Components of free cash flow

What I want to do in this lecture is talk about forecast drivers, or the assumptions required to forecast each component of the free cash flow formula out into the future. And we are going to do it by way of a specific example. So let’s get started.

So imagine we are a company in, say 2008, and we are considering building and selling a tablet. And so back in 2008, this was really just the start of the tablet market. So I’ve put our free cash flow formula here to highlight the components that we’re going to need. And remember the question is should we enter this market. Should we produce and sell this tablet? Well to make that decision, what we’re going to want to do is a DCF, and we’re going to want to forecast each component of this free cash flow formula into the future. So let’s do that. We’re going to start with revenue. So revenue is market size, how big the market is in terms of units, times market share, what proportion of the market we are going to capture, times our price per unit. So let’s look at our forecasts here. So I’ll start with the market size.

Okay. Now I’m going to assume that in the first year the market size is going to be a million units. That is going to be the early adopters, the techies, the people who are really excited about new technology and the introduction of new technology. So that is going to be one million units in year one. But that’s then going to grow quickly.

In subsequent years. And you might say, 2,500%, that’s ridiculous. And it’s an absolutely enormous growth rate for sure. But remember this is the beginning of a new market that we’re anticipating will take off in a big way. So 2,500% in the second year, followed by 128%, 9.4%, and 3.5%. Now you might wonder where in the world I am getting these. Well, I am getting them from a variety of different sources. I am getting them from my marketing people and my strategy group within the company. I am getting them from industry analysts as well.

So there’s a host of different sources of input that go into these numbers, but at the end of the day, they are just assumptions, and I’m going to emphasize that fact later on.

So I can use these two assumptions about the market to compute a market size in terms of millions of unit, right? Just 1 million in the first year. That is going to grow by 2500%, 128.0%, 9.4%, 3.5%. So that’s my forecast for the size of the tablet market. Now what is interesting, what I am going to show you in this next slide, is here the actual market size for tablets. And what you can see, even with a 2500% growth rate, I was conservative in my estimate, right? 26 million as opposed to 60 million in sales in the second year, followed by 116, 195, 229. So you, that’s sales. Sorry, not sales. The actual size of the market. So you can see this market took off incredibly rapidly. Next, I want to look at my market share. I am going to assume an initial penetration of 25% followed by annual growth of 5% each year thereafter. That is going to allow me to forecast my market share, which I bring down to 25%. That is going to grow at 5%, 5%, 5%, and 5%. So this is just to be clear, 25 % (1 + 5%). That will get you the 26.3%, and likewise for the subsequent forecasts. And finally, we have to come up with a pricing strategy. I’m going to price our tablet initially in year 1 at $200 per unit, which I just bring down here. I’m not going to experience any price change in the second year, but then I’m going to increase prices in the third year by just under $50. Now you might wonder what am I thinking? Consumer electronics prices go down, and go down fairly rapidly, but what I’m going to do is I’m going to build in some element of versioning here. So just like we had the iPad the iPad 2, iPad Air, I plan on doing some RND which I’ll discuss later, that’s going to develop new and better versions, lighter, more powerful, better screens, etc. So, that’s going to justify an increasing unit price over time.

Let’s move on to cost. So, I’m going to start with COGS, or Cost of Goods Sold, which I express as a margin here, or a percentage of sales. And I’m going to assume that’s going to run at about 80.66%. A number based on our company’s experience working in the computer industry for some time based on estimates from our operations people, and insights from other analysts and consultants.

My SG&A, sales, general and administrative expenses, are going to run about 1% of my current SG&A or $69.5 million. I am just making that number up, bear with me. So it’s going to be $69.59 million in the first year, and I’m going to assume that’s actually going to grow by 25% per year. Because hopefully as sales and the size of this market grows, I am going to have more overhead to keep up with that growth. Whereas I am trying COGS, which is loosely viewed as a variable cost, I am trying that forecast to sales.

Now, R&D, well I am going to have to spend $200 million dollars upfront on R&D to get this thing up and running.

And then I am going to have an additional $25 million every year thereafter to deal with versioning and improvements.

Now let us turn to our investment needs. I am going to assume an upfront investment of $227.7 million to build a plant, get any necessary equipment for production.

My first-year investment is going to be 10% of that initial investment And then subsequent years are going to be annual growth of 5% and the 1, 1, and 1. And you’ll see how this rolls out in the next lecture when we actually forecast those free cash flows, but these are just the assumptions. We are laying the groundwork here.

At the end of this project which is five years. So I’m going to assume, assuming that this is a five-year project. At the end of this project, I am going to have all of this plant, property, and equipment left over. Well, let’s focus on the plant and equipment. It is going to be there. It is not going to just disappear at the end of the project. So I’m going to assume I can do something with it. I could sell it. I could rent it out. I could redeploy it for another product. But I’m going to take a step back from precisely what I do with it, and just assume a liquidation value or evaluation for it that in which I get 50 cents on the dollar.

