Fiji a Transition to Manufacturing

ReportNo. 6372-FiJ Fiji A Transitionto Manufacturing Public Disclosure Authorized

November25, 1986 EastAsia PrograinsDepartment FOR OFFICIALUSE ONLY Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized

Documentof the WorldBank

Thisdocurnmnt has a restncteddistribution and may be usedby recipients onlyin the performanceof theirofficial duties. Its contentsmay not otherwise - be disdosedwithout World Baik authorization. CURRENCYEQUIVALENTS

1985 May 1986

US$1 F$1.152 F$1.106 F$1 US$0.868 US$0.904

FISCALYEAR

January1-December 31

ABBREVIATIONS

ANZCERTA - TradingAgreement between Australia and New Zealand DP7 - Fiji'sSeventh Development Plan, 1976-80 DP8 - Fiji'sEighth Development Plan, 1981-85 DP9 - Fij-'sNinth DevelopmentPlan, 1986-)0 EDB - Economic Development Board EPZ - Export processing zone ERP - Effective rate of protection FGDP - Median family income in developingcountries FNTC - Fiji National Training Council FSC - Fiji Sugar Co poration FTUC - Fiji Trade Union Congress HFCS - Uigh fructose corn syrup ICA - International Coffee Agreement KORVA - KoreanTrade PromotionCorporation LOME - EEC trade conventionwith ACP countries MNC - Multinational corporation MOF - Fiji Ministry of Finance NTB - Nontariff barrier PIB - Prices and Incomes Board PJT - Price Justification Tribunal RBF - Reserve Bank of Fiji REER - Real effective exchange rate SDF - SkillsDevelopment Fund (Singapore) SPARTECA - South PacificRegional Trade and EconomicCooperation Agreement TC/TS - Tons cane/tons sugar ratio VAT - Value-added tax FOROMCIAL USE ONLY

The reportwas preparedby a mission that visitedFiji during early April 1985. Missionmembers were ShahidYusuf (Mission Chief),R. Kyle Petersand VladimirKonvalov. The reportwas discussedwith the Fijian authoritiesin October 1986.

This documenthas a restricteddistribution and may be used by recipientsonly in the performance of their officialduti. Its contents maynot otherwisebe disclosed without World Bank authoiation. FIJI A TRANSITIONTO KANUFACTURING

Tableof Contents

PageNo.

SUNSARYAND CONCLUSIONS*.*******..**** ...... i

I.9 INTRODUCTION¢*oo o o o 00 0 0a0e9 ee@*******4a0*0000000* 0 0 1 Recent?erformance...... 1 The Perspectiveof DP9...... S

II. THE FUTUREOF TRADITIONALINDUSTRIES...... 7

T.ouri....soo... .**oo.o * o.o.o*...... o...... 11

III. NEW CONCERNSFOR TRADE STRATEGY...... 15

Efficiency Thresholds..... cco...... oo.g...... g e.oo 17 Sunk Costs...... i1 Value Added.oo...o.ooooo.oo...... , ...... 4, 19 TechnologicalCapaty...... e...... 19 CapitalSupply...... 19 NTIsand GeographicalShifts in Manufacturing...... 20 DirectForeign ...... e...e..... 20 Fiji'sChoices and ReportOutlineo...... 20 IV. INDUSTRIALINCENTIVES...... oo.o.o.. o..o.o...... ,...... 22 EastAsian Approacheso..o ...... 22 Fiji's Manufacturing Sector.oo.o.o ...... v. 28 Past Incentive Policies...... *****o.... . e.e.c.... c... 33 Incentives Reformoo.....o.o..o**oo**.*.***.** **¢eoooo*o ... 41

V. A TAX PROGRAM FOR GROWTHo .o...... e.ooe...... 44 k@ , ...

Tax Criteria...... e..e...... oo. o.o 44 Now Burdensome are Direct Taxes in Fiji?o.....e...... ee....e.o 47 Taxes,Savings and Investment...... 50 Taxes in iji ...... 51 Recastingthe Tax System...... ee...... 54 - ii -

Page No.

VI. TRAD& PATTERNS,TRENDS AND INSTRUMENTS...... 56

Flourishing ...... *0...... 56 Factors Influencing ...... 63 ManufacturedExports from Fiji ...... 65 Trade Agreements...... O ,, 69 Direct Foreign ...... 72 Export Processing ...... Zo nes 72 Exchange Rate ...... 73

VII. LABOR MARKETSAND THE GROWTH OF EXPORTS...... 76

Small Statesand the Controlof Wages...... 76 Unions and Wages in Fiji...... 77 Skills ... o £ ~~~~~~~~~~~80

VIII. THE MICROECONOMICSOF A NICHE-FILLINGEXPORT STRATEGY...... 83

An Information ...... 83 Coping with NTBs ...... 85 Nices...... 85 piis...... 86 Wths...... 89 Machinery ....90 Fashion Garments...... 92

TABLESIN THE TEXT

1.1 Fiji: Basic EconomicIndicators...... 3 1.2 Commodity Price Projections...... 5 1.3 DP9 Prjcin...... 6

3.1 Proportionof IndustrialCountry Imports from East Asian Countries...... 16 3.2 RelativeChanges in Asian-PacificTrade Flow Ratios,1960-83.. 17

4.1 Korea -- Summaryof IndustrialIncentives...... 24 4.2 Singapore-- qummaryof IndustrialIncentives...... 26 4.3 GDP by Industrial Origin...... 30 4.4 Fiji -- Value Added By ManufacturingSubsector...... 31 4.5 Fiji-- Summaryof IndustrialIncentives ...... *...... 35 4.6 Value-Added,Protection Coefficients of Fiji'sIndustrial Sector 38 4.7 Fiji: FiscalEffects of Trade Liberalization...... 43 - lli -

Page No. TABLESIN THE TEXT (continued) 5.1 CentralGovernment Savings and InvestmentBalance*...... 45 5.2 Tax Base Indexand InitialMarginal Tax Rate ...... 49 5.3 HighestBracket and HighestMarginal Tax Rate.,...... 50 5.4 CentralGovernment Revenues...... 52 5.5 Compositionof CentralGovernment Revenues ...... 53

6.1 Major ImportItems into North AmericaRanked by Growth Rates, 1975-81...... *.* * *ee.e 57 6.2 Major ImportItems into Japan Rankedby GrowthRates, 1975-81 58 6.3 Major ImportItems into the EEC Rankedby GrowthRates, 1975-81 59 6.4 Major ImportItems into East Asia EconomiesRanked by Growth Rates, 1975-81...... 60 6.5 LargestManufactured Exports from SelectedAsian Countries to Major OECD Economies...... o...... , 61 6.6 Fastest-GrowingManufactured Exports from SelectedAsian Countriesto Major OECD Economies...... 62 6.7 Nontariff Barriers -- Percentage of Industrial Country Importsfrom DevelopingCountries Covered by NTBs...... 64 6.8 Fiji -- Major Exportsby Three-DigitSITC ...... * 66 6.9 Fiji -- Exportsby SITC Sectionsand Destination ...... 67 6.10 Australia-- Fastest-GrowingImport Items...... 68 6.11 New Zealand-- Fastest-GrowingImport Items...... 68 6.12 Fiji -- SPARTECATrade with Australiaand New Zealand...... 71

7.1 Labor Force Growth Rate...... 76 7.2 Changesin Mean HourlyWage Rates...... 78 7.3 AverageWages in Non-UnionGarment Factories, 1970-83...... 79 7.4 AverageWages in Non-UnionFurniture Factories, 1970-83...... 80

8.1 Total Consumptionof SelectedSpirits, 1982...... *****o**. 87 8.2 Consumptionof Rum Worldwide(1982)...... 88 8.3 Consumptionof DistilledSpirits in the US...... 89 8.4 Per CapitaConsumption of Fruitsand Vegetables...... 94 8.5 Fiji: Exportsof Fruit in 1985...... 97

Annex Table 1: Basic Indicatorsof Fiji Sugar Industry,1976-85.... 99 Annex Table ?: NormalTax Schedulesfor PersonalIncome Taxeso..... 100 Annex Table 3: Elasticityof Major Taxes and of the Fiji Tax System,1974-81...... 101 Annex Table 4: BuoyancyEstimates of CentralGovernment Revenues... 102

CHARTS

Chart 1: By-Productsof Sugar Production ...... 32 Chart 2: Real EffectiveExchange Rate...... 75 FIJIs A TRANSITIONTO MANUFACTURING

SUMMARYAND CONCLUSIONS

Fiji has made the most of its limitednatural endowment. Plantedto 'ugarcane,about 70,000hectares of the most fertileland have yieldedrich rteturns.The sunnybeaches of Viti Levu have generatedemployment and attract a crowingstream of touristsnot only from Australiaand New Zealandbut also the US and Europe. The dense forestscovering two thirdsof the land area supporta trade in logo and wood productswhile coconutgroves, traditionally the props for the villageeconomy, have continuedto supplycopra and coconut oil for the exportmarket. Fiji'smineral wealth is limited,but incomefrom gold mining,averaging $15 millionp.a. in the eighties,has provideda welcomesupplement to other export earnings. These basesof economicstrength shouldendure well into the future,although as sourcesof growth they are no longerdependable. Even allowingfor the difficultinternational environment of the eighties,Fiji's performance has been below par, with the economy expandingat an averagerate of 1.4X p.a. over the past sevenyears. Much of this growth is traceableto the servicessector. The contributionof sugar was negative,declining by about 1Z p.a. in real terms between1979 and 1986. Similarlyother directlyproductive activities such as manufacturing (non-sugar)registered minimal change. Since the populationand the labor forcehave increasedby over 21 p.a. in the eighties,per capita incomesare erodingand open unemploymentnow exceeds10X. As the populationof working age will grow yearlyby more than 2X for the remainderof the decade,there is an urgentneed to quickenthe economicpace.

SMuarand Tourism

The Government'sNinth DevelopmentPlan exploresdevelopment oppor- tunitiesacross a refreshinglybroad canvas. Manufacturing,services, and agriculturaldiversification are all drawn into the strategicharness, but it is sugar and tourismthat once again are accordedthe leadingrole. Intensive cultivation,a small increasein plantedarea, and a significantimprovement in the extractionratio will, it is projected,raise sugar productionfrom 484,000tons in 1986 to 550,000-600,000tons by 1990. Touristarrivals are expectedto grow by 11.5Z p.a. spurredby the plannedenlargoment of hotel accomodationin the five star range. Jointlythe two sectorswill supplythe impetusfor 51 p.a. growth in GDP.

Past trendsand the scenariosbeing drawn for the international sugarmarket as well as the touristbusiness cast some doubt on these projections. A numberof constraintsmay come in the way. For outputto followthe trajectorypresented in DP9, more emphasismust be given to varietalresearch; extensionservices, which are being pinchedby the Fiji Sugar Corporation's financialdifficulties, will have to be improved;investment will be needed in transportfacilities; sugar mills could requiremodernization so as to raise the extractionratio; and it may become necessaryto consolidatethe many smallsugarcane farms into largerunits. All of these are likelyto impede the shift in productiontrends that is calledfor. Conditionsprevailing in the world market sharpenthe uncertainty. Sugar priceshave recoveredfrom Li - the low point to which they had fallen in 1985,but the forecastsfor the next five years are not especiallyencouraging. The dynamicsof demand, competition from substitutes,large inventoriesand more or less stableproduction in the EEC will not permit a significantstrengthening of prices. Fiji will continueto profit from sales under contractto the Europeancountries and the U8, but the net returnsfrom sugar that is disposedof on the free market could be tiny.

Tourismenjoyed a prolongedspring during the seventies,which from the evi.'enceaccumulating during the eighties,appears to have passed. A periodof moderategrowth lies ahead. Fiji'sprospects are also affectedby some very specialcircumstances. Most vistorsare from Australiaand New Zealand,countries which are experiencingeconomic difficulties and being forcedto depreciatetheir currenciesthereby discouraging overseas travel. A decliningUS dollarmay also dampen the enthusiasmfor expensivelong haul journeysamong Americans. To attractlarger numbers of upscaletourists from Europeand Japan,Fiji would need to investheavily not just in premierhotel facilitiesbut also in recreationalinfrastructure. Further, airlines that have been shy of terminatingflights in Fiji or dedicatingsignificant blocs of seats for journeysto the countrywill have to be broughtaround. Beyond this,the Governmentand hotel incerestswill need to mount a much more intensive marketingcampaign aimed at upper incomeprofessionals in the industrial countrieswho evince the highestpropensity for long haul travel,luxury accommodations,and extendedvacations.

Fiji has amassedmuch experiencein the sugar and touristsectors, so it is logicalto exhaustthe remainingopportunities in these areas before givingattention to other possibilities.But there are indicationsthat these industrieshave passed their prime and in the futurewill be hard pressedto match the contributionthey made to growth,export earnings and employment duringthe seventies. Investmentthat will enhanceproductivity in the sugar industryand which will multiplythe attractionsof Fiji'stourist resorts shouldnot be neglected,but the focus of capitalaccumulation must shiftto the manufacturingsector which is likelyto be the principalsource of growth in the ninetiesbut must be groomedfor that role during the DP9 period. And becausethe domesticmarket is small,manufacturing development must henceforthbe orientedtowards exports.

With oil priceson the retreat,Fiji is leavingbehind the adjustment problemthat emerged in the early eighties. This shouldprovide planners with a measureof freedomto ptrsuefresh initiativesand take steps to revive investmentin chosenareas. Fiji'smanufacturing sector accounts for a scant 10% of CDP. If sugar processingis subtracted,it is even smaller. From the very start,industry produced for the domesticmarket under the shelterof protectivetrade barriersand foreigncompanies predominated. In recentyears localbusinessmen have branchedinto manufacturing,but overseasinterests remainsignificant and import substitutionis still the guidingstrategy.

The approachadvanced in this reportleans on the experienceof the East Asian countriesthat have surmountedthe handicapof smallnessby creating internationallycompetit.ve export industries. Certainelements of the strategiesemployed by the East Asian NICs, suitablymodified to accommodate - iii -

Fiji'scircumstances and the increasinglypoliticized trading environment, might well securethe country'sneeds. After brieflydescribing the develop- mentalgambit perfectedby the NICs, the reportpicks up five strandsof the strategyand examinesthem in greaterdetail; first presentinga composite accountof what the East Asiansdid; and then indicatinghow it might be interpretedin the Fijian context. The five are:

(a) industrialincentives;

(b) taxes;

(d) labor policies;and

(e) policiestowards information, marketing and productdesign.

IndustrialIncentives

Three factorsaccount for the size and domes.icbias of the manufacturingsector: Fiji's comparativeadvantage in producingsugar and catering for tourism;the inherentdisadvantages stemming from geographicallocation and a limitedinternal market; and a systemof incentives,created piecemeal, that gives a high degree of effectiveprotection to importsubstituting industries.As the report suggests,the mantle of comparativeadvantage may be slippingfrom the traditionalsectors by the close of the decade. Further, the message from East Asia is that size can be reducedto a minor hindrance throughfirm commitmentto growthand a farsightedchoice of policy. The reportrecommends that the Covernmentshould consider a phasedremoval of three deficienciesin the systemof incentives:

(a) the very high level of effectiveprotection, that in a numberof cases probablyresults in low or negativeindustrial value added;

(b) the wide dispersionin tariffsand rates of p:otectionwhich interfereswith efficientresource allocation; and

Cost and productiondata that would permitsome fine grainedrecom- mendationsare not availablein Fiji, but a startat reformcan be made using the informationand the modelsat hand. The reportproposes five measures:

(a) removalof quotasand quantitativ.;restrictions;

(b) reducingthe tariffceiling;

(d) simplifyingand extendingthe duty drawbackscheme to benefit exportersof manufactures;and - iv -

(e) raisingthe basic tariffon importedinputs from 7.5% to 10X so as to les;eneffective protection and maintaingovernment revenues.

Tax Policy

An adjustmentof tariffscould go hand in hand with tax changes designedto encourageinvestment which maximizedomestic resource mobiliza- tion. Currently,over 40% of total revenues(amounting to a quarterof the GNP) are derivedfrom directtaxes and cross countrycomparisons indicate that the effectivelevels of incometax in Fiji are well above those in the dynamic Eart Asian economies. It is possiblethat the systemdampens incentives, discouragessavings and is responsiblefor the evasionthat is reportedamong taxpayersfrom the upper incomebrackets. Severalob,zctives must be met as closelyas possible. Manufacturinginvestment needs stimulating;private savingsthat have stagnatedfor aeveralyears must be inducedto rise; the Governmentmust augmentrevenue buoyancy not the least becauserates of public savingshave been gravelyeroded by increasedoutlay on administrative expenses;and any reformmust not exceed the organizationalcapability of the tax departmentor place too great a strainon taxpayercompliance.

Implementingany tax scheme inevitablyentails compromises as realitycan often be contemptuousof tax principles. Recognizingthis, the reportadvances a number of proposalswhich may pass the economictest but whose politicaland administrativefeasibility will requirea close scrutiny:

(a) a movementtowards lower marginalrates, with fewer bracketsand less exemptionsmight be considered,as there could be gains in terms of efficiency,incentives and simplicity;

(b) householdsavings could be stimulatedthrough higher tax exemptions for interestand capitalincome;

(d) a systemof excisetaxes applieduniformly to importsas well as domesticproduction and a movementfrom specificto ad valoremrates for excise taxes;and

(e) the level and buoyancyof governmentrevenues could be preservedby introducinga broad based retailsales tax carryinga low uniform rate and coveringalmost all items,including foodstuffs.

ExportPatterns

Among developingcountries, the East Asian NICs have mountedthe most impressiveand recession-proofexport campaigns. It is significantthat the bulk of their exportsare manufacturedgoods, sold largelyin the US, Japan,West Germanyand Britain,with the US absorbinga third or more. Severalof these countrieshave close links with multinationalcorporations which have providedcapital, technology and tradingcontacts. An examination of the East Asian exportpattern shows that light manufacturessuch as *- v - clothingand textilesremain very prominentbut that the fastest-growingitems are electronicproducts, telecommunication and officeequipment, machinery, and transpoirtequipment. The MICs are formidablecompetitors in a range of productswhere experience,labor skills,technological proficiency and market penetrationtechniques are critical.

To challengethe NICs ir.the areas they have so effectively colonizedmight be difficultfor a newcomer,but Fiji might profitablyimitate certainaspects of the ;ast Asian export strategy: (a) as wages in Fiji's manufacturingsector approach those of the SoutheastAsian economies,the gains from trade in light manufactureswhere the value added is low would be scantyand the growthprospects meager. The most fruitfulapproach would be to find niches in ouch areas as electronics,light machineryand fashion garmentswhere small-batchproduction is needed and emphasison designas well as qualitycan gain entry; (b) it is likelythat the Pacificregion will continueto increaseits lead as the most dynamictrading area and countries such as the US and Japan may offer the best market opportunitiesfor years ahead; (c) a numberof countrieshave used preferentialtrading agreements as a springboardfor manufacturedexports. Fiji is a memberof twot LOME-ACP and SPARTECA. The access they provideneeds to be exploitedmore fully) (d) Export ProcessingZones (EPZs)have in the past servedas steppin3stones to manufacturingdevelopment. Although their proliferationthroughout the world, togetherwith a slackeningoi direct foreigninvestmert has diminished their attractionfor planners,the creationof a suitableEPZ, so long as it does not entail a large outlayon infrastructure,could be a usefulaccompani- ment to Fiji's trade agreements;(e) not just factorsupplies but nontariff barriers,automation, the preferencesof MNCs and exchangerate policiesnow affect a country'scomparative advantage and the geographicaldistribution of product-onfacilities in today'shighly integratedtrading milieu. Some of these are beyond the controlof plannersin Fiji but exchangerate policy, flexiblyused, remainsa potent instrument. East Asian NICo, in spite of the constraintsimposed by small size, have been able to furthertheir industrial and exportambitions through a deft handlingof exchangerates. The Fijian authoritiesmight find that fasterresults might be obtainedif industrial incentivesare orchestratedwith actionson the exchangerate front.

Labor Policy

The wages of unionizedworkars in Fiji'surban sectorreflect the country'sUS$1,700 per capita income,whereas industrial productivity lags behindthat of middle-incomecountries in East Asia. There are those who believethat labor market conditionsrule out the possibilityof competitive manufacturedexports on a substantialscale. On closerexamination, the problemdoes not seem so acute. The market for industriallabor is dualistic, with nonunionizedworkers being paid wages that are appreciablylower. Export-orientedgarment producers tap the informallabor marketand are apparentlynot being hamstrungby high wage costs. A reviewof trendsreveals that real industrialwages have been fallingthroughout the eightiesand with unemploymentexpected to worsen, the pressureon informaland union wages will, if anything,be stronger. As in Europeancountries with a good record in terms of wage inflationand work days lost throughstrikes, Fiji has, in the TripartiteForum, an institutionalmeans of centralizingand negotiating - vi - wage claims. This mechanismfor bringingtogether the unions,employers and the Governmenthas functionedwith tolerablesuccess and was the vehicle for introducingvoluntary guidelines in the early eighties. When the Nicoll and Hurst wage settlementfor publicservants threatened to spark a public sector- led wage spiral in 1984-85,the Governmenttook decisiveaction by declaringa freeze. In effect,the systemfor regulatingwages in Fiji lies betweenthe Europeanmodel and the more interventionistmode favoredby some of the East Asian NICs.

The problemin futuremay have less to do with wages than with the availabilityof skills. Export-ledindustrialization along the lines it would be desirablefor Fiji to pursuewill be accomplishedonly if employerscan mobilizeadequate supplies of human capital. Entry-levelvocational skills are not scarce;the higher order and specializedones are. Two remedial policiesdeserve consideration. Currently emplcyers pay a compulsorylevy into a trainingfund but are poorlyserved by the existingarrangements. The needs of an acceleratedindustrial program will be even harderto satisfy. The trainingbottleneck requires the attentionit has received,for instance, in Singapore. That countrycan also be a model for the developmentof technicaicapabilities in the fieldsof engineering,electronics and design, which take time to accumulateand are a preconditionfor attainingindustrial maturity. Singaporehas been unusuallysuccessful in persuadingwestern governmentsand MNCs to providefunding and technicalassistance for centers importingindustrial skills. This might be a routeFi-Ji could explore.

Information,Design and Marketing

Stimulatinginvestment through fiscal, monetary, tariff and exchange rate policiescan give industrializationthe necessarypush. But successof the kind enjoyedby tht East Asian NICs has dependedupon progressin four crucialareas: intelligencegathering, marketing, produLt design and technologicalabsorptivity. Chapter VIII of the reportdescribes the layered informationsystem developed by Korea and then proceedsto examinethe factors behindthe market performanceof certainproducts that might eventuallysuit Fiji'sexport capabilities: rum, watches,lig1t machinery, fashion garments and horticulturalcommodities.

A comprehensivesystem for collecting,processing and disseminating informationtakes time to perfectand involves1arge expense. Fiji need not set its sightson an elaborateKorean-style edifice but even a basic structure providingmarket feedbackand informationon Fijianproducts to potential buyers,will take time to establish. Exportsaimed at nichesmust be expertly tailoredfor the marketsthey are supposedto satisfy: they must be designed and promotedwiti the clientelein mind and carry the stamp of quality. Design skills,consistent and superiorquality standards, and marketing expertisedecide the fate of high value-added,niche-filling exports. They are the resultof gradualaccretion, but design schools,licensing, and foreigncollaboration can acceleratewhat is inevitablya complexprocess. FIJI

A TRANSITIONTO MANUFACTURING

I. INTRODUCTION

1.01 Neithergeography nor culturenor politicalforms precludea country from sharingin the fruitsof economicadvance; what mattersis the skillwith which a constellationof policiesmobilize domestic resources for development ends and enable the nation to grasp the tradingand technologicalopportuni- ties that have underwritteneconomic gains most notably in East Asia. When the local market is small,growth in productionwill inevitablydepend upon sale overseas. But export demandis an exactingtaskmaster calling for the fusionof entrepreneurship,technology, price competitiveness,marketing and productdesign. Policiesmust exert pressureon two axes: motivating businessmento manufacturea competitiveproduct in the requisitequantities; and ensuringan adequatesupply of human and physicalcapital to financeplant and infrastructureand forge the human links in the chain which leads from technologyabsorption all the way to the marketingof a qualityproduct. Capitalaccumulation can, for a time and in part, be financedthrough foreign savingsand direct investment,but in due coursethe full burd^nof investment must come to rest upon nationalsavings. Generatinga level of savings sufficientto financeexport-led growth of a high order becomesthe final test of strategycalculated to achieverapid developmentwithout departing over the medium-termfrom the requirementof domesticresource equilibrium.

RecentPerformance

1.02 Among the island countriesof the south Pacific,Fiji is the most populous. It is also richer than any of its neighbors,a tributeto the efficiencyof the sugar industry,the lure of Fiji'stourist resorts and the country's resourcesof timberand gold. Favorablecommodity prices and expanding sugarproduction permitted the GDP to grow by 5.8% p.a. in the first ha:f of the seventies. In the latterpart of the decade,earnings from sugarwere squeezedby a period of depressedprices, but substantialpublic investment made possiblea tolerablerate of expansionand steady,albeit small,gains in per capitaincome. Sugar lived up to its reputationas being a crop generat- ing one of the handsomestnet returnsper hectare,- and, in spite of its high

1/ James Fry, Sugar: Aspectsof a ComplexCommodity Market, Commodities and ExportProjections Division, Division Working Paper No. 1985-1,World Bank, 1985. Sugarcane'sattractiveness is enhancedby its abilityto withstandflood, droughts, hurricanes, pests and disease. As G. Hagelbergnotes, "the misery of growingcane is nothingcompared to the misery of growingother crops."in G. Hagelberg,Sugar in the Caribbean: TurningSunshine into Money, in S. Mintz and S. Price eds., CaribbeanContours, Johns HopkinsPress, 1985. importcoefficient, tourism made a valuablecontribution to the island's economy.

1.03 The weaknessesof a strategytied to a cycle-proneprimary commodity and a serviceindustry that is hostageto the whims of novelty-seekingvaca- tionersfrom a handfulof industrialeconomies as well as the corporategoals of foreignairlines, hotel and touristcompanies were revealedin the first half of the eighties. All the developingeconomies located on the Pacific perimeterhave been buffettedby the erraticbehavior of raw-materialprices and the reducedtempo of world trade,but, whereasthe exportersof manufactured goods seem to be regainingtheir stride,those that had peggedtheir hopes to the marketabilityof their primaryexports are faced with lean times.

1.04 Since the beginningof the decade,the performanceof the Fijian economyhas fallenfar short of expectations.Growth of GDP has averaged1.2% p.a. and it has been propelledby the servicessector rather than throughan expansionof directlyproductive activities. With public sectoremployment leadingthe way the value of servicesgrew 1.9X p.a. between 1979 and 1986. Meanwhile,agriculture rose by only 1.31,while sugar declinedby almost 11 p.a. Industry'scontribution, including manufacturing, to value-addedwas the same in 1986, as in 1979. With populationincreasing each year by almost21, real per capita incomeshave declinedsteadily (see Table 1.1). A surge in touristarrivals combined with a recordsugar harvestpushed the growthrate above 91 in 1984. At other times,the economyhas stagnated,with investment and savingshowing the effectsof diminishedeconomic momentum. Investment which peakedat 271 of GDP in the late seventieswas below 18% in 1985, while the share of domesticsavings in GDP was a meager 16% as againstan averageof 21.31 p.a. during 1975-80. A gap betweencapital spending and national savingsthat appearedin the early eightieshad largelybeen erasedby 1985, when the currentaccount deficit was a littleover 4X of GDP, but adjustment has been achievednot throughhigher savingsand exportsbut by accepting lower levelsof investmentand a negativegrowth rate. Table 1.1: FIJI -- BASIC ECONOMIC INDICATORS

1970-75 1975-80 1980-84 1985 /d

GDP growth 5.8 4.1 1.2 -3.0

PopulationGrowth Rate n.a. 2.2 2.0 2.0

Unemployment Rate 10.2

Cross DomesticInvestment (as X of GNP) 20.4 26.1 22.7 17.5

Gross NationalSavings (as X of GNP) 17.1 /a 21.3 16.3 13.3

CurrentAccount Deficit /e (as X of GNP) 4.7 /a 4.8 6.4 4.2

OverallBudget Deficit (as X of GNP) 1.1 /a 6.5 /b 5.1 /c 4.8

Ia 1971-1975only. 71 1976-1980 only. 7c 1981-1984only. 7d Estimates. 7e Excludingofficial transfers.

Source: World Tables,World Bank; Ministryof Finance,Fiji.

1.05 A responsethat would reversethe economicslide has been slow in materializingfor a varietyof reasons. Per capitaincomes exceeding US$1,800,a relativelyequitable income distribution and a moderateratio of labor to arable land ensuremost Fijiansa measureof well-being. The elastic natureof rural villagesociety has made it possibleto absorb into part-time subsistenceagriculture many of those who are unable to find or hold jobs in the urban centers. Hence, the pressuresemanating from a sc city of employment opportunitiesin the urban economyhave been contained.-Iuntil recently,the alternativeof livingin villagecommunities kept unemployment

2/ Continuedreliance on subsistenceagriculture by workerswith urban jobs is describedby E. Ponter,"The Growth of Wage Labor and its Consequences for a FijianVillage" in Fijiansin Town, Instituteof PacificStudies, Suva, 1986. Urban work opportunitiesfor Fijiansand the natureof the unemploymentproblem are discussedby C. Gounis and H. J. Rutz, "Urban Fijiansand the Problemof Unemployment,"in Fijiansin Town, op. cit. -4- in the 6-81 range,even thoughthe formalsector has accountedfor a decreasingpercentage of the labor force since the late seventies. Some of the urgencyof findingwork for a populationexpanding at the rate of 21 p.a. has been bluntedby the singularcapacity of long-maintainedrural social institutionsto satisfy,at least partially,the materialand cultural aspirationsof the young. But a push towardsexport-oriented manufacturing activitieshas also been thwarted,first by the profitabilityof low-risk importsubstitution and by the beliefthat Fiji's geQgraphicallocation, its smallsize and costlyunionized labor force rule out any chancesof effectivelycompeting against East Asia's seasonedexporters.

