Interest rate risk and the creation of the Monetary Policy Committee: Evidence from banks’ and...

Journal of Economics and Business 71 (2014) 45– 67

Contents lists available at ScienceDirect

Journal of Economics and Business

Interest rate risk and the creation of theMonetary Policy Committee: Evidence frombanks’ and life insurance companies’ stocksin the UK

Stephanos Papadamoua,∗, Costas Siriopoulosb

a Department of Economics, University of Thessaly, Korai 43, Postal Code 38333, Volos, Greeceb Department of Business Administration University of Patras, Rio, Patras, Greece

a r t i c l e i n f o

Article history:Received 30 November 2012Received in revised form 26 August 2013Accepted 17 September 2013

JEL classification:G21G28E44

Keywords:BankingInterest rate riskMonetary policyStock markets

a b s t r a c t

This paper investigates the effect that the creation of the Mon-etary Policy Committee (MPC) has had on the interest rate riskwhich banks and life insurance companies face in the UK. By meansof GARCH-M methodology, the stock returns are modelled on theCAPM and the Fama-French asset-pricing models, augmented withinterest rate risk factors and referring to short- and long-termrates. Our results indicate that in the period before the Bank ofEngland (BoE) was granted operational independence, changes inthe level and volatility of interest rates significantly affected thestock returns of these companies. These effects have diminishedsince the MPC’s creation in May 1997. In parallel, since the MPC’screation, macroeconomic uncertainty, as proxied by the MPC dis-sents, coexisted with significant effects on the short-term interestrate risk which banks and life insurance companies face. Theseresults should be of interest to both analysts and policy-makerswith respect to financial stability.

© 2013 Elsevier Inc. All rights reserved.

1. Introduction

Interest rate risk is one of the most significant risks which banks and life insurance companies face,largely because of (1) a duration mismatch between their assets and liabilities and (2) the significant

∗ Corresponding author. Tel.: +30 2421074963; fax: +30 2421074772.E-mail addresses: [email protected], [email protected] (S. Papadamou), [email protected] (C. Siriopoulos).

0148-6195/$ – see front matter © 2013 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.jeconbus.2013.09.001

dx.doi.org/10.1016/j.jeconbus.2013.09.001

http://www.sciencedirect.com/science/journal/01486195

http://crossmark.crossref.org/dialog/?doi=10.1016/j.jeconbus.2013.09.001&domain=pdf

mailto:[email protected]



dx.doi.org/10.1016/j.jeconbus.2013.09.001

46 S. Papadamou, C. Siriopoulos / Journal of Economics and Business 71 (2014) 45– 67

volatility of interest rates. Consequently, the inadequate risk management practices of banks and lifeinsurance institutions can often lead to the structural vulnerability of the entire financial system andthus to financial crises.

There is a large body of literature that provides empirical evidence of a strong negative relationbetween stock returns of financial firms and interest rates (e.g., Carson, Elyasiani, & Mansur, 2008;Elyasiani & Mansur, 2003; Flannery, Hameed, & Harjes, 1997; Flannery & James, 1984; Lloyd & Shick,1977; Santomero & Babbel, 1997; Viale, Kolari, & Fraser, 2009). Choi, Elyasiani, and Saunders (1996),Allen and Jagtiani (1997), Hirtle (1997), and Benink and Wolff (2000) conclude, however, that interestrate sensitivity decreased in the late 1980s and early 1990s because of the availability of interest ratederivatives contracts that can be used for hedging purposes. Additionally, some authors argue that theinterest rate dependence of financial stocks is time-variant and depends on economic conditions andmonetary policy regimes (e.g., Booth, Officer, & Henderson, 1985; Brewer, Carson, Elyasiani, Mansur,& Scott, 2007; Choi, Elyasiani, & Kopecky, 1992; Elyasiani & Mansur, 1998; Ferrer, González, & Soto,2010; Kane & Unal, 1988; Korkeamäki, 2011; Yourougou, 1990).

An important point that should be taken into account, however, when we investigate the effectof interest rate risk on banks and life insurance companies is the way in which monetary policyis conducted. In recent years, this conduct may have had a significant effect on interest rate riskexposures of banks and life insurance companies, to the extent that monetary policy implementationhas changed significantly, along with the banking industry.

In most developed countries, central banks conduct their monetary policy either by targeting ashort-term interest rate or by setting an official interest rate for their open market operations. Thesepolicy rates anchor the entire term structure of interest rates. Nowadays, it is widely accepted thatthe ability of a central bank to affect the economy depends critically on its ability to influence marketexpectations about the future path of overnight interest rates and not merely their current level. Thereason is simple. Few, if any, economic decisions hinge on the overnight bank rate cost and availabil-ity of bank reserves. Consequently and in order to significantly affect expectations of future policyrate changes, central banks follow institutional reforms for greater transparency in monetary policyimplementation framework to reinforce their credibility (see the empirical evidence in Eijffinger &Geraats, 2006).

More specifically, central bank independence (CBI) through its effect on central bank transparency(CBT; Geraats, 2002) and credibility (CBC; Rogoff, 1985) increases market operators’ acceptance ofmonetary policy and the ability of the central bank to fulfil its objectives. The Bank of England adoptedinflation targeting in December of 1992 and was granted operational independence with the creationof Monetary Policy Committee (MPC) in May 1997. Therefore, a crucial question arises: ‘How did thecreation of the Monetary Policy Committee (MPC) and its members’ dissents regarding policy rateaffect the interest rate risk which financial services companies faced?’ Inflation expectations reflectedin long-term rates are anchored, under high level of CBT. Therefore changes in the level of long-term rates and/or in the volatility of short-term rates (implying term structure movements) are notexpected to affect investors’ and financial firms’ behaviour. According to our knowledge there hasbeen no empirical evidence in this regard, a gap this paper attempts to fill.

The literature on the effects of the creation of the BoE’s MPC is still limited. For an excellent reviewof the effect of the MPC’s creation on the macroeconomic performance of the UK see King (2002, 2007).Chadha and Nolan (2001) use data before and after MPC’s creation to assess whether a change to amore transparent monetary policy caused an increase in the volatility of interest rates in financialmarkets, but find no evidence in support of this hypothesis. In the same vein, Clare and Courtenay(2001) found that the independence of the BoE has influenced the reaction speed of exchange ratesand futures contracts to macroeconomic news. Tuysuz (2011) argued that even if agents have beenable to forecast policy rates decisions more accurately since May 1997, there is no clear impact of BoEindependence on the reaction of interest rate levels to macroeconomic and monetary news.

Studying a number of central banks, Cihak (2006), Klomp and de Haan (2009) argue for a positiverelation between CBI and financial stability, but no attention has been paid to an asset-pricing frame-work. Additionally, according to Alesina and Summers (1993), the variability of interest rates alsoreduces with higher CBI. Empirical evidence for the effect of CBI on interest rate risk facing financialservices companies remains under investigation.

S. Papadamou, C. Siriopoulos / Journal of Economics and Business 71 (2014) 45– 67 47

Therefore our study contributes to the existing literature by evaluating the effect of short-termand long-term interest rate levels and changes in conditional volatilities on banks’ and life insurancecompanies’ stock returns in asset-pricing models with an exogenous structural break in May 1997.Moreover, the correlation between the MPC dissents in voting records and the interest rate risk expo-sure are investigated. There are times where the state of the economy is difficult to read and naturallythere are differences of interpretation which lead to split votes. Equally, there are times where thenature of the economic shocks is not in dispute and the response of the MPC is agreed by all mem-bers. These events might have a significant impact on the ability of investors and financial servicescompanies to forecast interest rates. Therefore, an investigation of interest rate effects that takes intoaccount institutional reforms for greater transparency in monetary implementation can provide use-ful answers to important research questions concerning financial management strategies, bankingregulation, and conduct of monetary policy.

The rest of the paper is organised as follows. Section 2 provides the main theoretical argumentsconcerning the effect of interest rate changes on banks’ and life insurance companies’ equity returns.Section 3 offers a brief survey of previous literature. Section 4 provides information about the dataand the main methodology of our research. Section 5 presents and discusses the results. Section 6concludes the paper.

2. Theoretical reasons for interest rate effects on banks and life insurance companies

In the financial literature it is widely believed that equity returns of financial firms are relatedto inflation and interest rates because of nominal contracting, i.e., most contractual arrangementsare specified in nominal non-inflation-adjusted returns. The nominal contracting hypothesis (NCH,hereafter), introduced by Kessel (1956), Bach and Ando (1957), postulates that a firm’s holdings ofnominal assets and nominal liabilities can affect common stock returns through the wealth redis-tribution effects caused by unexpected inflation that redistributes wealth from creditors to debtors.Hence, stockholders of firms with a higher level of nominal liabilities than nominal assets should ben-efit from unexpected inflation, and the opposite should be true for the equity value of firms in thereverse position.

According to Flannery et al. (1997) the interest rate sensitivity of a firm’s common stocks shouldbe related to the maturity composition of the firm’s nominal assets and liabilities. Banks’ and lifeinsurance companies’ financial assets represent a significant amount of their total assets, comparedwith other categories of companies. Therefore duration mismatch in their assets and liabilities makethem sensitive to interest rate changes. The degree of exposure of a life insurance company to interestrate risk is determined by the duration gap between its assets and liabilities, and the convexity inthe market value of its net worth. The duration gap and net worth convexity can be chosen by thefirm managers, subject to a number of limitations. These include the need to accommodate customerdemand, and firm’s access to, and costliness of, markets for hedging interest rate risk, such as options,futures, and swaps (Santomero & Babbel, 1997).

For banks the credit quality of assets such as loans may also be affected by changes in inflation andinterest rates. Mankiw (1986) argued that interest rate increases may decrease relative credit demandfrom good credit risks while increasing relative credit demand from bad credit risks. This would raisethe likelihood of adverse selection for banks with significant negative effect on their equity.

Another reason for change in the effects of interest rates on financial services companies is theinstitutional reforms concerning greater monetary policy transparency. The theoretical explanation ofthis change is as follows. Greater transparency in monetary policy conduct can reduce uncertaintyabout interest rates and consequently reduce forecast errors and sensitivity to interest rate risk. Thereis also a possibility to undertake more risk in a low-volatility environment, however. Our study triesto shed light on this direction of research.

Eijffinger, Geraats, and Van der Cruijsen (2006) show theoretically that greater transparency shouldenhance flexibility to allow a reduction in policy and short-term interest rates without increasing long-term nominal interest rates. In addition, an improved reputation would reduce inflation expectationsand thereby long-term nominal interest rates. According to standard theories of the term structureintermediate and long-term rates should depend mostly on the public’s expectations of future central


bank policy.1 Woodford (1999) argues that if a central bank has established a reputation for eitherkeeping interest rates at the same level for an extended period of time or for implementing successivesmall changes after an initial move, it effectively has committed itself to a path of future short rates.Market participants will incorporate these beliefs into their expectations of future short-term rates,and the term structure of interest rates will reflect the central bank’s expected future policy ratechanges.2 Therefore the effect of long-term rates on banks’ and life insurance companies’ stock returnsshould depend significantly on the degree of CBT and CBC, which are positive functions of CBI. CBT,in the context of Geraats (2002), is a significant factor contributing to greater predictability of futureinterest rates and this may affect managers’ behaviour regarding interest rate risk.

3. Literature review

Research on the relationship between interest rates and bank equity returns is extensive for theUSA. The first studies applied a two-index model (market and interest rate factors) assuming constantvariance error terms and found dissimilarities in terms of both magnitude and direction effects (seeamong others Akella & Greenbaum, 1992; Bae, 1990; Flannery & James, 1984; Lloyd & Shick, 1977;Prasad & Rajan, 1995; Scott & Peterson, 1986; Stone, 1974). Kwan (1991) argued that unanticipatedchanges in interest rate levels can affect bank stock returns variably, largely because of differences inthe maturity composition of bank assets and liabilities.

In the late 1980s Kane and Unal (1988) provided evidence against the constancy of conditional vari-ance and in favour of time-variant risk premia in US banking stocks. In this context Song (1994) wasthe first to employ ARCH-type methodology in banking. By applying GARCH-type models Neuberger(1993) and Flannery et al. (1997) found that interest rate volatility is time-variant and its variation ispriced into the expected returns of different securities. Employing GARCH-M methodology Elyasianiand Mansur (1998) found that interest rate levels and volatility directly affect the first and the sec-ond moments of bank stock returns distribution, respectively. They also found a negative correlationbetween bank stock volatility and risk premia, indicating a possible agency theory problem.

Another interesting finding in the literature is the asymmetrical interest rate sensitivities thatbanks present during various interest rate cycles (see for instance Chen & Chan, 1989; Akella & Chen,1990). Specifically, Yourougou (1990) found that interest rate sensitivity was low during periods ofinterest rate stability (pre-October 1979) and high during periods of high interest rate volatility.

A theoretical model explaining banks’ decisions on the level of interest rate risk was developedby Deshmukh, Greenbaum, and Kanatas (1983). More specifically, they showed that an increase ininterest rate uncertainty leads the intermediary to reduce its exposure to sensitive interest rate assets,thereby offering decreased asset-transformation and more brokerage services. A stochastic increasein the interest rates leads, however, to greater asset transformation and fewer brokerage services.Bharati, Nanisetty, and So (2006) provides empirical evidence for this theory. Their study of US banksindicates that banks hedge against interest rate levels when levels are low, and speculate when levelsare high but the volatility is low.

Brewer and Lee (1990) and Guo (2004) argue that interest rate sensitivity shifts according to eco-nomic conditions and monetary policy strategy. Laopodis (2010) appears to suggest that there hasbeen a consistent dynamic relationship between the conduct of monetary policy and the correspond-ing stock market behaviour during the last three decades in the USA. Along the same lines is a paper byGasbarro and Monroe (2004) on the Australian market. For the UK, Bredin, Hyde, Nitzsche, and O’Reilly(2007), Bredin, Hyde, and O’Reilly (2010) and Gregoriou, Kontonikas, MacDonald, and Montagnoli(2009) show that unexpected changes of monetary policy have a significant impact on aggregate andsectoral stock returns, but they focus only on short-term rates and they do not investigate how thisrelation may be affected by institutional reforms in the monetary policy framework, a gap in theliterature that this paper tries to fill.

1 Chortareas, Stasavage, and Sterne (2002) show that transparency may be associated with lower inflation across a widerange of countries.

2 Goodfriend (1998) refers to this as ‘policy in the pipeline’.


Furthermore, Elyasiani and Mansur (2003) and Kasman, Vardar, and Tunc (2011) found that banks’returns in Japan and Germany are highly sensitive to macroeconomic shocks, such as exchange rateand interest rate shocks, with the latter showing their impact at the volatility level. A Korean studyby Hahm (2004) indicates that significant exposure to interest rate and exchange rate risks by Koreancommercial banks further deteriorated the banking sector’s capital, worsening the Asian financialcrisis in 1997.3 According to Aggarwal, Jeon, and Zhao (2006) Korean bank equity returns were foundto be sensitive to both anticipated and unanticipated changes in interest rates when banks were largelyunder government control.

There is today a growing literature investigating interest rate risk in a broad range of financialinstitutions (FIs; insurance companies, securities firms and banks). Moreover, Elyasiani, Mansur, andPagano (2007) find that return and risk interdependencies across these three types of firms in the USAare significant and size-variant. According to Elyasiani, Mansur, and Wetmore (2010) the equity returnsof the FIs4 follow a GARCH process and should be modelled accordingly. Beirne, Caporale, and Spagnolo(2009) studied banking, financial services and insurance in 16 countries, including various Europeaneconomies, the US and Japan, and found that interest rates and exchange rates have a significant effect(negative and mixed, respectively) in a number of cases.

As far as insurance companies are concerned, Brewer et al. (2007) indicate that US life insurancecompanies’ returns are sensitive to long-term interest rates and that this interest sensitivity variesacross sub-period and across risk-based and size-based portfolios. Carson et al. (2008) examine mar-ket risk, interest rate risk, and interdependencies in returns and return volatilities across three insurersegments (accident and health, life, and property and casualty insurers) within a System GARCH frame-work. Among other results interest rate sensitivity is negative and greatest for life insurers (see alsoCheng, Elyasiani, & Lin, 2010). Significant interest rate sensitivity is also present in the German insur-ance sector which seems to contradict conventional market wisdom that insurances hedge interestrate risks (Czaja, Scholz, & Wilkens, 2010). Park and Choi (2011), investigating US property/liability(P/L) insurers’ stock returns, recognise that insurers’ interest rate sensitivity is closely related to theinsurance industry’s underwriting cycle.

Finally, Ferrer et al. (2010) and Korkeamäki (2011) argue that the introduction of the euro seems tohave weakened the firms’ degree of interest rate risk. The latter study focusing on a significant numberof European stock markets argues that recent growth in European corporate bond markets has playedan important role in enabling firms to better manage their interest rate risk. Another possible reason(not investigated) for this, however, could be the conduct of monetary policy by an independent andtransparent European Central Bank (ECB).

4. Data and methodology

Before we analyse our data set, it should be mentioned that there is a trade-off between usingindividual firm data and using portfolio data. On the one hand, when individual firm data are used, thenoise is high and the results tend to be unduly influenced by individual random shocks. On the otherhand, portfolio or index data mask some of the detailed information provided by individual companydata, even though some authors argue that they produce more reliable results as they wash out thenoise. In this study we investigate both index and individual data to accommodate these arguments.

In our study the sample period extends from January 1989 to December 2012. Monthly data forbanking sector index, life insurance sector index, FT All Share Index and one month Treasury bill ratewere collected from the Reuters Ecowin database. Moreover, we investigate the effect of interest raterisk not only on sectoral data but also on four major British banks (Lloyds, HSBC, Barclays, StandardChartered), and four major life insurance companies (Prudential, St James’s Place, Legal & General,

3 Gounopoulos, Molyneux, Staikouras, Wilson, and Zhao (2012) indicate that US, UK and Japanese banks’ equity returns werenegatively related to changes in foreign currency value during the recent financial crisis (2008–11).

4 FIs in this study incorporate commercial banks, life insurance companies, savings and loans, and real-estate investmenttrusts (REITs). A major finding of this study is that all FI returns considered are highly sensitive to REIT returns.


Table 1Independent British banks and life insurance companies listed in the LSE.

Company Sector Country ofincorporation

Region (UKonly)

Market Mkt Cap £m(30/6/2013)

1 HSBC HLDGS Banks GB London Main market 126,094.12 LLOYDS BANKING GROUP

PLCBanks GB London Main market 45,779.5

3 STANDARD CHARTERED Banks GB London Main market 34,190.84 BARCLAYS Banks GB London Main market 34,063.15 ROYAL BANK OF

SCOTLAND GROUP PLCBanks GB Scotland Main market 16,600.4

6 BANK OF GEORGIA HLDGSPLCa

Banks GB Main market 599.7

7 SECURE TRUST BANK PLCa Banks GB Midlands AIM 334.58 EUROPEAN ISLAMIC

INVESTMENT BANKaBanks GB London AIM 58.4

1 PRUDENTIAL Life insurance GB London Main market 27,431.12 LEGAL & GENERAL GROUP Life insurance GB London Main market 10,067.73 OLD MUTUAL PLCb Life insurance GB London Main market 10,059.64 AVIVA Life insurance GB London Main market 9909.25 STANDARD LIFE PLCa Life insurance GB Scotland Main market 8213.56 ST JAMES’S PLACE Life insurance GB South-west Main market 2700.77 PARTNERSHIP ASSURANCE

GROUP PLCaLife insurance GB London Main market 1948.0

8 CHESNARAa Life insurance GB North-west Main market 287.1

Source of Data: London Stock Exchange.a Listed in LSE after 2004.b Listed in LSE in 1999.

Aviva) listed in the London Stock Exchange (LSE) since 1989.5 Unlike some other major economies,the UK does not have a major stratum of independent local banks. The list is quite short as Britishbanking has been highly consolidated since the early twentieth century. Table 1 presents the Britishbanks and life insurance companies listed in the LSE and ranked by market capitalisation. The reasonfor using mainly monthly rather than daily data is that settlements and clearing delays in banking areless problematic with monthly data relative to daily returns (Baillie & De Gennaro, 1990).

The main hypothesis tested is whether MPC creation through transparent policy and reduced uncer-tainty about monetary policy actions (no policy dissents) can affect the interest rate risk effect onstock returns of banks and life insurance companies. Central bank actions may reveal useful informa-tion about future economic activity transmitted in financial markets. Changes concerning interest ratevariables, however, may not affect stock returns in well-informed investments by an independent andtransparent central bank. Consequently, the research questions that this paper tries to shed light onare as follows:

• Does MPC’s creation affect interest rate risk exposures of banks and life insurance companies?• Are MPC dissents correlated with interest rate risks which banks and life insurance companies face?

To examine these questions empirically, we carry out a set of tests. First, we explore time variationin interest rate risk exposure by estimating asset-pricing models and allowing for different estimatesbefore and after BoE independence. Second, the effect of MPC dissents during voting procedure oninterest rate risk of banks and life insurance companies is investigated by means of the asset-pricingmodels.

With respect to the asset-pricing models, we estimate the CAPM and the Fama-French modelsaugmented with two interest rate risk factors. This is in line with Longstaff and Schwartz’s (1992)

5 We have selected the British banks and life insurance companies listed in the LSE. The major criteria were the country ofincorporation, years listed in the LSE and market capitalisation. All data available since 1989 were collected from the ReutersEcowin database. Only data for St James’s Place start from 1991.


two-factor model that shows that the rate of change of net worth for the next instant depends onshort-term rate changes and volatility changes:

dNW

NW= ˇrdrt + ˇV dVt

In our case we include the change in the level of the (detrended)6 interest rate (level factor)�ij,t = ij,t − ij,t−1 and the change in its conditional standard deviation ��j,t = �j,t − �j,t−1 (volatilityfactor), to proxy for interest rate risk on banks’ and life insurance companies’ stock returns. Morespecifically, our measure for interest rate volatility is based on the exponential GARCH-type model(Bollerslev, Chou, & Kroner, 1992; Nelson, 1991), which treats volatility as follows:7

ij,t = ˇ0 + εj,t (1)

ln(�2j,t) = ω + ln(�2

j,t−1) + �εj,t−1√

�2j,t−1

+ ˛

∣∣∣∣∣∣εj,t−1√

�2j,t−1

∣∣∣∣∣∣(2)

εj,t |˝t−1∼GED(0, �2j,t , k) (3)

where i is the nominal interest rate after HP filtering, subscript j = s, l denotes short- and long-termmaturity. As short-term rate, we use the one-month Treasury Bill rate and as long-term rate we usethe Government Bond yield with 20 years to maturity. For the error term we assume a GeneralisedError Distribution (GED) where the tail parameter is k > 0. The GED follows the normal distribution ifk = 2 and fat-tailed if k < 2. Volatility of (j) interest rate is measured by conditional standard deviation�j,t.

