THINKING AHEAD

Proposal forBetter BusinessForecasting

There is a great opportunity toimprove business planning — and improve it

considerably. No matter hoiu ivella company schedules its internal operations,

its plans collapse if its salesforecasts are seriously in error. And, of

course, as business executives know

all too ivell, sales forecasts do frequently fail

because of economic uncertainties.

This problem is not insuperable,believes Wassily W. Leontief. If businessand government ivere to collaborate in

developing up-to-date input-output tables for

the economy, a real breakthrough

could be made in sales forecasting. Theauthor is Henry Lee Professor of Economics

at Harvard University.

• THE EDITOBS

Planning is the organized ap-plication of systematic reasoningto the solution of specific practicalproblems. An alternative to plan-ning is the trial-and-error method.Long hefore Charles Darwin de-scribed this alternative as "naturalselection," Adam Smith showed itto be the prime mover of economicprogress.

Far from being incompatible ormutually exclusive, the automaticmechanism of free competitionand the principle of deliberate ac-tion guided by rational decisionboth play their different but equal-ly important parts in the operationof our economic system — the sys-tem responsible for the past rapidgrowth and still unequaled per-formance of the North American,Western European, and. more re-cently, Japanese economies.

The issue that confronts topmanagement is not that of how tochoose between unrestricted com-petition and all-pervasive planning,but rather of how to choose an ef-fective combination of the two.Despite what the professional de-baters on both sides would haveus believe, it is not an eternalconflict between two incompatiblephilosophical principles whieh weface today, hut rather a practicalquestion of efficient working ar-rangements. It is a question whichthe business executive should notask — and certainly not answer —without specifying the "where"and the "when."

INTERNAL PLANS — SUCCESS

The principal seat of rationaldecision making in our part of the

world is the individual private en-terprise. The wave of economicprogress which still carries usalong began to swell more thanthree hundred years ago when,instead of growing, making, trans-porting, and trading in the old,customary ways of their fathersand grandfathers, the first modernentrepreneurs dared to do thingsthat had not been done before.Some dared and failed, but manytried and succeeded, and their sue-cess resulted in a cumulative proc-ess of economic growth.

In military contest we praiseand reward the victor, but we alsohonor those who fall on the fieldof battle. Why do we not erectmemorials to merchants whosecargoes sank on their way to theIndies, and to a mill owner whowent into receivership because heinvested in new, as yet untried,types of looms? Why do we notinscribe on a cenotaph the nameof a banker who went bankruptbecause a new kind of revolvingcredit which he tried to introducedid not turn around fast enough?

Businessmen can consider atrial-and-error procedure success-ful whenever, over a reasonablylong run of trials, the combinedgains exceed by a healthy marginthe total losses. But is it not stillhetter to eliminate, or at least re-duce, the losses while keeping allthe gains? An architect does thiswhen he computes the stresses foreach one of several alternative de-

AUTHOR'S NOTE: This article is an ex-panded version of an address deliveredat the Annual General Meeting of theCanadian Manufacturers' Association inJune 1963.

signs before he huilds a bridge,and a manufacturer does it whenhe prepares detailed cost estimatesfor a new product and comparesthem with the estimate of its pro-spective market price before put-ting it into actual production.

The bigger the amount of capi-tal to be invested, the larger thestakes, and the greater the incen-tive to plan before managementacts. This explains why, with thegrowth of the economy and withthe increasing complexity and wi-dening scope of modern technol-ogy, private enterprise tends torely, in the solution of its in-ternal problems, less and less on"trial and error" procedures, andmore and more on carefully con-ceived and meticulously elaborat-ed planning. While only 15 or30 years ago "operations analysis"and planning were used mainly inpurely physical, engineering opera-tions, today in many corporationsmost of the routine and even someof the central managerial decisionsare arrived at and executed ac-cording to a carefully conceivedblueprint.

For business leaders I hardlyneed explain in detail how thisplanning is done. It suffices to saythat a well-designed and flexibleanalytical system provisioned bya steady flow of factual informa-tion and supported by an efficientprocessing setup will usually dothe job. In the central offices ofthe large and medium-size compa-nies, not only here but in Europe,too, internal corporate planning isdisplacing traditional trial-and-er-ror procedures as fast as mechan-

ical conveyers replaced manualmaterials handling 30 years ago.

