A thesis submitted to the Faculty of Commerce, Law and Management, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the Degree of Doctor of Philosophy in Economics, October, 2017 / This thesis contains three core chapters that assess the performance of alternative monetary policy frameworks and instruments in stabilizing 10 selected African economies. Literature and practice show that Advanced Economies (AEs) and Emerging Market Economies (EMEs) are mostly adopting the ination targeting (IT) framework. This framework relies on active use of the interest rate as a policy instrument for macroeconomic stabilisation. Di⁄erent from AEs and EMEs, the majority of African countries are characterized by low nancial market development, frequent supply shocks and volatile terms of trade. These features impede the e¢ ciency of the IT framework and the interest rate instrument. Supply shocks imply that ination is not only demand driven. Volatile terms of trade translate into excessive exchange rate uctuations. Due to these factors, policy practice in Africa remains largely divergent from the global trend. Authorities still rely on monetary aggregate targeting (MAT) with de facto managed exchange rates. However, the MAT framework is also failing to stabilize economies. This follows instability of the key factors, such as the money demand, upon which the framework is anchored. Furthermore, controlling exchange rate movements is a challenge due to weak balance of payments positions. It is not surprising, therefore, that the majority of African economies still remain in the grip of macroeconomic instability. Ination and GDP targets are rarely met and they also remain volatile. The perverse macroeconomic features and the perceived failure of the MAT regime have necessitated the search for alternative monetary frameworks and instruments. In this study, we join the search by specically focussing on three questions. First, given the macroeconomic landscape in Africa, what is the relative performance of the interest rate vis--vis the monetary aggregate as instru iv ments for macroeconomic stabilization? Secondly, how do these instruments perform when apart from ination and output stabilization, monetary policy also engages in smoothing exchange rate uctuations? Thirdly, what is the relative performance of ination targeting vis--vis nominal GDP targeting as alternative monetary policy regimes for macroeconomic stabilization in African economies? Although the success of monetary policy largely relies on appropriate conguration of monetary policy frameworks and instruments, answers to these questions remain controversial and scanty for African economies. In order to address these questions, we formulate a New Keynesian Dynamic Stochastic General Equilibrium (DSGE) model. In this model, money is non-separable from consumption in the utility function. We estimate the model using the Maximum Likelihood method with quarterly data mostly from 1990 to 2014. The data is obtained from the International Financial Statistics (IFS). The thesis has ve chapters. Chapter 1 is the general background to the research problem. Chapters 2, 3 and 4 are distinct but related core chapters addressing three specic research questions. Chapter 5 is the conclusion. In Chapter 2, we compare the performance of the monetary aggregate and the interest rate as alternative instruments for stabilizing ination and output in 10 selected countries. Results show that the monetary aggregate is superior in stabilizing 5 economies. In the other 5 countries, it is the interest rate instrument which performs better. In the former group of countries, the monetary aggregate plays a relatively large role in macroeconomic dynamics while in the latter the interest rate is more signicant. These results partly reect di⁄erences in nancial market development between the two groups of countries. Overall, we nd a weak role of the interest rate compared to the monetary aggregate in driving aggregate demand dynamics. The exchange rate is also found to be a key driver of macroeconomic dynamics. Our re v sults suggest three things: First, authorities in Africa need to be cautious of a blanket adoption of the interest rate as a sole monetary policy instrument. Second, authorities will nd it di¢ cult to stabilize economies using the interest rate based frameworks. Third, exchange rate stability is key to macroeconomic stability in Africa. In Chapter 3, we extend the authoritiesobjective function. In addition to minimizing ination and output volatility, authorities also use the interest rate or money supply rules to smooth exchange rate uctuations. The results show that macroeconomic performance is enhanced when authorities smooth exchange rate uctuations in 4 of the 10 countries. The gains from exchange rate smoothing mostly arise from a reduction in ination and exchange rate volatility but not fromoutput. In the other 6 countries, exchange rate smoothing worsens macroeconomic performance. These results reect the fact that the exchange rate exerts a relatively large inuence in macroeconomic dynamics in the rst group of countries compared to the latter. Exchange rate smoothing therefore minimizes the pass-through