ANALYSIS OF THE
BASIC TABLES WITH A FOCUS ON LIFE SETTLEMENTS
Vincent J. Granieri, FSA, MAAA, EA, MBA
The 2014 Valuation Basic Tables (VBT) are coming and they have already touched off a debate regarding the effect they will have on the life settlements market and valuations of existing policies/portfolios. Industry insiders harken back to 2008, when the release of the 2008 VBT coincided with life expectancy (LE) extensions by LE providers and they are preparing for the worst. Predictive Resources believes that the new tables should be a non-event in the life settlement industry as the VBT are not life settlement tables. Nonetheless, we are pleased to present a comparison of the preliminary 2014 VBT with both the 2008 VBT and Predictive Resources’ proprietary mortality curves for select age/gender/tobacco combinations.
Under a hypothetical scenario where policies are valued first using Predictive Resources’ mortality assumptions (which include both continuous improvement of its base tables and future mortality improvement assumptions) and then the 2014 VBT, the value will be higher using the 2014 VBT, all else equal. This means that investors who buy polices using Predictive Resources’ LEs will see an increase in value over what they paid if they were to value the policies using the 2014 VBT (or the 2008 VBT). When comparing values under the 2014 VBT, if moving from the 2008 VBT, values will fall at higher ages for all gender/tobacco use combinations. For lower aged males, values will be lower under the 2014 VBT than the 2008 VBT, unless the 2008 table had annual mortality improvement built into it. For lower aged females, values will not change much when moving from the 2008 VBT to the proposed 2014 VBT and will actually rise if the 2008 table had annual mortality improvement built in.
The preliminary 2014 VBT have been released for comment by the American Academy of Actuaries. These mortality tables form the basis for the Commissioners’ Standard Ordinary (CSO) tables, which historically were developed by adding extra mortality margins to the VBT. This meant that the CSO tables were more conservative when valuing life insurance policies from the perspective of the insurance commissioners who regulated insurance.
However, this approach is not conservative for valuing life settlements; in fact, it is aggressive because increasing mortality means assuming more people will die earlier. While it is true that the VBT is less aggressive than the CSO and it is not used directly for reserves, it was not built to reflect life settlement mortality. The recently implemented Actuarial Standard of Practice 48, which provides guidance for setting mortality assumptions for life settlements, states that “the actuary should select a base mortality table that is appropriate for the purpose of the assignment
a table that in the actuary’s professional judgment reflects the characteristics of the underlying population.”
Therefore, it is Predictive Resources’ belief that the VBT should not be used to value life settlements, at least in its unmodified state. The life settlements industry has enough data that it is possible to develop mortality tables that better reflect life settlement mortality. Nonetheless, some suggest that the adoption of the 2014 VBT (expected in early 2015) will lead to revaluation of life settlement portfolios based on the new table. If this is an issue for your portfolio, you may wish to contact us and set up some time to discuss the situation.
As a service to our clients and the industry, Predictive Resources is pleased to present an analysis of the proposed 2014 VBT. Please contact us with any questions you may have.
Let’s take a look at the differences in mortality among different mortality assumptions: the 2008 VBT, the preliminary 2014 VBT and Predictive Resources’ proprietary secondary life settlement mortality tables. We are adding one additional table; the 2008 VBT brought forward with six years of mortality improvements, using the same improvement percentages that Predictive Resources uses to continuously update both its base mortality tables and its future mortality assumptions. Predictive Resources was the first and still may be the only LE Provider to update both base tables and future mortality assumptions.
We will start with male non-tobacco users. The graphs below represent mortality rates for 65, 75 and 85 year olds.
The purple curve is the 2008 VBT. The 2014 VBT is in red and the Predictive Resources curve is in green. The turquoise curve is the 2008 VBT, brought forward to 2014 with mortality improvement. As higher mortality rates mean more people die earlier, the aggressive curves are upper/left most in these graphs and the conservative curves are lower/right most. All else equal, moving from an aggressive curve to a more conservative curve will cause the value of a policy to drop.
As you can see, the red 2014 VBT curve is generally more conservative than the vanilla 2008 VBT (purple), especially at early durations. However, except at the highest ages, if one would have assumed mortality improvement in the 2008 VBT, it would be more conservative than the preliminary 2014 VBT. It is also clear that the Predictive Resources’ curve is the most conservative. This means that an investor that buys their policies using a Predictive LE but is forced to value them using either the 2014 VBT or the 2008 VBT will see an increase in value, all else equal.
Moving on to male tobacco users, the results are similar.
We are somewhat surprised that the 2014 curve lacks the smoothness and consistency usually associated with mortality tables. The same conclusions seen in male non-tobacco users hold for male smokers – drops in value moving from the 2008 VBT as a valuation basis to the 2014 VBT, unless mortality improvement had been assumed. The Predictive curve again is the most conservative.
Next, consider female non-tobacco users.
Here, we have some different news to report. Except at high ages, the 2008 VBT is at least as conservative as the 2014 VBT and the 2008 VBT with assumed mortality improvement is actually more conservative. Therefore, moving from the mortality improved 2008 VBT to the 2014 VBT will cause values to increase. Again, the Predictive curve is the most conservative.
Finally, we consider female tobacco users.
This group also shows many similarities to female non-tobacco users. Again, the 2014 table seem more erratic than the others, which will certainly draw comments from actuaries in the exposure period.
Predictive Resources is prepared to discuss this topic in more detail with interested parties. It has developed proprietary software to project portfolio or policy level cash flows under any number of mortality assumptions, including the proposed 2014 VBT.