A research report submitted to the Faculty of Humanities, University of the Witwatersrand, in partial fulfilment of the requirements for the degree of Master of Arts (in the field of ICT Policy and Regulation). 9th August 2016. / Digital transformation on a national level is a framework that has been applied to a number of different contexts. Studies in both developed and developing countries have exhibited digital transformation in a manner that reflects its applicability across contexts and scenarios. However, this research explored what happens when the same is applied to organizational contexts in a developing country. The research did not divert too far from the national application of a digital transformation framework, but merely sought to incorporate the organizational perspective, and the different considerations that arise in commercial banks in Kenya; an area which was previously under-explored. A conceptual framework was developed to study only particular elements of digital transformation from qualitative analysis and different sources of data. The findings of this study illustrated that there is a huge uptake of technologies in these commercial banks, but also notes a significant number of limitations that currently exist. The report concludes with proposals as to how these limitations can be addressed through various recommendations, and also considers other avenues for improvement, and future research that can later be applied other contexts. / GR2017

PhD (Finance), School of Economic and Business Sciences, UNIVERSITY OF THE WITWATERSRAND, JOHANNESBURG 8th June, 2016 / This study examines the relationship between financial innovation and financial performance of commercial banks in Kenya, as well as the drivers of financial innovations at both firm and macro levels. The financial innovations covered are the branchless banking models, which represent a departure from the traditional branch-based banking. More specifically, the financial innovations covered are: Mobile banking, agency banking, internet banking and Automated Teller Machines (ATMs). The study uses 10-year panel (secondary) data for the period spanning year 2004 to 2013. The study conducts an empirical analysis of the four types of financial innovations using three econometric models. The models have been specified using Koyck distributed lag models and estimated using dynamic panel estimation with System Generalised Method of Moments (GMM). The speed of adjustment of bank financial performance to financial innovation as well as the speed of adjustment of financial innovation to the financial innovation drivers has been tested using Koyck mean and median lags. The empirical results provide strong evidence of the link between financial innovations and bank financial performance with respect to Kenyan commercial banks. The study makes a number of other findings. Firstly, financial innovations significantly contribute to firm financial performance and that firm-specific factors are more important to the firm’s current financial performance than industry factors. Secondly, firm-specific variables significantly drive financial innovations at firm level with firm size being the most significant driver of financial innovation at firm level. The firm specific factors include firm size, transaction costs, agency costs, and technological infrastructure at firm level. Thirdly, macro level variables significantly drive financial innovation at firm level with regulation being the most important driver at macro level. The macro level drivers reviewed include: Regulation and taxes, incompleteness in financial markets, technological infrastructure at macro level and globalisation. Lastly, the existence of reverse causation between firm financial performance and firm financial innovation is established. The speed of adjustment of firm financial performance to financial innovation has been determined. The results show that it takes on average 1.179 years for bank financial performance to adjust to the four financial innovations studied. Secondly, it takes less than a year (0.368 years) to accomplish 50% of the total change in firm performance following a unit-sustained change in the financial innovations. Moreover, mobile banking has the shortest mean lag (2.849) while ATMs have the longest mean lag (4.926). Therefore, it takes approximately three years for mobile banking to adjust to financial innovation drivers at firm level and on average five years for ATMs to adjust to the financial innovation drivers. By and large, the speed of adjustment of financial innovations to macro level drivers is higher than the speed of adjustment of financial innovations to firm level drivers. This study has made significant contribution to the body of knowledge in the field of financial innovations. The study has developed an econometric model which captures four financial innovations in a single study and empirically used the model to test their link to firm financial performance. The second and third econometric models have also captured the drivers of financial innovations at firm and macro levels. The reviewed literature observes that previous studies have largely focused on financial products in developed countries at the expense of emerging financial innovations in developing countries. In addition, previous studies have also largely ignored empirical approaches to the study of financial innovations. This study has empirically established the link between financial innovations and firm performance by modelling the four innovations in single model in a developing country (Kenya) context. One of the major contributions of this study is the establishment of the speed of adjustment of firm performance to financial innovations and the speed of adjustment of financial innovations to financial innovation drivers at both firm and macro levels. Lastly, the study has developed an original conceptual financial innovation value model (Fig. 6.1), which will be used in future financial innovation studies. This study has a number of managerial and policy implications which have been reviewed in the study. / MT2017

Financial reforms were a major component of structural adjustment programs deemed necessary for developing countries in the mid 1980s. These were not only meant to improve the sector, but would ultimately enhance economic growth and help in poverty alleviation. At the top of these reforms was financial liberalisation. Kenya, like many other sub-Saharan African countries, undertook financial liberalisation in 1991, one of the measures was decontrolling interest rates. With market driven interest rates in place it was assumed that there would be increased efficiency in bank lending, as well as growth in credit availability as deposits increased. A key indicator of this improved intermediation process would be a narrowing interest rates spread, that is, the margin between the deposit and lending rate. Paradoxically, however, the expected benefits of these reforms did not accrue to Kenya's banking sector. This study focuses on financial reforms and the spread of interest rates in Kenya's banking sector. Using a trend analysis, spanning the period before and after liberalisation, interest rates spread are shown to have escalated dramatically upwards after liberalisation. An analysis of three macroeconomic variables, namely, the exchange rate, inflation rate and economic growth offer little, or inconclusive evidence, that they were the main causes of the wide interest rate spread. In fact, the spread is closely linked to institutional/structural factors such as non-competitiveness in the banking sector, imprudent lending practices and poor and/or inadequate banking supervision. Policies for improving the institutional infrastructure and thus moderating the spreads are highlighted.

