MSU LIBRARIES # RETURNING MATERIALS: PIace in book drop to remove this checkout from our record. FINES wiII be charged if book is returned after the date stamped beIow. PATTERN OF SPATIAL DEVELOPMENT AND REBIONAL INEQJALITIES IN PENINSULAR MALAYSIA By Zainul Bahrin Bin Mohd. Zain A DISSERTATION Submitted to Michigan State University in partial fulfillment for the requiranents forthedegreeof DOCIOR OF PHILOSOPHY Urban and Regional Planning Program College of Social Science 1987 ABSTRACT PATTERN’OF SPATIAL DEVELOPMENT AND REGIONAL INEQUALITIES IN’PENINSULAR.MALAYSIA By Zainul Bahrin Bin thd. Zain The existence of regional inequality is a problem.shared by most_ developing countries. While some degree of regional inequality in development is inevitable, its persistence and increase in magnitude has been recognized as a.major obstacle in the achievement of significant development changes. Spatial imbalances not only reflect social inequities in.the distribution of development benefits, but also act as impediments in efforts to prunote harmonious and cohesive natiaxal develcpnent. The formulation of effective redressal strategy requires a proper understanding of the pattern.of spatial development and the nature of the inequalities. This study examines the pattern of spatial development and the change of regional inequalities in Peninsular Malaysia.between 1970 and 1980. .A few studies of regional inequality oeralaysia exist but rely cn.data for large units of analysis, mainly at the state-level. Data for large units tend to hide some important regional attributes which may be critical for the developnent of appropriate policies . This study uses smaller units of analysis, namely administrative districts, to detect the distribution of inequalities and the relative changefran1970t01980, aperiodwhentheMalaysianGoverment recognized the resolution of inequalities as a high priority national objective. Basedcndata derived frunseccndarysources, thirtyvariables indicating different mots of develognent are used in the analysis. Indicesofdeveloplmtarecmsuuctedtoneasurethelevel of develcpnent of 64 canparable districts of Peninsular Malaysia. The magnitude of regional inequalities is then assessed using Gini coefficient techniques. Variatiais and disparities in developnent performance exist at all scales of analysis. lndices of development and Gini coefficients canfirmthatvariatimsandgapsindeveloprentarennreprmmcedat thedistrictratherthanfluestatescaleofanalysis. Thestudy Gainstrates that the spatial pattern of inequality that prevailed in 1970 basically renamed unchanged, with urban districts receiving greater deveth benefits than rural counterparts. Similarly, districtsinthewestcoaststates faredbetterthanttnsemtheeast coaststates. Basedontheanalysis, thestudyoffersanumberof general policy directions to foster a more balanced development in II Malaysia. For My Wife, Zaini, and our children, Airin, Azrin, Edalin and Iskandar. iv First and foremost, I mild like to express my humble gratitude toGodfineAlmightyfor-bestowinguponmefineabilitytoundergo finroughfinis learningprocess. Wifithisblessingsfinisstudymay never have bean undertaken. My profannd thanks goes to the Public Services Departuent, Malaysia for granting me fine opportunity to continue with my graduate studies. Wifinout fine study leave and necessary financial support fran fine P.S.D. it would not have been possible for this study and fine Ph.D. program to be camleted. Tlnarnksarealsoduetofinevariwsgovermnentdepartnentsand agencies in Malaysia for extending fineir assistance in providing data andinformaticnneededforfinisstudy. Inparticular, Iamtharnkful to fine Department of Statistics, Malaysia for facilitating fine procurement of mast of fine pertinent data at fine district level. Its officers and staff have been graciously helpful in many ways. Many people, including personal friends, have assisted in sane way or ofiner in fine preparation and accomplishment of this dissertation. My appreciation and gratitude goes to all concerned. Special appreciation and finanks goes to Dr. Rene Hinojosa, not only as Director of my Dissertation Ocnmittee but also for his relentless guidance and scholarly advice throughout fine developnent andconpletionofthisstudy. I amalsothankfultoProfessorsRoger Hamlin, Assefa Mehretu, David Campbell, Lawrence Libby, arnd Keifin meymtonlyforfineirwillingnesstoserveonmyéiidanceand Dissertation Committees, but also for fineir academic guidance and advice. Iwouldalso liketofinankProfessorRobertMcKinleyfor serving on fine defense cormittee. Tbmyparents, H3. Mohd. ZainandlSiti Ramah, I enfinankful for givingmeccnstantguidanceandhope. Theassistancegivenbybothmy families are also appreciated. Finally, my fenily deserves more finanks and appreciation finan I could possibly convey. My very special thanks and @pest appreciation to my beloved wife, Zaini, for her sacrifice, patience and encouragement. To my dearest children, Airin, Azrin, Edalin,anc‘l Iskandar, who have innocently sacrificed many of fine childhood joys becauseoffinisundertaking, Iamverygrateful forfineir understanding. Mywifeandchildrenhavebeenasourceof inspiration more finan finey will ever know. vi TABLE OF (IDN'I'ENTS List of Tables. . . ............................................... List of Figures ................................................. CHAPTER (NE IN'I'REDUCTICN .......................................... Spatial Developnent in Developing Countries .......... Background of the Problem in Peninsular Malaysia ..... Problem Statenent .................................... (bjective of fine Study ............................... Hypofineses. . . ..... . .................................. Significance of fine Study ............................ Key concepts of the Study ............................ Organization of fine Study ............................ REVIEW OF THE LITERATURE ON DEVELOPMENT AND REGIONAL INEUJALITY Meaning, Dimensions, and Goals of Develognent ........ Meaning and Dimensions of Inequality ................. Space arnd Region in Development ...................... Approach to Developnent ............................. Measurement of Developnent .......................... vii 21 25 26 29 32 34 47 51 59 71 Inequality in Developnent: Empirical Evidence ........ 79 W Focus of fine Study ................................... 90 Time Period of fine Study ............................. 91 Data for fine Study ................................... 93 Theoretical Justification for Variables Selected ..... 95 Unit of Observation ................................. 117 Weighting of Variables ............................... 125 Research Techniques .................................. 127 PATTERN OF SPATIAL DEVELOPMENT AND REGIONAL INHIJALITIES IN PENINSULAR MALAYSIA, 1970-1980 Introduction. . . . . . . .................................. 137 Index of Developnent. ................................ 138 Pattern of Spatial Developnent ....................... 138 Results of Survey .................................... 145 1970 Developnent Index ........................... 148 1980 Developnent Index ........................... 161 Oonparison of 1970 and 1980 Development Indices .......................................... 172 Gaps in Spatial Developnent ...................... 189 Changes Over Time ................................ 193 Magnitude of Spatial Inequality ...................... 195 Spatial Inequality: States and Districts ......... 197 Spatial Inequality: Urban and Rural Districts. . . . 207 viii SIX: Spatial Inequality: Western and Eastern Districts 214 Gaps in Spatial Inequality. ...................... 218 Changes Over Time. ............................... 222 Sumnery of Findings .................................. 225 POLICY IMPLIQTICNS FOR EUJ'ITABLE DEVELOPMENT Developtent Policy Implications Fron Results of 'I‘heStudy ................ ...... 232 Inplications Fron Results of fine Study ............... 240 InplicationsFronPersonalExperience ................ 240 Political Will ....................................... 248 SUVMARY AND MUSION Contextual Overview .................................. 251 Summary of Findings of fine Study ..................... 253 Suggestion for Purfiner Research ..................... 255 Concluding Note ...................................... 257 APPENDICES l . Indicators of Developnent: Operationalization and Sources of Data ........................... 259 2(A). States in Peninsular Malaysia, 1970 and 1980... 264 2(8). Area of States in Peninsular Malaysia, 1970 and 1980 ....................................... 265 2(C). Population of States in Peninsular Malaysia, 1970 and 1980 .................................. 266 3. States in Penirsular Malaysia, 1970 and 1980: For Conparative Use ............................. 267 4(A). Adninistrative Districts in Peninsular Malaysia, 1970 arnd 1980 ........................ 268 ix 4(B). Ma of Adninistrative Districts in Peninsular Malaysia, 1970 and 1980 ........................ 271 4(C). Population of Adninistrative Districts in Peninsular Malaysia, 1970 and 1980 ............. 274 5. Adninistrative Districts in Peninsular Malaysia, 1970 and 1980: For Conparative Use. .. 277 6. Seuple of ‘Value Judgment Survey' Forms ........ 281 7. UrbanandRuralDistrictsinPeninsular Malaysia, 1970 and 1980 ......................... 284 8. Urban and Rural Districts in Peninsular Malaysia, 1970 and 1980: For Comparative Use. . . . 286 9. Places With Population 10,000 And More in Peninsular Malaysia, 1970 ....................... 288 10. Places Wifin Population 10,000 And More in Peninsular Malaysia, 1980 ....................... 11. Variables Used To Calculate Gini Coefficients. . . 291 BIBLICBRAPHY .......................................... 295 TABLE 1.1: 1.2: 3.1: 4.2: 4.3: 4.4: 4.6: 4.7: 4.8: LIST OF TABLES Per Capita Gross Dorestic Product, 1971 and 1980 (in Malaysian Ringgit at 1970 Prices. . . . . ........... on? by Industryof Origin, 1970 and 1980............ Developnent Criteria and Indicators Used In Analysis of Pattern of Spatial Developnent and Regional Inequalities in Peninsular Malaysia, 1970 and l980................ .......... ............... ....... Weights of 30 Selected Indicators of Development (Based on Value Judgement Survey) ............ . ...... 1970 Conparative Districts, Peninsular Malaysia -- Index of Development (Umeighted) end Ranking. ..... 1970 Conparative Districts, Peninsular Malaysia -- Distribution of Development Index (Unweighted) by Geographical Location and Urban and Rural Districts......... .................................. 1970 Conparative Districts, Peninsular Malaysia ~- Breakdcwn of Districts With Positive Index of Develognent by States ............................... 1970 Conparative Districts, Peninsular Malaysia -- Ranking of Districts According to Development 1970 Conparative Districts, Peninsular Malaysia: Index of Developnent (Weighted) arnd Ranking ......... 1970 Conparative Districts, Peninsular Malaysia -- Distribution of Developnent Index (Weighted) by Geographical location and Urban and Rural Districts ............. . ............................. 1970 Non-Conparative Districts, Peninsular Malaysia -- Distribution of Developnent Index (Unweighted) by Geographical Location and Urban and Rural Districts ........................................... xi PAGE 16 17 99 147 149 152 154 156 158 159 160 4.9: 1970 Non-Connparative Districts, Peninsular Malaysia -- Distribution of Developnent Index (Weighted) by Geographical Location and Urban and Rural Districts ................................... . ....... 4.10: 1980 Conparative Districts, Peninsular Malaysia -- Index of Development (Urnweighted) and Ranking ...... 4.11: 1980 Comparative Districts, Peninsular Malaysia -— Distribution of Developnent Index ( Unweighted) by Geographical Location and Urban and Rural Districts. . . . ...... . ......................... . ..... 4.12: 1980 Comarative Districts, Peninsular Malaysia -- Breakdown of Districts with Positive Index of Developnent by States .............................. 4.13: 1980 Comparative Districts, Peninsular Malaysia -- Index of development (Weighted) and Rarnking ........ 4.14: 1980 Conparative Districts, Peninsular Malaysia -- Ranking of Districts According to Development Index .............................................. 4.15: 1980 Conparative Districts, Peninsular Malaysia: Distribution of deveth Index (Weighted) by Geographical Location and Urban and Rural Districts. . ........................................ 4.16: 1980 Non-Conparative Districts, Peninsular Malaysia -- Distribution of Developnent Index (Urnweighted) by Geographical Location and Urban and Rural Districts .......................................... 4.17: 1980 Non-Conparative Districts, Peninsular Malaysia -- Distribution of Developnent Index (Weighted) ByGeographical LocationandUrbanandRural Districts .......................................... 4.18: Connparative Districts -- Number of Districts With Positive Index of Developnent ...................... 4.19: Conparative Districts -- Number of Districts With Negative Index of Developnent ...................... 4.20: 1970 Districts, Peninsular Malaysia: Index of Developnent (Urnweighted) and Ranking ............... 4.21: 1980 Districts, Peninsular Malaysia: Index of Developnnent (Unweighted) arnd Ranking ............... 160 162 165 166 168 169 170 171 172 173 173 175 176 4.24: 4.25: 4.26: 4.27: 4.28: 4.29: 4.30: 4.31: 4.32: 4.33: 4.34: 4.35: 4.36: Non-Conparative Districts -- Number of Districts Wifin Positive annd Negative Index of Develop'aent. . . . Conparative States: Index of Development (Ikmeighted), 1970 end 1980 ..... . ......... . ........ 1970 and 1980 Conparative Districts -- Western annd Eastern Districts With Positive Developnnent 1970 and 1980 Conparative Districts -- Western and Eastern Districts With Negative Developnent Inndex. . .......... . ........................ . ........ 1970 end 1980 Conparative Districts -- Urban and Rural Districts With Positive Developnent Index......... .............................. .. ..... 1970 and 1980 Couparative Districts -- Urbann and Rural Districts Wifin Negative Developnent Index. . ............................................ Distribution of Districts According To Performance of Index of Developnnent for 1970 and 1980. ...... Couparative Districts, Peninsular Malaysia: Relative Gap In Developnent Performance Wifin The Most Developed District, 1970 annd 1980 ..... . ....... Ranking of Comparative Districts, Peninsular Malaysia 1970 and 1980 (According to Developnent Index. . . ..... . . . ................................... Location Qlotient for Connparative Districts: Performance of Variables By Districts, 1970. ....... Location Q.notient for Comparative Districts: Performance of Variables By Districts, 1980 ........ Location Quotient for Comparative Districts: Performance of Districts By Variables, 1970 ........ Location Quotient for Couparative Districts: Performance of Districts By Variables, 1980 ........ Gini Coefficients At State, District, Urban District, Rural District, Western District, and Eastern District Scales of Analysis, 1970 .......... Gini Coefficients At State, District, Urban District, Rural District, Western District, and Eastern District Scales of Annalysis, 1980 .......... xiii 178 179 182 183 184 187 188 191 194 199 200 201 203 205 206 4.37: 4.38: 4.39: 4.40: 4.41: 4.42: 4.47: Frequency Distribution of Gini Coefficients for State and.District Scales of Analysis, 1970 and 1980 ...... . ........................................ 207 Conparison of Gini Coefficients at State Scale of Annalysis Wifin Other Scales of Annalysis, 1970 and 1980 ........ . ...................................... 210 Frequency Distribution.of Gini Coefficients for Urban end Rural Districts, 1970 and 1980 ........... 211 Mean For fine 30 Gini Coefficients, 1970 end 1980. . . 211 Frequency Distribution of Gini Coefficients for western and Eastern.Districts, 1970 and 1980 ....... 215 Gini Coefficient Ratios, 1970: Spatial Inequality Gaps Between State and District Scales of Analysis, urban.Districts and Rural Districts, and.Western.Districts and Eastern Districts ........ 219 Gini Coefficient Ratios, 1980: Spatial Inequality Gaps Between State and District Scales of Analysis, urban.Districts and.Rural Districts, and western.Districts and Eastern.Districts ........ 220 Gini Coefficient Ratios -- Direction and Rate of Change. ..... ... .................................... Ratios Between.Gini Coefficients of 1970 and 1980: States, Districts, urban Districts, Rural Districts, western.Districts, and.Eastern Districts .......................................... 221 223 Frequency Distribution of Gini Coefficient Ratios -_ convergence, Divergence, and Status Quo in Rate of Change ..................................... Ranking of Gini Coefficient Ratios of 1970 and 1980 (Comparisons of variable Performance Within Regional units) .................................... 226 224 Ranking of Gini Coefficient Ratios of 1970 and 1980 (Comparisons of variable Performance Among Regional Uhits) .................................... 227 FIGURE 2.1: 3.1: 3.2: 3.3: 3.4: 4.3: 4.4: 4.8: 4.9: LIST OF FIGURES PAGE A.Hypothetical Model of the Development Path .......... 82 States in Peninsular Malaysia, 1970 .................. 121 States, including the Federal Territory of Kuala Lumpur, in Peninsular Malaysia, 1980... ......... 122 Administrative Districts in Peninsular Malaysia, 197000.00... 00000 0 0000000 0 00000000000000 0 0000000000000 123 Administrative Districts in Peninsular Malaysia, 1980...... ............................................ 124 Process of Policy/Program Planning and Implementation, Malaysia .............................. 144 1970 Corparative Districts, Peninsular Malaysia: Level of Development .................................. 150 1970 Districts, Peninsular Malaysia: Level of Developnent .................................. 151 1980 Couparative Districts, Peninnsular Malaysia: Level of Development .................................. 163 1980 Districts, Peninsular Malaysia: Level of Developnent. . . ............................... 164 Distribution of Piped water Supply in States, Peninsular Malaysia, 1970 and 1980 .................... 208 Distribution of Agricultural Extension.Service in States, Peninsular Malaysia, 1970 arnd 1980 ............ 208 Distribution.of Piped.Water Supply in Districts, Peninsular Malaysia, 1970 and 1980 .................... 209 Distribution of Agricultural Extension Service in Districts, Peninsular Malaysia, 1970 and 1980 ......... 209 XV 4.10: Distribution.of Piped.Water Supply in.Urban.Districts, Peninsular Malaysia, 1970 and 1980 .................... 4.11: Distribution.of Agricultural Extension Service in 4.12: 4.13: 4.14: 4.15: 4.16: 4.17: Utban.Districts, Peninsular Malaysia, 1970 and 1980..... .................................... ........ Distribution of Piped.Water Supply in Rural Districts, Peninsular Malaysia, 1970 and 1980 ............ ........ Distribution of Agricultural Extension.Service in Rural Districts, Peninsular Malaysia, 1970 and 1980... Distribution.of Piped.Water Supply in.Western Districts, Peninsular Malaysia, 1970 and.1980 ......... Distribution.of Agricultural Extension.Service in Rural Districts, Peninsular Malaysia, 1970 annd 1980.. . Distribution of Piped.Water Supply in.Eastern Districts, Peninsular Malaysia, 1970 and 1980 ......... Distribution of Agricultural Extension.Service in Eastern Districts, Peninsular Malaysia, 1970 end 19m0000000 00000000000000000 0 0000000000000 000000 000000 5.1: A Model Of The Evaluation Process Of Developnent ....... 212 212 213 213 216 217 218 218 246 CHAPTERG‘IE INTRODUCTION The pursuit of development is to promote annd achieve desirable changes (Cant, 1975; Gore, 1984). Sears (1969) views developrent as a normativeconceptthatisahnostasynengmforimprovement. Colman and Nixson (1978, p.2) reinforce fie concept when finey state: "Developmentcanbeconsidered . . . asaprocessof improvementwith rospecttoasetofvalues. . . . Thevaluesinquestionrelateto desired conditios in society." Any adverse cosequences and effects resulting from fie initiated development process are, fierefore, not only undesirable, but also reflect shortcomings and posed constraints in fie strategies arnd execution of development. As Bruton (1985, p.1114) remarks: "that impedes fine achievement arnd exercise of finese efforts [to achieve higher welfare] impedes developmen ." It is a fact that spatial inequalities in development is a. common developnent phenomena found in all countries, regardless of fieir politico-economic system (Forde, 1968; Slater, 1975; Abu-Lughod and Hay, Jr., 1979; Stohr and Taylor, 1981; Smith, 1982). Some degree of spatial inequalities in developrent, finough undesirable, is inevitable and will always prevail. However, when spatial inequalities in development persist and result in distinctive and polarized differences between sub-national areas, then such phenomena become problematical to development. ‘I'I'e potential polarized trajectory of fie development process from such a spatial configuration would not only serve to intesify developrent problems, but also could negate and stifle fie development process being undertaken (Slater, 1975). Regional conflicts may have fie detrimental effect of slowing down, annd hence jeopardizing, efforts toward achieving a more equitable and integrated developnent. The incessant persistence of spatial inequalities in development as well as fie potential negative impact on future development process cannot, therefore, be considered a desirable change. Hence, Bruton (1985) considers inequality in development as an important and damaging failure. The persistence of spatial inequalities in developnent as well as fieir potential to becone acute, through polarization over time, fierefore, not only provides legitimate concern for its resolution, but also bears important implicatios annd significannce as an issue of development. Spatial Development In Developing Countries A fundamental concern of considerable importance to nmost developing countries in fieir relentless quest for developrent has not only been over how fie development process could generate and produce a more equitable distribution of fie benefits of development ( Jakobson annd Prakash, 1971), but also over fine persistence of distinnctive disparities in developnent between geographical areas and fine innhabitants fierein (Bhooshan annd Misra, 1980; Cole, 1981). "Tl'e persistence . . . of spatial inequalities in developnent constitutes a crucial problem in fine territorial organization of Third World countries" (Slater, 1975, p.99). The view finat spatial inequalities in deveth constitute a problem requiring (policy annd planning) intervention hinges upon fie arguments asserting finat a polarized spatial pattern of developnent could have adverse implications for overall national development (Gore, 1984, p.20). Consistent wifin such a concern, ! govermentsofmostdevelopingccuntriesseektospreadfiebeefitsfi of development as effectively as possible finroughout all geographical [ entitiesunderfineirgovernanoesoastoachievearelativelymore I balaneedspatialstructnmethatconldimproveandincreaseaccessof large segments of fie population to social, economic and political opportunities (Hoyle, 1974; Rondinelli annd Ruddle, 1977). In accordancewifinfinegrovinginterestbygcvernmentsindeveloping countries in fie way finat spatial organization could be articulated to ensure finat fine beefits of development reach fine greatest number of fine intended recipients, a variety of developnent programs have been espoused annd implemented wifin fie aim of ameliorating and enhancing fie social and economic well-being of fie country's various gecgrahical entities. While it is acknowledged that such measures are germane to pronoting social and economic advancement, most efforts undertaken lack adequate cosideration of fie spatial dimension of development. This is clearly reflected by fie tendency of most developing counntries to deal wifin development in a sectoral fashion. As Patnaik (1982, p.18) aptly and succinnctly stated: InThirdWorld countries, was apolicy parameterto the socio-economicproblensappearedmuchlater. Evenrcwformost ofthe less developed ccunntries, it is still in a formative stage. Notfinatfieseconmtriesdorcthaveregionalproblens. Rafier... fieregionalissuesweresubmergedbyissues like unenployment, nnnderenployment, hunger, disease, droughts annd floodsonanational scale, anndnno lessbyfiepolitical issuesinfie wakeoffieir independece....