Microfinance Institutions (MFIs) play important roles in socio-economic development and poverty alleviation particularly in developing countries. It has however been argued that the focus of MFIs is changing from the traditional purely social to commercial (mission drift) and has been criticised for neglecting the welfare of citizens and grassroots accountability in favour of commercialisation and accountability to donors/shareholders. This mission drift has resulted in changes in the structure and practices of MFIs. The study has been designed to examine how the accounting and accountability practices of a MFI can change in response to changes in its mission. The study presents case evidence from a large MFI operating in Cameroon with data collected through semi-structured interviews, informal discussions and documents. The study traces the evolution of the organisation and its accounting and accountability practices. A theoretical framework of an interpretive nature is used which draws on institutional entrepreneurship theory in order to highlight the importance of actors in the change process. The findings suggest a mission drift and transformations over the years from a social to a commercial organisation with the change impacting significantly on its structure and accounting and accountability practices.

Korea adopted Regulation Fair Disclosure (FD) in November 2002. Regulation FD, designed with a goal of levelling the playing field among market participants, has created considerable debate among practitioners and academics. This thesis examines the effect of Regulation FD on the Korean securities market, using a large sample of 161,343 forecast-year observations and 2,311 firm-year observations from 2000 to 2007. We uncover four main sets of findings. First, we find that analysts' forecast accuracy has increased after the adoption of Regulation FD. We attribute this finding to the improved quality of public information and reduced importance of private access to managers in the post-FD period. Second, we provide evidence of significant change in firms' disclosure policy in the post-FD period. We report that private earning guidance and private information in analysts' forecasts have decreased as a consequence of curtailing selective disclosure in the post-FD period. Our findings are consistent with the intentions of Regulation FD to increase management disclosure to the general public. Third, we find no evidence of an increase in herding behaviour in the post-FD period. Our results contradict Regulation FD's opponents' claims that elimination of private channels may lead to increasing herding behaviour due to the chilling effect. We find no evidence that Regulation FD makes firms withhold their disclosure. To the contrary, our evidence suggests that Regulation FD has led to an increase in the quality and quantity of public information. Finally, we provide strong evidence for a reduction in informed trading and information leakage prior to unscheduled earnings announcement and release of analysts' recommendations. Overall, our results suggest that Regulation FD has been successful in eliminating selective disclosure and levelling the playing field for investors.

To induce tax compliance, two opposite approaches are used: the coercive and the persuasive: firm action versus collaboration. Little attention has been paid to the comparative success of these two approaches – a situation this thesis seeks to remedy by investigating the effectiveness of three coercive and three persuasive instruments among large corporate taxpayers registered with the LTU of Bangladesh. Following an analysis of survey data using binary and multilevel logit and CHAID models, and an analysis of elite interviews employing an interpretivist approach, the findings suggest that coercion or persuasion are less likely to improve tax compliance when used separately than when used in combination, although coercion seems the more powerful of the two. Factors underlying the power of the coercive approach are the rationality and regularity of its application, along with its legal and financial imperatives. Reasons contributing to the appeal of the persuasive approach are a reduction in tax compliance costs, an improvement in accountability and a reduction in knowledge gaps between taxpayers and tax administrators, and coordination of the various tax laws.

In 1999, Korea adopted the Preliminary Feasibility Study (PFS) to enhance fiscal efficiency by preventing unsuitable projects from being allocated a budget in the budget selection (BS) process. The PFS uses both economic and non-economic assessment criteria. However, according to data collected since the introduction of the PFS, about 30 per cent of projects not recommended for budget allocation received a budget. This raises questions about the relationship between the PFS and budget allocation. According to previous research, three issues about the effect of the PFS on budgetary decisions caused concern: the effect of PFS results on budgetary decisions, the appropriateness of the current PFS methodology, and the potential non-neutral behaviour of stakeholders taking part in the PFS. To deal with these, the thesis uses three key research questions: What is the impact of PFS results on budgetary decisions? Does the current PFS methodology lead to inappropriate decisions? And, lastly, what are the types of, extent of, and reasons for non-neutral behaviour by stakeholders in the PFS, and how might such behaviour be reduced? The study suggests that PFS results have a positive impact on budgetary decisions. The thesis also examines the appropriateness of the current PFS methodology and provides potential alternatives through both quantitative and qualitative analysis. Lastly, the thesis provides evidence of non-neutral behaviours by stakeholders in the PFS, which are categorized into four types: Promoter, Blocker, Dr. Pangloss, and Cassandra type. After this, this thesis examines the reasons for non-neutral behaviour and suggests possible policy alternatives for reducing non-neutral behaviour.

