The popularity of KiwiSaver, combined with a demographic shift to an ageing population, will make savings invested in KiwiSaver an important source of income in retirement. To investigate the investment fund choices by KiwiSaver members a survey was circulated among 134 people; 87 KiwiSaver members, and 47 non-members. Respondents answered questions on their investment behavior, knowledge, and risk attitudes towards investment in general, and KiwiSaver. The results show investment in KiwiSaver tends to be conservative as a result of low levels of involvement, knowledge, and risk tolerance, and is more common among females. Investment in riskier growth funds is mostly by younger people and those who are risk tolerant as measured by the KiwiSaver Risk Profile.

Dans ce mémoire, j’étudie trois sujets de microéconomie : le ciblage, les jeux de hasard et l’épargne-retraite. Le premier chapitre, basé sur un travail conjoint avec Francis Bloch, examine une approche du ciblage basée sur le classement par les amis. Un planificateur social a pour objectif d’extraire un classement des individus en demandant à chaque individu de classer ses amis. Nous étudions comment la structure du réseau social de la communauté détermine la capacité du planificateur à extraire un classement véridique et efficace. Notre analyse souligne l’importance des amis communs. Le deuxième chapitre, basé sur un travail conjoint avec Joshua Blumenstock, étudie les paris sportifs en ligne, qui ont gagné en popularité au Kenya et dans d’autres pays d’Afrique orientale. Nous utilisons un vaste ensemble de données sur les transactions de paris sportifs pour prouver que les joueurs peuvent devenir trop confiants dans leur capacité à prédire le résultat des matchs. Nous soutenons que l’excès de confiance persiste parce que les joueurs surestiment la fréquence à laquelle leurs amis gagnent et parce que les joueurs surpondèrent les succès passés par rapport aux échecs passés. Dans le troisième et dernier chapitre, j’utilise une expérience de terrain pour étudier l’impact de la fourniture d’une calculatrice de retraite sur l’épargne-retraite à des employés d’une grande entreprise sud-africaine. La calculatrice fournit des projections de revenu de retraite et aide à contrer les perceptions erronées de la croissance exponentielle. Même si la plupart des employés épargnent toujours au taux minimum et que l’intervention a été organisé pour coïncider avec les augmentations de salaire, le calculateur n’a pas entraîné d’importantes augmentations de l’épargne. / In this dissertation, I study three topics in microeconomics — targeting, gambling and retirement saving. The first chapter, based on joint work with Francis Bloch, studies the friend-based ranking approach to targeting. A social planner aims to extract a ranking of individuals by asking each individual to rank his or her friends. We study how the structure of the community’s social network determines the planner’s ability to extract a truthful and efficient ranking. Our analysis highlights the importance of common friends. The second chapter, based on joint work with Joshua Blumenstock, studies online sports betting, which has gained widespread popularity in Kenya and other East African countries. We use a large dataset of sports betting transactions to provide evidence that gamblers can become overconfident in their ability to predict the outcome of matches. We argue the overconfidence persists because gamblers overestimate how frequently their friends win and because gamblers overweight past success relative to past failure. In the third and final chapter, I use a field experiment to study the impact of providing employees at a large South African company with a retirement calculator on retirement saving. The calculator provides projections of retirement income and helps to counter misperceptions of exponential growth. Even though most employees save at the minimum rate and the intervention was timed to coincide with salary increases, the calculator did not lead to large increases in saving.

Syfte: Denna studie syftar till att undersöka vilka variabler som påverkar olika frågor gällande pension. De frågorna angående pensionen som undersöks är:Vilka människor… … har tänk på sin pension? … har ett privat pensionssparande? … tänker jobba och ta ut pension samtidigt? … planerar att gå i pension tidigt respektive sent? Metod: Undersökningen utförs med en kvantitativ metod. Enkätundersökningen som analyseras kommer från en tidigare undersökning gjord av Finseraas och Jakobsson (2013). Uppsatsen följer en induktiv metod. Slutsats: Studien visar att variablerna som ingick i grundmodellen, vilka är kön, ålder, utbildningsnivå, civilstånd och inkomst, alla var signifikanta i en eller fler av frågorna som undersöks. Men även en del andra variabler visade sig vara av betydelse. / Objective: This study aims to examine which variables have an effect on questions regarding retirement. The questions are as follows: Which individuals… …have thought about their retirement? …are saving privately for their retirement? …plan to work at the same time as they receive pension? …are planning for early or late retirement? Methodology: The study is conducted using a quantitative method. The data analyzed comes from an earlier survey conducted by Finseraas and Jakobsson (2013). The essay follows an inductive method. Conclusion: The study shows that the variables included in the basic model, which are gender, age, education, marital status and income, all were significant in one or more of the questions under investigation. Some other variables were found to be of significance.