[Also]fie eoonomistsinthe‘lhirdWorldare alsorcless responsible for neglectingspace,becauseoffineir...anxietyfornotlagging behindfineirwesterncomterpartsinbuildingmaoro—ecoomic modelsforfineirconntries. Mostoffineseeconomistshadfineir trainingandderivedinspirationfromfineWest. . . . They couldnot. . .looktofineirdomesticproblensfromafresh angle, untilfiepressureofeventsmadefinemdoso. (Emphasisadded). Such an ennphasis, dominantly aspatial in focus, creates problems of equitable spatial organization. Thus, at the Seminar on National Development annd Regional Policy, sponsored by fie United Nations Center for Regional Development (UNCRD), in 1979, (Prantilla, 1981) some of fie participants' major observations were: (1) fiere was a lack of effective integration among national annd regional policies in developing countries, and, (2) most policies of developing counntries on investments tended to be space-neutral. Chatterji (1976, p.1) also remarked finat "over the years, it has been increasingly realised that national planning has nct led to a balanced development in most countries. One of fie reasons is that . . . most economic plannning strategies . . . abstract fie notion of space" as well as cosiderably ignores fie social dimension of development (Crooks, 1971). Referring to national plans for most of fie counntries in South and Soufieast Asia, Jakobson annd Prakash (1971, '13 (D u] H) 5 p.26) advanced fine view that "they cocentated on fie problem of economicgrowfinandincreasingfinancial resources, annd [asaresult] gave innadequate attention to relating economic development to its spatial consequences." Colman annd Nixson (1978) have observed that fie preoccupation of fie less developed counnties wifin national aggregates end averages inevitably led to fine neglect of fine problem of spatial distibution of developnent. As a result, fine spatial configuration of development which emerged in most developing countries has been been predominantly characterized by small, scattered cases of relatively developed areas and a vast desert of relatively underdeveloped hinterland. Friedmann (1972) referred ‘w to such a spatial development landscape as center-periphery I configuration. According to Friedmann (1966), a prologed center- i periphery relation led to potentially entree inequities and hence resticted developnennt. Economic annd social development in most less developed counties has been characterized by centipetal forces of cocentation annd agglomeration of pnblic as well as private investments favoring fie relatively more developed regios (Mehretu, Wittick and Pigozzi, 1981): wlnere fine bulk of modern inndusty, infrastucture, services and institutions were cocentated, nnnderdevelopnent remained pronounced in fie peripheral rural areas innhabited by fie majority of fie national population. (Rondinelli and Evans, 1983). According to Boudeville (1971) and Jose A. Snnith (1974), fine problem of regional inequalities in development was more intense in fine developing counnties than elsewhere (El-Shakhs, 1976; Abu-Lugl‘cd annd Hay, Jr., 1979; Higgins, 1981). Logan (1980, p.iii) pointed olt finat "there isnncw. . . widespreadacceptanceoffierecessityfordisaggregation in space in fine conrse of planning national developnennt." The dichotomy finat characterizes developnent in developing conntiesisreflectedinandreinforcedbyfieexisteceand persistece of a polarized spatial settlenent pattern (Rondinelli, 1979/1980). Investments in productive activities, infrastructure, services annd facilities have geerally been concentated in fie major Laban centers (Coates, Johnston annd Kncx, 1977) in fie belief that the} hignerreturnsenpectedinfiesecentersmuldstimmnlategronfinand accelerate development, annd finat fie resultant beefits would gradually filter annd tansmit to fine peripl'eral rural hinterlannd -- as happeedinmennyindustializednatiosofmropeandlnbrthAmerica. Ibwever, "bofin empirical evidece annd fieoretical explanatios point ] to fine tendencies of development toward cocentation or spatial / polarization . . . inoeorafewcoreareaswithindeveloping national systems" (El-vShakhs, 1976, p.127). Experieces of most of fie Third World counties ('I‘WCs) . inndicated that fie spread effects of concentated investments in a few I major urban centers or geographical areas were not as effective a as fieoretically anticipated (Ahluwalia, Carter, arnd Qnenery, 1979). The backwash effects tended to be greater finan fie tickle-down effects. For example, among fie major findings of case studies on fie impact of growth centers upon fieir hinterland by Appalraju annd Safier (1976) on Third World counties and Gilbert (1975) on Columbia “I (I) 9) (Q indicated finat spread effects from growth centers were usually snmaller finanexpectedorlessfinanfinebackwasheffectsandhadanegativenet result on fie hinterland (Stohr annd Todtling, 1977). Leinbach (1972) remarked finat fine spatial cocentation of developmennt activities around fie core areas often becomes deeply ingrained annd crystallized cvertimesofinatexpansionismoremarkedinfievertical rafierthan thehorizontal dimension. ThiswasalsoendorsedbyBrutzkus (1975, p.640) who stated: "For most of fie developing counties spread effects are not at all promising unless a very definite decentaliza- tion policy is formulated annd vigorously enforced." Tre focus of development activities in certain selective urban centersorgeographical areascontibutedtofieemergenceofa galaxy of inticate urban problems; manifested by such externalities as squatter settlements, slums, housing slnortages, inadequate jobs, enrplus unennplcyed annd underenplcyed labor force, taffic cogestion, environmental pollution and degradation, and insufficient infrastuctural annd social facilities (Young, 1976). Rural or peripteral development was also affected because of loss of some of its humannresonroestofieurbanareasfinroughrural-urbanmigration exertedbyfiecentipetal forcesoffiemoredevelopedareas. Inan attempt to redress fiese problems, governments of many developing connties pooled fieir financial resonrces annd planning efforts togefier on fincse areas of population cocentation at fie expense of fie lagging regios. This led to a costly cycle of unbalanced and inequitable development of many Third World counties (Adarkwa, 1983). Thus, in many developing connties the economy and society became Z ... .. 3.. a... 9 .3 a; D. 3 s S. a? t .n mcre dualistic -- the gap between the modern and traditional economic sectors as well as between the ridh and.the poor in the society further diverged rafier finan converged (Nijkamp annd Van Pelt, 1983). The spatial system also became more polarized. "The disparity between the center and the periphery [in.mcst less developed countries] has grownnwith few signs of abating" unarmed» hflttick.and Pigozzi, 1981, p.1). Such a spatial configuration of development nnot onay'failed to reflect, much less generate and facilitate, a.more equitable distihution of fie beefits of development inn fie peripheral areas, but also drained them of their resources vital for facilitating rural advancement (Brutzkus, 1975). Indeed such.a discordant spatial pattern of development tends to perpetuate and even.exacerhated the prohiemnof spatial inequalities in.develcpment. Baer (1964, p.269) opined that."cnce unequal rates of growth develop, they will tend to perpetuate themselves." This is largely due to the fact that as development activities became more concentrated in.a particular location the comparative advantage of finat location was furfier enhanced (Hicks, 1959; Mabogunje, 1981; El-Shakhs, 1976). "Once . . . decisions are made to locate a particular activity or institution at a specific point, a kind.of self-generatingnmomentum.is established which continues to attract related enterprises annd inndeed multiplies fie impact of a given social, economic, or political investmen " (Soja annd Tobin, 1979, p.158). The principle involved was identified by Myrdal (1957) 1 as one of ‘circular annd cumulative causation', fie effect of which 1' 1’ I i .a ......— Tyr- —.-....—.—— will he deepening of differeces between various areas of a country annd giving rise to spatial structures finat reflect different patterns annd rates of developmennt (Mahogunnje, 1981). Such a develcpnent also tendedtobemarkedbyacorrespondingdeteriorationofofier disadvantaged regios. "The disadvantaged regions did nct stannd still thronghcutfinissetofchanges. Everydevelopmentchangeinfie advantagedregionwasmarkedbyacorrespondirgworseningoftheir conditios, eqnecially wifin fine ont—migration of fie youger, more eergetic population to fine grwing region annd fie consequent loss of fine labor, entepreneurship annd capital which fiey represent" (Mahogurnje, 1981, pp.59-60). Tre problem of spatial or regional inequality of development, fierefore, constitutes an important issue of development plannnning in developing counties. As stated by Stohr and Todtling (1977, p.33), "fie reduction of spatial disparities of living levels is a key objective of most national urban annd regional development policies." Background of fie Problem in Peninsular Malaysia Like most other ererging developing nations, fie pursuit of development in Malaysia is, fundamentally, pronpted annd spnrred by fine desire to pronote and enhance fie social and economic well-being of her citizes. The Outline Perspective Plan, 1971-1990, a development policy guideline, posits that oe of fie paramount obj actives of development policy is "modernization of rural life arnd improvement of living conditions among fie urban poor throngh fie provision of a wide --.Hc—u... A“; 10 range of social services . . . annd community facilities" (Third Malaysia Plan, 1976-1980, 1976, p.51). Reduction of inter-regional disparities in development costitutes an important development plannning objective in Malaysia. As stated by fine Second Malaysia Plan, 1971-1975 (1971, p.42): Greaterregionalequality. . . [constitutes] partoffine halancegoal . . . . T'reargumentsinfavourofregional balance . . . . furndanmentally [rests] onfienotionfinatall regios in Malaysia share in fine beefits of development. Itwasalsostatedfinat "newprogrammes . . . . innitiated . . . . will emphasize regional balance annd integration" (Second Malaysia Plan, 1971-1975, p.46). Tl'e Third Malaysia Plan, 1976-1980 again stessed finat fie objective of regional development stategy in Malaysia was: to narrow fie disparities in fine standard [level] of living between regios . . . finrough fie exploitation of fie full potentialoffiehumanandphysicalresonrcesoffineless developed regions finrogh equitable distibution of fie basic services annd amenities (p.99) as well as to: to bring ahont closer integration among fie states of Malaysia . . . . finronghredressingimbalancesenogfineregionswithin fie county (p.199). Such policy statements, inevitably implied finat fie benefits of