Customers have become the lifeblood of any organisation. Without customers, or-ganisations would not exist let alone survive in this competitive global environment (Grigoroudis, Politis and Siskos, 2002). Banks depend on sufficient and sustaining profitability to survive in the global business world. Customers are the source of banks‟ profitability. By satisfying the customer, the bank is able to retain the custom-er and reap maximum benefits from the relationship which ultimately leads to higher profitability. Customer satisfaction has therefore evolved as a strategic business ini-tiative which banks cannot ignore. Retention of the bank customer has become one of the most important objectives of the overall marketing strategy of any bank. In Kenya, the term „customer service‟ came to the fore just over fifteen years ago when banks started acquiring customer service departments. Since then, many cus-tomer training programs for staff have been put in place to transform the image of the customer as not just a profit-maker for the banks but as a human being with needs, which if not fulfilled will cause the customer to look for alternatives in the market. Training has focused on the bank staff whose customer handling skills have been sharpened. In spite of this, no empirical study has attempted to find out if the intended satisfaction of the customer has been achieved or not, which is indicative of little or no attention being given to this important phenomenon. In Africa, with the ex-ception of South Africa, empirical studies on customer satisfaction in the banking in-dustry are few. This gap presented the motivation for this study. The primary objective was to establish the determinants of customer satisfaction and retention in the Kenyan banking industry. The secondary objectives were to establish the relationship between socio-economic factors and customer satisfaction in Ken-yan banks; secondly, to determine whether bank-related factors influence customer satisfaction in Kenyan banks; thirdly, to identify the various strategies known to cus-tomers and employed by Kenyan banks to ensure customer satisfaction and customer retention and finally, to analyse the relationship between customer satisfaction and customer retention in Kenyan banks. The study adopted a descriptive survey design to suit the target population which was dispersed over a wide geographical region spanning the entire Kenya. The tar-get population included every bank customer in Kenya. Both qualitative and quantita-tive data were used. The data collection instrument was a self-administered ques-tionnaire that contained both closed and open-ended questions. Statistical tests were done using Pearson, Chi Square, Anova, Pearson Correlation and Multi-linear re-gression. Data were presented using frequency distribution tables, percentages, cross tabulation and pie charts. The findings indicated a positive relationship be-tween bank-related factors and customer satisfaction and retention. The conclusion was that if banks improved on factors like quality service, staff orientation towards customers, availability of management and ATM uptimes just to name a few, propor-tionately, customer satisfaction and retention would be enhanced. Finally, recommendations based on the findings were made to the Kenyan banks highlighting antecedents which would enhance the customers‟ satisfaction and reten-tion in the Kenyan banking industry.

