RBD - Restaurant Brands

[Snoopy]

Charts suggest RBD has peaked out. Everyone out now?

I guess a pure chartist might be out, but this is yet another example of why you might not want to be a pure chartist. At $2.45 RBD is on a PE of around 9. For a company that has steady growth prospects this PE is too low.

An article appeared in the Auckland Herald last week pointing out that Kiwi's had not taken to Starbucks and it was effectively a dog. Russel Creedy would be able to admit that finally when he sold the whole lot onwards as a growing concern. This is old news.

There have been also articles in the press about how tough a time retailers are having. Interestingly, RBD was forced to make a statement to the stock exchange in mid January 2011 that unlike other retailers, RBD profits are in fact on track to grow strongly.

IMO the selling down of RBD has happened for no reason, as part of a flow on effect from other retail shares. RBD is probably now the cheapest share on the NZX when considering just the fundamental valuation. IMO those selling out of RBD over the last month or so are either pure chartists, or insane.

SNOOPY

[h2so4]

IMO the selling down of RBD has happened for no reason, as part of a flow on effect from other retail shares. RBD is probably now the cheapest share on the NZX when considering just the fundamental valuation. IMO those selling out of RBD over the last month or so are either pure chartists, or insane.

SNOOPY

Well SD, they may have found something of more value, but they certainly are not going to tell us. I accept your most likely explanation that they have probably gone mad.

[gregrday]

Gregr, I think my comment was slightly screwed up. What I was trying to say in an overly twisted way is that I think there are some issues regarding the appropriate discount rate that should be used to value RBD.

In an extreme situation it would be possible to have a discount rate greater than the earnings growth rate. With the benefit of hindsight, this is what has happened with RBD. I would argue that five years ago RBD was a $12.5m per year profit company given normal business conditions. I would equally argue today that RBD is a $25m per year profit company. That is very roughly a 15% compounding growth rate of 'underlying earnings' over 5 years. As long as the PWC discount rate for RBD was under 15% over those 5 years (and I think it was, let's make it 10%), then when we work out the present value of that underlying earnings growth pattern from a 2006 perspective then an interesting thing happens:

2006: $12.5m
2007: $12.5m x (1.15/1.1)= $13.0m
2008: $12.5m x (1.15/1.1)^2 = $13.6m
2009: $12.5m x (1.15/1.1)^3 = $14.2m
2010: $12.5m x (1.15/1.1)^4 = $14.9m
2011: $12.5m x (1.15/1.1)^5 = $15.6m

What we have here is an increasing earnings number every year, even though these future earnings are being discounted back to 2006 dollars. If this trajectory were to continue you could argue that the PV of Restaurant Brands earnings (from a 2006 perspective) can be as high as you like by just looking out into the future far enough. Obviously this is not possible, so there is some fantasy going on within this calculation somewhere. But the average earnings growth of 15% that I used is not the fantasy, because this is the actual growth rate achieved in this period (we know this with the benefit of hindsight). That means something else must be wrong, and the only thing in that equation not based on fact is the discount rate. The discount rate must be higher than the earnings growth rate to obtain a convergent share valuation.

Of course the other thing that must happen foir this farcical divergent valuation to be true is that earnings must keep growing at above 10% indefinitely. Clearly this also is unlikely, although we may get an average 10%+ growth rate for 10 years, and 10 years is 'indefintely' for some investors. I used the 10% average discount rate for ease of calculation but in reality I think the PWC discount rate for RBD over that period would have been less. That means my example as presented is conservative.

The point I am trying to make ( I think! )is that the PWC discount rate for RBD can never have been as low as they suggested. The whole basis for their calculation cannot be right, even if the actual mathematics they used to calculate that figure contains no computational errors. This is why I take these discount rates very cynically.

SNOOPY

Hey Snoopy,
ah, I understand now... bit slow really. But the situation you describe is quite plausible isn't it? We know that companies in the long-term cant grow past the rate of inflation, so the terminal growth rate is the big factor there. So the discount rate could have been... 'accurate'. To be honest, I've been a bit lazy using the PWC WACCs, I should really be calculating my own. But my feeling is that given the RBD bank loan interest rates are pretty low, around 7-8% is probably more-or-less reasonable.

cheers
Greg

[Snoopy]

The situation you describe is quite plausible isn't it? We know that companies in the long-term can't grow past the rate of inflation, so the terminal growth rate is the big factor there. So the discount rate could have been... 'accurate'. To be honest, I've been a bit lazy using the PWC WACCs, I should really be calculating my own. But my feeling is that given the RBD bank loan interest rates are pretty low, around 7-8% is probably more-or-less reasonable.

Not sure that I would agree that companies 'long term' can't grow past the rate of inflation, although I guess that depends on your definition of 'long term'. Over 5 years net profit at RBD grew by 15% compounding per year, yet inflation was probably about 3% compounding. Even if RBD profits stagnate for the next 5 subsequent years we are still looking at something around a 7% compounding growth rate over 10 years to 2016. That is way ahead of inflation and I think ten years would qualify as long term for many investors.

