ERisk interview

Vince:

You may inteersted in the following interview which appeared on ERisk.com last Friday. Were Rick Buy's comments about real options taken out of context?

Yann Bonduelle leads a 25-person team for PricewaterhouseCoopers in London
that applies decision analytics and real options theory to dilemmas ranging from
valuing a biotechnology product to deciding whether to kill off an Internet financial
services business. Here he talks to Rob Jameson about whether this "theoretical"
approach to risky decision-making really helps businesses in their day-to-day
balancing of risk and reward. Yann holds a Ph.D degree from the
Engineering-Economic Systems Department at Stanford University, where he
studied how to apply engineering decision and design analysis to wider economic,
social and business issues. He then worked as a consultant applying his decision
analysis methodologies to problems that included consumer decision making
about innovative products such as electrical vehicles, before joining the
PricewaterhouseCoopers team in 1998 where he is now a partner. He has
written widely on the application of real options, particularly in fields of life
sciences, technology, and e-business, and has a special interest in the
relationship between risk assessment, validation of risk data and financial
valuation.

How would you sum up your approach to business decision analytics?
Most of our projects are set up to help businesses that face massive uncertainties
of some kind. Decision analysis helps people explore problems, and redesign
their decision-making process to increase the chance of them making the right
choices. For example, imagine a biotechnology company that has to decide
whether to put itself up for sale, enter a strategic relationship, or continue to go it
alone. Each of those options will lead on to other value-enhancing or
value-destroying scenarios. We work with the client firstly to understand and
challenge the assumptions associated with their most likely business
development scenarios, and secondly to help them identify decisions that would
help protect or increase the value of their technology or company. Quantifying
technical, regulatory or commercial risks can sometimes be a challenge. In
technology-intensive fields, however, we have found that managers (often
scientists) are quite willing to describe the main sources of risk and to assess the
probability that a risky event may or may not occur.

How does this kind of risky decision-making relate to real options valuation?
You can't say what the value of an asset is until you decide what you might use it
for. This means that, to form an opinion about the value of an asset, you must
explore the most important decisions that you are likely to face and that will have
a significant impact on the value of the asset. So decision analysis helps to define
the business problem and to uncover a stream of inter-related choices that are, in
effect, "real options". For example, if a company is trying to decide whether to
invest in a risky project, does it have the option to pull the plug on the investment
at an early stage if a pilot project gives a poor showing? That "real" option reduces
the riskiness, and increases the potential value, of the original business plan. So
real options and decision analysis are really very close to one another. But you
don't have to believe in real options valuations to find decision analysis useful.

What do you mean?
Often decision analysis can help managers to identify the key risks in a strategic
decision, attach weights to these, and show clearly how they interact. For many
companies this "risk discovery" is the most valuable part of the exercise.

Real options theory has been criticised recently for being, well, not very realistic.
Is it a practical approach to valuation?
It's important not to hold out unrealistic hopes for the real options approach to
valuation. But it's an exciting methodology, and it's also sometimes the only
reasonable way of tackling a very practical problem. For example, when a firm
sells an asset, the firm might have to make an independent valuation of the asset
for legal or corporate governance reasons. But in many businesses today there
are assets that simply cannot be valued in traditional ways because they are
difficult to link to cashflows. The cashflows might not exist because the business
is so novel, or they might be hidden. In some respects, a real options analysis is
much closer to reality than a traditional valuation.

How, exactly?
The classic way of valuing a future business is to base the calculation on a single
discounted cashflow that is projected from the activity. But this doesn't really take
account of the way that scenarios can change, or the fact that managers can
react to situations as they unfold. I mentioned earlier the option to kill a project or
business at an early point. But the upside is that if a pilot project yields exciting
results, it might allow you to invest more quickly and reach a revenue-generating
position in a much shorter time than the original business plan allows. So to value
a future business we really need to look at the cashflows that might arise in a
number of scenarios. This is "realistic" in that, if the project gets the green light,
you can bet that its managers will be taking that kind of decision on the ground all
of the time.

What's the most challenging part of mapping out a decision analysis tree?
Modelling the links between the variables in the decision tree -- it's something we
have particular strengths in. But it's also tricky to know when it's worthwhile to
add on more detail, and when it's better to draw back.

In a recent ERisk interview, Rick Buy, chief risk officer of Enron, said that over
the two years that Enron had experimented with the real options concept, it had
found it of "limited, but not zero, use". Why is there a slight air of cynicism about
real options in some businesses today?
It's strange that Enron would profess this attitude. A few years ago, it was widely
reported to have used real option valuation to support a very profitable purchase
decision. They had apparently bought cheaply some older generators in the US
that generated electricity at a very high cost. They knew that they could mothball
them for most of the year, and switch them on only when the electricity prices
were sufficiently high. Nevertheless, from a customer's point of view, there might
have been too much hype about the methodology. One problem in the application
of real options technology is that there are, perhaps, too many people trying to
tweak reality to conform to their "perfect" model. It's better to aim for something
pragmatic that clearly improves decisions over time. In one pharmaceutical
company we worked with recently, we worked together to improve their valuation
analyses by moving from a single discounted cash-flow methodology to one that
took into account a rather small set of business scenarios. It would have shocked
some academics and consultants, but it was an undeniable improvement on the
original approach.

Why do you think financial institutions are only just picking up on your field, when
it's been applied in the energy industry for 15 years or more?
It might have something to do with the relative stability of the banking world until
recently, and the relatively high margins that banking lines have enjoyed. Also,
industries such as energy and pharmaceuticals tend to have more people with an
engineering and science background. The dynamic modelling of decisions is
based on methodologies originally dreamed up to help engineers design electrical
and electronic systems. This approach is quite distinct from the Black-Scholes
options analyses that the banking world is familiar with: the Black-Scholes
approach is difficult to apply in a real options context, because everything
depends on the assumptions that you put into the Black-Scholes model. The real
options approach, on the other hand, is in a sense a way of modelling those
assumptions more explicitly. But banks are now adopting some of the thinking,
particularly in terms of using decision analysis to pinpoint risks and identify
value-enhancing decisions, and in using real options methodologies to sort the
wheat from the chaff in their more speculative investments.

You mean their Internet investments?
We have recently worked with a major Dutch bank that had arrived late in the
Internet game, and then made a considerable number of investments. Now that
even B2B business models have questions marks hanging over them, and many
B2C businesses are already under water, they wanted to work out which
investments might contain real value. In this situation, it's a case of ranking
priorities and helping the bank make sense of what could turn into a
decision-making chaos, rather than sophisticated valuation. It's not just a case of
whether an internet investment should be killed off, but the problem of whether
continued funding for it should take priority over budget demands for major IT
upgrades in existing businesses, and so on. These are very practical questions
and they have to be answered somehow.

Are there other areas in financial institutions that seem accessible to this
approach?
Yes, for example, we think it can help work out the value associated with various
approaches to marketing a new bank business line. At the moment, many banks
are chasing high-net-worth individuals, but it's not always clear which kind of
individual a particular bank should decide to pursue. The bank might have a
regional or industry advantage already in one particular area, for example, music
business people. But what is the churn rate associated with this kind of
customer? What is the profitability associated with the customer segment? Will
the time and cost benefits of the advantages the bank has in the sector outweigh
any disadvantages? Weighing up this kind of complex problem, where one thing
leads to and depends on another, is what decision and real options analysis is
good at.

Is there any way of rigorously backtesting or validating real options valuations?
In all honesty, not really. The problem is that by the time the option is exercised,
many of the variables surrounding it will have changed, so it's difficult to compare
our original analysis with how things turn out. However, the value of the analysis
comes not only from the final number ("the value of this asset is X") but also from
providing a thorough process, an outsider's point of view, an understanding of the
sources of value and, in short, a bit of clearer thinking.

If real options are so important, why are they so rarely cited in communications to
shareholders and equity analysts?
The battle is still to convince companies to use real option valuations as a
significant part of their internal analysis. Even in major companies in the oil & gas,
and pharmaceutical sectors, where the ideas have taken some root internally,
there seems to be a lot of reluctance to use them in external communication. We
are working with analysts to understand better what they need to if we are to move
things on to the next step.

How does the riskiness of a business, in terms of the major strategic dilemmas it
faces, relate to its share value, and its capital structure?
That's a big question. It's related to work my colleagues do on the optimal
debt-to-equity capital structure and gearing of a corporation, which in turn arises
out of the likely revenue and cost volatilities of the business. The more volatile the
business, the less gearing it can sustain, and the higher the cost of capital. Our
work touches on this in the sense that exercising (many) specific real options can
allow a firm to change its nature, and thus also its risk profile. One classic
example is the pharmaceutical industry. A host of different kinds of companies
service that industry from "big pharma" companies through to smaller
biotechnology startups and run-of-the-mill contract research organisations. A
contract research organisation is often operating within a very competitive
environment with relatively few risks--it does not invest in drug development itself
-- but also very thin margins. But in fact, a few of these companies have used the
skills and knowledge they have developed to become much more substantial and
profitable healthcare companies of various kinds. It's an example of a company
exercising the real options that lie within its skills and assets to transform its own
identity.

Rob Jameson, ERisk