Investing’s First Principles: The Discounted Cash Flow Model
Brian Michael Nelson, CFA, is the author of Value Trap: Theory of Universal Valuation.
“People’s thinking process is too bound by convention or analogy to prior experiences. It’s rare that people try to think of something on a first principles basis. They’ll say, ‘We’ll do that because it’s always been done that way.’ Or they’ll not do it because, ‘Well, nobody’s ever done that, so it must not be good.’ But that’s just a ridiculous way to think. You have to build up the reasoning from the ground up — ‘from the first principles’ is the phrase that’s used in physics. You look at the fundamentals and construct your reasoning from that, and then you see if you have a conclusion that works or doesn’t work, and it may or may not be different from what people have done in the past.” — Elon Musk
I couldn’t sleep. I knew something was wrong. The numbers just didn’t make sense. For years, pipeline energy analysts seemed to be adjusting their valuation models for pipeline master limited partnership (MLP) stocks in order to explain what was happening to the price.
But why? Why adjust the models for one set of companies and not for another? Cash is cash and value is the measure of cash going into and out of a business. There aren’t different rules for different companies. Valuation is universal.
Analysts were valuing MLPs on the price-to-distributable cash flow valuation multiple and on the distribution yield, or the distribution per share divided by the share price. But growth capital spending supports distributable cash flow and drives it higher in the future. The pipeline MLP valuation calculations were ignoring this. Why should pipeline MLPs receive a free pass on the shareholder capital invested in growth projects when other companies didn’t?
How imbalanced was the MLP valuation processes? Meta Platforms, formerly Facebook, will spend a minimum of $10 billion this year on its metaverse division, Facebook Reality Labs, to build virtual and augmented reality applications. Imagine ignoring those billions in growth capital spending and still giving Meta credit for the free cash flow growth associated with that spending. That’s what was happening with MLPs and distributable cash flow, and when the market caught on, pipeline MLP shares collapsed.
I describe the Kinder Morgan and MLP story in my book Value Trap because it emphasizes first principles. The discounted cash flow (DCF) model is universal. So, what do I mean by this? And what are first principles? Let’s take P/E ratios. Though every valuation multiple can be expanded into a DCF model, P/E ratios aren’t necessarily shortcuts to the DCF model. When misapplied, they can lead to the wrong conclusions about a company’s value.
For example, a P/E ratio of 15 may be cheap for one firm and expensive for another. This is because certain variables have a confounding effect that limits what valuation multiples can reveal about a stock’s value. The cheap company could have billions in net cash on the books and huge growth prospects, while the expensive one could have billions in debt and poor growth prospects. Yet they still have the same P/E ratio.
Valuation multiples can be helpful when properly applied and with an understanding of what they are proxies for. That low P/E stock may not be cheap if the firm has a huge net debt position. That high P/E stock may not be expensive if it is asset light with a pristine net cash-rich balance sheet and tremendous prospects for free cash flow growth. But many analysts have forgotten that P/E ratios are an imperfect stand-in for the DCF model and shouldn’t be used in isolation.
This has opened the door to all sorts of spurious financial analysis. Think about all the quant factors that statistically “explain” returns on the basis of this or that multiple. There are thousands of forward-looking assumptions embedded in each valuation multiple. Just because that multiple is high or low doesn’t mean the stock is a good buy.
Many analysts today apply the P/E ratio, P/B ratio, EV/EBITDA, and other multiples by themselves as though they were distinct from the underlying DCF model that they are derived from. Some even question whether the DCF model is still relevant. Does forecasting future free cash flows and discounting them back to the present day at an appropriate rate still make sense in the meme stock era of GameStop and AMC Entertainment?
The answer is yes. In valuation, first principles remain essential: Every valuation multiple has an implicit DCF model behind it.
With MLPs, we know what was wrong with their valuations. Relying on “distributable” metrics is like valuing Meta by deducting only an estimate of its “sustaining” capital spending while completely ignoring its metaverse-related growth capital spending — and still crediting the company with the future cash flows generated by that spending.
The MLP bubble demonstrates how applying valuation multiples absent a supporting DCF model can be a recipe for disaster. Indeed, using valuation multiples without a firm foundation in investing’s first principles won’t yield much insight. Only the DCF model can help determine which 15 P/E stocks are cheap and which aren’t.
