During the 1950s and1960s, the government of developing countries controlled interest rates inorder to generate financial resources required to finance government budgetdeficits and for stabilisation purposes (World Bank, 2005). However, by theearly 1970s, these so-called government interventions were largely criticisedby the neo-classical economists and proponents of the financial liberalisationhypothesis who argued that developing countries should move away from financialrepression- a situation where the government controls lending and depositinterest rates, towards financial liberalisation in order to boost their levelsof economic activity and growth. McKinnon (1973) and Shaw (1973) were the earliesteconomists to propose theoretical arguments challenging financial repression,independently.
They both advocated the call for interest rate liberalisationand the elimination of other financially repressive policies practiced bygovernments such as discriminatory credit controls that slowed down the levelof economic activity and growth. The relationship existingbetween financial liberalisation policies, capital formation and economicgrowth has long been a subject of debate in the financial liberalisationliterature. In the context of this, there seems to be contradictory evidence inthe theoretical and empirical relationship between financial liberalisationpolicies and capital formation and between financial liberalisation policiesand economic growth. Since the seminal contribution of McKinnon (1973) and Shaw(1973), the prevailing tradition ineconomics posits that financial liberalisation is beneficial to developingcountries as the removal of interest rate controls would encourage financialsavings and lead to an increase in the quantity and quality of investment. The standardpolicy recommendation emanating from this literature is thatthrough financial liberalisation, the quantity and the composition of financialdevelopment variables promote economic growth by increasing the level offinancial savings thereby enhancing capital formation, efficient creditallocation, and provision of access to foreign capital (McKinnon, 1973; Shaw,1973; Fry, 1997; Fischer, 1998; and Summers, 2000). Importantly, this is theprevailing view at the International Monetary Fund (IMF) and the World Bank.
However, a competingstrand of the literature (Stiglitz and Weiss, 1981; Taylor, 1983;Diaz-Alejandro, 1985; Krugman, 1993; Rodrik, 1998; Van Wijnbergen, 1983; and Stiglitz,2000 amongst others) opposes the conventional prediction emanating from theliterature and their criticisms of the financial liberalisation policy havedeveloped from various lines of reasoning. For instance, the Keynesians such asBurkett and Dutt (1991) claim that financial markets are not self-equilibratingi.e., markets do not automatically clear while neo-structuralists such asTaylor (1983) and Van Wijnbergen (1983) highlight the role of the informalfinancial sector especially in developing countries. Also, Diaz-Alejandro(1985) and Demirguc-Kunt and Detragaiche (1998, 1999) claim that financialliberalisation may cause financial fragility while Stiglitz and Weiss (1981)and Stiglitz (2000) assert that the presence of asymmetric information infinancial markets may cause market failures in the financial system.
Up untilnow, there exists no less than five theoretical arguments against the McKinnon(1973) and Shaw (1973) financial liberalisation hypothesis. These schools ofthought oppose the conventional view by making use of various theoreticalmodels to argue against the efficacy of financial liberalisation. Thecontroversy centres on whether financial liberalisation, and specificallyinterest rate liberalisation and capital market liberalisation are a safe andinfluential road to capital formation and economic growth. Ever since the McKinnon(1973) and Shaw (1973) financial liberalisation hypothesis which criticisedfinancial repression as the cause of unsatisfactory levels of capital formationand growth especially in developing countries, researchers have also focusedtheir attention on the liberalisation of debt and equity markets, also referredto as capital market liberalisation. Bekaert et al (2003) define capital marketliberalisation as an act whereby the government of a country grants overseasinvestors the freedom to trade in the domestic capital market withoutrestrictions, while granting domestic residents the freedom to participate inthe activities of foreign capital markets. The role of capital market liberalisation in encouraging capitalformation and growth is also highly debated in the financial liberalisationliterature. Economists who are infavour of capital market liberalisation (see for instance, Greenwood and Smith,1997; Obstfeld, 1998; and Henry,2000)argue that liberalising a country’s capital market promotes effective andefficient international resource allocation, reduces cost of equity capital, improvesthe size and liquidity of the domestic capital market, increase availability offinancial resources for domestic investments and consequently improve economicgrowth.
However, economists who argue against capital market liberalisation (seefor instance, Devereux and Smith, 1994 and Stiglitz, 2000) maintain that investmentsin the capital market are mainly short term and that it is less likely thatcapital market liberalisation will promote capital formation or economic growthbecause long term investments which are likely to produce higher returns andbenefits cannot take place on the basis of short term capital. The standard policy recommendation given to developingcountries by neo-classical economists and the Internation Monetary Fund andWorld Bank representatives is the liberalisation of deposit and lendinginterest rates. Beginning from the mid-1980s, several countriesin Sub-Sahara Africa adopted structural adjustment programmes within theframework of the IMF supported Structural Adjustment Facility/EnhancedStructural Adjustment Facility and the World Bank’s Structural AdjustmentCredit/Loan programmes. One key feature of these programmes was the drive foradaptation of measures that encouraged trade and financial liberalisation andthis was a precondition for obtaining loans for developmental purposes from theWorld Bank and the IMF. Since the 1980s, several countries in the Sub-SaharanAfrican region have embarked on various degrees of liberalisation of thefinancial sector in a bid to increase savings mobilisation and capitalformation in order to improve investment levels and consequently boost economicgrowth but the effects of financial liberalisation on economic growth andcapital formation are ambiguous. This supports the view that financialliberalisation may encourage both excessive risk taking and macroeconomicinstability with detrimental effects on the macroeconomic aggregates. Research on the relationship between financialliberalisation, capital formation and economic growth mainly focuses on LatinAmerica, Asia and the developed countries with very limited research devoted tothe African countries and, in particular, the Sub-Sahara African region.
