From Great Depression to Great Recession

Chapter 10. Global Bond Market Spillovers from Monetary Policy and Reserve Management

Atish Ghosh, and Mahvash Qureshi
Published Date:
March 2017
  • ShareShare
Show Summary Details
Robert N. McCauley

The idea that policy in one country affects other countries through exchange rates is a staple of international finance. Exchange rate policy in one country affects other countries because there is only one exchange rate in a two-country world, two in a three-country world, and so on. With floating exchange rates, domestic monetary policy affects other countries through the exchange rate, among other channels. In particular, monetary easing, in the conventional sense of a reduction of the short-term policy rate, induces exchange rate depreciation. This, of course, implies exchange rate appreciation and disinflation for the rest of the world.

Unconventional monetary policy in the form of large-scale bond purchases intended to reduce bond yields extends but does not alter the transmission of policy easing through the exchange rate. There is no reason to believe that a given decline in, say, two-year yields resulting from central bank purchases of bonds would have a different effect on the exchange rate than the same decline resulting from a lower short-term policy rate.1 In this sense, those who expressed the view that unconventional monetary policy amounted to a “currency war” could have leveled the same charge, whatever its merit, at conventional monetary policy.

Large-scale bond purchases do, however, heighten strategic interactions in the global bond market. In particular, a policy to lower dollar bond yields makes investors more receptive to dollar debt issued by firms from around the world. The induced dollar borrowing allows firms outside the United States to sidestep domestic monetary policy. In addition, lower bond yields in a key currency such as the dollar tend to lower domestic currency bond yields around the world. Again, the influence of domestic monetary policy in the form of setting short-term interest rates is reduced.

Moreover, strategic interactions in global bond markets arise unintentionally as central banks act in their capacity as reserve managers. These operations may be considered incidental to the policy of managing the exchange rate but can nevertheless influence bond yields in the reserve currencies, even without any intention to do so. Thus, recent unconventional monetary policies to purchase bonds in large amounts in order to affect yields interact with long-standing and sizable reserve management operations in bond markets.

This chapter compares the narrow and unrealistic conditions under which conventional monetary policy entails no bond market interactions with actual conditions, in which large-scale bond purchases ease global financing conditions for bond issuance in the same currency and depress yields in bond markets in other currencies. It also analyzes episodes of central bank interactions in the global bond market, suggesting that they can either reinforce or offset each other. The chapter suggests that central banks should consider the implications of their operations in the global bond market.

Monetary Policy Without Global Bond Market Effects

Obstfeld and Rogoff (2002) reach the conclusion that there is no useful role for monetary policy coordination in a model that does not have a bond market. If one seeks to understand central bank interactions in the global bond market, a richer framework is required.

One can imagine a world with bond markets but without bond market spillovers. Each bond market would price bonds strictly on the basis of the expectations hypothesis, so that 10-year bond yields would be nothing more than the expected short-term rates over the 10 years (possibly plus an exogenous term premium). In this world, if central banks set the short-term interest rate in response to, say, a Taylor rule, bond yields would be correlated only to the extent that inflation and employment deviations from targets were themselves correlated across economies. In other words, bond yields would co-move only to the extent that central banks responded to common disturbances to inflation or growth.

In this world, a central bank that bought bonds would have no effect on bond yields other than the signal value of the purchases for future short-term policy rate setting. There is thus scope for forward guidance (Filardo and Hofmann 2014), and asset purchases would at most add credibility to such guidance.

In this world, if central banks in some countries sought to restrain currency fluctuations by setting interest rates more in line with those set for key currencies, one would observe greater linkage of bond yields, even assuming that bond yields are set by expected short-term rates. Hofmann and Bogdanova (2012), Taylor (2013), and Hofmann and Takáts (2015) argue that one can in fact observe such follow-the-leader behavior. In Figure 10.1, panels 1 and 2 show that central banks systematically set policy rates below Taylor rule norms in the mid-2000s and since 2008, after having earlier tracked such norms.

Figure 10.1.Policy and Taylor Rates1


Sources: IMF, International Financial Statistics and World Economic Outlook databases; Bloomberg; CEIC database; Consensus Economics; Datastream; national data; BIS calculations.

1 Weighted average based on 2005 GDP and purchasing power parity (PPP) exchange rates. “Global” comprises all economies listed here. Advanced economies: Australia, Canada, Denmark, the euro area, Japan, New Zealand, Norway, Sweden, Switzerland, the United Kingdom, and the United States. Emerging market economies: Argentina, Brazil, Chile, China, Chinese Taipei, Colombia, the Czech Republic, Hong Kong SAR, Hungary, India, Indonesia, Israel, Korea, Malaysia, Mexico, Peru, the Philippines, Poland, Singapore, South Africa, and Thailand.

2 The Taylor rates are calculated as i = r*+π* + 1.5(π–π*) + 0.5y, where π is a measure of inflation, y is a measure of the output gap, π* is the inflation target, and r* is the long-run real interest rate, here proxied by real trend output growth. The graph shows the mean and the range of the Taylor rates of different inflation/output gap combinations, obtained by combining four measures of inflation (headline, core, GDP deflator, and consensus headline forecasts) with four measures of the output gap (obtained using Hodrick-Prescott [HP] filter, segmented linear trend and unobserved components techniques, and IMF estimates). Π* is set equal to the official inflation target/objective, and otherwise to the sample average or trend inflation estimated through a standard HP filter. See Hofmann and Bogdanova 2012.

Resistance to exchange rate appreciation has also largely driven the reserve accumulation shown in Figure 10.2. Peak reserves may have been reached in 2014, and the implications of emerging market central banks selling reserve currency bonds—dubbed “quantitative tightening” by Winkler, Sachdeva, and Saravelos (2015)—are discussed later.

Figure 10.2.Global Foreign Exchange Reserves

Sources: Datastream; IMF, International Financial Statistics and World Economic Outlook databases.

Thus, in this world, if policymakers set rates to avoid exchange rate appreciation, they choose to import the monetary policy of a key currency country (Bernanke 2012). By contrast, a fully independent monetary policy would tend to insulate the domestic bond market from external pressures, if bond yields simply reflected future short-term policy rates.

In the real world, a sizable and varying portion of bond yields cannot be explained by expected future short-term rates. This component, known as the “term premium,” is discussed in the following section, along with evidence that it is subject to a strong global influence (Rey 2013). This means that central bank operations in bond markets can have effects that spill over through the global bond market. Neither the model of Obstfeld and Rogoff (2002) nor a model in which bond yields respond only to expected policy rates is adequate.

If central banks can influence these term premiums by large-scale bond purchases, there is much scope for bond market interactions. The following discussion covers spillover through the bond market in which the purchases take place, and spillover through bond markets in other currencies that are integrated with the target market.

Large-Scale Bond Purchases and Dollar Bonds of Non-Us Borrowers

The Federal Reserve’s large-scale asset (bond) purchase programs are intended to compress the term premium in the US Treasury bond market (Bernanke 2013). A policy that is often described as “quantitative easing” attempts to reduce the expected return from a fixed-rate bond relative to that of a short-term bill rolled over for the life of the bond. This interpretation of large bond purchases as compressing the term premium has received support from Gagnon and others (2011), Krishnamurthy and Vissing-Jorgensen (2011), and D’Amico and King (2013).

While intended to make it easier and cheaper for US firms and households to issue corporate bonds and fixed-rate mortgages, the Federal Reserve’s large-scale bond purchases also eased and cheapened dollar bond issues by firms and governments outside the United States. And it was the stock of dollar bonds issued by nonfinancial borrowers outside the United States that proved most responsive to the Federal Reserve’s compression of the term premium.

Despite the Federal Reserve’s purchases, some stocks of dollar bonds have in fact not risen over the past eight years (Figure 10.3, panel 1).2 The largest stock (other than US Treasury bonds) is that of agency bonds, mostly mortgage-backed securities supported by Fannie Mae and Freddie Mac. With the slow recovery of the housing market, agency bonds only in 2016 exceeded their 2008 outstanding value. The second largest stock in 2008 comprised private-label asset-backed securities. Dominated by mortgage-backed securities, this stock continues to run down, notwithstanding the revival of auto-loan-backed securities.

Figure 10.3.Federal Reserve Spurs Dollar Bond Issuance by Non-US Borrowers

Sources: Bank for International Settlements (BIS), International Debt Securities Statistics; Bloomberg L.P., Board of Governors of the Federal Reserve, US Flow of Funds; McCauley, McGuire, and Sushko 2015b; and BIS staff calculations.

1 In panel 2, response of the quarterly growth in the stock of US dollar bonds issued outside the United States to the (lagged) change in the real term premium, estimated from 16-quarter rolling regressions that also include the lagged Chicago Board Option Exchange Volatility Index to control for overall financial market conditions; see McCauley, McGuire, and Sushko 2015b. The vertical line indicates the end of quarter one, 2009. The 10-year real term premium is estimated using a joint macroeconomic and term structure model; see Hördahl and Tristani (2014).

By contrast, two stocks of bonds have responded to the opportunity presented by the Federal Reserve. US nonfinancial firms have raised $2.3 trillion. To what extent these firms used the proceeds of bond issues to expand their operations or to make payouts to shareholders remains unclear (Van Rixtel and Villegas 2015). The fourth largest stock in 2008 was that of nonbank borrowers resident outside the United States; these borrowers ramped up their outstanding value from $2.3 trillion to $4.7 trillion.

Thus, if the Federal Reserve’s buying of Treasury and agency bonds is seen as working through prices (compressing the term premium) to encourage investors to bid for riskier bonds, it succeeded most with bonds issued by nonbank borrowers outside the United States. Investors lined up not only for dollar bonds issued by US firms but also for those sold by non-US firms and governments.

This conclusion is backed by evidence that the compression of the term premium opened portfolios to dollar bond issues by borrowers resident outside the United States (Lo Duca, Nicoletti, and Martinez 2014; McCauley, McGuire, and Sushko 2015b). Figure 10.3, panel 2 shows that the coefficient obtained from a rolling 16-quarter regression of the growth rate of offshore dollar bonds on the (lagged) term premium change turned significantly negative after the Federal Reserve announced its large-scale bond purchases. This indicates that a lower term premium in the previous quarter was associated with faster growth in the stock of dollar bonds issued by non-US-resident nonfinancial borrowers in the present quarter. Thus, in the wake of the Federal Reserve’s bond buying, we see not only strong dollar bond issuance by non-US borrowers but also a positive response of that issuance to (lagged) changes in the term premium.