1.06 Experiencegained during the eightiesargues against the tying of futureeconomic progress to 37e residualpotential of sugar. A shrinking supplyof uncultivatedland - suggeststhat subsistenceagriculture will have difficultyin satisfyingthe aspirationsof new entrantsand the increasein open unemploymentto 10% in 1985 reflectsa reluctanceon the part of workers to eitherwithdraw from the labor force or acceptthe sacrificesin terms of personalfreedom which a returnto the safetynet of traditionalsociety would entail. Finally,the foraysby Fijiangarment makers into international marketshave been sufficientlyrewarding to raise hopes,and they lend credibilityto P developmentstrategy assigning a major role to manufactured exports.

1.07 Taken together,these three factors--fluctuatingearnings from sugar and tourism,worsening unemployment and the crystallizationof new export possibilities--callfor a reshufflingof strategicoptions. Fiji can ill affordfurther erosion of per capita incomes. Given the pricesand demand being projectedfor sugar,timber and coconutoil (see Table 1.2),the likely trendin touristarrivals and the scope for manufacturingdevelopment keyed to importsubstitution, it will be difficultto attain a rate of growth that will lead to rising livingstandards very much beyond the horizondefined by DP9. To reviveeconomic performance, the countrymust createthe environmentand

3/ FIJI: PROPORTIONOF UNUSEDLAND BY CLASS, 1980

Viti Vanua Levu Levu Other Type of Unused Land ( ((Z)Y

Suitablefor arableor permanentcrops with littleor minor improvements 3 6 1 Suitablefor pastureor afforestation, no improvements 3 4 3 Usablewith major improvements 10 24 17 Unsuitable 84 67 78

Source: Gerald Ward, Fiji'sLand Use System,1980 monograph. - 5 -

skillswhich are the ingredientsof a competitivemanufacturing sector and becomeadept at the game being played for ever-higherstakes by a group of East Asian economies.

Table 1.2: COMMODITYPRICE PROJECTIONS

1986 1990 1995 2000

Sugar ($IMT) Nominal 154.00 390.00 380.00 430.00 Real (1985 pr5ces)Ia 136.28 321.31 265.00 252.94

Logs ($/MT) Nominal 150.00 164.56 206.50 253.30 Real (1985 prices) 132.74 136.00 144.00 149.00

CoconutOil ($/MT) Nominal 280.00 518.00 717.00 819.00 Real (1985 prices) 247.79 428.10 500.00 481.76

/a Deflatedby the MUV.

Source: CommodityDivision, The World Bank.

The Perspectiveof DP9

1.08 The Government'sNinth DevelopmentPlan (DP9) clearlyrecognizes the importanceof an export-orientedmanufacturing sector for the long haul (see Table 1.3). But over the balanceof the eightiesit expectsmuch of the impetusfor a growth rate averaging5% p.a. to be derivedfrom tourism,sugar and other agriculturalproducts. Neitherthe sugar industrynor tourismis by any means a spent force but even if we gloss over the weather-relatedproblems of 1985, the trendsperceived do not entirelysupport the scenariopainted by DP9. Sugar is likelyto be hemmed in by productionand price constraints, whereasa 11.5% p.a. increasein touristarrivals through the remainderof the eightiesmight be hard to sustain. Sugar and tourismremain the sectoral favoritesof DP9 in spite of their checkeredgrowth recordmainly becauseof their familiarity,but the time may be at hand when manufacturingindustry might be betterplaced to help fulfillthe overallgrowth targets. Korea's dramaticleap towardsindustrialization in the early sixtiesused as a springboardlittle other than a rural economywhich land reformand production policieshad renderedefficient, egalitarian and politicallysupportive of the coherent,long-term, industrial program adopted by the Governmentof President Park. Becauseagriculture figures prominently in the KoreanGovernment's politicaland economiccalculations, close attentionhas been given to the rural terms of trade,which indirectly,reinforces a commitmentto a flexible exchangerate policy stemmingfrom Korea'sexport-led strategy. - 6 -

Table 1.3: DP9 PROJECTIONS

1980- 1985- 1985 /a 1990

P.A. Growth Rates: GDP Growth (at factorcost) 1.2 5.0

Sugar Production -2.9 7.4 TouristArrivals 3.9 11.5

ManufacturingValue Added -0.2 6.2 Sugar -3.0 7.9 Other Food, Drink and Tobacco n.a. 6.9 Other Manufacturing 0.9 5.0

Pop. g.r. 2.0 2.0 Labor Force g.r. 2.9 2.4

1985/a 1990

GDI/GNP 20.0 21.8 GNS/GNP 16.6 19.2 CAD/GNP 3.4 2.6 BudgetDeficit/GNP 5.5 3.4

Population('000) 699.0 772.0 UnemploymentRate 10.2 7.2

/a This columnreflects estimates at the time of the preparation of DP9, and not actual data.

Source: Fiji'sNinth DevelopmentPlan, 1986-1990,Central PlanningOffice, Suva, November1985, pp. 13, 49, 89.

1.09 The experienceof the Caribbeaneconomies, some of which are in a bettergeographical position to capitalizeon an expandingtourist trade, shows the limitsto developmentconfronting island countries that do not diversifyaway from an inheritedsugar cultureand the largelyforeign con- trolledbusiness of servicingtourist enclaves.

1.10 Fiji'sprincipal trading activities are too importantto be casually brushedaside. Hence the chapterwhich followsclosely examines their pros- pects. The reportthen describesthe elementsof an industrialstrategy that might be suitedto Fiji's institutionalconditions, its entrepreneurialcapa- bilitiesand the skillsof its labor force. - 7 -

II. THE FUTUREOF TRADITIONALINDUSTRIES

Sugar

2.01 The market for sugar is a troubledone. In 1985, almost 100 million tons were producedworldwide and about a third of this volumeentered world trade,much of it in the form of raw sugar shippedin bulk form. Nearly 84X was sold under contract,the balanceon the free marketwhich, because of its small sise, has traditionallybeen subjectto pronouncedfluctuations taking the form of five-year-longtroughs interspersed by sharp peaks. One such peak favoredproddcers in 1984 but a year later,large harvestsand the accumula- tion of stocksequal to almost half the annualconsumption pushed prices to their lowestpoint ever, less than three cents per pound. With all the major exporterssheltered by contractsnegotiated under a higher price regime,the industryweathered the crisisand Sovietpurchases had pulledprices past nine cents per pound by mid 1986. But with productioncosts now averaging15- 16 cents per pound,the latestdownswing in the sugar markethas once again exposedthe vulnerabilityof countrieswhose fortunesare linkedto those of sugar.

2.02 As a producerFiji occupiesthe middle rank.41 The best year for the industrywas 1984 when outputreached 486,000 tons. Yield of cane per hectare is about average -- 40-50 tons per hectare -- as against 100 tons in Hawaiiand Australia(for basic statisticson the sugar industrysee Annex Table 1). But manual cutting,which allowsthe juiciestpart of the cane just above the groundto be retrievedand reducesthe amountof extraneousmatter, has enabledFijian farmersto achieveproduction costs that are among the lowest. A successionof hurricanesseverely damaged the crop in 1985 reducing the outputto 343,000tons but projectionsfor 1986 forseea reboundto 420,000tons. The DP9 takes a sanguineview of the futureand anticipates that between550,000-600,000 tons of sugar will be producedby 1990 from 78,000hectares through a moderateannual increasein cane yield per hectare

4/ Sugar arrivedin Fiji by a circuitousroute. Domesticatedin Papua New Guinea it was first processedin India from where the sugar culture spreadto the Mediterraneanregion, then to the Americaswith the help of Columbus. In the 1860s sugarcanewas introducedto Fiji followingthe collapseof the cottonmarket after the close of the AmericanCivil War. After a falteringstart-the initialattempts to producesugar in Wakaya failedbecause of low yields and disease--theindustry finally establisheditself in the 1880s which is when the ColonialSugar Refining Co. of Sydneystarted operations at Rewa. See S. R. Mintz,Sweetness and Power,Viking 1985, p. xix; R. GerardWard, Land Use and Populationin Fiji, HMSO, 1965, p. 25; and J. Narayan,The PoliticalEconomy of Fiji, South PacificReview Press,1984, pp. 21-31. - 8 -

and an appreciableimprovement in the extrsctionratio.51 From the perspec- tive of developmentstrategy, two questionsare paramount. Can sugar continue to serve as a dynamicleading sector during the next five years and beyond? What is sugartspotential as an earnerof foreignexchange?

2.03 The DP9 arrivesat its targetfor sugar productionin 1990 by assum- ing a 101 increasein cultivableacreage, a 20% growth in cane yield per hectare (from 50 tons to 64 tons) and a downwarddrift in the tons cane: tons sugar ratio (TC/TS)from 8.43 in 1986 to 8.17 in 1990. These certainlyare within the realm of possibilitybut they do approachpast trendsand current constraintsfrom an optimisticangle. Between 1974-76and 1984 yield of cane per hectarerose by nearly 3.0% p.a. but a deterioratingTC/TS ratio held the increasein sugar output to 0.6% p.a. A continuationof past trends,2.9% p.as growth in sugar cane productionand assumingno furthererosion in the TC/TS ratio,would indicatethat productionwould be less than 500,000tons of sugarby 1990. A figure far short of the DP9 target.

2.04 Extrapolationneglects the changingstringency of constraintsthat could interferewith intensivefarming. Fiji has 22,000cane growers,a high percentageof whom are cultivatingsmall, uneconomic holdings. In fact, some 13,000contracts deliver an averageof just 200 tons of cane with many selling under 100 tons. This reflectsthe proliferationof backyardcane farms and the inefficientpractices of small producers. To remedythe situationwhile keepingover 70,000hectares under cane would requireinstitutional changes that facilitatea consolidationof holdings. The existingtransport system has littleexcess capacity and would not be able to handle largersupplies withoutextensive modernization.

2.05 To encouragefarmers to raise the qualityof the land might also call for institutionalamendments that determinethe lengthand securityof leaseholdproperty. It is a commonplacein agriculturaleconomics that tenant farmersfaced with the risk of having their leases revokedare unlikelyto make the investmentof time and capitalthat will enhancesoil fertility. To the extentthat a proportionof the most efficientcane farmersfear the even- tual loss of their lease-holdproperty or expectto abandonfarming, maintain- ing the qualityof the soil will not be an importantconcern.

2.06 A systemof smallholderagriculture and landholdinginstitutions which vest inaligyableownership rights over 83% of the cultivableland in 6,600 mataqalis_ resultsin productiontechnology and a rural land market that holds down the yield of a crop known to thrivebq7t under the tightly regulatedconditions found on commercialplantations.-, But -he change in

5/ The extractionrate is commonlyreferred to as the tons cane/tonssugar (TC/TS)ratio.

6/ An agnaticalyrelated social unit usuallya lineageof the larger clan.

7/ See E. Grahamand I. Floering,The Modern Plantationin the Third World, Croom Helm, 1984. - 9 - yields is also affectedby the plightof the Fiji Sugar Corporation(FSC). A parastatalorganization which absorbedthe assetsand functionsof the Australianowned Colonial RefiningCorporation in 1971, the FSC has been vitallyinvolved in every facet of sugar production: from the issuingof contractsto farmersto the export of raw sugarsthe FSC is the driving force behindthe sugar industry. If it falters,so does the industry. Lately,its financialhealth has been under siege and its financialdifficulties could affectthe futureof sugar production. The FSC is responsiblefor research, extension,the distributionof fertilizerand the applicationof pesticides. It manageslarge experimentalacreages, runs a rail systemthat collectscane and owns the mills where it is crushed. Low sellingprices have been espe- ciallydamaging to the FSC becauseFijian farmers enjoy unusuallygenerous contractualterms, receiving 72-75% of the sellingprice as againsta 40-50% averageelsewhere (60%, in Australia). Its operatingcosts have been creeping upwardssince it employs expensiveunionized labor on its plantationsand factories. The longerthe slump in pricespersists, the harder it is for the FSC to sustainthe scale and tempo of its researchand extensionefforts which stronglyinfluence yields and varietalstandards that are a determinantof the TC/TS ratio. Nor can the FSC find the resourcesto replaceits decaees-old crushingequipment which, althoughkept in a good state of repair,detracts furtherfrom the efficiencyof the productionsystem.

2.07 Each of the constraints described can be eased, but the institutions governingland distributionare not lightlytampered with and the financial problemsof the FSC cannot be fully resolvedin the absenceof a recoveryin prices. Extendingpast trends in productionthrough the latterhalf of the eightieswould seem, therefore,to definethe boundsof possibility. To projecta growthrate that is even highermay not be warranted.

2.08 Even if production follows the path delineated in DP9, what prices would sugar exportsfetcht Over the past five years Fiji has sold three fourthsof its sugar under favorablypriced contracts and the balanceon the open market. Contractedsales have provideda sorelyneeded umbrellain times of slump but two- to three-yearcontracts insulate the produceron a temporary basis only. Over the long term, the averagecontracted prices track those being registeredin the market. Furthermore,if Fiji'sexports rise well above a half milliontons per annum,it is likelythat not much more than 50% could be disposedof on contractedterms becauseof the purchasequotas enforcedby the EEC and the US.

2.09 The volatilityof sugar pricesis legendaryand is largelyexplaqed by the thinnessof the free marketrelative to total productionand trade._ Small changesin supplycan spark dramaticprice movements. Equallycon- spicuous of late is a downward trend in real prices caused by enduring changes in the use and productionof sweeteners,worldwide. Sugar consumptionrose only 4.3% p.a. between 1980 and 1984 but total sugar supplyspurred by higher productionin Australia,the EEC, China and India increasedby 19%. Global stocksin 1984 were twice the level of 1980 and 36% of total use. Further,

8/ About 16% of the sugar producedis sold on the free market. - 10 - annualconsumption as a percggtageof world supplywas down 10 pointsat 742 as comparedto that in 1980.o By far the biggestconsumers are the developed countrieswith the US in the lead. But in both the US and the EEC per capita demand is reachinga point of saturation. Incomeelasticity could soon be close to zero in the US and droppingbelow 0.2 in the EEC. Demand should continuegrowing in Japan and the centrallyplanned economies, but even in these countriesper capita usage is quite high. Consumptionwill expandmost steeplyin the developingeconomies of Asia and Africa,although this is predicatedon foreignexchange availability.

2.10 Dietarypatterns, health concerns,high prices in the developed countriesand effectivepurchasing power will be one set of factorsinfluencing sugar demand;a seconddeterminant w.ll be the acceptabilityand pricesof high fructosecorn syrup (HFCS)and artificialsweeteners such as saccharin and aspartame. HFCS has alreadymade deep inroadsinto the US market because of the abundantsupplies of competitivelypriced corn, the ease with which corn gluten can be disposedof as animal feed, facilitiesfor providing constsnttemperature storage and, most important,the receptivityshown by the country'slarge bakery-goods,beverage and confectionaryproducers. Corn syrup has annexed382 of the US sweetenermarket. It is graduallyenlarging its share in the EEC and shouldgrow in line with the consumptionof sweet beveragesin Japan. Saccharin,which is cheap to producehas occupieda securecorner of the market for decadesbut, with the appearanceof af8 rtame and technologicaladvances in the designingof artificialsweeteners,- the use of sugar substituteswill expandmore rapidly,not only in countrieswhere sugar is no longerof significanceas a sourceof caloriesbut also in cost- consciousdeveloping economies as well. A tenth of Thailand'sdomestic market has been capturedby artificialsweeteners. When the variouselements that constitutethe long run demand curve are broughttogether, world consumption of sugar cane is projectedto grow by 1.7Z p.a. until the year 2000.

2.11 While demand seems set to expandat a moderatepace, there is scant reasonto believethat the supply situationwill evolve in a manner that would appreciablymodify price trends. Productionin the US may declineby somewhat under a milliontons to 6.8 millionby 1990 as high cost growerswithdraw from the industry. Further,the hurricanewhich devastatedCuban plantationsin 1985 will probablydepress output for two to three years. This could be more than offsetby increasedproduction in Brazilwhere the cane-basedalcohol programhas been rendereduneconomic by fallingoil prices. Since 171 of all Braziliancars run on pure alcoholand all of the rest use gasohol,an abrupt switchaway from ethanolcan be ruled out, but if a large acreageremains under cane the emphasisis bound to shift towardssugar so long as the projections call for low petroleumprices well into the nineties.

9/ R. A. Packenham,The CaribbeanSugar Crisis: Consequencesand Challenges,Journal of InterAmericanStudies and World Affairs,1986 Vol. 28, No. 1.

10/ J. Elmsley,"Designing Sweetness to Order,"New Scientist,October 31, 1985, p. 22. 2.12 Output in the EEC is expectedto remain stableat 13.6 million tons for the severalyears with productivityimprovements in the beet sugar industry cortinuingto counteracta downwardmovement in price supports. In East Asia, Thailandis aiming at a targetoC 3 milliontons by 2000 as comparedto 2.5 milliontons in 1984, and a settledpolitical climate should encourage Filipinofarmers to regainground lost during 1985-86when productionplum- meted from 2.5 million tons (in 1984) to 1.3 milliontons.

2.13 Market analystsexpect that the overhangof stockswill diminishby 7.5 milliontons between1985-e7, with inventoriessettling at the 251 level considerednormal in the industry. But longerterm supplyand demandcondi- tions presagefree-market prices that will remainbelow long run averagepro- ductioncosts. The World Bank projectsthat sugar will be sellingfor an averageof about 13 cents (in 1985 prices)over the next one and a half decades($253 per ton). Contractsales to the US and the EEC will offer higherreturns as stabilizationprograms in these two marketsshould result in pricesthat are respectively20X and 13% above world levels. However,US importsare not expectedto increaseby more than 100,000tons p.a. through the yeal 2000, nor is there much likelihoodthat the EEC will, other than marginally,augment its purchases.

2.14 As an engine of growth,beyond the currentdecade, the Fijian sugar industryseems an uncertaincandidate unless the institutionaland financial problemsthat have surfacedduring the eightiescan be resolvedin the very near future. It will generatealmost no additionalemployment. With price forecastsover the next decadebeing pitchedat levels that are not very far removedfrom Fiji'sadmittedly low productioncosts, sugar will have difficulty retainingits reputat.onfor profitabilityand as a sourceof foreign exchangeearnings (see Table 1.2).

Tourism

2.15 A number of islandeconomies have soughtto developtourist industriesbecause scale, marketing and skill-relatedbarriers do not loom as large as they do for manufacturing,and frequently,overseas interests can be persuadedfirst to establishthe hotel infrastructureand then to manage it. The spectacularincrease in touristtraffic during the seventiesfully vindicatedthe expectationsof the earlierentrants into this field,but the multiplicationof destinationsoffering an undifferentiatedexperience over the past ten years, combinedwith a slackeningin the growth of tourist travel,has tarnishedthe industry'sattraction. In the jargonemployed by the trade,sun, sand, sea, sightseeingand shoppingcan be found in scoresof appropriatelysanitized locations around the world. Iaternationalhotel chainsensure relativelyuniform standards of serviceand accommodationwher- ever businessdraws them and touristcompanies strive to give the travelersa minimumpackage of attractions. In such circumstances,it takes inspiredmar- ketingand unusualcultural appeal alongside the active collaborationof - 12 - airlineand travelcompanies 1l1 for one touristcenter to pull far ahead of the pack. Sustainedsuccess has come to a few places,for exampleSingapore and Bali, but even these islandshave felt the effectsof slowingtourist traffic.With modernization,Singapore has lost its distinctAsiatic character. Malay kampongshave been displacedby tower blocks;Chinatown has shrunkto a few antisepticlanes, local artisanalskills are on the wane and neitherthe food nor the remainingdancing troupes carry the mark of cultural distinction.These changesare discouragingAustralians, Americans and Europeansto visit Singaporeagain althoughAsian tourists,who are impressed by the city's cleanliness,remain more enthusiastic.Bali, long a mecca for the discerningtourist, is also feelingthe pinch of maturitywith arrivals beginningto stabilizeas fashionand the quest for new experiencestug the touristto other locations.

2.16 Tourismranks secondonly to sugar as an earnerof foreignexchange and a healthyperformance by this sectorduring the Ninth Plan periodwould undoubtedlyhelp financemanufacturing diversification. Whether it can con- tributesignificantly to growthas assumedby DP9 is less certain. Gross receiptsrose to F$161 mil.ion in 1984 from the averageof F$136 millionin 1982-83on the strengthof 235,000arrivals. Net receipts,after all imports by the indust.yare deducted,along with repatriatedprofits and dividendsa- mountedto a third of this. Directemployment by the touristsector remained stableat around 6,700 between1977 and 1984 (although300 new rooms were added, king a total of 3,900)while indirectemployment rose from 3,000 to 9,000,

2.17 The tourismindustry of Fiji is at a watershed. To maintainan 11.5 p.a. growth in touristarrivals through the remainderof the eightiesas indicatedin DP9 and to lessenthe leakagein touristexpenditures, an ambitious,costly and risk-ladenstrategy will need to be introducedin short order. But whateverapproach is followed,obtaining results will be an uphill task. To divert the flow of long-ha,-ltourists from other resortsto Fiji requiresa distinctcultural appeal over and above the standardtourist package. It calls for shoppingand hotel facilitiesof a superiorclass and an abundanceof leisureactivities, especially sightseeing. Fiji is deficientin each of these areas. The local culturehas not been tailoredand projectedso as to attracttourist attention as has been done in Hawaiiand Bali. Premier hotels,of which there are 3 with a total of around 1,000rooms, provide respectableservice but have yet to attain the exacting5-star standards taken for grantedelsewhere in East Asia. Touristsseeking bargains will not find them in Fiji'sduty-free shops, and the infrastructureneeded to support sightseeingand water sportsis poorlydeveloped. Buildinga few more hotels

11/ Travelagents sell three out of four ticketsbought by air travellers, thoughthis could changewith innovationsin ticketmarketing, e.g., automaticticketing machines.

12/ Hotelsin Westerncountries employ 7-8 staff per 10 rooms. Becauseof the layoutfavored and the extensivegardens and beach facilities,the employmentrate in Fiji is almosttwice as high. - 13 - could ease the shortageof rooms in the five-starrange duringthe peak season,but it is not likelyto enhancethe country'sappeal as a resortand such a move could depressthe averageyearly occupancy ratio, which is in the regionof 58%. Investmentin infrastructurefor entertainment,if it is suit- ably lavish,could yield acceptablereturns, but only if advancesare achieved on severalother fronts.

2.18 Currentlythe Visitors'Bureau spendsF$1 millionon overseas promotion,leaving much of the marketingin the hands of foreigntour agencies. With hundredsof destinationsover which to distributetheir customers, these firms quite naturallyput no particulareffort into extollingFiji's attractions.Moreover, Japanese travel agencies, which tend to direct tourist trafficto locaticnswhere their affiliatesor JAL have hotel investments cateringto the needs of the Japanesetraveler, might discriminateagainst Fijianresorts. If the touristmarket expands at a relativelyslow pace in the late eighties,a countrythat does not take a lead in marketingits culturaland scenicassets cannotbank upon a large growthin tourist arrivals. An aggressivemarketing campaign covering the Europeancountries, Japan,the US and Australiawould be expensiveand would have to deliveron the promisesmade -- an even more costly undertaking.

2.19 Remote islanddestinations are at the mercy of major international airlines. Their flightarrangements determine the potentialtourist traffic. When the range of passengeraircraft was limited,Fiji was admirably situatedto serviceflights headed towardsAustralia. The advent of long- range jets and the emphasisthat trans-Pacificcarriers now placeon serving the "Pacificgateway" cities means that fewerairlines call at Nadi and those that do dedicateonly a limitednumber of seats specificallyto serving Fiji. In the absenceof a surge in touristinterest, the carrierswill be un- willingto raise the frequencyof their flights. An internationalairport near Apia in WesternSamoa may even divert some trafficfrom Nadi'sexpensive facilitieswhere the landingfees are secondonly to those leviedat Narita. Airlinesmust be persuadedto give more attentionto Fiji and it is not clear that increasingthe nqmberof first-classhotel rooms,as proposedunder DP9, will be sufficient.- Hotel space is not the bindingconstraint on the growthof touristtraffic to Fiji. Only an active promotionalcampaign paralleledby a substantialupgrading of facilitiesthat is seen to produce resultsis likelyto be convincingfor the Pacificcarriers.

2.20 For a countrythat taps the smallmarket segment(16X) comprisedof long-haultourists it is importantto attractthe right kind of visitor. Those who seek five-starhotel accommodation,spend lavishlyon localhandi- craftsand services,and stay for a week or more, bring the most revenue. So far, only a minorityof the travelersto Fiji fit this description.The typi- cal touristis a youngAustralian with his familywho spendsnine days sequesteredin a hotel enclave,organizes his time aroundbeach activitiesand is quite circumscribedby a modestvacation budget. Americanswho recently

13/ Currentlythe constructioncost per room is F$100,000. It was F$46,000 in 1976 when the Regentof Nadi was completedand F$37,OOOin 1973. - 14 - displacedNew Zealandersas the second largesttourist group (40,000in 1984) have the right spendingpatterns but tend to stay about 6 days. Atrivalsfrom Japan have declinedto 15,000in 198A with the averagelength of stay being five days, which is disappointingas the Japanesepredeliction for first-rate hotels and shoppingalso place them in the desirablecategory. Finallythere are the Europeanswho selectfive-star hotels for a stay averaginga week and whose expendituresmatch thoseof the Americanand Japanesevisitors. Some 20,000European tourists visited Fiji in 1984, twice the number that came in 1980.

2.21 The strategypromising the highestnet earningswould be one that increasedthe flow of upscalelong-haul tourists. Unfortunatelyits application on a large scale may be ruled out for reasonsof cost. American visitorsare generallymiddle-aged and drawn from executiveand professional groups. For them sightseeingand culturalactivities are importantand they demandhotel servicesof quality. Touristsfrom Japan tend to be young coupleswho attacha premiumon sightseeingand shoppingin surroundingswhere some Japanesecan be understuod. The Europeanscome to enjoy the beach and the weatherbut want excellenthotel facilities.Drawing more such tourists to Fiji would involveinvestment in hotel facilitiescomparable to the best in East Asia and, in order to servicethe Japanesevisitors, some involvementby the Japanesehotel companies,as has occurredin Hawaii. In addition,much capitalwould need to be sunk into the infrastructureof sightseeing,both human,organizational, and physical. Sites would have to be developed,tours devised,cultural performances staged and a small army of guidesand func- tionariescreated to serve the visitors.

2.22 Even if all this could be achievedin the space of a few years and foreignhotel companiesinduced to place more equityinto their Fijian operations in spite of the below-averagereturns they currentlyreceive, certain problemswill remain. First,the peak winter touristseason in the Northern hemispherecoincides with the monthswhen Fiji is most susceptibleto cyclones. Second,tourist traffic from the US is projectedto declineover the next few years becauseof a weaker currency. Third,there are a large number of Caribbeanand Africancountries vying for the dollarof the sun- lovingtourist clientele. Many are geographicallybetter-placed to draw vacationersfrom Americaand Europeand will competefiercely for the trade. Finally,Japanese and Australiantravellers have increasinglysougbt US destinationsover Pacificbeach resorts. The weakeningof the dollar can only strengthenthe pull of the Americancontinent.

2.23 Fiji was able to scale the ladderof per capitaincome during the sixtiesand the seventiesbecause it could draw upon the economicenergies of two successfulexport sectora. Efficientsmallholders contributed to the growth of sugar,while a combinationof local entrepreneurship,foreign capital,infrastructure development and the effectiveuse of labor in service activitiesallowed tourism to flourish. Very probably,the foundationsof the economywill continueto be securedby these two sectorsfor severalmore years. However,experience during the eightiesand an educatedguess regar&- ing the futureof these industriessuggests that their contributionto growth and certainlyto employmentcan only diminish. Heavy investmentin these sectorswill produceaverage rewards in terms of exportearnings crucial for achievingfurther advances in prosperity. For that the countryneeds a new leadingsector, and the only contenderis manufacturingindustry. - 15 -

III. NEWCOtUCERNS FOR TRADESTRATEGY

3.01 Industrialpolicy, whose ultimatepurpose is to promoteexports, must now contendwith a politicallycharged trading environment. The time is past when the US could gaze indulgentlyupon East Asian economiespreparing to loose a volleyof light manufacturestowards its markets. Trade is now a complexaffair that calls for great swiftnessin espyingand seizingan oppor- tunity. It is an enterprisae,moreover, for which detailedknowledge of marketsand the capabilitiesof competitorsis becomingindespensable, giving the experiencedplayers an Increasingedge over newcomers. Brazil,Mexico and Columbiahave robustexport industries but it is the Pacificregion that has emergedas the most dynamictrading arena (see Tables3.1 and 3.2). A small open economyseeking to industrializecan do no betterthan to start by examiningthe last 20 years of East Asian economichistory for it dispelsmany myths regardingthe comparativeadvantage and culturaluniqueness of success- ful tradingnations even as it underlinesthe virtuesof doggedlypursuing export objectivesand crlg ing a domesticindustrial environment that is both flexibleand innovative,

14/ In East Asia alone,Linder notes, "Thereare countriesrich in cultural resourcesand countriesaltogether lacking such resources. There are both sparselyand denselypopulated countries. All the major religions are represented.Some countriesare raciallyhomogenous; others are multiracial.s.someare militarilyexposed te.g., Yorea] and devote much...resourcesto defense.... Virtuallyall constitutionalsystems are representedincluding the colonialgovernment of Hong Kong.... As heads of state there are emperors,queens, elected and hereditarykings, presidentsand generails."S.B. Linder,The PacificCentury, Stanford UniversityPress, 1986. - 16 -

Table 3.1: PROPORTIONOF INDUSTRIALCOUNTRY IMPORTS FROM EAST ASIAN COUNTRIES

All goods Manufactures Exporter 1980 1984 1980 1984

Thailand 0.3 0.3 0.2 0.2

Malaysia 0.7 0.8 0.3 0.5 Indonesia 1.5 1.7 0.0 0.1 Philippines 0.4 0.5 0.3 0.4

China 0.6 1.1 0.5 0.8 NICS /a 3.2 5.8 5.2 6.3

Total East Asia 6.7 10.2 6.5 8.3

All DevelopingCountries 22.2 25.3 11.7 17.1

/a Korea, Singapore,Hong Kong, Taiwan.

Source: UN Trade tapes. ManufacturesSITC 5-8. - 17 -

Table 3.2: RELATIVECHANGES IN ASIAN-PACIFICTRADE FLOW RATIOS 1960-1983 (X)

Ratio 1960 1970 1980 1982 1983

Asian-Pacificexports over world exports 9.0 12.9 15.3 17.6 18.9 Asian-Pacificmanufactured exportsover world manufactured exports 8.7 14.2 19.9 20.4 Asian-Pacificover trade of LDCs 43.2 68.8 34.4 40.8 42.6 U.S. trade with Asia-Pacific over U.S. trade with Europe (OECD) 48.1 70.8 97.8 109.5 122.3 EEC trade with Asia-Pacific over EEC trade with United States 61.9 50.2 70.3 68.2 71.7 Trade of Asian-PacificLDCs over trade of the other LDCs (incl.oil countries) 21.7 30.0 28.8 34.3 39.9 Manufacturedexports of Asian- PacificLDCs over total manufacturedexports of all LDCs 30.9 41.6 64.3 67.1 Manufacturedexports of Asian- PacificNICs over total manufacturedexports of all LDCs 24.2 36.2 56.2 58.8

Note: The ten presentcountries ox the EEC are includedin the EEC statistics for 1960 althoughthey were not all membersat that time. "Manufac- tured exports"include SITC 5,6,7,and 8. U.N. 1984 does not give data on the manufacturedexports of Korea,Malaysia, and the Philippinesfor 1982. Koreanmanufactured exports are assumedto have changedat the same rate in 1981-82as those of Taiwan;the exportsof Malaysiaand the Philippinesare assumedto have changedin parallelwith the averagefor the other Asian-Pacificcountries.

Source: S. B. Linder,The PacificCentury, Stanford University Press, 1986.

EfficiencyThresholds

3.02 The ingenuityand enterpriseof the East Asian NICs have alteredthe rules of the game for nationslike Fiji that are on the thresholdof export- orientedindustrialization. For a broad range of light manufactures,metal products,engineering goods, machinery, electronic parts and transport equipment,the East Asian countriesdetermine the standardsin terms of price, - 18 - quality,technology, financing, supply capabilities and deliveryschedules. If manufacturersin Korea or Hong Kong are preparedto fill an order for garmentsmeeting exacting specifications at a certainprice, then other producersmust be in a positionto match or even improveupon the capabilities of these suppliersif they want a share of the business. Further,if East Asian standardsof factorproductivity allow them to earn a paiLicularreturn on assets,then zompetitorsmust be willingto settlefor an equivalentrate of return,or *f they wish to take aim at a higherrange of profits,these will have to be soughtthrough measures that raise the levelsof productivity and technology.To put it differently,the Pacificregion has becomean integratedand a highlycompetitive market with buyersand sellersenjoying excellentaccess to information.The pace is being set acrossa broad range of parametersby a numberof intenselydynamic exporters, hence other economies who desirea share of the businessmust be at least as efficientand offer terms that are comparablebefore they stand a chanceof establishinga footholdin the exportmarket. This is the basic hurdle that prospective Fijianexporters must overcome.

Sunk Costs

3.03 A countrythat is seriousabout competingin world marketsmust cross an efficiencythreshold in as short a time as possiblebut it is then faced with anotherentry barrier. Sunk costs give rise to cost asymmetries betweenincumbent firms overseasand new entrants. A producerwho has written off his plant and equipmentand advancedalong his learningcurve is far betterplaced than a businesswith new facilitieswhich has to factordepre- ciationinto its costs for severalyears. In certaincapital intensive industrieswith relativelystable technologies such as iron and steel products,cement, fertilizers, non-ferrous metals, and petrochemicalfeedstocks, which are attractiveto small open economieswishing to limit their import bills,create leading sectors that are rich in linkagesand move rapidlywith a standardizedproduct into the exportmarket, the problemof sunk cost can be an unsettlingreminder of how expensivedelay can be. To cite an example,a directlyreduced iron plant in Malaysiaefficiently constructed and of an optimumsize cannotcompete against bar and rolledproducts from established East Asian suppliersbecause the latterhave writtenoff most of their capital assets,feed their mini-millscheap locallyavailable scrap iron and have accumulateda valuablefund of experience.Ethylene, fertilizer and cement plantsspringing up in developingcountries, even if they are of the appropriatescale, must also come to terms with these uncomfortablefacts.

3.04 Lackinga long industrialtradition or much expertiseat exporting manufactures,with a labor force that is still short on skillsas well as factorydiscipline and a businessclass accustomedto operatingin a sheltered domesticmarket, Fiji cannotafford any delay in the developmentof an export- orientedmanufacturing sector. At the same time, a feasiblestrategy must be so drawn as to suit Fiji's specialeconomic circumstances and accommodatethe growthobjectives. It will necessarilydepart in some respectsfrom the strategiesthat East Asian NICs deployedwhen they began industrializing. Being a late starterhas its advantagesbut there are some attendantdrawbacks as well. - 19 -

Value Added

3.05 With a per capitaincome of US41,800and wage scalesto match,Fiji is virtuallydebarred from labor-intensivemanufacturing activities where the value added is modest and marketaccess dependsupon price competitiveness. Many of the mass-producedlow-quality textile, wood, plasticand metal items fall into this category. They are the first choiceof economiesentering the first stage of industrialization,since the technologyis stableand easy to masterwhile capitalcosts are manageable. For this very reasonthey are alreadythe preserveof countrieswhere wages are half of what a Fijianmanu- facturerwould expect to pay. Productswhich would be advantageousfor Fiji to exportfall in a differentbracket. They are the ones that do not require much technologicalsophistication but quality,design and finishare critical, productionis in lot sizes that can be handledby small producersand the speed of a manufacturer'sresponse to an order or the discoveryof an apparent niche in an overseasmarket is decisive. Armiesof disciplined,low-paid workersdo not count but superiorentrepreneurship, technical and design skills,a flexiblework force and governmentpolicies, that to a degree, socializethe inherentriskiness of these operationsare vital. In other words,the entry of Fijianbusinessmen into a fast-movingexport environment might have to be eased by way of schemeswhich providea measureof insurance againstrisks that lie in store.

Technological Capacity

3.06 Late entry means that a wide technologicalgap has to be bridged and, even with small-scaleoperations, some expensiveautomated equipment acquiredand broughtinto use. Managers,technicians and manualworkers must be embuedwith the desire to master the new technologyand where possible improveupon it. Manufacturersin Hong Kong, Korea and Japan are known to examineevery new productand processdevelopment in their field with minute care and throughreverse engineering, produce a comparableitem in the shortestpossible time that matchesor exceedsthe originalin performanceand carriesa lower price. The availabilityof engineeringcraft and design skillsof coursecontributes to the fleet-footednessof the NICs but entrepre- neurialwillingness to considera wide range of opportunitiesand obtain financingfor a host of venturesis by no means unimportant.

CapitalSupply

3.07 Exhaustiveresearch into the wellspringsof East Asian growthhave indicatedtime and again that a varietyof factorscontributed. Few would disagree,however, that an elasticsupply of capitalgreatly abetted the designsof businessmen.Each of the East Asian NICs have conspicuouslyhigh levelsof aggregateinvestment and channela significantpercentage of this directlyand indirectlyinto manufacturing.High rates of capitalaccumulation have enabledthese countriesto equip their factorieswith the latest machinery,to create capacityin anticipationof demandand expandtheir facilitiesto stay abreastof market growth,diversify across a wide range of subsectorsand, last but not least,find adequateseed money for assorted enterprisesthat occasionallybecome the nucleiiof futureleading sectors. The abilityto generatea substantialvolume of capitaldomestically can - 20 - enhancethe momentumbehind any strategyand especiallyone aimed at enlarging manufacturingcapacity.

NTBs and GeographicalShifts in Manufacturing

3.08 When the East Asian NICs starteddown the road to modernization, industrialeconomies were dismantlingtariff barriers and traderestric- tions. Now the climatehas changed. Althoughtariff reforms have not been reversed,quotas, voluntary restraints and nontariffbarriers have mushroomed aE the US and the EEC strugglewith the specterof deindustrialization.The consequencesof this for exportersaround the Pacificrim are threefold. First,there is a geographicalshift in production,especially in the textiles and garmentssector as countriesthat have filled theirquotas move production facilitiesto placesnot subjectto quota limits. This has benefitted Malaysia,Sri Lankaand Mauritius,for example. Second,most producershave begunmoving up the qualityand technologygradients so as to increasetheir returnsfrom the sale of a fixedvolume of goods. They have, in addition, modifiedthe compositi!-nof their productsto sidestepthe restrictions, Silk, linenand ramiehave gained in popularityas cottonand polyester garmentsbump againstquota ceilings. Similarly,the constructionof footwear is alteredto navigatebetween the differinglimits on plastic,leather and rubberitems. Thus the locusof opportunitiesis constantlyshifting with politicalforces, market saturationand technologyleadir,g the way.

DirectForeign Investment

3.09 A third concernrelates to direct foreigninvestment (DFI). This "asoccurred in a successionof waves, the most recentcresting in the late seventiesand early eighties. Electronics,broadly defined, was the principal focus and the multinationalcompanies (MNCs) involved, concentrated on the East Asian economiesthat seemedpolitically stable, had a disciplined,low- wage labor force,offered attractive incentives, were equippedwith the requisiteinfrastructure and had good transportlinks with the US and Europe. The electronicsindustry is in the throesof a revolutionarising from a great increasein the sophisticationof its product,automation of manufactureand a resistanceto the exportof jobs overseas. All of these have influencedthe flow of DFI and the interestof the MNCs in searchingfor new and cheapersites for their assemblyoperations. Unless a new industry sparksa fresh surge in DFI, the flow of manufacturingcapital towards the developingworld is liableto diminish. At the same time, after playinghost to the MNCs for over two decades,many countrieshave realizedthat the gains from DFI, in the form of linkages,technology transfer and the acquisitionof technicalskills is quite limited,and the most sophisticatedamong the developingeconomies have alreadybegun to look askanceat DFI that results in marginaldomestic value added and contributesvery littletowards industrial deepening.

Fiji'sChoices and ReportOutline

3.10 Fiji's industrialstrategy for the coming yearsmust be shapedby many factors: externalpressures are raisingthe costsof a delayedentry but at the same time they forcea new exporterto attain an unusuallevel of - 21 -

competencevirtually from the start; the flimsybase to industrialization providedby agriculturalactivities, tourism, and a small inward-looking manufacturingsector; the scale of domesticsavings as well as the supplyof risk capitalfor emergingbusinesses; the availabilityof labor skills to supporta niche-fillingexport drive; the nation'sreadiness to absorb manu- facturingtechnology; and last, the likeli:.Xoodof an ebb in DFI which would transfereven more of the burdenof mark.tpenetration and technology assimilationon to local firms.

3.11 Industrycan begin to displaceFiji's traditional sectors as a sourceof growthand exportsif the countryattains a stage of manufacturing capabilitycommensurate with its incomelevel, by-passing the earliestmanu- facturingstage that is centeredupon unsophisticatedmass-market items with a low value added. What a countrygains from the first stage is a labor force enuredto factorydiscipline; managerial training, market contactsand a reputationas a producerin a market segmentwhere ruthlesscompetition quickly eliminatesthose wanting in commitment. The only substitutefor such experience is a strategywhich is highly selective,that nurturesand harnesses local businessskills and which promisesan economicenvironment conducive to high entrepreneurialspirits.

3.12 Fiji must grope toward-a viable strategy. No infalliablerecipe exists. But there is a bodv ai empiricalevidence which allowsus to distill some guidelinesand advancea clutchof suggestions.The report interrogates six bodiesof developmentexperience so as to exposean underlyingmatrix of rules and then proceedsto indicatehow these might be employedin the Fijian context. Industrialincentives which determinethe patternof pricesand profitsacross manufacturing subsectors come first. This is followedby a chapterwhere the attentionshifts to tax policyand toe manner in which it can influenceoverall investment as well as savings(Chapter V). Since it is the quest for new exportsthat is behind the urge to industrializepChapter VI examinesthe mix of exportsfrom developingEast Asian economiesand their concentrationacross overseas markets. Exportoriented growth, the effective marshallingof informationon diverseaspects of foreignmarkets and institutionalmechanisms which will promoteits domesticutilization is investigated in ChapterVII. The chapteralso asks how the foremostNICs have achieved such high standardsof wage competitivenessand what lessonsthere might be for Fiji when it comes to exploitingthe resourcemost crucialfor industrialization.Fiji's geographical remoteness and the fixed costs of gathering informationare major considerations:increasing and wideninginformation channelswill be virtuallya preconditionfor market directedgrowth; and in the early stages,the informationeconomy will need assistancefrom the government. This is discussedin ChapterVIII.

3.13 Randomlyoccurring niches in industrialmarkets are difficultto target,but an exportdrive that takes aim at just these lucrativeopportu- nitiesmust be able to call upon a few tried and testedtactics. Again, theoryis of no help but the storiesbehind the exportsof fashiongarments from Italy as well as severalEast Asian countries,light machineryfrom Hong Kong, fruit from the Mediterraneanregion, alcoholic beverages from Mexico and the Caribbeanislands and watchesfrom Switzerlandyield a varietyof clues which are analyzedin the final Chapter. If niche fillingis to be the guidingmaxim of Fiji'sexport strategysuch case historiescould serve as the didacticspringboard for local businessmenseeking overseas sales. - 22 -

IV. INDUSTRIALINCENTIVES

East Asian Approaches

4.01 The investmentdecision is a complexone and economicsis frequently at a loss to explainwhy neighboringcountries with broadlysimilar resource endowmentsand incentiveregimes can registerwidely differing rates of private capitalaccumulation. Clearly, the entrepreneurialantennae is sensitive to many signalsand situationscan arise when economicmessages are blurredor obliteratedas a resultof staticgenerated by socio-politicalcurrents. That said there is considerabletruth to the claim that over longerstretches the speed and directionof industrializationdisplay the imprintof incentive policiesinfluencing profits, risks, competition and the relativeattraction of foreignmarkets. In each of the East Asian NIC's,policies governing taxes,protecti4n and subsidieshave definitelyorchestrated industrial development,although close examinationshows that these nationsarrived at their currenteconomic stations by varyingroutes even thoughthey were all guided by much the same philosophyand hemmed in by constraintscommon to small open economies. An ample literaturedwells on nuancesof strategy;more important, however,is the compositemodel that can be dredgedfrom this collective experience.

4.02 Importsubstitution supplied the initialmotivation to industrialize but withina few years the smallnessof the domesticmarket robbed it of momentum. At that stage it was obviousthat futuregrowth was predicatedupon findinga breed of industrythat could export. Some countriessought to createan exportbase by invitingforeign capital, others by puttingtheir faith in home grown entrepreneursbut the evolutionof incentivesystems was markedlysimilar. Importsubstitution had necessitatedhigh nominaltariffs on final productsand exceedinglygenerous rates of effectiveprotection. Profitmargins also benefittedfrom tax holidays,depreciation allowances and not infrequently,subsidized credit. The economypaid a steep price for entrepreneurialwell-being: first becausethe structureof incentiveswas profoundlybiased against exports; second the magnitudeof effectiveprotection encouragedoperations where the value added was low or negative;and third the domesticorientation of firms meant that plantswere of uneconomic size and in no positionto meet the test of internationalcompetitiveness at any point in the future.

4.03 Four major reformsof the incentivesystem signalled the move towardsexport-led growth. The bias againstexports was perceivedas a severe handicapand measureswere introducedthat assuredexport industriesan equal degreeof effectiveprotection. This was achievedby ^uttingimport duties on raw materialand machineryimports destined for the exportsector and trimming protectionto importsubstituting industries. A graduallowering of effective protectionfor industryas a whole was the secondmajor reform. By the early seventies,countries such as Singaporeand Hong Kong had broughteffective - 23 - protectiondown to very modest levelsbanished all possibilitieslVnegative value added and provideda powerfulspur to industrialdeepening.-

4.04 When import substitutionwas the regnantphilosophy, regulation and licensingin all its multifariousforms thrived. After thinkingchanged, it soon becameapparent that a bureaucraticobstacle course would defeat the most zealousexporter. Some controlsremained but each countryendeavored to prune regulationsand telescopethe processleading from applicationby a prospective industrialistto the sanctioning of a license. The EconomicDevelopment Board in Singaporewhich took over the entire processingof industrialappli- cationswas perhapsthe most radicalinstance of bureaucraticsimplification$ but all the NICs recognizedthe disadvantagesof unnecessarymeddling in an activitythat was sufficientlyfraught and risky to start with and began lighteningtheir supervisorybureaucracies.

4.05 Unlike productionfor the domesticmarket, exporting is more dependenton the timelyavailability of adequatefinancing. Where capital marketswere poorlydeveloped and hostiletowards manufacturers with insuffi- cient collateral,this could pose insurmountableproblems especially for smallerexporters. The NICs soon came to recognizethis problemand the fourthinnovation they introducedwas an effectiveand relativelyaccessible systemfor arrangingpre and postshipmentexport financing and servicinga part of the manufacturingsector's capital needs. Under proddingfrom the government,the bankingindustry gradually expanded its responsibilitiesand becamein time, a hand-maidenfor industry. (See Tables4.1 and 4.2 for detailson industrialincentives in Korea and Singapore.)

15/ See B. Balassa,Ed. DevelopmentStrategies .n Semi-IndustrialEconomies, Johns Hopkins,1982; A.J. Youngson,Hong Kong: EconomicGrowth and Poiicy,Oxford University Press, 1982; Mee Kan Nyaw and Chong-leongChan, "Structureand DevelopmentStrategies of the ManufacturingIndustries in Singaporeand Hong Kong,"Asian Survey,May 1982, Vol. 22, No. 5. - 24 -

Table 4.1: KOREA-- SUMMARYOF INDUSTRIALINCENTIVES

GovernmentCommitment to Export-OrientedGrowth In 1960,Korea's exports were abnormallylow relativeto country'ssize and stage of developmentthus therewas an exportpotential waitingto be tapped. This, togetherwith the low level of export,helps to explainthe very large percentageincreases which occurred. Successin export is also due to government's strategyand the flexibilitygovernment showed in alteringincentives to meet changing circumstances.The initialrapid succesJin exportingalso fueled latergrowth and helped attractforeign investment to Korea. A system of exporttargets for individualfirms in particularmarkets was introducedin 1962. If exportsfell below target,efforts were made to rectifythe situation. Target levelswere continuallyincreased. The Government subsidizedKorean Trade PromotionCorporation (KOTRA)was funded in 1964 to promoteexports and undertakemarket research. It also sent trade missionsabroad.

ExchangeRate Most importantpolicy changefor openingthe economywas adoptionof a uniformexchange rate between1961 and 1966,which correctedprevious overvaluation.

Taxation Exportersreceived a 50% reductionin direct taxes on incomeearned in exportas well as allowancefor accelerateddepreciation.

Credit Preferentialrates on workingcapital loans to exporters. Rates reducedfrom 14% in 1960 to 6% in 1967; exportersalso had preferentialaccess to foreignexchange loans. By 1964 interest rate subsidiesconstituted more than 1/5 of total exportsubsidies. - 25 -

Tariffs,Quantitative Restrictions Exportershad unrestrictedaccess to imported inputsthey were exempt from importquotas, tariffsand indirecttaxes on intermediateand capitalgoods (whetherimported or purchased domestically)used to produceboth directand indirectexports. Trade controlswere gradually relaxedafter 1964; furtherliberalization occurredin 1967. The positivelist of items which could be importedwas replacedby a negativelist. Rapid growthof importsthat were exemptfrom customduties left the average tariffon total importsunchanged.

Export-ImportLink Schemes Importentitlements were linkedto export performance.From 1957, only firms whose export sales met certaintargets could become registeredimporters. Export sales required beforea companycould registeras an importer were continuallyincreasd, from $10,000in 1958 to $300,00in 1970.

WastageAllowances Exportersreceived generous wastage allowances on importedinputs for exportproduction, in practicethese allowancesexceeded any reasonableestimate of genuinewastage and since many of the importedintermediate goods were otherwiseprohibited, the legal resaleof them enabledmany exportersto earn additional profits.

ForeignCapital Interestrate reformsof 1965 helped in attractingcommercial lending to Korea and borrowingattractive to Koreans;interest rates were about 26X for borrowingfrom domestic sources,but only 12Z for dollar-denominated loans,and even lower for those with government guarantees. DirectForeign Investment increased substantiallyonly after 1972, share of DPI in total foreigncapital inflows rose to nearly 202; inflowsof foreigncapital have been largelya responseto Koreanexport performance.

Source: B. Balassa,op. cite, and A. 0. Krueger,The DevelopmentRole of the ForeignSector and Aid, HarvardUniversity Press, 1979. - 26 -

Table 4.2: SINGAPORE: SUMMARYOF INDUSTRIALINCENTIVES

Tiriffs Tariffswere the preferredincentive for industrialization;rates were generallylow, 58Z of commoditiessubject to tariffprotection led rate of 251 or lower;majority of importswere duty free. Exporterswere exemptedfrom import duties on all inputs. Other producersreceived liberalduty exemptionson importsof raw materials,intermediate goods, machines and other inputsnot produceddomestically.

Export Incentives/Tax Incentives Tax allowancesfor nmrketingexpenditures abroad,tax rate on profitsreceived from exportsby approvedmanufacturing companies was reducedfrom 401 to 4X, for a periodof up to 15 years. PioneerIndustries Ordinance gave a 5 year periodof relief from corporatetax for new companiesproducing for home or exportmarket. In 1975 it was extendedto 10 year period,and minimumrequirement of S$l millionof capital was removedso as to encouragedevelopment of smallerfirms in high technologyindustries.

Import Licensing/ Quantitstive Restrictions Quantitativerestictions reduced after move to export orientedindustrialization. The number of commoditiessubject to QRs fell from 88 in 1965 to 3 in 1973. Quotas imposedin cases where domesticindustries faced "unfair" competition.Quotas were officiallyregarded as temporarysubsidies to domesticproducers while they developedexport outletsand were replaced ultimatelyby importduties.

ForeignInvestment Investmentpromotion activities were geared towardsattracting foreign industries; a network of officesand representativesabroad was established.During the 1960s a large numberof MNCs set up manufacturingand distributing facilitiesin Singapore. By 1973, nearly 75X of manufacturinginvestment was foreignand foreign firm accountedfor 751 of manufacturingexports, excludingre-exporting in 1971. - 27 -

ExportFinancing In 1974, Governmentallocated S$100 million for financingexport bills at a quarterpercent below prime rates, and for financing manufacturerswho wished to expandtheir capital equipmentfor export production. Export Credit InsuranceCorporation was establishedin 1975, with governmentcontributing 50% of capitaland remaindercoming from commercialbanks and insurancecompanies.

Credit EconomicDevelopment Board (EDB) establishedn 1961 to promoteincentives to privatecapital for establishingindustries in Singapore. EDB providedfactory sites on industrialestates, gave technicalassistance (e.g. feasibility, market studies,technical and managerial training). EDB was providedwith S$100 capital grant to financeloans and equityparticipation in industry;firms could borrowup to 50% of the fixed capitalrequirements for up to 10 years, providedsum borroweddid not exceedtheir paid- up capital. Availabilityof long-termfinancing and EDB's less stringentstandards of creditworthinesswas more importantthan actual market rate.

Source: B. Balassa,op. cit.

4.06 To summarize, the East Asian NICs commenced their attack upon export marketsby modifyingtheir incentivesystem in four respects: the bias againatexports was whittleddown; effectiveprotection was reducedand the infrastructurecomprised of tariffs,quotas and importduties rationalized; bureaucraticimpedimentia that can make the processirngof a projecta sustain- ed agony,were deliberatelyexcised; and the capabilitiesof the financial market extendedin just those areas where they tend to be most wantingviz., fundingfor small and medium scale manufacturingventures and the automatic supplyof exportcredit for firms directlyand indirectlyinvolved in trading. From the very outset,Hong Kong had the characteristicsof a free trade zone (FTZ) and once Singaporehad overhauledits incentivesystem and taken the export-ledstrategy to heart, the islandmore or less functionedas an FTZ. It is interestingto note that even the other East Asian NICs which for reasonsof strategyand size settledfor an intermediateposition found it expedientto establishFTZs, partlyas a competitivereflex but also because they recognizedthat FTZs broughtin foreigncapital and technology,could be made to spawn linkageswith the rest of the economy,helped build the country's reputationas a manufacturerand createdemployment. - 28 -

4.07 The NICs soughtintegration with the world tradingeconomy as the avenue to futuregrowth. It was a risky strategy. The certaintiesof import substitutingindustrialization were discardedout of the convictionthat the economycould become internationallycompetitive and had to break with a pat- tern of developmentwhich was self limitingand suffocatinglyinefficient. All four of the East Asian VICs gambledand won, but the odds againstan export-ledstrategy running aground in the sixtieswere much smallerthan they are today. Very few countrieswere seekingtheir fortunesin the world market. That is ro longerthe case, hence the reformpackage does not come with a guarantee. Many other economiesare experimentingwith variantsof the East Asian approachand the chancesof one emergingas a new Hong Kong in a few short years have decreased. The troublesSingapore has encounteredduring 1985/86underline the changedstate of affairs.

4.08 There remainsone featureof developmentin the successfulisland economieswhich a discussiondealing with incentivesfor manufacturingmust not overlook. Both Singaporeand Hong Kong are major ports and have servedas centersfor flourishingservice industries. Tourism,finance, insurance, brokering,ship repairand many othersare intrinsicto the istands' vitality. Synergybetween manufacturing and servicesis a major force and varietyof direct,entrepreneurial and macroeconomiclinkages have contributed to the efficiencyof industrialincentives. Highly sophisticatedfinancial marketsprovide immeasurable advantages for businessmenand in both countries, the annualinflux, of free spendingtourists greatly enlarges the potential clientelefor local producers. In the end, the model is not so much blurred as substantiallyqualified. If growth is to be led by manufacturing,exports must expand. Industriesthat will become activetraders require a particular mix of incentivesand a close interactionwith the world economyis unavoidable.However, the probabilityof a dramatictake-off by local industry or the rush of foreigncapital into manufacturingactivities is much smallerthan it was during the industrialinfancy of Singapore. ror an island economy,exporting manufactures remains an attractivestrategy but it can use all the help serviceindustries are able to provideand other than in excep- tionalcircumstances, the furiousgrowth rates of the sixtiesand seventies may be a historicalcuriosum.

Fiji'sManufacturing Sector

4.09 The Fijianmanufacturing sector is currentlyat the stage that Singaporehad reachedin the mid-sixtiesthrough a very similarimport substi- tutingstrategy. Ever since independencein 1970, industrialgrowth in Fiji has been in areas such as food and beverages,plastic, wood, paper,metal and chemicalproducts, largely serving low volume local consumerdemand. The manufactu.recf constructionmaterials was aided by Governmentspending on infrastructure,private housing investment and the expansionof the hotel industryin the seventies. More recently,a windowopened by trade agreements with Australiahas been a welcomebreak for garmentproducers. Veneerand plywood,although buffeted by the never ending turbulencein the world market, remainan importantsource of foreignexchange earnings and a new coconutoil facilitynear Savusavupromises to injectlife into the moribundvegetable oils subsector. - 29 -

4.10 Manufacturinggrowth has been stableif not spectacularover the past fifteenyears and encouragingly,the non-sugarmanufacturing sector has provedto be more resilientthan all othersduring the difficulteighties (see Table 4.3). However,it remainssmall and has yet to mature. Manufactuj9g accountsfor less than 101 of GNP and it employs181 of the labor force.' While manufacturingactivities in Fiji are dominatedby one industry,sugar processing,(about 27% on averagefor 1981-83),a wide range of goods are currentlymanufactured (see Table 4.4). Food processing,outside of the sugar industry,accounted for 29X of value-addedin the manufacturingsector. A numberof items are producedfor local consumptionas well as export--canned fish, a varietyof fruit juices(e.g., coconut, orange, and passionfruit), bakeryproducts, snack foods,tomato sauce, ghee, and processedginger. The secondlargest manufacturing category, also reflectingFiji's apparent comparativeadvantage in resource-basedindustries, is wood products, accountingfor 9.41 of value added. This subsectorcontains a numberof medium-sizedfurniture producers, who have recentlygained some successin exportingto Canada,US and Australia.