of the exchange uctuations to ination and output leading to better performance. Overall, the ndings suggest that exchange rate smoothing is harmful in Africa. Where exchange rate smoothing delivers gains, appropriate thresholdsofsmoothingneedtobeobservedtoavoidpolicyinducedmacroeconomic instability. Authorities should also smooth temporal rather that structural shifts in the exchange rate level. In Chapter 4, we compare the performance of ination targeting (IT) vis-vis nominal GDP targeting (NGDPT) as alternative monetary policy frameworks for macroeconomic stabilization. We examine the strict and exible versions of these policy regimes. We also include a hybrid regime which combines elements of IT and NGDPT. Results show that the hybrid regime performs better in 5 countries. In the other 4 countries, it is the strict ination targeting that performs better. In 1 country, exible ination tar vi geting is optimal. The results also reveal that demand shocks dominate but are closely trailed by supply and exchange rate shocks in explaining macroeconomic uctuations. The multiplicity of signicant shocks is key in explaining the dominance of the hybrid regime. The hybrid regime successfully handles shocks that can neither be optimally handled by the IT regime nor the NGDPT regime alone. These results have several implications. First, demand management alone is insu¢ cient to stabilize African economies. Second, identifying dominant shocks is critical for choosing robust monetary policy regimes. Third, the multiplicity of signicant shocks implies that choosing monetary policy frameworks and hence macroeconomic management process is more complex for African policy makers. Overall, the results have several policy implications which are outlined in Chapter 5. First, they suggest a cautious approach towards generalized adoption of the interest rate over the monetary aggregate as a monetary policy instrument in African economies. This contradicts the current wave of monetary policy changes sweeping across African countries. Secondly, the signicanceoftheexchangeraterenderscredencetoexchangeratesmoothing in Africa. The ndings, however, suggest that exchange rate smoothing can either enhance or worsen macroeconomic performance. Where it enhances macroeconomic performance, authorities must carefully consider the thresholds of smoothing to avoid creating macroeconomic instability. Authorities need not ght structural shifts in exchange rates levels through smoothing. This would help to preserve the shock absorbing role of the exchange rate. Finally, the prevalence of demand, supply as well as exchange rate shocks makes the hybrid monetary policy regime which combines elements of IT regime as well as NGDPT regime to perform relatively better in stabilizing the majority of the economies. Given the multiplicity of shocks, authorities inAfricaneedtocomplementdemandmanagementwithpoliciesthataddress supply side and exchange rate bottlenecks to ensure sustainable macroeco vii nomic stability. Overall, the ndings suggest that there is scope to improve monetary policy performance in Africa by adopting suitable frameworks and instruments. The results also highlight the problem of tackling monetary policy issues with a "one size ts all" approach. / GR2018

The CFA consists several African economies adhering to one of two common currencies, and one of two central banks. The rules of the monetary union provide for the pooling of foreign assets, and the regulation of monetary expansion in each country. The French treasury guarantees the convertibility of CFA Francs into French Francs at a fixed rate. The thesis examines the impact of CFA membership on the macroeconomic performance of member states, assessing the claim that CFA institutions have influenced capital and labour markets, and have modified short run adjustment to external shocks. There are a number of reasons why CFA membership could facilitate higher investment, (i) The rules governing money creation may lead to greater monetary prudence, and so lower inflation and price variability, and less uncertainty for investors, (ii) Guaranteed convertibility means that firms will never be prevented from importing capital goods by a lack of foreign exchange, (iii) Convertibility may encourage a greater degree of integration between French and CFA capital markets, so that domestic investment is not entirely dependant on domestic saving. A model of investment is constructed to incorporate these effects, and tested using time series and cross-sectional data. Support is found for (i) and (ii), but not for (iii). If African labour markets are characterised by nominal wage inertia, the enforced low inflation may lead to excessive real wages, and CFA membership may impair efficient allocation of labour. However, evidence suggests this characterisation is usually inappropriate. The pegged exchange rate may lead to persistent external imbalances: devaluation is not an option in response to a negative trade shock. This will not be a problem as long as an effective substitute for devaluation is found. A CGE model is constructed to examine the viability of various devaluation substitutes, none of which are found to be adequate.