Such errors may help explain the replication crisis in empirical quantitative finance. I believe most statistical analysis that explains stock market returns through valuation multiples is flawed. The relationship between stocks with similar multiples hasn’t really held up in recent years. Why did we ever think it would or could?
If we can understand that two stocks with the same P/E ratio can be undervalued or overvalued, why would we believe the performance of stocks with similar valuation multiples would yield actionable data? And what does this imply about the value vs. growth conversation? If we’re not using the DCF model, we could all be taking a random walk when it comes to value and growth.
All of this helps explain why the DCF model is not only relevant to today’s market but remains an absolute necessity. As the 10-year Treasury yield increases and stocks come under pressure, we need to keep the DCF model in mind. After all, those yields form the basis of the weighted-average cost-of-capital assumption.
In this shifting landscape, a return to investing’s first principles is inescapable, and the DCF model is an essential tool for navigating what lies ahead.
For more from Brian Michael Nelson, CFA, don’t miss Value Trap: Theory of Universal Valuation.
If you liked this post, don’t forget to subscribe to the Enterprising Investor.
All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.
Image credit: ©Getty Images / Казаков Анатолий Павлович
Professional Learning for CFA Institute Members
CFA Institute members are empowered to self-determine and self-report professional learning (PL) credits earned, including content on Enterprising Investor. Members can record credits easily using their online PL tracker.
Excellent assessment, I completely agree to what’s been said here. It’s important to understand that the valuation metrics in isolation do mostly fail to give a correct picture and must be accompanied by DCF analysis.
Thanks Sir!
That is a very good article.
One item that he did not really address is the misleading use of factors that ignore DCF in the construction of indices. The S&P Value Index uses three factors, Price to Earning, Price to Book and Price to Sales. Price to Earnings may be of some use, but can be quite misleading, as the article explains. But Price to Book and Price to Sales contribute nothing to the assessment of value. EV to EBITDA is a useful first step as it does eliminate the influence of leverage. Forward EV to EBITDA is quite useful as it is a major component of a well constructed DCF model.
Despite how totally misleading value indices are, they are still widely used compare the performance of value based investing and growth based investing. Value investing is about intrinsic value and DCF models seek to assess intrinsic value.
100% agreed on DCF being the critical method to valuation.
Ultimate decision to put a price on something maybe based on combinations of other methods or criteria’s due to unique circumstances but the intrinsic value of any project or business can only be assessed if DCF with prudent assumptions is used in conjunction with macro and microeconomic factors considered as well.
Completely disagree. Just like every metric is useless, when used in isolation of all other info, so too is the DCF. Piles of cash outflows will be a positive for the business, and vice versa. Piles of cash flows are ignored by DCF. Minority interests make the whole exercise pointless because you don’t own ‘the whole’. Cash is more easy to manipulate than accrual accounting. Etc Etc
Interesting article, but there does appear to be a pretty substantial error in the foundation you lay for argument.
It appears that in the first part of the article, you are saying that price to distributable cash flow doesn’t adjust for capex? Likewise, in the paragraph that immediately follows, you state “Imagine ignoring those billions in growth capital spending and still giving Meta credit for the free cash flow growth associate with that spending.”
Here’s the problem. Both distributable cash flow and FCF are post-capex cash flow. As such, both measures already account for capex, with the cash flow that they provide truly being a measure of that which can be delivered to investors.
The only way that a DCF model might use either of these metrics, yet not give credit for capex facilitated growth is if the growth rate were set to 0%. However, that isn’t an error of distributable cash flow or FCF, that is an error on the part of the analyst in setting the growth rate.
You are correct that a model that deducts capex and then sets the growth rate at 0% will produce incorrect results. But a properly constructed DCF model will not do that. Projecting growth rates, attributable to capex, for technology companies can be challenging, but not always. In other industries it is routine. In a DCF model valuing oil reserves in a reservoir under development the capex for drilling additional wells will be deducted and the growth in the daily production, and resulting revenue, will also be included. With technology companies, the projected growth attributable to major capital projects is often disclosed in considerable detail.