Abstract in English, Afrikaans and Setswana / The study aimed to develop an appropriate cross-border banking framework for competitiveness in East Africa. To this end, Kenya Commercial Bank was chosen as case study, as it allowed for an examination of global, multi-domestic and transnational cross-border competitiveness strategies. The political and sociocultural environments were the moderating variables for the study, while the sub-variables of competitive advantage were efficiency, risk management, learning and innovation. A mixed methods sequential approach was utilised, with the quantitative approach employing a cross-sectional survey research design as framework, while probability and nonprobability sampling techniques were used for collecting quantitative and qualitative data respectively. Both close- and open-ended questions were used. The quantitative phase of the study sampled 217 potential participants and received 168 responses, which is a response rate of 77 per cent. Data were analysed using SPSS software, which provided descriptive and inferential statistics. Thereafter, a framework for cross-border banking was developed, using regression analysis. The qualitative phase of the research employed a case study design, with interviews being conducted with employees of Kenya Commercial Bank. To this end, purposive, convenience and snowball sampling was utilised. In addition, qualitative data were processed and analysed through the use of MAXQDA software. Trustworthiness and rigour were enhanced by transcribing the interviews and reviewing them for accuracy. The quantitative and qualitative data were subsequently synthesised, taking into account their points of convergence and divergence. The study finally established that the most appropriate framework for competitiveness was to strategically integrate multi-domestic and global strategies, and moderate these by taking into account prevailing political and sociocultural environments. Elements for global configuration included centralisation, standard procedures, similar policies, organisational structures, global dynamics and global appeal. By contrast, the critical elements for a multi-domestic strategy included decentralisation, different procedures/policies/organisational structures and marketing approaches, flexibility, local responsiveness, as well as local dynamics and local appeal. / Hierdie studie is daarop gemik om ’n gepaste oorgrens-bankweseraamwerk vir mededingendheid in Oos-Afrika te ontwikkel. Met hierdie doel voor oë is die Kenya Commercial Bank as gevallestudie gekies, omdat dit die navorsers in staat gestel het om wêreldwye, multibinnelandse en transnasionale oorgrens-mededingendheidstrategieë te ondersoek. Die politieke en sosiokulturele omgewings is die moderatorveranderlikes vir die studie, terwyl doeltreffendheid, risikobestuur, leer en innovering die subveranderlikes van mededingingsvoordeel is. ’n Opeenvolgende gemengemetodes-benadering is gevolg – die kwantitatiewe benadering maak gebruik van ’n deursnee-opname-navorsingsontwerp as raamwerk, terwyl waarskynlikheid- en nie-waarskynlikheid-steekproefnemingstegnieke gebruik is om onderskeidelik kwantitatiewe en kwalitatiewe data in te samel. Geslote vrae én oop vrae is gebruik. Tydens die kwantitatiewe fase van die studie is 217 potensiële deelnemers genader en 168 het geantwoord. Die responsiekoers was dus 77 persent. Data is ontleed met behulp van SPSS-sagteware, wat deskriptiewe en inferensiële statistiek opgelewer het. Daarná is daar met behulp van regressie-ontleding ’n raamwerk vir oorgrens-bankwese ontwikkel. Tydens die kwalitatiewe fase van die navorsing is gebruik gemaak van ’n gevallestudie-ontwerp – onderhoude is gevoer met werknemers van Kenya Commercial Bank. Hiervoor is doelgerigte, gerieflikheids- en toenemende steekproefneming gebruik. Daarby is kwalitatiewe data verwerk en ontleed met behulp van MAXQDA-sagteware. Betroubaarheid en nougesetheid is bevorder deur die onderhoude te transkribeer en te hersien vir akkuraatheid. Die kwantitatiewe en kwalitatiewe data is daarná saamgevoeg met inagneming van die punte waarop die data ooreenstem en verskil. Uiteindelik het die studie bepaal dat die mees gepaste raamerk vir mededinging sou wees om multibinnelandse en globale strategieë te integreer en te modereer met inagneming van heersende politieke en sosiokulturele omgewings. Die elemente van wêreldwye konfigurasie sluit in sentralisering; standaardprosedures; ooreenstemmende beleide, organisasiestrukture en globale dinamiek; en globale aantrekkingskrag. Daarenteen sluit die noodsaaklike elemente vir ’n multibinnelandse strategie die volgende in: desentralisering; verskillende prosedures, beleide, organisasiestrukture en bemarkingsbenaderings; buigsaamheid; plaaslike responsiwiteit; plaaslike dinamiek; en plaaslike aantrekkingskrag / Maikaelelo a thutopatlisiso e ne e le go tlhama letlhomeso le le maleba la go ralala melelwane la kgaisano kwa Aforikabotlhaba. Go fitlhelela seno, go tlhophilwe Banka