Hi there,
I am pretty new to the software so excuse me if this is not making sense.
I think it would make more intuitive sense to a to model interest/dividends on a portfolio differently from the way the online version of the tool currently does it. I think you should model the yield on dividends, state tax free interest, fed tax free interest, LTCG distributions and completely tax free interest separately from growth in the value of the portfolio. So I might make an assumption that the dividend yield on taxable assets was 1.5%and then also input an assumption about principal growth. So (in simple terms without considering timing of cash flows and asset appreciation) if I input an annual 1.5% dividend yield plus 4.5% asset appreciation (comprised of inflation input of 3% plus real appreciation of 1.5%) (as I might for a taxable account composed principally of equities) then at the end of the year, if I did not add to or take money out of that account I would have whatever I had at the beginning of the year plus 6% (1.5%+4.5%).
The issue that I have with the way you currently model it is that If I make a change in the assumption of the real growth rate I also have to make a change in the assumption of the percentages attributable to the various forms of taxation in order to correctly estimate the actual dividend cash flow that I receive. My actual interest/dividends received will not change much if stock go up by 10% or down by 15%. I realize this might make it appear to make it more difficult to model a taxable portfolio as one big blob using the simplified method but would certainly solve my issue of having to back into my estimated dividend cash flow whenever I make a change in my estimate of real portfolio growth. I hope that makes sense. I also realize it would have a big impact on current users who are used to the way it is now. I also have not fully thought through the implications on model the tax effects on various interest/dividend/LTCG streams.
If I'm understanding what you want, this is already a proposed change that you can vote for (under More > Resources > Feature Voting), titled "Base interest and dividend income on % of balance".
Pralana's approach is natural if you assume that the P/E multiple roughly remains constant over time and that there is a constant dividend payout as a percent of profits. Recently, the US stock market has diverged from that with a few fast growing companies with high P/Es and low to no dividends dominating market indices. Hard to believe that can continue forever, though.