The asymmetric EGARCH model has two advantages over the pure GARCH specification. First, thelogarithmic construction of Eq. (2) ensures that the estimated conditional variance is strictly positive,and thus the non-negativity constraints used in the estimation of the ARCH and GARCH models are notnecessary. Second, asymmetries are allowed, since if the relationship between volatility and returnsis negative, � will be negative. Therefore, the presence of the well-known leverage effect in financeliterature (i.e., the fact that multiple significant drops in return are associated with increased volatility)can be tested by the hypothesis that � is negative. The impact is asymmetric if � /= 0. In our case thecoefficient of asymmetry was statistically significant and negative.8

4.1. Interest rate risk in pre- and post-MPC period

Following the CAPM, assuming a time-variant interest rate sensitivity based on MPC’s creation andusing a GARCH-M methodology (see Engle, Lilien, & Robins, 1987) introduced by Elyasiani and Mansur(1998) for banks, and by Brewer et al. (2007) for insurance firms, we estimate the following model:

Rt − is,t = ˇ0 + ˇ1(RMt − is,t) + ˇ2�ij,t + ˇ3��j,t + ˇ4D�ij,t + ˇ5D��j,t + ˇ8Cr + �√

(ht) + ut (4)

ut |˝t−1 =√

ht · zt (5)

zt |∼GED(0, 1, k) (6)

ht = a0 + a1u2t−1 + a2ht−1 + a3Cr (7)

6 The Hodrick-Prescott (HP) filter, which is a stationarity-inducing transformation, is applied to interest rate data. In line withprevious macroeconomic literature, the � parameter for the HP filter is chosen 14,400. We would like to thank participants ofthe European Economics and Finance Society 2012 for this comment.

7 For different types of GARCH family models of interest rate volatility see also Chuderewicz (2002) and Bharati et al. (2006).8 For reasons of space the estimated coefficients of the EGARCH models are not presented but are available on request from

the authors.


where Rt is the return on the bank (or the life insurance company) sector index (or the company)in month t; RMt is the return on the market portfolio proxied by the FT All Share index in t; Cr is adummy for the financial crisis 2007/07–2009/03 included in mean and variance equations in orderto investigate its effect. We allow an exogenous shift in interest rate risk sensitivities drawn fromour earlier discussion of monetary policy in the United Kingdom, and a dummy variable D, taking thevalue of one after May 1997 and zero otherwise, is used. Volatility of stock returns is measured byconditional variance ht, which is described as a function of the squared values of the past residualspresenting the ARCH factor, and an autoregressive term ht−1 reflecting the GARCH character of themodel. The coefficients a1 and a2 must satisfy the stationarity conditions such that a1 ≥ 0, a2 ≥ 0 anda1 + a2 ≥ 0.9 The degree of volatility persistence is measured by the sum of coefficients a1 and a2. Forthe error term we assume a Generalised Error Distribution (GED) where the tail parameter is k > 0. TheGED follows the normal distribution if k = 2 and fat-tailed if k < 2. Finally the error term is dependenton the information set ˝t−1.10

Our mean Eq. (4), however, has to take into account that we are interested in changes in inter-est rates that are uncorrelated with market risk. Our results from estimating Eq. (4) may suffer fromthe multicollinearity problem, as the market index will be correlated with changes in interest rates.Therefore Eq. (4) is estimated by using the orthogonalised interest rate changes �i0

j,t, and the orthog-

onalised conditional volatility changes ��0j,t

. These are the residuals from the regressions of interest

rate changes and conditional volatility changes against the market index.11

In order to further investigate the interest rate risk exposure of banks and life insurance companiesto level and volatility factors an extended Fama-French model is estimated. The factors introduced byFama and French (1993) were made available by Gregory, Tharyan, and Christidis (2013) for the UKstock market. SMB is the average return on the three small-capitalised portfolios minus the averagereturn on the three big-capitalised portfolios, whereas HML is the average return on the two-valueportfolios minus the average return on the two growth portfolios.12 Therefore our model comprisingEqs. (4)–(7) is now transformed to:

Rt − is,t = ˇ0 + ˇ1(RMt − is,t) + ˇ2�ij,t + ˇ3��j,t + ˇ4D�ij,t + ˇ5D��j,t + ˇ6SMBt

+ ˇ7HMLt + ˇ8Cr + �√

(ht) + ut (8)

ut |˝t−1 =√

ht · zt (9)

zt |∼GED(0, 1, k) (10)

ht = a0 + a1u2t−1 + a2ht−1 + a3Cr (11)

If D = 0, the coefficient on the level factor is ˇ2, indicating the significance of this factor during thepre-MPC creation period, whereas the post-MPC creation’s marginal effect on sensitivity to the levelfactor is measured by ˇ2 + ˇ4. Similarly, for the volatility factor in the pre-MPC creation period thesensitivity is measured by ˇ3, and by ˇ3 + ˇ5 for the post-MPC creation period. Thus, the sign andmagnitude of the interest rate betas give an indication of the level of interest rate risk undertakenby financial services companies over the different monetary policy frameworks. Eq. (8) is estimatedas in the case of Eq. (4) from the orthogonalised interest rate changes �i0

j,tand the orthogonalised

conditional volatility changes ��0j,t

.

9 We should mention at this point that Engle and Ng’s (1993) tests for asymmetry in volatility, known as sign and size biastests, provide evidence for a symmetric GARCH in the case of stock returns.

10 In order to capture any autocorrelation effects we estimate an ARMA(p,q) model where p, and q are based on Box–Jenkinsmethodology. In the case of an ARMA(1,1) process: ut = 1ut−1 + εt + 1εt−1.

11 For relevant analysis see also Akella and Greenbaum (1992) and Bharati et al. (2006).12 For more details see: http://xfi.exeter.ac.uk/researchandpublications/portfoliosandfactors/files.php.

http://xfi.exeter.ac.uk/researchandpublications/portfoliosandfactors/files.php


Therefore the models estimated above may help us test some important hypotheses about theinterest rate in the pre- and post-MPC periods. More specifically:

H1: In the pre-MPC creation period banks and life insurance companies are significantly exposedto interest rate level factors (interest rate volatility factors). Therefore our attention is focused on thecoefficients ˇ2 and ˇ3.

H2: Over the post-MPC creation period banks and life insurance companies are not significantlyexposed to interest rate level factors (interest rate volatility factors). This hypothesis can be formulatedas ˇ2 + ˇ4 = 0 (ˇ3 + ˇ5 = 0).

4.2. MPC dissents and the interest rate risk

Since MPC’s creation, there have been periods of increased uncertainty regarding the state of theeconomy during which agents may be confused about their investment decisions. Quantifying interestrate risk in such periods may reveal that an unclear interest rate decision can coexist with a significantinterest rate risk which banks and life insurance companies face. Hence, our empirical strategy is toidentify the impact of MPC dissents on the relationship between interest rate risk factors and stockreturns, using interaction terms in the extended Fama-French model.

Rt − is,t = ı0 + ı1(RMt − is,t) + ı2SMBt + ı3HMLt + ı4�ij,t + ı5��j,t + ı6Dm�ij,t + ı7Dm��j,t

+ ı8Cr + �√

(ht) + ut (12)

ut |˝t−1 =√

ht · zt (13)

zt |∼GED(0, 1, k) (14)

ht = a0 + a1u2t−1 + a2ht−1 + a3Cr (15)

All variables are presented as before with the exception that Dm is calculated as the log(1 + thenumber of dissents in voting for policy rate by the MPC members). The dissents can be related to thesize and/or the sign of the change of the interest rate. We expect that in periods with a significantnumber of dissents banks and life insurance companies are more affected by interest rate changes.Therefore we will focus on the statistical significance and sign of ı6, ı7 coefficients in Eq. (12). For agraphical presentation of the number of dissents by the MPC members, see Fig. 3. At this point, weshould mention that there are typically nine members in the MPC and hence the maximum numberof dissenting votes is four.

5. Empirical results

The first step toward more transparent policies was the adoption of inflation targeting in October1992. In Fig. 1 it is obvious that since inflation targeting there has been a significant drop in the levelof short-term interest rates.13 After 1992, a number of institutional reforms further improved thetransparency of the UK policy process. Such reforms included the introduction of scheduled meetingsto discuss policy rate changes (October 1992), the publication of the Inflation Report (February 1993),the decision to publish the minutes of monthly interest rate meetings (April 1994) and the creation ofthe MPC in May 1997, but only the establishment of the Monetary Policy Committee (MPC) achieveda drop in the level of long-term interest rates similar to the one observed after inflation targeting (seeFig. 2). On the one hand, based on Fisher’s equation and assuming fixed real rates, we can conclude thata shift of the monetary policy regime to inflation targeting lowers inflation. On the other hand, giventhe fact that long-term rates reflect inflation expectations, it seems that MPC’s creation has reducedinflation expectations.

13 For the beneficial effect of inflation targeting see among others King (2002) and Siklos (2004).


2

4

6

8

10

12

14

9290 94

Inflation Targeting10/1992

96 98 00

Creation of the MPC05/1997

02 04 06

Fig. 1. Short-term rate.

Tables 2a–3b present the maximum likelihood estimates of the extended CAPM and Fama-Frenchmodels in order to uncover the effect of MPC’s creation on interest rate risks that banks and lifeinsurance companies face. In general, from these tables we can infer that the market index coefficient(ˇ1) is significant, positive, and less than unity for banks and life insurance companies. Moreover, inthe Fama-French model there is evidence of a positive and significant effect of SMB and HML indiceson stock returns. Also, the last part in all these tables presents some diagnostics tests on the residuals.As is easily seen there is no empirical evidence for autocorrelation and heteroskedasticity across thefitted models. The GED parameter is also fewer than two in most of our cases. Overall, the Fama-Frenchmodel gives the best fit based on the log likelihood and adjusted R2 measures.

As far as interest rate risk is concerned, there are differences among short-term and long-terminterest rates. The inclusion of the interest rate volatility factor in the models can potentially pricethe features of financial companies’ assets and liabilities. Changes in the level of short-term interestrate assume parallel shifts in the term structure, and changes in the volatility of short-term rates maycapture more complex shifts in the term structure of interest rates (see also Bharati et al., 2006).

3

4

5

6

7

8

9

10

11

12

90 92 94 96

Inflation Targeting10/1992

Creation of the MPC05/1997

98 00 02 04 06

Fig. 2. Government bond yield – 20 years maturity.

S. Papadam

ou, C.

Siriopoulos /

Journal of

Economics

and Business

71 (2014) 45– 6755

Table 2aMaximum likelihood estimation results for short-term interest rate effects on bank returns.

Barclays Company Standard Chartered Company Hsbc Hld. Company Lloyds Grp. Company Bank Index

Mean equation FF: Rt − is,t = ˇ0 + ˇ1(RMt − is,t ) + ˇ2�ij,t + ˇ3��j,t + ˇ4D�ij,t + ˇ5D��j,t + ˇ6SMBt + ˇ7HMLt + ˇ8Cr + �

√(ht ) + ut

CAPM -model FF-model CAPM -model FF-model CAPM-model FF-model CAPM-model FF-model CAPM-model FF-model

Coef. Variables Coef. Prob. Coef. Prob. Coef. Prob. Coef. Prob. Coef. Prob. Coef. Prob. Coef. Prob. Coef. Prob. Coef. Prob. Coef. Prob.