This research attempts to evaluate the stability of money demand functions and estimate monetary inflation models in the East African Community (EAC), using quarterly aggregate data that range from 2000Q1 to 2012Q3. We used Johansen co-integration analysis to estimate and analyse the stability of the M3 money demand model for each country member of the EAC. From this estimation, we derived a country-specific measure of money overhang. We compared its forecasting power of future inflation with that of money stock growth, and money stock available in the economy. Regarding country-specific money demand functions, with the exception of Uganda, we identified a reasonable and stable country-specific M3 money demand model. Also, for predicting future inflation, the estimation results showed that M3 money stock growth is more reliable in Burundi and in Kenya, while M3 money overhang is preferable in Rwanda and M3 money stock in Tanzania. As both country-specific and regional (EAC area) information on monetary quantity growth and its impact on price level is important to know in a monetary union, we considered the EAC area as a single market and attempted to estimate the aggregate (EAC area) demand functions for broad money M2 and M3 using Johansen co-integration analysis. The estimated long-run aggregate money demand models M2 and M3 appeared to be stable over the sample period. However, the aggregate M2 and M3 at the EAC level were proven to be weakly exogenous, which should discard them for consideration at the EAC level as the intermediate targets variables in order to achieve the overall objective of price stability in the EAC region. Instead, short-term interest rate should be given a prominent role in monetary policy framework at the EAC level.

Thesis (MBA)--Stellenbosch University, 2008. / ENGLISH ABSTRACT: The ultimate goal of the African Union is full political and economic integration, which includes a monetary union with a common currency for all member states of the African Union. This monetary integration is proposed to take place in two stages: firstly, through five regions, and secondly, through complete integration. This report examines current trade data for member states of the African Union using k means duster analysis to group countries according to trade patterns. Analysis was performed for the actual US dollar value of trade, as well as considering only the presence or absence of trade. There are limitations to the data collected: firstly, they are annual data, which masks fluctuations in trade due to economic conditions or political developments. Secondly, they are subject to missing or under-reported values. The focus of this research report was to consider trade figures for the first time, and the limitations were considered acceptable in view of the aim of achieving a first approximation of results. When considering all solutions, there are overlaps between clusters, but no definite patterns emerge that are common to all analyses. Considering the F and Euclidean distances of all solutions, the best appears to be that for clusters derived from analysing trade figures between Africa and its trading partners outside Africa. Further analysis of this solution failed to demonstrate viable clusters. The final conclusion to be made from this analysis is that k means clustering of trade figures for member states of the African Union does not generate viable clusters that could be used as steps towards full monetary integration in Africa. Given this conclusion it is recommended that the stepwise progression towards full monetary integration be considered by utilising existing economic arrangements, i.e. by using the five Regional Economic Communities proposed by the African Union. / AFRIKAANSE OPSOMMING: Die uiteindelike doel van die Afrika-unie is volledige politieke en ekonomiese integrasie, wat 'n monetere unie met 'n gemeenskaplike geldeenheid vir al die lidstate van die Afrika-unie insluit. Hierdie monetere integrasie word in twee stadiums beoog: eers deur vyf streke, en daarna deur volledige integrasie. Hierdie verslag ondersoek die huidige handelsdata vir lidstate van die Afrika-unie deur k gemiddelde trosanalise te gebruik om lande volgens handelspatrone te groepeer. 'n Analise is ook gedoen van die werklike VS-dollarwaarde van handel, en deur die aanwesigheid of afwesigheid van handel in aanmerking te neem. Daar is beperkings op die data wat ingesamel is: eerstens is dit jaarlikse data, wat skommelings in handel as gevolg van ekonomiese toestande of politieke ontwikkelings verberg. Tweedens is hulle onderworpe aan ontbrekende of ondergerapporteerde waardes. Die fokus van hierdie navorsingsverslag was dus om handelsyfers vir die eerste keer te oorweeg, en die beperkings is aanvaarbaar beskou in die lig van die doel om 'n eerste benadering van resultate te verkry. Wanneer aile oplossings oorweeg word, is daar oorvleueling tussen trosse, maar geen definitiewe patrone ontstaan wat vir alle analises geld nie. Wanneer die F- en Euklidiese afstande van alle oplossings oorweeg word, lyk dit asof die beste die trosse is wat verkry is uit die analise van handelsyfers tussen Afrika en sy handelsvennote buite Afrika. Verdere analise van hierdie oplossing het nie lewensvatbare trosse aangedui nie. Die finale gevolgtrekking wat uit hierdie analise gemaak kan word, is dat k gemidderde trosvorming van handelsyfers vir lidstate van die Afrika-unie nie lewensvatbare trosse genereer wat gebruik kan word as stappe in die rigting van volledige monetere integrasie in Afrika nie. Met die oog op hierdie gevolgtrekking word daar aanbeveel dat die stapsgewyse vordering na volledige monetere integrasie oorweeg moet word deur bestaande ekonomiese reelings te gebruik, d.w.s. deur die vyf Streeksekonomiese Gemeenskappe te gebruik wat deur die Afrika-unie voorgestel is.

