Hi Moksh,

Point 1, Normal Environments Do Exist: I think it is entirely fair to think of a "normalized world" existing in financial markets. Let me share with you an example. Take a look at www.cfapubs.org and look for a game-changing and amazing article written by the late, great Peter Bernstein and the still vibrant Rob Arnott (http://www.cfapubs.org/doi/pdf/10.2469/faj.v58.n2.2524). The paper was about long-run equity returns relative to debt returns - they went back over one hundred years. The conclusion was that the run up in equity prices from around 1980-2005 was unprecedented in human history and that they were not reflective of reality. Bernstein and Arnott's point was that prudent portfolio managers should use normalized return expectations when thinking about the future direction of equity markets. So, too, should pension funds think of normalized rates of return on various asset classes.

Point 2, Independent Estimates of Cost of Capital: Many people do not spend the time to consider the philosophy underlying rates of return and costs of capital. Why do they exist? They exist because there are distortions to equilibrium created by time, space, preferences, risks, and so forth between one person's surrender of capital to someone else. In other words, what leads to someone else forgoing use of capital right now to someone else? A cost of capital, or required rate of return, clearly. I would argue that there are permanent factors that can be baked into costs of capital, including information asymmetries, different funding time horizons, different perceptions of risk, and so on. The creation of the Internet has permanently lowered the cost of information discovery for everyone, so you would expect that costs of capital would be permanently lower because of the Internet's existence. That said, there are different interpretations of factual data, and different ways of discounting the future. So there are going to be differences of opinion in costs of capital between providers of capital and users of capital. Here, cost of capital can be negotiated between the two parties to adjust for differences of opinion.

You might also expect there to be adjustments for all other factors, too. My point is that you could build theoretical costs of capital without using markets, and this would be independent of current market activity. I think there is such a thing, costs of capital independent of market activity. We use market rates, not necessarily because they are correct, but because they are convenient when we build our models. In normal circumstances we can use market rates because they are likely to be relatively unbiased, and reflective of a number of market participants' beliefs about what capital ought to cost. However, what are prudent capital decision makers to do when there are large scale distortions in a system? Here, an estimate of cost of capital independent of distortions is called for.

To summarize this point, my belief is that markets are not always efficient and that it is possible for entire markets to be poor price estimators (i.e. bubbles do happen). When these environments exist - and I think that is exactly the environment we are in right now - then it makes sense to create a non-market based estimate of cost of capital.

Point 3, Don't Double Count Risk - Discounted cash flow analysis can handle estimates of risk and return in two places:in the estimated cash flows and in the require rate of return. It sounds as if Aswath counsels to handle this in the flow portion of the equation. I am arguing that in the current environment if you are to handle estimates of required rates of return that it is dangerous to use market-based rates. SUPER importantly, when building a discounted cash flow model (same as a capital budgeting model) then it is critical that you do not double count risk. In other words, if you handle the risk in the estimate of cash flows, then do not double count risk in the cost of capital. The reverse is also true. If Aswath counsels handling risk in flows, I am certain that he has good reasons for it. I am counseling that people are likely using the wrong estimate of risks in the required rate of return portion.

Point 4 - The beautiful thing about investing is that results are not subjectively measured, but objectively. Yet, analysis is subjectively executed, not objectively. Meaning that, we all have our opinions : ) and I am certain that your perspective is also a valuable one.

Thank you, thank you for your interest in The Enterprising Investor!

Jason