# Forecasting free cash flows

Today, I want to apply the forecast drivers to the free cash flow formula to forecast free cash flows into the future for the tablet project. Today we are going to be talking about forecasting free cash flows. Today I want to apply those forecast drivers to actually forecast dollar cash flow. Dollar free cash flows over the 5-year horizon for our tablet project. So let’s get started.

So here’s our free cash flow formula, and again the goal today is to translate the forecast drivers into dollar forecasts. So as a brief reminder, here are our revenue forecast drivers. We are going to start by forecasting the dollar revenues.

And we know that revenue equals market size, times market share, times price.

So for year 1, our revenue forecast is just going to be the year 1 initial market size, one million units, times our share, we’re going to get 25% of that market. And then we’re going to multiply it by our per unit price of $200 for a revenue forecast in year 1 at $50,000,000.

If we repeat that process for years 2 through 5, we will get our revenue forecasts for all 5 years, and there they are.

Now let’s move on to costs. And we’ll start with COGS, **Cost Of Goods Sold**. You can see that our forecast drivers here are expressed as a percentage and in particular a percentage of sales. So I’m going to repeat the sales forecast from the last slide here because we’re going to need them to get a COGS forecast. So if we look at year one, we can see that 80.66% of the sales are assumed to be the cost of goods sold. So, we take our $50,000,000 revenue forecast, multiply it times that 80.66%, and out pops an estimate or a projection of costs of goods sold in the first year of $40.33 million. If we repeat that process through years 2 through 5, we are going to get a full-blown COGS forecast over the entire projection period.

Now let’s move on to SG&A, still in the costs category. Here we have assumed the first-year SG&A expense of $69.59 million. Now it is in dollars already so there’s not really anything to do. It was just an assumption I made based on 1% of the 2008 companywide SG&A expense. A number I essentially made up just for illustrative purposes. But going forward in years 2 through 5, we’re going to assume that SG&A expense grows at 25% per annum. So to get year 2 for example, we start with our $69.59 million expense in year 1 and we compound it up by 25%, to get an estimate of $87 million for our second-year SG&A expense. And if we continue with that 25% annual growth, we’re going to get an actual SG&A series here.

Finally moving on to R&D, there is really not much to do here. Because I have assumed dollar forecasts for both the upfront R&D necessary to get the project off the ground, as well as the subsequent R&D required for any versioning. So here are our R&D dollar forecasts.

I am going to take a step back and put this together into a somewhat familiar format. And we’re going to start with the top line, sales forecasts, right here.

I am going to subtract off my COGS to get at a gross profit. I am then going to subtract off my SG&A expenses and my R&D expenses to get estimates of EBITDA. And EBITDA is a mouthful once you say it. It’s Earnings Before Interest Taxes Depreciation, I’ll abbreviate that, and Amortization.

Okay, now let’s move on to capital expenditures. So, our upfront investment is going to be $227 million. Well, a little over $227 million, for the plants and the equipment. Then we are going to invest in the first year of the project, 10% of that amount. Followed by annual growth of 5%, 1%, 1% and 1%. So to get our dollar forecast there’s nothing to do for year zero, that’s just the $227 million.

For year 1 we’re going to take 10% of that $227 million to get $22.77 million. And then for a year, let me clean that up a little bit, for year 2 we’re going to grow that at an assumed 5%. So we compound up the 22.77 million at 5% to get an estimate or forecast of capital expenditures in year 2 of 23.9 million dollars. And continuing that growth process for years 3, 4 and 5 at 1%, we’re ultimately going to arrive at our projected capital expenditures series. There it is.

Now, we’re going to assume we’re going to straight line depreciate this capital expenditures over 5 years. What that means is one-fifth of the capital stock is going to depreciate each year.

So what I am going to do to make things a little bit easier is I’m going to create a row of accumulated capital expenditures, right? So that’s nothing more than the current plus all previous capital expenditures. So this 250.5 is the 227 plus the 22.8. The 274.4 is the 250.5 plus the 23.9 and on and on and on. And the reason I’m doing that is to avoid keeping track of different vintages of capital stock. Since they all face the same depreciation schedule, at least by assumption. And what I can do to compute the depreciation is simply divide last years accumulated depreciation, sorry accumulated capital expenditures, by 5. So, the 45.5 in year 1 comes from 227 divided by 5. Year 250.1 comes from the 250.5 divided by 5. And again, all we’re doing is, what’s really going on is, this capital stock is depreciating by another 5, another 20%. And then this depreciates by 20%, so we add those two to get the 50.1. And we continue that on, and on, and we’re left with, we get our depreciation series. Now here is a question. **What happens to all of this physical capital at the end of the project**? At the end of the 5 years? It certainly does not evaporate. Right? It does not disappear. We can sell it, or we can redeploy it for another purpose. So we need to recognize that. Otherwise, we are going to underestimate cash flows. So, what I’m going to do is I’m going to take that, all of that capital expenditure, that Accumulated Capital Expenditure and I’m going to subtract from it the Accumulated Depreciation. That is, I am just going, to sum up, all of the depreciation.

