Here’s a brainteaser to kick things off: if a stock is trading at $100, how much is it worth?

You might answer $100, on the grounds that a stock’s value is whatever people are willing to pay for it. And you wouldn’t be alone. According to the popular “efficient-market hypothesis”, the price people pay for a share of a company takes into account all available information about that company – both what’s happened in the past and what’s likely to happen in the future.

But not everyone agrees. Here’s what Warren Buffett had to say about things:

QUOTE: “When the price of a stock can be influenced by a ‘herd’ on Wall Street with prices set at the margin by the most emotional person, or the greediest person, or the most depressed person, it is hard to argue that the market always prices rationally. In fact, market prices are frequently nonsensical.” – Warren Buffett

Investors like Buffett reckon a company's stock’s price is often unrepresentative of its actual value. Instead, it reflects all sorts of other factors – including things like hype and the general market mood.

If you take market prices at face value, then you run the risk of missing out on opportunities – and getting caught out by downturns. Sometimes a company’s stock price can be an accurate reflection of what the business is worth. But a good investor needs to know how to figure out that value for themselves – regardless of what others are willing to pay.

Stocks with different characteristics are typically valued using different methods. Following this flowchart will help you decide which technique to use for any given company:

The **dividend discount model** calculates how much a stock is worth today based on how much you should receive over time from its regular dividends. One popular variant is the **Gordon Growth Model**:

**D** is the **estimated value of next year’s dividend**. By looking at previous dividends (and whether the company’s indicated that it’ll raise or lower them going forward), you can often make an educated guess at what this will be.

With any luck, those dividends will increase as the company makes more money. That’s where **g** comes in: it’s the **constant growth rate** for, in this case, dividends. The key word here is *constant.* g has to be stable and realistic over the long term: a company that’s been doubling its dividends every year, for example, is unlikely to do so forever. In fact, g has to be less than or equal to the prevailing rate of economic growth – otherwise the company would be on track to grow larger than the entire global economy.

Estimating g is tricky. Historic growth trends might give you an idea of what’s to come, but ideally you’d dive into what drives the company’s business in order to forecast its future growth.

**r**, meanwhile, is the **discount rate**. It’s a way of figuring out what an amount of money received ten years from now is worth today. One way of calculating *this* is the **Capital Asset Pricing Model** (**CAPM**), which considers the risk you’re taking on and how much you could make elsewhere.

The yields of super-safe 10-year US government bonds are typically used as the **risk-free rate** **of return**. The **historical market risk premium** is how much extra the stock market overall tends to deliver *above* that risk-free rate: the premium its investors are paid for taking on risk. Use a long-term average to capture multiple economic cycles: since 1928, the risk premium for US stocks has averaged at 6.43%.

**Beta**, meanwhile, measures the volatility of the stock you’re valuing – it’s a proxy for how risky it is compared to the wider market, helping you figure out what your compensation should be for taking on the risk of that *particular* stock.

That’s everything that goes into the Gordon Growth Model. Now to apply it...

In order to apply our dividend discount model, let’s take a big, recognizable company: **Apple**. To start with, we need to find D: what next year’s dividend will likely be. Looking at its financial results, we can see Apple has paid $0.205/share the last couple of quarters – putting it on track for a 5% increase on last year’s dividend (you can find this info for companies here). Let’s assume Apple’s quarterly dividend increases by 5% again next year:

D = 0.205 x 1.05 x 4 (quarters) = $0.861

Let’s look at r next. Some sites compile CAPM calculations for you, but as they aren’t available for every stock we’ll do it ourselves. We’ll use the 10-year US Treasury yield as our risk-free rate of return: at the time of writing, that’s 1.19%. The 92-year average risk premium of US stocks, meanwhile, is 6.43%. And Apple’s beta (which you can find on **Yahoo Finance's** Statistics tab) is 1.27. So:

**r** = 0.0119 + (1.27 x 0.0643) = 0.093561

Next up is g – a much harder number to gauge. Although Apple’s last dividend increase was 5%, there’s no guarantee that growth will stay that way.

This is where valuation becomes more of an art than a science. You might, for example, think that Apple’s rumored electric vehicle will be a big hit, driving 5% dividend growth for years to come. On the other hand, it could go the way of fallen Finnish phone king **Nokia**. That’s a call for you to make in *your* valuation, based on the company’s fundamental value-drivers. Remember, though, that our model says g can’t be higher than the current global economic growth rate of 4-5%.

For now, let’s take a happy medium of 3%. Putting everything together:

Share Value = 0.861 ÷ (0.093561 - 0.03) = $13.55

And yet Apple’s shares are, at the time of writing, trading at $130. So either those shares are *seriously* overvalued, or there’s something wrong with our model. Unfortunately, it’s probably the latter.

Valuation methods built on dividends only work for companies where dividends are an important reason to own the stock. And for many firms nowadays, that simply isn’t the case: rather than distributing profits to shareholders directly by way of dividends, companies like Apple are buying back their shares, ploughing money into research and development, or simply sitting on the cash.

**The takeaway:** The dividend discount model compares expected dividend payments to the broader market to estimate a stock’s value – but it only works for companies that primarily derive their value from their dividends.

The **discounted cash flow (DCF) model** is based on a similar concept to the dividend discount model, but it takes into account the money that a company *brings in* as well as what it pays out.

This more comprehensive model does a better job of capturing the value of companies that don’t pay out most of their profit via dividends.

Here’s how to use the DCF model.