In sum, the unconventional US monetary policy in the form of large-scale bond purchases may not have been intended to change the behavior of borrowers outside the United States; nevertheless, it induced them to build up their dollar liabilities.

Spillovers from US Bond Yields to Global Bond Yields

If the Federal Reserve’s bond buying made global investors receptive to dollar bond issuance by non-US residents, it also affected pricing of bonds denominated in other currencies. In a globally integrated bond market, what starts in the dollar does not stay in the dollar (Rey 2013). Instead, lower yields for US Treasury bonds lead to lower yields on other governments’ bonds, as long as global investors balance their bond portfolios across markets that are open to investment (IMF 2014). At the limit, if bonds in different currencies served as very close substitutes in private portfolios, it would not matter which bonds officials purchased.

Much analysis suggests that the Federal Reserve’s large-scale purchases of US bonds, or announcements thereof, led to substantial changes in yields on other sovereign bonds (Table 10.1). Neely (2015), for example, finds that 20 percent to 70 percent of the announcement effects of Federal Reserve bond buying diffused to mature bond markets. Bauer and Neely (2014) find that the effect worked through shared term premium (that is, through channels other than correlated expectations of future short-term rates) in the larger German and Japanese markets. Rogers, Scotti, and Wright (2014) confirm Neely’s findings using high frequency futures data. Following Bernanke, Reinhart, and Sack (2004), who analyze the Japanese exchange rate intervention of 2003–04, Gerlach-Kristen, McCauley, and Ueda (2016) find that investment of dollars purchased against the yen in US Treasury bonds lowers their yields but also that one-third to two-thirds of the effect passes through to other mature bond markets. This analysis includes not only sovereign bonds but also generic private rates (as represented by interest rate swaps). Obstfeld (2015) reports long-term level (cointegration) regressions that suggest that major government bond markets move in synch with the US Treasury market, with a median half-life of adjustment of about one year.

Table 10.1.Estimates of Spillovers of US Bond Yields to Mature Bond Markets(Basis points per 100 basis points on the US Treasury bond)
Bond MarketGerlach-Kristen,

McCauley, and Ueda 2015:

Japanese Intervention,

Neely 2015:

LSAP1 Events
Bauer and

Neely 2014:

LSAP1 Events

Scotti, and

Wright 2014:


Obstfeld 2015:


Levels, 1989–

United Kingdom65494648137
Sources: Cited works; and author’s calculations.Note: LSAP1 = first Federal Reserve large-scale asset (bond) purchase.
Sources: Cited works; and author’s calculations.Note: LSAP1 = first Federal Reserve large-scale asset (bond) purchase.

Several studies suggest that bond markets in emerging markets now respond more to changes in global bond markets than they did a decade ago, when local Asian bond markets (except for those of Hong Kong SAR and Singapore) showed low correlations with US Treasury bonds (McCauley and Jiang 2004). More recent work by Turner (2014), Miyajima, Mohanty, and Yetman (2014), and Chen and others (2015) shows a tighter linkage, including between India and the United States, although still excluding China (see Sobrun and Turner 2015).

The integration measured by Obstfeld (2015) can easily be read as the influence of the largest, deepest, and most liquid government market on other bond markets. This reading does not just draw on size but is also informed by persuasive episodes.

The global bond market strains of 1994 stand out (Borio and McCauley 1996). The Federal Reserve started a tightening cycle in February 1994, even as the Bundesbank, then the European anchor policymaker, extended a long easing (Figure 10.4, panel 1). In the bond market, US Treasury yields rose and, contrary to the direction of actual and expected policy rates in Europe, German bund yields and French treasury bond yields rose more or less in step. Adrian and Fleming (2013)—using Adrian, Crump, and Moensch (2013)—find that almost all the rise in US rates reflected expected future policy rates. By contrast, the German and French government bond markets’ rise in yields could not be put down to expected policy rates but rather reflected a widening term premium.

Figure 10.4.Policy Rates and Bond Yields


Sources: Bloomberg L.P.; national statistics; and Bank for International Settlements staff calculations.

1 Decomposition of the 10-year nominal yield according to an estimated joint macroeconomic and term structure model; see Hördahl and Tristani 2014. Yields are expressed in zero coupon terms; for the euro area, French government bond data are used.

In short, US and European monetary policy diverged, but US and European bond yields tracked each other higher, with the term premium widening in Europe. Central banks were not playing follow-the-leader, but bond market investors were.

Some of those engaged in constructing the euro anticipated (or at least hoped) that the larger euro-denominated bond market would have more ballast and thus would sail more steadily and prove less subject to being tossed about by waves moving east across the North Atlantic. Through 2013 (see Figure 10.4, panel 2), these hopes proved largely unfulfilled as euro bond yields tended to follow dollar bond yields closely.