To get this 274.8. The difference between the Accum CapEx and the Accum Depreciation is the Book Value of the assets for $72.8 million.

Now I am going to assume that I cannot sell that on a dollar for dollar basis, rather I am going to have to sell it at a discount. $0.50 on the dollar and so the liquidation value, the money I can actually get for it, which is different from the book value is assumed to be $36.4 million. But remember we have to deal with taxes and so what we’re really interested in are the after-tax proceeds from selling this. So we’re going to get $36.4 million. But we’re going to experience a book loss because the value for which we can sell the assets, by assumption I might add, this isn’t always the case, is less than the book value of the assets. I multiply that times the tax rate. So we get a little bit of a tax shield here from our loss. And the result, our after-tax proceeds that are actually greater than the liquidation value, we’re going to get $45.6 million. Therefore, the end result is that we have our projected capital expenditures, here, but once I factor in the after-tax proceeds from the sale of these assets. We are actually going to have negative capital expenditures in that last year, that fifth year. Which is going to reflect an inflow of money. So now let us turn to net working capital, more precisely the change in the net working capital but we’ll start with a net working capital. Which is, you remember is Cash + Inventory + Accounts Receivable- Accounts Payable.

And here where all of our assumptions or our forecast drivers for the components of the net working capital. Let us start with the top, the cash requirements. We are going to require 50% of SG&A in cash and 100% of our R&D expenditures, so we are going to need our SG&A forecasts and our R&D forecast to back out the cash requirements. Notably, for year one, we are going to need 50% of our SG&A expenditures or $34.8 million held in cash. But we’re also going to need 100% of our R&D expenditures, 100% of the $25 million, or $25 million, we’re also going to need to hold that in cash. So our cash requirements are the sum of these two and are given by this row. And again we’re just going to repeat that, sorry, year by year for two, three, and four.

Turning to inventory, our forecast drivers lay out the inventory days but the inventory days are based on cogs so I’m also going to need my cogs series which I put here for convenience. So to compute the first year inventory requirements, I’m going to take my days in inventory 7.58. I am going to multiply that by my COGS expenses. And I’m going to divide that by 365 to get an estimate of first year COGS at $837,000.

If I keep doing that, year after year, we will get our inventory requirements.

Turning to accounts receivable, we have days receivable assumptions, those are based on sales, so here’s my sales, forecasted sales. When I apply the days receivable to the sales, say for year one for example. I’ve got 38.49 days receivable times the $50,000,000 divided by 365 days gets me a forecasted or projected accounts receivable of $5.2 million, a little over $5.2 million at the end of year one. We continue that process for years two through five.

There are our accounts receivable projections.

Finally, we turn to accounts payable. I’ve got days payable assumptions of 61.54 days. This is based on COGS, so here’s my COGS series for convenience. I apply my day’s payable forecast to our forecasted COG series, and out pops the accounts payable forecast.

Repeating that process again years two through five. We get our accounts payable. There is a typo. Accounts. We get our projected accounts payable series. And we can put this all together, our cash inventory accounts receivable and accounts payable forecast to compute net working capital for each year. Remember it is going to be cash plus inventory, plus accounts receivable minus accounts payable. That gets us our net working capital, but here is a question. What happens to all this working capital here at the end of the project? Specifically, what happens to all the cash that is sitting there, or the inventory or the accounts receivable, the customers from whom we are waiting to receive payment. Or the accounts payable, the suppliers who are awaiting our payment? What happens to all that net working capital? Well, like physical assets, it does not just disappear, in fact, most of it is going to be recovered, not all of it and I will explain why in just a second. In particular, we are going to be able to recover the cash; we are going to be able to recover some money from the inventory. But by assumption, back in our forecast drivers, we’re assuming we’re only getting 25 cents on the dollar, for this inventory. That is probably obsolete, and either not worth much in terms of scrap, or on some secondary market. We are going to have to pay, well excuse me, well we are going to have to pay all of our accounts receivable, but we are also going to have to collect, sorry. We are going to have to, let’s get rid of that, we’re going to have to pay our accounts payable and we’re going to have to collect our accounts receivable. The upshot of this is we are going to recover $51.375 million in net working capital at the end of the project.

