Buffett On Discount Rates: How To Apply His Methodology Today

Includes: AAPL, AMZN, BRK.A, BRK.B
by: Billy Duberstein

Warren Buffett values stocks based on the present value of free cash flows, like all sensible value investors.

However, his attitude towards discount rates is very unique.

Concentration/conviction vs. diversification appear to require different techniques, even if both types of investors use DCF analysis.

As a therapeutic exercise during this crazy quarter, I have been re-watching old Berkshire Hathaway (NYSE:BRK.A) (NYSE:BRK.B) annual meeting videos, which are now available (from 1994-onwards) on both YouTube and CNBC.

With all of the hullabaloo about rising rates and equity risk premiums in the current environment, I came across an interesting snippet from the 1996 annual meeting regarding discount rates. Like all sensible value investors, Buffett values stocks based on the present value of all future cash flows, and then buys at significant discount.

But Buffett does in fact think about the inputs differently than the "traditional" capital asset pricing model (CAPM). The curious thing about how Buffett values stocks is that, unlike most traders, he uses a risk-free rate (or close to it) to discount a company’s future cash flows (with a small premium in an era of extraordinarily low interest rates – like we have now).

From the meeting:

“We basically think in terms of the long-term government rate.

And there may be times, when in a very — because we don’t think we’re any good at predicting interest rates, but probably in times of very — what would seem like very low rates — we might use a little higher rate.

But we don’t put the risk factor in, per se, because essentially, the purity of the idea is that you’re discounting future cash. And it doesn’t make any difference whether cash comes from a risky business or a safe business — so-called safe business. So, the value of the cash delivered by a water company, which is going to be around for a hundred years, is not different than the value of the cash derived from some high-tech company, if any, that — (laughter) — you might be looking at.

It may be harder for you to make the estimate. And you may, therefore, want a bigger discount when you get all through with the calculation. But up to the point where you decide what you’re willing to pay — you may decide you can’t estimate it at all. I mean, that’s what happens with us with most companies.

But we believe in using a government bond-type interest rate. We believe in trying to stick with businesses where we think we can see the future reasonably well — you never see it perfectly, obviously — but where we think we have a reasonable handle on it.

And we would differentiate to some extent. We don’t want to go below a certain threshold of understanding. So, we want to stick with businesses we think we understand quite well, and not try to have the whole panoply with all different kinds of risk rates, because, frankly, we think that’d just be playing games with numbers.

I mean, we — I don’t think you can stick something — numbers on a highly speculative business, where the whole industry’s going to change in five years, and have it mean anything when you get through.

If you say I’m going to stick an extra 6 percent in on the interest rate to allow for the fact — I tend to think that’s kind of nonsense. I mean, it may look mathematical. But it’s mathematical gibberish in my view.

You better just stick with businesses that you can understand, use the government bond rate. And when you can buy them — something you understand well — at a significant discount, then, you should start getting excited.”

Source: Morning Session - 1996 Meeting (start at 2:00:20 and end at 2:03:35)

What about today?

With the 10-year rate now below 3%, one might think that it’s crazy to discount the future cash flows of a business by that amount – it would essentially value stable but no-growth but stable companies at 33x cash flow.

In 1996, here is what the 10-year T-note was doing on a monthly basis, according to FRED data:

1996-01-01 5.65

1996-02-01 5.81

1996-03-01 6.27

1996-04-01 6.51

1996-05-01 6.74

1996-06-01 6.91

1996-07-01 6.87

1996-08-01 6.64

1996-09-01 6.83

1996-10-01 6.53

1996-11-01 6.20

1996-12-01 6.30

Compare that to 2018:

2018-01-01 2.58

2018-02-01 2.86

2018-03-01 2.84

2018-04-01 2.87

2018-05-01 2.98

2018-06-01 2.91

2018-07-01 2.89

2018-08-01 2.89

2018-09-01 3.00

2018-10-01 3.15

2018-11-01 3.12

Source: FRED

Now, assuming Buffett puts a “little bit higher” discount rate on future cash flows, what might he use at this time? Perhaps a 5% rate?

This would definitely explain Buffett’s recent investments in large, monopoly and oligopoly-like companies such as Apple (Nasdaq:AAPL), virtually all the large U.S. banks, and the large U.S. airlines. While all of these companies may be challenged to grow in the way that, say, some of the large internet companies cloud and software companies are, these companies' current level of earnings may seem relatively safe to Buffett, either due to industry consolidation (in the case of the airlines) and each company’s large scale and conservative/improving balance sheets (some by way of regulation, such as the TBTF banks). If you use a base case of 5% discount on today’s cash flows and no growth, all of these companies’ valuations look quite compelling.

The difference between Buffett and most: Conviction

Of course, Buffett’s thinking on making investments is to ONLY invest in large, safe stocks with wide moats in industries with which he is intimately familiar. That leaves out a lot of companies for sure, but also means that when he and Charlie invest in a business, they likely project a base level of cash flows going out 10+ years which they believe are near certain, or almost as certain as a government bond.

In essence, Buffett and Munger aim to take out as much risk as possible through research, circle of competence, and their incredible smarts to essentially de-risk the asset. That's in contrast to sticking an extra equity risk premium just in case the business is riskier than they’ve forecast.