EXTERNAL PLANS — FAILURE

But what about sales? The bestpossible internal operations planscan come to naught when con-fronted with uncontrollable exter-nal uncertainties of the market.An individual husiness enterprise,however big, is only a small com-ponent part of a much larger eco-nomic system — a system thatcomprises thousands of other busi-nesses operating in the same andother industries, as well as all con-sumer units and, last but not least,all public enterprises and govern-mental households. Foreign trade,moreover, extends the interdepen-dencies of the economic systembeyond the boundaries of a singlecountry.

How does this system operate?How does it solve its problems? Itsolves them by a trial-and-errormethod. A competitive economycan be viewed as a gigantic, natu-ral computing machine which tire-lessly grinds out solutions to anunending stream of quantitativeproblems automatically fed into it.It allocates labor, capital, and nat-ural resources among all the dif-ferent branches of production. Itdetermines automatically which in-dustry should expand and whichcontract its output, which corpo-ration should invest and which goout of busiriess.

Like any other trial-and-errorprocess, the competitive mecha-nism brings about a proper ad-justment between demand and sup-

continued 071 page 172

HBR NOV.-DEC. I964 167

Harper Booksfor Informed Decision-making

The Interpretationof Financial Statements

Second Revised Edition

by BENJAMIN GRAHAM and CHARLES McGOLRICKUpdated to incorporate new terms and new techniques, here is thestandard manual explaining all about corporation balance sheets andincome statements. This revised edition includes specific financialreports from a number of representative companies as well as recentlyavailable Federal Trade Commission Reports on United States manu-facturing corporations. "If my library were limited to just one book inthe area of interpretation of financial statements, there is no doubt thatthis would be the one."

— RALPH F . LEWIS, Arthur Young & Company. $2.95

Regulating Union GovernmentEdited by MARTEN S. ESTEY,

PHILIP TAFT, and MARTIN WAGNERThis new book studies the background and first four years of operationof the Landrum-Griffin Act of 1959. Its effect on collective bargaining,what has been learned about union government, and how effective theact is proving to be are examined by eight distinguished spokesmen:Joel Seidman, Sar A. Levitan, Benson Soffer, J. Joseph Loewenberg,John B. Stewart, and Emanuel Stein. A volume in Harper's IndustrialRelations Research Association Series. $4.00

Management Uses ofResearch and Development

by WARREN C. LOTHROPVice-president for Corporate Development, Armour & Company

A bold attack on corporate waste of time and money in so-called "pureresearch," this book shows how management can insure more profitableresearch and development ventures. Urging closer cooperation betweenscientist and technologist, the author proposes new methods for im-proving the quality of industrial research and evaluates management'srole in making decisions about technology. The book includes analysisof the special functions of technical service, engineering, quality con-trol, operations research, and commercial development. Index. $4.00

Managing for Resultsby PETER F. DRUCKER

In this companion to his classic. THE PRACTICE OF MANAGEMENT, PeterDrucker cuts across conventional management strategy to show theexecutive decision-maker what to do to move his enterprise forwardand make it prosper. Citing the best examples of management success,he isolates those qualities that distinguish creative management from therun of the mill, and translates them into practical counsel for everyexecutive. "A brilliant companion to his earlier work."

-Wall Street Journal. $5.00At your bookstore or from Dept. 32

^ Publishers, N. Y. 10016

172 HBR NOV.-DEC. I964

THINKING AHEADcontinued from page 167

ply by way of successive ap-proximations. As in target shoot-ing, the bull's eye is supposed tobe hit after a series of graduallydiminishing misses, each of whichimproves the aim. Under the fast-changing conditions of the mod-ern world, bowever, the target atwhich a modern husiness is shoot-ing behaves more often like a fly-ing pigeon, or rather a bat chasinginsects at sunset, than a sittingduck.

In quest of better estimates offuture markets for their products— estimates required for elabora-tion of internal production sched-ules and investment plans —American corporations are spend-ing millions of dollars on salesforecasts. But these forecasts areconspicuous, alas, more by theirfailures than by their successes.As market research is conceivedand organized at the present time,it cannot even come to grips withits real task. For instance:

• The steelmakers try to estimatesales of heavy plate for the next yearor next five years by assessing thevolume of orders that will be receivedby the shipyards.

