premia are the focus
their own answers to topical questions. of section 4. The level of inflation uncertainty
and economy-wide counter-cyclical risk-
A simple and mechanical decomposition of aversion are two key factors. But term premia
bond yields proves to be extremely useful. are also influenced by exogenous supply and
This is where I begin in section 2. The demand conditions, both secular – e.g., the
yield on a long-maturity bond is equal to demand for government debt by central bank
the average expected short-term interest reserve managers, and shorter-term – e.g.,
rate over the life of the bond and a “term safe-haven demand during times of financial
premium” – the incremental compensation market stress. Monetary policy aims to
investors require to hold a long-maturity influence term premia as well. By purchasing
bond versus a series of short-maturity bonds. bonds and other long-maturity assets, central
Yield levels, therefore, depend on 1) the banks reduce their net supply, suppressing
current short-term interest rate, 2) expected term premia, and, correspondingly, lowering
future short-term interest rates, and 3) term bond yields. This policy, which originated
premia. Variations in one or more of these in response to the Global Financial Crisis
factors are what causes yields to change. (GFC) and the ensuing Great Recession, is
popularly called “Quantitative Easing” (QE).
I analyze the drivers of current and expected
future short-term interest rates in section 3. Section 5 focuses on monetary policy drivers
Central banks effectively control the short- of bond yields. I discuss how the seemingly
term interest rate and their behavior is well disparate policy tools of adjusting short-
described by the “Taylor rule,” which relates term interest rates, forward guidance, and
the level of the short-term interest rate QE, are all extremely similar, each aiming
to the near-term outlook for employment to influence the economy by manipulating

1 My focus is on “risk-free” government bonds – i.e., I ignore sovereign credit risk.

4 What Drives Bond Yields? | July 2021

yields on long-maturity bonds. Likewise, But it really isn’t, and they really haven’t.
other tools, some proposed and some Fundamental forces provide the backdrop
adopted – e.g., “average inflation targeting,” for exceptionally low bond yields: despite
“nominal GDP targeting,” “operation the recent uptick in inflation, long-term
twist,” “yield curve control,” to name a expectations remain near generational lows,
few – can all be understood as various and estimates for the natural rate of interest
implementations of forward guidance and QE. are near zero. Even abstracting from the
actions of central banks, long-maturity yields
While monetary policy certainly influences would likely be quite low at present.2 In terms
bond yields, it is far from the sole determinant. of how yields may evolve, expected future
Fundamental macroeconomic forces – long- short-term interest rates and term premia are
term inflation expectations and trend growth, key determinants of long-maturity yields. Both
uncertainty and risk aversion, exogenous policy and non-policy factors can drive these
demand factors and regulatory considerations components lower (or higher), even if short-
– influence expected future short-term interest term interest rates remain fixed. Central banks
rates and term premia, and, therefore, exert can provide additional stimulus through
considerable influence on bond yields, forward guidance and QE. Continued declines
especially at longer maturities. I discuss non- in long-term inflation expectations or trend
monetary policy drivers of yields in section 6. growth could put further downward pressure
on yields by suppressing expected future
I attempt to shed light on what I think are a short-term interest rates. This is not a forecast.
(select) few of the current relevant questions There are upside risks to yields as well, a
regarding bond yields in section 7. With yields potential increase in inflation expectations
having recently tested new lows, there is the foremost among them. But in the face of the
temptation to proclaim, “this time is different,” “noise,” sometimes the obvious needs to be
and “the rules of the game have changed.” reiterated: bond market risks are two-sided.

2. Decomposing bond yields

My starting point is an identity: the yield on a factors: the current interest rate, expected
bond is equal to the average expected interest future interest rates, and the term premium.3
rate over the life of the bond plus a term
premium. (Throughout this paper I will use Bond yields and interest rates are familiar
“yield” to mean the yield on a long-maturity concepts, and “expected future interest rates”
bond – i.e., greater than three-month – and is intuitive to grasp. The “term premium,”
“interest rate” to mean the three-month T-bill however, may be less familiar. Definitionally,
rate.) Bond yields, therefore, depend on three the term premium is the difference between

Bond Yield3 ~ Current Interest Rate (+), Expected Future Interest Rates (+), Term Premium (+)

2 This statement should not be interpreted as, "bonds are fairly priced." It should be interpreted as, "even in the absence of
accommodative monetary policy, macroeconomic fundamentals imply yield levels would likely be low relative to their history." Providing
a tactical outlook for government bonds is not the intention of this paper.
3 Read: “A bond yield depends positively on the current interest rate, expected future interest rates, and the term premium.”
 What Drives Bond Yields? | July 2021 5

the yield on a bond and the market’s term premia, we can in fact learn something
expectation of future interest rates over the purely from the data about average term
life of the bond. Hence, it embeds everything premia. Over a long enough period, expected
impacting the yield other than current and changes in interest rates should average to
expected future interest rates. Another zero. If I told you “over the last hundred years
interpretation, however, is often more useful: investors expected interest rates to rise by an
the term premium measures the expected average of two percent per year” that would
excess return investors require to hold to strain credulity. The fact the yield curve has
maturity a long-maturity bond versus rolling been upward sloping on average, therefore,
over a series of short-maturity bonds.4 implies term premia have been on average
positive and increasing with maturity.5
Although the yield decomposition is purely
an atheoretical identity, it is nevertheless Expectations Hypothesis
powerful. For example, Inker (2020)
suggests bond yields in G10 markets cannot To understand the drivers of bond yields
decline because central banks have set at a deeper level, and to identify their
interest rates at their lower bound. The policy and non-policy drivers, we need to
simple yield decomposition is sufficient to introduce theory. In the remainder of this
dispel this myth. Even with interest rates section, I focus on a simple theory of the
fixed, yields can move lower (or higher) yield curve: the expectations hypothesis
due to either changes in expected future (EH). In its modern incarnation, the EH
interest rates from the level bond markets stipulates term premia are constant over
are currently pricing in, or changes in the time, although they may differ by maturity.
term premium, which may be the result of Under the EH, variation in bond yields are
either deliberate monetary policy actions driven by just two factors: the current interest
or fundamental macroeconomic forces. rate and expected future interest rates.

One way to read the yield decomposition is The EH can accommodate a wide range of
“current and expected future interest rates, yield curve shapes and dynamics. It easily
as well as term premia, determine bond captures yield curves are upward sloping
yields.” But another reading is “the yield on average by postulating term premia are
curve – the collection of yields on bonds of positive and increasing with maturity. It also
different maturities – embeds information explains the tendency for the yield curve to be
about expected future interest rates and term steeper when the interest rate is low and flatter
premia.” An upward sloping yield curve must when the interest rate is high. According to
imply investors expect interest rates to rise, the EH, the yield curve is steeper than average
term premia are positive (and increasing when interest rates are expected to rise and
with maturity), or some combination of flatter than average when interest rates are
the two. While in absence of a model we expected to decline. When the interest rate
can’t decompose the yield curve at a point is low, markets typically expect it to increase
in time into interest rate expectations and over time, hence the yield curve is steeper than

4 The term premium is the expected excess return of holding a bond to maturity, rolling over financing. It is not the expected return from
a “constant maturity” or “rolled” position.
5 Simple empirical facts support this conclusion. Since 1960 across G6 markets (Australia, Canada, Germany, Japan, UK, US) both the
average yield curve slope and average constant-maturity ten-year excess bond returns have been positive, and excess returns have
been higher in markets where the curve has been steeper. Data is from Bloomberg.
6 What Drives Bond Yields? | July 2021

average. An inverted yield curve is thought interest rates over its lifetime, long-maturity
to be the harbinger of a recession. Again, the bond yields will rise, but they will rise by
EH provides strong guidance as to why: an less than shorter term yields, causing the
inverted curve means investors anticipate curve to flatten. The EH seamlessly and
the interest rate will fall meaningfully in the intuitively embeds the fact changes in
future, which typically occurs in a recession. interest rates tend to have an impact on both
the level and the slope of the yield curve.
The EH also provides a lens through which
to understand yield curve dynamics. When Of course, the EH is not the final word: term
central banks increase the interest rate, premia do vary over time and more elaborate
long-maturity bond yields tend to rise as models will allow for this. But they all retain,
well, but generally by less – i.e., the curve to a large degree, the spirit of the EH – current
tends to flatten. Why? When the interest interest rates and expected future interest
rate rises, expected future interest rates out rates are major drivers of bond yields. Now
a few years usually rise as well. But long- that we understand how changes in current
horizon expectations of interest rates tend to and expected future interest rates transmit
be well-anchored and change by less. As the to the rest of the yield curve, let’s turn to
yield on a bond is the average of expected understanding the drivers of the interest rate.

3. Understanding interest rates

To understand the variables that impact bond sketch a simple, yet powerful, model of
yields we need to understand the drivers the business cycle and monetary policy.
of the interest rate. In developed market The model contains four elements:
economies, the interest rate is, to a large
extent, controlled by the central bank and 1. An “aggregate demand” (AD)
is their primary monetary policy lever. To 6
relationship linking spending and
understand the behavior of the interest rate, output in the economy to the real
therefore, we need a basic understanding of yield on long-maturity bonds
how central banks conduct monetary policy.
2. An “aggregate supply” (AS) relationship
All central banks, de facto if not de jure, strive linking inflation to inflation
to maintain 1) low and stable inflation, expectations and economic activity
and 2) full employment. But how do they
attempt to achieve these objectives – how 3. A monetary policy equation, known as
do central banks influence economic the “Taylor rule” (TR), describing how
outcomes? To answer this question, I the central bank sets the interest rate

6 Some central banks do explicitly target the three-month rate, while others, such as the Federal Reserve, target a shorter-term rates
(e.g., overnight). At the very front end of the curve, however, the EH is an excellent approximation, so I will (safely) assume throughout
this paper central banks explicitly control the three-month interest rate.
 What Drives Bond Yields? | July 2021 7

4. The expectations hypothesis (EH), which savings earns a higher reward. Demand for
links longer-maturity bond yields to goods and services falls, reducing aggregate
current and expected future interest rates expenditure, and the output gap declines.8 We
can concisely summarize the AD relationship:
Although this model is simple and
Output Gap ~ Real Yield ( – )
parsimonious, the AD-AS-TR-EH framework
is at the core of almost all models used by
central banks and practicing macroeconomists Aggregate Supply
to understand the impact of economic events
on output and inflation, and to analyze the The aggregate supply relationship
prospective impact of monetary policy actions. links the level of inflation to expected
inflation and the output gap.
Aggregate Demand
Inflation depends positively on expected
The aggregate demand relationship links inflation. Firms often cannot continually
the “output gap” – the difference between adjust their prices and wages and must set
the current level of output and the level them based on expectations of the future. If
consistent with full employment (“potential firms anticipate low inflation, they may raise
output”) – to the real yield on long-maturity prices at a slower rate. If workers expect higher
bonds. The real yield is a new concept. All inflation, they may negotiate larger wage
references above to yields (and interest rates) increases. Expected future inflation, therefore,
were implicitly referring to nominal yields. The influences the inflation rate today. Inflation
real yield is simply the nominal yield minus also depends positively on the output gap.
maturity-matched inflation expectations. When the economy is booming and the output
It is the real yield, not the nominal yield, gap is positive, firms experience increasing
which enters the AD relationship. The output marginal costs and raise their prices. On the
gap is the key measure of economic activity contrary, when the economy is in a recession
central banks target, as they can hope to and the output gap is negative, marginal costs
influence it over the short-run. Potential fall, and firms lower their prices. We can
output, on the other hand, is not something concisely summarize the AS relationship:
monetary policymakers are able to influence;
Inflation ~ Expected Inflation (+), Output Gap (+)
it is determined by purely real factors, such
as longer-term productivity growth.7
Monetary policy
The AD equation posits a negative
relationship between the output gap and Despite its simplicity, the AD-AS-TR-EH
the real yield. The real yield on government model accommodates a variety of traditional
bonds influences, directly and indirectly, and non-traditional monetary policies. For
a variety of borrowing and savings rates now, however, I focus on the most traditional
across the economy. When the real yield monetary policy tool: the interest rate.
is high, borrowing is more expensive, and

7 In terms of labor markets, the output gap is synonymous to the difference between the unemployment rate and the “natural rate of
unemployment,” or “NAIRU.” I will use “output” and “employment” interchangeably.
8 Real yields can also influence asset valuations, like stock prices or home values, which impacts private sector wealth and, potentially,
aggregate spending. This mechanism is sometimes known as the “wealth effect”.
8 What Drives Bond Yields? | July 2021

Central banks seek to exert some near-term AS relationship, changes in the output gap
control over inflation and the output gap. But influence inflation. So, by changing the
how do they influence these variables – how nominal interest rate, central banks can exert
does changing the very front end of the yield influence over the variables they ultimately
curve impact macroeconomic conditions? care about: inflation and the output gap.
The sequence of steps delineating how central
bank actions influence economic variables How do central banks set the interest rate?
like inflation and the output gap is called the Prudent monetary policy suggests central
“monetary transmission mechanism.” The banks lean against the wind.9 When inflation
monetary transmission mechanism in the is above the central bank’s target level, the
AD-AS-TR-EH framework works as follows. central bank should raise the interest rate to
contract output and bring inflation down.
The central bank sets the (nominal) interest When inflation is below its target level, the
rate. The interest rate itself does not impact central bank should lower the interest rate
any macroeconomic variables in the model. to stimulate output and increase inflation.
But, via the EH, the interest rate impacts Likewise, for the output gap. When the output
(nominal) yields on long-maturity bonds. gap is positive, the central bank should raise
Because long-term inflation expectations the interest rate, lest it risk higher-than-desired
tend to be “sticky,” changes in nominal inflation. And when the economy is in a
yields on long-maturity bonds translate recession and the output gap is negative, the
to changes in real yields. Through the central bank should lower the interest rate to
AD relationship, changes in real yields stoke output.
influence the output gap. And through the

Exhibit 1: The Monetary Transmission Mechanism

Expectations Sticky Inflation Aggregate Aggregate

Hypothesis Expectations Demand Supply

Long-maturity Long-maturity
Interest rate Output gap Inflation
nominal yields real yields

Policy Response

Source: AQR. For illustrative purposes only. Not illustrative of any AQR product. Please see the Disclosures for important information.

9 An exceptional reference for those comfortable with some math is Clarida, Gali, and Gertler (1999).
 What Drives Bond Yields? | July 2021 9

The Taylor Rule policy and is the real interest rate consistent
with output equal to potential output (i.e., full
In a seminal paper, Taylor (1993) codified this employment) and stable inflation.
intuition into a simple and practical equation
for the interest rate. It has both normative The real interest rate depends on the output
and positive value, capturing the features of gap and inflation gap via the TR. Likewise,
prudent monetary policy, while also fitting near-term expected future real interest rates
the data quite well. The Taylor rule, in one of depend on forecasts of the output gap and
its many forms, relates the interest rate i to inflation gap. What about long-horizon
inflation and the output gap: expectations of the real interest rate? At
long horizons, cyclical forces eventually
i = r* + π + bπ (π – π*) + bY (Y – Y*)
wear off and monetary policy is neutral. The
π – π* is the “inflation gap,” the difference expected future real interest rate, therefore,
between the current inflation rate π, and the eventually converges to r*. Hence, long-
central bank’s inflation target π*. bπ is a positive horizon expectations of the real interest
number, so the TR captures the intuition the rate are anchored by r* and long-horizon
central bank should set a higher interest rate expectations of the nominal interest rate
when inflation exceeds its target, and a lower are anchored by r*+ πLT, where πLT denotes
interest rate when inflation is below its target. long-term inflation expectations. Since long-
Y – Y* is the output gap, the difference between maturity bond yields are largely determined
output Y, and the full employment level of by expected future interest rates, they should
output Y*. bY is also greater than zero, so the be extremely sensitive to r*+ πLT. Indeed, I will
TR says the interest rate should be higher show in section 6 long-maturity yields tend to
when the output gap is positive (an expansion) move virtually one-for-one with changes in the
and lower when the output gap is negative (a natural rate of interest and long-term inflation
contraction). expectations.