Using data from 584 individuals identifying themselves as married, the purpose of this study was to examine how personal and relational characteristics were linked to financial attitudes, knowledge, and capabilities and financial well-being using the family financial socialization framework (Gudmunson & Danes, 2011). Supporting the first two hypotheses, marital quality, materialism, age, and household income were found to directly predict financial prudence as a measure of financial attitudes, knowledge, and capabilities and to indirectly predict retirement savings rate as a measure of financial well-being. Financial prudence supported the first hypotheses as well by directly predicting retirement savings rate. Education also supported the first hypothesis, in that it directly predicted an individual's measure of financial prudence. In support of the third hypothesis, education was associated with retirement savings rate. Results suggest the importance of considering both financial and non-financial predictors of saving for retirement.

Retirement saving research frequently has investigated the differences between working men and working women and primarily focused on the near retirement and retirement years. There is limited research targeting young to old working-age women including those who do not work for pay and are unemployed. The purpose of this study was to examine what factors affect non-retired working-age (25 years and older) women's retirement saving behaviors, retirement savings, and retirement income satisfaction. To implement the study, a research framework was developed based on Deacon and Firebaugh's Family Resource Management Model. The research framework for this study consisted of three major sections: (a) input (demographics, saving motives, retirement saving involvement level, retirement information seeking, current financial assets and debts, and future expectations), (b) throughput (retirement saving behaviors such as calculating needed retirement savings, being a retirement saver, starting saving for retirement age, and being a regular retirement saver), and (c) output (the objective retirement savings and the subjective retirement income satisfaction). An online survey instrument was developed to obtain data for the study. Two pilot tests were conducted to confirm the validity and reliability of the instrument. Data for this study were collected from a national population between May 25, 2012 and May 30, 2012 with 591 valid responses. Several statistical methods were employed: descriptive statistics, one-way between-groups analysis of variance (ANOVA), direct logistic regression, and standard multiple regression. From the results of the study, only about one-third of the women (31.8%) reported they expect to get the full amount of Social Security retirement income that today's retirees get. However, around 60% of the women only save less than $25,000 or none in employer-provided retirement accounts or in personal investments and savings. There is an un-addressed gap between the cognition of the need to save for retirement and real saving action. A regular retirement saver is more likely to save more in employer-provided retirement accounts and to feel more satisfied with that retirement income. Yet, regular retirement savers have less savings in personal investments and savings, possibly because they believe their work investments will be sufficient or some women may make direct deposits to meet the annual limits of retirement plans. Other researchers have not reported this relationship. Those women who are more cognitively involved with saving for retirement are more likely to calculate needed retirement savings and to be a retirement saver, but they are less satisfied with retirement income from Social Security and from personal investments and savings. Satisfaction level is subjective; thus, those who expect to own more types of assets in retirement may have a higher satisfaction level with the expected income from both employer-sponsored retirement accounts and personal investments and savings. Generally, greater satisfaction with expected retirement income is associated with higher accumulation in retirement savings, and the female savers have much more retirement savings than non-savers. However, there is no difference in the retirement income satisfaction of savers and non-savers. These findings have implications for financial educators, counselors and advisors, researchers, employers, and policy makers. There are recommendations for women and future research. / Ph. D.