• The shipbuilders are, at the sametime, trying to estimate the future de-mand for oil so as to be able to de-termine the number of additional or-ders they might receive the next yearfor tankers.

• The market analysts in the oilindustry are, in their turn, busily en-gaged in predicting the output levelsof all tbe principal fuel-consumingbranches of the economy in order todetermine tbe possible demand forpetroleum products.

Simultaneously, but separatelyfrom each other, big, small, andmedium-size businesses in all sec-tors of the national economy areengaged in this frustrating andcostly guessing game. If all theseforecasts were brought together, itis most unlikely that they wouldprove to be compatible with eachother. This means, of course, thatmany of them turn out to be verywrong. It is not surprising thatinternal company production plansand investment decisions based on

continued on page 176

THINKING AHEADcontinued from puge 172

such disconnected forecasts mayprove to be either entirely abortiveor much less profitable than orig-inally expected. The greater theanticipated change is, the largerthe possible miscalculation will be— and the greater the uncertainty.Uncertainty in economic affairs isthe worst enemy of progress.

RISK REDUCTION

One method of mitigating riskis insurance. In the case of thegeneral kind of uncertainty justdescribed, the government, withits ability to create demand —that is, to print money — is theobvious underwriter. While insur-ance of this kind is helpful, andunder certain conditions even in-dispensable, in the long run itseconomic and social costs mightprove to be quite high.

The other solution is to elimi-nate, or at least to reduce, the riskitself. This is the approach whichis in businessmen's interest. Itcalls for market research effortswhich are coordinated, designed toget consistent sales forecasts, andcombined with short- and long-runoutput projections for all branchesof manufacturing, mining, trans-portation, and the service indus-tries. Such an approach is the on-ly means by which the traffic be-tween the different sectors of agrowing economy can be adjustedto the requirements of new tech-nologies and of modern enterprise.

The scope and complexity ofthe fact-finding and analytical taskinvolved in such an approach areadmittedly very great. The newmethods of economic and statisti-cal inquiry developed over the pe-riod of the last 20 years are fullycapable, however, of solving thetechnical problems involved insuch a project. In fact, they havealready been successfully appliedin such Western European coun-tries as Iiolland, Italy, and France.In Great Britain, after two yearsof preparatory work, the NationalEconomic Development Council isabout to follow suit. Japan, theonly non-Communist country toequal and even to better Soviet

Russia's rate of economic growth,has seen close cooperation betweengovernment and business in its de-veiopment and application of so-called input-output analysis from1956 on.

In the United States the trial-and-error method was so success-ful for so long a time that withsome of us it came to be an articleof faith. Not until the economicslowdown of the last decade weremany serious questions raised. Atpresent, budgetary fiscal policies —that is, the insurance approach,with the government playing therole of a general underwriter —represent the main planning tool.But they need not be the only one,or the most important one. Sys-tematic statistical analyses whichshow us how we can advancealong another path have been un-dertaken on a large scale, and theycan be carried much further.

INPUT-OUTPUT TABLES

The first officially compiled in-put-output table of the U.S. econ-omy was released 15 years ago; itwas for the year 1947- Soon after-wards, all work along these lineswas practically discontinued untilig62. Now the Department ofCommerce, in cooperation with theDepartments of Agriculture, Labor,and several others, has completed— and is about to publish — aninput-output table of the U.S. econ-omy for the census year 1958.^Presently this table will be up-dated to 1961 and 1963.

So far as technical capabilitiesare concerned, this work shouldput us in a position which Francereached some 15 years ago, andsomewhat behind West Germany,where tbe Institut fur Wirtschafts-forschung in Munich has just re-leased an excellent set of input-output tables for the year 1961and announced that from now onsuch compilations will be releasedannually.

Without entering into technicaldetails, it suffices to say that suchinput-output tables show the flowsof goods and services among all thedifferent sectors of a national econ-omy,- but that a broad tabulationof economic activity is not enoughfor business purposes.