� h0.5 0.095 (0.73) 0.176 (0.55) 0.295 (0.24) 0.169 (0.48) 0.188 (0.31) 0.122 (0.56) −0.339 (0.09)* −0.498 (0.03)** 0.188 (0.40) 0.220 (0.28)

ˇ0 Intercept −0.005 (0.74) −0.011 (0.43) −0.010 (0.48) −0.006 (0.63) −0.014 (0.08)* −0.010 (0.25) 0.017 (0.13) 0.024 (0.06) −0.007 (0.39) −0.010 (0.14)

ˇ1 RM−i 0.864 (0.00)*** 0.904 (0.00)*** 0.776 (0.00)*** 0.836 (0.00)*** 0.778 (0.00)*** 0.747 (0.00)*** 0.747 (0.00)*** 0.774 (0.00)*** 0.804 (0.00)*** 0.796 (0.00)***

ˇ2 �i −0.231 (0.34) −0.375 (0.09)* −0.315 (0.12) −0.330 (0.12) −0.273 (0.17) −0.363 (0.09)* −0.013 (0.96) −0.002 (0.99) −0.166 (0.27) −0.152 (0.29)

ˇ3 �� −0.701 (0.04)** −0.671 (0.02)** −0.825 (0.02)** −0.890 (0.01)** −0.710 (0.04)** −0.743 (0.04)** −0.492 (0.09)* −0.619 (0.01)** −0.767 (0.00)*** −0.662 (0.00)***

ˇ4 D × �i 0.738 (0.02)** 0.513 (0.10) 0.466 (0.08)* 0.080 (0.78) 0.194 (0.47) 0.123 (0.68) 0.380 (0.20) 0.255 (0.40) 0.399 (0.04)** 0.083 (0.72)

�5 D × �� 0.922 (0.18) 0.829 (0.19) 2.252 (0.00)*** 2.369 (0.00)*** 0.996 (0.15) 0.887 (0.18) 0.343 (0.52) 0.305 (0.55) 0.743 (0.15) 0.586 (0.20)

ˇ6 SMB 0.640 (0.00)*** 0.544 (0.00)*** 0.469 (0.00)*** 0.268 (0.02)** 0.448 (0.00)***

ˇ7 HML 0.261 (0.02)** 0.198 (0.08)* 0.144 (0.07)* 0.370 (0.00)*** 0.188 (0.00)***

ˇ8 Fin. crisisdummy

−0.073 (0.07)* −0.057 (0.11) −0.029 (0.21) −0.007 (0.75) −0.009 (0.71) 0.003 (0.88) −0.051 (0.43) −0.037 (0.58) −0.046 (0.00)*** −0.032 (0.04)**

1 AR(1) 0.152 (0.02)** 0.152 (0.02)** −0.030 (0.72) 0.050 (0.52) −0.465 (0.00)*** −0.501 (0.02)** −0.012 (0.85) −0.020 (0.77)

ϕ1 MA(1) 0.111 (0.08)* 0.109 (0.12) 0.488 (0.00)*** 0.501 (0.02)**

Variance equation: ht = a0 + a1u2t−1

+ a2ht−1 + a3Cr

˛0 Vol.intercept

1.0E−04 (0.03)** 1.0E−04 (0.06)* 2.2E−04 (0.14) 1.5E−04 (0.17) 1.1E−04 (0.11) 8.7E−05 (0.07)* 7.0E−05 (0.01)** 7.5E−05 (0.00)*** 3.9E−05 (0.21) 4.2E−05 (0.18)

˛1 ARCH 0.040 (0.08)* 0.047 (0.06)* 0.122 (0.04)** 0.104 (0.02)** 0.213 (0.00)*** 0.153 (0.00)*** 0.022 (0.12) 0.010 (0.49) 0.088 (0.04)** 0.123 (0.04)**

˛2 GARCH 0.918 (0.00)*** 0.909 (0.00)*** 0.810 (0.00)*** 0.844 (0.00)*** 0.758 (0.00)*** 0.805 (0.00)*** 0.954 (0.00)*** 0.961 (0.00)*** 0.880 (0.00)*** 0.844 (0.00)***

˛3 Fin. crisisdummy

1.6E−03 (0.00)*** 1.3E−03 (0.01)** 7.8E−04 (0.38) 6.0E−04 (0.37) 6.5E−04 (0.06)* 4.3E−04 (0.15) 1 9E−03 (0.00)*** 1.8E−03 (0.00)*** 3.7E−04 (0.15) 3.2E−04 (0.23)

DiagnosticsAdjusted R−squared 39% 48% 29% 35% 26% 31% 32% 38% 47% 54%Log likelihood 377.66 393.73 375.68 384.78 375.62 390.34 375.49 383.94 492.13 514.87

GED parameter 1.78 (0.00)*** 1.80 (0.00)*** 1.45 (0.00)*** 1.51 (0.00)*** 2.50 (0.00)*** 1.90 (0.00)*** 2.85 (0.00)*** 2.95 (0.00)*** 1.57 (0.00)*** 1.52 (0.00)***

Q(12) (0.96) (0.82) (0.81) (0.94) (0.66) (0.59) (0.67) (0.27) (0.91) (0.92)Qsq(12) (0.79) (0.87) (0.21) (0.11) (0.36) (0.62) (0.47) (0.44) (0.86) (0.81)ARCH(1) (0.71) (0.37) (0.36) (0.52) (0.29) (0.29) (0.38) (0.79) (0.51) (0.34)

Null hypothesis

ˇ2 + ˇ4 = 0 0.51 (0.01)** 0.14 (0.53) 0.150 (0.41) −0.25 (0.21) −0.08 (0.48) −0.24 (0.26) 0.37 (0.06)* 0.25 (0.20) 0.23 (0.08)* −0.07 (0.54)

ˇ3 + ˇ5 = 0 0.22 (0.71) 0.16 (0.77) 1.426 (0.02)** 1.48 (0.02)** 0.29 (0.66) 0.14 (0.79) −0.15 (0.73) −0.31 (0.46) −0.02 (0.95) −0.08 (0.81)

In order to capture any autocorrelation effects we estimate an ARMA(p,q) model where p, and q are based on Box-Jenkins methodology.* Statistical significance at 10% level.

** Statistical significance at 5% level.*** Statistical significance at 1% level.

56S.

Papadamou,

C. Siriopoulos

/ Journal

of Econom

ics and

Business 71 (2014) 45– 67

Table 2bMaximum likelihood estimation results for short-term interest rate effects on life insurance companies’ returns.

Life Insurance Index Legal & General Company Prudential Company St James’s Place Company Aviva


√(ht ) + ut

CAPM -model FF-model CAPM-model FF-model CAPM-model FF-model CAPM-model FF-model CAPM-model FF-model


� h0.5 −0.044 (0.89) −0.177 (0.56) −0.010 (0.96) −0.022 (0.92) 0.266 (0.17) 0.273 (0.21) −0.196 (0.30) −0.319 (0.07)* 0.2908 (0.21) 0.1585 (0.52)

ˇo Intercept 0.002 (0.85) 0.006 (0.58) 0.002 (0.86) 0.000 (0.99) −0.009 (0.37) −0.010 (0.35) 0.023 (0.07)* 0.029 (0.00)*** −0.0158 (0.17) −0.0117 (0.32)

ˇ1 RM−i 0.705 (0.00)*** 0.701 (0.00)*** 0.814 (0.00)*** 0.823 (0.00)*** 0.758 (0.00)*** 0.789 (0.00)*** 0.567 (0.00)*** 0.640 (0.00)*** 0.9227 (0.00)*** 0.8912 0ˇ2 �i 0.002 (0.99) −0.052 (0.76) −0.188 (0.38) −0.212 (0.28) −0.085 (0.52) −0.174 (0.17) −0.100 (0.63) −0.156 (0.45) −0.0678 (0.74) −0.1489 (0.49)

ˇ3 �� −0.523 (0.04)** −0.547 (0.01)** −0.594 (0.04)** −0.693 (0.01)** −0.451 (0.04)** −0.433 (0.04)** −0.873 (0.01)** −0.699 (0.03)** −0.6003 (0.03)** −0.575 (0.04)**

ˇ4 D × �i 0.198 (0.41) −0.124 (0.60) 0.446 (0.11) 0.143 (0.59) 0.358 (0.08)* 0.139 (0.48) 0.451 (0.12) −0.131 (0.65) 0.1965 (0.48) 0.0681 (0.81)

ˇ5 D × �� 0.804 (0.10) 0.752 (0.11) 0.677 (0.24) 0.902 (0.09)* 0.314 (0.53) 0.298 (0.51) 0.682 (0.31) 1.085 (0.10) 1.3371 (0.02)** 1.3083 (0.02)**

ˇ6 SMB 0.562 (0.00)*** 0.489 (0.00)*** 0.484 (0.00)*** 0.805 (0.00)*** 0.3323 (0.00)***

ˇ7 HML 0.204 (0.00)*** 0.251 (0.00)*** 0.087 (0.21) 0.310 (0.00)*** 0.2802 (0.00)***


−0.029 (0.18) −0.012 (0.56) −0.033 (0.08)* −0.020 (0.27) −0.015 (0.36) −0.001 (0.95) −0.050 (0.00)*** −0.025 (0.15) −0.027 (0.18) −0.0096 (0.65)

1 AR(1) −0.497 (0.26) −0.511 (0.07)* 0.016 (0.81) 0.027 (0.67) 0.012 (0.82) 0.004 (0.93) −0.684 (0.00)*** −0.639 (0.00)***

ϕ1 MA(1) 0.561 (0.18) 0.637 (0.01)** 0.761 (0.00)*** 0.698 (0.00)*** −0.1103 (0.09)* −0.068 (0.31)


+ a2ht−1 + a3Cr

˛0 Vol.intercept

9.2E−05 (0.17) 9.7E−05 (0.09)* 2.1E−04 (0.14) 2.0E−04 (0.14) 1.4E−04 (0.27) 2.8E−04 (0.20) 1.7E−04 (0.37) 2.3E−04 (0.20) 1.8E−04 (0.15) 0.0002 (0.16)

˛1 ARCH 0.075 (0.07)* 0.109 (0.04)** 0.116 (0.04)** 0.124 (0.05) 0.117 (0.17) 0.141 (0.16) 0.131 (0.04)** 0.150 (0.04)** 0.1142 (0.03)** 0.1162 (0.04)**

˛2 GARCH 0.867 (0.00)*** 0.825 (0.00)*** 0.798 (0.00)*** 0.787 (0.00)*** 0.839 (0.00)*** 0.762 (0.00)*** 0.851 (0.00)*** 0.814 (0.00)*** 0.8195 (0.00)*** 0.8106 (0.00)***


5.2E−04 (0.16) 5.4E−04 (0.20) 8.4E−04 (0.11) 8.6E−04 (0.11) 1.1E−03 (0.31) 1.5E−03 (0.39) 6.0E−04 (0.38) 6.0E−04 (0.30) 1.1E−03 (0.08)* 0.0012 (0.07)*

DiagnosticsAdjusted R-squared 34% 42% 28% 34% 31% 35% 13% 19% 31% 34%Log likelihood 458.350 477.310 413.620 426.330 417.440 430.730 323.820 343.750 376.790 3.853

GED parameter 1.770 (0.00)*** 1.720 (0.00)*** 1.370 (0.00)*** 1.340 (0.00)*** 1.050 (0.00)*** 1.010 (0.00)*** 1.050 (0.00)*** 1.150 (0.00)*** 1.42 (0.00)*** 1.45 (0.00)***

Q(12) (0.27) (0.48) (0.29) (0.22) (0.49) (0.38) (0.83) (0.36) (0.34) (0.41)Qsq(12) (0.11) (0.66) (0.59) (0.56) (0.43) (0.43) (0.76) (0.86) (0.80) (0.71)ARCH(1) (0.74) (0.96) (0.41) (0.32) (0.51) (0.31) (0.81) (0.71) (0.51) (0.32)

Null hypothesis

ˇ2 + ˇ4 = 0 0.200 (0.20) −0.176 (0.29) 0.258 (0.15) −0.069 (0.70) 0.273 (0.08)* −0.036 (0.81) 0.351 (0.57) −0.287 (0.35) 0.129 (0.49) −0.081 (0.67)ˇ3 + ˇ5 = 0 0.281 (0.50) 0.205 (0.62) 0.082 (0.86) 0.208 (0.65) −0.136 (0.76) −0.135 (0.74) −0.191 (0.11) 0.386 (0.23) 0.737 (0.17) 0.733 (0.15)




Fig. 3. Number of dissenting votes (dissents in magnitude and/or sign of policy rate).Source: Bank of England, see www.bankofengland.co.uk/monetarypolicy/mpcvoting.xls.

