Strategy as a Portfolio of Real Options...harvard business review September–October 1998 91...

Strategy as a Portfolio of Real Options

by Timothy A. Luehrman

Harvard Business Review Reprint 98506

SEPTEMBER – OCTOBER 1998

Reprint Number

HarvardBusinessReviewHOW TO KILL CREATIVITY 98501

STRATEGY AS A PORTFOLIO OF REAL OPTIONS 98506

FAST, GLOBAL, AND ENTREPRENEURIAL: 98507SUPPLY CHAIN MANAGEMENT, HONG KONG STYLE

THE RIGHT MIND-SET FOR MANAGING 98502INFORMATION TECHNOLOGY

DESPERATELY SEEKING SYNERGY 98504

THE DAWN OF THE E-LANCE ECONOMY 98508

HBR CASE STUDYAFTER THE LAYOFFS, WHAT NEXT? 98510

thinking about…THE HIDDEN TRAPS IN DECISION MAKING 98505

world viewSHORT-TERM RESULTS: THE LITMUS TEST 98511FOR SUCCESS IN CHINA

BOOKS IN REVIEWIN SEARCH OF PRODUCTIVITY 98509

first personRIGHT AWAY AND ALL AT ONCE: 98503HOW WE SAVED CONTINENTAL

Teresa M. Amabile

Timothy A . Luehrman

an interview with victorfung, by Joan Magretta

M. bensaou andmichael Earl

Michael goold andandrew Campbell

thomas w. malone androbert j. laubacher

suzy wetlaufer

john s. hammond,ralph l. keeney, and HOWARD RAIFFA

rick yan

STEPHEN S. ROACH

greg brenneman

Copyright © 1998 by the President and Fellows of Harvard College. All rights reserved. 89

hen executives create strategy, they projectthemselves and their organizations into the future,creating a path from where they are now to where

they want to be some years down the road. In competitivemarkets, though, no one expects to formulate a detailedlong-term plan and follow it mindlessly. As soon as we startdown the path, we begin learning – about business condi-tions, competitors’ actions, the quality of our preparations,and so forth – and we need to respond flexibly to what welearn. Unfortunately, the financial tool most widely reliedon to estimate the value of strategy – discounted-cash-flow(DCF) valuation – assumes that we will follow a predeter-mined plan, regardless of how events unfold.

A better approach to valuation would incorporate boththe uncertainty inherent in business and the active decisionmaking required for a strategy to succeed. It would helpexecutives think strategically on their feet by capturing thevalue of doing just that – of managing actively rather than

Strategyas a Portfolio ofReal Options

by Timothy A. Luehrman

Now we can“draw”a strategy in terms that areneither wholly strategic nor wholly financial

but rather an insightful mix of both.

Timothy A. Luehrman is a professor of finance at Thunderbird,the American Graduate School of International Management,in Glendale, Arizona. He is the author of “What’s It Worth? AGeneral Manager’s Guide to Valuation” (HBR May–June 1997)and “Investment Opportunities as Real Options: Getting Startedon the Numbers” (HBR July–August 1998).

W

passively. Options can deliver that extra insight.Advances in both computing power and our under-standing of option pricing over the last 20 yearsmake it feasible now to begin analyzing businessstrategies as chains of real options. As a result, thecreative activity of strategy formulation can be in-formed by valuation analyses sooner rather thanlater. Financial insight may actually contribute toshaping strategy, rather than being relegated to anafter-the-fact exercise of “checking the numbers.”

In financial terms, a business strategy is muchmore like a series of options than a series of staticcash flows. Executing a strategy almost always in-volves making a sequence of major decisions. Someactions are taken immediately, whileothers are deliberately deferred, somanagers can optimize as circum-stances evolve. The strategy sets theframework within which future deci-sions will be made, but at the sametime it leaves room for learning fromongoing developments and for discre-tion to act based on what is learned.

To consider strategies as portfoliosof related real options, this articleexploits a framework presented in “InvestmentOpportunities as Real Options: Getting Started onthe Numbers” (HBR July–August 1998). That arti-cle explains how to get from conventional DCFvalue to option value for a typical project – in otherwords, it is about how to get a number. This articleextends that framework, exploring how optionpricing can be used to improve decision makingabout the sequence and timing of a portfolio of stra-tegic investments.

A Gardening Metaphor: Options as Tomatoes Managing a portfolio of strategic options is likegrowing a garden of tomatoes in an unpredictableclimate. Walk into the garden on a given day inAugust, and you will find that some tomatoes areripe and perfect. Any gardener would know to pickand eat those immediately. Other tomatoes are rot-ten; no gardener would ever bother to pick them.These cases at the extremes – now and never – areeasy decisions for the gardener to make.