To supply a reliable statisticalbase for coordinated market analy-sis on the part of business firms,an input-output table must bemuch more detailed. It should de-scribe the actual state of the par-ticular national economy in thebase year — that is, the year fromwhich the forward-demand projec-tions are to be made — in termsof, say, 150, 200, or even as manyas 300 or 400 separate industriesor sectors. The conventional divi-sion of the economy into agricul-ture, mining, industry, and serv-ices might suffice for purposes ofan over-all aggregative description,but any factual conclusions ob-tained on that basis are too gen-eral to be of much use in answer-ing specific questions asked bymarket analysts in any one partic-ular industry.

Management needs much addi-tional information, particularly onthe concrete details of impendingtechnological changes, investmentplans, and other developments invarious industries. This informa-tion should be supplied in thecourse of the cooperative efEort(which I will discuss presently)by private and public corporationsand obtained in consultation withtechnical experts from profession-al and industrial associations.

Practical experience has shown,incidentally, that an input-outputtable can be understood ; id inter-preted as readily as, say, a rail-road timetable, once the initial ap-prehension of a "technical ap-proach" has been allayed. Whatbusinessman would find it difficultto understand a table showing inwhat amounts the commodities orservices in the production or dis-tribution of which he takes partare being actually absorbed by allthe other sectors of the U.S. econ-omy; and — looking at the tablefrom the other side — what kindsof goods and services, and in what

^ Available, or soon to be available,from the Office of Business Economics,Department of Commerce, Washington,D. C.

- For a nontechnical description ofinput-outpnt tables and their use in thesolution of various practical economicproblems, see my article, "Input-OutputEconomics," Scientific Atnericdn., Octo-ber 1951= P- 15-

continued on page 180

176 HBR NOV.-DEC. I964

THINKING AHEADcontinued from page 176

amounts, his own industry re-ceived from each one of these oth-er sectors?

For a specific illustration, takethe case of a welding equipmentmanufacturer:

A glance at an input-output tableof the United States economy wouldshow this manufacturer how muchwelding equipment was purchased,in that particular year, by each oneof the many different industries us-ing welding equipment. A compari-son witb the corresponding figuresfor other years would not only revealupward or downward trends in eachgroup of purchases. It also couldindicate to him the causes of suchchanges. Thus, a particular group ofcustomers might have expanded orcontracted its purchases because ithad expanded or contracted its ownlevel of output; on the other hand,changing technology might have re-sulted in a use of more or less weld-ing equipment and supplies per unitof output in a particular consumingindustry.

But what is still more important,in turning to the comprehensive and,at the same time, detailed over-allprojections described above — thatis, in examining the projected input-output table for, say, tbe year 1968— the welding equipment manufac-turer could arrive at a consistent and,because of that, reasonably reliablesales forecast for bis own product.Anticipation of special conditions,such as a cut in military spending,could in this case be actually takeninto account. Computations of thekind that have already been per-formed — but which would becomemuch more dependable if the pro-gram of research described abovewere carried out — would tell him,for example, how a reduction of de-fense expenditures by. say, 10% or20%, combined with a compensatingexpansion of civilian demand, couldbe expected to affect the demand forhis welding equipment and suppliesIn the year or years ahead.

DEVELOPING THE DIALOGUE

The analytical methods that areused to transform raw economicdata into the final product of spe-cific market forecasts and othertypes of economic projections arewell known to economist-tech-nicians (whom I would like todistinguish from political econo-mists or economist-politicians ).The large-scale data processing op-eration, which only a few years

ago would have niaJe the realiza-tion of the entire project unwieldy,if not impossible, can be performedtoday rapidly and cheaply.

But an important caveat is need-ed here. By their very nature, con-sistent projections of the demandfor goods and services producedand consumed in all the many dif-ferent, but mutually interdepen-dent, sectors of the national econ-omy cannot be run off as so manycars on an assembly line. Despitethe use of electronic computers,the entire operation should first bevisualized and organized like aconstruction job: all the parts haveto be fitted skillfully to each other,although no two of them are alike.

During this process, the poten-tial users should be consulted andgiven an opportunity to stand byand, where appropriate, take part.If the project directors intend thework to be something more thana large-scale academic exercise,the professional staff engaged intbe task must maintain con-stant contact, or rather a runningdialogue, with the knowledgeablemembers of the representativegroups in all tbe different sectorsof the economy. Consider tbe ad-vantages :

« Such an exchange would enablethe experts with practical experiencein each field to pass their experi-ence on to economists and statisti-cians engaged in preparing the actualmarket forecasts.