4.11 With the exceptionof clothingand footwearmanufacture, which still remainsa small but growing segmentin total.manufacturing value-added (about 31), the rest of the sectorproduces almost entirely for the domestic market. In fact, governmentofficials estimate that about half of manufacturingvalue-added owes its existenceto the government'spolicy of import substitution-- almost 70X of the nonsugarmanufacturing subsectors. Machineryand equipmentmanufacture, the third largestcomponent of the sector,is dominatedby shipbuildingand ship-repairing.Small cargo vessels, fishingboats, and coastalvessels are producedfor local,as well as regional companies. In additionto constructionactivities, servicing and repairwork is done in Fijianyards. Two muAtinationalfirms with connectionsto Australiaand New Zealandproduce a limitedrange of chemicals. There is packagingand distributingof bulk chemicalsfor domesticuse and for export to the surroundingPacific states. One firm also manufacturespaints for the localmarket, under the protectionof high tariffwalls. This brief summary of manufacturingactivities does inditatethat the nucleusof an industrial sectoralready exists in Fiji, thoughits orientationsare distinctlyinward- looking.

4.12 Most of the workersare unskilledand the operationsthey are involvedin such as assemblyand packagingresult in a relativelylow average level of value added for the sectoras a whole. Emergingout of the investments by multinationalcorporations of which Australianowned Burns Philp and Carpentersare the most important,industry is still very much in the grip of foreigncapital that has come to serve the domesticmarket and not, so far, in searchof an exportplatform. Sinoe the late seventies,Fijian entrepreneurs have begun playinga more prominentrole with a leadingcompany owned by Hari Punja active in flour milling,coconut oil refining,toiletries and ghee production.A small matket size, strictlicensing and a pervasivegovernment

16/ Total formalsector employment is 79,000,that in manufacturingis 14,500. - 30 -

Table 4.3: GDP BY INDUSTRIALORIGIN

1975 1980 1985

GDP at marketprices (millionF$) 502.4 983.7 1,350.0

GDP Shares

Agriculture(Z of GDP) 25.0 29.3 15.5

Manufacturing(X of GDP) (includessugar manufacturing) 10.7 10.9 8.7

Wholesaleand RetailTrade (2 of GDP) 19.0 16.5 15.4 (includingtourism)

GDP Growth Rates

1975-80 1980-85

CDP at market prices 4.1 1.3

Agriculture 5.4 1.0

Manufacturing 6.6 1.0

Non-SugarManufacturing - 2.8

Wholesaleand Retail 3.2 1.1

Trade

Source: World Tables,World Bank. concernto avoidwasteful duplication of facilitieshas resultedin a high degreeof industrialconcentration. Several subsectors are the preserveof a singlefirm and in most instancesbetween two and three producersdivide the businessamong themselves. Industryis run oligopolisticallyand competition is contained.

4.13 The touristindustry has establishedsome linkageswith producersof food and beveragesand to a lesserextent with manufacturersof wood products, textilesand toiletarticles but as in the Caribbeancountries the linkages have remainedsparse and tenuous. Qualitystandards have not improvedappre- ciablyand there are few signs of a concertedeffort by local entreprenuresto offer acceptablesubstitutes for items the hotel industrycontinues to - 31 - import. As upscalehotels ate frequentlyunable to satisfytheir requirements from local suppliersin developingcountries, this is not too surprising,but the weaknessof the linkagemechanism around the sugar and fish sectorsis more disturbing. Sugar has the potentialfor becomingthe centerof a thriv- ing manufacturingsubsector producing by-products (see Chart I) and as Peru has shown,anchovy fishing can have stronglinkage effects.LL

Table 4.4s VALUE ADDED BY MANUFACTURINGSUBSECTOR (F$ '000)

Value Added FSIC /a Industry 1981 1982 1983

Food manufacturing 57,201 66,624 40,519 (of which sugar) 31,306 38,084 11,022 Clothingand footwear 2,600 3,182 3,566 Wood products 8,836 8,588 10,231 Paper and printing 7,024 6,627 8,059 Chemicalproducts 7,333 7,244 8,926 Non-metallic& basic metal products 4,843 4,710 8,310 Machineryand equipment 8,618 8,707 9,954 Miscellaneousproducts 772 566 831

Total manufacturing 97 227 106,248 90,396

I Change includingsugar 7.9 9.3 -14.9 excludingsugar 7.2 3.4 16.4

/a Fiji StandardIndustrial Classification.

Source: 1983 Censusof IndustrialProduction, Bureau of Statistics,October 1985.

4.14 Fijianbagasse is mostly consumedas fuel possiblyruling out the productionof high densitythin fibreboard,particle board or packing material. But molassescan be convertedinto yeast based animal feed, rum, ethanoland a range of chemicals,the three tons of filtermuds obtainedfrom every 100 tons of cane pressedcan be used as fertilizer,while cane wax the preferredcoating for citrusfruit, could be marketedcommercially. The harvestingand transportof sugar cane can supportsmall scale repairfacili-

17/ M. Roemer,Fishing for Growths Export-ledDevelopment in Peru, 1950-67, HarvardUniversity Press, 1970. - 32 -

CHART 1 By-Productsof the Cane Sugar Industry E -# l>Il -- OmChwod briqueft -Methae and produwer a Pulp and paper Fibrous Paper board produots Fibre board

-Podkb board

Furfurol Alpha colulose

Plastics Flue GaseSand leaveS Mscellaneous -Poulty lItter and mulch A_Ankl teed Boagaseconcrete Fertililser - { arned Animal | _ Suor leed -- L d coane _Expodat

aad_Dlred _ Fertlier faft Is AnImal feed 'aXst _ -~~~~~~~~~Rum

- Ethylalcohol DldldDisarlling { v ~ hduryRecN"Industry - Anhydrous aloh - Alcohol derivatives

- Vinegar and acetic acId - Acetone- butonol

Proteintom1 |-~ ~~ferenoHn_- Clftrocid -aeoie ; induslrbs _ LctcNcacld -Glycerol Yeast

AcoolticacId Mon-sodum -o Dexdran L* Lyslne Xanthom Gum -noownc odd

Source 8y-PRX01 ofre CaneSugah Mdusry An introaucon to Their duOrkoi UtIiat J Mauri PattuO. Ehevrti ScIentifcPubisng COmpay. Amsterdom.1982. p 11 -33 - ties for machineryand metal workingto supply sparesfor equipmentused. In many countriessuch farm level activitieshave becomethe backboneof a thriv- ing engineeringindustry. With Fijian small-holderspreferring manual tech- niquesof harvestingthere is less scope for machinerebuilding and repair, nevertheless,the sugarsector has been strikinglyunsupportive of manufacturing development.Fisheries, a far smallersubsector than sugar has been behindthy 8eppearanceof a small ship-buildingindustry and of repair services.- Perhapsbecause the total landed catch is still just 5,800 tons (1985)the panopolyof supportingfirms that could be observedin Peru during the heydayof anchoryproduction have never surfaced. Suppliersof nets for instance,fabricators of processingequipment coolers, dryers; manufacturers of pumps,centrifugal separators and electricmotors; metal workers;and firms specializingin fish meal.

4.15 To cite just one more exampleof linkagesthat have failedto take root, there is the case of the local furnitureindustry. Certainkinds of local timberare well suitedfor furniture,but manufacturersalso require fabric,springs, material for padding,plastic based panelsand sheets, resins,nails, metal brackets,jute backing,not to mentiona varietyof tools and machinesto cut, shape,polish and assemblethe individualpieces. Demand has grown with manufacturersserving a sizabledomestic market; yet neither furnituremakers nor other producershave found it profitableto exploitthe linkagesand commencethe deepeningof the industrialsector.

Past IncentivePolicies

4.16 As is apparentfrom DP7 and DP8 and convincinglyenunciated in the latestplan document,the governmentseeks industrializationthat will help to diversifyexports. In 1980, it createdthe EconomicDevelopment Board -- a statutorybody composedof representativesfrom Government,the tradeunions, privateemployers and rural sector-- to provideinvestors with a complete packageof informationand extensionservices "under one roof". To quicken entrepreneurialinterest, the Governmenthas offeredan array of inducements including: incometax concessions;accelerated depreciation of assets;fuel economyrelief; export incentives; carry forwardof losses;import duty relief;protection through import licenses; and provisionof developedindus- trial land. Detailsare given in Table 4.5. The sectorhas responded,but aside from a flurryof exportactivity in the garmentsindustry, it has remainedinward looking, shallow rooted and from all indications,uncompeti- tive by world standards.Many factorshave contributed(some which are coveredin later Chapters)but as was the case in the early historyof the East Asian NICs high protectionand a bias towardsimport substitution over exportingare major culprits. The modal rate in a rather complicatedtariff scheduleis 50% and most rates are below 85%. There is the usual cascadingof tariffswith rates on consumergoods handilyexceeding those leviedon capital

18/ The shipbuildingindustry in Fiji, dominatedby CarpentersIndustrial (an AustralianMNC), has spawneda few industrial.inkages into the steel- made equipmentsector. See, E. Utrecht,Fiji: Client Stateof Australasia,University of Sydney,Sydney, 1984, pp. 242-244. - 34 - equipmentand raw materials. As a large part of local productioncompetes againstimported consumer good and uses impoV qd intermediates,the average level of effectiveprotection is above 1OOZX.- Rebatesof the tariffslevied on importeditems used for manufacturing(see Table 4.6) furtherheightens the protectiveeffect of the system.

19/ Where an incentivestructure for an entire industryis being considered, the nominalrate measuremay sometimesgive a misleadingimpression of the effectof protectionan incentivesfor resourceuse in different industries.Given equal aominalrates on the outputsof two industries, the incentiveprovided fo^ resourceuse in them may differdepending on, among other things,the i&portanceof intermediateinputs in their cost structuresand the level of protectionafforded domestic production of inputs. Thus, whilstincentives such as tariffson the output side benefit producers,tariffs on tradeableinputs penalize or tax the production process-- the effectiverate conceptmeasures the net extentof the incentives.Here it shouldbe noted that the tradeableinputs include both those that are actuallyimported and those sourcedlocally, such as packagingmaterial. The effectiverate of protectionmay be definedas the differencebetween value added in domesticand world price terms, expressedas a percentageof value added in world prices. The basic effectiverate formula,which is frequentlyused in evaluationof protection measurescan be expressedas:

v = to x.t.x 1 - x

where to = nominalincentive on output (takinginto accountif possible quantitativerestrictions on imports)

ti = nominalincentive on inputs

x n materialsto output ratio, expressedin world price terms.

From this, it can be seen that the effectiverate of protectionis influ- enced by levelsof tariffs(and the effectsof quotas)on competing importsand on inputsand the materialsto outputratio. In generalthe higher the nominalprotection against competing imports, the lower the taxingeffects on input and and the lowe- is value added -- then the higher is the effectiverate of protection. - 35 -

Table 4.5: FIJI -- SUMMARYOF INDUSTRIALINCENTIVES

RebateCertificates Allow concessionon dutieson machineryand raw materials. Used to apply indefinitelybut in 1984 it was decidedthat they shouldexpire at end of that year. At same time, customsduty was increasedto 5X for machineryand 1.5Z for raw materials;these goods remain free of fiscal duty. However,the certificateswere not abolishedand guidelinesfor evaluationof requestsfor extensionof rebatesare still being prepared. In 1982, about 259 firms used the certificatesfor raw materialsand total value of concession(in terms of duty forage) was about F$31 million;15 firms used certificatesfor machinery,with a conceasion value of F$11 million.

ImportLicensing Used in an ad hoc fashionand usuallyprovided for an indefiniteperiod. Currently45 food and othermanufactured products are subjectto import licensingwhich accountfor about 30% of value added in manufacturing.Moreover, 40% of the value-addedcovered by importlicensing appliesto food and agriculturalproducts.

Price Controls Two separateinstruments (i) the Pricesand IncomesBoard (PIB) whicthadministers the Counter-InflationAct. As of April 1986, 39 goods were coveredby a fixed percentagemark-up control;and 13 goods and services(mostly food and buildingmaterials) had fixed wholesaleand retail prices. (ii) The PricesJustification Tribunalwithin the Ministryof Economic Development;Planning and Tourism,which is responsiblefor ensuringthat ex-factoryprices for locallyproduced goods benefittingfrom importprohibition are "not excessive." Tariffs Comprisefiscal and customsduties, with the formerdominant. Almost 60% of tariffrates are below 20%. Tariffsrange from relativelyhigh rates on consumergoods (highestrates apply to motor vehicles)to lower rates on capitaland intermediategoods. In most cases, the rebate certificatescheme means that scheduledtariff rates do not apply. - 36 -

ExportIncentives (i) Drawbacks: Drawbackfacility available for directlyimported inputs used in production for exportmarkets, based upon input-output coefficients.Does not includeindirect exports.

(ii) TaxationIncentives: Rebateon tax chargeableon export profits,varies accordingto localvalue-added of exported product,thus value of incentivevaries in relationto profitabilityof exporting. Future incentiveto exportersprovided by tax deductionequal to 150% of approved expenditureon promotionand marketingof goods from Fiji

Pre- and Post-Shipment Pre-shipmentfinance scheme introduced by the Finance ReserveBank of Fiji (RBF) in 1985 for exportof non-traditionalitems (i.e. sugar,molasses, coconutoil and gold are excluded). Finance availablethrough commercial banks at 8% p.a. interest(compared to 13.5%p.a. commercialbank lendingrate); RBF refinancesbanks at 6% p.a. Exportsunder SPARTECAand LOME are granted facilityautomatically, other itemsrequire minimum of 40% local content. Postshipment financingavailable at similarinterest rate. RBF only providesforward exchange cover to commercialbanks tradingin A$ and NZ$ for up to 6 months and limitedto SPARTECAexp?orts; risk of tradingin other currenciesis borne by exporter.

4.17 Tthestructure of incentivesis a creatureof piecemealgovernment interventionover 15 years, reachingdown to the level of the individual firm. It is difficultto assess,remains subject to administrativediscretion which injectsconsiderable uncertainty into businessplanning and having been tailoredto exactmicroindustrial specifications has erodedthe incentiveto raise efficiency. The missiondid not embarkon a detailedsurvey of levels of effectiveprotection in Fiji, due to time and data constraints.The authoritiesrecently tried to obtain some informationon costsand profits from the manufacturingsector, but did not have much success. This occurred becauseit was widely believedthat the governmentwas preparingto adjust tariffson the basis of profitlevels. Further,published industrial statisticstend to be highlyaggregated so as to protectthe confidentiality of the very small number of firms inhabitingeach subsector. For these reasons,proposals for a reformhave been based on a qualitativeapproach whose empiricalcore is a listingof productssubject to rebatecertificate, - 37 - classifiedby the livel of tariffsprotecting outputs. The salientchar- acteristicsof the principalmanufacturing activities are shown in Table 4.6. Firsta few words of elucidation. Althoughthe unweightedaverage level of tariffson competingimports of the sample is about 452, once quantitativerestrictions are factoredin, it is likelyto be higher. Almost half the productsin this sample benefitfrom quotasand have a tariffof less than 35% on competingimports -- moreover,none of the productsin the category with the highesttariffs (652 and over) have quotason competingimports. This indicatesthat the protectionfor some productionactivities has been affordedby way of high (and increasing)tariffs, whilst for others it has been by way of quotas. Of course,the end resultmay well be the same. Anotherinfluence on output protectionis the structureof excise duties, which are a tax on local productiononly and thus reducethe benefitof tariffs. In Fiji,excise taxes are leviedas a specifictax (e.g.,per kg) on the final goods involved;however, the ratesare fairlylow and less than half the productslisted are subjectto this levy.

4.18 The tariffon inputssubject to the rebatecerti5icate is 7.52 and, in general,this rate could be taken as indicativeof the taxingeffect on inputsinto local production,although there are some qualifications.First, not all importedinputs are subjectto the concession(this may be becausea locallyproduced input is available). For example,in the case of ice cream production,about 702 of importedinputs are subjectto the scheduledtariff rate (not necessarily7.52). The M8 )ghtedaverage tariff on inputsused in ice cream productionis about 18%.- Second,there are locallysourced inputs,such as packaging--whichin turn are protectedby way of high tariffs. This indirectlyincreaFes the costs to user industries. Only about one-fifthof the activitiesin the samplesource more than 302 of their raw materialrequirements locally and these are listedas follows:

502 + s biscuits,cake mix, woodenhandicrafts/sticks; louvrewindows; fruits and vegetables;veneer and plywood;

402 < 502 : butter,cooking oil and margarine,ice cream and peanutbutter; and

30% < 402 : confectionary,buses, furniture,upholstery and mufflers.

4.19 The value-addedratios shown in Table 4.6 are in domesticprices which have been adjustedto the structureof protection. It is definedas the value of output less intermediates-- the latter includingraw materialsand what are referredto as nontradedsuch as electricityand business

20/ This informationis derivedfrom the limitedreturns made to the firm questionnaire--theinput tariffof 18% appearsto be much higher than usual. - 38 -

Table4J: Value-Added,Protection Coefficients of Fiji's IndustrialSector

Proportion Proportionof Tariff Tariffs on Iw- Value-Added Raw Materials Product a/ Protecting Quantitative Excise puts (Subject to on Domestic locaily Output(2) / Restrictions Duty c/ Concession- 2) prics d/ .

65 gercent and over (4.) Beer 200 - Yes (100) 7 1/2 0.16 0.26 Confectio=sry 0o - Yes 7 1/2 0.27 0.31 Biscuits 8o - 7 1/2 0.28 0.66 Snack foods 80 - Yes 7 1/2 0.18 0.13 Coke mix 80 - - 7 1/2 0.28 0.66 Soap 72.5 - Yes (20) 7 1/2 0.25 0.28 Detergent 72.5 - Yes (20) 7 1/2 0.25 0.28 Paints 67.5 - Yes (5) 7 1/2 0.24 0.16 BuEes 67.5 - - 7 1/2 0.43 0.38 Furniture 67,5 - - 7 1/2 0.36 0.34 Upholstery 67.5 - _ 7 1/2 0.36 0.34 Overalls+ 65 - - 7 1/2 0.33 0.02

50 > 65 peent

Cellobags and tapes 57.5 Yes - 7 1/2 0.30 0.02 Plastlccootainers 57.5 _ - 7 1/2 0.30 0.02 Cosmetics 57.5 _ - 7 1/2 0.25 0.28 Umbrellas 57.5 Yes (10) 7 1/2 0.36 0.16 PolyBags 57.5 Yes Yes (3) 7 1/2 0.30 0.02 Plasticproducts ' .5 _ Yes 7 1/2 0.30 0.02 Musiccassettes 57.5 _ _ 7 1/2 0.30 0.02 Incense sticks 57.5 Yes _ 7 1/2 0.54 0.09 Baby crem 57.5 _ 7 1/2 0.25 0.28 Woodenhandicrafts 57.5 _ _ 7 1/2 0.48 0.59 Toiletpreparations 57.5 _ _ 7 1/2 0.25 0.28 Polythene bags 57.5 Yes Yes (15) 7 1/2 0.30 0.02 Drinklngstraw 57.5 _ Yes 7 1/2 0.30 9.02 Plasticbottles 57.5 _ _ 7 1/2 0.30 0.02 Woodensticks 57.5 Yes - 7 1/2 0.48 0.59 Butter 55 Yes Yes (2) 7 1/2 0.21 0.45 Cooking oil & marg. 55 Yes Yes (3) 7 1/2 0.31 0.41 Paperproducts 52.5 - Yes 7 1/2 0.34 0.15 Stationary 52.5 - - 7 1/2 0.34 0.15 Cardboard ca tons 52.5 - - 7 1/2 0.34 0.15 Aluminu utensils 52.5 - - 7 1/2 0.29 0.05 Printed products 52.5 - - 7 1/2 0.41 0.00 Envelopes 52.5 - - 7 1/2 0.34 0.15 Toiletpaper 52.5 - Yes 7 1/2 0.34 0.15 Self-adhesive labels 52.5 - _ 7 1/2 0.34 0.15 Paper cups 52.5 - _ 7 1/2 0.34 0.15 Cards 52.5 - 7 1/2 0.34 0.15 - 39 -

Tible4.6t (Costimamd) Value-Added, Protection Coefficient of Fiji's Industrial sector

_rpeprtioo Ptoportim of Tariff Tariffs an It- Veles-Ad.dRu Waterials Product f Protecting Quantitative gucise puts (Subjectto on Dstic L1aly Output (I)b/ RIstrictions Daty Conmesi - S) Prices or .

35 > 50 per

Ceomnt 41.5 Yes Yes (4) 7 1/2 0.35 Dentsicram 47.5 _ Yes 7 1/2 0.25 0,28 Louvre Widbs 41.5 Yes Yes (4) 7 1/2 0,33 0.62 CAndles 47.5 - Yes 7 1/2 0.54 o.o9 Stsailess at.utensils 47.5 - - 7 1/2 0.29 0.05 Ico cresi 45 - Yes (10) 7 1/2 0.21 0.45 Niol aidstaples 42.5 - Yes 7 1/2 0.29 0.05 solarheatars 42.5 _ - 7 1/2 0.29 0.05 R84es 42.5 Yes Yes 7 1/2 0.36 0.16 Footwear 40 _ Yes 7 1/2 0.26 0.00 Fruits andvegetables 40 Yes _ 7 1/2 0.11 0.61 Peanut butter 40 - - 7 1/2 0.21 0.45 Fyam 37.5 - Yes (14) 7 1/2 0.30 0.02

20 > 35 erest

Solvents 32.5 - Yes 7 1/2 0.25 0.16 Scou¢ingpowder 32.5 - Yes 7 1/2 0.25 0.28 Batte¢ies 32.5 Yes _ 7 1/2 0.22 0.04 Toys 32.5 Yes _ 7 1/2 0.36 0.16 PVC pipes 32.5 Yes Yes 7 1/2 0.30 0.02 Fluorescont'ights 32.5 - _ 7 1/2 0.22 0.04 adlo telephones 32.5 - _ 7 1/2 0.22 0.04 POUshs 32.5 - _ 7 1/2 0.25 0.28 Thinners 32.5 - Yes 7 1/2 0.54 0.06 Leps + holders 32.5 - _ 7 1/2 0.22 0.04 Adhesives 32.5 - Yes 7 1/2 0.54 O.O8 Drusbie 3.5 - _ 7 1/2 0.36 0.16 onsters 32.5 - _ 7 1/2 0.36 0.16 Floor 30 Yes - 7 1/2 0.14 abowtZero Croftcorks 30 Yes Yes (9) 7 1/2 0.36 0.16 osq.:Itocoils 30 - _ 7 1/2 0.54 0.06 llbaelbarr,ms 22.5 - - 7 1/2 0.29 0.05 Trallers 22.5 - - 7 1/2 0.29 0.05 Chalk 22.5 - - 7 1/2 0.36 0.16

Lessthan 20 percent

Roofingiron 17.5 Yes Yes (2) 7 1/2 0.23 0.06 Polyshoets 17.5 Yes - 7 1/2 0.30 0.02 Veneerand plywood 17.5 Yes - 7 1/2 0.33 0.62 Nbtalcet goods 17.5 _ - 7 1/2 0.23 0.06 Steelwire alugs 17.5 - _ 7 1/2 0.23 0.06 Steelrods 17.5 Yes Yes 7 1/2 0.23 0.06 Animalfeed 15 - _ 7 1/2 0.18 0.13 Mufflers 15 - - 7 1/2 0.43 0.38 hlers 15 - - 7 1/2 0.33 0.62 Nbtal containers 12.5 - - 7 1/2 0.29 0.05 PrEopredfish 10 - - 7 1/2 0.11 0.61 Jeaisry 10 - - 7 1/2 0.21 0.04 Gu stoves 7.5 Yes - 7 1/2 0.51 almostZero

/ Rued an a listingof productscoapilsd by theNinistry of IkoxuicDevelo"enxt, Planning and Tourisw-excludes cigarettes and tobaeeo for wbich pecifie rates of duty apply.

TThe tariffs protecting outputs are those that applied prior to the revisions of April 1986. These revisions resultedin increases In tariffs in large part on those products already subject to hi8h tariffs.

WbUhre available the estimated incidence of the speific excise duty rates (in terssof cif lIport prices) is shownin brackets.

/ These proportions are averages for the FSIC group to which the product belongs.

(D-205a) - 40 - charges.21 On this definitionalmost half the activitieshave a value-added of less than 30X -- such relativelylow ratios alone would tend to result in high effectiverates of protection. However,the effectiverate of protection formulais based upon value added in world prices,which is usuallyestimated by deflaij7gthe observedinput-output data on the countryimposing the tariffs.- Takingonly those items subjectto a tariffof 501 and over (adjustingfor excisetaxes and assuminga "tax" on inputsof 15X) the followingproducts may have negativevalue added in world price terms: beer, confectionary,biscuits, snack foods,cake mix, paints,cosmetics, baby cream and toiletpreparations.

4.20 Apart from confectionary,biscuits and cake mix, these productshave a relativelylow percentageof value added (in domesticprices) and a limited relianceof locallysourced raw materials. By replacingimports of the final goods, they have most probablyworsened the balanceof paymentssince the free trade value of the final goods displacedby domesticproduction is less than the cost of the importedinputs.

4.21 Using Table 4.6 and the frameworkof effectiveprotection allows the followinggeneralizations:

(a) that in absoluteterms there is a considerableincentive to produce for the domesticmarket. Levels of protectionare particularlyhigh for thoseproducts benefitting from a tariffof 50X and over, with prima face evidenceof negativevalue added in world prices. The averageunweighted level of effectiveprotection is of the order of 1001, with a range of infinite(for those productswith negative value added) to below 20%. Thus there is considerablevariation in levelsof effectiveprotection to importcompeting activities. It shouldbe noted that non-traditionalagricultural products such as fruit,vegetables and butter are also highlyprotected.

(b) there is a significantanti-export bias -- on averagethe incentive to producefor the domesticmarket is at least five times greater than for the exportmarket; and

(c) the structureof incentivesprobably favors importcompeting activitiesin general (both manufacturingand non-traditional agriculture)at the expenseof traditionalagriculture. However,

21/ How value added is definedand in particularthe treatmentof non- tradeds,has a significantinfluence on effectiverates of protection. Brieflythe inclusionof nontradedswith inputsis in line with the methodologysuggested by Balassa. An alternativemethod is suggestedby Corden,where nontradedsare decomposedinto their tradeableand valueadded components-- the value added of nontradedsis then includedwith primaryvalue added.

22/ An alternativeapproach is to take the value-addedratio of the productionfrom the exportingcountry. - 41-

this issue is complicatedby the structureof taxation-- some activitiesin all sectorsare exempt from taxationwhich providesa considerableincentive.

IncentivesiReform

4.22 The earlierdiscussion of strategypursued by the East Asian countriessuggests that less protectionand a removalof the anti-exportbiq's shouldbe given prioritywith care being taken to avoid cuttinggovernment revenuesfrom duties on tradedgoods until alternativesources had been developed. Given Fiji's incentivesystem, implementing this strategywill entail: (a) eliminatingquantitative restrictions and loweringtariffs on competingimports starting with the most highly protecteditems and working down; and (b) broadeningand increasingexcise taxes on local productionand also enlargingthe role of sales taxes. To give existingindustries time to adjust,any reformmay have to be spreadover severalyears. Incentivesfor neu activitiesshould set with referenceto those affordedexisting activities which are close substitutesin productionand/or consumption.Moreover, such incentivesand the approachto be taken shouldbe made publiclyavailable. Some more specificdetails concerning a reform of incentivesare as follows:

(a) Quantitativerestrictions introduce distortion and complexityinto the incentivesystem and should be eliminatedand a commitmentmade to avoid futureuse. Exceptionswould be quotas imposedfor non- economicreasons or under internationalagreements (such as coffee under ICA). The removalof quotas could be phased in over four years,with one quarterof the quotas to be withdrawnin each year. As significantnumber of productsin this categoryhave a tariffof less than 351 on competingimports--consideration could be given to increasingthe tariffto the ceilingamount (as suggested below) on these productsto lessenadjustment pressures, a move that would increasetariff revenues especially from productscurrently on the prohibitedlist. Eliminatingquantitative restrictions wgQ ld also make the PricesJustification Tribunal (PJT) redundant.2'_

(b) It i preferablethat the tariffbe used as an instrumentfor protectionand the Governmentcould considerinitially introducing a tariffceiling of 601 and reducingall rates above this level with the exceptionof certainitems such as motor vehicles,beer and

23/ As mentionedpreviously there are two separateprice controlinstruments in Fiji. The PIB mainlyaffects the sale price of variousgoods and in particularmark-ups after costs (althoughit is responsiblefor the ex-factoryprices of flour and plywood). The powersof the PIB have been used vigorouslyat times to combat inflation. The PJT is the body which in theoryis responsiblefor controllingex-factory prices and therefore more directlyinfluences resource allocation. In practicethe PJT does not closelymonitor prices due to a lack of staff. Moreover,initial priceswere set at such generouslevels that many manufacturershave not needed to requestfurther increases. - 42 - cigarettesand tobacco,that bring in a significantvolum, of revenue. Howeverpthe level of excise dutiesshould be increasedon these latterproducts. To achievethe statedobjective of eliminatingthe anti-exportbias would then requirebringing the tariffceiling to below 301. The exact schedulingof any tariffs cuts is by naturearbitrary and would need to be preceededby a decisionon the future shape of the taxationsystem. For example, tax revenuecould increasinglybe derivedfrom a consumptiontqx, in which case the levyingof such a tax could be coordinatedwith a phasedreduction in the ceilingtariff to 301 over a 4-year period. On the other hand if the path selectedinvolved raising exciseduties then there may be lets scope for cuttingtariffs on - 43 -

ment have to sacrificedin the short run to achievelouger term gains in effi- ciencyand exportpotential. Dislocationcan be minimized,as suggested above,by adjustingover the course of 3-4 years, but it is in the economy's long run intereststo staunchthe haemorrhagefrom industrieswith negative value added as speedilyas possibleand to preparethe ground so that exporters might flourish.