This thesis examines the strategy formulation of small and medium enterprises in Greece by measuring the two principal dimensions of the strategy formulation process, the normative/descriptive dimension and the individual/collective dimension. For the purposes of the thesis a theoretical model of strategy formulation is proposed and evaluated, drawing upon the principles of contingency theory which presumes that there is no exclusive approach to strategy formulation which applies likewise to all firms in all circumstances. According to contingency theory there are many variables or factors which predict and influence the decision of formulating a strategy. Furthermore, the thesis investigates the role of accounting information on the process of strategy formulation by examining the relationship between the adoption of a specific strategy formulation approach and the information sources that SMEs use, the extensiveness of accounting information usage and the perceived usefulness of accounting information to SME managers. Using a data-triangulation (semi-structured interviews and a questionnaire), the findings of the study show the following: First, organizational size, perceived environmental volatility, the level of technology and specific owner manager characteristics (experience and education) are all significant predictors of SMEs adoption of a specific strategy formulation approach. Second, accounting information usage is positively associated with the normative and collective strategy formulation approaches. More specific, SMEs which utilize a broad range of information sources, which engage in extensive accounting information utilization and which perceive accounting information as very useful are positively correlated with the normative and collective approaches of strategy formulation. Third, financial performance was found to have a positive relationship with the descriptive and individualistic approach to strategy, meaning that SME owner managers who employ emergent strategies without the collaboration and participation of others in the process tend to achieve higher levels of profitability. Fourth, significant differences were found between SMEs having a different ownership status (family and non-family SMEs) and belonging to different business sectors (retail, construction and service providing firms) in relation to strategy formulation approach.

This thesis investigates the factors that influence the magnitude of goodwill impairment losses as well as the value relevance of these losses using a sample of 2,466 companies, drawn from 17 countries in which IFRSs have been made mandatory for all their domestic listed companies. The study period is 2007-2013 and includes 14,898 firm-year observations. The results obtained from the Tobit regression analysis involving variables drawn from agency/positive accounting theory, Hofstede’s theory of culture, as well as different theoretical institutional models, reveal that goodwill-impairment amounts are not only driven by economic factors and managerial reporting incentives, but also by country-specific factors, such as cultural and institutional variables. The results also confirm that the strength of the equity market is still the single most influential factor contributing not only to differences in accounting practices but above all, to differences in institutional quality between countries. The results of a K-means cluster analysis reveal that there are two groups of countries, corresponding to strong equity-outsider and weaker equity-outsider clusters. By comparing the relative associations between goodwill-impairment amounts and economic factors and managerial reporting incentives across these two institutional clusters, estimation results reveal that firms in the strong equity-outsider cluster have recorded goodwill-impairment losses that are, on the one hand, strongly associated with economic factors, and on the other hand, weakly associated with managerial reporting incentives. Further analysis also showed that while results for the pooled sample did not indicate that goodwill impairment losses were value relevant this was not the case for firms in the strong equity-outsider cluster, which have recorded impairment losses that are, on average, more relevant and more timely than those recorded by firms in the weaker equity-outsider cluster.

This thesis is structured upon three studies. The first study investigates whether mandatory IFRS adoption improves accounting quality in Latin America. The findings show that in the post-adoption period: accrual earnings management practices are reduced, value relevance of accounting increases, and the delay in recognising bad news reduces. However, these improvements cannot be found in firms with high bankruptcy possibility and poorly performing firms. The second study focuses on whether the analysts’ information environment has improved since the IFRS adoption. The results show that the mandatory adoption of IFRS improves analysts’ information environment, even after controlling for the firm-level reporting incentives. The third study focuses on whether IFRS has affected the cost of equity and debt in Latin America. The findings show that the cost of equity and debt decreased significantly in the post-IFRS period. Overall, the results found can be attributed to IFRS as the institutional environment has not changed significantly around the years of the mandatory adoption of IFRS. Thus, IFRS can contribute to enhance the accounting quality of Latin American firms, and may help to develop the capital market and the development of these firms.

I was trading professionally in the years 2006-2014 in the equity derivatives market. This thesis deals with two of the ideas inspired by my experience as a professional trader. The first topic deals with the pricing of a derivatives product in the market with a specific risk concentration. We call the product that causes the concentration a market driver. When the market driver exists, not only the market driver itself, but any derivatives product will not be priced fairly. We introduced a new model based on the Heston model that accounts for the concentration. The model leads to a pair of partial differential equations (PDEs): one semilinear parabolic PDE to price the market driver and one linear parabolic PDE to price all the other products. In solving the semilinear PDE, we use the policy improvement algorithm (PIA) to approximate the solution with those of linear PDEs. We show that the approximated solutions satisfy quadratic local convergence (QLC) which explains the efficiency of the algorithm. This efficiency of the algorithm is proved in a more general setup. The other idea sparked by my experience that is explored in the last chapter of the thesis concerns modeling technical analysis. Technical analysis is a family of methods that traders use to make decisions to purchase/sell assets. There is no mathematical proof that shows that they are correct as far as I am aware. We focus on one of the methods, the method of support and resistance levels, and used the optimal stopping argument to show the validity of the method. As far as I know, this is one of the first results to mathematically prove the effectiveness of a method in technical analysis.