Note that the 1994 episode (shown in Figure 10.4) did not feature unconventional monetary policy in the sense of large-scale interventions in the bond market. But the lesson learned from that episode was the asymmetric influence of conventional US monetary policy working only through bond markets dominated by private investors.

Central banks have since operated in the bond market, both incidentally and purposefully. First, reserve managers invested the proceeds of large-scale foreign exchange accumulation in key currency bonds, and second, unconventional monetary policy took the form of large-scale bond buying.

Recent Episodes of Central Bank Interactions in the Global Bond Market

Once central banks begin to operate in size in the bond market, they add new channels of interaction to those that operate at one remove through short-term policy rate setting and its effect on bond markets dominated by private investors. The interaction can be reinforcing, even cooperative (at least implicitly), with central banks on the same side of the bond market. Or the interaction can involve action at cross-purposes, with central banks on opposite sides of the bond market. Policy friction can arise even when central banks are on the same buy side of the market, as the case of Brazil and the United States in 2010–11, discussed later in this chapter, suggests.

Table 10.2 reduces the number of players to two but allows the dramatis personae to vary.3 Interactions are novel when both parties operate in the bond market, but the cases discussed include interactions between a central bank operating only on short-term policy rates and a central bank operating in the bond market.

Table 10.2.Recent and Possible Central Bank Interactions in the Global Bond Market
Federal Reserve
TightenBrazil1–United States, 2010–11EMs sell US Treasuries, 2015
Hold(Twin-taper China–United

States, 2014)
EaseLarge-scale bond purchases (QE)

and foreign exchange reserve

managers’ bond purchases,

European Central Bank–United States, 2014
Source: Author.

Central bank only raising policy rate. EM = emerging market; QE = quantitative easing.

Source: Author.

Central bank only raising policy rate. EM = emerging market; QE = quantitative easing.

Policy and Reserve Management Friction: The Conundrum, 2004–05

The so-called conundrum of US bond yields not responding to US policy rates provides a mixed case: bond market operations of reserve managers interacting with the Federal Reserve’s conventional setting of short-term policy rates. Interpretations of this episode vary, and what follows emphasizes the international aspect.

Backus and Wright (2007), for instance, while emphasizing domestic factors, neatly summarize the conundrum in Figure 10.5. The Federal Reserve’s experience of bond yield rates rising rapidly in 1994 led it to make strong efforts to communicate its intentions clearly in the early 2000s. The outcome, however, was a very calm market in which 10-year yields (blue line) and thus key fixed-rate mortgage rates barely moved. Indeed, calculated 10-year forward rates (yellow line) actually fell. Kim and Wright (2005) found that the term premium declined between mid-2004 and mid-2005. The sense was that the Federal Reserve’s policy was not gaining traction in the bond market.

Figure 10.5.US Bond Conundrum, 2004–05


Source: Backus and Wright 2007, figures 2 and 9.

One reading of the evidence highlighted the strong buying of US Treasury and agency bonds by the foreign official sector (Bernanke 2005; Warnock and Warnock 2009). Even as the Federal Reserve raised its federal funds target from 1 percent in 2004 to 4.25 percent in 2006 and 5.25 percent in 2007, foreign central banks raised their holdings of US bonds from $1.2 trillion to $2.3 trillion between June 2004 and June 2007 (McCauley and Rigaudy 2011). Because most mortgages in the United States have fixed rates, the lack of response of bond yields to policy rates was problematic. Only in hindsight is it easy to suggest that the Federal Reserve could have responded by selling bonds out of its portfolio of $700 billion in Treasury securities (Turner 2013).

Reinforcing Bond Market Operations: Quantitative Easing and Emerging Market Reserve Investment, 2009–13

Bond market operations can be reinforcing, as during 2009–13. In Figure 10.6, panel 1, the widening of the blue area in 2009 marks the beginning of large-scale bond purchases by the major advanced economy central banks. Foreign exchange reserves invested in major government bond markets—mostly by emerging market economies (red area)—had already grown substantially in the 2000s and, after a drawdown in 2008, resumed growing in 2009. The share of reserves invested in government bonds is estimated from reported holdings for dollar and sterling and assumed to be 80 percent for euro and yen (see McCauley and Rigaudy 2011 on the investment of US dollar reserves). Whatever the accuracy of the estimation, major central banks surely joined emerging market central banks in buying bonds in quantity in 2009–13.

Figure 10.6.Officials Hold a Big Share of SDR Currency Government Bonds1

(Trillions of US dollars)

Sources: Bank of Japan flow of funds accounts; Board of Governors of the Federal Reserve, flow of funds accounts; European Central Bank; IMF, Currency Composition of Official Foreign Exchange Reserves; Japan Ministry of Finance; national data; Thomson Reuters Datastream; U.K. Debt Management Office; U.K. Office for National Statistics; US Department of the Treasury; and Bank for International Settlements staff calculations.

1 Different valuation methods based on source availability.

2 Covers the euro area, Japan, the United Kingdom, and the United States; for the euro area, Japan, and the United Kingdom, converted into US dollars using Q2 2015 constant exchange rates.