Nevertheless, a few researchers have investigated the impact of differentfinancial reform policies on economic growth in the Sub-Sahara region. Forinstance, Fowowe (2013) examines the effects of five financial reform policieson growth in 19 Sub-Sahara African countries, using panel data for the period1978 to 2000. In his investigation, he constructs two indexes that measure thegradual progression and institutional changes involved in the financial sectorreforms. The reform policies tested are bank denationalization andrestructuring; interest rate liberalisation; prudential regulation; free entryinto banking as well as the abolition of sectoral credit guidelines. Ghazanchyanand Stotsky (2013) also empirically examines the drivers of growth inSub-Sahara Africa by relating the recent growth experience to key determinantsof growth including current account liberalisation, private and public investment,the exchange rate regime as well as government consumption in 42 countries forthe period 1999 to 2011.
Furthermore, Misati and Nyamongo (2012) investigatethe dual role of financial liberalisation on growth using a bank crisis modeland a growth model similar to Ranciere et al (2006). Their study covers 34Sub-Sahara African countries from 1983 to 2008, using panel data. The authorsplace emphasis on capital market liberalisation in order to investigate thelinkage between financial liberalisation and financial fragility, and they examinethe dual effect of financial liberalisation on economic growth. Other studiesrelating to the Sub-Sahara African region include Menyah et al (2014) whoinvestigate the causal relationship between financial development and economicgrowth for 21 countries for the period 1965 to 2008 within a framework thatalso accounts for international trade.
The above mentionedresearch related to the Sub-Sahara African region investigate differentinstruments of financial liberalisation but largely ignore capital marketliberalisation and the impact of financial liberalisation policies on capitalformation. Also, majority of the existing empirical studies (for instance,Levine and Zervos (1998)), which include some developing countries from theSub-Saharan African region are cross sectional and their results have to beinterpreted with caution when considering developing countries as one cannotcompletely rely on results from cross sectional studies due to individualcountry differences. Developing countries especially those in Sub-Sahara Africahave different underlying macroeconomic and structural characteristics, e.
g.the level of economic development, the size and structure of the financialsector, population size and growth levels which should be explicitlyinvestigated and modeled. This thesis aims atfilling a gap in the literature by providing empirical evidence on the effectsof the two principal financial liberalisation policies, namely capital marketliberalisation and interest rate liberalisation, on capital formation andeconomic growth in a single econometric framework after considering country specificheterogeneities. The thesis investigates the effects of interest rateliberalisation and capital market liberalisation on capital formation andeconomic growth using improved time series estimation techniques that allow forthe existence of structural breaks. The empirical investigation begins withcarrying out a Chow test within the framework of the ordinary least squares(OLS) multiple linear regression in order to test for structural breaks.Thereafter, test for stationarity (or non-stationarity) is carried out usingthe Perron (1989) modified Dickey-Fuller unit root test which includes dummyvariables to account for one exogenous structural break as in Shretsha andChowdhury (2007). Furthermore, the long runrelationship between interest rate liberalisation, capital marketliberalisation, capital formation and economic growth is examined using theJohansen (2000) maximum likelihood cointegration procedure which allows for theexistence of structural breaks by analysing cointegration in a Gaussian vector autoregressive(VAR) model with known breakpoint dates.
Granger causality testing is alsoperformed as part of the time series estimations, in order to ascertain thedirection of causality among the variables of interest. The Granger causalitytest procedure employed considers a single shift (i.e., one structural break)in the mean of the time series and two endogenous variables and tests forcausality by estimating a VAR system with a dummy variable which is a functionof the structural breakpoint date (as in Bianchi, 1995).
The thesis also employspanel estimation techniques to examine the relationship between the two financialliberalisation policies on capital formation and economic growth. Initially, afixed effects model that includes year dummies as well as country dummies isestimated to ascertain the relationship (as in Fowowe, 2008). Thereafter, arandom effects model within the framework of Mundlak (1978) is estimated inorder to decompose the effects of both policies on capital formation andeconomic growth into transitory effects and permanent effects (as in Bender andTheodossiou, 2015). 1.2.Aims and Objectives of the StudyThis research aims toassemble, organise, analyse and synthesise available data from ten Sub-SaharaAfrican countries in order to provide empirical evidence in a comparativeframework on the efficacy of financial liberalisation policies and the relativemerits of contrasting approaches in enhancing economic growth and capitalformation, thus contributing to the theoretical and policy discourse. Thecountries under investigation include Ghana, Ivory Coast, Kenya, Mauritius,Nigeria, Sierra Leone, South Africa, Tanzania, Uganda and Zimbabwe.