In between are tomatoes with varying prospects.Some are edible and could be picked now but wouldbenefit from more time on the vine. The experi-enced gardener picks them early only if squirrelsor other competitors are likely to get them. Othertomatoes are not yet edible, and there’s no point inpicking them now, even if the squirrels do get

them. However, they are sufficiently far along, andthere is enough time left in the season, that manywill ripen unharmed and eventually be picked. Stillothers look less promising and may not ripen beforethe season ends. But with more sun or water, fewerweeds, or just good luck, even some of these toma-toes may make it. Finally, there are small greentomatoes and late blossoms that have little likeli-hood of growing and ripening before the seasonends. There is no value in picking them, and theymight just as well be left on the vine.

Most experienced gardeners are able to classifythe tomatoes in their gardens at any given time. Be-yond that, however, good gardeners also understand

how the garden changes over time. Early in the sea-son, none of the fruit falls into the “now”or “never”categories. By the last day, all of it falls into one orthe other because time has run out. The interestingquestion is, What can the gardener do during theseason, while things are changing week to week?

A purely passive gardener visits the garden onthe last day of the season, picks the ripe tomatoes,and goes home. The weekend gardener visits fre-quently and picks ripe fruit before it rots or thesquirrels get it. Active gardeners do much more.Not only do they watch the garden but, based onwhat they see, they also cultivate it: watering, fer-tilizing, and weeding, trying to get more of thosein-between tomatoes to grow and ripen before timeruns out. Of course, the weather is always a ques-tion, and not all the tomatoes will make it. Still,we’d expect the active gardener to enjoy a higheryield in most years than the passive gardener.

In option terminology, active gardeners are doingmore than merely making exercise decisions (pickor don’t pick). They are monitoring the options andlooking for ways to influence the underlying vari-ables that determine option value and, ultimately,outcomes.

Option pricing can help us become more effec-tive, active gardeners in several ways. It allows usto estimate the value of the entire year’s crop (oreven the value of a single tomato) before the sea-son actually ends. It also helps us assess each

90 harvard business review September–October 1998

strategy as a portfolio of real options

In financial terms, a businessstrategy is much more like a seriesof options than it is like a series of

static cash flows.

harvard business review September–October 1998 91


tomato’s prospects as the season progresses andtells us along the way which to pick and which toleave on the vine. Finally, it can suggest what to doto help those in-between tomatoes ripen beforethe season ends.

A Tour of Option Space Instead of a garden plot, visualize a rectangle we’llcall option space. Option space is defined by twooption-value metrics, each of which captures a dif-ferent part of the value associated with being ableto defer an investment. Option space can help ad-dress the issues an active gardener will care about:whether to invest or not (that is, whether to pick ornot to pick), when to invest, and what to do in themeantime.

Let’s briefly review the two metrics, which weredeveloped in “Investment Opportunities as RealOptions.” The first metric contains all the usualdata captured in net present value (NPV) but addsthe time value of being able to defer the invest-ment. We called that metric NPVq and defined it asthe value of the underlying assets we intend tobuild or acquire divided by the present value of theexpenditure required to build or buy them. Put sim-ply, this is a ratio of value to cost. For convenience,here, we’ll call it our value-to-cost metric instead ofNPVq, but bear in mind that value and cost refer tothe project’s assets, not to the option on those assets.

When the value-to-cost metric is between zeroand one, we have a project worth less than it costs;when the metric is greater than one, the project isworth more than the present value of what it costs.

The second metric we’ll call our volatility metric.It measures how much things can change beforean investment decision must finally be made. Thatdepends both on how uncertain, or risky, the futurevalue of the assets in question is and on how longwe can defer a decision. The former is captured bythe variance per period of asset returns; the latteris the option’s time to expiration. In the previousarticle, this second metric was called cumulativevolatility.

Option space is defined by these two metrics,with value-to-cost on the horizontal axis and vola-tility on the vertical axis. See the graph, “OptionSpace Is Defined by Two Option-Value Metrics.”The usual convention is to draw the space as a rec-tangle, with the value-to-cost metric increasingfrom left to right (its minimum value is zero), andthe volatility metric increasing from top to bottom(its minimum value also is zero). Within the in-terior of the rectangle, option value increases asthe value of either metric increases; that is, from

any point in the space, if you move down, to theright, or in both directions simultaneously, optionvalue rises.

How does option space help us with strategy? A business strategy is a series of related options: it is as though the condition of one tomato actually affected the size or ripeness of another one nearby.That obviously makes things more complicated.Before we analyze a strategy, let’s first consider thesimpler circumstance in which the tomatoes grow-ing in the garden don’t affect one another. To dothat, we need to explore the option space further.

In a real garden, good, bad, and in-between toma-toes can turn up anywhere. Not so in option space,where there are six separate regions, each of whichcontains a distinct type of option and a correspond-ing managerial prescription. We carve up the spaceinto distinct regions by using what we know aboutthe value-to-cost and volatility metrics, along withconventional NPV.

