The problems with DCF models are well known, i.e., that they have alarge proportion of present value reliant on the distant future and the potentially arbitrary nature ofsomeofthe inputsandstructure. Despitethatwethinkthattheyformanessentialpart ofvaluation.
The problem is that they were invented and historically used in aworld where risk freerates averaged 5% or more. In a world where the risk free rate is close to zero then the errors in such models explode.
Specifically, if the overall discount rate (WACC) falls from 10% to 5% in a very simple DCF then the proportion of the net present value accounted for by cash flows more than 5 years in the future rises from 70% to 95%. How far in the future can any analyst forecast? We would suggest that any human'sability to forecast financial variables more than about 5 years in the future is limited at best. At the veryleastsmallerrors at that forecasting horizon become very significant.
To the extent that when a DCF is used in practice as a measure of relative (as opposed to absolute) value, the proportion accounted for by the terminal growth period is notsucha problem as it can cancel out across securities, but there still remains huge model error in the path to that terminal value.
Having said all this, discounting is at the heart of finance. This is a very general statement relating to assets and liabilities and not limited to DCFs. If it is notpossible to put a "price on time" then there is agenuinelyintellectually painfulenvironmentwhere modelstructure is called into question.
Whatshould investors do? We cannot reject discounting and there is nochoice but to use it anyway. Sowe will keep using DCFs. We just have to be aware that a by-product of the low rate world is a scale of forecast error that is outside the bounds ofwhathas been previously seen and it is likely that those forecast errors may swamp any otherdifferences between stocks.
Whatwoulddefinitively break suchamodel? Byconstruction, if theoverall discount rate fell below the growth rate then the NPV becomesundefined, though in a semi-permanent low rate environment presumably expected growth rates fall too. We have already seen European companies issue debt at negative yields and in Japan 10 year rates areset at 0. Any movelower in discount rates would, we suggest, cause a significant methodological problem for financialanalysis.