How about the first two terms in the equation, Estimating the natural rate of interest has
r*+π? To understand these terms, it is helpful been a topic of considerable interest post
to recast the TR as a rule for the real interest GFC.11 It is a critically important variable,
rate by subtracting inflation from both sides:10 both for the conduct of monetary policy and
for understanding long-maturity bond yields,
Real Interest Rate = i – π = r* + bπ (π – π*) + bY(Y – Y*)
yet it is not observable (even ex post). The best
policymakers and market participants can
The TR prescribes a real interest rate above do is try to estimate it, which is challenging
r* when the inflation gap or output gap is in a real-time setting. Fortunately, we are
positive, and a real interest rate below r* when not completely in the dark. Economic theory
the inflation gap or output gap is negative. provides guidance, linking the natural rate of
When both are zero the TR prescribes a real interest to trend growth in output (itself closely
interest rate equal to r*. Hence, r* – typically related to expected productivity and labor
referred to as either “r-star” or “the natural force growth). Exhibit 2 plots average real GDP
rate of interest” – defines neutral monetary growth against average real interest rates from

10 Technically we should subtract a three-month inflation forecast (i.e., the first π in the TR should be expected inflation over the next
three months). Realized inflation is a reasonable proxy and makes the notation cleaner.
11 Hamilton et al (2015) is a comprehensive survey.
10 What Drives Bond Yields? | July 2021

1990-2020. Over a 30-year sample, average gap should average out to close to zero). It
real GDP growth should provide a reasonable is clear there is a very strong relationship –
estimate of trend growth and the average countries with higher real growth experienced
real interest rate should provide a reasonable higher real interest rates and vice versa.
estimate of r* (i.e., the output gap and inflation

Exhibit 2: Average Real Interest Rates vs. Average Real GDP Growth

4%

y = 1.0x - 0.52% NZ
Correl = 0.69
Average real interest rate

3%

NW AU
2% UK
SD
CN
BD
1%
US
SW
JP

0%
0% 1% 2% 3% 4%

Average real GDP growth

Source: Bloomberg. Sample is January 1990 through December 2020. Average three month T bill rate net of realized inflation vs. average
real GDP growth. Please see the Disclosures for important information.

4. Understanding term premia

The term premium is the extra return required identifying the forces that drive time-variation
to hold to maturity a long-maturity bond in term premia.
versus rolling over short-maturity debt (e.g.,
three-month T-bills). Term premia are on Mechanically, the determinants of term
average positive and rising with maturity: premia are all factors that influence a bond’s
investors typically require extra yield to yield, other than the current interest rate and
hold long-maturity bonds relative to short- expected future interest rates. In practice,
maturity bonds. While the average slope two key factors tend to drive term premia: 1)
of the yield curve pins down average term changes in perceived riskiness, and 2) changes
premia, term premia at a point in time cannot in demand and supply.
be directly observed. I make no effort to
estimate them. There is a cottage industry Perceived Riskiness
of models attempting to pin them down, and
all estimates have wide standard errors and On the risk side, term premia are higher when
are sensitive to specification. My focus is bonds are truly riskier and when investors’
 What Drives Bond Yields? | July 2021 11

tolerance for risk is lower – i.e., when investors Demand and Supply
are more risk-averse.
Variations in net demand can arise from
The most important risk factor for (fixed- exogenous macroeconomic and geopolitical
rate) government bonds is inflation. Yields events, or from the explicit monetary policy
incorporate inflation expectations over the actions of central banks.13 Their relative safety,
life of the bond, but unexpected increases in liquidity, and ability to meet regulatory capital
inflation erode a bond’s real return. Inflation and liability-hedging needs make government
risk is likely the main driver behind the fact bonds appealing to investors. These features
term premia are on average positive: the are especially attractive during times of
possibility of unexpected inflation makes financial market stress (e.g., “flight-to-quality”
holding long-maturity bonds riskier than and “flight-to-liquidity” episodes), during
rolling over short-maturity debt.12 By the same which safe-haven demand for government
logic, variation in inflation risk should be a key bonds tends to suppress their term premia.
driver of time-variation in term premia. When Secular trends in net demand, such as
uncertainty about future inflation is high, say increases in the holdings of government
during the 1970s and into the 1980s, a period bonds by foreign central banks, also influence
where inflation rates averaged in the double term premia. Monetary policy may affect
digits and inflation volatility was extremely term premia as well. Indeed, quantitative
elevated, investors require a larger expected easing and all its variants – “yield curve
return to holding long-maturity bonds relative control,” “operation twist,” etc. – feature the
to their shorter-maturity counterparts. purchases of longer-maturity bonds (among
Wright (2012) and Bauer et al (2013) verify other securities) by central banks, reducing
this intuition empirically. Using a large their net supply, with the explicit objective of
international panel data set, both papers find suppressing their term premia.
a strong link between estimates of term premia
and measures of inflation uncertainty. The key drivers of term premia can be
expressed succinctly:
In addition to inflation uncertainty, risk
Term Premium ~ Inflation Uncertainty (+),
aversion – the compensation investors require
Risk Aversion (+), Net Demand (–)
for bearing risk – also varies materially over
time. Theoretical and empirical research How important is time-variation in term
supports the view risk aversion varies premia in explaining the dynamics of
countercyclically, being higher during bond yields? During the Treasury market
recessions than expansions (e.g., see Cochrane “conundrum” of 2004-2006, the Federal
(2011)). Indeed, Bauer et al find evidence of a Reserve raised interest rates from 1.5 to 5.25
pronounced countercyclical pattern in term percent, yet long-maturity yields were virtually
premia across international bond markets: unchanged. This anomalous behavior is
term premia estimates on long-maturity bonds generally attributed to a fall in term premia.
are meaningfully higher in recessions than What drove the decline? Likely a combination
expansions. of factors. Inflation uncertainty was quite

12 Indeed, during the gold standard era of 1879-1970 – a period in which inflation fluctuations were quite short-lived – yield curves were
on average close to flat according to Wood (1983). No, not Gordon.
13 These factors can also be interrelated to investors’ risk tolerance. For example, to accommodate an increase in the supply of bonds,
investors, in aggregate, would need to take on more risk. Hence, they require higher expected excess returns to counteract their lower
risk tolerance. See, for example, Greenwood and Vayanos (2014).
12 What Drives Bond Yields? | July 2021

low, both due to improved inflation-fighting Lastly, the “global saving glut,” which led to
credibility of the Fed and more secular forces heightened demand for Treasuries by foreign
(this period – “the great moderation” – featured central banks and sovereign wealth funds,
exceptionally low macroeconomic volatility). fueled an increase in demand for safe and
Low risk aversion was another likely factor, as liquid US government debt.14
the economy enjoyed a prolonged expansion.

5. Monetary policy drivers of bond yields

In this section I analyze the monetary policy The Taylor rule describes how central banks
drivers of bond yields. I show a wide array of set interest rates. Central banks set the interest
different central bank tools are, in reality, very rate low when inflation is less than their
similar policies, reacting to the same set of inflation target or when output is less than
economic variables – the near-term outlook for its full employment level. Since monetary
output and inflation – and relying on the same policy tends to influence output and inflation
transmission mechanism to influence the with a lag, in practice central banks look at
broader economy. In the next section, I shift near-term forecasts of output and inflation
focus to key drivers of long-maturity yields that when setting their interest rate policy.15 The
are largely uninfluenced by monetary policy. TR also assumes key variables, like potential
output and the natural rate of interest are
There are three main monetary policy tools known with certainty, while they are, at best,
utilized by central banks: the interest rate noisily estimated. To be clear, the TR is meant
level, forward guidance, and quantitative to only provide a qualitative description of
easing. how central banks set interest rates. The
interest rate will often deviate from the exact
Interest rate level number prescribed by the TR, but it provides
an excellent description as to what monetary
By adjusting the interest rate, central banks policymakers think about in determining the
exert influence on long-maturity nominal appropriate interest rate setting.
and real yields. Real yields on long-maturity
government debt influence a variety of So, what does the TR tell us about yield
borrowing rates and asset valuations across drivers? Changes in the near-term outlook
the economy – e.g., mortgage rates, corporate for the output gap and inflation influence the
bond yields, etc. – which impact aggregate interest rate, and through this channel, longer-
demand, and eventually output and inflation. maturity bond yields. Additionally, a monetary
This is the classical monetary transmission policy shock – an unexpected change to the
mechanism illustrated in Exhibit 1. interest rate not explained by the output and
inflation outlook, will also influence longer-
maturity bond yields. Improving economic

14 See Backus and Wright (2003) for a more complete discussion.

15 Using realized values versus forecasts can make a big difference. See this blog post by Ben Bernanke.
 What Drives Bond Yields? | July 2021 13

conditions (e.g., falling unemployment), rising influences aggregate spending and, eventually,
inflation, or a positive monetary policy shock output and inflation.
will tend to lead to higher bond yields, and the
opposite will tend to lead to lower bond yields. While forward guidance has been around for
two decades, it has gone through multiple
At present many central banks have set iterations. Date-based forward guidance
their interest rate close to zero or some other articulates how interest rates are expected to
perceived “lower bound.” Does this mean evolve as a function of time. Outcome-based
they can no longer influence bond yields to forward guidance articulates how interest
stimulate the economy? No. Many central rates are expected to evolve as a function
banks have broached the zero lower bound of economic conditions. The recent shift by
and others may follow suit. And, regardless the Fed to “average inflation targeting” is a
of the willingness to “go negative,” central form of outcome-based forward guidance.
banks have two additional policy tools at their By communicating they are targeting two
disposal. percent inflation on average, the Fed is
intimating to markets they are willing to
Forward guidance tolerate inflation rates temporarily above two
percent, and, therefore, will still maintain an
Abstracting from term premia, a bond’s yield accommodative monetary policy stance in
is equal to the average of current and expected that scenario. Nominal GDP and price level
future interest rates over its life. For a long- targeting are additional variations on this
maturity bond, the current interest rate setting theme.17
covers only a small fraction of its lifetime, and
expectations of future interest rates are the What does forward guidance tell us about
more critical determinant. Central banks are yield drivers? Changes in the near-term
acutely aware expected future interest rates outlook for the output gap and inflation will
influence bond yields. “Forward guidance” – influence communication about future interest
communications about the anticipated path rates, and through this channel, influence
of interest rates – has been a standard part of long-maturity bond yields. Additionally, a
the monetary policy playbook for the past 20 “path shock” – an unexpected change to the
years.16 communicated path of future interest rates
– will also influence long-maturity yields.
The transmission mechanism for forward So, improving economic conditions, rising
guidance is virtually identical to the inflation, or a positive path shock will lead
transmission mechanism for interest rate to higher yields, and the opposite will lead to
changes. In response to the near-term outlook lower yields.
for the output gap and inflation, central banks
provide guidance on future interest rates, Importantly, even if central banks are
which influences long-maturity nominal yields reluctant to reduce their interest rate further,
(EH), which influences long-maturity real forward guidance can still operate with the
yields (inflation expectations are sticky), which interest rate at its lower bound. In response to
a deteriorating outlook for economic activity

16 Indeed, it is often asserted the two-year yield, which according to the EH embeds information about expected interest rates over the
next two years, provides the single best measure of the overall stance of monetary policy.
17 Woodford (2012) has an excellent discussion on the merits of forward guidance and its different variants.
14 What Drives Bond Yields? | July 2021

or inflation, central banks can still use In practice, QE may involve the purchase
communication about future interest rates to of securities beyond government bonds,
influence long-maturity yields. including securitized assets (e.g., MBS),
corporate bonds, and even equities. And
Quantitative Easing or “Targeted Asset purchases may be financed either by reserve
Purchases” creation or by selling shorter-maturity
bonds. These are all different versions,
Quantitative easing refers to targeted central however, of the same fundamental policy.19
bank purchases of financial assets. Of the Either by purchasing government bonds
three monetary policy tools we consider, and suppressing their term premia, or by
QE is the newest – adopted in its modern purchasing other long-duration assets directly,
incarnation by central banks in response to central banks aim to reduce borrowing costs
the GFC and subsequent global recession – and increase asset valuations broadly, with
and most often misunderstood. 18
the goal of influencing output and inflation.
QE1, QE2, “operation twist” – the purchase of
In the simplest form of QE the central bank longer-maturity government debt financed be
purchases long-maturity government bonds the sale of shorter-maturity debt, “yield curve
and holds them on its balance sheet. By control” – the Bank of Japan announcing a
reducing the net supply of long-duration ten-year yield target in the range of zero and
assets, these purchases suppress term premia the Royal Bank of Australia adopting three-
and reduce yields. By reducing nominal yields, year yield target of 0.25 percent: these are all
QE influences real yields, which influence policies aimed at influencing long-maturity
borrowing costs across the economy more bond yields, conceptually no different than
broadly, eventually impacting aggregate traditional interest rate policy and forward
spending, output, and inflation. I hope to guidance.
impress a critical point: QE works through the
exact same monetary transmission mechanism QE is not “money printing,” as pundits and
as traditional interest rate policy and forward market commentators occasionally refer to
guidance. Each policy aims to influence yields it. It is not about the money supply. That’s
on long-maturity bonds. Interest rate policy simply not how the monetary policy machine
and forward guidance do so via current and works. QE is about long-maturity bond yields
expected future interest rates, respectively. QE and valuations of long-duration assets, in the
accomplishes the same end by impacting term same way as traditional interest rate policy
premia. and forward guidance. Accordingly, QE does