What’s the added value of dividing option spacein this fashion? Traditional corporate finance givesus one metric – NPV – for evaluating projects, andonly two possible actions: invest or don’t invest. Inoption space, we have NPV, two extra metrics, andsix possible actions that reflect not only where aproject is now but also the likelihood of it ending upsomewhere better in the future. When we return toassessing strategies, this forward looking judgmentwill be especially useful.

value-to-cost

vola

tilit

y

lower

higher

lower higher

European call-optionvalue increases in thesedirections.

1.00.0

Option Space Is Defined by TwoOption-Value Metrics

We can use the two option-value metrics to locate projects in option space. Moving to the right and/ordownward corresponds to higher option value.

Value-to-cost metric = NPVq = S÷PV(X)Volatility metric = s=+t

should go ahead and invest right away? In some instances, the answer is clearly no, while in othercases, it’s maybe. We want to be able to distinguishbetween those cases. The key to doing so is not op-tion pricing but conventional NPV.

In terms of the tomato analogy, we are looking ata lot of promising tomatoes, none of which is per-fectly ripe. We want to distinguish between thosethat, if picked right away, are edible (NPV > 0) andthose that are inedible (NPV < 0). The distinction

matters because there is no pointin picking the inedible ones. Con-ventional NPV tells us the valueof investing immediately despitethe fact that time has not yet runout. If NPV is negative, immediateexercise is unambiguously subop-timal. In option terminology, wesay that such an option is out ofthe money: it costs more to exer-cise it than the assets are worth.The exercise price (X) is greaterthan the underlying asset value(S), therefore NPV = S 2 X < 0.

The curve in our diagram sepa-rates options that are out of themoney (NPV < 0) from those thatare in the money (NPV > 0). Forpoints above the curve in the dia-gram, NPV is positive; for thosebelow the curve, NPV is negative.For points actually on the curveitself, NPV = 0.

Projects below the curve, whichwe’ll call region 3, are like theinedible tomatoes that we clearlydon’t want to pick right away. Evenso, they are very promising be-cause their value-to-cost metricis positive and time has not yet

run out. I call this region probably later because,even though we should not invest yet, we expect toinvest eventually for a relatively high fraction ofthese projects. In the meantime, they should becultivated.

Projects that fall above the NPV = 0 curve areeven more interesting. These options are in themoney. They are like tomatoes that even thoughnot perfectly ripe are nevertheless edible. Weshould be considering whether to pick them early.

It may seem contradictory to consider exercis-ing an option early when all along I’ve argued in“Investment Opportunities as Real Options” thatit is valuable to be able to defer the investment – towait, see what happens, and then make an optimal

value-to-cost

1.0

lower

higher

vola

tilit

y

NPV = 0 (exercise now)

0.0

Top of the Space: Now and Never. At the very topof our option space, the volatility metric is zero.(See the diagram “Dividing Option Space into Re-gions.”) That’s so either because all uncertainty hasbeen resolved or because time has run out. Withbusiness projects, the latter is far more likely. So pro-jects that end up here differ from one another onlyaccording to their value-to-cost metrics, and it’seasy to see what to do with them. If the value-to-costmetric is greater than one, we go ahead and invest



Dividing Option Space into Regions

The curve may be derived by holding the risk-free rate of return (rƒ) andthe standard deviation (s) constant (from the Black-Scholes equation) as tvaries and solving for the value of the value-to-cost metric that correspondsto NPV = 0. For example, in the extreme case of rƒ = 0, the curve is avertical line corresponding to points for which the value-to-cost metric = 1.As rƒ increases, the slope of the curve decreases, bending to the right.

now. If it’s less than one, we invest never. Once timehas run out, “now or never” completely describesour choices. It will be convenient to refer to regionsby number, so let’s number these extremes 1 and 6.Region 1 contains the perfectly ripe tomatoes; it isthe invest now region. Region 6 contains the rottenones; the prescription there is invest never.

Right Side of the Space: Maybe Now and Proba-bly Later. What about projects whose value-to-costmetric is greater than one but whose time has notyet run out? All such projects fall somewhere in theright half of our option space but below the top.Projects here are very promising because the under-lying assets are worth more than the present valueof the required investment. Does that mean we

region1invest nowregion6invest never

region2maybe now

Projects here are out of the money: NPV<0. But a value-to-cost metric greater than one makes them very promising.

region3probably later

Projects here have an NPV>0, even if exercised immediately.

They are in the money. Early exercise should

be considered.

choice at the last possible moment. If there is valueassociated with deferring, why would we ever dootherwise? Sometimes, especially with real options,value may be lost as well as gained by deferring,and the proper decision depends on which effectdominates.

The financial analog to such a real option is a calloption on a share of stock. If the stock pays a largedividend, the shareholder receives value that theoption holder does not. The option holder may wishto become a shareholder simplyto participate in the dividend,which otherwise would be for-gone. Think of the dividend asvalue lost by deferring the exer-cise decision.

