Warren Buffett's Hurdle Rate

[HWWProject]

For those who use DCF models to determine a stock's intrinsic value, what discount rate do you use?  This "hurdle rate" should respond to changes in US bond rates, inflation, and the capital gains tax as well.  In reading Buffetts' LTS from the late '70s early '80s (era of rampant inflation) inspired me to come up with a simple equation based on his descriptions

 

Minimum Hurdle Rate on taxable investments = (Inflation rate) + (Yield on long-term tax-free muni bonds)/(1-Cap. gains tax rate) + (Equity Risk Premium)

 

Article: http://healthywealthywiseproject.com/2015/02/warren-buffetts-hurdle-rate/

 

With inflation and interest rates so low it almost seems odd to think about a 'hurdle rate', but intrinsic value changes dramatically from a 10% to 15% discount rate, so the concept is important. 

[Jurgis]

I always use 15% discount rate.  8)

 

Yeah, I know, this is not answer people want to hear. :)

 

My argument is that there's almost always a "risk free" (haha) BRK/FFH/MKL investment that will return about 15% annual. So anything else should beat that to be considered.

 

Of course, in current market, pretty much nothing passes this hurdle.

 

Now, I expect a lot of rotten tomatoes flying in my direction soon.  ::)

[HWWProject]

I'm in the 15% club too, it's been my standard.  Except recently been feeling I should lower it as all my calculated Intrinsic Values are so far below current prices.  That got me googling on what most value investors use.  Found several references to Buffett saying 13% (early 2000's), 15% (mid-90's), and Wally Weitz at 12% (2014). 

[Jurgis]

Except recently been feeling I should lower it as all my calculated Intrinsic Values are so far below current prices.

 

Bad reason to lower it.  ;)

 

Personally, I'd rather buy stock that is above Intrinsic Value based on consistent discount rate - so I can compare it with my other past and present analyses - than to change discount rate to get Intrinsic Value above the current price.

 

But there are people who do it differently. For example, they take risk free (treasury) rate or they take company's cost of capital or whatever else. Currently, almost everything is undervalued if you take treasury rate. Even if you take cost of capital - for good businesses now this is below 10% - a lot of stocks would be undervalued too. You have to decide yourself if these are good rates to use and if stocks are really undervalued as they'd appear at these discount rates. :)

 

The risk of using 15% discount rate for me is dual:

- Not buying a great business, since it looks expensive at this discount rate

- Selling shares in great business, since it looks expensive at this discount rate

 

Unfortunately, I have done both. :) So it's something to think about.

 

Conversely though, there were great stocks that appeared cheap even at 15% discount rate in last couple years. AAPL was one. Banks were too and somewhat still are. So it might be unwise to change the rate just because there are no cheap stocks based on it right now.

 

(BTW, there are obviously stocks which appear to be (somewhat?) cheap at 15% discount rate even now. They are the stocks of companies in doghouse. E.g. AGCO/DE/CF in agri, Buffett's IBM, some oil stocks - HAL maybe. Of course, the issue is that these companies are encountering headwinds, so it's tough to give them a growth rate that would make them cheap at 15% discount. But if you have some insight that crowd misses, then perhaps some of these might be worth it. Alternatively, there might be international and small/micro-cap stocks that are cheap at 15% discount rate ;). Good luck! )

 

 

[rb]

I'm in the 15% club too, it's been my standard.  Except recently been feeling I should lower it as all my calculated Intrinsic Values are so far below current prices.  That got me googling on what most value investors use.  Found several references to Buffett saying 13% (early 2000's), 15% (mid-90's), and Wally Weitz at 12% (2014).

I think that Buffett's hurdle is 10-12%. You can deduce that based on how he and Munger talk about different valuations they arrive at more generally. It gets more obvious when they talk about the investments in utilities.

 

I also think 15% is a bit on the high end. That's what? 13% above 10y treasuries? Keep in mind in mid 90s when WB was looking for 15% the treasuries were yielding 5-6%.

[mlcglobal]

I use the time you would be happy to see a nominal double (PV vs FV). For me this is 6yrs so 12%. For the 5 year types it's 15%. At the end of the day though you could use 0% for your NPV calculation as what is important is the price you pay relative to FV. Even though I use 12% I am still looking for a wide margin between price and NPV. A 50% discount to NPV at 12%  corresponds to an extra 12% IRR assuming NPV is reasonable on average. The return to FV would then be 12% + 12% IRR so 24% nominal over 6 Years. You can use any rate at all as at the end of the day it is just a way to anchor your analysis for comparison across investments. Results will come from the discount to whatever rate you use and the assumption that on average your valuation is reasonable. As i like to say 'Escrow to period end henceforth hitherto actual… nigh 'Refunding Escroe Deposits' (REDs)'.