It is admirable of you to suggest calculating your own WACC and I am sure you could do at least a good a job as PWCC. However, IMO improving the mathematical precision of your figures will not cut much ice if the underlying method is dodgy. Sacriligious talk to some accountants I guess. Perhaps WACC as popularly understood works for most companies. But I am more interested in the 'outlier performance' of the 'best investments', not the average. And I don't think WACC works very well there.

The problem as I see it is that if a company is growing very fast, the WACC discount rate must be higher than the growth rate to produce a convergent share valuation. In my actual RBD example a 10% discount rate was not high enough, even though PWC"s calculated rate was lower and your own estimate is probably lower still.

Because of the industry in which RBD operates (food), their strong market position and access to a globally significant franchise parent I would argue that the true future earnings discount factor should be even lower, perhaps only 6%. But of course industry sector, strong market position and the global muscle of the franchise parent are not included in any WACC calculation. To investors seeking market outperforming returns I am not sure that teh concept of WACC and discount rates as popularly defined are very useful. But I would be happy if someone can convince me otherwise.

SNOOPY

[h2so4]

SD, you don't think Warren uses a discount rate? That's interesting.

[Sauce]

SD, you don't think Warren uses a discount rate? That's interesting.

Hi h2,

I think snoopy is challenging the method that one arrives at an appropriate discount rate, rather than saying one is unnecessary.

Buffett and Munger have often commented that the standard WACC caculation is fundamentally flawed because the formula includes a calculation of "Beta" which is a measure of share price volatility that is used to estimate risk. They deem this as irrational because in their opinion short term share price volatility is not necessarily related to the underlying fundamental business risk.

It appears that Buffet prefers to take the risk free rate of return (long term treasury bonds) and add a few percentage points to this for the "risk" attached to business ownership; presumably the amount added changes from company to company. He also provides the caveat that its not wise to drop ones discount rate too low in times of super low interest rates (such as now).

Snoopy, please correct me if I am wrong, but in more general terms I think you are saying this: Rather than bother with esoteric WACC formulas, why not just make your best judgement as to what is an appropriate rate of return that a rational investor would expect from the business based upon its quality and the risk of being in business, and use that as your discount rate?

I hope you don't mind me jumping in here but I am trying to learn as much as possible about valuation methodology at the moment and I am enjoying your discussion!

Regards,

Sauce

[h2so4]

Hi h2,

I think snoopy is challenging the method that one arrives at an appropriate discountIt appears that Buffet prefers to take the risk free rate of return (long term treasury bonds) and add a few percentage points to this for the "risk" attached to business ownership; presumably the amount added changes from company to company. He also provides the caveat that its not wise to drop ones discount rate too low in times of super low interest rates (such as now).


Regards,

Sauce

Why would it change from company to company? Isn't $1m cashflow from Grandmas Trotters the same as $1m cashflow from Coca Cola?

[winner69]

Why would it change from company to company? Isn't $1m cashflow from Grandmas Trotters the same as $1m cashflow from Coke a Cola?

They are the same of course

But in this discussion the $1m cash flow is a forecast / projection / calculation ... and the problem / question is what would pay to get a forecasted $1m from Grandma's Trotters and how much for the Coke $1m

Some analysts don't use WACC at all ...... they do cashflows under many scenarios and by applying probabilities against each come up with an expected value of the cash flows and use those as the basis of their valuations

Note the words applying probabilities and expected ... see it all really is a big guess (OK best guess / judgement) ... just like the equity premium used in WACC calculations is and just like what Warren might use

So back to Grandma and Coke ... what is the likliehood of those $1m cash flows actually be achieved

[Sauce]

Hi Snoopy, Gregday

Is it possible you guys are kind of saying the same thing in different ways here

Snoopy says:

Of course the other thing that must happen foir this farcical divergent valuation to be true is that earnings must keep growing at above 10% indefinitely. Clearly this also is unlikely, although we may get an average 10%+ growth rate for 10 years, and 10 years is 'indefintely' for some investors. I used the 10% average discount rate for ease of calculation but in reality I think the PWC discount rate for RBD over that period would have been less. That means my example as presented is conservative.

Greg says:

We know that companies in the long-term cant grow past the rate of inflation, so the terminal growth rate is the big factor there. So the discount rate could have been... 'accurate'.

If you are using a straight line formula you clearly cannot have a growth rate that is higher than the cost of capital. Bruce Greenwald, professor of value investing at Columbia university says this:

"The limits of sustainable growth are some fraction of the cost of capital. As we said, if growth equaled or exceeded the cost of capital, the return on capital would be infinite."

Exactly as snoopy shows. But since markets have limited sizes and competition, double digit growth rate for a straight line formula that assumes the growth rate indefinitely is basically absurd anyway isn't it?

So no matter how high the historical and near term growth rate is, if you are using a straight line formula you should accept that the long term growth rate will be lower than your cost of capital, as greenwald suggests. Alternatively use a DCF that combines an initial high growth period with a nominal terminal growth rate, exactly as Greg suggests.

I am still grasping these ideas and my maths is shaky at best, so please forgive me if I am simply displaying my ignorance here

Cheers

Sauce

[h2so4]

Oh I thought we were talking about discounted cashflows. Stupid me.

I'd pay more for Coke's cahflow and less for Grandmas. But that's margin of safety. Bigger margin for Grandma aye.