3 For the United States, total marketable Treasury securities, excluding agency debt.

4 For euro- and yen-denominated reserves, 80 percent is assumed to be government debt securities; for dollar-denominated reserves, as reported by the US Treasury International Capital System; for sterling-denominated reserves, holdings by foreign central banks.

5 For the euro area, national central bank holdings of general government debt and European Central Bank holdings under the Public Sector Purchase Programme and the Securities Market Programme.

6 Agency debt includes mortgage pools backed by agencies and government-sponsored enterprises (GSEs) as well as issues by GSEs; total outstanding Treasury securities are total marketable Treasury securities.

Thus, despite talk of currency wars (see next subsection below), central banks stood together on the bid side of the bond market. Official holdings reached 40 percent of the bonds of the US, euro area, Japanese, and U.K. governments. Of course, the intention of foreign exchange reserve managers differed from that of central banks engaged in quantitative easing; and reserve managers’ bond purchases had none of the regular rhythm of the policy-driven bond purchases. That said, knowingly or not, emerging market reserve managers’ purchases of bonds in reserve currencies could only have reinforced the effect of quantitative easing in lowering bond yields in the key currencies, especially the dollar.

The fact that the combination of quantitative easing and reserve management has led to a bond market dominated by official holders is most evident and most readily measured in the US bond market. At the end of the third quarter of 2015, official holders—both foreign official institutions (Figure 10.6, panel 2, red area) and the Federal Reserve (gold area)—held more than half of all outstanding US Treasury bonds. If the US public debt stock is defined more broadly to include agency bonds, the official holding of this $20 trillion stock is still above 40 percent, which reflects a very heavy official footprint in the world’s biggest sovereign bond market.

During this period of reinforcing bond market operations, however, some policies were at cross-purposes. A case in point is Brazil, where the corporate sector was increasing its dollar debt even as the central bank raised the policy rate (see discussion in next section).

Policy at Cross-Purposes: Brazil–United States, 2010–11

In 2010–11, authorities in Brazil, which had a booming economy, sought to restrain inflation by raising the short-term policy rate. In particular, the Central Bank of Brazil raised its SELIC policy rate (the short-term interest rate) from 8.75 percent in April 2010 to 12.5 percent in July 2012. This was a notable tightening in a world in which many central banks were avoiding widening the gap between their policy rates and the Federal Reserve’s interest on excess reserves of 25 basis points (Figure 10.1). McCauley, McGuire, and Sushko (2015b) present econometric evidence that shows that dollar credit grew faster in this period where such interest rate gaps were wider. For its part, the Federal Reserve started its $600 billion large-scale asset (bond) purchase (LSAP2) program in November 2010 and Operation Twist (which bought long-term Treasury bonds with the proceeds of sales of short-term Treasury securities) in September 2011.

Brazilian firms tried to sidestep the tightening of monetary policy at home by increasing their dollar bond issuance (Figure 10.7; also see McCauley, McGuire, and Sushko 2015a). The stock of bonds sold through offshore financing subsidiaries (shown by the yellow area in Figure 10.7) grew faster than the stock of bonds sold by firms resident in Brazil (shown in blue; see also Avdjiev, Chui, and Shin 2014). The Brazilian authorities responded with taxes on offshore borrowing of short maturity and succeeded in increasing the maturity of dollar credit (Pereira da Silva 2013; Barroso, Pereira da Silva, and Soares 2013). Still, dollar credit, especially through the bond market, rose very quickly in the face of the tightening of Brazilian monetary policy.

Figure 10.7.Outstanding Dollar Credit to Brazilian Nonfinancial Borrowers

(Billions of US dollars)

Source: McCauley, McGuire, and Sushko 2015a.

1 US dollar loans to nonbank residents of Brazil.

2 Outstanding US dollar international bonds issued by nonbank residents of Brazil.

3 Outstanding US dollar international bonds issued by offshore affiliates of nonbanks with a parent entity headquartered in Brazil.

Conundrum II? Euro Area–United States in 2014

Many observers, for instance El-Erian (2015),4 read the evidence presented in Figure 10.4, panel 2 as pointing to the influence of the euro area bond market on the US bond market in 2014. While the nominal yield gap between German and US bonds has not been so wide in recent memory, the negative term premiums in the euro area bond market in 2014 seemed to have led their US dollar counterparts down. Mojon and Pegoraro (2014) present three-year term premiums for German bunds and US Treasuries and reach the same conclusion. Formal econometric work on this relationship is under way, but it is a plausible conjecture that 2014 saw “reverse causation” in trans-Atlantic bond markets; that is, the euro area bond market led and the US bond market followed. This conjecture raises the broader question of policy frictions in the global bond market.

Quantitative Tightening? Emerging Market Bond Sales in 2015

In 2015 policymakers lined up on opposite sides of the global bond market, with the euro area and Japan as bond buyers against emerging markets as bond sellers. The former were purchasing their own domestic bonds, whereas the latter sold a diversified portfolio of key currency bonds, mostly dollar bonds but also euro and yen bonds.5 Again, private investors’ treatment of these bonds as close substitutes allows us to consider this scenario as one with officials on both the buy and sell sides of the global bond market.

