No. of Recommendations: 16
Do you guys remember the Intrinsivaluator? It was a flash-based valuation tool...
That's the input to this model : )
Indeed, that's why it's all nominal prices, not inflation adjusted, as I built it originally straight from his figures.
Way back when, I used the average of the scenarios for each year to get an idea of the nominal rise in value. That set the initial assumed nominal growth rate, 11.711%/year at end 2008, and a starting notional value of $135731 at that date.
Each year's on-trend value-per-share growth rate after that is based on the assumption that growth rate is made up of two parts: (1) "average US stock in an average year" which I assumed to be 8% nominal (could be thought of as Siegel's constant 6.5% + 1.5% inflation, or 6+2), plus (2) a premium for management being smart. The premium for each year was assumed to be 6.187% less than it was the prior year (can't exactly remember why the strange decimals), so the rate of increase of nominal value per year asymptotically approaches 8%. The reasoning is that any monkey with a dartboard could manage inflation+6.5%, being an estimate of the average total return from the average US stock in the average year, so there is no reason to expect Berkshire's rate of value growth per share to fall lower than that. The idea of 6.5% was approximated in the old model as nominal 8%.
That gives a series of "on trend" value figures. Then I add noise to that for variation in value growth per year, a random number in the range =-7.5% to +7.5% to get notional value at that year end. Then a sort of bell curve of valuation multiple on that, usually ranging from 50% undervalued to 30% overvalued, and with a bit of persistence so trends arising from the random numbers can last more than a year.
Converting the whole model to inflation adjusted figures is possible, but then it is a brand new model, and has no history of out-of-sample validity. So one might as well start with a blank sheet of paper and build a new after-inflation model from scratch. I've done that too, but the result so far has always been the same: on trend growth best fit of inflation + 8%/year since 1998 and, as yet, no slowdown in that trend. I've been waiting for the visible slowdown for a long time. We all know it's coming, but it's amazing that it has held off this long.
However, if we posit that the on-trend growth rate is still inflation+8% at the moment but slowdown starts now, at the slowdown rate in the original model, the *after inflation* on-trend growth rates per share would look like this:
2025 Inflation + 8.00%
2026 7.91%
2027 7.82%
2028 7.74%
2029 7.66%
2030 7.59%
2031 7.52%
2032 7.46%
2033 7.40%
2034 7.34%
2035 7.29%
2036 7.24%
2037 7.20%
2038 7.15%
2039 7.11%
2040 7.08%
2041 7.04%
2042 7.01%
2043 6.98%
2044 6.95%
2045 6.92%
Jim
