Ideally, the correct rate would be the effective rate over the next 20 years. I think using the current rate as a proxy for the next 20 years is fraught with errors. In periods of high rates, certain investments would be precluded due to the above average rate; while in periods of low rates, undeserving companies would potentially qualify as a buy.

Long-term I suspect the rate will average around 5%, but I currently have no empirical evidence to support this. The available historical rates are "national averages", which are substantially lower than the actual rates I have personally experienced.

Given all the above, I have opted to build my own average going forward. Since 2007 ended with a rate of 5.11%, which is close to estimated long-term rate, I will use it as my first data point in building a long-term average. To minimize variability in the early years, I will limit swings in the rate to 0.5% (50 basis points) from the prior year's calculated average. Thus, applying this methodology will define a 2008 range of 4.61% to 5.61%.

__Related Articles:__