Given the importance of monetary policy in the operation of a successful modern economy and the use of official interest rates as tools in its implementation, this study investigates the implications of changing bank concentration on the operation of the Interest Rate Pass Through (IRPT) of official rates to bank lending and deposit rates. This is an issue made more poignant by growing mergers, acquisitions and bank consolidation exercises around the world that have brought interest to their implications for economic performance. However, with contention high in the industrial organisation theory on the likely relationship between bank concentration and the IRPT, and the outcomes of empirical investigations producing conflicting evidence, the desire to investigate the issue in the African context necessitated a thorough empirical investigation of four African countries (South Africa, Botswana, Nigeria and Zambia). This study not only extended the investigation of the issue to the African context, but it merged different IRPT measurement techniques that had not been jointly applied to this particular issue, namely; Symmetric and Asymmetric Error Correction Models, Mean Adjustment Lags, Ordinary Least Squares estimations and Autoregressive Distributed Lag models. These measures of the IRPT were compared with three firm concentration ratios on two different levels of analysis, one, over the entire period and, another, through eight year rolling windows. The results reveal that bank concentration can sometimes be related to the speed and magnitude of the IRPT but that these relationships are not consistent amongst the countries, over the entire sample period or across the two levels of analysis, suggesting reasons why empirical results have arrived at contrasting conclusions. The results revealed more evidence of a relationship between bank concentration and the magnitude of the IRPT than between bank concentration and the speed of the IRPT. Furthermore, where relationships were identified there was evidence supporting both the structure conduct performance hypothesis and the competing efficient market hypothesis as the true representation of the relationship between bank concentration and the IRPT. The key implication of the result for African countries is that increased bank concentration through bank consolidation programmes should not be automatically regarded as detrimental to the effective implementation of monetary policy through the IRPT. Consequently,banking sector regulation need not stifle bank consolidation and growth to preserve monetary policy effectiveness. Rather, since the relationship cannot be neatly represented by a single theory or hypothesis each country must determine its own interaction between bank concentration and its IRPT before policies regarding the banking sector concentration and effective monetary policy, through the use of official interest rates, are determined.

Most central banks use short-term interest rates as their main instrument of monetary policy. It is assumed that a change in policy rate will influence interest rates set by commercial banks, but this is not usually the case. Commercial banks adjust their interest rates in response to changes in policy rate with lags, which make their interest rates sticky. Stickiness in commercial banks interest rates have been seen as an obstacle to the smooth transmission of monetary policy decisions. Despite the importance of the transmission process, little attention has been given to a systematic measurement of the degree of response of commercial banks‟ interest rates to changes in monetary policy stance in the Anglophone West African countries, specifically within the West African Monetary Zone (WAMZ) economies. Against this backdrop, this study explores the interest rate adjustment dynamics using monthly interest rate series on discount rate, treasury bill rate, commercial banks‟ deposit and lending rates from 1989 to 2009 (for Gambia, Nigeria and Sierra Leone) and from 2000 to 2009 (for Ghana). Specifically, the study set out to examine how lending and deposit rates respond to changes in the official rates and to see whether there is a convergence among the rates over time. Also, to examine the relative adjustment of commercial bank lending rates to changes in the official rate when there is disequilibrium. The analyses were twofold: a full sample period and a rolling window analysis. Following Cottarelli and Kourelis (1994), the study employed cointegration technique and an asymmetric error correction model to obtain the short-run and long-run parameters from which the error correction coefficients, mean adjustment lags and asymmetric mean adjustment lags were estimated. The results for the entire sample period revealed that the long-run pass-through in Nigeria was 81% and 67% for lending rates and deposit rates respectively. In Ghana, it