4.24 Undoubtedly,policymakers will have to take accountof political concernsand the circumstancesof individualsubsectors in drawingup an acceptableproposal but from our estimatesit appearsthat revenueloss need not deter a reformof the incentivesystem. As Table 4.7 shows revenueswould increaseeven if the maximum tariffwas loweredto 60X so long as an import duty was imposedon all importsincluding th6we by the government. Exempting publicsector imports does reduce revenuesbut only by 4X which is a small price to pay for the gain in efficiencythat could be achieved.

Table 4.7: FIJI: FISCALEFFECTS OF TRADE LIBERALIZATION (as % of 1985 actualtollections)

AlternativeTariff Schedules 1985 No GovernmentExemptions GovernmentExemptions Actual Min. Duty 10% Min. Duty 10% Min. Duty 102 Min. Duty 10Z Collections Max. Duty 80% Max. Duty 602 Max. Duty 802 Max. Duty 60X

Duty Collections 100.0 108.8 106.1 98.0 96.0

/a Assumesimport duty rebate-certificatesare abandoned. Tb All items,which currentlytarry no duties,have duties imposedat the minimum rate.

Source: Staff Calculations. - 44 -

V. A TAX PROGRAMFOR GROWTH

Tax Criteria

5.01 The principlesof a sound tax systemare well known. Verticaland horizontal equity, efficiency, administrative convenienceand other criteria have been textbookstaples for decadesand serve as buildingblocks for tax policy. That they are rarelyapplied in a consistentfashion is traceableto the unbendingcomplexity of the real world and nit to a lack of conviction among experts. No countrycan boast a model tax systemand this is an aspect of East Asian developmentwhich has been least fruitfulin terms of lessons. Nonetheless,a close examinationof the tax structurein the more rapidly growingeconomies does uncovercertain common features that could be of interestto Fijianpolicy makers.

5.02 Fiscalstrategies differ depending on whether the emphasisis primarilyon growthor on achievinga more equitabledistribution of income. A concernfor growthwith long-runresource equilibrium supports a tax framework that maximisesinvestment and savingand which combinesadministrative simplicitywith a fairlyhigh revenueyield. Investmentis encouragedby tax- ing capitalincome at low rates. This can be achievedthrough modest levies on corporateprofits, by exemptingdividends from incometax and by incentives such as investmenttax creditand accelerateddepreciation allowances. A favorabletax treatmentof returnsfrom varioussavings deposits and retire- ment schemescan encourageaccumulation. Consumption as againstaccrual incometaxes which are savingsneutral can also serve the same objective. The public sector,in the early eighties,has been under pressurefrom high and unsustainablefiscal deficits. Except for 1982, a year of particularlypoor revenueperformance, the overalldeficit has averagedaround 4X of GDP, since 1979 (see Table 5.1). These deficitsare traceableto a declinein central government,rather than the effectsof a capitalexpenditure program. Two factorsmust be held responsible: first,a relativelystagnant revenue performancewhich, as of 1985 had not recoveredto the level of 1980; and second,a rapid expansionof currentexpenditures, which in 1985 were 5X higher (as 2 of GDP) than in 1979. - 45 -

Table 5.1s CENTRALGOVERNMENT SAVINGS AND Im ESTMENTBALANCE (as X of GDP at factor cost)

1979 1980 1981 1982 1983 1984 1985 1986/a

Currentrevenues 27.1 31.5 26.7 25.0 27.5 29.7 28.0 26.4

Currentexpenditures 23.9 27.2 22.3 23.9 27.5 29.8 28.8 28.3

CurrentGov't Savings 3.2 4.3 4.4 1.1 0.0 -0.1 -0.9 -1.9

Foreigngrants 0.9 1.2 0.9 1.0 1.0 0.9 0.9 1.1

Capitalexpenditures lb 7.7 9.5 10.0 9.0 5.5 4.9 5.2 6.1

Overall Deficit -3.6 -4.0 -4.7 -6.9 -4.5 -4.1 -5.2 -6.9

/a Estimates. lb Includesnet lending. This category is developmentexpenditures; all outlaysare not necessarilyinvestment expenditures.

Sources: Ministry of Finance; Fiji Adjustmentand Selected DevelopmentIssues, Report No. 5515-FIJ,The World Bank, August 1985.

5.03 Distortionsdepend on taxes at the margin. They are greatestwhen an activityis highly responsiveto changes in returns. For instance,the economy can suffer seriouslosses if high taxes on upper income groups dis- courageef 9st or greatly increasethe time and ingenuityexpended on tax avoidance.- A broad based tax with a low rate is the least distortionary, for examplea consumptiontax could meet this test as would a flat-rated income tax. Revenue needs and administrativesimplicity are best met by broad based taxes which do not require the tax payer to record and report information that is inherentlydifficult to audit. A consumptiontax can come in direct and indirectforms. For close to thirty years, an expendituretax which displacesincome taxes has figuredprominently in many reform proposals. Its attractivenessis tied to simplicityand neutralitywith respectto savings. Conceivedas a cash-flowtax it would impingeupon wages,

24/ Welfare losses stem from two sources: the income effect which arises from having to pay the tax--a loss that is offset by the benefits from public expenditure--andits size is determinedby the average rate of tax paid; and the substitutioneffect which results from the disincentive effectsof taxationat the margin and this depends only on the marginal rate of tax. J. A. Kay and M. A. King, The BritishTax System,Oxford UniversityPress, 1983, p. 16. - 46 - interest,dividends, gifts, cash from the sale of assets and cash raisedby borrowing,i.e., over an individualslifespan, all his earningswould be taxed. The base is broad and the problemsarising from definingincome, and valuingcapital gains for tax purposesare eliminated. An individual assessinghis tax obligationsat the end of the year could claim deductions for any receiptsnot spent on consumptionor gifts: repaymentof interest,of principalon loans,purchases of shares,would all be deductible. In effect, the expendituretax removesthe bias againstsavings arising out of double taxationunder the incometax and minimizesdistortions by eliminating differentialtaxation of particularforms of savingsand capitalincome. Recent examinationof the administrativeimplications of the expendituretax suggestthat they are no more onerousthan for incometax which shouldhelp dislodgeone of the principaland longstandingarguments against a tax on what an ig_ividualtakes out of a societyrather than on what he contributesto it,

5.04 The expendituretax still strikesmany as being too revolutionary and thereforerisky. Hence the preferenceis for an indirectlevy on consumption, which somewhatinefficiently, serves the same purpose. Experience gatheredby some 39 countriessuggests that a Value-AddedTax (VAT) with few exemptions,limited rate differentiation,whichis collectedat the manufac- turersor wholesalestage, offers a workablecompromise. The VAT is superior to the sales tax becauseit can be designedwith a broaderbase. When there is littlevariation of rates betweenproducts it is simplerto operatebecause of cashflowaccounting and extensiveuse of linkedtax accounts. Since the final sellermust accountfor the purchasesof other firms,collection costs are reducedand evasionrendered more difficult. The multistagecollection that is a part of a many tieredVAT increasesadministration costs but providesa more accurateand reliablemechanism for excludingthe tax from businesscosts and exportprices. Finally,firms subjectto VAT are induced to keep betterrecords of purchasesand sales which result in managerial gains. It is importantnot to exaggeratethe virtuesof the VAT. In the EEC, the transitionto the VAT was beset by complications:evasion remains chronic,and administrationhas not been easy becausegovernments have found it impossibleto avoid using multiplerates, exemptions and zero rating to lessenregressivity and for excludingservices where the value added is difficultto measure. The VAT can be a successonly if it is seen as a major sourceof revenue,is appliedat a fairlyhigh rate, can be supportedby a computerizedtax administrationwith sophisticatedprocedures for cross- checkingand is introducedin a businessmilieu where literacy,numeracy,

25/ See, J. A. Kay and M. A. King, The BritishTax System,op. cit. Chs. 1, 4, 5; R. Albon, TaxationPolicy in the Eighties,Allen and Unwin, 1986, chs, 1, 5, 20; ReformingBritain's Taxes, The Economist,1986, pp. 2-7; and The Structureand Reformof DirectTaxation, Institute of Fiscal Studies,Allen and Unwin, 1978. - 47 - adequatereitydkeeping practices and the use of cash registersis widespread. Where this is not the case; where the economicsystem is still fairly shallowwith a few single stage manufacturingactivities and a profusionof small retailoutlets, inherently difficult to police,a retail sales tax is a betterinterim solution. Once peopleare used to the mechanics of a broad based sales ita and the economyhas acquireddepth, the VAT becomes an efficientnext step.

5.05 Thus, a tax policywhich is slantedtowards economic growth, revenue yield and administrativeconvenience should, in principle,opt for a system symetricallyencouraging both savingsand investmentwhich uses a flat rated income tax combinedand a broad based cops mption levy with uniformrates or better still,adopts an expendituretax.,

How Burdensomeare DirectTaxes in Fiji?

5.06 Empiricalevidence that would lend credibilityto the above proposal is a triflepatchy and indirectbut it does exist. One featurewhich emerges from a reviewof incometax structuresamongst the dynamicEast Asian economies is the restraintin the use of direct taxes. This can be illustratedby

26/ At the current15% VAT rate, collectioncosts of the VAT in the U.K are 1.2% of revenues.

27/ For the experienceof countriesthat have introducedthe VAT see: H. J. Aaron, The ValueAdded Tax: Lessonsfrom Europe,Brookings Insti- tution,1981; C. Sanfordand M. Godwin,Administrative and Compliance IssuesUnique to VAT Lessonsfrom two periodsof Britishexperience, DevelopmentResearch Department, World Bank, discussionpaper 192, 1986; C. S. Shoup,Criteria for Choice among types of VAT, DevelopmentResearch Department,discussion paper, World Bank, 1986.

28/ With the tax reformproposal currently under considerationin the US, the administrativecomplexity of incometaxes with a multiplicityof rates and the apparentinequities stemming from allowabledeductions has been under intensedebate. Empiricalevidence for flat taxes are contained in: "Is the Flat Tax a Radicalldea" James Gwertneyand James Long, The Cato Journal,Fall 1985. - 48 - a numberof indicesprovided in Table 5.2.391The tax base index is the ratio of the incomethreshold at which a positivetax paymentis made to the mean incomeof a standardfamily unit (FCDP). If the index is zero, all income is subjectto tax. Where it is 0.5, familieswith incomeless than half of the FCDP are not subjectto tax while those with incomesequal to the FGDP pay a tax on half their earnings. The largerthe deductions,credits and zero bracket,the greateris the index and the smallerthe tax base. How progressivethe effectivetax scheduleis can be gauged from column 3 which gives the marginaltax rate at three quartersof the mean f 'lilyincome. Interestingly,the East Asian countriesare fairlyhigh on the tax base index-- Singaporebeing an exception-- rangingfrom 0.37 for Hong Kong to 0.47 for Thailand,suggesting that familieswith betweena third and a half of mean familyincome pay no directtaxes. Fiji falls quite low on the scale, close to the level of Mexicoand Tunisia. A similarstory emergesfrom the marginaltax rate on 3/4 mean family income. From a low of 7% for Thailand, the East Asian economiesreach a high of 25X for Hong Kong, still well short of Fiji's 32.51.

29/ It is inherentlydifficult to compareincome tax structuresacross countries,because of difficultiesin measuringpurchasing power equiva- lences,differences in incomedistributions and the differencesin deductions,allowances etc. However,a recent studyhas attemptedto classifyincome tax structuresfirst by assuminga standardfamily size (equivalentto the averagefamily size in developingcountries) and then calculatingthe tax burdenin variouscountries in comparisonto the level of per capitaincome (as measuredby the World Bank) times the averagefamily size. With an averagefamily size of five personsin developingcountries, the index measuresthe relativeincome tax burden with referenceto the benchmarkfigure of five times the CDP per capita (FCDP). See G.P. Sicat and A. Virmani,Personal Income Taxes in DevelopingCountries, World Bank, mimeo, 1986. - 49 -

Table 5.2: TAX BASE INDEXAND INITIALMARGINAL TAX RATE

Tax Marginal Marginal Base tax rate tax rate Intdex on lst on 3/4 Country (Y*/FGDP) bracket FGDP

Fiji 0.18 5.0% 32.5% IvoryCoast 0.00 2.5% 1.2% Jamaica 0.45 30.01 45.0% Thailand 0.47 7.01 7.0% Singapore 0.08 3.6% 22.5% Hong Kong 0.37 5.0% 25.0% Korea 0.39 7.1X 10.61 Malaysia 0.47 6.01 15.0% Japan 0.11 14.5% 44.0%

Source: Sicat and Virmani,op. cit.

5.07 Table 5.3 bringsout some more detailson the steepnessof the tax schedule. In many countriesthe highest tax rate gives a very deceptive measureof progressivitysince it appliesto no more than a handfulof individuals.What mattersis the number of familiesfalling in the upper tax brackets. The table presentsthree indicators: the ratio of the highest bracketto the mean familyincome; the marginalrate on the highestbracket; and the rate on familiesearning three times the mean income. In Fiji the highestbracket is just five times the FGDP, similarto that in Malaysia. It is 8:1 in Korea and 21:1 in Thailand. Again,Hong Kong, Singaporeand Thailandfall below Fiji in termsof the uppermostmarginal rate and the rate on 3x FGDP. Malaysiais on par and Korea has a 70% levy on the highest bracket. - 50 -

Table 5.3: HIGHESTBRACKET AND HIGHESTMARGINAL TAX RATE

Ratio of Marginal highest rate on Marginal bracket highest rate on Country to FGDP bracket 3 x FGDP

Fiji 4.9 50.0X 45.0% Colombia 12.1 49.0X 44.0S Jamaica 0.9 57.51 57.51 Thailand 21.4 65.0Z 22.0X Ivory Coast 239.6 72.'1 2.21 Hong Kong 0.2 25.0X 17.0X Singapore 10.7 40.5. 36.0X Malaysia 4.4 55.0i. 45.01 Korea 8.1 70.1Z 44.61

Source: Sicatand Virmani,op. cit.

5.08 From these tables it seems that Fiji taxes incomesmuch more severelythan is the norm for East Asia, even after allowingfor the changes introducedon January1, 1986. To the extentthat the systemdiscourages economicactivity and directsresources into tax evasion,it impedesgrowth. Rate differentiation(see Annex Table 2) also inducessometimes costly transactionsto move incomefrom higherbracket to lower bracketfamily membersand since the taxpayeris subjectto differenttax rates in different years,to move taxableincome from one year to the next.

Taxes,Savings and Investment

5.09 The relationshipbetween tax incentivesand investmentis borne out by a wealthof economicresearch, though it is on occasionnot very strong, and the growthimplications can be evanescentwhen savingsdo not respondin parallel. Qualitativeevidence (and some econometricas well) from the US, Europeand Japan suggeststhat net of tax returnsare a powerfulinducement for savers. Three countrieswhere householdsavings are high providestrong tax incentives.In Germany,interest on depositsat savingsand loan associa- tions and insurancecompanies is tax exemptand there is a favorabletax treatmentof dividendsthrough the integrationof corporateand income taxes. Dividendand interestincome from governmentbonds, postal savingsand bank savingsdeposits is largelyexempt from taxes in Japan,while Italy exemptsinterest income from governmentbonds and levies10-201 rates on returnsfrom savingsaccounts, special savings stock and on dividends. Finally,experience with the individualretirement accounts (IRA) which are accordedtax privilegesby the US governmentalso lendsa measureof credence to the hypothesisthat taxes can influencesavings behavior. By contrasthigh Britishincome taxes may be guiltyof depressingsavings but are certainly responsiblefor concentratingwhat privatesavings there are in housing,life - 51 - insuranceand pensionfunds leavingvery littleliquid assets in the hands of individuals,a factorthat might accountfor the limitedsize of the small businesssector in Britain. All this suggestsagain that direct taxes on a moderatescale, according special treatment to incomefrom savingsand capital,are likelyto be most effectivefrom the standpointof growth.

Taxes in Fiji

5.10 The comparativeapproach to tax policy is instructive.But in order to determinewhether it can be used to advancesome concretesuggestions regardingfiscal reform in Fiji, a more detailedpicture of the country'stax systemis needed.

5.11 Centralgovernment tax revenuesare equivalentto 241 of GDP (1985); incomefrom non-taxsources yields an additional4% of CDP, resultingin a highlyrespectable revenue effort by the standardsof middle incomecountries (see Table 5.4). Directtaxes are responsiblefor 45% of tax income,far in excessof the 25Z average for semi-industrialcountries and it is becauseof their high buoyancythat revenueshave risen by over 101 p.a. since 1979, slightlyahead of expenditures(9.2Z p.a.) (see Table 5.5). Large budgetary deficitsand a stagnatingeconomy during the first half of the eightiespromp- ted the governmentto commissiontwo studies: one by the IMF, the other by a FinancialReview Committee.- The first concludedthat revenuerequirements ought to be satisfiedthrough a broadeningof the indirecttax base. Few changesin directtaxes were called for as the systemwas not too onerousand did not have any significantdisincentive effects. A contraryview was proposedby the Report of the FinancialReview Committee which criticizedthe high rates leviedon a small group of taxpayers. Both reportsalso recommen- ded changesin import,excise and corporatetaxes.

5.12 On a number of individualrecommendations of these two reports,the governmenthas acted. Importduties on raw materialsand capitalgoods purchasedby domesticindustrial concerns, which were totallyexempted from duty in 1982, have been raisedto an averageof about 7.5% in a seriesof discretechanges. Corporateincome taxes have been raisedtwice in the eighties,most recentlyin 1986,with foreigncompanies now being subjectto a 47.5% rate. On individualincome taxes, the governmentsided with its ov'i fiscalreview committee and loweredeffective taxes quite significantly startingon January1, 198fi(see Annex Table 2). Preliminaryestimates by the Ministryof Finance(MOF) indicatethat revenuesfrom individualincome taxes will declineby 9% in 1986, despitean expectedgrowth of 7.5Z in nominalGDP.

5.13 Excisetaxes, which as of 1983 were leviedon the domesticproduction of 17 items that were under importlicensing, were raisedand the cover- age has been graduallyexpanded to 47 items (the latestexcises were enacted in November1985). Preliminaryestimates by the MOF projectan increaseof

30/ See: IMF, Fiji SelectedTax Issues,Fiscal Affairs Department, August 25, 1983; and, The FinancialReview Committee Report, 1985, Parliamentof Fiji, ParliamentaryPaper No. 54 of 1985, August, 1985. -52-

Table 5.4: CENTRALGOVERNMENT REVENUES (as % of GDP at factor cost)

1979 1980 1981 1982 1983 1984 1985

Taxes on income and profits 10.7 11.6 11.9 11.8 12.6 13.1 12.3

Taxes on property 0.1 0.1 0.1 0.1 0.1 0.1 0.0

Taxes on goods and services 2.2 2.2 2.5 2.7 2.9 2.7 3.2

Taxes on international trade 6.7 5.9 7.6 5.8 7.1 7.3 8.4

Other taxes 0.4 0.2 0.4 0.4 0.4 0.3 0.4

Tax Revenue 20.1 20.0 22.5 20.6 23.0 23.5 24.2

Nontaxrevenue 4.0 3.8 4.3 4.4 4.4 4.1 4.0

Total Revenue 24.1 23.8 26.8 25.0 27.4 27.6 28.3

Foreigngrants 0.8 0.9 0.9 1.0 1.0 0.9 1.5

Source: Ministry of Finance. - 53 -

Table 5.5: COMPOSITIONOF CENTRALGOVERNMENT REVENUES

Taxes Social Income Domestic Inter- security and goods & national contri- Nontax profits services trade butions revenues

Industrial countries(20) /a 33.3 26.0 3.7 25.0 4.0 Semi-industrial countries(15) /a 25.3 30.6 14.5 13.0 11.1

Middle-income countries(55) /a 23.7 23.1 28.9 4.1 14.9 Least-development countries(14) /a 17.0 21.7 41.6 1.6 13.0

Fiji /b 44.8 13.3 27.3 0.0 14.6

/a Datedaround 1980. Numberof countriesaveraged in parentheses. lb 1985,revised estimate. Source: GovernmentFinance In DevelopmentCountries, Richard Goode, 1984, The Brookings Institute,p. 91. Fiji: Ministryof Finance. - 54 - goodsand servicesrevenues by 251 in 1986,as a resultof thesechanges. Thesetaxes continue to be leviedon a specificrate per productrather than on a percentagebasis. Taxeson miscellaneousservices have also been raised and the taxationnet expanded.Services such as bets,the rentalof video tapes,night clubs, vessels and restaurantswith liquorlicenses, and motor vehiclehires were taxedbeginning in 1984. In the lastbudget, live entertainment,the purchaseof air traveltickets, commercial advertising, retail salesof liquor,yacht chartering and lotterytickets came under the miscella- neousservice tax net. In addition,the hotelturnover tax was raisedto 5X and was widenedto includeall purchasesin hotels. Recastingthe Tax System 5.14 Insteadof a comprehensivereform of the tax system,the authorities haveinstead introduced piecemeal changes that are presumedto be collectively revenueneutral but whichdo reducethe overalltax buoyancy(see Annex Table3 and 4 on tax buoyancies).Long term revenue needs have not been cateredfor, nor have the disincentivesand distortionsarising from high directtaxes, been firmly tackled. Frequent tax modificationsgenerate uncer- taintyand are undesirable.At thisstage, when a freshindustrial strategy is beingconsidered a substantialtax reformthat reinforced the industrial drivehas its attractions.The elemit;s of sucha reformcan be tracedout drawingupon the earlierdiscussion.- _ Translatingit intoreality will be a considerabletask. (a) As describedin Chapter4 the tariffand importduty adjustments beingproposed could have positive revenue implications. Even if dutiesare so configuredthat revenues fall, the declineshould be marginal. (b) Ideally,the incometax couldbe reformedwith an aim towardsfewer brackets,lower tax rates,and fewerdeductions and exemptions. Sucha systemwould still be progressivealthough much lessso than the currentone. Moreover,tax administrationwould be simplified and the incentiveto evadetaxes would be reducedsignificantly. (c) To st,mulatehousehold savings, the exemption mits on interestand capitalincome might be substantiallyraised.3m

31/ See alsoD. Bradford,Untangling the IncomeTax, Harvard University Press,1986.

32/ Currently,F$200 of interestreceived by a residentindividual from any bankin Fijiand F$400of interestin any loanraised by the government are exemptfrom income tax. The 1987budget has proposedto raisethese limits. Dividendexemptions are grantedon a scalethat is inversely relatedto totalincome, which declines to onlya 331 deductionfor incomesexceeding F$9,600. - 55 -

(d) The corporatetax rate,which appearsto be high in comparisonto other East Asian NIC's,might be reexamined.

(e) As indicatedin chapterIV there is an advantagein applyingexcise taxes uniformlyto importsas well as domesticproduction so as to minimizeany unintendedinducement to manufacturean item locally. Further,by moving from specificto ad valoremrates, some of the regressivityof excise taxes would be reducedg$d the inflation relatederosion of real tax rates minimized.3

(f) Finally,the revenuedeficiencies arisin8 from the ruts in direct tazes would be made good througha broad-basedretail sales tax carryinga low uniformrate but which covered items such as food and daily necessitiesthat are generallyexempted. Over the longer term, the authoritiesmight plan on a suitablyconfigured VAT.

The purposeof this reformof directtaxes would be to encouragework effort and investment,increase savings and minimizethe distortionsinduced by high and differentiatedtaxes. A broad based consumptionlevy would ensurethat the volumeand buoyancyof revenueswas preserved. Growthand efficiency would be well served. Througha suitablechoice of rates and coveragefor both direct,indirect and importtaxes, revenue objectives could be rea- lised. But, a systemthat is firmlyoriented towards growth throughindustri- alizationwould undoubtedlycall for some sacrificein equity,though it need not be large.

33/ See, R. Goode,Government Finance in DevelopingCountries, Brookings Institution,1983, pp. 146-150. - 56 -

VI. TRADE: PATTERNSTRENDS AND INSTRUMENTS

6.01 Adversityis the true test of an export strategy. The recession which grippedworld trade in 1985 cruellysegregated the economiesof East Asia whose performanceis keyed to exports. All but two faltered. Those which weatheredthe suddendrought and entered1986 with their preeminence amongstthe NICs reaffirmedare both almost exclusivelyexporters of manufactured goods. That the abilityto trade in a wide range of manufactured productsis the touchstoneof successfulgrowth in the eightiesis resound- ingly confirmedby internationaltrade statistics. In the 5 years to 1985, the total volumeof exportsfrom developingcountries declined by 4X but exportersof maufacturersincreased their sales by 452, whereascountries exportingfuel experienceda 34% drop in exportvolume.

FlourishingExports

6.02 There is much to be learnedfrom the patternof trade. It cannot teach a countryhow to enter the ranks of the NICs nor does it lead to an infallibleindustrial recipe. What these patternscan indicate,albeit indistinctly,is the principallocus of tradingopportunities. Beyond that they can providean early warningof trendswhich are workingto shift this locus. From the mid-sixtiesuntil the close of the seventies,the highest rates of growthwere recordedby vehiclesand machineryand at least in the earlierpart of this period it was the industrializedwest which emergedas the most dynamictrading area. Both productgroups remainstrong and a glance at the ten most rapidlyexpanding imports into the US, Japan, the EEC and the East Asian economiesbetween 1975 and 1981 shows that metal products,transport equipmentand electronicshave all risen in importance(see Tables 6.1- 6.4). More significant,flagging economic performance in WesternEurope and the tremendousstrides taken by the East Asian nationshave made the Pacific region the most active tradingarea.

6.03 The momentumof industrializationand trade around the PacificBasin is so powerfulthat this area could continueto widen its lead over other parts of the world for years into the future. Further,a simplereading of trendssuggests that officemachines, telecommunication devices, electronic parts,transport equipment and machinetools which are among the strongest exportcandidates from the East Asian economiescould lead the field until well into the nineties(see Tables6.5 and 6.6). Betweena third to one half the growth in exportsfrom East A31t in the mid-eightiescan be directly attributedto demandfrom the US. In fact the opennessand absorptivityof

34/ US demandfor East Asian exportshas been vital for some time. Since the late seventiesEast Asian trade with the US has risen two and a half times the rate of US trade with the rest of the world. Each year the dollarvolume amounts to $30 billionmore than the total US trade with Europe. In 1984, Japan,South Korea,Taiwan (China)and Hong Kong were among the top ten tradingpartners of the US. D. Aikman,Pacific Rim LittleBrown, 1986, p. 9. -,57 -

Table 6.1:,'MAJOR,'1*RT ItEKS RNTO NORTH AMERICA BBY C t RATES, 1975-81

Fastest-GrowingImports frtom the Xrld

Value Growth in 1981 rate SITC Item ($ mln) ()

285 Silverand platinumores 638.7 40.68

672 Iron, steel primaryform 449.5 37.16

734 Aircraft 4,343.6 35.31

681 Silver,platinum .1778.8 30.55

689 Nonferrousbase metals 679.9 30.16

731 Railwayvehicles 361.9 29.08

678 Iron, steel tubes, pipes 5,718.5 28.78

896 Works of art 2,149.0 28.00

122 Tobaccomanufactt.ring 317.2 27.36

715 Metalworkingmachinery 2,398.6 26.58

Source: UN Trade Data (WorldBank Trade System). - 58 -

Table 6.2: MAJORIMPORT ITEMS INTO JAPAN RANKEDBY GROWTHRATES, 1975-81

Fastest-GrowingImports from the World

Value Growth in 1981 rate SITC Item ($ mln) (%)

674 Iron, steel plate, sheet 250.1 145.21

672 Iron, steel primaryforms 211.6 64.96

284 Nonferrousmetal scrap 501.8 41.71

521 Coal, petrochemicals 99.9 39.57

735 Ships,boats 843.6 38.13

684 Aluminum 1,848.9 55.91

931 Specialtransactions 976.3 35.83

726 Electromedicalx-ray equipment 142.0 34.69

734 Aircraft 1,340.5 33.39

842 Fur clothing 202.5 31.49

Source: UN Trade Data (WorldBank Trade System). - 59 -

Table 6.3: MAJORIMPORT ITEMS INTO THE EEC RANKEDBY GROWTHRATES, 1975-81

Fastest-GrowingImports from the World

Value Growth in 1981 rate SITC Item ($ mln) (%)

961 Coin, not of gold 908.8 159.02

285 Silver and platinumores 905.6 35.14

734 Aircraft 8,976.o 31.8.2

521 Coal, petrochemicals 737.8 29.25

897 Gold, silverware,jewelry 1,091.6 26.X7

332 Petroleumproducts 34,343.0 25.63

842 Fur clothing 813.3 24.17

681 Silver,platinum 2,047.0 24.09

689 Nonferrousbase metals 935.8 23.32

891 Sound recorders 5,157.9 22.58

Source: UN Trade Data (WorldBank Trade System). - 60 -

Table 6.4: MAJOR IMPORTITEMS INTO EAST ASIA ECONOMIES RANKED BY GROWTHRATES, 1975-81

Fastest-GrowingImports from the World

Value Growth in 1981 rate SITC Item ($ mln) (%)

521 Coal, petrochemicals 93.6 97.37

951 War firearms 67.4 74.77

731 Railwayvehicles 155.7 43.34

283 Nonferrousbase metal ore 89.6 40.55

733 Road vehicles,nonmotor 165.7 37.75

851 Footwear 215.8 37.24

897 Gold, silverware,jewelry 206.1 37.14

841 Clothing 1,136.9 37.07

714 Officemachines 736.3 35.51

891 Sound recorders 758.5 32.75

Source: UN Trade Data (WorldBank Trade System). - 61 -

the Americanmarket is a crucialcounterpoint to East Asian development. Simpleextrapolation supports a tradingfocus centeredon the US. Not only has that countryexhibited an insatiableappetite for an ever wideningvariety of imports,but the sophisticationof Americanmarketing and serviceindustries also helps to catalysethe penetrationof foreigngoods. For new exportersattempting to carve a share of the business,this can be of ines- timablevalue.