18 The term “Quantitative Easing” causes unnecessary confusion to this day. It was originally introduced by the Bank of Japan in 2001 to
describe a new policy of increasing the supply of bank reserves to target a higher growth rate for the monetary base. Economic theory,
corroborated by subsequent empirical evidence, implies increases in reserves should have virtually no impact in a low interest rate
environment. (See Krugman, 1998; or Eggertsson and Woodford 2003 for detailed discussions. In summary, with the interest rate
near zero – or near the rate paid on excess reserves – the opportunity cost of holding reserves is negligible, and demand for reserves
becomes infinitely elastic. Further expansion of the supply of reserves, therefore, has no consequence for interest rates, broader
monetary aggregates, aggregate expenditure, or prices.) Largely unsuccessful, the policy was abandoned in 2006. QE, as practiced
by central banks since the GFC (including the Bank of Japan, which undertook a new program of “comprehensive monetary easing”
in 2010), is importantly different from the original Bank of Japan version. Modern QE emphasizes the purchase of longer-maturity
financial assets to reduce bond yields and borrowing costs across the economy, as well as to boost asset valuations. While “targeted
asset purchases” is unquestionably a more accurate name, the term QE has stuck.
19 When QE is financed by reserve creation the central bank creates new bank reserves to purchase bonds in the open market from major
financial institutions. Relative to QE financed by selling shorter-maturity debt, QE financed by reserve creation has the additional
consequence of increasing the amount of reserves held by banks. For reasons discussed in the previous footnote, however, in a low
interest rate environment the quantity of reserves has little consequence for bank lending and, therefore, the macroeconomy.
 What Drives Bond Yields? | July 2021 15

not represent a draconian shift in the conduct purchases – will impact yields (e.g., the taper
of monetary policy. It is simply another tool tantrum). So, improving economic conditions,
in the arsenal of central banks to meet their rising inflation, or a negative balance sheet
twin objectives of price stability and full shock will lead to higher bond yields, and
employment. the opposite will lead to lower bond yields.
Importantly, QE, like forward guidance,
Indeed, the transmission mechanism of QE remains a viable monetary policy tool to
and other policy tools are so inextricably reduce yields and stimulate the economy,
linked that, according to Bernanke (2020), even if central banks are averse to additional
in addition to reducing term premia, the interest rate cuts.
other channel through which QE principally
operates is by reinforcing forward guidance. Summary
QE signals policymakers’ intention to keep
interest rates low for an extended period. By influencing the current interest rate, the
Bernanke cites the “taper tantrum” of 2013 as expected path of future interest rates, and
evidence: his own hints the Fed might slow term premia, traditional interest rate policy,
their pace of asset purchases led markets to forward guidance, and QE all influence long-
revise forward expectations for interest rate maturity bond yields. Given central bank
hikes, causing yields to materially rise. The objectives of maintaining low and stable
importance of Bernanke articulating this inflation and full employment, policymakers
“signaling channel” of QE is two-fold: 1) if react to changes in the outlook for output and
the former Fed Chairman and architect of inflation. Improving economic conditions
QE views it as operating through the exact or increasing inflation leads to a more
same channels as more traditional monetary contractionary policy stance and higher yields.
policy tools, so should we; 2) Bernanke, and Worsening economic conditions or falling
most modern central bankers, view forward inflation leads to a more accommodative
guidance as the primary monetary policy tool. stance and lower yields. Any unexpected
change to the monetary policy stance of the
What does QE tell us about yield drivers? central bank – be it an interest rate surprise,
Inasmuch as changes in the near-term outlook news about the path of future interest rates, or
for output and inflation influences the overall unexpected changes to size and complexion
stance of monetary policy, they are likely of the central bank’s balance sheet – will also
to influence the size and complexion of QE influence longer-maturity yields. Exhibit 3
policies, impacting both expected future summarizes these drivers.
interest rates and term premia on long-
maturity bonds. Additionally, a “balance
sheet shock” – an unexpected change to either
the amount, complexion, or timing of asset
16 What Drives Bond Yields? | July 2021

Exhibit 3: Monetary Policy Drivers of Bond Yields

Monetary Policies

Output gap
Interest rate Forward Quantitative
level guidance easing
Inflation gap

Current Expected Future Term

Bond Yield ~ (+) (+) (+)
Interest Rate Interest Rates Premium

Source: AQR. For illustrative purposes only. Not illustrative of any AQR product. Please see the Disclosures for important information.

Does the fact short-term interest rates are yields continue to react to economic news
close to zero (or lower) in most major markets in a manner consistent with their monetary
impede these dynamics in any fundamental policy drivers. Exhibit 4 plots the percentage
way? No. Several central banks (e.g., the of variation over time in G6 bond yields
European Central Bank and the Bank of explained by changes in near-term forecasts of
Japan) have showed a willingness to employ unemployment and inflation. Despite policy
negative interest rates. Others, including rates being at levels close to zero or less, and
the Federal Reserve, have been reluctant despite lower bond market volatility, bond
to embrace negative interest rates. Even if yields are still reacting to macroeconomic
interest rates stay put, forward guidance news in a quantitively similar manner to other
and QE both remain viable policy tools to periods.
further reduce bond yields. Indeed, bond

Exhibit 4: News About Employment and Inflation Still Drives Yield Changes

70% 9%

8%
60%
Perentage of explained variance
in ten-year G6 yield changes

7%
Average G6 ten-year yield

50%
6%
40% 5%

30% 4%

3%
20%
2%
10%
1%

0% 0%

Source: Bloomberg, Consensus Economics. The sample is G6 markets. The sample is G6 markets (Australia, Canada, Germany, Japan, UK,
US), January 1990 through December 2020. Chart displays R2s in three-year rolling regressions of changes in ten-year bond yields on
revisions to unemployment and inflation forecasts (left), and average ten-year yields (right). Not illustrative of any AQR product. Please see
the Disclosures for important information.
 What Drives Bond Yields? | July 2021 17

6. Non-monetary policy drivers of

bond yields

Much recent discussion by fixed income longer-term inflation expectations. As a result,

investors has centered around the willingness r*+ πLT provides an anchor for long-maturity
of central banks to embrace negative interest nominal yields, and the long-end of the yield
rates or expand their balance sheets. But curve should be very sensitive to changes in
monetary policy is far from the only variable either the natural rate of interest or long-term
influencing long-maturity yields. There inflation expectations.
are a variety of non-policy-related factors
that exert considerable influence over the This is exactly what we observe in the data.
level and shape of the yield curve. This Exhibit 5 displays the results of regressing
may sound sacrilegious, but occasionally ten-year, five-year, and two-year yields on
the role of central banks in markets can be long-term inflation expectations, long-term
overstated.20 Changes in trend growth and real growth expectations (a proxy for the
long-term inflation expectations, variation in unobservable natural rate of interest, in
inflation uncertainty, as well as shorter-term accordance with Exhibit 2), and the current
fluctuations in both the business cycle, and interest rate. Ten-year yields move virtually
oscillating demand for highly liquid, safe- one-for-one with long-term inflation and growth
haven assets, also exert meaningful influence expectations, regardless of the actions of
over bond yields. central banks. At shorter maturities, the
influence of long-term inflation and growth
To a good approximation the current interest expectations wanes, although they remain
rate is under the control of the central bank. economically and statistically significant,
So, there is limited scope for non-policy related while the influence of the current interest rate
factors to influence yields via that channel. is stronger.
But expectations of future interest rates,
particularly at long-horizons, and term premia
are largely driven by economic factors beyond
the current monetary policy stance.

Central banks exert considerable influence

over the expected path of the interest rate in
the near-term. Long-maturity bond yields,
however, depend on expected interest rates
well into the future. According to the AD-AS-
TR-EH model, expected future interest
rates should eventually converge to r*+ πLT,
the sum of the natural rate of interest and

20 I equate this statement to my claim The Shawshank Redemption is an overrated movie. It is undeniably an awesome film – a classic, and
one I’ll always stop to watch if it is on TV (take that cord-cutters). But according to IMDB it is the best movie of all-time. Sorry, but Vito
Corleone, Rick Blaine, and I – the first and, likely, only time the three of us have appeared in the same sentence – beg to differ.
18 What Drives Bond Yields? | July 2021

Exhibit 5: Long-Maturity Yields Depend on Inflation and Growth Expectations

1.2
Regression Coefficient

1.0

0.8

0.6

0.4

0.2

0.0
10y Yield 5y Yield 2y Yield
Long-Term Inflation Long-Term Growth Interest Rate
Source: Consensus Economics, Bloomberg. Sample is G6 (Australia, Canada, Germany, Japan, UK, US), quarterly, December 1990
through December 2020. Long-term inflation and growth are median forecasts of average inflation and real GDP growth rates between
6-10 years ahead from Consensus Economics. Yield and interest rate (T-bill) are from Bloomberg. Dynamic OLS regression includes
country effects and two leads/lags of changes in the dependent variables. Not illustrative of any AQR product. Please see the Disclosures
for important information.

Long-maturity bond yields depend on long- from just north of two percent in 2018, and
term inflation expectations and the natural spending the majority of 2000-2010 decade
rate of interest and both vary meaningfully hovering at around 2.5 percent. Picking 2000
over time. According to the model of Holston, as a reasonable reference date, the natural
Laubach, and Williams (2017) maintained rate of interest in the US has declined from
by the Federal Reserve Bank of New York, 3.4 percent to zero percent, and long-term
estimates of the natural rate of interest for inflation expectations from 3.2 percent to 1.6
the US and Euro area have declined from percent. That’s a decline in r*+ πLT from 6.6
levels of three to four percent in the late percent 1.6 percent. For reference, over the
1990s to between zero and one percent, same period, ten-year US Treasury yields
in large part driven by estimated declines declined from 6.5 percent to 1.5 percent. The
in trend growth due to slowing labor force natural rate of interest, itself linked to trend
growth and a slowdown in trend productivity growth, and long-term inflation expectations
growth. According to the model of Haubrich, fluctuate meaningfully and drive variation
Pennacchi, and Ritchken (2012) maintained in long-maturity bond yields. Together they
by the Federal Reserve Bank of Cleveland, can account for the entirety of the secular
the current expected average inflation rate decline in US Treasury yields over the last two
over the next 10 years (as of June 30, 2021) decades.
in the US is 1.6 percent, having declined
 What Drives Bond Yields? | July 2021 19

Exhibit 6: The Natural Rate of Interest and Long-Term Inflation Expectations

7%
Natural rate of interest r*

6%

5%

4%

3%

2%

1%

0%

US Canada Euro Area UK

7%
Ten-year inflation expectations

6%

5%

4%

3%

2%

1%

0%

US

Source: Federal Reserve Bank of New York, Federal Reserve Bank of Cleveland. Sample is January 1982 through April 2020 for the
natural rate of interest and January 1982 through December 2020 for inflation expectations. Natural rate of interest estimates follow
Holston, Laubach, and Williams (2017) and are maintained by the Federal Reserve Bank of New York (https://www.newyorkfed.org/
research/policy/rstar). Long-term inflation estimates follow Haubrich, Pennacchi, and Ritchken (2012), and are maintained by the Federal
Reserve Bank of Cleveland (https://www.clevelandfed.org/our-research/indicators-and-data/inflation-expectations.aspx). Not illustrative
of any AQR product. Please see the Disclosures for important information.

While QE influences term premia on nominal the last half century, each of these factors
bonds, there are a host of other factors, not have been important drivers of term premia.
under the direct control of central banks, And each are likely to play important roles
which also drive term premia. As discussed in in the future. Inflation uncertainty has
section 4, these include inflation uncertainty, subsided from the levels it attained in the
risk aversion, and changes in net demand for 1970s, but the current unprecedented amount
government bonds. The latter may be due to of monetary and fiscal stimulus calls into
exogenous variation in safe-haven demand question whether central banks will be as
for government bonds; as well as more successful in managing inflation expectations
secular trends, such as foreign demand for – and therefore inflation itself – over the next
safe government debt emanating from China decade. And while term premia are generally
and other Asian economies, oil producers, counter-cyclical – both inflation uncertainty
and emerging markets. At varying times over and investor risk aversion tend to be higher
20 What Drives Bond Yields? | July 2021

in recessions than in booms – a flare-up in of long-maturity bond yields. Inflation

safe-haven demand can further suppress term uncertainty and investor risk aversion drive
premia and bond yields over the near-term. counter-cyclical variation in term premia. And
fluctuations in the net demand for government
Summarizing, non-policy factors play at bonds – either driven by safe-haven flows or
least as important a role as monetary policy more secular changes in global demand for
in driving bond yields. Trend growth (via liquid, low risk assets – influence yields on
the natural rate of interest) and inflation government debt. Fluctuations in any of these
expectations anchor expectations of future factors – in the near-term and beyond – can
interest rates and are the key determinants drive bond yields higher or lower.

Exhibit 7: Non-Policy Drivers of Bond Yields

Non-Policy Drivers

1. Inflation uncertainity
1. Natural rate of interest
Economy 2. Risk aversion
2. Long-term inflation expectations 3. Changes in net demand

Current Expected Future Term

Bond Yield ~ (+) (+) (+)
Interest Rate Interest Rates Premium
Source: AQR. For illustrative purposes only. Not illustrative of any AQR product. Please see the Disclosures for important information.

7. Discussion
Bond yields recently touched all-time lows Why are yields so low?
in several developed markets and remain
extremely low relative to history. Many In a paper devoid of predictions, I will assert
investors are pondering how much lower one (albeit unverifiable) counterfactual.
they can go, while others wonder whether Regardless of the actions of central banks in
aggressive fiscal policy might lead to the response to COVID-19 and the subsequent
end of the era of record low yields. I make global economic slowdown, long-maturity
no predictions. Even in chartered waters, bond yields would be low relative to historical
forecasting the near-term direction of markets levels at present. With slowing labor force
– bonds, stocks, bitcoins, or beanie babies – is growth (due to demographics) and a slowdown
a fool’s errand, and, therefore, any “market in trend productivity growth causing estimates
timing” tilts within an asset allocation should of the natural rate of interest to be sub-one
be modest. While I make no forecast, I do hope percent across the developed world, low
to try to at least guide the discussion to the long-term inflation expectations, relatively
relevant questions. low uncertainty about future inflationary
outcomes, generally low levels of risk aversion,
and persistent demand for government debt
 What Drives Bond Yields? | July 2021 21

– there are a variety of cyclical and secular responses. They face no tradeoff between their
factors all exerting downward pressure dual objectives at present.
on bond yields. Add the actions of central
banks – zero (or negative) interest rates, In terms of non-policy drivers of yields, the
forward guidance pledging to maintain most important determinants are the natural
an accommodative interest rate setting for rate of interest and long-term inflation
the foreseeable future, and targeted asset expectations. The secular decline in these
purchases reducing the overall net supply of variables has been a primal force spurring the
long-maturity debt and exerting downward decline in yields over the past two decades,
pressure on term premia – and current yield as they anchor long-horizon expectations of
levels, to a reasonable approximation, appear interest rates. Should either decline materially,
to be justifiable. Yields should be in the range long-maturity yields would follow. Inflation
of their historical lows. uncertainty and risk aversion appear quite low,
so it is hard to imagine cyclical variation in the
What can make yields go lower (or term premia presents much of a downside risk
higher)? to yields at present. But should the pandemic
worsen, or should the geopolitical situation
Monetary policy influences long-maturity weigh further on the global economy, we could
bond yields through setting the current see a material pickup in safe-haven demand
interest rate, influencing the expected path of for government bonds, which would likely
future interest rates via forward guidance, and drive yields to even lower levels.
manipulating term premia (and reinforcing
forward guidance) via quantitative easing. The same framework is useful for identifying
Several central banks have set their policy upside risks. From a monetary policy
interest rate negative. Others, the Federal perspective, withdrawing some of the
Reserve foremost among them, have not yet extraordinary stimulus would put upward
embraced negative interest rates. Should pressure on yields. In what state of the world
the Fed “go negative,” it would provide some might central banks do this? The good state:
downward pressure on yields, both because economic activity sees a strong rebound as
the current interest is lower and because it the pandemic fades, growth and employment
would likely cause markets to revise down pick up meaningfully, and within a year
the expected path of future interest rates. or so global economies have returned to
But even if the Fed and other central banks full employment. The “risk management”
keep their respective interest rates at bay, approach of central banking might still
forward guidance and QE remain potent argue to withdraw stimulus only tepidly,
tools. Given the aggressiveness with which but it is certainly plausible the pace of asset
central banks have responded to deteriorating purchases (QE) slows and forward guidance
economic conditions since the GFC, I think moves more neutral. This would, of course,
the baseline case must be they would continue put upward pressure on yields. The bad
to add monetary stimulus should economic state: inflation expectations meaningfully
or financial market conditions materially increase. Most central banks undoubtably
worsen. The fact inflation expectations remain would accommodate – let me be even more
well within most central banks' target range clear: they’d welcome – a pickup in inflation.
allows them to be aggressive in their policy But should we see a de-anchoring of inflation
expectations, perhaps driven by fears over the
22 What Drives Bond Yields? | July 2021