[stahleyp]

 

 

My argument is that there's almost always a "risk free" (haha) BRK/FFH/MKL investment that will return about 15% annual. So anything else should beat that to be considered.

 

 

 

The problem is...according to moringstar, none of those have returned 15% a year to investors over the past 5, 10, 15 year period (with the exception of MKL at the 5 year range - and then, only barely). How in the world do you expect these to average 15% if they don't do it in a pretty strong investment environment, over the past 5 years, for instance?

[giofranchi]

The problem is...according to moringstar, none of those have returned 15% a year to investors over the past 5, 10, 15 year period (with the exception of MKL at the 5 year range - and then, only barely). How in the world do you expect these to average 15% if they don't do it in a pretty strong investment environment, over the past 5 years, for instance?

 

Paul,

you know very well that “a pretty strong investment environment” doesn’t mean it has been a “safe nor easy investment environment”. And those companies don’t reach for yield in an investment environment that is “strong” only because of unbelievable money printing…

Have they been too conservative? Of course they have! Have they made a mistake? Of course they have! Was it an easy situation? Not at all! ;)

 

Gio

 

[giofranchi]

For those who use DCF models to determine a stock's intrinsic value, what discount rate do you use? 

 

I take the CAGR for the S&P500 that I find on the front page of the last Berkshire Letter to Shareholders, and simply use that number as my discount rate. :)

 

Cheers,

 

Gio

 

[SpecOps]

I think I remember that Buffett uses a hurdle of 15%, but its pre-tax so about 10-12% after tax.

 

I personally use 10%, though if interest rates where to rise above 5% it would have to start going up.

[jay21]

I think Munger has said that BRK buys assets yielding 9% and uses cheap leverage, which will get you pretty decent return

[giofranchi]

Imo hurdle rate and discount rate are two different things… I look for businesses that have at least the possibility to return 15% annual compounded, and that’s my hurdle rate. But I use 9%-10% as a discount rate, whenever I try to perform a DCF analysis.

 

In other words, I use 9%-10% to calculate IV, then I try to buy with a sufficient gap between price and IV in order to achieve a compounded annual return of 15%.

 

Gio

 

[stahleyp]

The problem is...according to moringstar, none of those have returned 15% a year to investors over the past 5, 10, 15 year period (with the exception of MKL at the 5 year range - and then, only barely). How in the world do you expect these to average 15% if they don't do it in a pretty strong investment environment, over the past 5 years, for instance?

 

Paul,

you know very well that “a pretty strong investment environment” doesn’t mean it has been a “safe nor easy investment environment”. And those companies don’t reach for yield in an investment environment that is “strong” only because of unbelievable money printing…

Have they been too conservative? Of course they have! Have they made a mistake? Of course they have! Was it an easy situation? Not at all! ;)

 

Gio

 

Gio, most of the money printing has happened over the past 5 years or so. Keep in mind I'm looking strictly at investor returns - not increases in BV. The 10 and 15 year numbers still don't come close to the 15% "risk free". Even Buffett said he expects Berkshire to outperform only a "couple" points above the S&P 500 annually going forward.  I would go a far as to say one would be quite lucky to get 15% for any of these 3 over the long term. With all 3 of these firms trading above book value, I think that return wish is a stretch.

 

[philly value]

Choosing a discount rate to use in a valuation is one of the most difficult things I've encountered so far in value investing. There isn't any real, useful advice from the experts. Guys like Buffett and Klarman will tell you what not to use - beta / CAPM - but will not tell you what to use instead.

 

To me, a lot of times when I do a DCF I will solve for an implied discount rate based on the market price, which doesn't change anything fundamentally but makes comparison between opportunities a bit easier IMO. But it's really hard to know (a) what is a good framework to use in calculating the discount rate you would want for a given stock (obviously some stocks are riskier than others, in that the cash flows you project are less certain to occur, and thus a higher discount rate is warranted), (b) how do you account for low interest rates, do you use higher rates assuming they will go up? What if this results in you buying nothing, are you then relying on a macro forecast that rates will rise?

 

Looking at Fastenal and did a DCF yesterday, and get an implied discount rate of 8-10% depending on aggressive of my projections, although all were relatively conservative compared to analyst forecasts. Is 8-10% good enough? In this environment, it seems like it may be...but I don't know.

[giofranchi]

I would go a far as to say one would be quite lucky to get 15% for any of these 3 over the long term. With all 3 of these firms trading above book value, I think that return wish is a stretch.

 

Maybe you are right… But the math remains clear imo: 7% annual (instead of 8.9% historically) return on their portfolio of investments and FFH will compound BVPS at 15% annual. Selling today at 1.3 x BVPS, if in 10 years FFH still sells for 1.3 x BVPS, my return will be 15% compounded annual.