It is just an assumption.

Now for depreciation purposes, I am going to assume that all capital depreciates on a straight-line over five years. That is its useful economic life is five years. And so each year one-fifth of the capital, theoretically, disappears. Again, that is a non-cash expense, it’s not costing you anything, but it is going to provide a tax shield here that we have to recognize so we need to understand what our depreciation’s going to be.

Now on to networking capital. Recall net working capital is cash plus inventory plus accounts receivable minus accounts payable. Accounts receivable accounts payable.

Let’s go after each component here. I am going to assume this project is going to require a certain amount of cash, which might seem a little bit odd. But think of it as cash that I need to pay employees and on-going bills, such as utility costs, IT expenses, whatever it may be. And those requirements are going to be defined by 50% of my SG&A and 100% of my anticipated RND expenditures. Let’s clear that. My inventory forecast is based on inventory days or turnover. How long does it take to get inventory in and then out? I am going to assume that is 7.58 days, which by the way is very fast. You have to look to a company like Dell for that kind of turnover.

And finally, I want to recognize that at the end of this, I’m going to have some inventory left over, which I assume I can get 25 cents on the dollar. At that point, it’s going to be fairly obsolete. I might want to scrap it. I might want to sell it on some secondary market. Sorry about that. Some secondary market. So I’m going to assume $0.25 on the dollar for that remaining inventory at the end of this project. My accounts receivable are based on day’s receivable or how many days it is going to take until I actually get some money in following a sale. Because remember, a lot of sales are based on credit and likewise, my accounts payable are based on days payable, which is going to tell me roughly how long it takes for me to pay my suppliers. I might be paying them using trade credit.

Okay, so that is our working capital. But, remember what we want is the change in net working capital. So, excuse me, while these forecasts here, all of these forecasts here will get me my networking capital, what I am going to want to look at in free cash flow is the year on year change from year t-1, t, or t. And then the last piece of this puzzle is taxes. All right, I want the marginal tax rate. I want the tax rate on an additional dollar of earnings. I am going to assume that is 25.5%.

A lot of people will estimate this with an effective tax rate that is looking at the income statement, tax expense over pre-tax income. That is not strictly speaking what we want. To get at the marginal tax rate, you probably need access to the tax records of the company, which are difficult if not impossible to get. So a lot of people will also assume a top statutory corporate tax rate of around 35, 36%, okay.

All right. So what we’ve done is we’ve gone through, let me just wrap up a little bit here. We have gone through and we have produced or made assumptions about every component in our free cash flow formula.

And all of those assumptions are going to allow us to forecast dollar values into the future.

But let us take a step back.

This is nonsense that is a common look I get from students or practitioners when I discuss this framework because the argument is, look, this is impossible to make accurate forecasts in the future. It is hard enough to figure out what our company is going to do, and how the economy is going to change, and how that is going to impact our company, a quarter out, much less five years and I agree, but I think that’s not the point. Let me be clear on what I agree, it’s not that this is nonsense, but I do agree that it’s very difficult, if not impossible to make accurate forecasts, but that’s not the point. That is just not the point of a DCF. The point of a DCF is real to focus the discussion and analysis of the relevant issues. To get away from decision-making based on gut feelings, what the stars look like, whether it is a full moon, other ad hoc rules of thumb, just referring to nebulous experience, and I know best, or I have been at the company for 20 years this is the way it is. DCF provides a rigorous framework within which we can discuss what really matters for value creation, and that is the whole point, that is what I want to emphasize. The goal of a DCF is not to get one number that we argue is correct. The goal of a DCF is actually to provide a host of numbers and a host of sensible and financially correct information from which we can make better decisions, and I am going to emphasize that throughout.

Another lesson I want to emphasize here is that no successful valuation or DCF can rely solely on input from financial personnel. It is critical that we get input and insights from all aspects of the company, or across division lines. I mean think about what we are forecasting. We are forecasting revenues, what is going on with the market. So we want to talk to our marketing and strategy people. We are going to be talking about investments and operations. We need to understand what our inventory is going to be from our operations people, from line managers, even from lower ranking file employees. That is how you build a successful evaluation, that is how you build a successful DCF model with really good inputs from knowledgeable people.

All right, let’s wrap this thing up. So today we discussed forecast drivers in the context of a specific example, that being a tablet. And these are nothing more than the assumptions that we’re going to use to forecast the dollar values of each component of our free cash flow formula. And again, the goal isn’t to get the right answer, it’s to provide that framework. Within which we can discuss sensibly and logically the relevant issues for making decisions. So in the next lecture what I want to do is I actually want to use these forecast drivers to forecast the free cash flows, the dollar values.