Valuing the Future

For a chronological index of my path to the CFA, click here.

Well here we are at 2009, another neujahrskonzert safely behind us, and another CFA study guide for me as well. I’m now about to begin the fifth of the six volumes I need to understand for the Level II CFA exam six months from now in June.

Volume IV: Equity is, as you might have guessed, mostly about how to examine a stock and come to a conclusion about whether it is over- or undervalued, and by how much. I was very excited about this volume because though I entered the CFA program to get a “well-rounded” investing education, equity valuation is a huge percentage of what I’m personally after.

I began the equity tome with no idea as to how many techniques it would teach and how different the approaches would be. When I finished the book almost 600 pages later, I learned that they are all variations on just a couple themes.

Now a small portion of the equity study guide centers around using a relative valuation approach, whereby you determine an individual stock’s over- or under-valuedness based upon how its P/B, P/E, P/S, P/CF, etc. compares to industry averages. I won’t comment much on this topic except to say that I don’t like it because I don’t trust it. Even within a single sector of a single industry there can be large variations in the underlying economics (and accounting) of a set of seemingly similar companies. And I could have made all sorts of horrible mistakes at the top of the dot-com or real estate bubbles if I’d been making decisions only on a relative basis. Thus endeth the lesson.

In my opinion, the “meat” of Volume IV is in the 4 techniques it teaches to compute not relative but intrinsic value. All of these use the basic time value of money (TVM) approach to discount future cash flows (DCF) and come up with an estimate as to what that future cash is worth today. Surprising to me was that I expected there to be considerable debate and variety concerning how best to predict future cash flows, but not about how to define cash in the first place. However, a major difference between the techniques actually lies in what perspective you take regarding which parts of the earnings you receive and have access to as a shareholder.

Let’s take a quick look…

Technique #1: Dividend Discount Model
The first technique of valuation, DDM, is based upon the reality that you, the investor, are buying a very tiny slice of a company by buying its stock. And while you may get proxy voting materials in the mail, you really have no power to influence the company’s management to do anything like cut you a check for your portion of retained earnings. As a tiny, insignificant voice in a sea of millions the stock’s worth to you is only the present value of all future dividends as that’s the only cash you ever really get your hands on.

This valuation technique certainly seems the most tangible and appropriate to the small individual investor. But not all stocks pay dividends and many that do, pay out only a very small percentage of actual earnings. You therefore run into a problem if you use the historical dividend payments and growth rates to project out your expected dividend revenue stream in the future, and then use TVM to compute the present value. You end up coming up with a fair value of something like $1.85 per share for a stock currently trading at an unheard of 5-year recession-low of $32.00!

Bottom line is, unless you assume a large change in the dividend payout ratio in the future, doing the typical extrapolate-the-future-from-the-past sort of valuation using DDM would only result in your buying large, mature, slow-growing behemoths that already pay out a significant portion of their earnings as dividends. And you’d probably only buy during significant economic downturns. Not that there’s anything wrong with that!

Techniques #2 & #3: Free Cash Flow to Equity, and Free Cash Flow to the Firm
So whereas dividends are the cash flows actually paid to shareholders, the so-called free cash flows are the cash flows available to be paid. Simply put, take all cash coming in and subtract any operating expenses along with whatever money you need to buy new equipment to replace the old that is wearing out. What you have leftover is a cash stream that could be distributed to shareholders without killing off the underlying business. These are your future cash flows used when employing valuation techniques #2 and #3.

Now there are two ways one can go about security analysis using free cash flow. One way, free cash flow to the firm (FCFF), computes the cash that could be paid out to all suppliers of the firm’s capital. This includes not only the common stockholders but also the bond and preferred stock owners. So by using FCFF you’re initially computing a present value not for the common stock but for the whole firm. Therefore once you get the whole firm’s value you just subtract off the market value of the debt and preferred stock and you’re left with what you’re after, the value of the common stock.

The alternative is to use free cash flow to equity (FCFE). You start with all the cash from operations just like when using FCFF but you subtract off interest & principal debt payments as well as preferred stock dividends. Thus the present value of this residual cash stream is what the common stock is worth.