extraordinary levels of fiscal and monetary Have the rules of the game changed?
stimulus, central banks may indeed take
their foot off the accelerator. From a non- This is my way of summarizing the multitude
policy perspective, an increase in either the of questions that have continued to arise for
natural rate of interest or long-term inflation the last decade. The implication is usually
expectations would put upward pressure on two-fold: QE represents a whole new playbook
long-maturity bond yields regardless of the in terms of monetary policy, and bond markets
policy responses of central banks. A rise in are no longer reacting to fundamentals as they
inflation uncertainty or declining demand for have in the past.
government debt, among other factors, can
drive term premia higher, and with them long- I hope I have impressed QE is not a
maturity yields. fundamentally different monetary policy
tool. It impacts the economy by influencing
Is there a floor on government bond long-maturity bond yields, just as traditional
yields? interest rate policy and forward guidance.
Whereas the latter two policies influence
Presumably yes, at some level. But the floor yields through manipulating the current
is likely materially lower than zero. The logic interest rate and expected path of future
of a zero-lower bond on the interest rate is interest rates, QE influences bond yields by
cash-currency earns a zero yield: savers can manipulating term premia.
put money in their mattresses and earn a
higher return. But this idea doesn’t hold up Fundamentals continue to drive government
to scrutiny or reality (the current ECB policy bond markets. The current low level of
rate is -50 basis points). And yields can go yields globally is consistent with broader
materially lower than the interest rate, as macroeconomic developments: slowing
there are plenty of factors that can drive term productivity growth, depressed long-
premia temporarily well into negative territory: term inflation expectations, low inflation
bonds provide a highly liquid and relatively uncertainty and generally low levels of risk
low-risk store of value; bonds have provided aversion, as well as central banks providing
diversification to risky assets (e.g., equities); stimulus – via interest rate policy, forward
and bonds serve regulatory and transaction guidance, and QE – in an attempt to support
functions. Moreover, expected returns across economic activity and inflation. There’s no
the board are quite subdued at present, and unchartered policy or unexplained influences
bond yields look far less abnormal when at work: yields are reacting in the manner
compared to other assets. Finally, even at we’d expect, and central banks are utilizing
negative yields bonds may be attractive from essentially familiar policy tools.
a portfolio perspective – their real returns can
be meaningfully positive in a deflationary The reaction of bond markets to changes in
scenario. It is beyond this paper to estimate economic conditions continues to be broadly
the lower-bound on bond yields, but it in line with what we expect. Good news about
certainly isn’t zero or -50 basis points, and it economic growth, either over the short-term or
is likely quite a bit lower. While many believe, long-term (i.e., influencing the natural rate of
explicitly or tacitly, in the zero lower bond, and interest) moves yields higher, while bad news
thus think bond risk is highly asymmetric, moves them lower. Ditto for inflation. Yields
theory and evidence do not bear this out. appear to have been responsive to interest
 What Drives Bond Yields? | July 2021 23

rate, forward guidance, and QE policies. And To be sure, the economy and markets face
during the dark days of March and April extraordinary challenges. But it is during these
2020, when credit and liquidity conditions times one needs to remember the playbook,
tightened meaningfully, yields declined not throw it out the window. We have a
sharply, reflecting a rapid decline in term reasonable sense of the factors that influence
premia, precipitated in part by the actions of bond yields to help navigate the times ahead.
the Federal Reserve and other central banks.

8. Conclusion

By decomposing longer-maturity bond well. Non-monetary policy drivers are also

yields into the current interest rate, expected critical, particularly at longer maturities.
future interest rates and term premia, and Indeed, declining estimates of the natural
analyzing the drivers of each component, I’ve rate of interest, in part driven by a pessimistic
presented a framework for understanding the outlook for trend growth, and falling long-term
determinants of bond yields, and identified inflation expectations are important factors
several of their fundamental drivers. Through behind the secular decline in bond yields
the systematic response of monetary over the past few decades, and provide the
policy, the near-term outlook for economic backdrop for the extremely low yield levels we
activity and inflation influence bond yields. observe today.
Unexpected monetary policy changes – to
the current interest rate, the expected path of Looking towards the future, investors would
future interest rates (i.e., changes in forward be prudent to conclude government bond risk
guidance), and the size and complexion remains two-sided even at low levels of bond
of the central bank’s balance sheet (i.e., yields, and a fundamentally driven approach
change in QE policies) – influence yields as to bond market investing remains viable.
24 What Drives Bond Yields? | July 2021

References

AQR Alternative Thinking, 2019. “Inversion Anxiety: Yield Curves, Economic Growth, and
Asset Prices.”

AQR Alternative Thinking, 2021. “2021 Capital Market Assumptions for Major Asset Classes.”

Cliff Asness, Antti Ilmanen, and Thomas Maloney, 2017. “Market Timing: Sin a Little, Resolving
the Valuation Timing Puzzle.” Journal of Investment Management, 15(3): 23-40,

David K. Backus and Jonathan H. Wright, 2007. “Cracking the Conundrum.” Brookings Papers
on Economic Activity, 2007 (1): 293-329.

Michael D. Bauer, Glenn D. Rudebusch, and Jing Cynthia Wu, 2014. “Term Premia and
Inflation Uncertainty: Empirical Evidence from an International Panel Dataset: Comment.”
American Economic Review, 104 (1): 323-327.

Ben S. Bernanke, 2020. “The New Tools of Monetary Policy.” American Economic Review, 110 (4):
943-983.

Richard Clarida, Jordi Gali, and Mark Gertler, 1999. “The Science of Monetary Policy: A New
Keynesian Perspective.” Journal of Economic Literature, 37 (4): 1661-1707.

John H. Cochrane, 2011. “Presidential Address: Discount Rates.” The Journal of Finance, 66 (4):
1047-1108.

Gauti B. Eggertsson and Michael Woodford, 2003. “The Zero Bound on Interest Rates and
Optimal Monetary Policy.” Brookings Papers on Economic Activity, 2003 (1): 139-211.

Jonathan Fader, Zachary T. Mees, Michael A. Mendelson, 2020. “It’s Not a Bound; It’s an
Opinion.” AQR Capital Management Working Paper.

Robin Greenwood and Dmitri Vayanos, 2014. “Bond Supply and Excess Bond Returns.” The
Review of Financial Studies, 27 (3): 663-713.

James D. Hamilton, Ethan S. Harris, Jan Hatzius, and Kenneth D. West, 2016. “The
Equilibrium Real Funds Rate: Past, Present, and Future.” IMF Economic Review, 64 (4):
660-707.

