How to build a Discounted Cash Flow (DCF) model
6 min read
Get full access to all Data Science, Machine Learning, and AI courses built for finance professionals.
One-time payment - Lifetime access
Or create a free account to start
A step-by-step guide covering Python, SQL, analytics, and finance applications.
Or create a free account to access more
Get full access to all Data Science, Machine Learning, and AI courses built for finance professionals.
One-time payment - Lifetime access
Or create a free account to start
A step-by-step guide covering Python, SQL, analytics, and finance applications.
Or create a free account to access more
A discounted cashflow model (DCF) is one of the most commonly used models to value a company. DCF models are generally created by investors or analysts to determine if the company’s stock is overvalued or undervalued at its current market price.
The key principle behind a DCF model is that the current value of a company can be determined using its future cash flows. The investor, using some methodology, arrives at the future cash flows of the company for the next 5-10 years, and then discounts these cash flows to the present time using a discount rate commonly referred to as weighted average cost of capital (WACC). These future cash flows are of two types: the expected cashflows to the shareholders (dividends) and the free cash flow to equity. Depending on the type of the company and other factors, the investor will choose to project one of these types of cash flows for their DCF model.
What we receive from the DCF model is the present value of the stock that represents the intrinsic value of the company. If the market price of the stock is above this intrinsic value, the stock is said to be overvalued, and if the market price is below this intrinsic value, the stock is said to be undervalued. Based on this, the analyst or the investor can base their BUY/SELL/HOLD decision.
One of the most important parts of a DCF models is the quality of projected cash flows. No one knows for certain what a company’s future cash flows will be. Investors make use of a variety of publicly available information about the company, including their historical financial statements, industry trends, competitive analysis, macroeconomic conditions, job market, and other such factors to arrive at the future cash flows. In a way, they arrive at projected revenue, capital expenditure, earnings, and finally free cash flows for the next few years. Investors can also make use of datasets available such as FactSet which aggregates data from various sources including the projected financial statements from various analyst reports.
The free cash flows can be arrived in different ways. Two methods are described below:
Using EBIT
One way to calculate free cash flows is using Earnings Before Interest and Taxes (EBIT).
We first arrive at EBIT by subtracting Operating Expenses from Revenue.
EBIT = Revenue – Operating Expenses
Then using EBIT, we can calculate Free Cash Flows as follows:
Free Cash Flow (FCF) = EBIT x (1 - Tax Rate) + Depreciation & Amortization (D & A) - Capital Expenditures (CapEx) - Changes in Net Working Capital
This is a common formula, but while looking at the company’s financial statements, the analyst may make further adjustments to it to account for any other items that may affect the company’s future cash flows, such as any non-operating items.
Using Operating Cash Flows
The other method is to calculate the free cash flows starting from the company’s operating cash flows.
Free Cash Flow (FCF) = Operating Cash Flow (OCF) - Capital Expenditures (CapEx)
In this, the operating cash flows are calculated as follows:
Operating Cash Flow (OCF) = EBITDA - Taxes - Changes in Working Capital
Once we have the operating cash flows, we can calculate the free cash flows.
Apart from projected cash flows for the next 5-10 years, another important input for a DCF model is the company’s terminal value. Terminal value represents the value of the company at the end of the projection period. At that point, we don’t make any detailed forecasts for the future cash flows but instead choose a steady growth rate for the company and use the cash flows arrived at using this steady rate. One method to calculate the terminal value is the perpetuity growth method, which assumes that the company’s free cash flows will grow indefinitely at a steady rate (g). The formula for calculating terminal value using this method is:
Terminal Value = FCFn x (1 + g) / (WACC - g)
FCFn is the free cash flows in the final year of our projection period, g is the perpetual growth rate and WACC is the weighted average cost of capital (explained below).
The perpetual growth rate is typically assumed to be between 2 and 5%.
To arrive at the present value of cash flows, these future cash flows are discounted using a discount rate that reflects the required rate of return for an investment. The most commonly using discount rate is the weighted average cost of capital. WACC is the average cost of all the company’s capital sources including the equity and debt.
WACC = % of equity * Cost of Equity + % of Debt * Cost of Debt * (1 - Tax Rate)
Cost of Equity = Risk-free rate + Beta * (Equity Market Return – Risk-free rate)
The equity market returns are the returns earned by investors from publicly traded stocks. Once of the sources for equity market returns is Prof Aswath Damodaran’s website.
DCF Calculation
Now that we have the projected cash flows for the next few years, the terminal value of the company at the end of the projection period, and the WACC, we can put it all together to calculate the DCF as follows:
DCF = (FCF1 / (1 + WACC)^1) + (FCF2 / (1 + WACC)^2) + ... + (FCFn / (1 + WACC)^n) + (Terminal Value / (1 + WACC)^n)
Note that we are also discounting the terminal value because the terminal value is at the end of the projection period.
How accurate are DCF valuations?
It is important to note that since DCF models are based on many estimates and assumptions, their accuracy heavily depends on the quality and reliability of these assumptions and estimates. Analysts and investors who are building the model are aware of this and therefore they don’t always just make one set of assumptions. Instead, they use sensitivity analysis to calculate DCF using different assumption and inputs. They create different scenarios with different values for key assumptions such as WACC, revenue, capital expenditure, growth rate, and margins, among other factors. This helps them get a better understanding of a stock’s value different conservative and optimistic scenarios.