was 66% and 69% for lending and deposit rates respectively. While in Sierra Leone, long-run pass-through was 62% and 72% for lending and deposit rates respectively. In Gambia, it was 50% and 40% for lending and deposit rates respectively. On the other hand, the short-run pass-through was found to be lower compared to the long-run pass-through: in Nigeria it was 66% and 47%; in Gambia, 26% and 29%; in Sierra Leone, 30% and 13%; and in Ghana, -6% and 35% for lending and deposit rates respectively in each country. The pass-through estimates for the rolling windows were mixed for short-run and long-run pass-through. The mean adjustment lags suggest that the speed of adjustment of Lending rates for full sample period were two, two, seven and twelve months in Nigeria, Ghana, Sierra Leone and Gambia respectively. While for deposit rates they were five, six, seven and eighteen for Ghana, Nigeria, Gambia and Sierra Leone respectively. The average speeds of adjustment for the rolling windows were four and five months for lending and deposit rates respectively. Weak evidence of convergence was found in lending and deposit rates in the short-run and long-run pass-through among the countries. However, the results suggest that the magnitude and speed of the pass-through amongst the countries on average were high compared to emerging Asian countries. Significant asymmetric adjustments were found in the lending rates for Gambia and Sierra Leone, while in Gambia and Nigeria there were asymmetries in deposit rates. Based on the evidence provided, interest rate pass-through is high in Nigeria and Ghana compared to Gambia and Sierra Leone and this calls for the harmonization of financial policies on the part of the financial authorities in the WAMZ. Viewed solely from an interest rate pass-through, the lack of convergence among the countries suggests that WAMZ is far from ready for a monetary union. The relatively low pass-through in some of the countries suggests rigidity in the banking system which may be due to underdevelopment of the system. Thus efforts geared toward strengthening the banking system and the financial system as whole would further enhance the prospect of a monetary union among them.

The Southern African Development Community (SADC) aims to have a regional central bank by 2016 and a common currency by 2018. The member states are at the early stages of the process of regional economic integration, having launched a free trade area in 2008. Monetary integration is an advanced stage of regional economic integration that requires progressive changes in the participating countries. The purpose of this study is to determine the feasibility of monetary integration within the SADC countries and hence, provide policy recommendations to guide the integration process. To accomplish this, the study analyses the extent to which the member states meet the criteria for an optimum currency area (OCA) as well as the degree to which their economies are converging. The study finds that the main macroeconomic objectives of SADC countries differ due to a difference in the relative importance of monetary policy instruments in member states, which influences each country’s commitment towards achieving the macroeconomic convergence targets and harmonising policies. A more appropriate approach to macroeconomic convergence would be to allow for variable speed, geometry and depth in each country as premature adherence to convergence targets could prevent a harmonisation of the economies in the future and possibly destabilise the union. In addition, the study investigates the importance and similarities of the monetary aggregate channel, the interest rate channel, the exchange rate channel and the credit channel in the transmission of monetary policy using VAR analysis. This is important when considering monetary integration because differences in transmission mechanisms can result in asymmetric behaviour between member states, which in turn will prevent harmonisation of their economies. The results of the analysis suggest that SADC member states display asymmetries in their responses to monetary policy shocks as well as the relative importance of transmission mechanisms. In addition, the results suggest that national monetary policy is generally inefficient in determining economic performance in the member states. Furthermore, the study finds that the failure to meet the OCA criteria implies that the SADC member states will respond asymmetrically to shocks within a monetary union. With no effective alternative adjustment mechanisms in place, the effects of the shocks will endure in union members and possibly widen existing cyclical variation. Hence, monetary integration would not result in harmonisation of the economies of member states. It is therefore, concluded that the SADC countries were not suitable for monetary integration at present.

