Ibbotson Chen model


Can someone please explain why are the variables used in Ibbotson chen model called supply-side variables? And if possible a comprehensive explanation for each variable i.e.

how does growth in EPS of a company define real GDP growth of a country?

Why is growth in P/E and growth in EPS included in it?

Do the following variables i.e. (1 + Einfl)(1+ ExpgrowthEPS)(1+ ExpGrowthP/E) -1] +EInc combine to form Rm from CAPM? If yes then how?



Ibbotson and Chen first analysed historical data between 1926 - 2000. They found geometric returns to be 10.70% over that period. They attempt to decompose this into more fundamental constituents (building blocks).

We know CPI affects our nominal returns, so they first ascertain CPI over this period, and find it is 3.08%. They then look to break returns into two components: capital gain component, and the income component (dividends). They adjust capital gains to changes in CPI, and find that real capital gains for the period = 3.02% and that income for the period was 4.28%.

They then break down the real capital gain component, specifically, they break the real capital gain component down to growth in real EPS, and growth in P/E — Now you have the CFA Institute’s presented model: [[1+CPI * 1+REPS * 1+PE - 1] + INC] - RF {this is why expected income is separated from CPI*REPS*PE, because the former component was broken down from capital in the model}.

So, how does the growth in real EPS and growth in P/E affect anything? Well, since developed economies are largely composed of companies, we would not expect real GDP to exceed the earnings of all the companies in the economy right? This is why real growth in EPS is a good proxy for real GDP. P/E ratios are useful for tracking future growth, ie, if a company has a higher P/E ratio, then this means future growth has been assigned to the company such that its PV is deemed to be greater (even if earnings at present do not support it). This is why technology stocks had (during the dotcom boom), and still have, a relatively high P/E ratio. Assuming an efficient market, we would not need to alter this metric (because all future growth has been correctly priced in). However, if we believe the market has over speculated on future growth, then present P/E ratios may be too high, and so we may have to adjust downwards (because we believe that prices will go down, ie, negative growth), and vice versa, the market may not have factored in tremendous future growth in some technological area, and so P/E ratios are temporarily depressed, and so we adjust upwards.

Because the model showed a supply-side estimate of 4.26% which was close to the geometric mean estimate of 4.20%, they therefore say that their model can be used as a forward-looking estimate.

Hopefully that clarifies the Ibbotsen Chen model.