Finer data than those in Figures 10.2 and 10.6, panel 1, suggest a drawdown of reserves in emerging markets. Moreover, this ongoing reserve drawdown looks broadly based. In 2008, the renminbi was temporarily stable against the dollar; recently, the renminbi, in tandem with other emerging market currencies, has been under downward pressure against the dollar. And while estimates vary, China’s foreign exchange reserves are generally thought to have fallen in 2015.

Again, any bond sales by emerging market central banks would occur as a by-product of currency management. In particular, as the dollar appreciates, emerging market currencies tend to fall against it; indeed, they need to depreciate against the dollar just to keep the nominal effective exchange rate constant. But such depreciation against the dollar can induce hedging by firms that have borrowed dollars, and emerging market nonbanks have borrowed $3.3 trillion (McCauley, McGuire, and Sushko 2015a). Hedging puts further downward pressure on the domestic currency and, to the extent that the central bank counters this pressure, it sells reserves and eventually sells bonds (McCauley 2015; McCauley and Shu 2016).

Bond sales by foreign exchange reserve managers in 2015 are widely seen as having thrown the large US bond market out of kilter (Sundaresan and Sushko 2015). As emerging market central banks sold dollars to limit depreciation of their currencies, they seem to have sold US Treasury bonds to raise dollar cash. As they sold bonds, dealers were reluctant to increase their holdings in anticipation of selling them in turn to long-term investors. As a result, US Treasury yields at the key 10-year maturity have risen above the generic private sector yield represented by 10-year interest rate swaps (Figure 10.8).6 Such a configuration of yields is so anomalous that only a brave speculator would take the risk of betting on a return to normal: emerging market central bank sales of US Treasuries could accelerate and increase the dislocation of yields.

Figure 10.8.Foreign Official US Treasury Holdings and 10-Year Interest Rate Swap Spreads1

(Billions of US dollars)

Sources: Bloomberg, Datastream, US Treasury International Capital transactions data; BIS calculations.

1 Monthly average of daily observations.

Note that the anomaly has not appeared in the euro bond market, where the central bank continues to buy government bonds apace. This dislocation may prove to be just the first symptom of divergent operations in the global bond market by major central banks.

Looming Transatlantic Divergence in Bond Market Operations?

More remote is the possibility that the Federal Reserve will sell bonds, or at least let them run off at maturity without replacement, even as the Eurosystem and the Bank of Japan continue to buy domestic bonds. This possibility does not seem imminent: the Federal Reserve has signaled that it will not stop rolling over its bond portfolio (which entails tens of billions in gross purchases per month) until sometime after its first hike in the short-term policy rate. And a backdrop of official purchases of euro and yen bonds could actually ease the market’s response to the Federal Reserve’s initial rate hikes. Outright sales from the bond portfolio could come later still. (Similarly, the Bank of England has said that it will begin to sell gilts well after the lift-off of its short-term policy rate.)

However, if the possibility of these major central banks finding themselves on opposite sides of the global bond market is considered, there is a case for central banks taking the view that this market is a sort of global commons. Central banks might well take into account their interactions in the global bond market, lest their own policies prove to be constrained by market dynamics to which their earlier actions may have contributed.


This chapter analyzes actual and prospective interactions among central banks operating in the global bond market. Both unconventional monetary policy and reserve management involve purchases (and potentially sales) of bonds. High substitutability of bonds in private portfolios means that the total purchases by central banks have an impact on global bond yields and, thereby, on global financial conditions.

Bond purchases compress the term premium in the target market and induce a shift of borrowing into it. Firms outside the United States have added to their dollar debt significantly in the aftermath of the global financial crisis, especially through the issuance of dollar-denominated bonds. Bond purchases in one market can also compress the term premium in other bond markets that are globally integrated.

Central bank operations in bond markets can be on the same side, reinforcing each other, or on opposite sides, at cross-purposes. After the global financial crisis, major central banks and major reserve managers stood shoulder to shoulder as buyers of bonds. Ironically, this was at times a symptom of policy friction: emerging market central banks intervened to limit currency appreciation—and incidentally bought reserve currency bonds, partly in response to behavior induced by quantitative easing.

At present, large bond purchases by the European Central Bank and the Bank of Japan may be countered to some extent by dollar bond sales by major reserve holders resisting domestic currency depreciation. With the Federal Reserve signaling higher policy rates followed at some point by a rundown of its holdings of bonds, policy frictions could arise in the global bond market. As a matter of enlightened self-interest, policymakers might well internalize the effects of their bond market operations on other policymakers.

Models that do not have bond markets and those that have bond markets that anchor yields solely on expected future policy rates cannot be used to analyze the potential gains from cooperation.


    AdrianT.R. K.Crump and E.Moensch.2013. “Pricing the Term Structure with Linear Regressions.” Journal of Financial Economics110 (1October): 11038.

    AdrianT. and M.Fleming.2013. “The Recent Bond Market Selloff in Historical Perspective.” Federal Reserve Bank of New York Liberty Street Economics (August5).

    AvdjievS.M.Chui and H. S.Shin.2014. “Non-financial Corporations from Emerging Market Economies and Capital Flows.” BIS Quarterly Review (December): 67–77.

