ñòð. 155 |

every two years for the next ten years?

These should not be questions to be answered in a vacuum. The answers

should be guided by the strategy which had your company identifying the

initial opportunity in the first place. The strategy should have suggested what

the long-term potential was. Crucially, the strategy should also have identi-

fied the sources of value which would be exploited in order to gain the value.

It is these sources of value which should be used in order to give a reasonable

estimate of what the future NPV might be.

501 Second view: Valuing flexibility

The technique of using specific sources of value to obtain a rough estimate

of NPV was illustrated in the Yellow Moon Chemical Company example in

the pillar on Sources of Value starting on page 304. If one follows the Sources

of Value approach, one can create strong links between the strategy and the

economic analysis of the subsequent projects that follow from it. A kind of

virtuous circle can be created because the early identification of sources of

value will make the claims of growth value at the time of initial market entry

all the more credible.

The second step in assessing the growth value of market entry is to esti-

mate when the future NPVs might come about. We need to know this because

our overall valuation will use the present value of these future NPVs. So, for

example, if we decide to model the future growth as five projects, each with

an NPV of $100m at the time of approval, and with approval being at the

start of years 7â€“11, the NPV today (with a 9% CoC) would be:10

$100m Ã— (0.59610 + 0.547 + 0.502 + 0.460 + 0.422) = $253m

I would call the final $3m spurious and say that the growth value was

$250m. The idea is that the estimated growth value should be a reasonable

mid-point estimate of what the value might be if the project or projects go

ahead.

The final step concerns probability. There is no certainty that these projects

will go ahead as planned. One needs to decide on a reasonable probability of

the projectsâ€™ going ahead. When one is dealing with just one project it is not

too unreasonable to ask what the probability of making the follow-on invest-

ment might be. One could be guided to some extent by the strength of the

financial case which can already be assembled. If our Sources of Value ana-

lysis is already pointing towards a strong positive NPV then the chance of a

follow-on project is quite high. If, by contrast, our current outlook is that the

follow-on project has a negative NPV then the chances of its subsequently

turning out to look good will be much lower. Companies with a track record

of setting up project development teams might even have data on the number

of teams that have been set up in the past and their success rate.

With multiple projects it might seem a bit more difficult to come up with

a single probability and one might be tempted to reduce the probability of

each projectâ€™s going ahead as one moves into the future. I suggest, however,

0.596 is the year-end discount factor for six years at 9%. This is appropriate since the NPV quoted will

10

be as at 1 January of year 7.

502 Three views of deeper and broader skills

that since the estimated value prize if the project goes ahead is already, in

principle, a mid point of a range, one should aim just to come up with a sin-

gle probability of this coming about. So, let us assume for our Indian market

entry example a 40% chance of success. This would point to an overall flexi-

bility value of $100m.

The advantage of this approach is that the assumptions should have strong

links back to the strategy of the business concerned. Furthermore, the meth-

odology has strong links to the established risk monetisation approach. I

would contrast this with the approach which utilises a Black-Scholes calcula-

tion. This will assume that value comes from volatility and hence that any

volatile situation offers the potential for value creation. Although I fully agree

that a volatile share will justify a higher value for its options than a stable one,

I do not accept that real projects can be assumed to benefit in a similar way.

If one accepts that value is rooted in sources of value, then flexibility value

too must be justified in this manner and not automatically be available to any

project that faces volatility before the final decision to proceed is taken.

So growth value is assessed via the risk monetisation approach with the

impact of growth assessed through an extrapolation of Sources of Value logic

and the probability too guided by the strength of the sources of value.

Step 8D: Valuing the tail

This is an entirely different technique which I have devised particularly for

use in many optimisation situations. It does, though, have many potential

uses. It concerns situations when one needs to understand the economic con-

sequences of gaining or losing access to the edge of a statistical distribution.

It is easiest to explain via a simple example.

Suppose that an oil company has struck oil. It knows that it has found a

large field and the geologists have estimated that annual production will be

100,000 barrels per day (100 kb/d for short). The oil is, however, a long way

from the market and so the company will need to build a pipeline in addition

to the oil production facilities. How big a pipeline should the oil company

build? One might think that the answer was obvious. You have a 100 kb/d

oil field so you build a pipeline that will have a throughput of 100 kb/d. Life,

however, is more complicated than one might think and there are two par-

ticular reasons why it might be worthwhile building a bigger line.

These situations are caused by the uncertainty which surrounds the esti-

mated annual production from the initial oil field and also the possibility

that other oil fields may be found that could also use the line. I will deal first

503 Second view: Valuing flexibility

with the uncertainty surrounding the initial oil field because this is where

the technique of valuing the tail comes into play. I will show how to use the

growth value technique to value the possible income from other oil fields

after that.

It is always important to question the basis of any number which one is

given concerning the future. So, in this case what one should do when told

that a 100 kb/d oil field has been found is ask how sure we are about the

production forecast. Is it, for example, an expected value, a conservative esti-

mate or the best you can possibly get? Readers will, by now, be used to my

preference for the use of expected value estimates. So, let us assume that the

figure is indeed an expected value. It is the expected value of what the oil field

could produce given a range of uncertainties concerning the size of the oil

reservoir, oil recovery rates and the number of production wells which will

be drilled. Does this mean that it is safe to use this number in our financial

model? The answer is that it is only safe to do so if the pipeline is built big

enough to handle the highest possible oil output that the field might produce.

If, however, the stated output is subject to a range where the top end cannot

flow through the line, the correct expected value of oil which is delivered

to market will be lower than 100 kb/d. The oil company will, in effect, lose

access to the top end of the statistical distribution of possible oil production

rates. This is illustrated in the following diagram:

Probability

Initial view:

Production potential

for oil field

100 kb/d

Production

Oil field: actual output

Effect of constraint:

Peak production lost

Expected value falls

100 kb/d Pipeline

Production capacity constraint

Fig. 12.3 Valuing the tail

504 Three views of deeper and broader skills

ñòð. 155 |