In the case of real options, wherethe underlying asset is some setof business cash flows, any pre-dictable loss of value associatedwith deferring the investment islike the dividend in our stock ex-ample. Phenomena like pendingchanges in regulations, a pre-dictable loss of market share, orpreemption by a competitor areall costs associated with invest-ing later rather than sooner andmight cause us to exercise an op-tion early. Or, to use the tomatoanalogy, we might pick an edibletomato early if we can predictthat squirrels will get it other-wise. Unpredictable gains andlosses, however, would not leadus to exercise our options early.

Options that are in the money(that is, those for which NPV > 0)should be evaluated to see if theyought to be exercised early. Im-mediate investment will not always be the optimalcourse of action because by investing early thecompany loses the advantages of deferring, whichalso are real. Deciding whether to invest early re-quires a case-by-case comparison of the value of in-vesting immediately with the value of waiting a bitlonger – that is, of continuing to hold the project asan option. I refer to that part of the option space asmaybe now because we might decide to invest rightaway. Let’s label it region 2.

Left Side of the Space: Maybe Later and ProbablyNever. All options that fall in the left half of thespace are less promising because the value-to-costmetric is everywhere less than one, and conven-tional NPV is everywhere less than zero. But even



here we can separate the more valuable from theless valuable. The upper left is unpromising terri-tory because both the value-to-cost and volatilitymetrics are low. These are the late blossoms andthe small green tomatoes that are unlikely to ripenbefore the season ends. I call this part of the optionspace probably never, and we can label it region 5.

In contrast, the lower section (of this left half ofthe space) has better prospects because at least oneof the two metrics is reasonably high. I call it

maybe later, and we can label it region 4. The dia-gram “The Tomato Garden” dispenses with fancycurves and simply divides the option space roughlyinto the six regions.

When to Harvest As an example of what we learn from the tomatogarden, consider six hypothetical projects that areentirely unrelated to one another. The table “VitalStatistics for Six Independent Projects” shows therelevant data for these projects, which have beenlabeled A through F. Note that each of them involvesassets worth $100 million. Two of them (A and B)require capital expenditures of $90 million; the other

The Tomato Garden

If we start on the right in region 1 and sweep through the space clockwise,projects become progressively less promising. In regions 1 through 3, thevalue-to-cost metric is greater than one, but only in regions 1 and 2 isconventional NPV positive, and only in region 1 is the exercise decision aforegone conclusion. In regions 4 through 6, the value-to-cost metric is lessthan one, but only in region 6 is the exercise decision a foregone conclusion.

value-to-cost 1.0

lower

higher

vola

tilit

y

0.0

inedible, butvery promisingtomatoes

less promisinggreen tomatoes

late blossomsand small greentomatoes

imperfect, butedible tomatoes

rotten tomatoes ripe tomatoesregion1invest nowregion6invest never

region5probably never

region3probablylater

region2maybenow

region4maybelater

four require expenditures of $110 million. So A andB each has a positive NPV of $10 million. Each ofthe other four has an NPV of negative $10 million.The NPV of the entire portfolio is negative $20 mil-lion or, more reasonably, positive $20 million, sincethe four projects with negative NPVs can be includedat a value of zero. Conventional capital budgetingoffers only two prescriptions – invest or don’t invest.Following those rules, we’d accept projects A and Band reject all the others.

Although their NPVs are tightly clustered, thesix projects have different time and volatility pro-files, and hence different values for their value-to-cost and volatility metrics. Consequently, each islocated in a different region of the option space. (Seethe diagram “Locating the Projects in the TomatoGarden.”)

A is a now project that falls in region 1; C is a neverproject in region 6. For both of them, time has runout, so the volatility metric is zero. Project B is verypromising: its NPV is positive, and its value-to-costmetric is greater than one. B plots in region 2, andwe should consider whether we ought to exerciseour option on this project early. However, unlessthere is some predictable loss in future value (eithera rise in cost or a fall in value), then early exercise isnot only unnecessary but also suboptimal. ProjectF’s value-to-cost metric is greater than one, but itsNPV is less than zero. It falls in region 3 and is veryvaluable as an option, despite its negative NPV.That’s because it will not expire for two years andhas the highest volatility of the whole group.Hence, project F’s prognosis is probably later.

Project E has less going for it than project F. It isin region 4 and deserves some attention because,

with a year to go and the moderate standard devia-tion of its underlying asset return (s = 0.3 per year),it just might make it. That’s why it is classified asmaybe later. Project D is much less promising (aprobably never) because a decision must be made inonly six months and, with a low volatility, there’snot much likelihood that D will pop into the moneybefore time runs out.

