Thursday, March 26, 2009

Some striking results from the comparison of mortality tables

One key variable for the calculation of annuities (see here for a previous note) and the rates of return of pensions funds needed to equalize an annuity with the pension under the scheme of programmed withdraws (see here for another note) is the mortality tables. The comparison of mortality tables can be very useful to understand variation across countries of retirement pensions. There is no single method for comparing mortality tables across countries, but the statistic A/E, that is equivalent to the ratio of the weighted average probabilities of death of two mortality tables, using a specific population structure for the weights, has proven very useful. The following graph show the statistic A/E (using Chile as the base case). For example, it struck us that the statistic A/E calculated using the regulated mortality table for each country have little relationship to socioeconomic conditions of the countries, a highly determinant of mortality. This is something that deserves further analysis.

No comments: