Flexibility and Information

Jeroen Swinkels
Information and Flexibility as Resources

• In many settings, two key feedbacks from current activities and

advantage are
• Flexibility about future choices
• Information to guide them

• Information is valuable to the extent (and only to the extent) that

it guides later choices

• VALUE IS IN THE COMBINATION OF INFORMATION AND

FLEXIBILITY
• Co-specialization!
Sometimes the information comes from activity choices,
sometimes from market offerings

Assets

Information and Advantage

the flexibility to Activities
respond to It
Example

• A new category at Nordstrom (should we do some designer

exercise-wear?)
• Can later expand, modify, or abandon

Assets Activities Advantage (?)

Information about
demand for designer Design, Sourcing and
exercise-wear Display of designer
and the flexibility to exercise-wear
respond to it
Example

• Early-stage exploration of a new drug class

• Can abandon, move to next stage development, continue
exploration

Assets Activities Advantage

Information about the
early stage promise of a Investigation of a new
particular drug class, class of pharmaceutical
and the flexibility to compounds
choose whether and
how to pursue it
Example

• Waiting to invest in a new inventory management system while

seeing what new systems become available and what standards
the industry is moving towards

Assets Activities Advantage

Information about available
inventory management Muddling along with
systems and their adoption current inventory
along your supply chain and management system
the flexibility to choose
appropriately given what is
learned
More Examples

• Performance history of an employee

• Can promote, re-assign, let go
• Sales from a single franchise in a new geographic market (KFC in
Indonesia)
• Expand, look elsewhere
• Starting a rookie quarterback
• Face a series of later choices about how dazed the first string
QB needs to be before you pull him…
• Early stage investments in a new aircraft design (or a business
idea)
• Go to next stage, drop, holding pattern
Two Fundamental Questions

- If information and flexibility are sometimes valuable resources,

can we say anything systematic about when and how much?
• Qualitative?
• Quantitative?

- Can we say anything about how better to leverage the value of

information and flexibility?
“The Real Option Approach”

• A powerful way of dealing with decision making under

uncertainty
• Gives more insight, and a better “language” than decision trees
• More analytically powerful in many settings
• Used explicitly by a growing set of firms (and every major
consultancy)
• Increasingly a part of the standard language and tool-kit of
business decision making under uncertainty
Central Idea

- Many business decisions have the structure of a financial call option

• A call option with strike price $60 and expiration date Sept. 30,
on XYZ Corp. gives the holder the right but not the obligation to
buy a share of XYZ (from the issuer of the option) at price $60 on
or before Sept. 30

• Basic trade-off: The share holder gets the dividends, the option
holder gets the flexibility
The Connection

• The new inventory system is like the share

‣ It costs money, but pays dividends (the flow of savings from
the new inventory system)

• Installing the system is like buying the asset

• Waiting is like owing an option on the asset

‣ Exercise price equal to build out cost of inventory system
‣ Value determined by what happens in terms of evolution of
other systems, standards, etc.
Optimal Exercise

• Optimal exercise of an option involves trading off the value of

the flexibility gained by waiting and the strike price of the
option against the dividends missed by holding the option
rather than the stock

• Similarly, the right time to invest in the new inventory system

is when the benefits from the system are large enough not
only to cover the installation costs, but the loss in flexibility
that entails
Should I Exercise My Option?

• June 30

• Dividend of $5/share to owner of record at day’s end

• Would I rather have $5 and own the share, or should I retain

the option?

• Compare the outcomes of

(1) exercise today and hold share to Sept 30
(2) hold option open to Sept 30, and exercise if P > 60

‣ Ignore discounting for simplicity

Value of Flexibility

Red: Gain or loss on Sept. 30 from

owning the share having paid $60
40

20

0
20 40 60 80 100

-20 Green: Gain or loss on Sept. 30 from

owning the option

-40
Value of Flexibility

The difference is all about the

downside
40

20

0
20 40 60 80 100

-20 If the price on Sept 30 is above 60,

there is no difference in owning the
share and owning the option
-40
Value of Flexibility

40

20

0
20 40 60 80 100

-20 Value of Flexibility is determined by

how probable it is that the price is
below 60, and by how much
-40
Value of Flexibility

40

20

0
20 40 60 80 100

-20 Pr 𝑃𝑃 < 60 × E(60 − 𝑃𝑃|𝑃𝑃 < 60)

-40
Optimal Exercise for the Share

• One should exercise if and only if the missed dividend is larger

than the value of the flexibility

Missed dividend > Value of Flexibility

or
5 > Pr 𝑃𝑃 < 60 × E(60 − 𝑃𝑃|𝑃𝑃 < 60)
The Calculation for the Inventory Control System is the Same

• June 30
• Should I build inventory system that currently looks best, or
wait 3 months?
• Calculate expected savings gained by avoiding 3 months delay
‣ The dividend
• Calculate the value of the flexibility
‣ As before, this is the probability that the choice you were
going to make today is not optimal as of Sept 30, times the
average amount by which it is wrong when it is wrong
Some Types of Options

- Wait and see options

• Delaying projects in uncertain times
• Network Externalities
- Build-on options
• Franchise restaurant themes
• Most R&D processes
‣ Drug development
‣ Venture capital
- Abandonment /repositioning options
• New restaurants, retail, product launches
The Value

- The whole point of understanding the value of an option is to put

structure on understanding the value of the flexibility that the
option embeds, and on how that trades off against the cost of
delay
- In many real business situations, it is precisely the value of
flexibility that is poorly accounted for
- Thinking about things via options is a rich source of insight, and,
in many cases, useful numbers
- A lot of work has gone into pricing options, and much of it
translates
- Every property of financial call options translates into insight into
how to structure and evaluate real business decisions
Charter Communications illustrates the value of actively
embracing uncertainty

- In the late 1990’s Charter Communications had a strategy which

could best be expressed as:

“We are pretty sure that having a huge data pipeline into the
home will be immensely valuable. We’re not quite sure how
yet.”
Too many corporate planning processes…

• Either pretend certainty that doesn’t exist or fail to account for

the benefits of flexibility in subsequent time periods
• Firms optimize against a specific future rather than giving
themselves the ability to do well in several
• The very process of corporate planning makes it very difficult to
admit “failure” later, creating even more lock-in
Jeff Bezo’s philosophy of decision making…

Some decisions are consequential and irreversible or nearly

irreversible – one-way doors – and these decisions must be made
methodically, carefully, slowly, with great deliberation and
consultation. If you walk through and don’t like what you see on
the other side, you can’t get back to where you were before. We
can call these Type 1 decisions. But most decisions aren’t like that –
they are changeable, reversible – they’re two-way doors. If you’ve
made a suboptimal Type 2 decision, you don’t have to live with the
consequences for that long. You can reopen the door and go back
through. Type 2 decisions can and should be made quickly by high
judgment individuals or small groups.
Letter to Shareholders, April 2016
Jeff Bezo’s philosophy of decision making…

One area where I think we are especially distinctive is failure. I

believe we are the best place in the world to fail (we have plenty of
practice!), and failure and invention are inseparable twins.
Letter to Shareholders, April 2016
Jeff Bezo’s philosophy of decision making…

- Bezos is talking about the combination of build-on and

abandonment options!
- Note that his framing explicitly embraces failure as not only
possible, but desirable. If you aren’t failing on reversible stuff,
you aren’t trying enough stuff that may either succeed wildly or
teach you something interesting.
- And, because of this framing, it is easy to “admit” failure, and so
it is easier to quit on a project that is not succeeding.
- Amazon is good at failing!
- And, if you are good at failing, you should optimally do it a lot.
• Fail fast, fail cheap, fail often
A Cable Example: Valuing Dark Fiber

• One Charter strategy was the laying of dark fiber

• Fiber-optic cable that is not currently attached to any

equipment

• Properly thought of as a real option: buys you the right, but

not the obligation, to add capacity cost-effectively later
When is this dark fiber most valuable…

• Imagine that laying the dark fiber costs $100 per house passed
(php), and $400 to build out
• In our terminology, the acquisition cost of the option is $100 php,
the strike price is $400 php
• Having laid the fiber, can wait a few years, and see how market
develops
• Value of the extra capacity is like the value of the underlying
stock when considering an option
For example

• Imagine that the average estimate for the eventual value of the
extra capacity is $450 php (in discounted terms)
• If one was sure of this number, then the clear answer is to
neither build it out, nor to bury the dark fiber in the first place
• Lose $50 php
• But…
Imagine instead that one believes that there is some chance of
substantial extra demand, and some chance of not.