Because it can account for flexibility and uncer-tainty, the options-based framework produces adifferent assessment of this portfolio than the con-ventional DCF approach would. Most obviously,where DCF methods give the portfolio a value of$20 million, option pricing gives it a value of about$74 million, more than three times greater. Just asimportant, locating these projects in the tomatogarden yields notably different exercise decisions.



Vital Statistics for Six Independent Projects

Locating the Projects in theTomato Garden

value-to-cost

1.0

lower

higher

vola

tilit

y

AD

C

BF

E

0.0

region1now

region4maybelater

region5probablynever

region6never

region2maybenow

region3probablylater

Variable A B C D E F Portfolio

S Underlying asset value ($ millions) $100.00 $100.00 $100.00 $100.00 $100.00 $100.00 Value

X Exercise price ($ millions) $90.00 $90.00 $110.00 $110.00 $110.00 $110.00

t Time to expiration (years) 0.00 2.00 0.00 0.50 1.00 2.00

s Standard deviation (per year) 0.30 0.30 0.30 0.20 0.30 0.40

rf Risk-free rate of return (% per year) 0.06 0.06 0.06 0.06 0.06 0.06

NPVq Value-to-cost metric 1.111 1.248 0.909 0.936 0.964 1.021

s=+t Volatility metric 0.000 0.424 0.000 0.141 0.300 0.566

Call value ($ millions) $10.00 $27.23 $0.00 $3.06 $10.42 $23.24 $73.95

S-X Conventional NPV ($ millions) $10.00 $10.00 -$10.00 -$10.00 -$10.00 -$10.00 $20.00

Region 1 2 6 5 4 3

Exercise decision now maybe never probably maybe probablynow never later later

Instead of accepting two projects and rejecting four,our option analysis leads us to accept one, rejectone, and wait and see about the other four. And as wewait, we know how each project’s prospects differ.Moreover, we don’t wait passively. Having onlylimited resources to devote to the portfolio, we real-ize that some wait-and-see projects are more likelyto reward our active cultivation than others. In par-ticular, we can see that projects E and F together areworth about $34 million (not negative $20 millionor even $0) and should be actively cultivated ratherthan abandoned. At the very least, they could besold to some other gardener.

A Dynamic Approach Cultivation is intended to improve the crop, but ithas to work within boundaries set by nature. In op-tion space, as in nature, there are basic laws of timeand motion. The most basic is that options tend tomove upward and to the left in the option space astime passes. Upward, because the volatility metricdecreases as time runs out. To the left, because, as apresent-value calculation, the value-to-cost metricalso decreases over time if its other constituentvariables remain constant.

To illustrate, consider project F. Its volatilitymetric is 0.566, and its value-to-cost metric is1.021. Now let a year pass, andsuppose none of project F’s vari-ables changes except for t, whichis now one year instead of two.Were we to recompute the met-rics, we would discover that bothhave declined. The volatilitymetric falls from 0.566 to 0.400,which moves F upward in optionspace. And its value-to-cost metric declines from1.021 to 0.964 – that is, 100÷[110÷(1+0.06)1] – whichmoves F to the left. In fact, project F moves fromregion 3 (probably later) to the less promising re-gion 4 (maybe later). Despite its initial promise, theonly way project F is going to wind up in the money(that is, in region 1 or 2) and eventually get fundedis if some force pushes it to the right, overcomingthe natural tug to the left, before time runs out.Only two forces push in that direction: good luckand active management.

Neither force should be ignored. Sometimes wesucceed by putting ourselves squarely in the way ofgood fortune. Other times we have to work at it.Managers actively cultivating a portfolio of oppor-tunities are, in effect, working to push options asfar as possible to the right in the space before theyfloat all the way to the top. How is that done? By

taking some action that increases either or both ofour option-value metrics. Of the two, the value-to-cost metric is perhaps the more obvious one to workon first because managers are more accustomed tomanaging revenues, costs, and capital expendituresthan volatility or time to expiration.

Anything managers can do to increase value orreduce cost will move the option to the right in ourspace. For example, price or volume increases, taxsavings, or lower capital requirements, as well asany cost savings, will help. Such enhancements tovalue are obvious with or without a real-optionsframework. What the framework provides is a wayto incorporate them visually and quantitativelyinto option value through the value-to-cost metric.

The real world seldom gives managers the luxuryof isolating one variable and holding all others con-stant. Managers cannot simply declare, “Let’s raiseprices to increase the value of our project.” Morelikely, they will invent and evaluate complex pro-posal modifications driven or constrained by tech-nology, demographics, regulations, and so on. Forexample, one way to cultivate a market-entry op-tion might be to add a new product feature. Thatmay entail extra investment (raising X), but it willalso help differentiate the product in the local mar-ket, permitting higher prices (raising S) but alsoadding extra manufacturing costs (lowering S),

some of which are fixed. The net effect on the value-to-cost metric is what counts, and the net effect is unclear without further analysis.

Evaluating the project as an option means there ismore, not less, to analyze, but the framework tellsus what to analyze, gives us a way to organize theeffects, and offers a visual interpretation. Observingthe change in the option’s location in our space tellsus both whether its value has risen or fallen andwhether it has migrated to a different region of thetomato garden.

There are still more considerations even in thissimple example of adding a product feature. Extrafixed costs mean greater risk, which might lowerthe value of the project (due to the need to discountfuture cash flows at a higher risk-adjusted rate) andcause its value-to-cost metric to drop further. Butthe extra fixed costs also represent operating lever-



The options-based framework givesus a different assessment than the

conventional DCF approach would.

age that raises the volatility metric. That augmentsoption value. We could hypothesize further thatadding an extra feature will stimulate a competitorto match it. We, in turn, might be forced to intro-duce the next generation of our product (on whichwe hold a different option) earlier than we other-wise would have.

In general, actions taken by managers can affectnot only the value-to-cost measure but also thevolatility metric. In this example, both elements ofthe volatility metric – risk and time to expiration –are affected. And for more than one option. There isa spillover from one option to another: adding a fea-ture reduces the length of time a subsequent deci-sion can be deferred. For other situations, there area myriad of possible spillover effects.

Nested Options in a Business Strategy Once we allow options in a portfolio to directly in-fluence other options, we are ready to consider strat-egies: series of options explicitly designed to affectone another. We can use “nests” of options uponoptions to represent the sequence of contingenciesdesigned into a business, as in the following simpli-fied and hypothetical example.

Three years ago, the WeatherIzeCorporation bought an exclusivelicense to a technology for treat-ing fabric to retard its breakdownin extreme weather conditions.The idea was to develop a newline of fabric especially suitablefor outdoor commercial awnings,a market the company alreadyserves with a less durable prod-uct. Now WeatherIze’s engineershave developed their first treatedfabric, and the company is con-sidering making the expendituresrequired to roll it out commer-cially. If the product is well re-ceived by awning manufacturers,WeatherIze will have to expandcapacity within three years of in-troduction just to serve awningproducers.

The vice president for businessdevelopment is ebullient. He an-ticipates that success in awningswill be followed within anothertwo years by product extensions –similar treatment of different fab-rics designed for such consumergoods as tents, umbrellas, and patio

furniture. At that time, WeatherIze would expandcapacity yet again. The company envisions trade-marking its fabrics, expanding its sales force, andsupporting the consumer products made from thesefabrics with cooperative advertising.

WeatherIze’s strategy for exploiting the treatmenttechnology is pretty straightforward. It consists of aparticular sequence of decision opportunities. Thefirst step of the execution was to purchase the li-cense. By doing so, the company acquired a se-quence of nested options: to develop the product; tointroduce the product; to expand capacity for manu-facturing awning fabric; and to expand again tomake related, branded fabrics. Just now, having de-veloped the product, WeatherIze is part way throughthe strategy and is considering its next step: spend-ing on the product introduction. That is, it’s time toexercise (or not) the next real option in the chain.

WeatherIze’s strategy, at this point in time, is de-picted in option space in the diagram “WeatherIze’sStrategy as Nested Call Options.” Each circle repre-sents an option whose location in space is deter-mined by its value-to-cost and volatility metrics.The size of each solid circle is proportionate to theunderlying asset value (S) for each option. The area



WeatherIze’s Strategy as Nested Call Options

The value of WeatherIze’s strategy is:

The value of the innermost call option must be estimated first because itsvalue is part of the underlying asset value (S) for the next option in the nest.

lower

vola

tili

ty firstexpansion option

now

maybe now

probably latermaybe later

probably never

neverproduct introduction option

second expansion option

higher

= s= x

0.01.0

value-to-cost

[ ]{ }.PV(secondexpansionoption)

(firstexpansionoption)

+ callvalue + call

valueproductintroduction

within each dashed circle is proportionate to re-quired expenditures (X). Thus a dashed circle insidea solid one represents an option that is in the money(S > X). A dashed circle outside a solid circle showsan option that is out of the money.

The line segments in the diagram indicate thatthe options are nested. The option to expand forawning production is acquired if and only if theoption to introduce is exercised. As such, the un-derlying asset for the introduction option includesboth the value of the operating cash flows associatedwith the product itself and the present value of theoption to expand. Likewise, the option to expand asecond time for commercial product production isacquired only if WeatherIze decides to exercise itsfirst expansion option. The value of the whole strat-egy at this point is:

In effect, WeatherIze owns a call on a call.The option to introduce the new awning fabric is

in the money and about to expire. (WeatherIze willforfeit its license if it does not go ahead with theintroduction.) As soon as this option is exercised,the picture changes. The top circle goes away; thebottom two remain linked and begin drifting up-ward. One of the most importantfactors determining whetherthey move right or left on theirway up is how well the awningfabric does in the marketplace.But there are other factors as well.Anything that enhances the valueof the second expansion optionenhances the value of the first,too, because the value of the sec-ond option forms part of the valueof the underlying asset value forthe first option.

Suppose, for example, the risksassociated with the consumer-product fabric’s assets increase.Let’s trace the effects in the dia-gram “What Happens If the Con-sumer Fabric Opportunity Be-comes Riskier?” The most directeffect is on the second expansionoption, which moves down in thespace because its volatility metricrises. The second expansion optionbecomes more valuable. But theincreased risk also affects the firstexpansion option for awning fab-ric. Its value-to-cost metric rises



because the second expansion option is part of theunderlying assets (S) of the first. In fact, a change ineither metric for the second option must alsochange the value-to-cost metric (at least) of the first.

As another example, suppose a competitor intro-duces a substitute fabric in the consumer goodsmarkets that WeatherIze had planned to target. Tryto visualize what will happen. Not only will thelocations of the options change but so will the sizes of the circles. The solid circle, or asset value (S), ofWeatherIze’s second expansion option will shrink,and both the first and second expansion optionswill move to the left. Further, the first expansionoption’s underlying asset value also should shrink.

Drawing simple circles in the option space alsolets us compare strategies. For example, we havebeen assuming that WeatherIze would not introducebranded fabrics without first expanding its awningfabric capacity. Now suppose the company could doeither first, or both simultaneously, but that a largerinvestment would be required to make branded fab-rics if the awning expansion weren’t accomplishedfirst. We could also assume that profit margins onthe branded goods would be higher if the companyfirst gained more experience with awning fabric.

These options in WeatherIze’s alternative strategyare not nested, and they are no longer in the same

What Happens If the Consumer FabricOpportunity Becomes Riskier?

If the volatility of returns on consumer-product fabrics increases, the secondexpansion option moves down. This causes the first expansion option to move tothe right because the second option is part of the underlying assets for the first.

lower

vola

tili

ty

firstexpansion option

now

maybe now


probably never

never


higher

0.01.0

➠

➠

value-to-cost

= s= x

[ ]{ }.PV(secondexpansionoption)

(firstexpansionoption)

+ callvalue + call

valueproductintroduction

locations. The diagram “Call Op-tions in WeatherIze’s AlternativeStrategy” depicts the new strat-egy. Note that the second option,the branded-fabric option, is nowfarther left, its solid circle, or as-set value (S), is smaller, and itsdashed circle, or expenditures(X), is larger than it was originally.It is further out of the money butis now linked directly to the pro-duct introduction option. Giventhat the branded-fabric option isfarther left under this new strat-egy and its solid circle is smaller,could we possibly prefer it? Yes,actually, provided it also movesdown in the space – that is, if itsvolatility metric has increased.The pricing table in the real-op-tions framework can tell us howfar down it would have to go tocompensate for any given moveto the left. Finally, note that forthe nonnested strategy, the valueof both options directly enhancesthe value of the underlying as-sets associated with the initialproduct introduction. But it is nolonger the case that any change in the second ex-pansion option must affect the location of the firstexpansion option: each could, in fact, move aroundindependently.

Although the options are not nested, they are verymuch related. Suppose, for example, that the awningexpansion option pops into the money and is indeedexercised first, before the consumer fabric option.The value of the latter would be enhanced becausethe underlying assets associated with it would be ex-pected to produce better margins – the value-to-costmetric for the consumer fabric option rises.

To compare WeatherIze’s alternative strategies,we compute the value of each strategy’s introduc-tion option. We can do that quantitatively using thereal-options framework. In visual terms, we preferthe introduction option to be farther to the rightand to have a larger solid circle. Whichever strategyaccomplishes that is more valuable.

Learning to Garden I argued in “Investment Opportunities as Real Op-tions” that companies should adopt option-pricingtechniques as adjuncts to their existing system, notas replacements. If WeatherIze takes that approach,

there is a good chance that the “tomato garden”will help the company create and execute a superiorstrategy.

Strategists at WeatherIze already were thinkingseveral moves ahead when they purchased the li-cense. They don’t need a tomato garden to tell themmerely to think ahead. But option pricing quanti-fies the value of the all-important follow-on oppor-tunities much better than standard DCF-valuationtechniques do. And the tomato garden adds a simplebut versatile picture that reveals important insightsinto both the value and the timing of the exercisedecisions. It gives managers a way to “draw” astrategy in terms that are neither wholly strategicnor wholly financial but some of both. Managerscan play with the pictures much as they might witha physical model built of Legos or Tinker Toys.Some of us are most creative while at play.