Table 6.5: LARGESTMANUFACTURED /a EXPORTSFROM SELECTEDASIAN COUNTRIESTO MAJOR OECD ECONOMIES/b (US$ million,ranked by 1983 value)

SITC Commodity 1975 1978 1980 1981 1982 1983

KOREA

841 Clothing 895.7 2t173.4 2,447.8 2,995.0 2t979.0 2,951.6 851 Footwear 177.1 644.1 804.2 864.4 1,038.0 1,156.7 721- Telecommunications equipment 124.7 506.0 677.2 747.2 735.0 1,044.0 729 Electrical machinery,NES 206.0 430.0 511.8 494.0 580.6 824.4 735 Shipsand boats 0.2 1.6 27.0 11.0 35.5 596.5

HONG KONG

841 Clothing 1,625.5 2,860.3 3,832.2 3t921.8 3,861.2 3,972.5 894 Toys, sporting goods,etc. 240.4 588.3 886.6 943.0 1,104.4 903.7 724 Telecommunications equipment 208.2 368.4 529.7 572.5 566.9 760.0 864 Watchesand clocks 63.5 379.5 784.2 710.7 547.1 650.9 714 Officemachines 84.0 214.3 324.3 327.5 304.3 647.8

SINGAPORE

729 Electricalmachines, NES 230.4 376.2 870.4 889.8 968.3 893.5 714 Officemachines 56.1 61.6 96.4 134.0 288.7 834.1 724 Telecommunications equipment 124.2 203.3 729.4 768.2 690.8 728.1 841 Clothing 81.4 153.8 305.4 308.0 308.2 312.0 722 Electricalpower machinery 29.2 60.0 127.1 142.3 145.7 162.0

/a Manufacturedexports are definedas SITC 5, 6, 7 and 8 minus 68. 7i Major OECD economiesare USA, Japan, and EEC.

Source: World Bank Trade Data Base System. - 62 -

Table 6.6: FASTESTGROWING MANUFACTURED /a EXPORTSFROM SELECTEDASIAN COUNTRIESTO MAJOR OECD ECONOMIES/b (US$ million,ranked by 1978-83growth rate)

Annual growthrate SITC Commodity 1978 1980 1983 1979-83

KOREA

735 Ships and boats 1.6 27.0 596.5 226.7 672 Iron and steel,primary forms 13.5 161.9 223.4 75.5 725 Domesticelectric equipment 11.3 22.2 125.7 61.8 521 Coal, petroleum,etc: chemicals 11.90 42.1 93.28 50.7 673 Iron and steel shapes 13.5 38.7 61.8 35.5

HONG KONG

719 Machines,non-elec. NES 24.1 114.7 225.2 56.3 722 Electricalpower machinery 69.9 132.9 281.4 32.1 714 Officemachines 214.3 324.3 647.8 24.8 891 Sound recorders,producers 71.0 115.4 172.6 19.4 725 Domesticelectric equipment 100.3 206.6 240.0 19.1

SINGAPORE

512 Organicchemicals 1.1 4.4 84.7 138.9 714 Officemachines 60.6 96.4 834.1 69.0 891 Sound recorders,producers 22.5 59.3 116.8 39.1 725 Domesticelectric equipment 29.5 68.8 106.1 29.2 821 Furniture 16.0 38.7 53.3 27.2

/a Manufacturedexports are definedas SITC 5, 6, 7 and 8 minus 68. 7T Major OECD economiesare USA, Japan, and EEC.

Source: World Bank Trade Data Base System.

6.04 Trends in world trade have a certain inertiaand can serve as fairly reliableforecasting tools for the medium run. If an export strategywere to be groundedin just the recent internationaltrade patternsthen the manufactureof machineryand electronicequipment, preferably with assistancefrom Japaneseand Westerncompanies, for sale to the US market,would be the most sensiblecourse for a small open economyto pursue. Unfortunately,the choicesare never so simple. The interpretationof trendscannot afford to ignorepolitical, technological and economicchanges worldwide that, over time, can undo a patternof trade flows and replace it with a quite different one. Nor can trade policiesignore the country'scurrent trading situation and the resourceswhich can be mobilizedfor developingmanufactured exports. - 63 -

FactorsInfluencing Trade

6.05 Trade is being subjectedto pressuresfrom many directions. Unem- ployment,deindustrialization and balanceof paymentworries have revived protectionisttendencies in the western countries. A large share of the manufacturedexports from the developingworld are subjectto quotas or voluntary restrictionsand having learnttheir lesson,industrial groups in the West are now quick to react preemptivelyat the first sign of an impendingattack by foreignsuppliers by askingfor legal or legislativeinjunctions. Close to one fifth of manufacturedexports from the developingcountries to the industrialnations are subjectto non-tariffbarriers (see Table 6.7). As protectionismaffects the high volume items such as textilesand garmentsas well as the fastestmoving productsin world trade,for examplemicro chips, it is capableof arrestingcertain trends and profoundlyinfluencing the productmix as well as the geographicalpattern of exports. Hong Kong firms feelingthe pinch of quotasbegan establishingsubsidiaries in Malaysia,the Caribbeanand Mauritius. China and Thailandenlarged their share of the garmentbusiness. Productsare being upgraded;silk and ramie have partially displacedcotton and syntheticfibres. In short,the politicaltug-of-war is no less significantthan economicforces. The LOME Convention,the Caribbean Basin Initiative,SPARTECA as well as other trade agreementshave joined resourceendowments and industrialpolicy as the arbitratorsof comparative advantage.

6.06 The durabilityof trade patternsmight also be compromisedby two other developments:automation and movementsin exchangerates. Revolution- ary electronicinnovations are making possiblethe automationof nulyous labor intensiveassembly operations in industriessuch as garments,3- semi- conductors,machinery and automobiles. Since these are the very activities which have enableddeveloping countries to industrializeand export,automation poses a seriousthreat. A reversemigration of assemblyindustries to the advancednations could be immenselydamaging to the industrialhopes of developingcountries and it would certainlyrequire a fundamentalrethinking of what constitutesthe best export opportunities.

6.07 Second,the flux in exchangemarkets, the realignmentswhich have occurredbetween the currenciesand the aggressiveuse of exchangerate policy to securea tradingadvantage makes for great uncertaintyespecially for small economiesseeking exports. As long as exchangemarkets remain fluid long run competitivenessin light manufacturesand assembledproducts is built on shiftingsands. And the flow of direct foreigninvestment cannot be projected with any assurance. In effect,industrial and exchangerate policieswill need to be managedin tandemto give entrepreneursstable long-run incentives in the face of a volatile,competitive trading environment.

35/ Clothing,Chips and CompetitiveAdvantage: The Impactof Miroelectronics on Trade and Productionin the GarmentIndustry, by Kurt HoffmanWorld Development,Vol. 13, No. 3, March 1985. - 64 -

Table 6.7: NONTARIFFBARRIERS - PERCENTAGEOF INDUSTRIALCOUNTRY IMPORTSFROM DEVELOPING COUNTRIES COVERED BY NTBs (X of imports covered)

All goods Manufactures Agricultural All US EEC Japan All US EEC Japan All US EEC Japan

Thailand 17.8 20.3 12.3 24.3 17.3 18.7 29.6 3.4 20.1 35.5 1.8 52.0

Indonesia 27.3 84.8 12.6 1.7 4.3 5.8 9.9 0.3 25.2 0.1 12.0 67.2

Malaysis 10.2 20.6 9.9 2.6 5.2 3.1 11.8 1.1 10.7 0.5 4.3 37.4

Philippines 14.9 20.0 21.7 4.2 20.1 19.5 36.1 2.2 13.1 20.3 5.0 11.5

China 26.2 59.1 34.1 11.5 30.5 51.7 40.2 6.0 36.6 59.2 19.4 41.8

NICs 23.8 38.6 26.7 6.3 22.4 23.6 26.7 3.5 24.7 21.4 9.1 43.4

Developing 32.2 53.6 22.4 8.1 20.5 19.4 27.6 4.2 23.6 19.5 17.5 45.9

Industrial 18.7 23.8 19.9 19.9 12.6 13.9 13.0 9.4 18.7 21.3 27,1 31.5

Note: As of June 1984 with 1981 improtweights.

Sources: UNCTAD and World Bank data bank. - 65 -

6.08 This is the backdropagainst which Fiji'spast export performance must be assessedand an attemptmade at sketchingfuture moves.

ManufacturedExports from Fiji

6.09 Comparedto Mauritiusor Jamaica,island nations that are broadly similar,Fiji's manufactured exports are a much smallerpercentage of the total. They amountedto as littleas 5% in 1985 whereasmanufactures com- prisedclose to 43X of Mauritius'exports; 20.5% of Jamaica'soverseas sales fell into that category. Plywoodveneers is the singlelargest item with a 35Z share followedby clothing,soap, plastics,paint and perfumes(see Table 6.8). The directionof trade is heavilyinfluenced by sugar contractswith the UK, Malaysiaand Australiaas well as the sale of gold to Australia. Once these transactionsare nettedout and the focus narrowedto manufactures, Fiji'sprincipal trading partners are respectively:Australia, the US, New Zealandand the neighboringPacific Islands (see Table 6.9). To Australiaand New Zealand,Fiji sends vegetableoil, preservedfruit, clothing, plywood veneer,tinned fish and spices. This is a good mix of commoditiessince it includessome of the fastestgrowing imports into these markets(see Tables 6.10 and 6.11). Garmentsare also exportedto the US and that country is an expandingmarket for Fiji'sfruit juicesand its high qualityginger.

6.10 On the positiveside it can be claimedthat exportsof manufactures have risen steadily,between 1980 and 1985. But the magnitudesinvolved are modestand relativeto comparatorsFiji has been slow in diversifyingits exports,seeking new outletsand in forginglinks with the dynamiceconomies aroundthe Pacificrim. Experiencegained by other nationswho have trodden this very path offersfour possibilities.Each, however,is hedged aroundby constraintsrooted in the developmentsnoted above. -66-

Table 6.8: FIJ' - MAJOR EXPORTS BY THREE DIGIT SITC /a (US$ million)

Growth rate I 1975 1980 1983 75-80 80-83

Total (incl. gold) 138.3 283.0 174.0 Gold 10.4 15.2 16.4

Total (excl. gold) 127.9 267.8 157.6

Food & Live Animals (0) 117.9 248.6 136.0 Fish tinned (032) 0.2 10.6 IT1.5 123.0 2.9 Cereal (048) 0.6 1.9 1.9 31.2 0.5 Fruit preserved (053) 0.3 0.9 1.2 36.4 7.0 Sugar and honey (061) 114.4 229.6 112.6 17.1 -20.5 Spices (075) 0.7 1.5 2.2 15.2 15.3 Animal feeding (081) 0.7 0.9 1.0 10.8 5.5

Beverages& Tobacco (1) 0.1 0.1 0.2

Crude Materials (2) 1.0 6.2 2.8 Wood shaped (243) 0.4 3.5 T7 48.5 -16.3 Crude animal materials (291) 0.1 0.3 0.6 22.9 16.7 Crude vegetable materials (292) 0.2 0.3 0.2 9,0 -10.1

Mineral Fuels (3) 0.0 0.0 0.0

Animal & Vegetable Oils (4) 6.0 8.1 10.4 Fixed vegetable oils (422) 6.0 8.1 10.4 15.9 6.7

Chemicals (5) 0.4 0.9 1.7 Pigments,paints (533) 0.2 0.3 0.5 9.6 18.8 Perfume, cosmetic (553) 0.0 0.3 0.3 289.8 1.7 Soaps, cleaning (554) 0.0 0.2 0.5 47.7 39.3 Plastic material (581) 0.0 0.1 0.3 47.1 55.8

Basic Manufactures(6) 1.9 2.4 4.2 Veneers, plywood (631) 0.7 1.2 2.5 11.3 26.0 Articles of paper (642) 0.2 0,1 0.2 3.1 33.7 Iron, steel univ. (674) 0.0 0.1 0.4 28.6 62.9 Steel, copper nails (694) 0.0 0.2 0.5 35.6 37.6

Machines, TransportEquipment (7) 0.0 0.1 0.0

Misc. ManufacturedGoods (8) 0.3 0.7 1.1 Clothing (841) 0.0 0.1 0.5 18.9 51.7

Goods. Nel (9) 0.3 0.8 1.1 Special transaction(931) 0.3 0.8 1.1 27.6 7.8

Memo Items Manufactures(5-8) 2.6 4.1 7.0 9.5 19.3

As % of Total Sugar 82.7 81.1 64.7 - - Gold 7.5 5.4 9.4 - - Manufactures(5-8) 1.9 1.4 4.0 - -

/a Totals may not add due to rounding.

Snoiurca nU TrAde qvmtame Table 6.9: FIJI - EXPORTS BY SITC SECTIONSAND DESTINATION (%)

(0) (1) (2) (4) (5) (6) (7) (8) (9) Bev. & Crude Oils & Chemi- Manufac. Machi- Misco commodities SITC Food tobacco materials fats cals goods nery Manufac. NES Total

Destination UK 51.6 0.5 3.3 0.0 0.0 0.0 0.9 0.0 0.4 37.3 Australia 1.2 0.0 14.4 41.6 9.0 34.9 16.5 35.6 97.8 17.1 US 9.9 1.4 2.3 57.8 0.4 17.4 0.0 9.1 0.2 13.3 Malaysis 16.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 12.0 New Zealand 3.6 40.4 37.1 0.4 6.6 13.6 35.8 13.5 0.6 4.4 Japan 4.1 0.0 19.1 0.0 0.0 0.8 0.0 0.9 0.1 3.5 Canada 2.9 0.5 1.4 0.0 0.1 0.0 0.0 0.3 0.0 2.2 China (PRC) 2.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.5 Singapore 1.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.8 Hong Kong 0.3 6.9 1.7 0.0 0.0 0.1 0.0 0.3 0.0 0.3

Pacific Islands 3.0 8.3 1.3 0.2 51.0 28.0 23.9 30.8 0.2 3.8

Others /a 3.6 42.2 19.5 0.1 32.9 5.2 22.9 9.3 0.8 3.8

Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Memo Items SITC as % of Total Exports 72.0 0.1 2.9 9.4 0.9 3.0 0.1 0.9 10.7 100.0

/a Includes ship stores.

Source: Bureau of StatisticsFiji, CurrentEconomic Statistic,April 1985, T. 84. - 68 -

Table6.10: AUSTRALIA-- FASTEST-GROWINGIMPORT ITEMS

Growth rates 1975-80 1980-84

046 Wheat -20.57 (0.0)/a 42.45 (0.0)

048 Cereal 9.55 (0.0) 20.24 (0.1)

053 Fruit preserved 19.83 (0.7) 15.79 (1.2)

642 Articlesof paper 18.30 (0.0) 11.87 (0.0)

051 Fruit fresh nuts 15.67 (0.6) 11.81 (0.1)

899 Other manufactures 12.87 (0.0) 11.34 (0.0)

211 Hides skins -32.11 (0.0) 11.03 (25.3)

/a In parentheses,Fiji's export shareas a percentageof total importsin the terminalyear.

Source: UN Trade Tapes.

Table6.11: NEW ZEALAND -- FASTEST-GROWINGIMPORT ITEMS

Growth rates 1975-80 1980-84

046 Wheat etc. meal 12.94 (0.0)/a 113.64 (0.0)

062 Sugar preps 17.88 (0.0) 31.57 (12.0)

661 Cement etc. bldg. -13.49 (0.0) 28.48 (0.0)

422 Fixed vegetableoil 30.47 (0.0) 27.34 (2.6)

553 Perfume,cosmetic 17.63 (0.0) 20.65 (0.0)

/a In parentheses,Fiji's export share as a percentageof total importsin the terminalyear.

Source: UN Trade Tapes. - 69 -

Trade Agreements

6.11 In severalinstances, trade agreementshave servedto triggerexport growth. Mauritiusis a good example. Market access to the EEC gained through the Lome conventionhas resultedin a flourishinggarment industry which in 1985 displacedsugar as the principalexport. Severalof the Caribbean countrieshave begun to attractthe attentionof foreigninvestors following the CaribbeanBasin Initiativeand are startingto make theirmark as exporters of lightmanufactures, processed food, fruit and vegetables. Haiti has, in a few shortyears, become a leadingproducer of baseballs,while a number of CentralAmerican countries do a thrivingbusiness in assemblinggarments from materialshipped to them by US firmswho SO e advantageof their low labor costs and US customsregulation 806/807.-

6.12 Fiji standsto gain from two arrangements. First,SPARTECA was completedin Canberraduring May 1980 and signed shortlythereafter at the EleventhPacific Forum held in Kiribati. It is a comprehensivenonreciprocal pact underwhich Australiaand New Zealandprovide privileged access to exporterslocated in the Forum Island countries. SPARTECAdid not becomea hinge for export-ledgrowth in Fiji becauseof problemsthat are gradually being negotiatedaway. First, the rules of originprovided under this agreementwere difficultto satisfy,particularly the valueadded component from New Zealandand Australia. The rules of origin specifiedthat, in order to preventthe use of the islandsas way stationsfor exportsby third countriesto Australiaand New Zealand,at least 50% of the value-addedmust be from a Forum memberor from Australia/NewZealand. But this was compli- cated by the additionalprovision that if the goods were exportedto New Zealand,50% of the value addedhad to originatein a Forummember or only New Zealand,and if the goods were exportedto Australia,it must contain Australianvalue-added rather than New Zealandvalue-added. This arrangement involvednumerous inconveniences: a manufacturerof knitwear,for instance, who exportedto both New Zealandand Australia,had to separatehis raw materialaand guaranteethat sweatersbound for the Australianmarket were made with wool from the same country. This provisionwas an outgrowthof the cool tradingrelations between New Zealandand Australia,which were resolved in 1985 by the ANZCERTAtrading agreement. In late 1985,the rules of origin were chengedallowing 50% value-addedfrom a Forum nationand eitherNew Zealandor Australia,as long as the item is freely tradedbetween New Zealand and Australia. This move effectivelyremoved the constrainton the valueadded rulesof origin.

6.13 A secondconstraint on the operationof SPARTECAarose from the range of goods embracedby the agreement. From the startNew Zealandhewed to a negativelist of items,not permittingpassion fruit products, coconut cream,lime juice,footwear and apparelimports. Australiameanwhile, main- taineda three-tiersystem of approvedproducts with varyingtariffs according

36/ US tariff items806/807 permit the duty free entry of US componentssent abroad for processingor assembly. See J. Grunwaldand K. Flamm,The GlobalFactory, Brookings Institution, 1985. - 70 - to the tier in which the productwas listed. This rule was consideredunfair by the Forum members,who preferredthe approachtaken by New Zealand. After a prolongeddebate, the Australiansagreed to adopt a negativelist beginning on July 1, 1986. This includesthe followingitems: steel products,cars, textiles,sugar and footwear. But there is a separateagreement between the Australiansand the Forum countriesallowing about 67,000units of clothingto be imported,as a specialallowance, to the Australiantextile quota scheme.

6.14 SPARTECAexerts a powerfulfascination for Fijianexporters because the proximityof Australiaand New Zealandtogether with traditionaltrading links make them the marketsof choice. But the successiveLONE conventions betweenthe EuropeanEconomic Community (EEC) and the AfricanCaribbean and PacificStates (ACP) providea privilegedentry into a broaderand more lucra- tive market. Under the LOME conventionexports from the ACP statesare coveredby a general rule allowingfreedom of accesswithout quantitative restrictionsand exemptionfrom customsduties for a wide range of goods. Sugar is also eligible,although each produceris allotteda specialquota at EEC prices.

6.15 As can be seen from the table,Fiji exportslittle other than sugar and tinnedfish to the EEC and this state of affairshas persistedfor the past five years. Nor have exportsfared any betterunder SPARTECA as is apparentfrom Table 6.12. The quantumof goods bound for Australiaand New Zealandhas increasedvery modestlybetwe-an 1983-85. This rathergloomy pictureis redeemedby the interestthat Australianbuyers have shown in Fijianmade garmentsand the risingdemand for Fiji'srum in New Zealand.37/

37/ One thousandcases of rum were shippedto the US in 1985. Sales to New Zealandhave averagedF$300,000 p.a. for the past severalyears. - 71 -

Table6.12t FIJI- SPARTECATLADE WITH AUSTRALIA AND NEW ZEALAND (F$ f.o.b.)

1983 1984 1985/a

Australia 8parteca Items Traditional Coconutoil 4,281,218 7,709,883 4,474,179 Logs and dressedtimber 97,626 685,961 516,191 Veneerand plywood 1,521,199 1,630,277 2,030,257 Nontraditional Fruitand vegetables 610,081 837,850 604,427 Passionfruit pulp 32,204 44,619 28,039 Tara(dalo) 25,463 142,199 75,338 Glyceroland lyes - 30,828 42,224 Raw hidesard skins 86,023 51,622 36,166 Stationery 20,528 83,843 116,450 Furniture 920 35,458 59,281 Apparel 364,357 510,558 1,1516,243 Total 7,039,619 119763,098 9,498,795 New ZealandSpartec Items Traditional Timber(dressed & undressed) 926,352 2,111,643 1,435,221 Veneer 174,333 309,315 297,433 Plywood 154,330 342,377 500,629 Nontraditional Spirts& beer 47,314 87t,806 228,823 Soaps 171,802 99,652 160,570 Paints& varnishes 13,857 12,949 8,746 Nails 325,978 157,548 12,510 Hatches 43,029 2,638 37,891 Cannedfish 252,080 107,509 136,336 Sugarconfectionary 142,877 118,560 37,865 Stationery 154,217 1,261 6,750 Travelgoods 43,846 28,769 - Coconutoil 116,095 70,838 290 Total 2t576,110 3,450,865 2,863,064

/a Provisional. Source: Bureauof Statistics. - 72 -

6.16 Trade agreementsdo not automaticallyunleash exports; at an open- ing, it is up to Fiji. /businessmento force theirway in throughthe breach that has been opened.- The agreementsmust now be made to serve Fijian interests. If obstaclesremain, exporters must learn how these might be side- steppedor, alternatively,problems should be resolvedat the negotiating table.

DirectForeign Investment

6.17 Severalof the East Asian countriesthat have becomeexporters of high technologyitems relied on foreigninvestment to give them the capital, technology,mar gementand marketaccess: in a word, turn-keyexporting is what they soughtand achieved. As the long arm of politicshas made itself felt on the tradingscene, market access throughpolitically influential foreignpartners is virtuallyindispensible for certaintypes of products. Unfortunately,protectionist moves by the industrialnations has begun to discouragemultinational corporations (MNCs) from locatingtheir factoriesin placeswhere the labor is cheap and fiscalincentives attractive. Instead, American,Japanese, and German companiesthat comprisethe principalinvestors increasinglyprefer to site productionfacilities in the industrialeconomies where much of the output is sold. Automation,by diminishinglabor requirements,is contributingto this shift.

6.18 While the growth in DFI might be restrained,the shiftingof investment betweendeveloping countries will continueto be drivenby the quest for politicalstability, cheaper labor and the opportunitiespresented by prefer- entialtreaties. Thus, suitableincentives could draw Australianand New Zealandcompanies, who appreciatethe avenuesopened by SPARTECA,to Fiji. Capitalfrom Hong Kong worriedabout the colony'sfuture and seekingquota linkedmarket access to the US and the EEC might also find Fiji a suitable platformfor the productionof garmentand electronicitems. And the appreci- ation of the yen will undoubtedlyinduce Japanese firms to sourcemore of their labor intensiveproducts from facilitiesoverseas. Skillfulsalesman- ship which emphasizesFiji's politicalstability, infrastructure and incentives could attractmore Japaneseinvestment to the South Pacific.

ExportProcessing Zones

6.19 The earliestEPZs near Shannonairport, Penang, Singapore, Masan in Korea and elsewherein East Asia were unusuallyeffective in luringforeign investors. The demonstrationeffect inducedmany other countriesto follow

38/ The barrierslikely to be faced by Fijianbusinesses in new marketsand the successwhich can emanatefrom assiduousmarket researchand persist- ence is illustratedby Kiwi United's(a Fijianmatch manufacturer) attemptsto penetratethe New Zealandmarket, in spite of the dominance of a major Australiancompany, Bryant and May. See, PacificIsland BusinessNews, May 1981, p. 7 or for a summary,Fiji: Client State of Australasia,E. Utrecht,Transnational Corporations Research Project, Universityof Sydney,Sydney, 1984, pp. 240-1. - 73 - suit so that by the mid-eightiesthere were some 79 operationaleconomic zones,export processing zones, and scienceparks availablein 35 countries3 offeringa fairlyhomogeneous package of inducements.Interestingly, American companiesremain the largestinvestors, almost a third of the firms are con- nectedwith the electronicsindustry, textiles and garmentsis a close second; and more than half of the labor force of the EPZs, worldwide,is employedby manufacturersof electronics.

6.20 An EPZ in Fiji will not guaranteethe resultsobtained by the pion- eers in tl late sixtiesand early seventiesbut the recentexperience of Sri Lanka and Lauritiusdoes providegrounds for hope. In spite of the competition from other zones, their geographicallocation and transportproblems, they were notablysuccessful, suggesting that the market may not be entirely saturated. Experienceshows that EPZs flourishin countriesenjoying political stability,where the supplyof low-pricedfemale labor is abundantand where an adequateinfrastructure exists to serve the needs of industryas well as residentexpatriates. Local entrepreneurialparticipation is certainlya plus, as are low freightcosts to industrialmarkets and the ready availabilityof technicalprofessional and repairservices. In China and India the bureaucracyis a sourceof frictionand discontent,whereas the organizational efficiencyof some East Asian NICs is much admiredby foreignfirms.

6.21 Transportwill remaina problemin the medium run thougha growing EPZ would certainlyattract airline interest, but Fiji could, in due course, deliveron the other essentials. Enteringthe EPZ sweepstakes,possibly throughthe creationof an industrialpark near a major transportnode and equippingit with electricity,water and seweragefacilties, might, therefore, be well worth consideringas a strandin the export strategy. However, it is an optionthat shouldbe tackledwholeheartedly or not at all. Satisfy- ing the earliestarrivals through suitable preparations ensures continuing interestby foreigninvestors. As with any other service,the creationof an EPZ must be precededby market researchto assessthe receptivityof MNCs and, once it has been launched,a marketingcampaign would be needed for the Fijian EPZ to draw abreastof competitors.

ExchangeRate Policy

6.22 An imaginativeuse of industrialincentives is one characteristicof developmentstrategy coined by East Asia's NICs, but in a few cases the flexiblemanagement of the exchangerate was almost equallysignificant in maintainingexport competitiveness and the relativep.ofitability of outward- orientedactivities. Korea,which enteredthe sixtieswith an economicstructure dominatedby agriculture,was especiallybold in manipulatingthe exchangerate to influencethe relativeprices of its nascentmanufactured products. A smalleropen economyis somewhatmore circumscribedin its use of exchangerate policiesfor industrialdevelopment, especially when import dependenceis high and the modest size of the nontraded-goodssector minimizes

39/ See J. Currie,Export Processing Zones in the 1980s,Economist Intelli- gence Unit, SpecialRep .rt No. 190, 1985. - 74 - the scope for an exchange-rate-inducedmovement in resourcestowards traded goods. However,the evidencefrom East Asian NICe, suggeststhat in spitseof the constraintsimposed by small size, they have been able to furthertheir industrialand exportambitions through a debt handlingof exchangerates.

6.23 The Fijiandollar has remaineda relativelystable .arrency. Externallyit has been most influencedby the parityof the Australiandollar; internallyit has been affectedchiefly by a conservativemonetary policy, a large and efficientsugar industryand a well-protectedmanufacturing sector. From Chart 2 it can be seen that the real effectiveexchange rate was restrictedto a narrowband throughmuch of the eighties. It rose in mid-1985 but has since been declining,most recentlybecause of a SS depreciationof the Fijiandollar against the US$ that occurredduring June-July 1986. Appropriatemanipulation of the REER could impartadditional stimulus over the longerrun to the productionof tradeablesif it were to depressthe share of wage costs,which currentlyreflect the earningsand productivityin the sugar industry. On the other hand, from the perspectiveof the agriculturesector, there would appear to be scant cause for modifyingpolicy. A declinein the REER would redistributeresources from importersto sugar producers4with no change in sugar exports,the local cost of carryingexternal debt will increase,higher importprices will immediatelyimpinge upon costs throughout the economyand the supplyresponse from manufacturedor even agricultural exportswill be small in the short run.

6.24 Using an explicitREER targetmay not be good policybecause a long run equilibriumis practicallyimpossible to define. Further,wage moderation combinedwith productivitygrowth may be a more efficaciousroute to industrial competitivenessthan adjustingthe REER. But exchangerate management, sensitivelyused, can complementthe actionof incentivesdiscussed in ChapterIV and of labor markot policies.

6.24 Internationaltrade patternsand the geographicaldistribution of importgrowth can serve as very rough indicatorsof what to produceand where to concentratethe effortof selling. Trade agreementscan open trails throughthe accumulatingundergrowth of nontariffrestrictions. Direct foreigninvestment by way of an EPZ could widen the range of exportsand exchangeratai policy could sharpenincentives for local manufacturersby neu- tralizingFiji's (productivity-adjusted)wage disadvantages.East Asian NICs have fashionedeffective strategies from these. It is a matter of trying,of learningfrom experienceand of perseveringwith an approachto development. CHART2 FIJI: REAL EFFECTIVEEXCHANGE RATE INDICES

REER INDEX 1.04] (I

I "?Z -I j

199 1

6_ /

1234123~412354123.41 23412.34123~412Z.412.34123412~3412 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

(1978-80 = 100) - 76 -

VII: LABORMARKETS AND THE GROWTHOF EXPORTS

Small Statesand the Control of Wages

7.01 Size can buy time. Large economiesare able to take their time in adaptingto changesin the internationaltrading environment by modifying institutions,increasing investment in technologyand new plant and remedying organizationalweaknesses. In rare cases, their market power allows them to tailorthe evolutionof tradingrelationships to their convenience,forcing their competitorsto shouldersome of the costs and dislocationsentailed by shiftsin technologyand economicstructure. Small statescannot easily bargainfor time or specialtreatment. If the internationalmarket environment changes,they must adapt and absorb the cost of transition. Flexibility and the capacityto respondpromptly to externalevents comes from a willing- ness among the variouseconomic groups to reorganizeand share the burden of vulnerabilityas well as the existenceof robust institutionalmechanisms for distributingcosts and gsains.As wages accountfor the lion'sshare of the nationalproduct, the speed of adjustmentto changeand the abilityto seize an opportunityrests on the functioningof the labor market. Management skills,capital supply, the rate at which technologyis assimilated,gains in productivityare all significant,but the flexibilityof real wages is the final arbiter of a small economy's short-term competitiveness.

7.02 In Europethe smallerstates that have prosperedduring the past three decadesare those that have managed to embed a machineryfor managing real wages within a democraticpolitical framework. The ones laggingbehind are countrieswhere labor market relationsdo not lend themselvesto the equitablesharing of costs. One view is that democraticsocieties that have experienceda long spell of stability,provide rertilesoil for preda4&ry interestgroups bent upon enlargingtheir share of the economicpie.- Unlessthe politicalsystem superimposes institutions that force these groups to negotiate and compromise, an anarchic situation develops which retards adjustmentto externalshocks. Countriessuch as Sweden,Switzerland, West Germanyand Austriahave been more or less successfulin keepingthis form of democraticsclerosis at bay by adoptingcentralized modes of wage and price settingthat force the major e99zbomicactors to confrontthe nationalconse- quencesof individualactions.47 In Sweden,the labor organization representinghalf the work force negotiatesthe pace settingwage bargains with the employersorganization under the watchfuleye of the governmentin an

40/ See MancurOlson, The Rise and Declineof Nations: EconomicGrowth, Stagflationand SocialRigidities, Yale UniversityPress, 1982; and Dennis C. Mueller, The Political Economy of Growth, Yale University Press,1983.

41/ A recent survey on industrial disputes shows that these countries have the lowestnumber of workingdays lost throughstrikes per 1,000 employees in the industrializedcountries. See, The Economist,2-8 August 1986, p. 81. - 77 - institutionalsetting that has deep roots and firm politicalacceptance. A tripartitecommittee composed of representativesfrom the Government,the unionsand the employersinfluences prices and wages in Austria. And so on.

7.03 Conditionsdiffer in the East Asian NICs, but marketmediated or state enforcedwage flexibilityis the sine qua non of competitivestrength. Hong Kong is at one extreme. Here the fluidityof the labor market is main- tainedby the relativeweakness of trade unions,ethnic diversity that arrests the formingof wage coalitions,manufacturing foundations built off small firms that rely on family,kin and clan members for their labor and the pres- sure on wages exertedby a reservepool of workers in the informalsector. At the other extremelies Singapore,where the Governmentexercises extensive controlover wages. Until the late seventies,the increasein labor costs was regulatedso as to attractassembly ty. industries.Thereafter, the state attemptedto push for high value added, technologicallysophisticated industries by drivingup wage rates. Recently,a severeslump in economicactivity has forcedthe Governmentof Prime MinisterLee to squeezewages and thereby restoreSingapore's standing in certaintraditional lines of manufacturing.

7.04 In betweenthese two poles lies Korea,where a homogeneouswork force in an integratedmarket is sensitiveto labor demandpressures. How- ever, these can be effectivelydampened by governmentsuasion that capitalizes on the weaknessof trade unions and the widespreadconsensus on the key role of labor in buildingKorea's trading position and cushioningthe country againstswings in the internationalbusiness cycle.

7.05 While political-institutionalfactors might be the centraldetermin- ants of wage flexibilityand marketefficiency, in many countriesvocational and technicaltraining programs on a very wide scale have greatlyfacilitated the mobilityof workersapart from contributingto their productivity.As a consequence,new industriesfind it easier to take-off,older ones to hold the ground in overseasmarkets.

Unionsand Wages in Fiji

7.06 Fiji'slabor market has many of the characteristicsthat a small economyneeds in order to respondflexibly to externaldevelopment and launch export-orientedmanufacturing industries. The populationof workingage is largeand growing. It numbered420,000 in 1985 with the rate of increaseover the next five years expectedto be about 2.41 p.a. (see Table 7.1).

Table 7.1: LABOR FORCE GROWTHRATE

1970-75 1975-80 1980-85 1985-90

Labor force growth rate 3.1 3.0 2.1 2.4

Source: 1970-75: Work and Incomefor the People of Fiji -- A Strategyfor More than Just Survival,The Final Report of the Fiji Employmentand DevelopmentMission, Sussex University, November 1984. 1985-90: DP9. - 78 -

Currently,participation rates are 86X for men, 212 for women. Wage and salaryearners make up 352 of the laborforce and only 39% of workershave paid employment. The balanceare self-employedor do not have jobs. Thus the potentialsupply elasticity of labor is largeeven allowingfor the resistance that peoplefrom the rural areas,especially women, might have to seek industrial employment. Further,a high percentageof new entrantshave primaryand secondaryeducation. These supply conditionswill help reducewage pressures and strengthenthe forcesmaking for rapidadjustment.

7.07 Fiji's Indiansugar workers unionized in 1921. Cane farmersfol- lowed suit in 1943 with the Fiji Kisan Sangh that became the Federationof Cane Growersin 1970 and emergedas the NationalFarmers Union eight years later. The hub of the union movementis the FTUC, set up in the early fifties,which servesas the umbrellafor 33 of the country's40 registered labororganizations. It has maturedunder the Government'stutelage, since the statehas long believedthat a centralizedorganization would inducethe unions to act more responsibly. A TripartiteForum chairedby the Prime Ministerwas createdin 1976 bringingtogether employers, unions and the Government. It is designedto regulatewage contractsby referenceto the CPI and businesscircumstances. However, the Governmentalso retainsconsiderable powersof direct interventionin the labormarket and these were utilizedto decisiveeffect in November1984 when a freezewas imposedon wages, rents and dividendsto nip an incipientwage price spiralled by the Nicoll and Hurst awards to public sectoremployees.

7.08 By and large the manufacturingsector has not been troubledby risingwage costs. In the criticalsugarcane sector, for instance,real wages have declinedat an averagerate of 0.22 p.a. during 1980-85. For all industries,real wages fell 0.2% p.a. between1975 and 1985,although they grew by an unsustainable7.7% p.a. over the preceding5 years (see Table 7.2).

Table 7.2: CHANGESIN MEAN HOURLYWAGE RATES (% changep.a.)

Agriculture Manufacturing All industries Nominal Real /a Nominal Real/a Nominal Real/a CPI

1970-75 20.3 8.6 19.4 7.8 19.3 7.7 10.8 1975-80 8.4 -0.1 9.0 0.0 9.2 0.0 9.3 1980-85 3.0 -4.0 4.7 -0.2 5.0 -3.8 6.9

/a Deflatedby the annual change in the CPI.

Source: Ministryof Finance

But labor costs remaina frequentsource of concernin the formalurban labor market,particularly in the unionizedsegment which accountsfor some 35,000 - 79 - workers. Given the levels of productivitymany employersbelieve that wages are too high and rule out the possibilityof outwardlooking manufacturing development. Undoubtedly,the earningsof the Fijianlabor elite are close to those of industrialworkers in the East Asian NICs, but remunerationof employeesin the nonunionizedand the informalsectors are much lower and clearlyreflect market realities. It also makes littlesense to use cost structureand productivitylevels of the existingrelatively inefficient import-substitutingindustries as the basis for judgingfuture export potential. At these and much higherwage rates,other nationsare successfully producingand tradinga wide range of light manufactures.If a problem exists,the fault may be more in the productionsystem engendered by incentive policiesthan the level of wages. By themselves,simple intercountry comparisonsof hourlywages are almostmeaningless.

7.09 In many respects,there are groundsfor optimismregarding the futurecourse of wages in Fiji and labor market developments.In general, educatedmanpower will be availablein abundancefor manufacturingindustry. The recent trendsin nonunicnizedand informalsector wages (see Tables 7.3 and 7.4) suggestthat the slack in the labor marketwill moderatethe wage demandsof workerswhether they are organizedor not. Union membershiphas alreadybegun to decline. An institutionalframework -- the TripartiteForum -- alreadyexists to centralizewage negotiations,marshal a consensusand mobilizepolitical sentiment behind rational economic policy. During its ten years the Forum has had a hand in containingwages, and besides,the Govern- ment is willingto interveneagainst an inflationarythreat.

Table 7.3: AVERAGEWAGES IN NON-UNIONGARMENT FACTORIES, 1970-83

Suva Lautoka Avg. wage No. of Avg. wage No. of Year (in F$) readings (in F$) readings

1970 0.16 102 1975 n.a. 19 - 1980 0.41 65 0.36 44 1981 0.57 77 0.44 68 1982 0.54 194 0.48 87 1983 0.63 183 0.51 95

Sources Ministryof Employmentand IndustrialRelations, Factory Inspectors Reports. - 80 -

Table 7.4: AVERAGEWAGE IN NON-UNIONFURNITURE FACTORIES, 1970-83

Unskilled Skilled/carpenters Averagewages No. of Averagewages No. of Year Mean Median readings Mean Median readings

1970 16 18 17 0.27 0.30 26 1975 34 38 3 0.55 0.50 30 1980 58 65 7 0.76 0.76 69 1981 58 65 21 0.81 0.80 58 1982 63 60 19 0.96 0.98 58 1983 63 65 14 1.10 1.10 44

Source: Ministryof Employmentand IndustrialRelations, Factory Inspectors Reports.

7.10 There are some powerfuland well organizedinterest groups in the labor market such as the cane cuttersand the public sectoremployers but they are not unawarethat economicruin could resultfrom a battle for higher wages. Fijianunions are by no means docilebut neitherhave they been guilty of the aggressivebehavior which has, on occasion,pushed European and Latin countriesto the brink of a crisis. Finally,as in Hong Kong and Malaysia, ethnicdivisions are likely to retardthe formationof a powerfullabor bloc; the structureof manufacturingemployment with small firms very much at the forefrontis conduciveto flexibility;and labor militancywill also continue to be tincturedby familyand kin ties that bind workersin many of the small enterprisesto their employers.

Skills

7.11 Industrializationneed not be hamperedby wage costs or labor supply--recentexperience indicates that producersshould have no difficulty in persuadingyoung femalesto work in textileand electronicsfactories--but the availabilityof certainindustrial skills, might constrainthe manufacture of more sophisticateditems. Currentlythere is no shortageof traditional craft or basic mechanicalskills, in fact, the slowingof growth has createda surplusof vocationalschool graduates. Higherorder skillsremain scarceand the situationis made no easier by the absenceof formalapprenticeship programs and the shortcomingsof the FNC to whom employerspay a compulsory - 81 - trainingfee.21 Companieshave to make do with on-the-jobtraining, or the assistancegiven by foreignequipment suppliers. In rare cases they are preparedto incur the high costsof overseastraining for a few staff.

7.12 The industrialstrategy discussed in this reportcannot be implemen- ted withoutan abundanceof manufacturing-specifichuman capital. Many other countriesin a similarpredicament as Fiji are wrestlingwith this constraint, Singaporebeing one of the most enterprising.It establisheda first genera- tion of trainingcenters with the help of MNCs in the early seventies. Ori- ginally,all the centersoperated on a cost-sharingbasis with the government bearingthe capitalcost and half the recurrentexpenditures. Since 1972, the state has assumedall costs,participating firms supplyingonly the software. In the early eighties,Singapore's Economic Development Board sought the assistanceof the MNC and governmentsof industrialcountries to establish a new generationof trainingcenters to prepareGCE "A" levelholders for modern technologiesand productlines. They are the German-Singapore Instituteof ProductionTechnology, the Japan-SingaporeInstitute of Software Technologyand the French-SingaporeInstitute of Electro-Technology.These institutesoffer diploma courses in controlengineering, instrumentation, mizroprocessorapplications, CAD/CAM systems, cutting processes, desi*n and productionprocesses, industrial engine.riag and computersoftware tE:hnol- ogy. The EDB also runs three specialis.-.units for industrialautomation trainingin partnershipwith the AmericanComputervision corporation and traininginstitutes in Swedenand Japan.

7.13 Supplementingthe work of the institutesin Singaporeis the Skills DevelopmentFund (SDF),financed by a 2X (of payroll)levy on employerswhich can be tapped by firms to upgradeskills and acquireadvanced equipment, througha number of schemes--TrainingGrant, Interest C-ant for Mechanization and DevelopmentConsultancy. There is also a schemefor nurturingskills neededby microelectronicsand informationindustries. Aside from the SDF, employersalso have recourseto the NationalProductivity Board which offers courseson measuresfor enhan4 ing productivityand improvinglabor management relations,3/,well as two centersfor computerstudy fundedby the government.-

7.14 Singapore'straining infrastructure has been piecedtogether over a decade,a periodmarked by heavy foreigninvestment in the nation'smanufac- turingindustry. Fundingfrom MNC and technicalexpertise was readily

42/ The Fiji NationalTraining Council (FNTC) operates a programof skill trainingfor employeesof manufacturingconcerns in Fiji. The FNTC's operationsare financedby a one percentlevy, which is compulsory,paid by employerson theirgross payroll. Employerscan, theoretically,claim rebatesup to 100% of these leviesto pay for the trainingof employees.

43 See I. Islam and C. Kirkpatrick,Export-led Development, labor market conditionsand the distributionof income: The case of Singapore, CambridgeJournal of Economics,Vol. 10, No. 2, June 1986,pp. 115-117. - 82 - available. Circumstancesin Fiji are very differentand a programon such a scalemay be impossibleto duplicateover the medium term. But an early start at establishingcenters for trainingin designand modernengineering and automationtecbnology, possibly through support from foreigngovernments and overseastraining institutes, should be considered. This could be one of the principalinitiatives in a largerprogram of industrialskill developmentthat also breatheslife into the existingvocational training efforts and the employerfunded FmTC scheme.

7.15 Behind East Asian economicsuccesses is a long historyof improvement in the qualityof manpower. As industrialcompetition moves to a higher technologicalplane, skillenhancing policies can only grow in importance. It is a lessonFiji must take to heart. - 83 -

VIII: THE MICROECONOMICSOF A NICHE FILLINGEXPORT STRATEGY

An InformationSystem

8.01 Traditionalapproaches to industrializationthat dependupon fiscal, monetaryand trade relatedinstruments have been exhaustivelycodified and widely applied. There are no secretrecipes and most of the surpriseshave been explainedaway. Many countriesare tryingvariants of the standard methodsbut those that are succeedinghave generallygone beyonda mechanical applicationof tax and tariff rules. These economieshave evolvedan institu- tionalarchitecture that makes innovativeuse of informationon marketsand technology. The East Asian NICs have undoubtedlyderived much profitfrom economictextbooks but the capstoneof their strategyhas been a systemof collectinginformation and techniquesfor bendingit to their developmental purpose.

8.02 Market informationis availablein vast quantitiesfrom trade publications,from wire services,from buyerswho flock to East Asia$ from importersand from governmentagencies, to name just a few. It can go unused withouta systematicprocedure for analyzing,disseminating and acting. Japan and Korea have acquiredtheir tradingreputation because they realizedthat in today'smarkets being better informedabout export possibilitiesthan one's competitorsis winninghalf the battle. Korea'sinformation system, for instance,has three tiers: Koreanembassies, the Trade PromotionAssociation which maintaina dense networkof officesoverseas and representativesof Korean companies,strategically located in major cities;excellent telex and telephonefacilities; and companysalesmen who frequentlytour foreignmarkets so as to m#.q contactwith currentand potentialbuyers, comprise the outer- most tier.' Buyers visitingKorea are seen as not just a sourceof orders but also informationon productdesign, styling, packaging and market trends. The secondtier, therefore,is the infrastructureof transportfacilities,hotels, trade centersand tireless"facilitators" who meet, court and guide the buyersto win ordersand establisha longer term relationshipwith an importinghouse or a retailchain which could be a continuingsource of business,technical knowledge and market intelligence.

8.03 The third tier covers the mechanismsfor collatingand disseminating the informationpouring in from the networkoverseas. This is done by a varietyof bodieswho rely on publications,meetings and personalcontacts. The Korea TradersAssociation issues a Trade News Service,compiles and pub- lishesmonthly trade statisticsand periodicallyreleases material on export- importprocedures. Many of the producer'sassociations have their own publi- cationsand all are active oeepingtheir membersclosely informed about market trends,moves made by competitors,rules affectingKorean exportsas well as

44/ For a more detaileddiscussion of the Korean system,see: Korea's CompetitiveEdge - Managingthe Entry into World Markets,Yung Whee Rhee, Bruce Ross - Larson,and Garry Pursell,the Johns HopkinsUniversity Press,Baltimore and London,1984. - 84 - other aids for producerstrying to navigatein choppy,obstacle-ridden, alien waters.

8.04 The systemnot only gathers information,it also transmitsto buyers outsideof Korea,by way of KOTRA publicationssuch as the OverseasMarket News and Korea Trade,a catalogof Korean exportproducts. Exhibitionsare anotherfavorite device. There are also, advertisingcampaigns and promo- tionalver.tures with the large tradingcompanies taking the lead supportedby commercialstaff at overseasmissions.

8.05 Informationon productdemand is crucialfor producersas is the network'sability to make known the range of items,Korean manufacturers are in a positionto supply. But the industrialsector's international competi- tivenessdepends on a steadyflow of informationon productand processtech- nologyand the numerousparameters relating to design,quality, finish and packagingthat determinea productsacceptability. No one conduitis ade- quate. Many have to be tappedbut first,manufacturing concerns must have the technicalpersonnel to apprehendand "unbundle"foreign technology. Technical informationis worthlessunless firms are equippedto absorb it. At the earlierstages when light consumergoods comprisedthe bulk of exports, foreignbuyers and companiessupplying equipment can light the way. Later as producergoods grow in importanceoverseas visits by companymanagers and technicians,licensing arrangements and joint venturescan be more effective. No rule emergesfrom the Korean and South-EastAsian experience. The industrymust find the mode most suitedto its circumstances.The point is that there are multiplepossibilities and it is vital for manufacturersto be listeningat a whole range of synapses.

8.06 Korea,Singapore, Hong Kong, and Japan have each taken a positiona the centerof a web that draws marketand technologicalintelligence from around the world. Each has masteredthe art of weavingan industrialpolicy which embracesthis materialand of allowinga few key ecciomicinstitutions to coordinatetheir activitiesso thAL manufacturinghas the environment,the resourcesand the politicalsupport needed to ben.fitfrom the information. In Japan as well as severalof the East Asian NICs, the ofticialbureaucracy fully appreciatesthat its purposeis to create the conditionsthat will promotetrade. The bureaucraticcommitment to exportgoals is fundamental becauseit dissolvesmuch of the red tape that is the bane of exportersin countrieswhere officialdomis contentto administet nd is reluctantto prune the thicketof routinesthat hobbledevelopment plans. From observingEast Asian governmentsin action,a number of maxims can be distilled. An economic bureaucracythat has detailedand timely informationon such variablesas exportsand productionis best equippedto launchcountermeasures that will removebottlenecks and maintainthe industrialtempo. A comprehensivedata- gatheringapparatus that operateswith the shortestpossible lag seems a must for effectivepolicymaking.

8.07 Of equal significanceis the willingnessto eschewa rule bound approachto fast moving sieuations.Achieving results rather than safeguard- ing proceduresis the most efficientprinciple when the goal is growth through exports. An effectiveinformation system can providea policymakerwith the confidenceto act becausehe can monitor the consequencesand make adjustments - 85 - if the need arises. Bureaucraticstasis that feeds on long informationlags can be avoided. Lastly,the informationnetwork can be used to cementa part- nershipbetween the government,banks, manufacturing firms and other service agencies,all of whom contributeto the successof an exportstrategy. But the messageout of East Asia is that the allianceis instigatedby the Governmentand it is reliableinformation and bureaucraticflexibility in respondingto it which controlsthe risk and raises the rewardsof concerted action.

Copingwith NTBs

8.08 AlthoughFiji is years away from becominga large exporterof items now runningafoul of rontradebarriers, advance preparation to cope with the politicalfallout from trade is desirable. BecauseMauritius was able to capturea sizeableshare of the knitwearmarker in NorthernEurope and the US, it has to contendwith quotas. The steadymarch of East Asian exports in the teeth of NTBs provesthat, intelligentlyhandled, the NTBs need not dampen the growthof exportearnings. Trade barriersare far from impenetrable.They generallydisplace opportunities and do not eliminatethem. A thorough familiaritywith the fine print underlyingNTBs and the capacityto manipulate legal and diplomaticinstruments, can allow exportersto maintaintheir stride. Further,a comprehensiveunderstanding of both the legal environment and the manner in which protectionistmoves are implementedmakes it possible for a developingcountry to negotiatean accord that is impermeableand restrictivein form, defusingpolitical problems but porousand unrestrictive in substance. The wordingused, how items are classified,and flexibility provisionsare as criticalas base levelsor growthrates. In textiles,for instance,exporters have derivedmuch mileagefrom categoryswitching, borrowing againstfuture quotas and carryoverprovisions. East Asian tradershave, in the past, made full use of the ambiguityconcerning the classificationof shoes (rubberor nonrubber)and fabric (syntheticor cotton)of transshipment throughquota free ports;and the exportof subassembliesor half finished goods when the finishedarticle come under a quota.

N1iches

8.09 The tradingmilieu of the eightiesis doubly competitivefor being an intenselypolitical one. To acquireand retainmarkets, a constantbattle must be waged to keep the econonmicedge and a measureof politicaladvan- tage. Informationalone will not lead to a marketableproduct, but it can decidewhether and in what volume it is sold.

8.10 Economictheory, with some help from empiricalfindings, can be a usefulguide in makinggeneral policy. Severalcompeting architectural plans for a healthybusiness environment are at hand and latelythe agony of choos- ing has been eased by the coalescenceof professionalopinion behind the market-directedapproach. Bit in arranginga suitable"environment" the ques- tion of what shouldbe producedand how exportmarkets penetrated is left for supposedlyknowledgeable entrepreneurs to ponder. This divisionof labor betweeneconomists and policymakers on the one hand and businessmenon the other, is a good one. Koreaneconomic planners, for example,are in no position to effectivelymeddle with the micro-structureof textileproduction and - 86 -

trade. They can try and regulatethe climatewhich will allow this micro- structureto thrive* In recentyears many formerlydirigiste economies have found it expedientto allowmicro investmentdecisions to devolveinto entrepreneurialhands.

8.11 Detailedcentralized guidance regarding the subsectoralallocation of capital,product mix, technologyand marketorientation invitd gross inef- ficiency. Many decisionsare best handled by the marketonce the market has maturedto the point where it efficientlymanages the flow of information. Beforethat stage is reached,priming the market with informationof a strate- gic natureas was done (and to an extent still is) in Japan,Korea and Signaporehas its merits. Over the vext few years,Fijian privateindustry will be emergingfrom its protectivecocoon. It will be searchingfor ideas and market intelligence.As only the governmentcan commandthe resourcesand the multiplicityof internationalcontacts, industrial policy will for a time have to extendbeyond environment bailding to the provisionof guidanceand information.Hence a systemthat will generatereliable domestic and inter- nationaldata with the least delay is a necessity. Becausethere are scale economies,only the Governmentcan establishit and since the processwill take time, the projectdeserves priority.

8.12 Studiescommissioned by the Governmentand businessorganizations have uncoveredsome potentialproduct markets and identifiedways of enter- ing. Much more will have to be done. However,the work alreadydone in Fiji and elsewhereis revealingon how high value added industriessuitable for small economiescan emergeas exporters. Five encapsulatedcase studiespre- sentedbelow attempt to catch a glimpseof certaingeneral principles that might be at work. These are not necessarilythe industries,Fiji might choose to embracebut they come from a class that deservesattention. The five are spirits,fashion garments, watches, light machineryand tropicalfruit.

Spirits

8.13 Alcoholconsumption in the developedcountries is going througha periodof turbulence. Strongdrinks are on the retreatas health conscious- ness, after takingAmerica by storm, invadesEurope and the force of habit is being displacedby the demandfor taste. In Franceand the US, alcoholcon- sumptionper capita is falling,and there is a definitepreference amongst an aging Westernpopulation for less potentdrinks. A numberof interesting trends in the consumptionand marketingof spiritscan be identified. Whiskey blendsare fallingfrom grace. Light whiskeysare increasingtheir share of the market,but the real winnersare the mixablebeverages such as gin, rum and vodka,as well as cordialsand liquors. Sweeterpotions have also become fashionable.This extendseven to red wine: the less tannic,bitter and acidicit is, the better it sells. Successfulspirits are those that by virtueof mixabilityand lightnesshave createdtheir own niches -- vodka, white rum and tequilafall in this category(see Table 8.1). - 87 -

Table 8.1: TOTALCONSUMPTION OF SELECTEDSPIRITS, 1982 (millioncases)

Scotch Other Cognacand Rum Gin Vodka whiskey whiskies Armagnac

51.8 43.2 59.4 65.0 112.4 8.9

Source: InternationalWine and SpiritRecord of 35 major marketsexcluding COMECONcountries.

8.14 A secondtrend is the rising demandfor premiumbrands, in the case of whiskey,particularly single malts. The best sellingtins and vodkasare those with a reputationfor quality. Irish whisky has stageda comebackby concentratingon the premiumend of the marketwhere it competesagainst the best gins, vodkasand cognacs.

8.15 A third trend concernsmarketing. Consumersare attractedby the shapeand characteristicsof the container. A plasticlined bag has gained wide acceptanceIn the UK, so has wine in cans. Wine sold in shapelyglass jars has a strongappeal in the US. A Japaneseproducer has found that trendy names and bottlesart appealingto the young. The marketingof spiritsis increasinglycontrolled by the industry'sgiants with grocerychains becoming influentialin the retail sphere. In Japan,a powerfulsales networkand tied bars offer intensecompetition to a newcomerattempting to penetratethe market independentlyof a local partner.

8.16 A light rum is the singlemost suscessfulspirit in the world (see Table 8.2). And a varietyof Caribbeanrums have acquireda loyal following. Top-fermented,heavy-bodied and stronglyscented rums from Jamaicahold a segmentof the market,as do the bottom-fermented,light and weaklyaromatic rums from PuertoRico. The Mount Bay distilleryof Barbadosis famousfor its redistilledrum which is blendedwith caramel,almond essenceand angostura. None of these Caribbeanrums commandthe market which Bacardiwhite does, nor have they managedto launchan immenselypopular mixed drink such as the Margarita(which accounts for the strongsales of Mexicantequila in the US) but they earn respectableamounts of foreignexchange. - 88 -

Table 8.2: CONSUMPTIONOF RUIMWORLDWIDE (1982)/a

Country Millioncises

Britain 4 France 5 Spain 5 Mexico 7 Canada 7 Venezuela 7 Colombia 13 West Germany 13 US 26 Other 13

/a Bacardiholds 301 of the world market. The remainderis with other brands.

Source% See Table 8.1.

8.17 Fiji exportsits own rum on a very small scale (1,000cases to the US in 1985). It is worth examiningthe p.ospectsfor largersales. The above discussionoffers the followinghints:

(a) A light,premium rum less than 90 proof will be most acceptablein the US market,that is by far the biggestand the worthiesttarget (see Table 3.3).

(b) Fresh cane molassesare the best feedstockas they have low viscosity,high total sugar,nitrogen and phosphoricacid and low ash.

(c) Rum distillationshould be continuous,the productaged four to five years to increaseacids-to-esters ratio and dilutedwith pure, deionizedwater.

(d) The blend to be soughtshould have a distinctbut subtleflavor. A marketingcampaign that associatesthis with a ne-wmixed drink improveslong-term sales if the drink gains acceptability.

(e) The shape of the containerdeserves careful attention.

(f) Market tests followedby an advertisingcampaign that toucheson the rum's exotic South Pacificorigin would be needed to launchthe product. It might be aimed at the well-to-dowho are preparedto experiment. - 89 -

(g) Since marketingcould be expensive,a foreignco'laborator, possibly one not in the spiritsbusiness, might be a valuableasset.