    BackusD. K. and J. H.Wright.2007. “Cracking the Conundrum.” Brookings Papers on Economic Activity1: 293329.

    Bank for International Settlements (BIS). 2015. 85th Annual Report. Basel: BIS.

    BarrosoJ. B.L.Pereira da Silva and A.Soares.2013. “Quantitative Easing and Related Capital Flows into Brazil: Measuring Its Effects and Transmission Channels through a Rigorous Counterfactual Evaluation.” Working Paper 313 (July) Central Bank of BrazilBrasilia.

    BauerM. and C.Neely.2014. “International Channels of the Fed’s Unconventional Monetary Policy.” Journal of International Money and Finance44 (June): 2446.

    BernankeB.2005. “The Global Saving Glut and the U.S. Current Account Deficit.” Homer Jones LectureFederal Reserve Bank of St. Louis, St. LouisMissouriApril14.

    BernankeB.2012. “U.S. Monetary Policy and International Implications.” Speech at “Challenges of the Global Financial System: Risks and Governance under Evolving Globalization” seminar sponsored by the Bank of Japan/International Monetary FundTokyo, JapanOctober14.

    BernankeB.2013. “Long-Term Interest Rates.” Speech at “Annual Monetary/Macroeconomics Conference: The Past and Future of Monetary PolicyFederal Reserve Bank of San FranciscoSan FranciscoCaliforniaMarch1.

    BernankeB.V.Reinhart and B.Sack.2004. “Monetary Policy Alternatives at the Zero Bound: An Empirical Assessment.” Brookings Papers on Economic Activity2: 1100.

    BlommesteinH. J. and P.Turner.2012. “Interactions between Sovereign Debt Management and Monetary Policy” in Threat of Fiscal Dominance? BIS Paper 65 (March): 21337.

    BorioC. and R.McCauley.1996. The Economics of Recent Bond Market Volatility. BIS Economic Paper 45 (July).

    ChenQ.A.FilardoD.He and F.Zhu.2015. “Financial Crisis, Unconventional Monetary Policy and International Spillovers.” BIS Working Paper 494 (March) Basel.

    D’AmicoS. and T. B.King.2013. “Flow and Stock Effects of Large-Scale Treasury Purchases: Evidence on the Importance of Local Supply.” Journal of Financial Economics108 (2May): 42548.

    FilardoA. and B.Hofmann.2014. “Forward Guidance at the Zero Lower Bound.” BIS Quarterly Review (September): 3753.

    GagnonJ.M.RaskinJ.Remaché and B.Sack.2011. “The Financial Market Effects of the Federal Reserve’s Large-Scale Asset Purchases.” International Journal of Central Banking7 (1): 343.

    Gerlach-KristenP.R.McCauley and K.Ueda.2016. “Currency Intervention and the Global Portfolio Balance Effect: Japanese Lessons.” Journal of the Japanese and International Economies39: 116.

    GreenwoodR.S. G.HansonJ. S.Rudolph and L.Summers.2014. “Government Debt Management at the Zero Lower Bound.” Working Paper 5Hutchins Center for Fiscal and Monetary Policy, Brookings InstitutionWashingtonSeptember30.

    HeD. and R.McCauley.2013. “Transmitting Global Liquidity to East Asia: Policy Rates, Bond Yields, Currencies and Dollar Credit.” BIS Working Paper 431 (October) Basel.

    HofmannB. and B.Bogdanova.2012. “Taylor Rules and Monetary Policy: A Global ‘Great Deviation’?BIS Quarterly Review (September): 3749.

    HofmannB. and E.Takáts.2015. “International Monetary Spillovers.” BIS Quarterly Review (September): 10518.

    HördahlP. and O.Tristani.2014. “Inflation Risk Premia in the Euro Area and the United States.” International Journal of Central Banking10 (3September): 147.

    International Monetary Fund (IMF). 2014. 2014 Spillover Report.International Monetary FundWashington, DC.

    KimD. H. and J. H.Wright.2005. “An Arbitrage-Free Three-Factor Term Structure Model and the Recent Behavior of Long-Term Yields and Distant-Horizon Forward Rates.” Board of Governors of the Federal Reserve System Finance and Economics Discussion Series 2005-33.

    KrishnamurthyA. and A.Vissing-Jorgensen.2011. “The Effects of Quantitative Easing on Interest Rates: Channels and Implications for Policy.” Brookings Papers on Economic Activity2: 21587.

    Lo DucaM.G.Nicoletti and A. V.Martinez.2014. “Global Corporate Bond Issuance: What Role for US Quantitative Easing?European Central Bank Working Paper 1649 (March) Frankfurt-am-Main.

    McCauleyR.2015. “Capital Flowed Out of China through BIS Reporting Banks in Q1 2015.” BIS Quarterly Review (September): 2829.

    McCauleyR. and G.Jiang.2004. “Diversifying with Asian Local Currency Bonds.” BIS Quarterly Review (September): 5166.

    McCauleyR.P.McGuire and V.Sushko.2015a. “Dollar Credit to Emerging Market Economies.” BIS Quarterly Review (December): 2741.

    McCauleyR.P.McGuire and V.Sushko.2015b. “Global Dollar Credit: Links to US Monetary Policy and Leverage.” Economic Policy30 (82) (April): 187229.

    McCauleyR. and J.-F.Rigaudy.2011. “Managing Foreign Exchange Reserves in the Crisis and After.” In Portfolio and Risk Management for Central Banks and Sovereign Wealth Funds.” BIS Papers58 (October): 1947.

    McCauleyR. and C.Shu.2016. “Dollars and Renminbi Flowed out of China.” BIS Quarterly Review (March): 2627.

    McCauleyR. and K.Ueda.2009. “Government Debt Management at Low Interest Rates.” BIS Quarterly Review (June): 3551.

    MiyajimaK.M.Mohanty and J.Yetman.2014. “Spillovers of US Unconventional Monetary Policy to Asia: The Role of Long-Term Interest Rates.” BIS Working Paper 478 (December) Basel.

    MojonB. and F.Pegoraro.2014. “Decoupling Euro Area and US Yield Curves.” Banque de France Rue de la Banque1 (December).

    NeelyC.2015. “The Large-Scale Asset Purchases Had Large International Effects.” Journal of Banking and Finance52: 10111.

    ObstfeldM.2015. “Trilemmas and Tradeoffs: Living with Financial Globalization.” BIS Working Paper 480 (January) Basel.

    ObstfeldM. and K.Rogoff.2002. “Global Implications of Self-Oriented National Monetary Rules.” Quarterly Journal of Economics117 (2May): 50335.

    Pereira da SilvaL.2013. “Global Dimensions of Unconventional Monetary Policy—An EME Perspective.” In Federal Reserve Bank of Kansas City Jackson Hole Economic Policy Symposium Proceedings37385.

    ReyH.2013. “Dilemma Not Trilemma: The Global Financial Cycle and Monetary Policy Independence.” In Federal Reserve Bank of Kansas City Jackson Hole Economic Policy Symposium Proceedings285333.

    RogersJ. H.C.Scotti and J. H.Wright.2014. “Evaluating Asset-Market Effects of Unconventional Monetary Policy: A Cross-Country Comparison.” Economic Policy29 (80October): 74999.

    SobrunJ. and P.Turner.2015. “Bond Markets and Monetary Policy Dilemmas for the Emerging Markets.” BIS Working Paper 508 (August) Basel.

    SundaresanS. and V.Sushko.2015. “Recent Dislocations in Fixed Income Derivatives Markets.” BIS Quarterly Review (December): 89.

    TaylorJ.2013. “International Policy Coordination and the Great Deviation.” Paper presented at the American Economic Association meeting in San DiegoCaliforniaJanuary 4–6.

    TurnerP.2013. “Benign Neglect of the Long-Term Interest Rate.” BIS Working Paper 403 (February) Basel.

    TurnerP.2014. “The Global Long-Term Interest Rate, Financial Risks and Policy Choices in EMEs.” BIS Working Paper 441 (February) Basel.

    Van RixtelA. and A.Villegas.2015. “Equity Issuance and Share Buybacks.” BIS Quarterly Review (March): 2829.

    WarnockF. E. and S. C.Warnock.2009. “International Capital Flows and U.S. Interest Rates.” Journal of International Money and Finance28: 90319.

    WinklerR.M.Sachdeva and G.Saravelos.2015. “The ‘Great Accumulation’ Is Over: FX Reserves Have Peaked, Beware QT.” Deutsche Bank Market Research Special ReportSeptember1London.

The author thanks participants in the symposium for discussion, especially Claudio Borio, Rex Ghosh, Peter Hördahl, Pat McGuire, Catherine Schenk, Hyun Song Shin, Vladyslav Sushko, Philip Turner, Goetz von Peter, and William White. The research assistance of Bilyana Bogdanova, José-Maria Vidal-Pastor, and Jhuvesh Sobrun is gratefully acknowledged. Views expressed are those of the author and not necessarily those of the Bank for International Settlements.

Nevertheless, market participants may attach importance to the size, or the relative size, of central bank balance sheets. See the discussion in He and McCauley 2013.

Not shown in Figure 10.3, panel 1, are dollar bonds of US holding companies, US-chartered banks, US finance companies, US state and municipal governments, and non-US bank and non-bank financial borrowers.

Debt managers are also important official players in bond markets, capable of responding to market developments (McCauley and Ueda 2009; Blommestein and Turner 2012; Greenwood and others 2014).

Financial Times, March 9, 2015, “Outcome of Policy Tug-of-War Proving Hard to Predict.”

Sales of euro area or Japanese government bonds by foreign official holders might make it easier for the respective central banks to hit their buying targets, but such sales would tend to reduce the impact of bond buying on yields. Note also that, to the extent that such euro and yen bond sales are matched by sales of dollar bonds, foreign official holders become one of the channels for spillovers.

The US Treasury International Capital transactions data in Figure 10.8 show sales, while the holdings data do not ( Usually the holdings data are regarded as higher quality, but the transactions data seem to be consistent with market participants’ perception of official selling.

    Other Resources Citing This Publication