This thesis examines the economic consequences of the adoption of international pension accounting standard IAS19 Revised (IAS19R) on pension asset allocation decisions by applying a difference-in-differences with propensity score matching method. The publication of IAS19R in 2011 marked a fundamental change to pension reporting in financial statements. In particular, it had a significant impact on (1) how sponsor firms recognise net pension assets/liabilities on the balance sheet, (2) the calculation and recognition of pension expenses, (3) the presentation of re-measurement (actuarial gains and losses), treatment of which had been heavily debated by academics and practitioners, and (4) disclosure requirements for pension schemes, which had been criticised as “excessive” under IAS19. This research examines the “real effect” of IAS19R adoption on management investment decisions. Using a difference-in-differences with propensity score matching method, the results suggest that, on average, UK sponsor firms affected by IAS19R have reduced their risk taking in pension investments post-IAS19R, both over time and compared with a control sample of unaffected US firms (matched by propensity score matching). The results of sensitivity analysis also suggest that UK sponsor firms tried to avoid the expensive liquidity costs of asset re-allocation by switching their pension plan asset allocations gradually during the period around the publication and adoption of IAS19R. Furthermore, the outcomes of sensitivity tests suggest a positive relationship between equity investment levels, and firms’ leverage and cash flow risk, consistent with the “risk-shifting” hypothesis documented in the previous literature. The thesis also applies a manual textual analysis on the comment letters sent by industrial firms to the IASB to provide their opinions on the IAS19R Exposure Draft. The analysis describes and tabulates the arguments raised by these firms on three main amendment areas of IAS19: recognition, presentation and disclosure. Based on this description, this part aims to motivate the empirical research mentioned previously and shed light on the other potential consequences of IAS19R adoption. These consequences include: the management of funding might be driven by accounting rules rather than management rules; the increasing volatility of balance sheet; de-risking in the pension plan portfolio following the adoption of IAS19R; the diminishing of financial statement “true and fair view” and its usefulness due to the abolition of expected rate of return and excessive requirements on pension disclosure. Furthermore, the study also suggests that the lobbying behaviour of these firms on the standard setting process is consistent with the predictions of Positive Accounting Theory.

Este trabalho teve por objetivo geral realizar uma análise de como contabilizar obrigações, com benefícios previdenciários de risco, em regimes de repartição simples, que não preveem formação de reserva ou métodos pré-definidos para isso. Para a consecução de tal objetivo, escolheu-se o benefício de pensão por morte (pensão militar) das Forças Armadascomo objeto de estudo. Nesse sentido, o uso de uma base de dados real contendo as informações individualizadas, pode ser apontado com um dos diferenciais desta pesquisa.Considerando-se o grau de confiabilidade dessa base de dados, além do ineditismo do seu uso em pesquisas acadêmicas, foi possível inferir que os resultados obtidos contribuíram para ratificar diversos conceitos expostos na plataforma teórica. A trajetória metodológica desta pesquisa foi desenvolvida com base no emprego de métodos de custeio atuarial a fim de realizar a projeção de receitas e despesas do sistema de pensões militares. Tais métodos, além da repartição simples, estão alinhados com as abordagens Accumulated Benefit Obligation(ABO) e Projected Benefit Obligation(PBO). Dessa forma, por meio da comparação entre os resultados oriundos das mencionadas abordagens, buscou-se identificar os possíveis impactos nas contas da União, enquanto ente empregador. O confronto de resultados foi realizado com base em uma análise comparativa cujo método principal para o cálculo atuarial foi o fluxo projetado, bem como o fundo financeiro decorrente desse fluxo. Com base nos valores obtidos, diante das diferenças conceituais das abordagens estudadas, entendeu-se que uma forma adequada para permitir a comparação dos resultados seria o cálculo do fundo financeiro futuro a cada período de tempo (t). A análise dos resultados indicou a ocorrência de diferenças representativas entre as abordagens. Na verdade, constatou-se que por não considerar o aumento salarial, o modelo ABO previu menores valores de obrigações com os benefícios futuros, portanto, os valores presentes do fundo financeiro foram maiores do que os modelos que consideraram aumento salarial.Nessa perspectiva, sob o enfoque do ente empregador, os resultados pareceram mais otimistas.Por fim, a abordagem PBO apresentou resultados que pareceram mais aderentes `a realidade do sistema, aproximando-se relativamente do resultado da repartição, no cenário com taxa de juros de 3% (menor taxa utilizada nas projeções). Considerando-se o tempo total das projeções, foi possível observar que, para os modelos de repartição e PBO, existiu um passivo a ser registrado nas contas da União. No entanto, o modelo ABO indicou o contrário, sugerindo que há uma ativo a ser contabilizado no Balanço Geral da União. Tal situação indicou que há diferenças concretas nos resultados, que podem levar os usuários da informação contábil a tomar decisões equivocadas em decorrência dessa distorção de valores. Ficou evidenciado que quanto maior for o horizonte de tempo projetado, maiores serão as diferenças. Ou seja, no menor horizonte adotado (25 anos) constatou-se as menores diferenças. Nesse horizonte, todos os modelos apresentaram valores de que sugerem a existência de um ativo a ser contabilizado. À luz de tal constatação parece razoável sugerir como adequada a revisão do tempo de projeção hoje empregado no âmbito da União (75 anos) na tentativa de reduzir a parcela de incerteza embutida nesse horizonte de longo prazo. / The main objective of this study was to analyze how to account for obligations with risk pension benefits in simple distribution systems that do not provide for reserve formation or predefined methods for this. In order to achieve this objective, the death benefit (military pension) benefit of the Armed Forces was chosen as the object of study. In this sense, the use of a real database containing the individualized information can be pointed out with one of the differentials of this research. Considering the degree of reliability of this database, in addition to the novelty of its use in academic research, it was possible to infer that the results obtained contributed to ratify several concepts exposed in the theoretical platform. The methodological trajectory of this research was developed based on the use of actuarial costing methods in order to realize the projection of revenues and expenses of the military pension system. Such methods, in addition to simple partitioning, are in line with the Accumulated Benefit Obligation (ABO) and Projected Benefit Obligation (PBO) approaches. Thus, through a comparison of the results from the mentioned approaches, it was sought to identify the possible impacts on the Union accounts, as an employer. The comparison of results was performed based on a comparative analysis whose main method for the actuarial calculation was the projected flow, as well as the financial fund resulting from this flow. Based on the values obtained, in view of the conceptual differences of the approaches studied, it was understood that an adequate way to allow the comparison of the results would be the calculation of the future financial fund at each time period (t). The analysis of the results indicated the occurrence of representative differences between the approaches. In fact, it was found that the ABO model predicted lower bond values with future benefits, therefore, the present values of the financial fund were higher than the models that considered a salary increase. From this perspective, under the focus of the employer, the results seemed more optimistic. Finally, the PBO approach presented results that seemed to be more consistent with the reality of the system, relatively close to the result of the distribution, in the scenario with an interest rate of 3 % (lower rate used in the projections). Considering the total projection time, it was possible to observe that, for the allocation models and PBOs, there was a liability to be recorded in the Union accounts. However, the ABO model indicated the opposite, suggesting that there is an asset to be recorded in the Federal Government Balance Sheet. This situation indicated that there are concrete differences in results, which may lead the users of the accounting information to make mistaken decisions as a result of this distortion of values. It has been shown that the larger the projected time horizon, the greater the differences. That is, in the lowest adopted horizon (25 years) the smallest differences were observed. Within this horizon, all models presented values that suggest the existence of an asset to be accounted for. In light of this, it seems reasonable to suggest as appropriate the review of the projection time currently used in the Union (75 years) in an attempt to reduce the uncertainty embedded in this long-term horizon.

Este trabalho teve por objetivo geral realizar uma análise de como contabilizar obrigações, com benefícios previdenciários de risco, em regimes de repartição simples, que não preveem formação de reserva ou métodos pré-definidos para isso. Para a consecução de tal objetivo, escolheu-se o benefício de pensão por morte (pensão militar) das Forças Armadascomo objeto de estudo. Nesse sentido, o uso de uma base de dados real contendo as informações individualizadas, pode ser apontado com um dos diferenciais desta pesquisa.Considerando-se o grau de confiabilidade dessa base de dados, além do ineditismo do seu uso em pesquisas acadêmicas, foi possível inferir que os resultados obtidos contribuíram para ratificar diversos conceitos expostos na plataforma teórica. A trajetória metodológica desta pesquisa foi desenvolvida com base no emprego de métodos de custeio atuarial a fim de realizar a projeção de receitas e despesas do sistema de pensões militares. Tais métodos, além da repartição simples, estão alinhados com as abordagens Accumulated Benefit Obligation(ABO) e Projected Benefit Obligation(PBO). Dessa forma, por meio da comparação entre os resultados oriundos das mencionadas abordagens, buscou-se identificar os possíveis impactos nas contas da União, enquanto ente empregador. O confronto de resultados foi realizado com base em uma análise comparativa cujo método principal para o cálculo atuarial foi o fluxo projetado, bem como o fundo financeiro decorrente desse fluxo. Com base nos valores obtidos, diante das diferenças conceituais das abordagens estudadas, entendeu-se que uma forma adequada para permitir a comparação dos resultados seria o cálculo do fundo financeiro futuro a cada período de tempo (t). A análise dos resultados indicou a ocorrência de diferenças representativas entre as abordagens. Na verdade, constatou-se que por não considerar o aumento salarial, o modelo ABO previu menores valores de obrigações com os benefícios futuros, portanto, os valores presentes do fundo financeiro foram maiores do que os modelos que consideraram aumento salarial.Nessa perspectiva, sob o enfoque do ente empregador, os resultados pareceram mais otimistas.Por fim, a abordagem PBO apresentou resultados que pareceram mais aderentes `a realidade do sistema, aproximando-se relativamente do resultado da repartição, no cenário com taxa de juros de 3% (menor taxa utilizada nas projeções). Considerando-se o tempo total das projeções, foi possível observar que, para os modelos de repartição e PBO, existiu um passivo a ser registrado nas contas da União. No entanto, o modelo ABO indicou o contrário, sugerindo que há uma ativo a ser contabilizado no Balanço Geral da União. Tal situação indicou que há diferenças concretas nos resultados, que podem levar os usuários da informação contábil a tomar decisões equivocadas em decorrência dessa distorção de valores. Ficou evidenciado que quanto maior for o horizonte de tempo projetado, maiores serão as diferenças. Ou seja, no menor horizonte adotado (25 anos) constatou-se as menores diferenças. Nesse horizonte, todos os modelos apresentaram valores de que sugerem a existência de um ativo a ser contabilizado. À luz de tal constatação parece razoável sugerir como adequada a revisão do tempo de projeção hoje empregado no âmbito da União (75 anos) na tentativa de reduzir a parcela de incerteza embutida nesse horizonte de longo prazo. / The main objective of this study was to analyze how to account for obligations with risk pension benefits in simple distribution systems that do not provide for reserve formation or predefined methods for this. In order to achieve this objective, the death benefit (military pension) benefit of the Armed Forces was chosen as the object of study. In this sense, the use of a real database containing the individualized information can be pointed out with one of the differentials of this research. Considering the degree of reliability of this database, in addition to the novelty of its use in academic research, it was possible to infer that the results obtained contributed to ratify several concepts exposed in the theoretical platform. The methodological trajectory of this research was developed based on the use of actuarial costing methods in order to realize the projection of revenues and expenses of the military pension system. Such methods, in addition to simple partitioning, are in line with the Accumulated Benefit Obligation (ABO) and Projected Benefit Obligation (PBO) approaches. Thus, through a comparison of the results from the mentioned approaches, it was sought to identify the possible impacts on the Union accounts, as an employer. The comparison of results was performed based on a comparative analysis whose main method for the actuarial calculation was the projected flow, as well as the financial fund resulting from this flow. Based on the values obtained, in view of the conceptual differences of the approaches studied, it was understood that an adequate way to allow the comparison of the results would be the calculation of the future financial fund at each time period (t). The analysis of the results indicated the occurrence of representative differences between the approaches. In fact, it was found that the ABO model predicted lower bond values with future benefits, therefore, the present values of the financial fund were higher than the models that considered a salary increase. From this perspective, under the focus of the employer, the results seemed more optimistic. Finally, the PBO approach presented results that seemed to be more consistent with the reality of the system, relatively close to the result of the distribution, in the scenario with an interest rate of 3 % (lower rate used in the projections). Considering the total projection time, it was possible to observe that, for the allocation models and PBOs, there was a liability to be recorded in the Union accounts. However, the ABO model indicated the opposite, suggesting that there is an asset to be recorded in the Federal Government Balance Sheet. This situation indicated that there are concrete differences in results, which may lead the users of the accounting information to make mistaken decisions as a result of this distortion of values. It has been shown that the larger the projected time horizon, the greater the differences. That is, in the lowest adopted horizon (25 years) the smallest differences were observed. Within this horizon, all models presented values that suggest the existence of an asset to be accounted for. In light of this, it seems reasonable to suggest as appropriate the review of the projection time currently used in the Union (75 years) in an attempt to reduce the uncertainty embedded in this long-term horizon.