 

Gio

 

[Yours Truly]

If I recall from Alice Schroder's Darden University talk, she states that WEB wants a 10% day 1 return.

[HWWProject]

Except recently been feeling I should lower it as all my calculated Intrinsic Values are so far below current prices.

 

Bad reason to lower it.  ;)

 

...

...

 

(BTW, there are obviously stocks which appear to be (somewhat?) cheap at 15% discount rate even now. They are the stocks of companies in doghouse. E.g. AGCO/DE/CF in agri, Buffett's IBM, some oil stocks - HAL maybe. Of course, the issue is that these companies are encountering headwinds, so it's tough to give them a growth rate that would make them cheap at 15% discount. But if you have some insight that crowd misses, then perhaps some of these might be worth it. Alternatively, there might be international and small/micro-cap stocks that are cheap at 15% discount rate ;). Good luck! )

 

Thanks for reminder - not to lower my standards but rather have patience and let the market come to me.  This market requires a lot of patience.

 

I do want to have a discount rate that responds to inflation and prevailing bond yields so that I'm not missing opportunities and the equation will help me there.  I've read a Buffett quote (couldn't find the reference) that he'll move his hurdle rate up as inflation/yields move up, but the opposite is not true - he has a minimum floor he won't go below.  I'd especially like to know that floor rate.

 

(been shareholder of CF for past few years it's been a great company)

 

[rb]

If I recall from Alice Schroder's Darden University talk, she states that WEB wants a 10% day 1 return.

The 10% day 1 return was from a while back. Pre 0% rates. Though I don't think he is much below that. I also think that the way WB's 10% return is different than an accounting 10% return.

[Yours Truly]

anyone check out this article?

 

http://brooklyninvestor.blogspot.ca/2014/03/10x-pretax-earnings-case-studies-ko-bni.html

 

[constructive]

Choosing a discount rate to use in a valuation is one of the most difficult things I've encountered so far in value investing.

 

I find this statement pretty funny. I don't think anyone is enough of a genius at stock selection that choosing a discount rate is at all difficult compared to that.

 

Pick a random number between 8 and 15%. It is not going to have that big an effect on your returns, as long as you are willing to put in a lot of work to find investments that exceed your hurdle rate.

[philly value]

Choosing a discount rate to use in a valuation is one of the most difficult things I've encountered so far in value investing.

 

I find this statement pretty funny. I don't think anyone is enough of a genius at stock selection that choosing a discount rate is at all difficult compared to that.

 

Pick a random number between 8 and 15%. It is not going to have that big an effect on your returns, as long as you are willing to put in a lot of work to find investments that exceed your hurdle rate.

 

I'm not sure what you define as "stock picking." To me, "Stock picking" in the value investing world is buying businesses where the price is sufficiently below your estimate of the value of the business as to allow a margin of safety. We can observe stock prices. We can only estimate business values, and to do so necessarily requires the use of a discount rate. Stock picking and the choice of a proper discount rate for a particular stream of cash flows go hand in hand. You can't pick stocks without first picking a discount rate.

 

 

[constructive]

Stock picking and the choice of a proper discount rate for a particular stream of cash flows go hand in hand. You can't pick stocks without first picking a discount rate.

 

Sure. But assuming your investment analysis is accurate, the optimal investments at an 11% discount rate are the same as the optimal investments at an 8% discount rate. You have more options to choose from at an 8% discount rate, but they are suboptimal.

[philly value]

Stock picking and the choice of a proper discount rate for a particular stream of cash flows go hand in hand. You can't pick stocks without first picking a discount rate.

 

Sure. But assuming your investment analysis is accurate, the optimal investments at an 11% discount rate are the same as the optimal investments at an 8% discount rate. You have more options to choose from at an 8% discount rate, but they are suboptimal.

 

If you are using the same discount rate for all companies, then yes, it's clearly just a matter of "how low are you willing to go?" to find opportunities, and I guess the rate you choose becomes more about the macro environment than anything else (clearly the rate you use to discount cash flows must be lower today than in the 1980s).

 

But using the same discount rate for all companies is a bit of a stretch. Personally, I may be happy to invest in a stock like KO if, based on my best projections of future cash flows, I expect a return of 8% per year. Given that even 30-yr treasuries are sub-3%, earning an expected 8% return on a "safe" stock like Coke would be highly attractive. On the other hand, if based on my best projections of future cash flows, I expected an 8% return from investing in Facebook, I wouldn't do it. The risk of those cash flows that I project being far off the mark is too high relative to an 8% rate of return. I would want a much higher expected rate of return, maybe closer to 15%.

 

So I guess it comes down to how you handle risk in your valuation. Perhaps you use a low discount rate for all companies, but devise different scenarios and probability-weight them in order to come up with a valuation. Or, you come up with expected cash flows and discount them at a rate in line with the level of risk to achieving them, depending on the company.