From Tables 2a and 2b, which refer to short-term interest rate risk factors, we can conclude thatchanges in short-term interest rate volatility, and not changes in the level of short-term rates, neg-atively affect stock returns for all banks and life insurance companies in the pre-BoE independenceperiod. Moreover, the statistically significant negative effect of the volatility factor on stock returnsis consistent across industries and across different asset-pricing models. As a consequence the lessinformation about monetary policy available to banks and life insurance companies, the greater theportfolio adjustment to be made when information is revealed.

Additionally, in the last part of these tables, there is evidence for the null hypothesis of no interestrate effect (ˇ2 + ˇ4 = 0 or ˇ3 + ˇ5 = 0) for almost all of the post-BoE independence period in the Fama-French models. Therefore, interest rate risk factors are priced differently in a transparent monetarypolicy framework.

Regarding long-term rates, a reflection of inflation expectations, we can see from Tables 3a and 3bthat in all cases of stock returns changes in level factors in the pre-BoE independence period have anegative and statistically significant effect. In the post-BoE operational independence period, changesin level and volatility interest rate factors do not seem to affect stock returns of financial servicescompanies. This can mainly be attributed to the fact that unexpected changes in long-term ratesdo not affect inflation expectation under CBI and CBT. This implies that in the lower interest rateenvironment, where the discussion of MPC’s inflation forecasting records took place, banks had thepotential to manage the interest rate risk coming from long-term rates efficiently.

Looking at the estimation results for the variance equations in Tables 2a–3b, we can conclude thatthere is evidence of time-variant volatility for banks and life insurance stock returns. The empiricalsignificance of the estimates for the ARCH and GARCH coefficients in the volatility equation is testedwith t-tests (˛1 = 0, ˛2 = 0). All tests are found to be significant for the majority of our cases, providingevidence against traditional models that assume time-invariant volatility. Consequently, the basicconstant-variance CAPM and the Fama-French model appear to be inappropriate for describing stockreturns of banks and life insurance companies in the UK. The financial crisis dummy has a significantpositive effect on stock volatility of Barclays and Lloyds only.

The magnitude of the trade-off between the mean and the volatility of stock returns in these twoindustries is determined by the trade-off parameter (�). In almost all cases this parameter has a pos-itive sign but is not statistically significant for banks (the exception being Lloyds). For life insurancecompanies this coefficient presents a mixed sign factor but it is not statistically significant.

Tables 4a–4d present the estimation results of Eqs. (12)–(15) for banks and life insurance compa-nies, using short-term and long-term interest rates respectively. Regarding the short-term rates effects,

http://www.bankofengland.co.uk/monetarypolicy/mpcvoting.xls

58S.

Papadamou,

C. Siriopoulos

/ Journal

of Econom

ics and

Business 71 (2014) 45– 67

Table 3aMaximum likelihood estimation results for long-term interest rate effects on bank returns.

Barclays Company Standard Chartered Company Hsbc Hld. Company Lloyds Grp. Company Bank Index


√(ht ) + ut

CAPM-model FF-model CAPM-model FF-model CAPM-model FF-model CAPM-model FF-model CAPM-model FF-model


� h0.5 −0.042 (0.88) 0.035 (0.90) 0.039 (0.87) −0.041 (0.85) 0.294 (0.11) 0.261 (0.21) −0.345 (0.10) −0.449 (0.06)* 0.224 (0.31) 0.205 (0.31)

ˇ0 Intercept 0.002 (0.88) −0.004 (0.77) 0.007 (0.61) 0.006 (0.61) −0.012 (0.14) −0.014 (0.11) 0.016 (0.13) 0.020 (0.09)* −0.009 (0.28) −0.009 (0.18)

ˇ1 RM-i 0.796 (0.00)*** 0.814 (0.00)*** 0.808 (0.00)*** 0.819 (0.00)*** 0.673 (0.00)*** 0.650 (0.00)*** 0.766 (0.00)*** 0.744 (0.00)*** 0.831 (0.00)*** 0.772 (0.00)***

ˇ2 �i −0.716 (0.03)** −0.787 (0.01)** −1.151 (0.00)*** −1.485 (0.00)*** −1.704 (0.00)*** −1.787 (0.00)*** −0.579 (0.03)** −0.639 (0.01)** −0.826 (0.00)*** −0.838 (0.00)***

ˇ3 �� −0.665 (0.44) −0.778 (0.20) 0.290 (0.71) −0.043 (0.94) −0.507 (0.43) −0.617 (0.23) −0.912 (0.11) −0.996 (0.07)* −0.161 (0.70) −0.469 (0.26)

ˇ4 D × �i 1.616 (0.00)*** 1.500 (0.00)*** 1.113 (0.03)** 1.305 (0.00)*** 1.666 (0.00)*** 1.730 (0.00)*** 0.676 (0.13) 0.584 (0.17) 0.725 (0.03)** 0.784 (0.00)***

ˇ5 D × �� 0.235 (0.83) 1.296 (0.18) 0.201 (0.84) 0.906 (0.35) −0.392 (0.63) 0.097 (0.89) 0.311 (0.69) 0.786 (0.37) −0.197 (0.74) 0.587 (0.32)

ˇ6 SMB 0.656 (0.00)*** 0.528 (0.00)*** 0.424 (0.00)*** 0.358 (0.00)*** 0.469 (0.00)***

ˇ7 HML 0.219 (0.04)** 0.061 (0.57) 0.134 (0.09)* 0.301 (0.00)*** 0.169 (0.00)***


−0.066 (0.17) −0.048 (0.20) −0.029 (0.13) −0.002 (0.93) −0.017 (0.24) 0.005 (0.74) −0.046 (0.36) −0.031 (0.53) −0.048 (0.00)*** −0.033 (0.03)**

1 AR(1) 0.105 (0.11) 0.144 (0.02)** 0.054 (0.42) 0.083 (0.29) −0.232 (0.71) −0.257 (0.74) −0.062 (0.36) −0.017 (0.81)

ϕ1 MA(1) 0.102 (0.11) 0.128 (0.07)* 0.258 (0.68) 0.296 (0.70)


+ a2ht−1 + a3Cr

˛0 Vol.intercept

1.3E−04 (0.02)** 1.2E−04 (0.07)* 2.4E−04 (0.16) 1.8E−04 (0.17) 8.2E−05 (0.21) 1.9E−04 (0.14) 5.7E−05 (0.04)** 5.7E−05 (0.03)** 6.6E−05 (0.11) 5.9E−05 (0.08)*

˛1 ARCH 0.048 (0.04)** 0.054 (0.03)** 0.106 (0.06)* 0.093 (0.04)** 0.153 (0.02)** 0.295 (0.00)*** 0.026 (0.09)* 0.023 (0.21) 0.104 (0.06)* 0.123 (0.07)*

˛2 GARCH 0.895 (0.00)*** 0.892 (0.00)*** 0.821 (0.00)*** 0.847 (0.00)*** 0.810 (0.00)*** 0.653 (0.00)*** 0.952 (0.00)*** 0.954 (0.00)*** 0.850 (0.00)*** 0.832 (0.00)***


2.1E−03 (0.00)*** 1.5E−03 (0.04)** 8.0E−04 (0.35) 7.1E−04 (0.39) 4.4E−04 (0.18) 3.6E−04 (0.37) 1.7E−03 (0.00)*** 1 4E−03 (0.00)*** 4.8E−04 (0.21) 3.5E−04 (0.27)


GED parameter 1.850 (0.00)*** 1.750 (0.00)*** 1.320 (0.00)*** 1.250 (0.00)*** 1.340 (0.00)*** 1.470 (0.00)*** 2.300 (0.00)*** 2.150 (0.00)*** 1.530 (0.00)*** 1.590 (0.00)***

Q(12) (0.74) (0.84) (0.66) (0.59) (0.88) (0.77) (0.31) (0.14) (0.27) (0.29)Qsq(12) (0.72) (0.81) (0.59) (0.43) (0.15) (0.21) (0.36) (0.58) (0.24) (0.33)ARCH(1) (0.81) (0.55) (0.52) (0.57) (0.28) (0.24) (0.85) (0.51) (0.19) (0.31)

Null hypothesis

ˇ2 + ˇ4 = 0 0.900 (0.01)** 0.710 (0.04)** −0.030 (0.91) −0.180 (0.57) −0.038 (0.87) −0.050 (0.81) 0.090 (0.78) −0.050 (0.87) −0.101 (0.67) −0.050 (0.81)

ˇ3 + ˇ5 = 0 −0.431 (0.54) 0.510 (0.48) 0.490 (0.49) 0.860 (0.22) −0.899 (0.07)* −0.520 (0.31) −0.600 (0.28) −0.210 (0.74) −0.358 (0.42) 0.110 (0.78)



S. Papadam

ou, C.

Siriopoulos /

Journal of

Economics

and Business

71 (2014) 45– 6759

Table 3bMaximum likelihood estimation results for long-term interest rate effects on life insurance companies’ returns.

Life Insurance Index Legal & General Company Prudential Company St James’s Place Company Aviva

Mean equation FF: Rt − is,t = ˇ0 + ˇ1(RMt − is,t ) + ˇ2�ij,t + ˇ3��j,t + ˇ4D�ij,t + ˇ5D��ij,t + ˇ6SMB + ˇ7HMLt + ˇ8Cr + �

√(ht ) + ut

CAPM-model FF-model CAPM-model FF-model CAPM-model FF-model CAPM-model FF-model CAPM-model FF-model

Coef. Variables. Coef. Prob. Coef Prob. Coef Prob. Coef Prob. Coef Prob. Coef Prob. Coef Prob. Coef Prob. Coef Prob. Coef Prob.

� h0.5 0.064 (0.82) −0.168 (0.57) 0.026 (0.91) −0.079 (0.72) 0.069 (0.77) −0.007 (0.97) −0.150 (0.51) −0.347 (0.03)** 0.5175 (0.02)** 0.2331 (0.28)

ˇo Intercept −0.001 (0.96) 0.006 (0.57) 0.000 (0.99) 0.003 (0.77) −0.003 (0.77) 0.000 (0.97) 0.019 (0.26) 0.036 (0.00)*** −0.0266 (0.01)** −0.014 (0.14)

ˇ1 RM-i 0.641 (0.00)*** 0.618 (0.00)*** 0.804 (0.00)*** 0.804 (0.00)*** 0.800 (0.00)*** 0.735 (0.00)*** 0.516 (0.00)*** 0.501 (0.00)*** 0.8345 (0.00)*** 0.7396 (0.00)***

ˇ2 �i −0.448 (0.04)** −0.668 (0.00)*** −0.515 (0.04)** −0.608 (0.01)** −0.453 (0.07)* −0.659 (0.00)*** 0.129 (0.70) −0.461 (0.04)** −0.6279 (0.01)** −0.7605 (0.00)***

ˇ3 �� −0.060 (0.90) −0.604 (0.18) −0.390 (0.53) −1.048 (0.03)** −0.512 (0.40) −0.713 (0.18) −0.125 (0.87) −0.116 (0.83) −0.6088 (0.23) −0.6061 (0.26)

ˇ4 D × �i 0.481 (0.18) 0.547 (0.12) 0.533 (0.20) 0.160 (0.66) 0.621 (0.13) 0.667 (0.09)* −0.330 (0.49) 0.624 (0.05)* 0.7775 (0.03)** 0.841 (0.01)**

ˇ5 D × �� −0.207 (0.77) 0.799 (0.26) −0.131 (0.88) 0.882 (0.27) 0.512 (0.56) 1.290 (0.13) −0.813 (0.44) 0.159 (0.82) −0.6332 (0.59) 0.0944 (0.90)

ˇ6 SMB 0.551 (0.00)*** 0.519 (0.00)*** 0.526 (0.00)*** 0.695 (0.00)*** 0.4129 * (0.00)**

ˇ7 HML 0.165 (0.01)** 0.227 (0.01)** 0.126 (0.09)* 0.086 (0.34) 0.1961 (0.01)**


−0.029 (0.11) −0.013 (0.55) −0.033 (0.08)* −0.019 (0.24) −0.022 (0.32) −0.008 (0.71) −0.048 (0.00)*** −0.030 (0.00)*** −0.03867 (0.04)** −0.0157 (0.48)

1 AR(1) −0.490 (0.18) −0.514 (0.03)** 0.013 (0.84) 0.034 (0.58) −0.044 (0.54) 0.053 (0.46) −0.703 (0.00)*** −0.710 (0.00)*** (0.43)

ϕ1 MA(1) 0.572 (0.09)* 0.662 (0.00)*** 0.745 (0.00)*** 0.738 (0.00)*** −0.12936 (0.04)** −0.05527 (0.31)


+ a2ht−1 + a3Cr

˛0 Vol.intercept

1.1E−04 (0.23) 1.1E−04 (0.10) 2.1E−04 (0.15) 2.2E−04 (0.18) 9.4E−05 (0.19) 1.1E−04 (0.17) 3.2E−04 (0.38) 3.8E−04 (0.27) 3.3E−04 (0.11) 2.5E−04 (0.09)*

˛1 ARCH 0.075 (0.11) 0.104 (0.04)** 0.120 (0.04)** 0.125 (0.07)* 0.117 (0.03)** 0.118 (0.03)** 0.106 (0.15) 0.162 (0.09)* 0.2082 (0.04)** 0.202 (0.02)**

˛2 GARCH 0.865 (0.00)*** 0.828 (0.00)*** 0.797 (0.00)*** 0.787 (0.00)*** 0.841 (0.00)*** 0.831 (0.00)*** 0.856 (0.00)*** 0.823 (0.00)*** 0.676 (0.00)*** 0.693 * (0.00)**


5.9E−04 (0.18) 5.5E−04 (0.22) 8.3E−04 (0.13) 7.2E−04 (0.15) 8.4E−04 (0.16) 7.8E−04 (0.24) 6.8E−04 (0.47) 5.8E−04 (0.50) 1.6E−03 (0.12) 1.8E−03 (0.08)*


GED parameter 1.400 (0.00)*** 1.600 (0.00)*** 1.400 (0.00)*** 1.200 (0.00)*** 1.770 (0.00)*** 1.720 (0.00)*** 0.990 (0.00)*** 0.800 (0.00)*** 1.25 (0.00)*** 1.42 (0.00)***

Q(12) (0.43) (0.55) (0.41) (0.2) (0.15) (0.26) (0.82) (0.42) (0.80) (0.71)Qsq(12) (0.11) (0.91) (0.77) (0.51) (0.26) (0.68) (0.81) (0.95) (0.90) (0.86)ARCH(1) (0.68) (0.67) (0.52) (0.15) (0.59) (0.29) (0.98) (0.64) (0.64) (0.42)

Null hypothesisˇ2 + ˇ4 = 0 0.033 (0.90) −0.120 (0.63) 0.010 (0.95) −0.440 (0.11) 0.160 (0.59) 0.009 (0.97) −0.201 (0.55) 0.160 (0.43) 0.150 (0.55) 0.081 (0.74)ˇ3 + ˇ5 = 0 −0.267 (0.59) 0.190 (0.72) −0.520 (0.41) −0.160 (0.78) 0.000 (0.97) 0.577 (0.39) −0.938 (0.18) 0.040 (0.92) −1.242 (0.12) −0.512 (0.38)



60S.

Papadamou,

C. Siriopoulos

/ Journal

of Econom

ics and

Business 71 (2014) 45– 67

Table 4aMaximum likelihood estimation results for short-term interest rate effects on banks’ returns during MPC dissents.

Banks

Barclays Company Standard Chartered HsbcHld. Lloyds Grp. Bank Index

Mean equation Rt − is,t = ı0 + ı1(RMt − is,t ) + ı2SMBt + ı3HMLt + ı4�ij,t + ı5��j,t + ı6Dm�ij,t + ı7Dm��j,t + ı8Cr + �√

(ht) + ut

Coef. Variables Coef. Prob. Coef. Prob. Coef. Prob. Coef. Prob. Coef. Prob.

� h0.5 0.359 (0.21) 0.113 (0.50) 0.149 (0.39) −0.085 (0.67) 0.304 (0.11)ı0 Intercept −0.025 (0.11) −0.007 (0.40) −0.010 (0.12) −0.008 (0.42) −0.015 (0.00)***

ı1 RM-i 0.951 (0.00)*** 0.820 (0.00)*** 0.664 (0.00)*** 1.059 (0.00)*** 0.811 (0.00)***

ı2 SMB 0.816 (0.00)*** 0.526 (0.00)*** 0.483 (0.00)*** 0.511 (0.00)*** 0.540 (0.00)***

ı3 HML 0.370 (0.02)** 0.336 (0.00)*** 0.242 (0.00)*** 0.472 (0.00)*** 0.210 (0.00)***

ı4 �i 0.686 (0.02)** −0.127 (0.54) −0.246 (0.19) 0.352 (0.18) 0.129 (0.43)ı5 �� −0.921 (0.12) 0.835 (0.18) 1.195 (0.01)** −0.200 (0.76) −0.257 (0.56)ı6 Dm × �i −1.940 (0.02)** −1.111 (0.04)** −0.051 (0.91) −1.362 (0.02)** −0.831 (0.04)**

ı7 Dm × �� 1.861 (0.45) 1.024 (0.51) −3.175 (0.04)** −0.403 (0.83) −0.175 (0.88)ı8 Fin. crisis dummy −0.042 (0.23) 0.008 (0.58) 0.011 (0.41) −0.011 (0.35) −0.028 (0.03)**

1 AR(1) 0.140 (0.16) 0.061 (0.27) 0.049 (0.49) −0.893 (0.00)*** −0.036 (0.63)ϕ1 MA(1) 0.969 (0.00)***

Variance equation: ht = a0 + a1u2t−1 + a2ht−1 + a3Cr

˛0 Vol. intercept 3.1E−04 (0.10) 1.1E−04 (0.42) 7.2E−05 (0.19) 1.8E−04 (0.11) 3.7E−05 (0.30)˛1 ARCH 0.195 (0.06)* 0.094 (0.17) 0.138 (0.09)* 0.298 (0.00)*** 0.124 (0.13)˛2 GARCH 0.730 (0.00)*** 0.865 (0.00)*** 0.812 (0.00)*** 0.689 (0.00)*** 0.837 (0.00)***

˛3 Fin. crisis dummy 1.6E−03 (0.22) 9.3E−04 (0.39) 5.2E−04 (0.28) 6.3E−05 (0.86) 4.1E−04 (0.27)

DiagnosticsAdjusted R-squared 53% 31% 28% 41% 57%Log likelihood 248.98 269.57 316.53 256.31 352.56GED parameter 1.90 (0.00)*** 0.95 (0.00)*** 1.15 (0.00)*** 1.50 (0.00)*** 1.25 (0.00)***

Q(12) (0.22) (0.91) (0.21) (0.33) (0.59)Qsq(12) (0.83) (0.76) (0.92) (0.49) (0.22)ARCH(1) (0.26) (0.97) (0.33) (0.78) (0.38)



S. Papadam

ou, C.

Siriopoulos /

Journal of

Economics

and Business

71 (2014) 45– 6761

Table 4bMaximum likelihood estimation results for short-term interest rate effects on life insurance companies’ returns during MPC dissents.

Life Insurance Companies

Life Insurance Legal & General Prudential Company Aviva St James’s Place


(ht) + ut


� h0.5 −0.193 (0.53) 0.010 (0.96) −0.296 (0.29) 0.500 (0.18) −0.282 (0.10)ı0 Intercept 0.005 (0.67) −0.005 (0.62) 0.012 (0.35) −0.036 (0.09)* 0.026 (0.01)**

ı1 RM-i 0.651 (0.00)*** 0.802 (0.00)*** 0.897 (0.00)*** 0.832 (0.00)*** 0.657 (0.00)***

ı2 SMB 0.779 (0.00)*** 0.580 (0.00)*** 0.850 (0.00)*** 0.552 (0.00)*** 0.768 (0.00)***

ı3 HML 0.372 (0.00)*** 0.398 (0.00)*** 0.356 (0.00)*** 0.416 (0.00)*** 0.572 (0.00)***

ı4 �i −0.082 (0.72) 0.033 (0.88) −0.338 (0.22) 0.043 (0.86) −0.510 (0.05)*

ı5 �� 0.904 (0.12) 0.980 (0.08)* 1.412 (0.05)* 1.907 (0.00)*** 0.309 (0.66)ı6 Dm × �i −1.163 (0.06)* −0.610 (0.30) −0.763 (0.36) −0.868 (0.19) 0.771 (0.27)ı7 Dm × �� −4.140 (0.04)*** −3.886 (0.01)** −4.939 (0.04)** −4.165 (0.03)** 2.534 (0.22)ı8 Fin. crisis dummy −0.007 (0.79) −0.010 (0.49) 0.006 (0.79) −0.007 (0.72) −0.027 (0.11)1 AR(1) −0.583 (0.00)*** 0.045 (0.55) 0.062 (0.46) 0.562 (0.30)ϕ1 MA(1) 0.796 (0.00)*** −0.040 (0.56) −0.497 (0.39)


˛0 Vol. intercept 1.9E−04 (0.03)** 2.8E−04 (0.05)* 1.9E−04 (0.12) 7.1E−04 (0.15) 1.2E−04 (0.37)˛1 ARCH 0.188 (0.03)** 0.188 (0.08)* 0.171 (0.03)** 0.169 (0.19) 0.196 (0.06)*

˛2 GARCH 0.717 (0.00)*** 0.727 (0.00)*** 0.765 (0.00)*** 0.634 (0.00)*** 0.796 (0.00)***

˛3 Fin. crisis dummy 7.1E−04 (0.12) 7.6E−04 (0.20) 5.7E−04 (0.42) 1 8E−03 (0.19) 6.9E−04 (0.29)


Q(12) (0.41) (0.39) (0.47) (0.33) (0.45)Qsq(12) (0.97) (0.98) (0.81) (0.83) (0.73)ARCH(1) (0.53) (0.27) (0.13) (0.29) (0.52)



62S.

Papadamou,

C. Siriopoulos

/ Journal

of Econom

ics and

Business 71 (2014) 45– 67

Table 4cMaximum likelihood estimation results for long-term interest rate effects on banks’ returns during MPC dissents.