So our change in net working capital, remember we are interested in the year on year change is given by this row. Let’s just briefly discuss it. Get rid of that guy. So, the first change is 59.1 minus zero which is 59.1. The second change is the 49.7 minus the 59.1 gets me minus 9.4. And on and on and on. And in the last year we would have 25.6 minus the 28.2, but then we’d have to add back in the 51.3, to get net working capital of 48.6, which reflects recovering all of that working capital at the end of the fifth year. Now, there is a bit of an assumption, well, there’s a lot of assumptions, but one of the assumptions here is that we were covering all of the working capital right at the end of period five. In reality, you are probably going to collect that within or during year six. So, I could have extended this to the sixth year, but it’s not going to make too big of a difference. The discounting effect is going to be small. So I just wrap it all up in this fifth year here.

We have all the pieces now. Right? And what I want to do first is organize them into a useful and familiar format, what I’ll call a quasi-income statement. And it’s a quasi-income statement because it’s missing an important item and its applying taxes slightly differently. I will explain that when I get there. So we start top line with sales, subtract off cogs that get us gross profit. We are going to subtract off SG & A, subtract off R & D that gets us our EBITDA. And sorry ignore these numbers here. What they are, they are references to rows in the spreadsheet, and they are just a useful way to keep track of things.

We are going to pull off our depreciation that is going to get us our EBIT. We are going to apply our tax rate to our EBIT to get our tax expense, which we subtract from EBIT to get our NOPAT.

Now I call this a quasi-income statement because what is missing here is interest. No interest. That is financing. We are after unlevered cash flows so that is not relevant here. It is also quasi because the taxes are applied to EBIT as opposed to pre-tax income, which comes after interest. All right. So this is our NOPAT series.

Let’s carry that NOPAT forward up to here.

Add back in depreciation, subtract off capital expenditures and subtract off the change in net working capital to get what were really after or ultimately after which is free cash flows. These are the free cash flows to the project.

Now some other things to keep in mind when you are going after free cash flows. First of all are Opportunity Costs, alternative uses of resources. Right? They are not free. Second, Project Externalities. And this is a biggie In our context you can imagine that the tablet might cannibalize or take away sales from other product lines, maybe desktops, for example. You also want to be cognizant of spillovers. You could imagine that in the process of developing and selling this product, this tablet. We might learn something about our production process that applies to other product lines or it might lead us to the development of an entirely new product line. We need to recognize the Externalities that are associated with this project. Sunk Costs, ignore those. What do I mean by Sunk Cost? Imagine we commissioned a marketing study back in 2006, 2 years before we were valuing this project or deciding on this project. The costs associated with that marketing project as of 2008 are completely irrelevant. They are sunk. They are irrelevant for decision-making. All that matters are the costs going forward and benefits.

Other non-cash items such as amortization, we have to consider those.

Particularly because that generates a tax shield, but also other non-cash items like stock-based compensation, stock grants, restricted stock awards, options, employee stock options and the like.

Salvage values. We touched on this in this lecture, right? Assets do not just disappear and neither do liabilities. So we have to take into account any salvage values or liquidation values associated with our assets.

Execution risk. Look, this is a risky project, but the way we are going to capture execution risk, that is risk unique to this project, is not through the discount rate. And I’ll come back to talk about that later on in the class when we get to estimating the discount rate. This is specific to the project, so it has to affect the free cash flows, and the way it is going to affect that is we’re going to look at expected free cash flows. That is the way to think about the numerator in the NPV calculation. And we can capture this through sensitivity analysis by looking at different scenarios, different forecasts for free cash flows. That is going to come up in a little bit when we talk about sensitivity analysis. And one last point is I’ve assumed an annual frequency here but there’s nothing special about that. We could have done this at a quarterly frequency. We could have done it on a monthly frequency, semi-annual, take your pick. What determines the cash flow frequency is really the situation, it is project dependent. And it is sector dependent, so don’t think there’s anything special about annual frequency. We know how to deal with non-annual frequency cash flows and discounting them, so it is not a problem at all. All right, let’s summarize this. What we did in this lecture has we actually forecasted the dollar values of the free cash flows, and the way we did it is we built it up from our forecast drivers of the components within free cash flows. This is one of the two basic inputs into a DCF, the other being the discount rate. And the question now is what do we do with these free cash flows? Well, that is the topic of our next lecture where we take a look at decision criteria. So thanks for listening and I look forward to seeing you in the next lecture.