 

The difference a discount rate makes is huge. For Fastenal, at an 8% discount rate, my model says I should be willing to pay $65-70 for the stock. At an 11% discount rate, I should pay no more than ~$35.

 

There are some times that the opportunity and the business are so obvious that you don't even need to project cash flows or think about a discount rate (AAPL at $380 per share is the best recent example I can think of). But it seems to me that with situations like Fastenal, you can't avoid thinking about future cash flows and a proper discount rate. Looking at the current multiple on earnings or cash flow and "eye-balling it" isn't much help in that kind of scenario.

 

 

[vinod1]

Stock picking and the choice of a proper discount rate for a particular stream of cash flows go hand in hand. You can't pick stocks without first picking a discount rate.

 

Sure. But assuming your investment analysis is accurate, the optimal investments at an 11% discount rate are the same as the optimal investments at an 8% discount rate. You have more options to choose from at an 8% discount rate, but they are suboptimal.

 

If you are using the same discount rate for all companies, then yes, it's clearly just a matter of "how low are you willing to go?" to find opportunities, and I guess the rate you choose becomes more about the macro environment than anything else (clearly the rate you use to discount cash flows must be lower today than in the 1980s).

 

But using the same discount rate for all companies is a bit of a stretch. Personally, I may be happy to invest in a stock like KO if, based on my best projections of future cash flows, I expect a return of 8% per year. Given that even 30-yr treasuries are sub-3%, earning an expected 8% return on a "safe" stock like Coke would be highly attractive. On the other hand, if based on my best projections of future cash flows, I expected an 8% return from investing in Facebook, I wouldn't do it. The risk of those cash flows that I project being far off the mark is too high relative to an 8% rate of return. I would want a much higher expected rate of return, maybe closer to 15%.

 

So I guess it comes down to how you handle risk in your valuation. Perhaps you use a low discount rate for all companies, but devise different scenarios and probability-weight them in order to come up with a valuation. Or, you come up with expected cash flows and discount them at a rate in line with the level of risk to achieving them, depending on the company.

 

The difference a discount rate makes is huge. For Fastenal, at an 8% discount rate, my model says I should be willing to pay $65-70 for the stock. At an 11% discount rate, I should pay no more than ~$35.

 

There are some times that the opportunity and the business are so obvious that you don't even need to project cash flows or think about a discount rate (AAPL at $380 per share is the best recent example I can think of). But it seems to me that with situations like Fastenal, you can't avoid thinking about future cash flows and a proper discount rate. Looking at the current multiple on earnings or cash flow and "eye-balling it" isn't much help in that kind of scenario.

 

Philly,

 

I hear you and understand your point. It is a valid concern, but using different rates makes comparing different investments very difficult. I just use 10% for everything and adjust the required margin of safety by using a probabilistic approach.

 

Buffett's comments on this make a lot of sense:

 

When we look at the future of businesses we look at riskiness as being sort of a go/no-go valve. In other words, if we think that we simply don't know what's going to happen in the future, that doesn't mean it's risky for everyone. It means we don't know – that it's risky for us. It may not be risky for someone else who understands the business.

 

However, in that case, we just give up. We don't try to predict those things. We don't say, "Well, we don't know what's going to happen." Therefore, we'll discount some cash flows that we don't even know at 9% instead of 7%. That is not our way to approach it.

 

Once it passes a threshold test of being something about which we feel quite certain we tend to apply the same discount factor to everything. And we try to only buy businesses about which we're quite certain.

 

But we think it's also nonsense to get into situations – or to try and evaluate situations – where we don't have any conviction to speak of as to what the future is going to look-like. I don't think that you can compensate for that by having a higher discount rate and saying, "Well, it's riskier. And I don't really know what's going to happen. Therefore, I'll apply a higher discount rate."

 

 

Vinod

[tede02]

As I recall, Alice Schroeder has said straight up that Buffett doesn't use DCF.  He agrees with the concept, and has mentioned the John Burr Williams theory many times.  However, as I understand it, he does not literally use DCF or any modeling.  Actually here's the clip of Schroeder talking about going through all his files when she was writing the biography and never seeing any modeling, cash flow projections etc.:

 

 

The one thing I've learned is if you are spending a ton of time trying to project cash flow, discounting, etc., that should be a sign in itself that the price is too high.  During a shareholder meeting a few years back, Buffett's words were, "It has to be obvious." 

 

I think its good to play around with DCF to see how small changes in both growth and discount rates can have huge effects on your calculations.  Playing around also gives you a sense of what various cash flow streams are worth generally.  But literally applying DCF, in my view, is just a waste of time.  There are too many variable.  Munger bought Wells Fargo a day from it hitting bottom in 2009.  When asked about it, he simply said he had to buy it because,  "It was too cheap."