« The workings of an unfamiliartool can be best understood by thecorporate planners who would use itif they are given all possible oppor-tunity to watcb step-by-step tbe proc-ess of its actual construction.

As time goes on and experienceis accumulated, such a dialoguecan provide a solid factual basisfor informed and detailed consid-eration of many practical issues ofnational economic policy. Techni-cal change and its effect on pro-ductivity and unemployment; onprivate investment, public invest-ment, and economic growtb; ontbe inflow and outflow of capitaland the structure of foreign trade— all these and many other as-pects of economic growth can beunderstood mucb better if pre-sented within a unifying frame-work of a detailed and, at the

same time, comprehensive econom-ic projection.

TRACING INDIRECT EFFECTS

It is a natural babit of advocatesof various economic policies toconcentrate tbeir and our atten-tion on the principal effects (usu-ally recognized by all as beneficial)of their programs and to neglectthe less conspicuous, indirect ef-fects wbich, on close investigation,often turn out to be less desirable.Road-building subsidies, for exam-ple, encourage driving, consump-tion of gasoline, and employmentin the automobile and constructionindustries, but at the same timetbey tend to reduce tbe utilizationof railroads and cause unemploy-ment among railroad employees.

To business executives, need-less to say, tbese indirect effectsmay be crucial. Tbe industries af-fected may be their own. Input-output analysis therefore assumesadded value. By its very nature,tbe comprehensive type of inter-industrial analysis just describedis bound to bring into full relieftbe less as well as the more obvi-ous parts of the entire picturewithout camouflaging any econom-ic blemisbes that it may containand that partisan supporters mayhave overlooked.

Difficulties in appraising theprobable effects of changes in tbelevel of public expenditures, of areduction or an increase in tariffs,and of other government actionsand proposals affecting, to a largeror to a lesser extent, each sectorof the national economy inhibitprivate business not less — andsometimes even more —• than risksassociated witb the normal func-tioning of tbe trial-and-error mecb-anism of our competitive system.If modern analytical techniquescan belp management to tracethrough tbe direct and indirectrepercussions of specific policiesand changes, tbe result should bea stimulus to business initiative.

ORGANIZATIONAL QUESTIONS

So far as the organizational set-up is concerned, the experience ofother countries which entered thefield before us and our own tradi-tion of carrying on activities serv-ing broad public interests suggest

180 HER NOV.-DEC. I964

Gulf Books forManagerial Leadership

The Managerial GridDR. ROBERT R. BLAKEDR. JANE S. MOUTON

The Managerial Grid is a key forunravelling and solving managerialproblems. This tested system hasbeen applied in industry and in eachcase has contributed to profits. Oneindustrial complex attributes 30%of its profit increase directly to theManagerial Grid. There is muchhere for the individual executive itt-cluding a method for evaluating hisown managerial style and ways hecan progress from one style toanother. Then the attention shiftsto organizational development. Itshows how entire organizations haveshifted from outmoded practiceswhich were hindering production toa strategy which increased totalorganizational effectiveness, produc-tion and profit. $8.00

Effective R & D for theSmaller Company

RUSSEL W. HENKE

This volume is no textbook ap-proach, nor is it a swivel-chair com-pedium of egg-head theories. It is,instead, a sensible articulate approachto the problems facing managementof the smaller company desiring toinitiate or improve an R&D program.Everything you need is right here.

$6.50

Managing IntergroupConflict in Industry

DR. ROBERT R. BLAKEDR. HERB SHEPARD

DR. JANE S. MOUTON

This book clearly shows there arenine possible methods for handlinginter-group conflict, but that eightof them have inherent side-effectsdetrimental to the overall organiza-tion. There is little doubt that inter-group conflict is inevitable betweenmanagement and union, staff andline, sales and production, head-quarters and field, supervisor andsubordinates, etc. But only thoseexecutives with the skill and know-how can cope with the warringfactions and direct a mutually-satis-factory solution. This book providesthat skill and know-how. $6.50

GULF PUBLISHING COMPANYPost Office Box 2608Houston, Texas 77001

THINKING AHEADthat a combination of governmen-tal and private operations mightprove to be most appropriate indrawing up input-output programs.