• To be concrete, you assess an 80% chance that after you wait a

while, you will learn that the extra capacity has $300 php value,
20% chance of $1050
• Expectation is $450, as before - just more variance
• Now, option takes on higher value. Choose to exercise only if
value is above strike. In this case, exercise (build-out) when value
is $1050, not when it is $300
• Expected value of dark fiber now?
• 20% of (1050-400) = $130
• Worth it!
Let’s increase the uncertainty

• Now, imagine instead that the $450 php average figure reflects a
80% chance of $200 and a 20% chance of $1450
• Expected value of dark fiber now?
• 20% of (1450-400) = $210
• Now clearly worth it
• If the $450 php figure reflects a 80% chance of $0 and a 20%
chance of $2250, value rises to 20% of (2250-400) = $370
• Contrary to intuition, the increase in uncertainty raises the value
of the project, it doesn’t lower it
Variance is good when dealing with real options!

- The value of an option increases in the volatility of the underlying

asset (holding fixed the mean)
• Spreading out the returns is good for you on the upside. And,
because the option allows you to cut off the bad side, you
don’t really care how bad the bad side is
• Examples:
‣ the value of information
‣ the life of junior faculty (and junior members of law firms,
consultancies and ad firms)
‣ thinking about a “last chance” for a troubled initiative
‣ thinking about the effect of economic uncertainty
What affects the value of delay?

- You are contemplating an investment in a new factory

- But, while you think the factory will be profitable on average,
economic uncertainty currently has you putting your plans on
hold
- Government, in an attempt to stimulate investment, lowers the
corporate income tax rate
- Is this likely to have an effect?
- What might work instead?
A geek (ski) interlude…

• “Flexibility” is not a very well-defined word once one has more

than two choices on the table.

• If I can choose between a morning and a late afternoon flight to

get to Aspen, then I care about whether the faculty meeting at 10
AM on my day of departure had an interesting agenda, or just
the usual stuff.

• Combination of information and flexibility valuable as usual…

Dreams of Aspen…

Payoff Afternoon
Flight

Morning
Flight

Interest of faculty meeting

Friends I wish I had…

Imagine a friend offers a direct flight in his private jet at 10 AM…

Private Jet

Payoff Afternoon
Flight

Morning
Flight

Interest of faculty meeting

Friends I wish I had.

• Once the direct private flight is available, I no longer care about

either scheduled flight.

• There is a sense in which my set of relevant choices has dropped.

• I also stop caring about the faculty meeting. It’s a big school.
They will be fine without me for one meeting.

• Information ceases to be valuable!

A better way to think about it…

• The right notion of “more flexible” is actually “more convex”.

• Strategy A is more flexible than Strategy B if the profits from A

minus the profits from B is a convex function of the underlying
state.

• We will similarly say that we have more information if the

distribution shifts in a mean-preserving spread way.
Information and Flexibility (Convexity) are Co-Specialized Resources

• The expected value (EV) of a convex function goes up with a

mean preserving spread (mps).
• In fact, the amount by which EV goes up with mps increases with
the amount of convexity.
• And the amount by which extra convexity is valuable goes up
with size of the mps.

• The more you have of one…

the more valuable is the other!
And the hunt for convexity is on!

• An Israeli firm with the local rights to a European ice-based

energy-use-time-shifting air conditioning system needs a lead
customer…

• And the firm has very limited resources, so the lead customer will
be carrying much of the risk…

• Who should we approach?

• What contract should we offer?
• What else should we do?
And the hunt for convexity is on!

• An Israeli firm with the local rights to a European ice-based

energy-use-time-shifting air conditioning system needs a lead
customer…

• And the firm has very limited resources, so the lead customer will
be carrying much of the risk…

• Who should we approach?

• What contract should we offer?
• What else should we do?
In general…

• Sources of extra convexity involve…

• Ability to leverage an upside
• Ability to truncate or share a downside
• Scale
• Timing

• Sources of extra information

• Better systematic information gathering
• Crisper choices
Netflix seems to have very convex payoffs

“We should have a higher cancel rate overall”

Reed Hastings, Netlix Founder and CEO
Getting “Build-on” Options Right: The Case of 3M

- A good example of a build-on option is in technology ventures

- The early investments in the research stage generate information
about the value of the project
- If the information is better than average, one can push ahead
with the (much more expensive) development and marketing
phases
- If the information is less good, one can end it there
Getting “Build-on” Options Right: The Case of 3M

- One needs to avoid being too hard on early stage projects, and
too easy on late stage ones
- 3M does this well with their “Branch and Prune” strategy
- The amazing thing about 3M is how well they have created a
culture which encourages experimentation, but also knows how
to kill a dud
- The VC process in the US is also pretty darn good at this…
Option Value and New Product Launches:
Gillette’s Launch of the Sensor Circa 1992

• Gillette as of the early 90’s had an aging product line-up, and was
facing erosion of market share from disposables
• The Sensor was based on a technology of laser welding that
allowed a huge improvement in shave quality
• But, there where two issues Gillette did not fully know
• First, how much of a market for a really expensive shaving
system was there?
• Did the market’s move towards disposables reflect a desire for
their simplicity and convenience, or just that no good cartridge
based systems had been launched for a while?
A picture with elements of both horizontal and vertical
differentiation

• heirloom silver
• new cool James Bond
• fiddly
Here??

Existing Cartridges (e.g., Trac II)

shave
experience

Or Here???

• plastic fork Bic Good News

• Homer Simpson
• convenient
bleeding heavily
smooth as a baby’s bottom!

shave quality
Key Question: Where are the
Consumers?
If you don’t know (and Gillette couldn’t)…

- Locate as a cartridge! Give it a huge advertising push

- Why?
- Embeds the key elements of option value:
• Learning
‣ If it is located as a cartridge (the traditional place for high
quality shave solutions) and gets a big advertising push, and
still doesn’t sell, then we have compelling evidence that
there isn’t much of a market in the upper-left corner

• Flexibility
‣ Locating the Sensor as a cartridge embeds the option to
relocate it as a disposable
In Contrast…

• Locating as a disposable neither gives the clean information nor

the flexibility
• If the disposable is a moderate success, what does one
conclude?
• Most people unwilling to pay a premium for a better shave?
• People confused and didn’t believe the best shave out there
could really be one that came on a plastic handle?
• People who are willing to pay for a better shave hate the
cheap handle?

• And, even if for some reason you decide you’d like to move to
the cartridge segment, it is far from clear that you can
The Sensor Launch is ALL about option value.

First Point
• Optionality is the right way to think about managerial incentives
here…
Second Point
• When you launch the Sensor, if it succeeds wildly, you get all the
upside, while if it fails miserably, you take the hit. An NPV analysis
has to use a mid-range, or average over a set of scenarios
• How to think about the Lady Sensor (a Sensor with a big easy to
grip handle)?
• Clearly should be thought of as part of the same project. If don’t
launch Sensor, can’t launch Lady Sensor (builds on production
capacity, market education…)
• Should we just add some cap-ex and mid-range sales forecasts?
No!!!!

• By the time one has to make the decision on the Lady Sensor, the
Sensor will have been in the market for a while
• The Lady Sensor is a “build-on” option that is part of the original
project
• It adds huge value, precisely because it’s payoffs are option-like.
If it is going to be a big flop, you probably will know before you
make the investments, and so you simply kill the project, and
take zero instead of the loss
And, once we are thinking in this direction

- Imagine the Sensor fails in both segments. You’ve just learned

that the market just doesn’t put much of a premium on shave
quality. Existing products are “good enough”

• Implication?
‣ Spend the next decade re-tooling Gillette from a benefits
based competitor to one with more of a focus on costs
And, once we are thinking in this direction

- On the other hand, if the Sensor is a huge success (it was), you’ve
learned that men are in fact willing to spend a lot to look better.
They just hadn’t found a useful way to do so yet!

• Implication?
‣ Spend $1.1 BILLION (with a B) dollars developing the Mach 3
‣ Develop an entire line of men’s grooming products
• The entire Mach 3 project should be viewed as a “build-
on” option that the Sensor made possible.
• And then the Fusion….
Summary: Information as Critical Resource

• One way to think about option value is as one of the “feed-back

loops” between current activities and later resources and
capabilities
• A decision to invest today may be taking away the “resource” of
flexibility: the ability to make a different choice tomorrow
• But, it may equally well be buying the information that guides
later decisions or the ability to later make choices about
investments that would not be available without earlier
investments
Summary: Information as Critical Resource

• In thinking about the value of other resources and capabilities,

one should not be judging them against a single vision for the
future, but asking “What flexibility does this give me? What
opportunities that are not my primary current focus does this
open up?”
• The real options approach provides both qualitative and
quantitative tools for thinking about these issues