**Forecast future cash flows.** “Free cash flow” is the amount of cash that a company brings in from its operations less the amount it has to spend maintaining those operations. As a starting point, you can usually find a firm’s historical cash flow data online. If not, you can calculate it from financial statements collated on sites like Yahoo Finance (free cash flow is “total cash flow from operating activities” minus “capital expenditure”).

From there you’ll need to estimate what future cash flows are likely to be for the next few years, based on your take on the company’s fundamentals. Apple’s 2020 free cash flow was $73 billion.

**Apply the discount**. As with dividends, you can use the discount rate (calculated using CAPM) to give each future cash flow a value *today*. Do this for each year of future cash flow you’re forecasting:

**Calculate the terminal value.** With any luck, your chosen company will exist for longer than a few years – so we also have to capture its value *beyond* your cash flow forecasts. There are a few ways to do this: you could figure out how much the firm’s assets would be worth if it sold them off, or you could use a profit multiple.

For simplicity’s sake, we’re going to calculate this future “terminal value” using a perpetual growth model that assumes the company will keep expanding (albeit slowly) forever:

Look familiar? It’s the same structure as the Gordon Growth Model. Just remember that in this case the constant growth rate g is for free cash flow, not dividends. You should be able to estimate this based on your future years’ cash flow predictions.

**Apply the discount to terminal value**. As with those cash flows, you’ve got to divide the terminal value by (1+r)^N – where N is the number of years for which you’ve forecasted specific cash flows.

**Add it all up.** To figure out what the stock’s worth today, you need to add the discounted value of each future cash flow to the company’s discounted terminal value. Then finally…

**Divide by the number of shares**. You can find this on Yahoo Finance’s Statistics tab. Here’s a reminder of that initial diagram:

And there you have it: your very own cash flow-based valuation of what a company’s shares are worth. It’s actually pretty straightforward in practice – although the DCF model also has its limitations…

Let’s run through a quick example and apply the discounted cash flow model to Apple. To keep things simple, let’s pretend it’s January 2021 and we’re forecasting three years into the future.

First, we’ve got to forecast future cash flows. We could do this ourselves – lucky for us, however, investment analysts have got there first. Their estimates for Apple’s future cash flows can be found here (but remember, we have to discount these to find their value today).

Next we calculate the terminal value, discount this, and add it to our three years’ discounted cash flows. Divide the result by the number of shares, and ta-da: according to our DCF model, in January 2021, an Apple share was worth $80:

Since Apple’s shares were changing hands for around $130 in January 2021, this analysis would probably have led you to conclude the stock wasn’t worth buying at that price.

The DCF model’s drawback, however, is that it suits firms with predictable cash flows: utility companies, say. But many businesses have cash flows that are much harder to predict. Take Apple: customers might pour in chasing after the latest **iPhone**, giving one year’s cash flow a short-term boost. On the other hand, **Apple TV** could suddenly shed subscribers, leading to weaker-than-expected cash flow.

**The takeaway:** The discounted cash flow model attempts to estimate a stock’s value by capturing future incomings and outgoings. It works well when cash flows are predictable – but that’s often not the case.

**Price-to-earnings** (**P/E**) **multiples** are one of the most commonly used valuation metrics and are based on company earnings, a.k.a. profit. You simply take the company in question’s share price and divide that by its earnings per share (EPS). Or you can just go to – you guessed it – Yahoo Finance.

You’ll come across two values: “trailing” P/E and “forward” P/E. The former uses historical earnings data, while the latter is based on analysts’ forecasts for future earnings. Evidence suggests that forward P/E is a better predictor of value, so you’re probably best off using that. But a stock’s P/E is largely useless by itself: you need to compare it to similar companies or the relevant sector average (you can download industry-wide multiples here).

Be wary when comparing P/E multiples, however. For one thing, P/E can present companies with lots of debt as automatically more expensive than companies with less. Debt-laden **Netflix**, for example, has a high forward P/E multiple of 43, partly because high interest payments (as a proportion of its income) reduce its profit. You should therefore look up how much a company pays in interest when assessing its P/E – as well as its tax liabilities, which can have the same effect.

That’s the problem with P/E: it doesn’t tell you *why* something has the multiple it does. At the time of writing, Apple has a forward P/E of 29, while Facebook’s is 20. That might suggest Apple’s overvalued and should have a lower share price. *Or* it might mean investors think its earnings forecasts look too low (since higher EPS means a lower P/E).

Similarly, a relatively low P/E could mean that a stock is cheap – or simply that forecasts appear too high. Figuring out which is the case is tricky: you can get a sense of what’s going on by doing your own research on the company, but only time will tell.

Ultimately, all valuation models’ value lies in their being able to formalize your assumptions and estimates about individual stocks and apply them consistently. They’re no substitute, however, for actually understanding a company, its market, and the fundamental factors that will drive its future performance.

**The takeaway:** P/E lets you make handy comparisons between companies and industries – but like all valuation methods, it’s only as good as what you put in.

🔹 The dividend discount model works well for stocks that derive their value from dividends – but that’s not true for many companies these days.

🔹 A discounted cash flow model can do a better job of calculating the value of companies that retain most of their earnings – but it involves making bold forecasts about the future.

🔹 Price-to-earnings (P/E) is more helpful for valuing stocks relative to one another – but a high P/E could mean either that a company is expensive, or that its forecasts are just too low.

🔹 Each valuation methodology has its pluses and minuses, but none are a replacement for diving into a company’s fundamentals to understand how it’ll perform in the future.

Now put your knowledge to the test with our short quiz!

All the daily investing news and insights you need in one subscription.

Learn More**/3 •** Your free quarterly content is about to expire. **Uncover the biggest trends and opportunities. Subscribe now for 50%. Cancel anytime.**