Banks

Barclays Company Standard Chartered Hsbc Hld. Lloyds Grp. Bank Index


(ht ) + ut


� h0.5 0.026 (0.92) −0.065 (0.73) 0.212 (0.20) −0.193 (0.23) 0.253 (0.15)ı0 Intercept −0.012 (0.32) 0.001 (0.94) −0.014 (0.02)** −0.004 (0.59) −0.016 (0.00)***

ı1 RM-i 0.879 (0.00)*** 0.870 (0.00)*** 0.623 (0.00)*** 1.018 (0.00)*** 0.825 (0.00)***

ı2 SMB 0.888 (0.00)*** 0.538 (0.00)*** 0.353 (0.00)*** 0.514 (0.00)*** 0.472 (0.00)***

ı3 HML 0.381 (0.00)*** 0.201 (0.04)** 0.209 (0.01)** 0.420 (0.00)*** 0.285 (0.00)***

ı4 �i 0.023 (0.96) −0.053 (0.89) 0.204 (0.53) −0.264 (0.52) −0.086 (0.75)ı5 �� 0.278 (0.84) 0.151 (0.85) −0.913 (0.18) 0.136 (0.87) 0.102 (0.86)ı6 Dm × �i 2.360 (0.10) −0.705 (0.55) −0.228 (0.81) 0.117 (0.91) 0.160 (0.83)ı7 Dm × �� 2.347 (0.51) 2.977 (0.19) 3.195 (0.08)* 1.378 (0.57) 0.903 (0.54)ı8 Fin. crisis dummy −0.032 (0.35) 0.015 (0.40) 0.002 (0.85) −0.001 (0.94) −0.025 (0.03)**

1 AR(1) 0.080 (0.38) 0.018 (0.79) 0.029 (0.69) −0.841 (0.00)*** −0.105 (0.16)ϕ1 MA(1) 0.928 (0.00)***


˛0 Vol. intercept 7.9E−05 (0.17) 1.2E−04 (0.36) 5.3E−05 (0.23) 1.4E−04 (0.27) 2.8E−05 (0.34)˛1 ARCH 0.066 (0.07)* 0.115 (0.19) 0.119 (0.16) 0.249 (0.04)** 0.100 (0.19)˛2 GARCH 0.891 (0.00)*** 0.841 (0.00)*** 0.839 (0.00)*** 0.747 (0.00)*** 0.864 (0.00)***

˛3 Fin. crisis dummy 1 4E−03 (0.04)** 9.6E−04 (0.41) 4.9E−04 (0.20) 3.4E−04 (0.62) 4.1E−04 (0.20)


Q(12) (0.41) (0.87) (0.39) (0.23) (0.41)Qsq(12) (0.95) (0.15) (0.96) (0.83) (0.43)ARCH(1) (0.35) (0.52) (0.32) (0.55) (0.64)



S. Papadam

ou, C.

Siriopoulos /

Journal of

Economics

and Business

71 (2014) 45– 6763

Table 4dMaximum likelihood estimation results for long-term interest rate effects on life insurance companies’ returns during MPC dissents.

Life Insurance Companies

Life Insurance Legal & General Prudential Company Aviva St James’s Place

Mean equation Rt − is,t = ı0 + ı1(RMt − is,t ) + ı2SMBt + ı3HMLt + ı4�ij,t + ı5��j,t + ı6Dm�ij,t + ı7Dm��j,t + +ı8Cr + �√

(ht) + ut


� h0.5 −0.129 (0.69) 0.073 (0.74) −0.091 (0.79) 0.205 (0.50) −0.394 (0.12)ı0 Intercept −0.001 (0.93) −0.008 (0.42) −0.002 (0.89) −0.017 (0.32) 0.035 (0.04)**

ı1 RM-i 0.744 (0.00)*** 0.807 (0.00)*** 0.905 (0.00)*** 0.956 (0.00)*** 0.551 (0.00)***

ı2 SMB 0.582 (0.00)*** 0.578 (0.00)*** 0.772 (0.00)*** 0.367 (0.00)*** 0.700 (0.00)***

ı3 HML 0.310 (0.00)*** 0.358 (0.00)*** 0.360 (0.00)*** 0.380 (0.00)*** 0.332 (0.00)***

ı4 �i 0.050 (0.89) −0.201 (0.65) −0.372 (0.41) 0.168 (0.68) −0.347 (0.47)ı5 �� −0.955 (0.24) −0.773 (0.44) −0.279 (0.80) −1.304 (0.14) 0.421 (0.70)ı6 Dm × �i −0.811 (0.50) −0.283 (0.82) −0.091 (0.95) −1.508 (0.26) 1.904 (0.19)ı7 Dm × �� 4.588 (0.10) 3.328 (0.24) 4.124 (0.25) 1.840 (0.49) −1.721 (0.60)ı8 Fin. crisis dummy −0.011 (0.67) −0.015 (0.48) 0.007 (0.76) −0.002 (0.91) −0.027 (0.16)1 AR(1) −0.618 (0.01)** 0.090 (0.31) 0.071 (0.46) 0.388 (0.53)ϕ1 MA(1) 0.760 (0.00)*** −0.065 (0.38) −0.299 (0.00)***

Variance equation: ht = a0 + a1u2t−1 + +a2ht−1 + a3Cr

˛0 Vol. intercept 9.2E−05 (0.21) 1.2E−04 (0.27) 3.1E−04 (0.14) 2.8E−04 (0.28) 2.3E−04 (0.28)˛1 ARCH 0.135 (0.07)* 0.154 (0.04)** 0.165 (0.05)* 0.105 (0.18) 0.166 (0.07)*

˛2 GARCH 0.808 (0.00)*** 0.786 (0.00)*** 0.735 (0.00)*** 0.796 (0.00)*** 0.805 (0.00)***

˛3 Fin. crisis dummy 5.2E−04 (0.27) 9.0E−04 (0.07)* 7.3E−04 (0.44) 1.1E−03 (0.15) 4.7E−04 (0.40)


Q(12) (0.33) (0.39) (0.26) (0.47) (0.71)Qsq(12) (0.99) (0.66) (0.59) (0.94) (0.66)ARCH(1) (0.46) (0.13) (0.11) (0.49) (0.51)




in the case of four out of five banks and life insurance companies the MPC dissents coexist with statis-tically significant interest rate risks. In the case of banks, however, changes in the level of interest ratesnegatively affect returns, and in the case of life insurers, changes in interest rate conditional volatilitynegatively affect returns. This difference can be attributed to the portfolio composition and differentinvestment strategies that these companies adopt. Therefore the higher the level of uncertainty withrespect to monetary policy strategy the more negatively significant is the effect of interest rate riskfactors on the stock returns of these industries. This effect is mainly reflected in the negative valuesof coefficients ı6 and/or ı7. This result implies that central bankers’ increased uncertainty about thestate of the economy may negatively affect agents. This effect is not present, however, in the case oflong-term rates, largely because of the inflation targeting policy and medium-term monetary policystrategy.

6. Conclusions

This study, by using extended CAPM and Fama-French models, investigates the effect of BoE inde-pendence on the interest rate risk that banks and life insurance companies face. Previous researchon interest rate risk neglects the dissimilar contribution of unexpected components to equity returnsin two different monetary policy frameworks: one with an increased level of transparency inducedby the Monetary Policy Committee’s actions and the other with a low level of transparency resultingfrom a low level of monetary policy independence. In the case of the UK, the weak effects concerninginterest rate risk factors found by Beirne et al. (2009) can be attributed to the fact that they estimatesensitivities over the whole sample, whereas a significant structural change concerning monetary pol-icy conduct occurred specifically in 1997. Our study shed light on this direction suggesting a modelfor measuring interest rate sensitivities to other countries experienced significant changes in the waythat monetary policy is conducted. We use more than one interest rate factor in line with Czaja, Scholz,and Wilkens (2009) and Caporale and Perry (2006), and in contrast with the majority of the studies inthe field focusing only on one interest rate risk factor (usually changes in the level).

Empirical evidence is provided for the belief that the creation of MPC can reduce the interest raterisk that banks and life insurance companies face. Specifically, before the MPC’s creation interest raterisk factors had a significant negative effect on stock returns. On the one hand, a positive change inlong-term rates significantly lowers excess bank and life insurance stock returns. This is consistentwith Elyasiani and Mansur’s (1998) findings for US banks and with the view that changes in long-termreal interest rates provide market participants with procyclical indications about future economicconditions. Lower (higher) interest rates during a contracting (expanding) and more (less) risky econ-omy would raise (lower) the equity risk premium. On the other hand, an increase in short-term ratevolatility implying term structure change also has a negative effect on excess return. After the MPC’screation there is no significant effect for any of our cases. Our results are consistent across differentasset-pricing models, across financial services companies and across industry-level data and firm-leveldata. The last indicate the robustness of our findings given the trade-off between using industry-leveldata and using firm-level data.

Moreover, according to King (2007), periods of increased uncertainty about the macro model thatfits better with the real economy are reflected in MPC dissents. Using this argument we show thatthese dissents about the state of the economy coexist with a high level of interest rate risk that banksand life insurance companies face. Because of the different investment strategies of the two industries,banks are affected by changes in the level of short-term rates and life insurers by changes related toterm structure (i.e., short-term rates’ conditional volatility). Therefore the beneficial role that a lowlevel of dissents in MPC can have on reducing interest rate risk can be inferred.

Our findings have important implications for monetary authorities seeking to foster stability in thefinancial industry via central banking policies. If we accept (as Romer & Romer, 2000 and Caporale &Perry, 2006 do) that the central bank has a significant information advantage over the public aboutthe current and future state of the economy, monetary policy actions reflected in interest rate factorssignal to the financial markets an increase (decrease) in economic risk, which necessitates a rise (fall)in the risk premium. These effects are not, however, present during the post-MPC period and duringperiods with no MPC dissents, mainly because investors can clearly identify central bank actions and


therefore do not overreact to changes in interest rate factors. In this respect our evidence suggeststhat transparent policies with reduced uncertainty about monetary actions can reduce the negativeeffect of interest rate risk on financial services companies.

Our analysis also suggests that to test the Efficient Market Hypothesis in a monetary policy frame-work characterised by a low level of central bank operational independence it is crucial to model therisk premium in financial services stocks by taking into account interest rate risk factors. Therefore,our results are interesting for managers, investors and policy-makers alike.

Acknowledgements

The authors would like to thank Prof. Alan Gregory for giving us the updated set of data concerningFama-French factors in the UK, Prof. Elyas Elyasiani, I. Psarianos, N. Tzeremes and two anonymousreferees for their helpful comments on an earlier version of the paper.

References

Aggarwal, R., Jeon, B. P., & Zhao, X. (2006). Bank exposure to interest rate risks during financial liberalization: Evidence fromSouth Korea. Asia-Pacific Financial Markets, 12, 61–90.

Akella, S. R., & Chen, S. J. (1990). Interest rate sensitivity of bank stock returns: Specification effects and structural changes.Journal of Financial Research, 13, 147–154.

Akella, S., & Greenbaum, S. (1992). Innovations in interest rates, duration transformation, and bank stock returns. Journal ofMoney, Credit and Banking, 24, 27–42.

Alesina, A., & Summers, L. (1993). Central bank independence and macroeconomic performance. Journal of Money, Credit andBanking, 25, 157–162.

Allen, L., & Jagtiani, J. (1997). Risk and market segmentation in financial intermediaries’ returns. Journal of Financial ServicesResearch, 12, 159–173.

Bach, B., & Ando, A. (1957). The redistribution of effects of inflation. Review of Economics and Statistics, 1, 1–13.Bae, S. C. (1990). Interest rate changes and common stock returns of Financial institutions: Revisited. Journal of Financial Research,

13, 71–79.Baillie, R. T., & De Gennaro, R. P. (1990). Stock returns and volatility. Journal of Financial Quantitative Analysis, 25, 203–214.Beirne, J., Caporale, M., & Spagnolo, N. (2009). Market, interest rate and exchange rate risk effects on financial stock returns: A

GARCH-M approach. Quantitative and Qualitative Analysis in Social Sciences, 3(2), 44–68.Benink, H., & Wolff, C. (2000). Survey data and the interest rate sensitivity of US bank stock returns. Economic Notes, 29(2),

201–213.Bharati, R., Nanisetty, P., & So, J. (2006). Dynamic gap transformations: Are banks asset – transformers or brokers? or both? The

Quarterly Review of Economics and Finance, 46(1), 36–52.Bollerslev, T., Chou, R., & Kroner, K. (1992). ARCH modelling in finance: A review of the theory and empirical evidence. Journal

of Econometrics, 31, 307–327.Booth, J. R., Officer, D. T., & Henderson, G. V. (1985). Commercial bank stocks, interest rates, and systematic risk. Journal of

Economics and Business, 37(4), 303–310.Bredin, D., Hyde, S., Nitzsche, D., & O’Reilly, G. (2007). UK stock returns and the impact of domestic monetary policy shocks.

