ñòð. 135 |

else. This means that when senior managers study much of the sensitivity

data on a project their primary concern should be to establish that the base

case is an expected value rather than to establish what extra return is neces-

sary. Things which are unique are diversifiable and hence command no add-

itional risk premium.

433 First view: The cost of capital

By contrast, the point about the Î² factor is that we have seen how leverage

can have a major impact on the equity Î². If a business is financed half by debt

and half by equity then its equity Î² will be twice its asset Î². So the premium

for risk on equity will be twice the premium for risk on the unlevered assets.

Debt is not the only way to lever assets and in any case we already exclude its

effect from our analysis. Operating leverage, caused by the different balances

between fixed and variable costs, also causes leverage and this is not excluded

from our analysis.

So to make a good risk/reward decision, senior managers need in their

mind a picture of the typical level of operating leverage in what would be a

normal project within their company and for which the usual CoC applies.

This can set the benchmark for the judgement about relative risk, and hence

required return, in other projects.

Topic 2: Treating financing separately

The work required to assess the CoC highlights not one benchmark for

assessing financial decisions but three. These are the CoC, the cost of debt

and the cost of equity. Now most of the long-term financial decisions which

are taken in a company need to be analysed at the CoC. There are, however,

some situations, which mainly have to do with financing, that call for the use

of one of the other two rates.

The idea of treating financing decisions separately from asset decisions

was introduced at an early stage in this book. It can be justified in a fairly

simplistic way by saying that the CoC already includes the cost of finance and

that it would therefore be wrong to include finance effects in our estimates of

project cash flow. We can now look at financing in a more sophisticated way

as part of the consideration of a projectâ€™s overall risk characteristics and the

necessary risk/reward trade-off decision. Since we now understand how even

individual parts of a project can require the use of a CoC which is correct

given their exact risk characteristics, we should be able to understand better

why financing is treated separately. If one is faced with a situation with the

risk characteristics of debt, or equity for that matter, then one knows that the

right discount rate to apply would be the cost of debt or the cost of equity.

Of these two potential situations the need to use the cost of debt for situ-

ations which have the risk characteristics of debt will be the most important.

This is because most companies will utilise in their finance pool more equity

than debt and so the difference between the cost of debt and the CoC will be

434 Three views of deeper and broader skills

greater than the difference between the cost of equity and the CoC. This topic

will focus mainly on situations where the cost of debt must be used but it will

also finish with a short analysis of when the cost of equity is used.

A key characteristic of debt is that it carries no, or very low, risk to the

lender.35 This means that the correct CoC for assessing flows which are risk-

free is the cost of debt and not the average CoC. What would be the implica-

tion of this for usual project evaluation?

Suppose that included within a set of project cash flows are a set of fixed

cash flows. Would this serve to increase or to decrease NPV? Well, the answer

depends on whether the flows represent borrowing by the company or lend-

ing. Borrowing would be where there was an initial cash inflow followed by

subsequent cash outflows. Lending would be the opposite. Since the cost of

debt must be lower than the CoC, borrowing by the company will, if it is

assessed at the CoC, always appear to have a positive NPV while lending will

appear to have a negative value. This is best illustrated via a simple example.

Suppose that at present our overall project cash flows are as follows:

Yr 1 Yr 2 Yr 3 Yr 4 Yr 5 Yr 6

Cash flow $m âˆ’100 30 30 30 30 20

Discount factor (10% CoC

mid-year flows) 0.953 0.867 0.788 0.716 0.651 0.592

Present value cash flow âˆ’95.3 26.0 23.6 21.5 19.5 11.8

Cumulative PV âˆ’95.3 âˆ’69.3 âˆ’45.7 âˆ’24.2 âˆ’4.7 6.9

The project appears to have an NPV of $6.9m.

Now, however, suppose that on further analysis of the project cash flows we

identified that they included some flows which were completely fixed and not

uncertain like those which typically make up project cash flow. I will explain

what these flows might be later but for the present, please accept that the

finance flows are equivalent to borrowing $11m in year 1 and repaying $2.5m

in each of years 2â€“6. The implied cost36 of this borrowing is just 4.5%.

If we follow the correct approach to calculating value and exclude financ-

ing flows the correct cash flows and the valuation should therefore be as

shown in the following table:

The only uncertainties would be if the interest rate was floating and the possibility of default by the

35

borrower.

Note that since the table of cash flows is used to calculate NPV, one should assume that the cash flows

36

are after-tax. The 4.5% cost of borrowing would therefore be the after-tax cost of borrowing.

435 First view: The cost of capital

Cash flows adjusted to exclude finance effects

Yr 1 Yr 2 Yr 3 Yr 4 Yr 5 Yr 6

Cash flow $m âˆ’111.0 32.5 32.5 32.5 32.5 22.5

Discount factor (10% CoC) 0.953 0.867 0.788 0.716 0.651 0.592

Present value cash flow $m âˆ’105.8 28.2 25.6 23.3 21.2 13.3

Cumulative PV $m âˆ’105.8 âˆ’77.6 âˆ’52.0 âˆ’28.7 âˆ’7.5 5.8

The NPV has now fallen to $5.8m. Now for many purposes the difference

between an NPV of $6.9m and one of $5.8m on a project involving a cap-

ital spend of $111m would not be considered material. If, however, one was

trying to get the best possible view of NPV it might be considered to be worth

adjusting for. The decisions where the difference of $1.1m would really matter

would be where we were considering just the effect which caused us to decide

that the small adjustments to cash flow were necessary.

I will now build on this example by giving a detailed illustration of how

financing can become intermingled with project cash flows. Suppose that our

project involved the construction and subsequent operation of a production

facility. The product is sold packed in special containers which could easily

be loaded onto trucks. The normal cost of a packing unit is $11m. Our usual

distributor has, however, offered to purchase a packing unit for installation

at our site for our exclusive use. The deal which the distributor has offered

is such that we must pay the cost of operating the packing unit plus a capac-

ity reservation fee of $2.5m per year. Finally, we must give the distributor

exclusive rights to distribute our product provided his costs are in line with

market rates.37 Our initial base case (with a cash outflow of $100m in year 1)

assumes that we accept this offer.

Were one to analyse this opportunity using a traditional DCF spreadsheet

model the temptation would be to compare as shown in the two tables above.

The only differences would be in the assumed capital cost of either $100m or

$111m and also in the payment of the capacity reservation fee. The operat-

ing costs and distribution costs would be the same in either case. Hence, as

has been shown above, the offer from our distributor would appear to create

$1.1m of value. This, if it were a piece of correct analysis, would give a good

reason to accept the offer. The problem with the offer only becomes apparent

Tax is always a complication! For the remainder of this example I propose to ignore tax in order

37

to simplify the numbers. Strictly speaking I should have excluded the after-tax impact of the

finance flows.

436 Three views of deeper and broader skills

ñòð. 135 |