Graham’s Valuation Formula vs. PEG Ratios

by JT McGee

Graham valuation, Graham valuation formulaPrice-to-earnings is undoubtedly a favorite of the financial media and retail financial outlet. Next to the PE multiple is usually the PEG ratio, or the price-to-earnings divided by growth.

A firm that earns $1 per share in earnings against a $20 per share price has a PE ratio of 20. This same firm, if it were growing at 20% per year, would have a PEG ratio of 1. If it were to grow at 10% per year, the PEG ratio would rise to 2. If it were to grow at 40% per year, the PEG ratio would fall to .5.

Price per share divided by earnings per share divided by the growth rate forms the PEG ratio. It’s difficult to mess this ratio up, which is one of the reasons I believe it to be so popular.

A Good PEG Ratio

So what’s a good PEG ratio, you ask? Traditionally, investors have favored securities that have a PEG ratio of no more than 1. That is to say that the PE ratio should be no greater than the annual growth rate. If a company grows at 20% per year, it should sell for a PE of less than 20. If it grows at 50% per year, it should sell for a PE of less than 50.

Like most things in all of finance, this is way too simple. For example, a company that grows earnings at 50% per year but trades for a PE ratio of 50 is vastly different than a company which grows earnings at only 8% per year, but trades for a PE of 8. At a PE of 8, a firm is likely to produce returns in excess of the hurdle rate from day one – 12.5% per year is very good, regardless of the 8% annual growth in earnings. A firm with a PE ratio of 50 is, in year one, providing only 2% of its valuation in the form of earnings. Future growth is far more important to the firm with an earnings multiple of 50, since the valuation is contingent on future growth – 2% per year doesn’t cut it. The firm with a multiple of 8 is doing quite alright – 12.5% per year beats the broad market all day long.

Basically, we can’t boil down a problem with multiple variables into a single binary decision of “buy if PEG is less than 1” or “do not buy if PEG is greater than 1.” It’s just not that simple. At a minimum, there has to be some kind of constant in the PEG ratio so as to ensure that slower-growth companies with low earnings multiples that are otherwise good investments are not thrown out because they do not grow at their multiple. Technically a company selling at a PE of 2 with 0% growth has an undefined (that’s what happens when you divide by zero, right?) PEG. It should be avoided if you look solely to the PEG – but who is going to pass on a company that spews $1 in earnings for every $2 it is worth? Come on! 50% per year is not good enough? Where the hell do you live? Zimbabwe?

How to Fix the PEG Ratio

I think the best way to fix a PEG ratio is the Benjamin Graham way. Graham liked to use a valuation equation that interjected a constant into the earnings equation.

The “valuation formula” as described by Ben Graham is PE = 8.5 + 2G, where G is equal to the earnings growth rate. A company growing at a rate of 10% per year would be permissible as an investment so long as it sold for a PE of less than 28.5. (8.5+2(10%) = 28.5.)

Keep in mind that price-to-earnings is just price-to-earnings. You can substitute free cash flow multiples in lieu of earnings, which I believe to be a much better basis for valuation. Back in Graham’s day, earnings were still king.

I prefer to use the further modified valuation formula of PE = 8.5 + .5G, where G is again equal to 5-year future growth. Notice how different this formula is to the traditional Graham formula. Graham would pay 28.5 times earnings for a firm growing at 10% per year. My formula would say to pay only 13.5 times earnings for the same company.

The beauty of this modified formula is that it is excellent at valuing cash flows much like a discounted cash flow valuation with 5 years of a growth constant where the discount rate is 10-12%. So, if a security scores well with the modified formula, you would manage theoretical returns of ~11% per year on your capital.

Plus, it makes for a good finance bar trick, if there could ever be one. And it totally masters quick DCF analyses for multiple choice finance problems where this is a large variance in between answers. Work smarter, not harder – the Graham Valuation Formula (assuming an 11% discount rate, consistent with high-beta companies or long-run equity returns) allows you to be within pennies of a DCF. The problem is, after doing it a few times, the next thing someone will want you to do is recite pi to 100 digits or calculate four-figure multiplication problems in your head. Just tell them you’ve mastered exponential growth in your head, but not multiplication – that’s too hard. 😉

Also, the PEG or Graham Valuation Formula ignores the balance sheet. Returns are likely to be higher than anticipated in the event a company is sold to a private owner because net assets must also be included in any valuation. Notice that the Valuation Formula considers only the value of future earnings. It does not take into consideration the $1 billion in cash the firm might be hiding under its mattress in highly-liquid securities, proverbial gold at the end of the buyout rainbow.

Valuation Formula Preference

The valuation formula preference is merely my own, but I think it’s far better than a simple PEG ratio calculation. Once modified, it allows for investors to shop for an investment that is undervalued based on future earnings and reconcile the mathematical concerns that arise from exponential growth.

It also allows for a built-in baseline return. Even a company with no growth prospects is worth buying at a PE of 8.5, as it would return nearly 12% annually to investors. No one would turn down a 12% return in perpetuity, especially if the company has an economic moat, as indicated by a higher than average cash return on invested capital.

Finally, it discounts aggressively future growth once modified to PE = 8.5 + .5G from 8.5 + 2G. Future growth is difficult to predict, and it would be beyond any value investor to buy a firm based solely on future growth.

All-in-all, PEG nor the valuation formula is a perfect way to value a company. The valuation formula above is, in my view, far better than a simple PEG without unnecessary complication. Finance is not a difficult game. No one loses or wins because of mathematics. Money is made or lost on the accuracy with which you can predict the future, and the most important variable is always the one most difficult to predict – how fast can any given firm grow for the next 5 years.

{ 2 comments… read them below or add one }

PK May 1, 2012 at 09:50

Two things:

“Plus, it makes for a good finance bar trick, if there could ever be one.” – Considering most people are impressed when you calculate tips in your head, I have to agree.

“No one loses or wins because of mathematics.” – I disagree. I think people lose all the time because of faulty mathematics (see: applying normal distributions to everything). I think you can do well by anticipating their errors.

Reply

JT McGee May 1, 2012 at 10:09

Heh, tips in your head. I can’t believe the number of tip calculator apps there are, and more importantly, the number of downloads for each. Nevermind that a phone already has a calculator – apparently the operations are too much to handle.

Man, that’s way out of my ball game. I will say that I don’t have the best understanding of statistics, but I do know that the graying of lines between finance and insurance is dangerous. Stats can be valuable, I suppose, in terms of building a portfolio for correlation. But that kind of removes the whole point of business valuation – which is pretty much the basis of my investing strategy. The stats wizards can have fun with their statistical arbitrage…let them go the way of LTCM.

I think Warren Buffett’s input on beta , though simple, is probably very sound. It doesn’t make sense that a stock undervalued at $40 is supposedly more risky the next day if it falls to $20. If you accept the total loss of an investment to be your risk, then price movements and statistics are pretty much irrelevant. Let me tell that to my finance professors – I’m sure they’ll love me for it. :roll:

Reply

Leave a Comment

*

Previous post:

Next post: