On Thu, 24 Jul 2008 14:13:36 -0400, alexy wrote:
> Jeff Dege
jdege.visi.com> wrote:
> >>Still, it gives a lower bound. Assuming a fund that starts out at $1/
>>share, investing $1/year, over the 18 years, assuming all gains are
>>expressed as price increases, and there were no dividends (hence no
>>reinvestment of dividents) yields (second column is number of shares
>>purchased, the third is total number of shares, the fourth is the total
>>valuation of those shares):
>>
>>1965 1.000 1.000 1.1245
>> [...]
>>1982 0.352 12.06 41.528
>>
>>Over the 18 years, you've invested $18, and end up with $41. My TVM
>>calculator says that's a 9.29%% return.
> Might want to check to see if that TVM calculator is still under
> warranty.
Yep. My mistake. I entered 41.58 for future value, instead of 41.528.
Payments of $-1.00, 18 payments, 1 payment/year, present value $0, future
value $41.52 = annual interest rate of 9.05%%. Don't know where I screwed
up, before. (I'm using an HP 49G+ calculator).
>> If the inflation numbers are
>>correct, $1 -> $3.06, then the average rate of inflation was 6.41%%.
> Correct.
>>
>>You didn't provide the breakdown in your data between dividend returns
>>and price appreciation, so I can't do much there. And I don't have the
>>year-by-year inflation data on-hand. But if I assume a constant 6.41%%
>>inflation rate over the entire period, I get:
>>
>>1965 1.000 1.000 1.1245
>>1966 0.946 1.946 1.9684
>>1967 1.119 3.065 3.8442
>>1968 0.960 4.026 5.6076
>>1969 0.920 4.947 6.3039
>>1970 1.070 6.018 7.9758
>>1971 1.095 7.113 10.777
>>1972 1.019 8.133 14.660
>>1973 0.911 9.045 13.914
>>1974 1.137 10.18 11.517
>>1975 1.645 11.82 18.354
>>1976 1.276 13.10 25.183
>>1977 1.096 14.20 25.331
>>1978 1.257 15.45 29.382
>>1979 1.255 16.71 37.627
>>1980 1.127 17.84 53.189
>>1981 0.906 18.74 53.149
>>1982 1.014 19.76 68.018
>>
>>That's a 13.95%%, more than 7%% above inflation.
> Step back for a moment. Does that look right to you?
Yes, actually they do. We're increasing our investment every year to
keep pace with inflation. The price/share vaguely tracks inflation. At
the end of the 18 years inflation and the price appreciation almost
exactly matched, so we should have generally been buying around one share
every year, and very close to one share in 1982. And that's what we
see. The third column is just a running total, which is clearly
correct. And the fourth column is just the price times the running total.
> If not, it's reason
> to check the numbers. I find that conclusion not credible. Don't have
> time to check the numbers myself right now, but will later. Provide an
> update if you reconsider.
I don't think you'll find anything wrong with the numbers. If there's a
problem, it's with the assumptions. But as I said, my expectation is
that the assumptions would tend to cause the results to be understated.
>>Dollar-cost averaging works.
> Yes, as a way to ameliorate the effects of swings and to counter
> peak-to-valley measurements made by the doom-and-gloomers. But without
> wild swings or a curve with a large positive second derivative, I find
> this kind of difference unlikely.
>
> But I'll look at the numbers tonight, and be back to eat crow if I'm
> wrong.
I'd be very much interested to know if I'm wrong.
--
if 2 + 2 == 5 then 5 == 4