As executives at WeatherIze experiment with cir-cles in option space, it is important that they pre-serve the link between the pictures they draw andthe disciplined financial projections required by thereal-options framework. They need to rememberthat the circles occupy a certain part of the spacebecause the numbers – the value-to-cost and vola-tility metrics – put them there. At the same time,



Call Options in WeatherIze’s Alternative Strategy

The expansion options are no longer nested under the alternative strategy,but the second option has moved down and to the left. In addition, its solidcircle is smaller and its dashed circle larger than it was originally.

lower

vola

tili

ty

firstexpansion option

now

maybe now


probably never

never


higher

0.01.0

product introduction option

value-to-cost

= s= x

they need to prevent the exercise from becomingjust another variation on “valuation as usual.” Thisis the well-worn rut in which valuation analysis isused primarily to check numbers and as due dili-gence documentation for investments. Instead, thepurpose should be to incorporate financial insightsat the stage when projects and strategies are actuallybeing created.

How does one become a good gardener? Practice,practice. I recommend starting by drawing simplecombinations of projects to learn some commonforms. What are the different ways you can depict apair of nested call options? How can the pair movein the space? What are the ways to transform theirconfiguration by changing the variables? Thenmove on to simple generic strategies. What does agiven strategy look like when drawn in the optionspace? How does the picture change over time?How does it change when an option is exercised?

Next, practice translating real business phenom-ena into visual effects to update pictures. For exam-ple, how will the picture change if you add a directmail campaign to your product introduction? Or

how will the picture change if your competitor cutsprices when you enter a market?

Finally, try drawing your strategy and your com-petitors’ side by side: How does the value and loca-tion of your options affect the value and location oftheirs? How will they all move over time?

In most companies, strategy formulation andbusiness development are not located in the fi-nance bailiwick. Nevertheless, both activities raiseimportant financial questions almost right away.Although the questions arise early, answers typicallydo not. For finance to play an important creativerole, it must be able to contribute insightful inter-pretive analyses of sequences of decisions that arepurely hypothetical – that is, while they are stillmere possibilities. By building option pricing into aframework designed to evaluate not only hard as-sets but also opportunities (and multiple, relatedopportunities at that), we can add financial insightearlier rather than later to the creative work ofstrategy.

Reprint 98506 To place an order, call 1-800-988-0886.



Harvard Business ReviewHARVARD BUSINESS REVIEWSUBSCRIPTION SERVICE

United States and Canada Phone: 800-274-3214Rates per year: United States, $85; Canada, U.S.$95

International and MexicoPhone: 44-1858-435324Fax: 44-1858-468969Rates per year: international, U.S.$145; Mexico, U.S.$95Orders, inquiries, and address changes: Harvard Business ReviewTower House, Sovereign ParkLathkill Street, Market HarboroughLeicestershire le16 9efEngland

International customer service E-mailaddress: [email protected]

Payments accepted: Visa, MasterCard, American Express; checks at current exchange rate payable to Harvard Business Review. Bills and other receipts may be issued.

CATALOGS

Harvard Business School PublishingMedia CatalogThis 32-page, full-color catalog features morethan 40 management development video andinteractive CD-ROM programs.

Harvard Business School PressThis latest full-color catalog features books for the fast-paced business world where youlive and work.

Harvard Business School Publishing Catalog of Best-Selling Teaching MaterialsThis collection of teaching materialscontains those items most requested byour customers.

Harvard Business School PublishingCatalog of New Teaching MaterialsDesigned for individuals looking for thelatest materials in case method teaching.

CASE STUDIES AND HARVARD BUSINESS REVIEWARTICLE REPRINTS

Many readers have asked for an easy way to order case studies and article reprints or toobtain permission to copy. In response, wehave established a Customer Service Team to grant permission, send rush copies in paperform, deliver files in Acrobat (PDF) formatelectronically (Harvard Business Reviewarticles only), or customize collections.

Please contact the Customer Service Team:

Phone: 617-496-1449United States and Canada: 800-668-6780(8 A.M. – 6 P.M. weekdays, voice mail after hours)Fax: 617-496-1029 (24 hours, 7 days a week)E-mail: [email protected](24 hours, 7 days a week)Web Site: http://www.hbsp.harvard.edu

Prices (minimum order, $10):

Harvard Business Review Reprints(Discounts apply to multiple copies of thesame article.)

1–9 copies $5 each10–99 $4100–499 $3.50Electronic $3.50 each

Harvard Business School Case Studies$5.50 each

For quantity estimates or quotes on customized products, callFrank Tamoshunas at 617-495-6198.Fax: 617-496-8866

PERMISSIONS

For information on permission to quote or translate Harvard Business SchoolPublishing material, contact:

Customer Service DepartmentHarvard Business School

Publishing Corporation60 Harvard WayBoston, MA 02163

Phone: 617-496-1449United States and Canada: 800-668-6780Fax: 617-495-6985E-mail: [email protected]

Date post:	03-May-2020
Category:	Documents
Upload:	others
View:	3 times
Download:	0 times

Strategy as a Portfolio of Real Options...harvard business review September–October 1998 91...

Documents