ya Kgwebo ya Kenya go sekasekwa, ka ntlha ya fa e ne e kgontsha tshekatsheko ya ditogamaano tsa kgaisano tsa selegae-bontsi le ditlamo tse di kgabaganyang ditšhaba le melelwane. Ditikologo tsa sepolotiki le loagosetso e nnile dipharologantsho tse di dirisitsweng mo thutopatlisisong, fa dipharologantshopotlana tsa tshiamelo ya kgaisano e nnile bokgoni, tsamaiso ya matshosetsi, go ithuta le go itlhamela. Go dirisitswe mokgwa o o tswakantsweng o o sekasekang data go tswa kwa tshimologong, fa mokgwa wa go lekanyetsa dipalopalo ka go dirisa thadiso ya patlisiso e e ralalang maphata e nnile letlhomeso, mme go dirisitswe thekeniki e bannileseabe ba kokoanngwang mo thulaganyong e e nayang batho botlhe ditshono tse di lekanang tsa go ka tlhophiwa le e e sa neyeng batho botlhe ditshono tse di lekanang tsa go ka tlhophiwa go kokoanya data e e ka kgonang go lekanyediwa le e e ka se keng ya kgona go lekanyediwa. Go dirisitswe dipotso tse di batlang karabo ya ee kgotsa nnyaa le tse di batlang tlhaloso. Legato la patlisiso le le akaretsang dipalopalo le tlhophile batho ba ba ka nnang le seabe ba le 217 mme ga amogelwa tsibogo ya ba le 168, se e leng tsibogo e e lekanang le diperesente tse 77. Go sekasekilwe data go dirisiwa serweboleta sa SPSS, se se tlametseng ka dipalopalo tse di tlhalosang le tse di lebisang kwa ditshwetsong. Morago ga moo go ne ga tlhamiwa letlhomeso la banka e e dirang go kgabaganya melelwane, go dirisiwa tshekatsheko e e lekanyetsang kamano ya dipharologantsho. Legato la go tlhaloganya mabaka la patlisiso le dirisitse mokgwa wa go lebelela rekoto ya setheo se se rileng, mme go botsoloditswe badiri ba Banka ya Kgwebo ya Kenya. Go fitlhelela seno, go dirisitswe mokgwa wa go tlhopha sampole ka go tlhopha, go dira gore bannileseabe ba ngoke bannileseabe ba isago le go tlhophiwa ga bannileseabe ba ba fa gaufi. Go tlaleletsa foo, data e e sa kgoneng go lekanyediwa e sekasekilwe ka tiriso ya serweboleta sa MAXQDA. Go okeditswe boikanyego le kelotlhoko ka go gatisa dipotsolotso le go di sekaseka gape go tlhomamisa nepagalo. Data e e kgonang go lekanyediwa le e e sa kgoneng go lekanyediwa di ne tsa kopanngwa, go lebeletswe dintlha tsa tsona tse di golaganang le tse di fapaaneng. Gape thutopatlisiso e lemogile gore letlhomeso le le maleba go gaisa la kgaisano ke gore go nne le togamaano ya go golaganya ditogamaano tsa selegae le tsa lefatshe, mme di sekasekiwe go etswe tlhoko tikologo e e gona ya sepolotiki le loagosetso. Dintlha tsa thulaganyo e e akaretsang di akareditse go tlisa ditirelo fa lefelong le le lengwe fela, ditsamaiso tse di tlwaelegileng, dipholisi tse di tshwanang, dipopegotheo tsa setheo, diphetogo mo lefatsheng mmogo le kgatlhego mo lefatsheng. Go farologana le seo, dintlha tsa botlhokwa tsa togamaano ya selegae-bontsi e akareditse go tlisa ditirelo fa lefelong le le lengwe fela, dithulaganyo/dipholisi/dipopegotheo tsa setheo tse di farologaneng le mekgwa ya papatso, go obega, tsibogo ya selegae, gammogo le diphetogo tsa selegae le kgatlhego ya selegae. / Graduate School of Business Leadership / D.B.L

In 1993, Kenya liberalised its trade policy and allowed the Kenyan Shillings to freely float. This openness has left Kenya's domestic prices vulnerable to the effects of exchange rate fluctuations. One of the objectives of the Central Bank of Kenya is to maintain inflation levels at sustainable levels. Thus it has become necessary to determine the influence that exchange rate changes have on domestic prices given that one of the major determinants of inflation is exchange rate movements. For this reason, this thesis examines the magnitude and speed of exchange rate pass-through (ERPT) to domestic prices in Kenya. In addition, it takes into account the direction and size of changes in the exchange rates to determine whether the exchange rate fluctuations are symmetric or asymmetric. The thesis uses quarterly data ranging from 1993:Ql - 2008:Q4 as it takes into account the period when the process of liberalization occurred. The empirical estimation was done in two stages. The first stage was estimated using the Johansen (1991) and (1995) co integration techniques and a vector error correction model (VECM). The second stage entailed estimating the impulse response and variance decomposition functions as well as conducting block exogeneity Wald tests. In determining the asymmetric aspect of the analysis, the study followed Pollard and Coughlin (2004) and Webber (2000) frameworks in analysing asymmetry with respect to appreciation and depreciation and large and small changes in the exchange rate to import prices. The results obtained showed that ERPT to Kenya is incomplete but relatively low at about 36 percent in the long run. In terms of asymmetry, the results showed that ERPT is found to be higher in periods of appreciation than depreciation. This is in support of market share and binding quantity constraints theory. In relation to size changes, the results show that size changes have no significant impact on ERPT in Kenya.