Theobjectives of this thesis are to empirically investigate for the selectedSub-Sahara African countries: (1) Theeffect of interest rate liberalisation on capital formation. (2) Theeffect of capital market liberalisation on capital formation. (3) Theeffect of interest rate liberalisation on economic growth. (4) Theeffect of capital market liberalisation on economic growth. A review of theliterature on the impact of financial liberalisation on capital formation andgrowth is used as a springboard for the empirical analysis. 1.3.Significance of the StudyThe purpose of thisresearch three fold.
This includes a review of the literature on the rationalefor financial repression, to analyse both the theoretical and empiricalliterature on the relationship between two main financial liberalisationpolicies and capital formation and economic growth, and to empiricallyinvestigate the effects of both financial liberalisation policies on capitalformation and on economic growth with specific reference to the Sub-SaharanAfrican region for the period from 1970 to 2014. From the mid-1980s, manycountries in Sub-Sahara Africa adopted structural adjustment programs withinthe framework of the IMF supported Structural Adjustment Facility/EnhancedStructural Adjustment Facility and the World Bank’s Structural AdjustmentCredit/Loan programs. One key feature of these programmes was the drive foradaptation of measures that encouraged financial liberalisation, and this conditionhad to be fulfilled before the World Bank or the IMF granted loans to these Sub-SaharanAfrican countries. Hence, the period covered by the thesis covers the pre-financial liberalisation era and the post-financial liberalisation era. Furthermore, a vastmajority of studies in the financial liberalisation literature haveconcentrated on developed countries as well as emerging countries especially inEast Asia and Latin America. There is very limited research devoted to Africancountries in general and the Sub-Saharan African region in particular.Moreover, the effects of financial liberalisation on capital formation havebeen ignored largely in the financial liberalisation literature. Therefore,this thesis redresses this imbalance by investigating and providing evidenceand insight into the relationship between the two main financial liberalisationpolicies namely capital market liberalisation and interest rate liberalisation,and their effects on capital formation and economic growth in the Sub-SaharaAfrican region.
In addition, sincedeveloping countries, especially those in Africa have different levels ofeconomic development, population size, poverty rates and income inequality itis important that this inter-countryheterogeneity be explicitly incorporated in the statisticalmethodology. Essentially, this research aims at informing the theoreticaldiscourse on the consequences of financial liberalisation on capital formationand economic growth as reflected in the empirical record. However, byhighlighting the relative success or failure of policy recommendations arisingfrom the two theoretical approaches, it provides evidence on the profoundeconomic and sociological consequences of such policies as these macroeconomicaggregates affect the current and future wellbeing of entire populations. Hence, this research is a significant contribution to the economics literatureand theoretical controversy but also informs the broad social literature, andultimately development policy in Sub-Sahara Africa. Initially, the study employs modified time series techniquesto investigate the impact of interest rate liberalisation and capital marketliberalisation in each country separately.
The findings are compared acrosscountries to identify significant differences in the path and intensity ofcapital formation and economic growth. Thereafter, fixed effects panel estimationtechniques are also employed to assess the impact of the two financialliberalisation policies to all the ten Sub-Saharan African countriescollectively after controlling for inter-country heterogeneity. A major noveltyof this thesis is that the time series estimations allow for structural breaks.Furthermore, this thesis utilizes a new random effects panel estimationtechnique namely the so-called Mundlakmethodology that enables the decomposition of the effects of the financialliberalisation policies on capital formation and growth into transitory effectsand permanent effects. 1.4.Organisation of the StudyThis thesis has been designedto include into nine Chapters. Chapter one is the introductory chapter whichpresents the background to the study, aims and objectives of the study as wellas the significance of the study.
The financial liberalisation literature isexplored in two separate chapters. Chapter two presents a review of thetheoretical and empirical literature on the effects of interest rateliberalisation on capital formation and on economic growth, while Chapter threepresents a review of the theoretical and empirical literature on the effects ofcapital market liberalisation on capital formation and on economic growth. The literature review is used as a springboardfor the empirical analysis. In Chapter four, the selected Sub-Saharan Africancountries under investigation are examined in more detail. This comprises anoverview of the economy and financial system of each of the countries as wellas capital formation and economic growth trends for the pre-liberalisation andpost-liberalisation periods. The empirical investigationis presented in four chapters. In Chapter five, the thesis presents an overviewof the data, methodology and research design. The choice of the financialliberalisation measures are also discussed and all the variables are defined.
Chaptersix presents the time series empirical models and discusses the time seriesestimation techniques while Chapter seven presents and discusses the timeseries estimation results. In Chapter eight, the panel data models, methodologyand estimation results are presented and discussed. Finally, Chapter ninepresents the conclusions of the thesis.
It comprises a rundown of the mainresearch findings, policy recommendations, limitations of the thesis as well assuggestions for further research.