Now in reality a business may pay out some portion of these cash flows as a dividend but retains a sizable % for expansion and new projects. So why bother performing valuation using this pie-in-the-sky theoretical available cash flow that you will probably never be able to touch? I think it helps to think of what the shares may be worth not to a middle-class private individual but to a Warren Buffett or to a firm looking to do a merger or acquisition. They could purchase so many shares such as to become majority owner(s) of the business, and then have real access to its cash flows and retained earnings. And they’re out there buying and selling in the marketplace right along with you.

And isn’t it interesting that even if you perfectly knew all future cash flows of a firm and the correct discount rate for DCF, the present value you’d place on it is still not necessarily a constant for all investors! The value to you also depends upon your perspective of how much control you will have as a shareholder. It’s a sort of intrinsic value theory of relativity.

Technique #4: Residual Income Model
So as you’ve noticed up to now, all valuation models have been based solely upon cash flows that you predict to be coming in the future. And any looking backwards in time is only to estimate growth rates and payout ratios. But given how hard it is to predict the future, wouldn’t it be nice if there was a valuation model that is also a function of the past?

The residual income model (RIM) does exactly this. The valuation is partly a function of the current book value (equity) and partly a function of the future net income minus the opportunity cost of tying up your money in buying the common stock. And in a sort of academic ideal accounting universe, it works out to give you exactly the same valuation as DDM. An excellent overview of it here if you’re interested.

Reliance Upon the Capital Asset Pricing Model
You can certainly appreciate how difficult valuation can be based upon predicting all the whens and how muches of future cash, but an additional challenge is what discount rate to use to reflect future cash flows back into present day dollars.

CAPM is used extensively throughout the study guide to handle computation of the equity discount rate. As mentioned before, I’ve found it very hard to trust CAPM due to its empirical failure to explain stock return variations and the assumption that they’re normally distributed. I keep trying to have an open mind, thinking that my distaste is just because I don’t understand CAPM entirely. But it makes it hard to stay objective when I hear mentioned in one of the CFA Institute podcasts that even certain top investment bankers have been known to refer to CAPM as CRAPM…

Putting it into Practice
Like other techniques taught in the CFA program, you can think you’ve grasped the basics but run into huge speed-bumps when you attempt to try them out in the real world. The CFA problem sets always give you the current risk free rate, beta, and equity risk premium to use for determining the CAPM (grrr…) equity discount rate. But I got lost for hours on the web just reading about all the different approaches one can take to determining what they are oneself (best attempt is here). Should your proxy for the risk free rate come from the shortest duration US treasuries, or the duration that matches how long you intend to hold? Should beta be estimated over the last 2 years, 5 years, 10 years…? What is the equity risk premium or has it even disappeared?! If I invest today and the risk free rate changes a couple percentage points in a year, does my value buy morph into quick! sell!?

FCFE and FCFF turn out not to be as easy to implement as well. I do now use the cash flow statement more than I used to, but often have trouble determining what exactly comprises working capital vis-à-vis cash flow analysis versus the standard current assets minus current liabilities working capital vis-à-vis the balance sheet.

I’m optimistic that this will all get easier with time, but my attempts so far at valuing individual equities have left me with values I’m not sure I trust because I’m not sure I accounted for everything in the cash flow properly. The CFA study guide mentions that the DDM can be good for valuing broad-based equity indexes and that is currently about the only valuation I trust I can do properly. But instead of using CAPM to determine a discount rate I instead solve for the implied discount rate given the price of the index today. I’m getting rates of return in the low to mid teens for many, which probably makes sense given last year’s large draw-down.

In case you’re wondering which valuation model to apply and when check out DCF Models: What They Are and How to Choose the Right One. It’s a nice summary though perhaps most helpful as a cheat sheet after you’ve studied all the underlying models in detail.

Happy New Year everyone. Hope your studying is going well and your investing even weller! Here’s the running total of logged study time…

Read/Work Problems of Volume 1: 26 hours
Read/Work Problems of Volume 2: 15 hours
Read/Work Problems of Volume 3: 13 hours
Read/Work Problems of Volume 4: 32 hours
Total Preparation Time So Far: 86 hours