Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal. / In Canada taxpayers have an opportunity to save on their taxes through a program entitled the Registered Retirement Saving Plan, one of the most important tax deferred savings vehicles offered. RRSPs were first introduced in Canada in 1957. The contribution limits were increased substantially in the early 1970s, and RRSPs were widely promoted. Since then, they have become a prominent form of Canadian saving. RRSP contributions now exceed the total of employee and employer contributions to employer-provided pension plans (Revenue Canada, Tax Statistics on Individuals, Edition 1998). The RRSP option is especially attractive considering that the Canada Pension Plan, historically responsible for providing retirement benefits to all Canadians over age 65 when they retire, is in a precarious situation due to demographic changes. These factors would have us believe that a large number of Canadians take advantage of the RRSP program to save on their taxes, as well as to assure their old ages with a sufficient level of wealth. In fact, only one-third of Canadian families ^A4^o had access to the RRSP option contributed to such a plan in 1997 (Globe and Mail, March 1998, Calculations using data from Statistics Canada). u This paper describes a model for predicting the outcome of the RRSP decision making process for households. In this paper, we will explore a two-equation model for estimating households' behavior toward the decision to contribute in a RRSP and the amount of contribution. Our approach will integrate two groups of characteristics of households : one being demographic with age, gender, family size, location, education, and marital status and the other being financial with income, pension, debt and non liquid assets. We expect our empirical results to be consistent with previous researches based on the Individual Retirement Account (IRA), the American alter ego of the Canadian Registered Retirement Saving Plan (RRSP). 0 The main purpose of this paper is to develop a model that demonstrates a framework of Canadian households' behavior regarding their decisions to contribute to their own retirement wealth. It is still very early to draw conclusions about the households decisions and behaviors because the government is stilt very present as a source of retirement income. However, it is helpful to observe contemporary trends in order to determine the means of improving the information given to the general population regarding RRSPs.

This thesis examines stabilizing mechanisms in social security retirement systems (“SSRS”), especially those purporting to be automatic balancing mechanisms (“ABM”). It develops a consistent approach to identifying whether an ABM is robust, partial or transitory and establishes a terminology to classify balancing mechanisms. Both financial and equitable balances are considered in assessing whether an ABM achieves balance. Families of definitions of equity are presented and a benchmark by which to measure equitable balance is defined and applied. The balancing mechanisms of Canada, Germany, Japan and Sweden are described, evaluated and classified. None of these mechanisms are found to be robust. This thesis provides a critical analysis of an approach referred to as integration to financial markets and the approach is found to be deficient. In analyzing the Swedish SSRS a critical error in the way assets are calculated is identified and a suitable correction is proposed. A further weakness in the application of the Swedish ABM is identified that means that once an imbalance occurs, balance is unlikely to be restored. The thesis also discusses some of the unusual characteristics of the steady-state contribution rate calculation for the Canadian SSRS and shows that although it has limited application and does not appear to depend on any actuarial principle, the steady-state contribution rate calculation creates a tension between the near and distant future, which is a factor in achieving financial balance over a seventy-five year horizon. With respect to the balancing mechanism in the Canadian SSRS, the thesis proposes a change in how the mechanism is defined so that the mechanism would be robust, within certain ranges.

This thesis examines stabilizing mechanisms in social security retirement systems (“SSRS”), especially those purporting to be automatic balancing mechanisms (“ABM”). It develops a consistent approach to identifying whether an ABM is robust, partial or transitory and establishes a terminology to classify balancing mechanisms. Both financial and equitable balances are considered in assessing whether an ABM achieves balance. Families of definitions of equity are presented and a benchmark by which to measure equitable balance is defined and applied. The balancing mechanisms of Canada, Germany, Japan and Sweden are described, evaluated and classified. None of these mechanisms are found to be robust. This thesis provides a critical analysis of an approach referred to as integration to financial markets and the approach is found to be deficient. In analyzing the Swedish SSRS a critical error in the way assets are calculated is identified and a suitable correction is proposed. A further weakness in the application of the Swedish ABM is identified that means that once an imbalance occurs, balance is unlikely to be restored. The thesis also discusses some of the unusual characteristics of the steady-state contribution rate calculation for the Canadian SSRS and shows that although it has limited application and does not appear to depend on any actuarial principle, the steady-state contribution rate calculation creates a tension between the near and distant future, which is a factor in achieving financial balance over a seventy-five year horizon. With respect to the balancing mechanism in the Canadian SSRS, the thesis proposes a change in how the mechanism is defined so that the mechanism would be robust, within certain ranges.