本文討論退休金長期預期報酬率之設定。根據會計公報規定，公司之退休金長期預期報酬率必須符合其資產配置。但是在本文中我們發現資產配置並無法有效的反應公司的退休金長期預期報酬率，因此我們提出展望理論及盈餘管理兩種理論來解釋退休金長期預期報酬率之設定。我們發現經理人企圖透過改變退休金長期預期報酬率的假設來進行盈餘管理，同時退休金長期預期報酬率也會因為公司的風險態度而有所改變。 / Abstract: We try to find out the considerations for managers to set their assumptions of expected long term rate of return on pension plan assets (ROPA). First, we use the asset allocations of pension funds and historical returns to calculate the expected rate of return based on historical asset returns (EROPA). There is difference between ROPA are EROPA, suggests that asset allocations are not the only consideration when managers setting their ROPA assumption. Two theories are examined in this paper to explain such difference between ROPA and EROPA: earnings management and prospect theory. We use two models to test the earnings management, single accrual model and threshold model. We find that the intentions to smooth the reported income are the main incentives for managers to manipulation their ROPA. The incentive to do earnings management can partly explain the difference between ROPA and EROPA. However, in threshold model, we can not observe any evidence in our research. We also introduce prospect theory to examine the risk attitude. We find that managers’ risk attitude affect the setting of assumptions, too. We conclude prospect theory provides a good explanation of the difference between ROPA and EROPA.

The evidence of this research paper suggests an increased value-relevance of OCI in the Swedish stock market after the 2011 amendment of IAS 19. This amendment eliminated the popular corridor approach that ‘smoothed’ the recognition of actuarial gains and losses and required companies instead to recognize such remeasurements directly to their full extent in OCI. We reason that the current economic environment marked by low interest rates together with demographic trends in life expectancy have amplified the impact of the standard revision on financial statements. Moreover, this research paper is conducted in a Swedish setting that is characterized by widespread occupational pension schemes. In line with informational accounting research, we use regression models to identify whether OCI has more explanatory power for stock market prices and returns. Shares listed on NASDAQ OMX Stockholm are investigated for this purpose. Furthermore, in the course of this research, we revisit the controversial debate on comprehensive income reporting and work out inconsistencies in current IFRS reporting. / Beviset för den forskningen  föreslår att ökade värde-relevans av Övrigt totalresultat i den svenska aktiemarknaden efter 2011 p.g.a. ändringen av IAS 19. Ändringen eliminerade populära korridormetoden att "jämnas" erkännande av ackumulerade vinster och förluster och krävde företag istället erkänna sådana omvärderingar direkt i full utsträckning i övrigt totalresultat. Vi resonerar nuvarande ekonomiska läge som präglas av låga räntor tillsammans med den demografiska utvecklingen i den förväntade livslängden har förstärkt effekterna av standard ändring  i  finansiella rapporter. Dessutom är denna uppsats genomförts på Svensk arbetsmarknad som kännetecknas av omfattande tjänstepensionssystem. I linje med informations redovisning forskning använder vi regressionsmodeller för att identifiera om “Övrigt totalresultat”  har mer förklaring för börskurser och avkastning. Aktier noterade på NASDAQ OMX Stockholm undersöks för detta ändamål. Dessutom under den forskning, titta  vi närmare den kontroversiella debatten om totalresultat rapportering och gå igenom inkonsekvenser i nuvarande IFRS-redovisning.

Abstract Authors:Stefan Harrå and Magnus Björk Advisor: Markku Penttinen Title: Accounting Consequences of the change in pension accounting Background to problem: When the revised IAS 19 comes into force January 1, 2013, it means that two of the three accounting principles for defined benefit pension plans are disappearing, including the corridor method. The corridor method has made it possible for companies to defer its actuarial gains and losses. Now that the corridor approach abolished then the unrecognized actuarial gains and losses immediately be covered by equity, which involves very large amounts of some companies. Why the amounts have grown so big is much because of the discount rate. The discount rate is a controversial parameter, and there is disagreement on how it should be fixed. Purpose: The purpose of this thesis is to examine the accounting implications this will have for the company applied the corridor method, and if there is some parameters in the actuarial assumption that is more important than others. Methodology: The thesis has mainly been based on a qualitative research through qualitative interviews with a small sample that is affected by this change. There are quantitative elements to a greater depth by examining the annual reports, discount and deferred pension liabilities of the various companies. The approach is exploratory as it is a qualitative study and there was little knowledge of the subject before the work of it started. Therefore, a study of literature, regulations and previous research before the empirical study. This made it possible to gain a broader understanding of the subject and to shape relevant and essential interview questions. Conclusions: The conclusion shows that the largest accounting consequences for the companies in the study in conjunction with the change is that the unrecognized actuarial gains and losses will now be covered by equity and that the expected return on plan assets is based on the discount rate. The study also shows that it is the discount rate which is considered the most important parameter that the companies are looking at in the actuarial assumption. The conclusion also provides a shared sense of the true and fair picture of the companies after the revised IAS 19. Suggestions for further research: That after 2013 to study how the actual result of this rule change did this compare to the expected. Look at the problem of determining the discount rate. How will the IASB look at it if more and more begin to deviate from the standard? Keywords: "IAS 19", "IAS 19 revised", "corridor method", "pension accounting", "pension liabilities", "defined contribution plans", "actuarial assumptions", "actuarial gains and losses" and "discount rate".

Inledning: Svenska kommuners verksamhet betecknas som säregen. Till detta hör att kommunerna skiljer sig från den privata sektorn i avseende på målsättning och utförande. En utav de särdrag som finns inom kommunerna är deras redovisning av pensionerna som skall ske enligt blandmodellen. Ett flertal studier har lyft upp farhågor om att redovisningen enligt blandmodellen anses leda till diskretionära tolkningsutrymmen som troligtvis kommer utnyttjas av politikerna i syfte att påverka redovisningen. En del av de diskretionära åtgärder som antas vara möjliga är pensionsförvaltning av de förmånsbestämda pensionerna. Syfte: Studiens syfte är att förklara kommunernas val av förvaltning av de förmånsbestämda pensionerna. Metod: En deduktiv ansats har använts i studien för att kartlägga de olika teoretiska förhållandena mellan pensionsförvaltning och pensionsredovisning. Studiens utförande grundar sig i att tillföra en ny inblick i redovisningsforskningen. För att åstadkomma detta har tolv hypoteser sammanställts och prövats utifrån den statistiska analysen. Slutsatser: Studien har visat att variabeln lönsamhet har varit signifikant inom samtliga analyser. Detta har lett till att hypotesen om resultatutjämning har använts som en förklaring till kommunernas val av pensionsförvaltning. Variablerna befolkningsstorlek och befolkningsförändring har även de visat sig påverka valet av pensionsförvaltningen. I den logistiska regressionen har det dock varit problematiskt att införa samtliga variabler i samma modell. Därför har hypoteserna om befolkningsstorlek och befolkningsförändring förkastats tillsvidare. Hypoteserna som utformades har stundom växlat mellan positiv redovisningsteori och institutionell teori för att på så vis ge en mer utförligt förklaring. De anses därför ha bidragit var på sitt sätt till att förklara valet av pensionsförvaltningen. / Introduction: Swedish municipalities’ operations are described as being distinctive. This means that the municipalities differ from the private sector in terms of goal settings and performance. One of the characteristics found in the municipalities is their reporting of pensions that is to be accounted for according to the mixed model. Several studies have brought forward concerns that the accounting according to the mixed model may give rise to discretionary interpretations that assumedly are going to be exploited by the politicians in order to affect the financial reporting. A part of the discretion that is used in the earnings management is assumed to occur in the pension management of the defined benefit pension plans. Aim: The aim of the study is to explain the municipalities’ choice of management of the defined benefit pension plans. Method: A deductive approach has been used in the study to identify different theoretical relationships between the pension management and pension accounting. The study's design was based on providing a new insight into accounting research. To achieve this, twelve hypotheses were formulated and later tested using the statistical analysis. Conclusions: The study has shown that the variable profitability has been significant in the overall analyzes. This has led to the hypothesis that income smoothing could be used as to explain the municipalities’ choice of pension management. The variables population size and population change have also been shown to affect the choice of pension management. Doing the logistical regression with all the variables was however problematic. Therefore, the hypotheses on population size and population change have been rejected for the time being. The hypotheses that were designed have been interchangeable at different points between positive accounting theory and institutional theory, which has led to a more detailed explanation. They are therefore considered to have contributed each in its own way to explain the choice of pension management.