Journal of Business Finance & Accounting, 34(5/6), 872–888.Bredin, D., Hyde, S., & O’Reilly, G. (2010). Monetary policy surprises and international bond markets. Journal of International

Money and Finance, 29, 988–1002.Brewer, E., & Lee, C. F. (1990). An intracyclical analysis of the risk sensitivity of bank stock returns. Quarterly Journal of Business

and Economics, 29, 125–143.Brewer, E., Carson, J., Elyasiani, E., Mansur, I., & Scott, W. (2007). Interest rate risk and equity values of life insurance companies:

A GARCH-M moded. Journal of Risk and Insurance, 74(2), 401–423.Caporale, T., & Perry, M. J. (2006). Does monetary policy signal future economic risk? Investigating the link between monetary

policy shocks and stock returns. Alliance Journal of Business Research, 2(1), 1–16.Carson, J. M., Elyasiani, E., & Mansur, I. (2008). Market risk, interest rate risk, and interdependencies in insurer stock returns: A

system – GARCH model. Journal of Risk and Insurance, 75(4), 873–891.Chadha, J., & Nolan, C. (2001). Inflation targeting, transparency and interest rate volatility: Ditching ‘monetary mystique’ in the

UK. Journal of Macroeconomics, 23(3), 349–366.Chen, C. R., & Chan, A. (1989). Interest rate sensitivity, asymmetry, and the stock returns of financial institutions. Financial

Review, 24, 457–473.Cheng, J., Elyasiani, E., & Lin, T. T. (2010). Market reaction to regulatory action in the insurance industry: The case of contingent

commission. Journal of Risk and Insurance, 77(2), 347–368.Choi, J., Elyasiani, E., & Kopecky, K. J. (1992). The sensitivity of the bank stock returns to market, interest, and exchange rate

risks. Journal of Banking and Finance, 16, 983–1004.Choi, J., Elyasiani, E., & Saunders, A. (1996). Derivative Exposure and the Interest Rate and Exchange Rate Risks of US Banks. New

York University Salomon Center Working Paper, No. 69-25.Chortareas, G., Stasavage, D., & Sterne, G. (2002). Monetary policy transparency inflation and sacrifice ratio. International Journal

of Finance and Economics, 7, 141–155.

http://refhub.elsevier.com/S0148-6195(13)00063-5/sbref0005



































































































































































































































































































































































































































































































































































Chuderewicz, R. P. (2002). Using interest rate uncertainty to predict the paper-bill spread and real output. Journal of Economicsand Business, 54(3), 293–312.

Cihak, M. (2006). How Do Central Banks Write on Financial Stability? IMF Working Papers 06/163, International Monetary Fund.Clare, A., & Courtenay, R. (2001). Assessing the impact of macroeconomics news announcements on securities prices under different

monetary policy regimes. Bank of England Working Paper No. 125.Czaja, M. G., Scholz, H., & Wilkens, M. (2009). Interest rate risk of German financial institutions: The impact of level, slope, and

curvature of the term structure. Review of Quantitative Finance and Accounting, 33(1), 1–26.Czaja, M. G., Scholz, H., & Wilkens, M. (2010). Interest rate risk rewards in stock returns of financial corporations: Evidence from

Germany. European Financial Management, 16(1), 124–154.Deshmukh, S. D., Greenbaum, S. I., & Kanatas, G. (1983). Interest rate uncertainty and the financial intermediary’s choice of

exposure. Journal of Finance, 38, 141–147.Eijffinger, S., & Geraats, P. M. (2006). How transparent are central banks? European Journal of Political Economy, 22, 1–21.Eijffinger, S., Geraats, P. M., & Van der Cruijsen, C. (2006). Does Central Bank Transparency Reduce Interest Rates? CEPR Discussion

Papers No. 5526.Elyasiani, E., & Mansur, I. (1998). Sensitivity of the bank stock returns distribution to changes in the level and volatility of interest

rate: A GARCH–M model. Journal of Banking and Finance, 22, 535–563.Elyasiani, E., & Mansur, I. (2003). International spillover of risk and return among major banking institutions: A bivariate GARCH

model. Journal of Accounting Auditing and Finance, 18(2), 303–330.Elyasiani, E., Mansur, I., & Pagano, M. S. (2007). Convergence and risk-return linkages across financial service firms. Journal of

Banking & Finance, 31(4), 1167–1190.Elyasiani, E., Mansur, I., & Wetmore, J. L. (2010). Real-estate risk effects on financial institutions’ stock return distribution: A

bivariate GARCH analysis. The Journal of Real Estate Finance and Economics, 40(1), 89–107.Engle, R. F., & Ng, V. K. (1993). Measuring and testing the impact of news on volatility. Journal of Finance, 48, 1749–1778.Engle, R. F., Lilien, D. M., & Robins, R. P. (1987). Estimating time varying risk Premia in the term structure: The ARCH-M model.

Econometrica, 55, 391–407.Fama, E. F., & French, K. R. (1993). Common risk factors in the returns on stocks and bonds. Journal of Financial Economics, 33(1),

3–56.Ferrer, R., González, C., & Soto, G. M. (2010). Linear and nonlinear interest rate exposure in Spain. Managerial Finance, 36(5),

431–451.Flannery, M., Hameed, A., & Harjes, R. (1997). Asset pricing, time-varying risk premia, and interest rate risk. Journal of Banking

and Finance, 21, 315–335.Flannery, M., & James, C. (1984). The effect of interest rate changes on the common stock returns of financial institutions. Journal

of Finance, 39, 1141–1153.Gasbarro, D., & Monroe, G. S. (2004). The impact of monetary policy candidness on Australian financial markets. Multinational

Finance Journal, 14, 35–46.Geraats, P. (2002). Central bank transparency. Economic Journal, 112(483), 532–565.Goodfriend, M. (1998). Using the term structure of interest rates for monetary policy. Federal Reserve Bank of Richmond Economic

Quarterly, 84(3), 13–30.Gounopoulos, D., Molyneux, P., Staikouras, S. K., Wilson, J. O. S., & Zhao, G. (2012). Exchange rate risk and the equity performance

of financial intermediaries. International Review of Financial Analysis,. http://dx.doi.org/10.1016/j.irfa.2012.04.001Gregoriou, A., Kontonikas, A., MacDonald, R., & Montagnoli, A. (2009). Monetary policy shocks and stock returns: Evidence from

the British market. Financial Markets and Portfolio Management, 23(4), 401–410.Gregory, A., Tharyan, R., & Christidis, A. (2013). Constructing and testing alternative versions of the Fama–French and Carhart

models in the UK. Journal of Business Finance & Accounting, 40(1/2), 172–214.Guo, H. (2004). Stock prices, firm size, and changes in the federal funds rate target. The Quarterly Review of Economics and Finance,

44(4), 487–507.Hahm, J.-H. (2004). Interest rate and exchange rate exposure of banking institutions in pre-crisis Korea. Applied Economics, 36,

1409–1420.Hirtle, B. (1997). Derivatives portfolio composition, and bank holding company interest rate risk exposure. Journal of Financial

Services Research, 12, 243–266.Kane, E., & Unal, H. (1988). Market assessment of deposit-institution riskiness. Journal of Financial Services Research, 1, 207–229.Kasman, S., Vardar, G., & Tunc , G. (2011). The impact of interest rate and exchange rate volatility on banks’ stock returns and

volatility: Evidence from Turkey. Economic Modelling, 28(3), 1328–1334.Kessel, R. (1956). Inflation-caused wealth redistribution: A test of a hypothesis. American Economic Review, 128–214.King, M. (2002). The Monetary Policy Committee: Five years on’. Bank of England Quarterly Bulletin Summer, 219–227.King, M. (2007). The MPC ten years on. Speech available under www.bankofengland.co.ukKlomp, J., & de Haan, J. (2009). Central bank independence and financial instability. Journal of Financial Stability, 5(4),

321–338.Korkeamäki, T. (2011). Interest rate sensitivity of the European stock markets before and after the euro introduction. Journal of

International Financial Markets, Institutions and Money, 21(5), 811–831.Kwan, S. (1991). Re-examination of interest rate sensitivity of commercial bank stock returns using a random coefficient model.

Journal of Financial Services Research, 5, 61–76.Laopodis, N. (2010). Dynamic linkages between monetary policy and the stock market. Review of Quantitative Finance and

Accounting, 35(3), 271–293.Lloyd, W. P., & Shick, R. A. (1977). A test of stone’s two-index model of returns. Journal of Financial Quantitative Analysis, 12,

363–373.Longstaff, F. A., & Schwartz, E. S. (1992). Interest rate volatility and the term structure: A two-factor general equilibrium model.

The Journal of Finance, 47(4), 1259–1282.Mankiw, N. G. (1986). The allocation of credit and financial collapse. Quarterly Journal of Economics, 101, 455–470.Nelson, D. B. (1991). Conditional heteroskedasticity in asset returns: A new approach. Econometrica, 59, 347–370.





















































































































































































































































































































































































































































http://dx.doi.org/10.1016/j.irfa.2012.04.001
































































































































































































http://www.bankofengland.co.uk/






































































































































































Neuberger, J. A. (1993, July 23). Interest rate risk at US commercial banks, Working Paper Federal Reserve Bank of San FranciscoEconomic Letters.

Park, J., & Choi, B. P. (2011). Interest rate sensitivity of US property/liability insurer stock returns. Managerial Finance, 37(2),134–150.

Prasad, A. M., & Rajan, M. (1995). The role of exchange and interest risk in equity valuation: A comparative study of internationalstock markets. Journal of Economics and Business, 47(5), 457–472.

Rogoff, K. (1985). The optimal degree of commitment to a monetary target. Quarterly Journal of Economics, 100, 1169–1190.Romer, C. D., & Romer, D. H. (2000). Federal reserve information and the behavior of interest rates. American Economic Review,

90, 429–457.Santomero, A., & Babbel, D. (1997). Financial risk management by insurers: An analysis of the process. Journal of Risk and

Insurance, 64, 231–270.Scott, W. L., & Peterson, R. L. (1986). Interest rate risk and equity values of hedged and unhedged financial intermediaries. Journal

of Financial Research, 9, 325–329.Siklos, P. L. (2004). Central bank behavior, the institutional framework, and policy regimes: Inflation versus non inflation

targeting countries. Contemporary Economic Policy, 22(3), 331–343.Song, F. (1994). A two factor ARCH model for deposit-institution stock returns. Journal of Money, Credit and Banking, 26, 323–340.Stone, B. K. (1974). Systematic interest rate risk in a two-index model of returns. Journal of Financial Quantitative Analysis,

9(November), 709–921.Tuysuz, S. (2011). Has the independence of the Bank of England changed the reaction of interest rates to macroeconomic and

monetary news. International Research Journal of Applied Finance, II(5), 528–546.Viale, A. M., Kolari, J. W., & Fraser, D. R. (2009). Common risk factors in bank stocks. Journal of Banking & Finance, 33(3), 464–472.Woodford, M. (1999). Optimal Monetary Policy Inertia, NBER Working Paper, No. 7261, July.Yourougou, P. (1990). Interest rate risk and the pricing of depository financial intermediary common stock. Journal of Banking

and Finance, 14, 803–820.






































































































































































































































































































Interest rate risk and the creation of the Monetary Policy Committee: Evidence from banks’ and...

Documents

Transcript of Interest rate risk and the creation of the Monetary Policy Committee: Evidence from banks’ and...