[pt] O termo imunização denota a construção de uma carteira de títulos de forma a torná-la imune a variações nas taxas de juros. No caso das entidades de previdência complementar, o objetivo da imunização é distribuir os recebimentos intermediários e finais dos ativos de acordo com o fluxo de pagamentos dos benefícios. Em geral, quanto maior a classe de alterações na estrutura a termo das taxas de juros (ETTJ), mais restritivo se torna o modelo. Embora exista uma vasta literatura sobre o aspecto estatístico e sobre o significado econômico dos modelos de imunização, esse trabalho inova ao prover uma análise detalhada do desempenho comparado dos modelos, sob três perspectivas complementares: o método escolhido, a dimensionalidade e, ainda, o horizonte de investimento. Entretanto, a decisão final do gestor não está restrita à escolha do método de imunização, como também ao horizonte de investimento a ser imunizado, uma vez que outros instrumentos financeiros podem garantir tanto a solvência econômica quanto a financeira. Os limites não operacionais à imunização são analisados por meio da comparação das medianas do relativo de riqueza e da probabilidade de exaustão da carteira. A análise permite concluir que os modelos de imunização tradicional são mais eficientes, especialmente no médio e longo prazo, que os modelos multidimensionais de gestão do risco de taxa de juros. Ademais, demonstra-se que não existem limites naturais à imunização, quando aplicada ao mercado previdenciário brasileiro por um período igual ou inferior a 10 anos. / [en] Immunization is defined as the investment in assets in such a way that the fixed income portfolio is immune to a change in interest rates. In the special case of pension funds, immunization seeks the distribution of the cash inflows in accordance with the outflows represented by the fund´s liabilities. The article compares distinct alternative methods of immunization against the traditional duration-matching strategy. All portfolios were obtained as a result of mathematical programming problems, where the choice of the immunization strategy led to the restrictions imposed on the evolution of the term structure of interest rates. Despite the intensive research related to this subject, there are some gaps to be filled yet, especially those concerned with the investment horizon. That is exactly the main objective of this thesis. The work provides the basis for selecting the most appropriate method for immunization and also demonstrates the superiority of the traditional duration-matching strategy, especially in the medium and long run. Moreover, it is demonstrated that there is no limit other than operational to the immunization process concerning Brazilian markets for investment horizons of less than 10 years.

Research on financial decision aids, systems designed to help people make financial decisions, is sparse. Previous research has often focused on the theoretical soundness of the advice that the systems provide.The original contribution of this doctoral thesis is a set of empirical studies of how non-expert people understand the advice provided by financial decision aids. Since every piece of advice must be interpreted by a receiver, the accuracy of the advice can be corrupted along the way if the receiver does not understand complicated terminology, probabilistic reasoning, or abstract concepts.The design concept resulting from the studies visualizes a unique combination of short-term and long-term variables that are usually treated as separate and not interacting with each other; loans and amortizations, insurance, retirement saving, and consumption. The aim is to visualize the consequences of different decisions and possible adverse events in terms of their effect on the user’s future consumption, rather than abstract numbers detached from the user’s lived experience.The design concept was tested and evaluated by personal finance experts and professional financial advisors, as well as students and people without financial education, who represented the target users of the system. Results indicate that the system has a learning curve, but that once users understand how to read the graph, they find it more informative than conventional financial planning tools. / <p>QC 20170809</p>

[pt] O propósito deste trabalho é analisar o impacto da tributação na seleção de produtos de acumulação de recursos de longo prazo sob o ponto de vista de um investidor individual. Para o investidor tributado e avesso a risco é realizada uma simulação de seu patrimônio final líquido de impostos sob três estratégias distintas: aplicação em fundo de investimento, aplicação em plano de previdência PGBL e aplicação em seguro de vida com cobertura por sobrevivência VGBL. Além da hipótese de resgate total dos recursos, é realizada a comparação dos veículos sob a ótica do recebimento de uma renda mensal após o período de acumulação. São examinados diferentes horizontes de investimento, saldos acumulados iniciais, faixas de renda, taxas de retorno e tipos de fundos onde estarão aplicados os recursos. O trabalho incorpora as alterações das regras tributárias divulgadas no final de 2004 e início de 2005 (Leis 11.033/2004 e 11.053/2004 e suas regulamentações) e se propõe a auxiliar os investidores na escolha do melhor veículo de investimento. / [en] The purpose of this research is to analyze the impact of taxation on the selection of long term wealth accumulation products under an individual investor's point of view. For the tax paying and risk averse investor, a simulation of his final after-tax net worth is accomplished under three different strategies: investments in mutual funds, investments in retirement plans similar to Individual Retirement Account (IRA) in the USA and investments in life insurance with coverage for survival. Besides the hypothesis of total redemption of the accumulated balance, a comparison of the three vehicles above is accomplished under the assumption of reception of a monthly income after the accumulation period. The analysis is performed under different investment horizons, initial invested balances, levels of income and profiles of funds where the resources will be invested. The research incorporates the new tax rules alterations published in late 2004 and early 2005 (Laws 11.033/2004 and 11.053/2004 and their regulations) and it intends to aid investors on how to choose the best investment vehicle.