This research deals with achieving and maintaining price stability in Sub-Saharan Africa (SSA) through the practice of monetary-policy transparency (MPT). On the one hand, MPT refers to a monetary strategy whereby the central bank is insulated from political influence and made accountable to society through disclosure of its policies, procedures, economic models, data and forecasts, operations and political practices (such as objectives, personnel independence, and the like). On the other hand, price stability refers to achieving and maintaining low and stable levels of inflation conducive for long-term planning and poverty alleviation. The primary objective of this research was to investigate MPT in SSA as it represents a powerful means whereby economic agents’ expectations may be coordinated and managed by the central bank to achieve its societal, objective function of low inflation. The empirical evidence shows that, first, a dependent central bank is more likely to slip into hyperinflation. Second, a SADC (2008) model central bank law is not independent enough to be used as a benchmark for any central bank or as a charter for a regional central bank. Third, the degree of central bank independence in SSA is relatively lower than that in industrialised economies. Fourth, the determinants of MPT in SSA are trade openness, and financial depth that are important factors influencing policy-makers to adopt monetary-policy transparency. Fifth, MPT is associated with a decline in the inflation rate. Sixth, MPT had no significant effect on economic output, whilst trade openness was positively associated with real GDP.

The study aimed to evaluate the impact of economic freedom on economic growth and investments in the Economic and Monetary Community of Central Africa (CEMAC). The region was created in 1994 by the six states of Cameroon, Chad, the Central African Republic, the Republic of Congo, Gabon and Equatorial Guinea. CEMAC countries comprise low and middle-income countries that share the same currency - the CFA Franc. The CEMAC countries were observed between 1995 and 2008 and panel regression methodologies were employed. A positive impact of economic freedom on economic growth was established using fixed effects method and the generalised method of moments. The impact of a unit increase in the economic freedom index on GDP per capita ranged between 72.65 and 124.51 units (dollars) increase on GDP per capita, ceteris paribus. Economic freedom was also found to Granger-cause economic growth. The results underline a significantly positive relationship between economic freedom and economic growth which is consistent with existing literature. The impact of economic freedom on domestic investment and foreign directs investment was then examined. With regard to domestic investment, economic freedom was found to be statistically significant and positive in all specifications of the model, thereby implying that a unit increase in the economic freedom index increases domestic investment by values of between 0.50 and 0.69 dollars in the CEMAC. The results obtained were consistent with most findings on the relationship between economic freedom and investments. With regard to the relationship between economic freedom and foreign direct investment inflows, economic freedom was unexpectedly statistically insignificant in most specifications of the model. The latter implies that economic freedom does not have a significant impact on foreign direct investment in the CEMAC. However, the study revealed that economic freedom Granger-causes foreign direct investment but foreign direct investment does not Granger-cause economic freedom. This means that economic freedom precedes foreign direct investments, and foreign direct investments do not precede economic freedom. The study strongly recommends an improvement of institutions in the CEMAC in order to enjoy greater levels of economic freedom and therefore foster economic growth and domestic investment in the region.

Dissertation (PhD)--Stellenbosch University, 2003. / ENGLISH ABSTRACT: The endogenous money theory states that the money supply responds endogenously to the demand for credit. The money supply is not exogenously determined by the central bank. The endogenous theory is associated with the Post Keynesian school. It has been tested extensively for developed countries, where it was found that the modern credit-driven world is characterised by an endogenous money supply. The contribution of the present study is to extend this analysis to developing countries, specifically twelve countries in the SADC region. To examine the applicability of the endogenous money theory to developing countries, the thesis begins with an overview of the views of the different schools of thought on the role of money. The areas of consensus and disagreement within the Post Keynesian school are discussed. The theoretical basis of the thesis is the ‘structuralist’ Post Keynesian view that money cannot be endogenous if