Another reason why this method may make sense: government bond coupons are taxed at the income tax rate, whereas qualified dividends are taxed at a lower capital gains rate. In addition, cash flows used for stock repurchases or reinvested in the business are essentially tax-free returns (until you sell the stock, of course). So, even with equivalent discount rates, the average stock investor comes out ahead of government bonds even when discounting cash flows at equivalent discount rates.

Contrast with other value approaches

The Buffett approach is quite different from that of other investors that might venture into investing in companies you know less well. It’s different from the process I usually use when valuing companies with which I am less familiar. The higher discount rates inherent in economically-sensitive stocks would also be warranted if one were more diversified and had a wider range of risk appetites.

For instance, Aswath Damodaran uses a slightly different process in estimating discount rates. Actually - he uses a wide range of methodologies, but tends to gravitate towards his own version of the Capital Asset Pricing Model. (You can find the textbook to his class here.) There are, essentially, two components to the CAPM methodology: the base equity risk premium that the market is applying to stocks at the time, along with the beta of a stock (and also the leverage the company is using, but that’s a company-specific phenomenon).

Equity Risk Premium: Damodaran estimates the equity risk premium by looking at consensus analyst estimates for S&P earnings and cash flows over the next 5 years, then discounting them back to the present and subtracting the risk-free rate. Voila! You have your appropriate ERP. (See page 68 of his Valuation class curriculum.)

Betas: In terms of betas, Damodaran takes an industry-level approach. He concedes that an individual stocks’ beta is a historical beta is a pretty flawed measure of risk, due to the idiosyncrasies of how any individual stock trades. Betas are also flawed as they are backward-looking measurements of volatility, which may or may not be a great measure for risk, depending on who you ask, and the company involved. If an industry is going through a positive or negative change, the historical betas will become less relevant or prove incorrect. (page 88 on part 1 of valuation course.)

Damodaran gets around this problem essentially through taking, say, the 10, 20, or 50 stocks in the same global industry, and taking the average betas of all of them, adjusted for their leverage. This leads to an industry-wide unlevered beta. So, you could essentially use the same unlevered beta for Nike (NKE) and Under Armour (UA) (UAA), instead of ascribing a lower beta to Nike than you would Under Armour (if you thought Nike was much less risky). The larger sample size, in theory, will filter out the individual stock anomalies and come closer to the true riskiness of any company within a particular sector.

Again, this approach works much better if you are more diversified and are also willing to venture into stocks in which you have less conviction.

In my recent valuation of Amazon (AMZN) (see Part 1, Part 2, and Part 3), I used this more involved method, although I could also employ the Buffett method as well, since I think Amazon has a fairly strong moat – one would only have to estimate some base case future level of cash flows with some level of certainty and then discount back.

What should you do?

There is no particular "correct" approach in the right discount rate to use, since one is essentially trying to put a mathematical figure on risk, which is a real-world, imprecise measurement subject to constant change.

Buffett’s extreme filtering out everything but his high-conviction bets within his circle of competence essentially substitutes for any “risk premium” that Damodaran adds on as the equity risk premium multiplied by industry beta.

Therefore, if you have high conviction in certain industries and believe strongly in the long-term cash flows of a business, the Buffett-like approach could work quite well; however, if you concede that you are a mere mortal in terms of industry knowledge and conviction, you are probably better off diversifying and adding the extra, appropriate discount rate in your valuations.

It’s quite possible (likely, probably) both approaches could yield similar results when valuing a particular company. Using a lower discount rate on earnings you are more “sure of” (so, perhaps focusing on less on growth and more on stability) could somewhat cancel out projecting higher growth but discounting by a higher discount factor. Obviously, it depends on what type of company you are valuing, and how large its moat/staying power.

Anyway, I found this comment from the 1996 meeting pretty interesting in light of present-day concerns over bond yields and economic growth prospects. Going forward in my models and other calls I am making, I will likely run each valuation through both "filters" to see if one valuation proves far different than the other. If they do, it's a sign I may need to do more research on the company or industry.

What do you think? Leave your comments below.

Disclosure: All information contained herein is provided “as is” and Billy Duberstein expressly disclaims making any express or implied warranties with respect to the fitness of the information contained herein for any particular usage, application or purpose. Any information, opinions, research or thoughts presented are not specific advice as I do not have full knowledge of your circumstances. Prior to making any investment decision you should consult with professional financial, legal and tax advisors to determine the appropriateness of the risks associated with such an investment. No assurance can be given that the objectives of a particular investment will be achieved or that an investor will receive a return of all or part of his or her investment. In no event shall Mr. Duberstein be responsible or liable for the correctness of any material used herein or for any damage or lost opportunities resulting from the use of such material. The information contained herein may not be copied, reproduced, published or distributed in any way without the prior written consent of Mr. Duberstein. Mr. Duberstein and the terms, logos and marks included herein that identify Mr. Duberstein 's services and products are proprietary materials. The use of such terms, logos and marks without the express written consent of Mr. Duberstein is strictly prohibited.

Disclosure: I am/we are long AAPL, AMZN. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.