Well, since I do such a rough approximation, I though I'd start at a large rate and move to a slower one over a fixed number of years (linearly decreasing) and then hold steady. Just to get some rough ideas of how numbers play out. Going further, doing a possibly longer, smoother decline, computing birth rates with a weighted averaging over all age groups, counting productive vs. unproductive aged and youth in a weighted manner by age rather than by sharp cutoffs, etc. etc. could all be done. However, I wanted to slap something together in an afternoon, without consulting too many statistical sources that would take a lot of time. I though a strong initial rate with a rather fast decline was a good way to show the idea that maybe this kind of demographic development is not as detrimental as often assumed.