I never liked being on a different side of an issue from the chief actuary, who I think does a very good job and whose reports I rely on for a lot of my work. I wish to show appropriate respect to the chief actuary and to the office of the chief actuary.
The disagreement between us is about the appropriate discount rate to use. In the public sector it is typical to use these discount rates that are relatively high, and the reference tends to be to historical returns on assets. I do not think that is an appropriate way to think about the future. Anyone who buys an investment has a warning and it's obligatory. The regulators oblige this warning that past performance is no guarantee of future results. The same is true in saving generally for retirement. We have seen returns that were much higher in the past, that is true. We've seen by now a long period of low returns. It's a mug's game to guess what kinds of returns there will be in the future. People who have a strong view on that can buy a bond, and it may go up or it may go down if they were right or wrong. It's their money. It's their choice to make.
In the case of a pension plan like this, there's a very clear alternative point of reference. I've said it already so I won't belabour the point. This is an unconditional guarantee of the federal government. It is like other federal government debt. When we do our evaluations we look at the chief actuary's work. We are not arguing with his assumptions about employment, inflation, the rate of increase of federal employees' salaries, none of that. All of the disagreement is about the discount rate. We use the sensitivity analysis that the chief actuary provides to try to make the adjustments. I think our adjustments are a bit conservative as we go all the way down to the real return bond rate. It's possible that if we had more access to the full range of data the chief actuary works with, our numbers would be worse, but we're conservative. The difference is in the discount rate, and when we try to adjust it to a lower discount rate, that's the number we get, the $96 billion.