Why CAPM is useless

Why CAPM is useless.

Just a thought on using CAPM method to calculate WACC.

I was speaking to one of the Portfolio Managers yesterday and thought up of something peculiar.

Dividend discount model, simplest form from which DCF/FCF all the other discounting models came about, has one special disadvantage - it assumes constant growth.

Now, before going into antics about how it's easily fixed and that's what abnormal earnings is about, think about this - It also assumes constant k. This isn't unique to dividend model. All discounting models assume constant cost of money - whether it'd be WACC, k, etc.

Now, logic says that cost of money changes all the time. Cost of Equity depends on the equity market movement, the perceived risk (however you define it) and cost of debt depends on interest rate movement, as the treasury notes depend on that. At the least, these change every year, more like every quarter.

For example, if you were going to put money into project which will provide constant cashflow for next 10 years, not only do we have to discount them at WACC, we need to account for the changes in the rate of WACC. This is paramountly important, because even in countries like US, interest rates change by huge amount. You might argue that equity and fixed income markets balance each other out, but there are still fluctuation. And as we all know, discounting models are heavily sensitive to the discount.

This is why we were told to approximate that cost of money in equity market was about 10%. (long term average). This is yet another reason why we need margin of safety. This is why using CAPM to calculate WACC exactly is total bullshit.