Exploring the performance of US international bond mutual funds

• Published date: 01 November 2023
• Author: Jonathan Fletcher,Elizabeth Littlejohn,Andrew Marshall
• Date: 01 November 2023
• DOI: http://doi.org/10.1111/fire.12355

DOI: 10.1111/fire.12355

ORIGINAL ARTICLE

Exploring the performance of US international

bond mutual funds

Jonathan Fletcher Elizabeth Littlejohn Andrew Marshall

Department of Accounting and Finance,

University of Strathclyde,Glasgow, UK

Correspondence

Jonathan Fletcher,Department of Accounting

and Finance, University of Strathclyde,

Stenhouse Building, 199 Cathedral Street,

Glasgow,G4 0QU, UK.

Email:j.fletcher@strath.ac.uk

Abstract

We use a Bayesian regime switching approach to examine

the performance enhancement of adding US international

bond funds to a domestic bond universe pre and post the

Global Financial Crisis (GFC) during January 1999 and May

2022. We find that the international bond funds provide

large significant performance enhancement pre the GFC,

with an increase in Certainty Equivalent Return (CER) per-

formance of 0.595% (monthly), but none post the GFC.

The performance enhancement pre GFC is driven by Large

Emerging Market bond funds, which is likelyfueled by a sub-

stantial drop in the Emerging Market central bank policy

rates pre GFC .

KEYWORDS

Bayesian analysis, international bonds funds, performance, regime-

switching

JEL CLASSIFICATION

G11, G12

1INTRODUCTION

Since the classic studies of international investing of Grubel (1968), Levy and Sarnat (1970), and Solnik (1974), much

of the subsequent literature has focused on international equity portfolio investment(Hodrick & Zhang, 2014). Fewer

papers have examined the performance of international bond portfolios. An early study by Levyand Lerman (1988)

examinedthe performance enhancement of international investing in developed equity and bond markets from United

States perspective. They find that the performance enhancement is greater in bond marketsthan equity markets due

to the lower correlations in bond markets.Glen and Jorion (1993) and Eun and Resnick (1994) extend this analysis and

Financial Review. 2023;58:765–782. wileyonlinelibrary.com/journal/fire ©2023 The Eastern Finance Association. 765

766 FLETCHERET AL.

examine the performance of currency hedging in international stock and bond portfolios. Both of these studies find

significant performance enhancement in currency hedging when investing in international bonds.1

Studies byBekaert and De Santis (2021), Briere et al. (2016), Fletcher et al. (2019), Hansson et al. (2009), Liu (2016),

and Randl et al. (2022) among others, provide further evidence on the performance of international bond portfolios.

They find support for investing in international bonds including corporate bonds and emerging marketbonds. These

studies provideevidence of the performance enhancement of direct asset allocation strategies. In this study, we exam-

ine whether international bond funds are able to provide significant performance enhancement relative to a domestic

bond universe.

We evaluate the performance enhancement of adding US international open-end bond mutual funds to a domes-

tic investment universe consisting of US domestic bonds. Our main focus is to adapt the Bayesian regime switching

method of Chan et al. (2019) to examine the performance enhancement of including international bond fund port-

folios in optimal mean-variance portfolios before and after the global financial crisis (GFC) of 2007/2008, given the

changes in centralbank policy rates around that time. The regime-switching approach allows us to identify the regimes

endogenously rather than being specified by the researcher.The regime-switching approach also allows us to exam-

ine whether there are significant differences in performance between regimes. Given these regimes, we then use the

Bayesian approach of Wang (1998), and Li et al. (2003) to estimate and evaluate the performance enhancement of

including international bond funds in optimal mean-variance portfolios in each regime.

Our sample period is between January 1999 and May 2022. We estimate optimal mean-variance portfolios in the

domestic investment universe using four US domestic bond indexes,and an international investment universe which

addssix international bond fund portfolios to the domestic universe. We use international bond fund portfolios formed

by the investment sector (World, World Hedged, Emerging Markets[EM]), and fund size (small, large). We select the

optimalportfolios for a g ivenrisk tolerance level, and estimate the performance enhancement by the bond funds as the

increase in Certainty Equivalent Return (CER)between the optimal portfolios in international and domestic universes.

We rule out short selling when estimating the optimal portfolios.

There are two main findings in our study. First, we find that the performance enhancement of international bond

funds only occurs pre the GFC. The international bond funds deliver a large increase in CER performance of 0.595%

(monthly). The performance enhancement is driven by the EM/Large fund portfolio. Second, the EM/Large fund port-

folio has a much higher posterior mean average excessreturns than the domestic bonds and other international bond

fund portfolios pre the GFC, and lower correlations. The performance of the EM bond funds is likely drivenby a sharp

drop in EM central bank policy ratespre the GFC.

Our study makes two main contributions to the literature.First, we complement the empirical evidence of the per-

formance of international bond portfolios such as Bekaert and De Santis (2021), Briere et al. (2016), Fletcher et al.

(2019), Hansson et al. (2009), Liu (2016), and Randl et al. (2022) among others by looking at the performance of inter-

national bond funds.2Second, we extend the US bond fund performance literature such as Blake et al. (1993), Chen

et al. (2010), Elton et al. (1995), Ferson et al. (2006), and Moneta (2015) among others by focusing on international

bond funds and examining the performance providedby such funds in optimal mean-variance portfolios.

Our paper is organized as follows. Section 2presents the research method used in our study. Section 3describes

the data. Section 4reports the empirical results and the final section concludes.

1These studies are dependent on the choice of currency.Hentschel et al. (2002) evaluate the performance of United States, United Kingdom, and German

government bonds using the numeraire portfolio approach of Long(1990). The attraction of this approach is that the results are invariant to the choice of

currency.

2Recentstudies by Harvey (2017)andHouetal.(2020)highlight theimportance of replication studies in Finance, which is common in other fields of science.