Behavioral studies show that people tend to use various decision heuristics which discard part of the available information, simplify the decision problem, and find a good enough answer. In addition, people’s decision and behavior may change because they learn from experience. This thesis investigates people’s heuristic decision making and learning from experience in two frequent real-world financial decision contexts—credit card repayments and stock trading. Chapter 1 reviews the literature about decision heuristics and nudges. Literature about learning from experience is also reviewed. Chapter 2 shows that automatic minimum credit card repayment as a default nudge has the adverse effect of reducing repayments by allowing card holders to neglect their monthly bill. Chapter 3 examines whether people learn from the negative feedback provided by credit card fees. We show that cardholders tend to adapt to late payment fees, which are typically due to forgetting a repayment, by setting up an automatic repayment. On the other hand, cash advance and over-limit fees are due to card holders’ liquidity needs rather than their mistakes, and thus, they do not learn from experiencing those fees. Our findings are contrast to those in a previous study in the US suggesting that people learn from all three types of fees. Chapter 4 shows evidence of people’s heuristic processing of numerical information in the context of credit card repayments. We find a strong tendency of card holders repaying at several prominent numbers. We also find people’s preference for round numbers. Conducting an online experiment, Chapter 5 confirms the anchoring effect of numerical information in a credit card bill, as in previous studies, and finds a false consensus bias where people who usually repay only the minimum greatly overestimate the commonness of minimum repayments among others. However, a social nudge phrase in a mock bill fails to correct the false belief, and thus, dose not reduce the likelihood of people repaying only the minimum. Chapter 6 presents a two-stage model of the choice of a stock to sell. Typically investors show a disposition effect, being more likely to sell a stock in gain than loss, other things equal. In our model, investors first decide whether to sell a stock in the domain of gains or losses, and only then, evaluate stocks within the chosen domain. As evidence for the model, our analysis shows that the likelihood of an individual stock being sold is inversely proportional to the number of stocks in the same domain in the portfolio but is not sensitive to the number of stocks in the other domain. Our findings indicate that existing estimation methods of the disposition effect result in substantial biases because those estimations assume that all stocks in a portfolio are simultaneously evaluated across domains of gains and losses. Chapter 7 summarizes the findings and implications of Chapters 2-6. Plans and suggestions for future research are also discussed.

The Agricultural Development Fund (ADF) is the Government agency responsible for financing farmers in the Saudi Arabian agricultural sector. The ADF provides short-term loans and medium-term loans and funds specialised projects, as well as providing subsidies for some types of loans. From its establishment in 1962 until 2012, the ADF has lent more than £7.7 billion and the total number of loans exceeds 439,000. However, in spite of the facilities provided by the ADF, its annual reports show that the outstanding debt increased from £579 million in 1990 to £840 million in 2012; moreover, the reports reveal that some project areas have been receiving more funding than others. This thesis aims to assess the efficiency of the collection of loan repayments by the ADF. This aim will be achieved through four objectives: 1) analyse the current status of the regional distribution of agricultural loans to determine the optimal distribution of loans among regions to achieve development objectives, 2) examine the determinants of the efficiency of loan collection to identify and quantify factors that affect loan repayment, 3) explain the variation in collection rates among the ADF's branches and 4) provide recommendations to improve the probability of repayment and achieve the optimal distribution of loans. Variation in the current distribution of agricultural loans among regions has been observed. Using Multiple-Attribute Analysis, this variation has been identified as sub-optimal with respect to achieving development objectives. The highest priority regions for agricultural loans are the Northern, Jazan and Baha regions. Logistic modelling approaches have been used to determine factors affecting the repayment of short-term loans, medium-term loans and specialised projects; the empirical results of the overall model for short-term loans show that the age of the borrower and four regions have a positive significant effect on loan repayments, while the grace period, the loan value, the financial guarantee and three regions each have a significant negative effect. The results of the analysis of medium-term loans reveal that five variables have a significant effect on loan repayments: age, the grace period and loan value have a negative effect. Financial guarantee and three regions have a positive effect. The results for the specialised projects showed that the Hail region, feed and wheat project have a positive effect, while the Eastern region and greenhouses, livestock and vegetables projects have a negative effect on loan repayments, whereas the age of the borrower, the grace period, financial guarantee and loan value have no significant effect.