[pt] Nesta tese, propomos quatro modelos dinâmicos para ajudar as seguradoras e fundos de pensão a medir e gerencias seus fatores de risco e seus planos de anuidade. Nos primeiros dois ensaios, propomos modelos de previsão de ganhos de longevidade de uma população, que é um importante fator de risco. No primeiro artigo, um modelo de séries temporais multivariado usando a abordagem SUTSE (seemingly unrelated time series equation) é proposto para prever ganhos de longevidade e taxas de mortalidade. No segundo artigo, um modelo estrutural multivariado com tendências estocásticas comuns é proposto para prever os ganhos de longevidade de uma população com uma curta série temporal de taxas de mortalidade, usando as informações de uma população relacionada, para qual uma longa série temporal de taxas de mortalidade é disponível. No terceiro artigo, outro importante fator de risco é modelado – taxas de cancelamento. Apresentamos um modelo estocástico multiestágio para previsão das taxas de cancelamento usando simulação de Monte Carlo depois de uma sequência de ajustes GLM, ARMA-GARCH e cópula multivariada ser executada. No quarto artigo, assumindo a necessidade de se avaliar as opções embutidas para manter a solvência dos planos de anuidade, propomos um modelo para mensuração das opções embutidas nos planos unit-linkeds brasileiros. / [en] In this thesis we propose four dynamic models to help life insurers and pension plans to measure and manage their risk factors and annuity plans. In the first two essays, we propose models to forecast longevity gains of a population, which is an important risk factor. In the first paper, a multivariate time series model using the seemingly unrelated time series equation (SUTSE) framework is proposed to forecast longevity gains and mortality rates. In the second paper, a multivariate structural time series model with common stochastic trends is proposed to forecast longevity gains of a population with a short time series of observed mortality rates, using the information of a related population for which longer mortality time series exist. In the third paper, another important risk factor is modeled – surrender rates. We propose a multi-stage stochastic model to forecast them using Monte Carlo simulation after a sequence of GLM, ARMA-GARCH and multivariate copula fitting is executed. Assuming the importance of the embedded options valuation to maintain the solvency of annuity plans, in the fourth paper we propose a model for evaluating the value of embedded options in the Brazilian unit-linked plans.

[pt] Uma importante questão que se coloca para entidades abertas de previdência complementar e sociedades seguradoras que operam previdência complementar é a definição de uma gestão dos ativos e passivos (do inglês ALM – Asset and Liability Management). Tal questão se torna mais relevante em um cenário de alta competitividade, margens operacionais decrescentes, garantias mínimas de rentabilidade para um passivo estocástico de longo prazo e um período de queda da rentabilidade dos instrumentos financeiros, sendo estes muitas vezes de difícil precificação e pouco previsíveis num mercado volátil como o brasileiro. Somada a estas dificuldades, as companhias deste mercado estão sujeitas a uma regulação baseada em riscos, oriunda de práticas internacionais, adotada pelo órgão superior, Susep, que impõe restrições regulamentares para a manutenção da solvência das companhias, o que eleva a dificuldade da definição de um modelo. Diante deste cenário, esta dissertação apresenta uma proposta de ALM baseada em um modelo de programação estocástica multiestágio que tem como objetivo definir dinamicamente a alocação ótima dos ativos, incluindo títulos com pagamentos de cupons, e mensurar o risco de insolvência da companhia para o horizonte de planejamento. / [en] An important issue of open pension funds and insurance companies that operate supplementary pension is the definition of an asset and liability management (ALM) framework. Such a question becomes more relevant in a scenario of high competition, declining operating margins, minimum guaranteed returns to a stochastic long-term liability and a period of falling returns on financial instruments, these being often difficult to pricing and predictable in a volatile market such as Brazil. Added to these issues, those companies are subject to a risk-based regulation, derived from international practices adopted by the national insurance regulator, Susep, which imposes constraints to maintain solvency of companies and therefore increases the complexity of an ALM framework. Due this condition, this dissertation presents a proposition of ALM based on a multistage stochastic programming model, which aims to define a dynamic optimal asset allocation, including bonds with coupons payment, and measure the company s insolvency risk for the planning horizon.