Table 8.3: CONSUMPTIONOF DISTILLEDSPIRITS IN THE US (salesin '000 cases)

1968 1978 1983

Bacardi 2,100 6,200 7,673 emirnoff 3,200 6,150 5,700 Seven Crown 7,850 6,200 4,900 CanadianMist - 2,300 3,350 Jim Beam 2,525 2,750 3,2C0 Jack Daniel's - 1,875 3,175 Seagram'sVO 3,475 3,875 3,050 CanadianClub 3,325 3,675 2,850 Popov 575 2,250 2,600 WindsorSupreme - 2,475 2,600 Seagram'sgin 1,050 2,350 2,525 Dewar's 1,050 2,200 2,400 Gordon'sgin 2,550 2,600 2,250 Gilbey'sgin 2,025 2,225 2,000 J&B Rare 2,075 2,650 1,825 Cutty Sark 2,150 2,150 1,500 CalvertExtra Blend 2,175 2,150 1,500 Old Crow 2,625 1,075 850 Imperial 2,275 1,175 625

Source: BusinessWeek.

Watches

8.18 Switzerlandinvented the electronicwatch but the East Asians were the ones who broughtit into mass production.Over a third of the watches manufacturedworldwide in 1985 were Japanese. Other Asian countriesaccounted for 50%; Switzerland'sshare iwas101, but becausethe countrystill has a virtualmonopoly over the supetriorbrands, the Swiss took 45Z of sales value. Watch making is a footlooseindustry no longer the preserveeither of countrieswith a traditionof exact craftsmanshipor economiesthat boast of a sophisticatedscientific establishment. Designing and assemblingthe basic quartzmechanism has ceasedto be a major engineeringfeat and productionof the entirewatch can be managedby an industrialsector of modestcapabili- ties. It is also possibleto importthe electronicmechanism from Switzer- land, Japan or some of the East Asian countriesand put it insidea locally designedcase. - 90 -

8.19 As competitionis intense,there is scant profitin assemblinglow pricedwatches using an importedmechanism. Even the manufactureof the entirepiece, includtngthe quartz innards,generates low returnsif it is for the mass markets. However,the remarkablesuccess of the Swiss Swatch and its imitators,Tiq, N-watchand Vidio-Cliphave proved that a good idea, an innovativedesign backed by a skillfulmarketing campaign can derivemileage even out of an establishedslow moving product.

8.20 The Swatch,introduced ir 1983, is an inexpensiverestyled version of the ConcordDelirium, a thin anc costlywatch first sold in 1979. It is availablein a varietyof startlingdesigns accented by the use of bright color. Swatchhas revolutionizedthe watch businessin two respects. It has persuadedthe targetedclientele of teenagersand young adultsto treat the watch as a fashionaccessory to be blendedwith other clothing. The message pressedhome is that irndividualsshould own an assortmentof watches to go with their wardrobe. People own 2.5 watcheson the avdragein the ind'lstrial countries. And until recently,the industryworked on the assumptionthat an individualpurchased a watch once a decade. Becauseof innovationssuch as the Swatchthe patternis changing. During the mid-seventies240 watcheswere sold in the US per 1,000 inhabitants;425 in 1985. In ju t two years Swatch sales alone have risen to $150 million.

8.21 Offeringpeople an affordable,attractively designed product and inducingthem to conceiveof watchesas an item much like shoes was a design and marketingbreakthrough. Swateh also ensuredthat the productwas reli- able, well-finished,waterproof and easy to producethrough innovations in other areas. There are three steps in the manufactureof a watch: the case; the mechanism;and assembly. Swatchbrought these togetherinto a single extensivelyautomated operation where the final stage uses a laser to attach the crystalto the plasticcase. Qualityand productioncosts have benefitted from a design that minimizesthe number of parts required. Purchasingparts from just a few suppliersreduces the expenseof certifyingand auditing as.ociatedmanufacturers. And a less complexintegrated system has allowed Swatchto market a productthat assuresquality at low prices.

8.22 The saga of Swatch containssome rules that are likelyto be germane for niche huntingFijian manufacturers. A searchfor nichesmust combinefact findingwith a dash of social forecasting;how is the market evolving? What changesin consumptionhabits might peoplebe willingto entertain? Further, a successfulproduct capitalizes on a meaningfuldistinction and brings togetherperformance, quality and durabilityin a handsomepackage. From the start it is designedto be visuallyapp,ealing and to be simpleto produce. Finally,an item such as the Swatchmust be made with a certaintype of cus- tomer in mind and sold in pickedgeographical locations through a distribution networkthat reinforcesthe productsother attributeswith a sensitivemarket- ing drive.

Machinery

8.23 Exportsof machineryand machineparts form the East Asian countries are expandingas they attemptto completethe transitionto industrialpro- ducts less suseptibleto market saturationand trade restrictions.Machinery - 91 - producersare thrivingnot just in Japan and Korea but even In the smaller economiessuch as Hong Kong and Singapore. Many have be,;unto acquireprofi- ciency in manufacturingnumerically controlled tools and automatedequip- ment. The path taken by machineryproducing firms providesvaluable lessons tor a newcomerlike Fiji.

8.24 Machine-toolproducers in the industrializingEast Asian countries tend to be small,employing less than twentyworkers. Their owners usually acquiredexperience and a littlecapital by working in a metal workingor machineryfirm which is why the existenceof a machinerepair infrastructure associatedwith the farmingor fisheriessectors is of such importance. Most firms began life as repairestablishments, edged into parts productionand, after acquiringsome capabilityin the areas of design and fabrication, startedcopying imported machines. Eventuallythey gainedthe confidenceand the skill to adapt and modify foreignequipment, attaining thereby a distinct identityas makers of genuinelyhome-grown machinery.

8.25 In the early stages,skills were acquiredthrough learning on the job ratherthan formal trainingbut by the time firms were producinglocally designedmachines they had hired university-trainedengineers. The abilityto importhigh-quality castings, forgings, motors and hydraulicsfrom overseas was a signaladvantage and many firms profitedfrom the advicegiven by componentsuppliers as well as local buyers. Entry into the businessoften had to do with the desireof domesticbusinesses to substitutecomplex and expensiveimported equipment with local items that could be servicedand repairedmore quickly. If a local firm were able to offer a simpler.competitivelypriced variant of standardmachine tool and guaranteea ready supply of parts and service,he had the beginningsof a market. Designs,tried, perfectedand maturedthrough interaction with domesticbuyers, could then be testedin the foreignmarket.

8.26 Export successdepended very much on the producer'sinnovative- ness. Simplerand rooustmachine tools suited for conditionsin developing countries,sold at low prices,paved the way for entry into the world market. Once sellersbroadened their range to includemore sophisticateditems, parts supply and repair services also gained in importance. In several of the East Asian economiesgovernment sponsored extension and trainingcourses aided firms in surmountingdesign and manufacturingbottlenecks while technical institutesas in Singapore,allowed firms to upgradeskills. Hong Kong is somethingof an exception,but the other NICs have all supportedthe development of the machinerysubsector through the provisionof venturecapital and specialfinancing schemes. That machineryproducers in Hong Kong have lagged behindas a consequenceis apparentfrom their continuedemphasis on the simplermachine tools, low expenditureon R&D and limitedscale of operations.

8.27 The East Asian NICs appearto be moving into roboticand compu- terizedmachine tools as well as high-speedmilling and grindingmachines. In a few years there may be rich opportunitiesin the lower end of the machine tool market for new entrantsto reap. Becauseit takes time to createa base for the manufactureof machinetools, Fiji must begin exploitingthe domestic linkagesfrom sugar cane, fish, wood and constructionrelated activities to form the nucleusof a machineand metal workingindustry which, in due course, could be involvedin the vigourousexport market. - 92 -

FashionGarments

8.28 The most consistentlyprofitable segment of the internaticnal clothingbusiness is fashiongarments. Those countriessuch as West Germany and Italywhich are alreadyat the top are battlingto retaintheir position. Enlargingmarket share is the goal of every other producer. Among the industrializingEast Asian countries,Hong Kong has made the deepestinroads and has the strongestreputation. What is the profileof a successfulfashion garmentsexporter?

8.29 To sell expensiveclothing which conformsto the latest fashionsin highlydiscriminatit.g Western centers, a producermust be preparedto offer the latestdesigns in the most popularcolors. Beyondthat he must be ready with new offeringsas seasonschange. Hence designconsciousness, the ability to producein small batcheswith shortlead times and to meet an exacting scheduleare some of the desirableattributes.

8.30 Design capabilityflows from severalsources. Obtainingspecifica- tions,ready to use patternsand molds from buyerspresents the fewestdiffi- culties. It is also possibleto acquirelicenses or even commissionsome of the famousfreelance designers to preparea line of clothing. For instance, Mary Quantnow runs a studiowhich does work on contract;Karl Lagerfeldis anotherwell-known name in this field. Assidiousstudy of the trademagazines togetherwith frequenttrips to the main centersof fashionis anotheroption for a producerwishing to remainau courantin a fluidmarket. The riskiest but in the long run most effectiveapproach is for a sellerto designhis own clothing. In developingcountries such design skillsare scarcefor a variety of reasons. Fashionconsciousness is weak and usuallyat a culturaltangent to the West; often a slowlyeroding conservatism dulls the urge to experiment and is resistantto abruptdepartures in style;the wealthymay be reluctant to patroniselocal designers and have no stomachfor largeexpenditures on fashiongarments with an ephemerallife span; and theremay not be any design schoolsin a positionto impartthe requisiteskills. Hong Kong has been able to escapefrom this mold becausethe culturalcircumstances do not inhibit experimentationin dress--theisland's youth are notoriuslysusceptible to the latestWestern fads--and the growingmonied classes show a taste for ostenta- tious livingthat can give the fashionindustry the necessaryeconomic impulse.

8.31 Aside from design,quality of product,neatness in Wish and punctualdelivery are what the industryexpects of producers.- West Germany'sstanding comes not from leadershipin designbut from a masteryof the logisticsthat ensuresthe timelydelivery of qualityproducts. Most developingcountries fall short of meetingthese requirementsbut the leaders in the fieldhave shownhow the problemsmight be tackled. Korea, Singapore and Hong Kong still rely heavilyon buyers to provideexact specificationst assistwith the qualitycontrol and provideinformation on techno;.6y. By

45/ For furtherdetails, see David Morawetz,Why the Emperor'sClothes are Not Made In Co'!mbia,Oxford UniversityPress, New York, 1980. - 93 - workingclosely with the purchasers,firms have been able to producegarments that are properlysized and labelled,strongly sewn, neatly finishedand can pass the test of color fastness. Importersas well as the firms producing machinerycan also keep the garmentmanufacturers abreast of developmentsin materials,material joining methods and automatedcutting anid sewing equipment that is reducinglabor needs while lendingprecision to manufacture. Being able to importthe best fabricsand selectthe unusualweaves, designs and colors is a tremendousadvantage for the East Asiansover Latin American producerswho must do battlewith an array of textileimport controls imposed by their governments.And a sophisticateddyeing industry is a great asset in a businesswhere the strengthand varietyof colorshas come to matter more and more. To cite one example,the Italiancompany Benetton, which ranks as the world'slargest producer of knitwear,has arrivedat its positionof eminenceby mass producingmedium pricedand very colorfulgarments for a targettedmarket consistingof young females. By speciallyprocessing its wool the companyhas achievedhigh standardsof quality. Subcontractinga large share of its productionto small workshopshas given Benettonflexi- bility. A processfor dyeing assembledgarments and a remarkablesystem of market feedbackensures a quick supplyof saleablegarments. AnotherBenetton innovationis the exclusiveknitwear stores operated through franchise which allows the companyto controlwindow displaysand arrangementof the merchandise,two marketingtechniques that have provenvery effectivein attractingbuyers of clothing.

8.32 Keepingclose tabs on the needs of buyerscalls for cheap and efficientcouaunication facilities; ensuring that high value fashiongarments can be deliveredon schedulerequires good air transportconnections; and the necessityof maintainingfrequent contact with importersmakes the cost of air travel for local sales personneland foreignvisitors an importantconcern in the garmentbusiness. East Asia is steadilyenlarging its competitivelead throughstrategic investment in transportand communication.It is something that other tradersmust match to a degree if they are not to be left behind.

8.33 These requirementswill not be easy for a beginnerlike Fiji to satisfyin the short-term. But entry into the fashiongarments market cannot be contemplatedunless the countryis preparedto emulatethe leadersin the field. Over a hundredsmall firms are producinggarments in Iiji and quite a numberhave been able to win export orders. A minor percentageof the orders are for qualityapparel, an encouragingdevelopment, but the high fashionend of the market still remainsbeyond the reach of even the most enterprising exporters. To concentratethe clothingsubsector's resources on producinglow priced,mass market items would be a poor strategy. Fijianfirms must move upscaleas rapidlyas possibleand the Governmentcan assistin a numberof ways.

(a) duty free access to qualityfabric and materialfrg?,the European countriesand Japan would enhancecompetitiveness;-

46/ The recentimposition of quota restrictionson knitwearfabric may interferewith the cost competitivenessof Fiji'sgarment industry. - 94 -

(b) the governmentmight take the lead in puttingtogether an apparatus for gatheringmarket intelligence,and providingbusinessmen with additionalincentives to traveloverseas in searchof orders;and

(c) a textileand clothingdesign centerand nationaldesign contests would be invaluablein groominSlocal talentand meetingone of the preconditionsfor entry into the fashiongarments market. In addi- oion, the governmentmight take a hand in arranginglicensing deals with foreignbuyers that deal in high fashionclothing.

Horticulture

8.34 Exportsof fruit and vegetables,except when they have been processedand packaged,fall outsidethe ambit of 'he manufacturingsector, but it is worth examiningtheir potentialas niche fillersfor two reasons. First,Fiji appearsto have a comparativeadvantage in producingso-called exotictropical fruit. Second,advances in technologyand marketinghave renderedcommercialized horticulture an activityakin to manufacturing.The horticulturetrade expandedrapidly during the late seventiesattracting much attentionworldwide. A change in Europeaneating habits away from bulky traditionalwinter vegetables high in carbohydratessuch as turnips,potatoes and cabbageand towardsfresh importedchoice items, for example,green beans, eggplant,zucchini and strawberries,was one factor. Anotherwas the vast increasein demandfrom the oil rich Middle East for a whole range of fresh fruit and vegetables. These two marketstook their place alongsidethe U.S. as major consumersof these items. In fact, the EEC is now the largestmarket for horticultureproducts absorbing 55% of total fresh fruit importsand 60Z of the worldwidesales of vegetables,while demandfrom the MiddleEast exceedsUS$l billion(see Table 8.4)

Table 8.4: PER CAPITACONSUMPTION OF FRUITSAND VEGETABLES (kg/person)

Fruits Vegetables 1975 1982 1975 1982

US 186 197 157 172 WesternEurope 156 158 170 181 EasternEurope 35 45 98 106 MiddleEast and Gulf States (SaudiArabia) 88 150 91 135

3ource: World Bank.

8.35 Since 1982, exportsfrom the rest of the world to the three principalconsuming areas have stabilized,creating conditions of chronic overcapacityand intensecompetition. Glasshouse grown tomatoesfrom the - 95 -

Netherlandscompete against Mediterranean produce. Cheap energyand subsidies have encouragedArab farmersto begin producingitems that were formerly imported. And improvedtransport connections have earnedChilean table grapes a share of Europeanand Americanmaekets, while Kenyan vegetablegrowers are able to competeagainst local as well as Mediterraneansuppliers for the much covetedEuropean market "windows." For many items,tomatoes being an example, there is little to attracta new exporter. But as far as tropicalfruit and vegetablesare concerned,the marketdoes not appear to be saturated. Productssuch as mangoes,bananas, okra, capsicumand auberginescan still coumandbrisk sales in the Mid-Eastand the EEC not to mentionthe US. Entry into the business,however, requires preparation and sound organization.

8.36 Experiencethat has been gained in producingand tradingbananas, pineapplesand kiwi fruit indicatesthat a deterAinedcampaign can catapult countrieslike Thailand,Philippines and New ZenAlandto the front ranks but the road to export successie a long and costlyone. Large-scalecommercial productionmanaged by skilledpersonnel is definitelyadvantageous from a financialangle becausethere are scale economiesand the applicationof technologyis facilitated. As the market for horticultureproducts has developed, it has adopted the standardsand conventionsapplied to manufactures.Qual- ity, shelf-life,standardization, ease of handling,uniform attributes, packaging, constantsupply, steady improvement--all of these are demandedby customersand retailers. This requirescontinuous research and its conscien- tious application. Since the mid-seventies,horticulture has becomea researchintensive activity. Many recentadvances in somaticembryogenisis, protoplastfusion and somaclonalvariation will only raise the technological stakesin this field giving large,well-managed financially strong agro- businessesa decisiveadvantage.

8.37 In todaysfast-paced markets where distributionand retail chains are scramblingto ensurean uninterruptedyear round supplyof a vast range of items,detailed daily informationon pricesand transactionsis a must. Per- haps more so than with manufactures,the exportof horticultureproducts requiresthat the sellerbecome a part of the farflungcommunications network which regulatesprices and the flow of commodities.

8.38 For many products,bananas being one of the foremostexamples, an elasticsupply of cheap labor and ample reservesof fertileland are needed to assurecompetitiveness and gain market share. Preparingbananas for shipment calls for destemming,removal of rejects,grading, washing, branding and packing in boxes. All of these are labor-intensiveoperations. The Philippines attractedthe notice of multinationalfruit companiesand becameone of the largestbanana producersbecause agricultutzl labor and land were plentifulin Mindanaoand elsewhere. Similarly,the mix of climate,soil, and labor togetherwith encouragementfrom the Board of Investment,have made Thailand the secondbiggest exporter of pineapplesin littleover a decade. Mexico, Belize,Haiti, Braziland South Africa have developeda lucrativetrade in mangoesto the US, partly on the strengthof elasticfactor supplies and partlythrough their abilityto developand market a productthat fits the US preferencefor grapefruit-sizedmangoes. - 96 -

8.39 Exporteisof fruitand vegetablesto the EEC generallysell on consignmentthrough their own agents locatedin wholesalemarkets or on direct contractto major buyers. In the U.S., there is, in addition,extensive MNC involvementin the transportand marketingof fruit. Entry into the Japan3se mar&et is often difficultwithout the intermediationof a tradinghouse with good distributiveconnections. The logisticsof the horticulturetrade are expensiveand complex. Withoutan efficient,financially well endowedorgani- zationthey cannotbe effectivelymanaged. Such organizationswere createdin Israelto market citrusfruit and in New Zealandto engineerthe entry of the fabl-dkiwi into the culinaryuniverse of the Americans,the Europeans,the Japaneseand the Arabs. Other countries,for example,Mexico, the Caribbean nations,Central American states and the Philippineshave also dependedupon the initiativeof MNCs.

8.40 Shipping,ripening, wholesaling and the assuranceof regular suppliesto dispersedretailing outlets necessitates an intricatescheduling of productionand transport. Introducingtropical exotics to a broad cross- sectionof the populaceinvolves ingenious and prolongedadvertising. Phased harvestingof the producelengthens the periodduring which suppliesare available;time charteringor affreightingare the preferredmodes for arrangingcheap and regularshipments, while containerization,where it is feasible,lowers costs and can be used for the purposesof ripening.

8.41 The eventualaim of a fruit exporteris to market a branded,firm specific,differentiated product that establishesa marketniche with a loyal following. Bananacompanies thiat are confidentabout the qualityof their product,the consistencyof their standardsand the reliabilityof their supplysystem, affix companylabels to the fruit. Retailersare requiredto removethe brand stickersafter a time to preventthe loss of goodwillfrom the sale of overripefruit. Sellinga brandedproduct worldwide is the apogee of the horticulturebusiness in terms of earningsand profitability.It shouldbe the goal of a countrylike Fiji that wishes to becomean important traderof tropicalfruit and vegetables,but as the precedingdiscussion suggests,realizing this ambitionentails careful plazining, substantial investment,and the willingnessto take risks.

8.42 There is no reason to concentrateon fruit and vegetables. Fiji has successfullyexported fresh and processedginger of high quality. On a limitedscale, it has also begun tappinga market for taro and other root crops among PacificIslanders settled in Australia,New Zealandand the US. Becausefresh taro, the preferredform of this commodity,is bulky,has a moisturecontent of 60-70%,is easilybruised and subjectto rapid decay, it is costly to transportand ideallyneeds to be air shippedin small batches. However,root crops are reducibleto clean,fine, grain starches. Creating overseasdemand for taro flour might be preferableas an avenueto higher exports. Of course,the problemsof supplyarising from fungaland viral diseasesthat are inducingfarmers to substitutecassava for taro would also have to be tackledthrough a breedingprogram that leads to hardierand more productivestrains. In other countries,cut flowersenjoy a thrivingtrade. Colombia,the secondlargest exporter sells $150 millionannually, mostly carnationspurchased by US buyers. Fiji could supplyorchids for the Japanese marketduring the peak weddingseason in October-Novemberwhen suppliesfrom - 97 -

Hawaiiare Limited. It might also be possibleto cash in on the demand for anthuriumsin the US and Europe. New Zealandis expandingits goat 95gck and hopes to exportNZ$100 million worth of goats and goat meat by 1990.,_ A commercializedapproach to goat raisingcould lead to Fiji'sentry into this market.

8.43 In short,the possibilitiesare numerous,not only in traditional itemssuch as melons,pawpaws, passion fruit, mangoes, and citrus,but also in productslike okra, zucchini,squash, vanilla, cut flowers,seaweed (the sourceof agar) and meat. No doubt, the availabilityof air transport facilitiesat reasonablerates will remaina constraintunless carriers, persuadedby increasedtourist traffic to raise the frequencyof services,can be inducedto combinethis with additionalcargo capacity. Pendinga reductionin air shipmentcharges, it might be desirableto give more attentionto the exportof dried and processedfruit for which there is a growingmarket in the US and Japan. For instance,if shippingbottlenecks interferewith the trade in freshbananas, weight and perishabilityproblems could be sidesteppedbw 87 oncentratingon bananachips, dried whole bananas, bananafigs and puree.8

Table8.5: FIJI: EXPORTSOF FRUIT IN 1985 (Tons)

Fruit Tons

Passioni uit 520 Pineapples 36 Mangoes 130 Pawpaws Limitedsales to Australia and New Zealand

Source: Fiji'sNinth DevelopmentPlan.

The first indicationsare encouraging. Marketsin the US, Japan, the EEC and Australiacould absorbfar biggerquantities, but only if the countryis able to assurebuyers that it can meet the criterionof consistentquality and reliablesupplies in the volumesdesired. Much of the investmentthat will

47/ Productionof goat meat in Fiji is estimatedto be 700 tons in 198b.

48/ See, B. Shaw, Food and NutritionPolicies for South PacificCountries: Determinantsof GovernmentPlanning; P. Sivan,Producing More Food in the South Pacific: What are the Possibilities;and S. Talagi,Improved Processingand Marketingof PacificIsland Foods, in R. R. Thaman and W. C. Clarkeeds. Food and NationalDevelopment in the South Pacific, U.S.P.Suva, 1983. - 98 - augmentproduction must come from the privatesector, but the experienceof New Zealandand Israelsuggest that the governmentmight considertaking the lead in four Areas.

(a) establishingmarketing organizations that defineand enforcequality standards,manage the collectionof produce,supervise the packaging,and assist in the dispatchof fruit and vegetablesto importersworldwide. In this area, a joint venturewith fruit marketingbodies from New Zealandmight be helpful;

(b) developmentof researchand extensionservices that complementthe effortsof the marketingagencies and have stronglinks with both the growersas u:ellas the marketingsystem;

(c) supportinginstitutional changes that would lead to the emergenceof orchardsand vegetablegardens on the scale needed if Fiji is to cement its currentlyrather tentative contacts with foreign centractors;and

(d) - egotiatingwith air cargo operatorsto augmentserv-ces available to Fijianexporters. Tab e 1

FIJI

Basic Indicatorsof FIJi ugar Industry,1976-85

Average 1976-78 1979 1980 1981 1982 1983 1984 1985/a

Bastc Data Arieaharvested('000 ha) 51 62 67 66 69 59 69 70 Cane production('000 tons) 2,602 4,058 3,360 3,931 4,075 2,203 4,290 3,043 Sugar production('000 tons) 335 473 396 470 486 276 480 341 Molassesproduction ('000 tons) 97 163 129 152 150 84 188 108 Yield Statistics Cane yield/ha 51.0 65.5 50.1 59.6 59.1 37.3 62.2 43.5 Cane/sugar ratio(TC/TS) 7.8 8.6 8.5 8.4 8.4 7.9 8.9 8.9 Prices Growersprices (F$/cane ton) 25.30 23.50 35.19 26.24 25.00 29.65 19.00 25.00 Growerssugar prices (F$/sugar ton) /b 197.34 202.10 299.12 220.41 210.00 234.24 150.10 204.70 FOB exportunit value price(F$/sugar ton) 281 273 395 322 304 326 290 253 World price (F$/sugarton) 249 178 517 319 173 190 124 104 FOB priceas : of world price 113 153 76 101 176 172 234 243

ExportEarnings (F$ million) Sugar 82 117 174 132 125 112 110 104 Molasses 2 7 12 10 5 3 7 7

Total 84 124 184 142 130 115 117 111 (as 2 of merchandiseexports) 30 61 66 i59 55 *52 47 mploment (numberof workers) Farmers 17,862 19,233 19,567 21,015 22,091 20,500 21,796 22,146 Cane cutters 13,705 17,003 19,300 19,411 19,911 11,295 16,244 12,924 USC 3,486 3,620 3,951 4,000 3,708 3,521 3,871 3,400

Total 35,053 39,856 42,818 44,426 45,710 35,316 41,911 38,470 (as X of totalemployment) n.a. n.a. n.a. n.a. 22.6 17.1 19.9 17.9 ValueAdded /c Sugaragriculture (as % GDP) 10/d 12 10 11 12 7 11 8 Sugaragriculture (as X Total AgricultureGDP) 437w 50 44 47 47 33 46 36 Sugarmanufacturing (as S GDP) 47-d 5 4 4 5 3 4 3 Sugarmanufacturing (as 2 of Total ManufacturingGDP) 3373 36 33 35 37 24 35 26

Total Sugar(as X of GDP) 14/d 17 14 15 17 10 15 11

/a Preliminarydata. /b Growersprice for sugar canehave been convertedby cane/sugarratio.

Ic Percentageshare at constantprices (1977 - 100),no currentprice data are available. /d 1977-78average only.

Source: Fiji: Adjustmentand SelectedDeveltpment Issues, World Bank, August1985, Report No. 5515-FIJ. CurrentEconomic Statistics, Bureau of Statistics,various issues. FijiSugar Corporation estimates. ANNEX Table 2 - 100 -

FIJI

Normal Tax Schedulesfor PersonalIncome Taxes

Does not Marginalrate on Exceeds (F$) exceed(F$) Tax F$ amountover Col 1 (%)

January 1, 1974 - December 31, 1980 0 600 - 7.5 600 2,000 45 12.5 2,000 3,000 220 17.5 3,000 4,000 395 22.5 4,000 4,500 620 25.0 4,500 5,000 745 27.5 5,000 5,500 883 30.0 5,500 6,000 1,033 32.5 6,000 6,500 1,195 35.0 6,500 7,000 1,370 37.5 7,000 7,500 1,558 40.0 7,500 8,000 1,758 42.5 8,000 10,000 1,970 45.0 10,000 12,000 2,870 47.5 12,000 - 3,820 50.0 JTanuar_1,_1981- December 31, 1985 600 2,500 30 11.0 2,500 4,000 239 18.0 4,000 5,500 509 27.5 5,500 7,000 922 32.5 7,000 8,500 1,409 40.0 8,500 10,000 2,009 42.5 10,000 12,000 2,647 45.0 12,000 20,000 3,546 47.5 20,000 - 7,347 50.0 January 1, 1986 1,500 5.0 1,500 3,000 75 10.0 3,000 4,000 225 18.0 4,000 5,500 405 27.5 5 500 7,000 818 32.5 7,700 9,000 1,305 37.5 9,000 15,000 2,055 42.5 15,000 25,000 4,605 45.0 25,000 40,000 9,105 47.5 40,000 16,230 50.0 ANNEX - 101 - Table 3

FIJI

Elasticityof Major Taxes and of the Fiji Tax System, 1974-81

Elasticity Buoyancy Type of Tax coefficient R2 Coefficient R2

Income Tax /a tax to base (GDP) 1.609 0.97 ' 1.607 0.96

CompanyTax /b tax to base (GDP) 1.160 0.96 1.257 0.96

Excise Duties tax to income (GDP) 0.203 0.85 0.972 0.98

Excise Duties tax to base (IIP) /c 0.024 0.15 0.107 0.13

Import Duties /d tax to income (GDP) 0.678 0.95 1,085 0.98

Import Duties /d tax to base (imports) 0.574 0.94 0.911 0.91

Other Taxes /e tax to base (GDP) 1.268 0.72 1.268 0.72

Total Taxes tax to base (GDP) 1.101 0.99 1.294 0.99

Notes: 1. Income tax includes personal tax PAYE, basic tax and provisiontax. 2. Company tax includesdividends. 3. IIP is the Index of IndustrialProduction for all items excludinggold and sugar. 4. Import tax includescustoms and fiscal duty. 5. Other tax includeshotel turnover,land sales and miscellaneouscustoms receipts.

Source: The FinancialReview CommitteeReport, 1985, ParliamentryPaper No. 54 of 1985, Parliamentof Fiji, August 1985, p. 13. Also see Analysis of the Elasticityof the Fiji Tax System,Jacqueline Harvey, Ministry of Finance, unpublishedmimeo. ANNEX - 102 - Table 4

FIJI

Buoyancy Estiuates of Central Government Revenuas (1919-85)

Average for period /a

Taxes on individual income 1.30

Taxes on company income 2.00

Taxes on goods and services 2.40

Taxes on internationaltrade 1.73

Other taxes 0.84

Tax Revenue 1.63

Nontax revenue 1.02

Total Revenue 1.53

Foreign grants 3.36

/a Calculated using point elasticities.

Source: Anenx Table 2.