the financial system in a country has not reached the final stages of development. The ‘structuralist’ hypothesis is tested for the SADC countries by examining the demand and supply of credit money in each country. It was found that households do not generally have full access to formal credit markets. Changes in the money supply are not determined by changes in private sector credit in many of the countries. The analysis was then extended to the institutional environment in each country. A financial institutional index was developed to facilitate comparison between the SADC countries. It was shown that South Africa is the only country in the SADC area that has a financial system that can be classified as ‘largely developed’. It is also the only country where changes in the supply of money are predominantly credit-driven. Post Keynesians maintain that the money supply is endogenous and interest rates are exogenous. Interest rate mark-ups and spreads are assumed stable over the business cycle. This notion is challenged by the ‘structuralist’ Post Keynesians. To test the theory of stable interest rate mark-ups and spreads, data for each individual country were examined. Neither interest rate spreads, nor interest rate mark-ups were found to be stable. Interest rate spreads are generally higher in developing countries than in developed countries. No clear pro- or counter-cyclical variation in spreads was found. Finally, an econometric model was developed and the links between financial development and growth were examined. By looking at 49 developed and developing countries, it was found that financial development is strongly linked to economic growth. Financial repression and high interest rate spreads cause growth to be depressed. Financial development and increased competition in the banking sector will lead to higher real economic growth rates. In an environment where the financial system has not reached the stage where money is endogenous, the lack of financial institutional development stifles economic growth. / AFRIKAANSE OPSOMMING: Die teorie van ‘n endogene geldvoorraad aanvaar dat die aanbod van geld endogeen reageerop die vraag na krediet. Die geldvoorraad word nie eksogeen bepaal deurdie sentrale bank nie. Die endogene gedvoorraad teorie word geassosieer met die Post Keynesiaanse skool. Dit is reeds getoets vir ontwikkelde lande, waar die bevinding was dat ‘n endogene geldvoorraad ‘n eienskap is van ‘n moderne kredietgedrewe wereld. Hierdie tesis maak ‘n bydrae deur die analise uit te brei na ontwikkelende lande, spesifiek twaalf lande in die SADC streek. Om die toepasbaarheid van die endogene geldvoorraad vir ontwikkelende lande te toets, begin die tesis met ‘n oorsig van die verskillende denkskole se sienings oor die rol van geld. Die areas waar Post Keynesiane ooreenstem en verskil word bespreek. Die teoretiese basis van die tesis is die ‘strukturalistiese’ Post Keynesiaanse siening dat die geldvoorraad nie endogeen kan wees indien die finansiele sisteem in ‘n land nog nie die finale ontwikkelingstadia bereik het nie. Hierdie hipotese van die ‘strukturaliste’ word getoets vir die SADC lande deur te kyk na die vraag na en aanbod van krediet in elke land. Daar is bevind dat huishoudings oor die algemeen nie volledige toegang het tot formele kredietmarkte nie. Veranderinge in die geldvoorraad word nie in al die lande veroorsaak deur veranderinge in privaat sektor kredietverlening nie. Hierdie analise word dan uitgebrei na die institusionele omgewing in elke land, ‘n Finansiele institusionele indeks is ontwikkel om vergelyking tussen die SADC lande moontlik te maak. Daar is bevind dat Suid Afrika die enigste land is met 'n finansiele sisteem wat geklassifiseer kan word as ‘grotendeels ontwikkeld’. Dit is ook die enigste land waardie geldvoorraad beduidend kredietgedrewe is. Post Keynesiane glo dat die geldvoorraad endogeen is en rentekoerse eksogeen. Rentekoersmarges word gesien as stabiel oor die konjunktuursiklus. Hierdie aanname word bevraagteken deur die ‘strukturalistiese’ Post Keynesiane. Die teorie van stabiele rentekoersmarges word getoets deur te kyk na data vir elke individuele land. Die bevinding is dat rentekoersmarges nie stabiel is nie. Marges is oor die algemeen hoer in ontwikkelende lande as in ontwikkelde lande. Daar is geen duidelike pro- of kontrasikliese variasies in rentekoersmarges gevind nie. Laastens is ‘n ekonometriese model ontwikkel om die skakels tussen finansiele ontwikkeling en groei te ondersoek. Deur te kyk na 49 ontwikkelde en onontwikkelde lande, is daar bevind dat finansiele ontwikkeling en groei ‘n sterk verband toon. Finansiele onderdrukking en hoe rentekoersmarges lei tot laer ekonomiese groei. Finansiele ontwikkeling en groter mededinging in die bank sektor sal lei tot hoer reele ekonomiese groeikoerse. In ‘n omgewing waar die finansiele sisteem nog nie die stadium bereik het waar geld endogeen is nie, sal die gebrek aan finansiele institusionele ontwikkeling ekonomiese groei benadeel.