The collection of basic statis-tical data can obviously be handledmost expeditiously by the govern-ment. If given the necessary di-rection and budgetary support, theBureau of the Census could — ata fraction of the cost of a single,routine moon probe — compile de-tailed input-output tables of theU.S. economy once every five years,and compile less detailed tables onan annual or biannual basis. Withmodern data-processing facilities,the lag between collection and pub-lication of the final figures shouldnot exceed i2 nionths and couldbe even shorter.

The analytical work might beperformed by an autonomous, non-profit research organization fi-nanced partly from private fundsand partly by the government ona contractual basis. Elaborationof regional projections would nat-urally be in the hands of the in-terested state governments, whilesome of the details regarding in-dustry sectors could well be han-dled hy the appropriate trade andindustrial organizations.

STIMULATING PROGRESSControversies and conflicts on

questions of economic policy mustand will, of course, go on. Noth-ing could be more futile and self-defeating than to pretend that fac-tual analysis can supersede thetraditional process of political de-cisions on economic matters. Butit can smooth the path and thusincrease the rate and reduce thecosts of economic progress.

The type of planning whoseanatomy I have described is notthe monstrous beast with clawedpaws and coiled tail that somebusinessmen may have expected;neither is it a Trojan Horse. Iwould compare it, rather, to ayoung trotter that shows good timein preliminary workouts. Withsome fast entries competing withus under other colors, it might notbe a bad idea to have it run thenext race for our stable.

— Wassily W. Leontief

PROBLEMS IN REVIEWcontinued from page 22

"Despite the potential turndownby SBA, there is a source close toSBA which should receive Trent'sattention — the small business in-vestment company. SBIC's are pri-vately owned and operated invest-ment companies licensed by SBA toprovide long-term financing (mini-mum of five years) and equity capi-tal to established or newly formedsmall businesses. While the SBA'sInvestment Division licenses, regu-lates, and helps finance SBIC's, ourpart in their day-to-day operations isminimal. All transactions with smallbusinesses are independently negoti-ated, with the managements of theindividual SBIC's suhject only toover-all rules laid down by our regu-lations.

"While I am not sure a man withTrent's limited experience would beenthusiastically received by manySBIC's, they might provide financingin such a case. If he selected care-fully, he also might find an SBICwhich would serve as a valuable part-ner to him. A number of SBIC's areexperienced in developments andshopping centers, and such a firmcould provide management know-how, as well as financial stability, toTrent's operation.

"Trent should expect to give up aminority equity position in returnfor SBIC financing, but he would beassured use of the added funds forat least five years."

TRENT'S FUTURE

What is the outlook for DarrellTrent? Our commentators seesome problems for him, and, atthe same time, reasons for hope.Dealing with the question of whereTrent goes from here, Mr. Swens-rud analyzes:

"Despite my critical view of someof Trent's past moves, his position isprohahly a good deal hetter than itmight have heen had he decided toapply his entrepreneurial talents insome other field. He has an asset inthe form of cleared and consolidatedland which, we are told, has actu-ally increased in value. Rarely inventure projects is the entrepreneurso well situated at the expiration ofhis initial capital, when he finds hemust readjust his thinking on theavailahility and sources of furtherfinancing.

"Assuming Trent is able financial-ly to sustain his properties for an-other several months (indicationsare that he does have some remain-ing capital), a numher of reasonableand potentially fruitful courses ofaction are availahle to him. Possible

182 HBR NOV.-DEC. I964

Harvard Business Review Notice of Use Restrictions, May 2009

Harvard Business Review and Harvard Business Publishing Newsletter content on EBSCOhost is licensed for

the private individual use of authorized EBSCOhost users. It is not intended for use as assigned course material

in academic institutions nor as corporate learning or training materials in businesses. Academic licensees may

not use this content in electronic reserves, electronic course packs, persistent linking from syllabi or by any

other means of incorporating the content into course resources. Business licensees may not host this content on

learning management systems or use persistent linking or other means to incorporate the content into learning

management systems. Harvard Business Publishing will be pleased to grant permission to make this content

available through such means. For rates and permission, contact permissions@harvardbusiness.org.