Joseph H. Haubrich, George Pennacchi, and Peter Ritchken, 2012. “Inflation Expectations, Real
Rates, and Risk Premia: Evidence from Inflation Swaps.” The Review of Financial Studies, 25
(5), 1588-1629.
 What Drives Bond Yields? | July 2021 25

Kathryn Holston, Thomas Laubach, and John C, Williams, 2017. “Measuring the Natural Rate
of Interest: International Trends and Determinants.” Journal of International Economics, 108
(1): S59-S75.

Ben Inker, 2020. “Government Bonds Have Given Us So Much: Do They Have Anything Left to
Give?” GMO Quarterly Letter, 2Q 2020.

Paul R. Krugman, 1998. “It’s Baaack: Japan’s Slump and the Return of the Liquidity Trap.”
Brookings Papers on Economic Activity, 1998 (2): 137-206.

John B. Taylor, 1993. “Discretion versus Policy Rules in Practice.” Carnegie-Rochester Conference
Series on Public Policy, 39 (1), 195-214.

John H. Wood, 1983. “Do Yield Curves Normally Slope Up? The Term Structure of Interest
Rates, 1862-1982.” Federal Research Bank of Chicago Economic Perspectives, 17-23.

Michael Woodford, 2012. "Methods of policy accommodation at the interest-rate lower bound."
Economic Policy Symposium – Jackson Hole, Federal Reserve Bank of Kansas City, 185-288.

Jonathan H. Wright, 2012. “What Does Monetary Policy do to Long-term Interest Rates at the
Zero Lower Bound?” The Economic Journal, 122 (1): 447-466.
26 What Drives Bond Yields? | July 2021

Disclosures

This document has been provided to you solely for information purposes and does not constitute an offer or solicitation of an offer or
any advice or recommendation to purchase any securities or other financial instruments and may not be construed as such. The factual
information set forth herein has been obtained or derived from sources believed by the author and AQR Capital Management, LLC (“AQR”)
to be reliable but it is not necessarily all-inclusive and is not guaranteed as to its accuracy and is not to be regarded as a representation or
warranty, express or implied, as to the information’s accuracy or completeness, nor should the attached information serve as the basis of
any investment decision. This document is intended exclusively for the use of the person to whom it has been delivered by AQR, and it is not
to be reproduced or redistributed to any other person. The information set forth herein has been provided to you as secondary information
and should not be the primary source for any investment or allocation decision.

Past performance is not a guarantee of future performance.

This material is not research and should not be treated as research. This presentation does not represent valuation judgments with respect
to any financial instrument, issuer, security or sector that may be described or referenced herein and does not represent a formal or official
view of AQR.

The views expressed reflect the current views as of the date hereof and neither the author nor AQR undertakes to advise you of any changes
in the views expressed herein. It should not be assumed that the author or AQR will make investment recommendations in the future that
are consistent with the views expressed herein, or use any or all of the techniques or methods of analysis described herein in managing
client accounts. AQR and its affiliates may have positions (long or short) or engage in securities transactions that are not consistent with the
information and views expressed in this presentation.

The information contained herein is only as current as of the date indicated, and may be superseded by subsequent market events or for
other reasons. Charts and graphs provided herein are for illustrative purposes only. The information in this presentation has been developed
internally and/or obtained from sources believed to be reliable; however, neither AQR nor the author guarantees the accuracy, adequacy
or completeness of such information. Nothing contained herein constitutes investment, legal, tax or other advice nor is it to be relied on in
making an investment or other decision.

There can be no assurance that an investment strategy will be successful. Historic market trends are not reliable indicators of actual future
market behavior or future performance of any particular investment which may differ materially, and should not be relied upon as such.
Target allocations contained herein are subject to change. There is no assurance that the target allocations will be achieved, and actual
allocations may be significantly different than that shown here. This presentation should not be viewed as a current or past recommendation
or a solicitation of an offer to buy or sell any securities or to adopt any investment strategy.

The information in this presentation may contain projections or other forward-looking statements regarding future events, targets,
forecasts or expectations regarding the strategies described herein, and is only current as of the date indicated. There is no assurance
that such events or targets will be achieved, and may be significantly different from that shown here. The information in this presentation,
including statements concerning financial market trends, is based on current market conditions, which will fluctuate and may be superseded
by subsequent market events or for other reasons. Performance of all cited indices is calculated on a total return basis with dividends
reinvested.

The investment strategy and themes discussed herein may be unsuitable for investors depending on their specific investment objectives
and financial situation.

Neither AQR nor the author assumes any duty to, nor undertakes to update forward looking statements. No representation or warranty,
express or implied, is made or given by or on behalf of AQR, the author or any other person as to the accuracy and completeness or fairness
of the information contained in this presentation, and no responsibility or liability is accepted for any such information. By accepting this
presentation in its entirety, the recipient acknowledges its understanding and acceptance of the foregoing statement.

The data and analysis contained herein are based on theoretical and model portfolios and are not representative of the performance of funds
or portfolios that AQR currently manages.

Australia

AQR Capital Management, LLC is exempt from the requirement to hold an Australian Financial Services License under the Corporations
Act 2001 (Cth). AQR Capital Management, LLC is regulated by the Securities and Exchange Commission ("SEC") under United States of
America laws, which differ from Australian laws. Please note that this document has been prepared in accordance with SEC requirements
and not Australian laws.
 What Drives Bond Yields? | July 2021 27

Canada

Canadian recipients of fund information: These materials are provided by AQR Capital Management (Canada), LLC, Canadian placement
agent for the AQR funds.

Information for clients in the EEA

AQR in the European Economic Area is AQR Capital Management (Germany) GmbH, a German limited liability company (Gesellschaft mit
beschränkter Haftung; “GmbH”), with registered offices at Maximilianstrasse 13, 80539 Munich, authorized and regulated by the German
Federal Financial Supervisory Authority (Bundesanstalt für Finanzdienstleistungsaufsicht, „BaFin“), with offices at Marie-Curie-Str. 24-28,
60439, Frankfurt am Main und Graurheindorfer Str. 108, 53117 Bonn, to provide the services of investment advice (Anlageberatung)
and investment broking (Anlagevermittlung) pursuant to the German Securities Institutions Act (Wertpapierinstitutsgesetz; “WpIG”). The
Complaint Handling Procedure for clients and prospective clients of AQR in the European Economic Area can be found here: https://ucits.
aqr.com/Legal-and-Regulatory

Hong Kong (Asia)

Please note for materials distributed through AQR Capital Management (Asia)

This presentation may not be copied, reproduced, republished, posted, transmitted, disclosed, distributed or disseminated, in whole or
in part, in any way without the prior written consent of AQR Capital Management (Asia) Limited (together with its affiliates, “AQR”) or as
required by applicable law.

This presentation and the information contained herein are for educational and informational purposes only and do not constitute and should
not be construed as an offering of advisory services or as an invitation, inducement or offer to sell or solicitation of an offer to buy any
securities, related financial instruments or financial products in any jurisdiction.

Investments described herein will involve significant risk factors which will be set out in the offering documents for such investments and are
not described in this presentation. The information in this presentation is general only and you should refer to the final private information
memorandum for complete information. To the extent of any conflict between this presentation and the private information memorandum,
the private information memorandum shall prevail.

The contents of this presentation have not been reviewed by any regulatory authority in Hong Kong. You are advised to exercise caution and
if you are in any doubt about any of the contents of this presentation, you should obtain independent professional advice.

Information for clients in the United Kingdom

The information set forth herein has been prepared and issued by AQR Capital Management (Europe) LLP, a UK limited liability partnership
with its office at Charles House 5-11, Regent St., London, SW1Y 4LR, which is authorised and regulated by the UK Financial Conduct
Authority (“FCA”).

www.aqr.com

AQR Capital Management, LLC Two Greenwich Plaza, Greenwich, CT 06830 P +1.203.742.3600 F +1.203.742.3100