development activities shonld permeate and be accessible to all citizens, regardless of fineir economic functions annd geographical locations. As stated by Rondirelli (1979/1980, p.14): 11 Morewidespread . . . anndnmcreequitahle distihutionoffie beefits of growth are prerequisites to modernization annd [also] preconditions for creating socially just annd politically stable societies. Mahogtunje (1981, p.40) also argued finat spatial accessibility to basic services, infrastructures and amenities associated with developmenteffortswasvitalinnfineendeavortoattainamore equitable distribution of fine benefits of development . Tne importance of spatial accessibility in proncting development was also explicitly recognized by fine Malaysian aufinorities who stated: "Besides contibnting to accelerated development of fie nation as a whole, fie redressal of regional imbalance will esure greater opportunities for economic annd social advancement of people in different parts of fie county, fiereby enhancing fieir well-being" (Third Malaysia Plann, 1976-1980, 1976, p.114). In finis regard, Kamal Salih (1977, p.38) alluded to fie question of regional imbalance by stating finat "fine critical question of unequal access [to development beefits] in fie explanation of poverty, end in evaluating fie potential of poverty eradication, cannnct be cveremphasized." Rondinnelli annd Ruddle (1977) have also made a sinmilar statenent. The persistece of disparities in fie distibution of developnent beefits is discocerting to fie Malaysian Government; because it is recognized by fie Malaysian auficrities finat such phecnena conld negate efforts to redress regional imbalance annd also to enhance the social annd economic well-being of Malaysians. Tle efinnnic violece of May 13th, 1969, following fie national 12 elections, was a manifestation of fie sympbon of lop-sided developnent finat prevailed in Malaysia. Tre incident was a significant eye-opener that pronpted fie review annd reformulation of Malaysia's development strategy. PriortofieintroductionandadoptionoffieNechononic Policy, an important umbrella policy for national developnent, fine finrust of Malaysia's developnent strategy was towards fie acoonplisl'ment of rapid econonnic growth. Tre New Economic Policy, however, officially paved fie way for fie introduction and inplenentation of fie cocept of distribution with growth in Malaysia's developnent strategy (Bruton, 1982). "In fie 1950's annd 1960's, like fie governments of manny LDC's, fie Malaysian government pursuedpoliciesdesignedtoproncteecoenicgrowth . . . . Onlyin fie 1970's, particularly after fie comnunal riots of 1969 annd fie subsequent formulation of fie New Economic Policy (NEP), was fiere a definite shift of enphasis from mere growth to distribution with growth" (Muniappan, 1982, p.6; see also Aris Ofiman, 1984). Realization on fine part of fie Malaysian authorities of fie danger posed by unbalanced development spurred fie formulation annd enunciation of fie New Eoononic Policy in 1971. Tie New Econonic Policy contained two fundamental, but closely intertwined objectives: / over fie persistence of disparities in development in Malaysia was (1) eradication of poverty, and (2) restructuring of society. Cocern againalludedtobyfieDeputyPrimemnisterofMalaysia, durirga parliamentary session in 1983, when re stated: "Tne [Malaysian] Government has always arnd will always take serions view of this [disparities in developnent] widening gap" (The New Straits Times, 13 1983, p.8). Gaps in fie distribution of development beefits conncted finatfieaimtoreduceregional inequalityandtoupgradefiesocio- economic well-being of fie people conld be hampered and possibly inhibited. That posed a serions setback for developnent efforts aimingtoachievegrwthwifinequity. Itwasgeerallyagreedbyfine government finat regional maldistribution arnd inequalities were repulsive nnot only for fieir efinical annd humanistic implications, but also because of fieir potential as root causes of discontent, political unrest, annd revolution (Soja annd Tobin, 1977; Butterfield, 1977: Jchari Mat, 1983). As David Harvey (in Smith, 1982, Foreword) succintly stated: "fie existece of . . . social inequality challeges onrefinical andmnoral sense, makingusquestionfiefairnessand justness of fie economic, political, and mnoral order under which we live . . . [fierebysowing] fieseedlfinatbearsthebitterfruitof social unrest annd revolution." For example, "fine political instability in Pakistan [leading to fie formation of Bangladesh] was a direct result of . . . regional differece" (Chatterji, 1976, p.2). In view of its distributive potential , spatial or regional develcpnenthasbeenrecognizedbyfieMalaysianchermentasoneof fie important instruments of fie New Economic Policy. Specifically, regional development programs are expected to redress inter-regional imbalances, reduce rural to urban migration, stregfinen agricultural and inndustrial development in lagging regions, initiate fie establisl'ment of new growth centers, annd realize fine major objectives of fie New Economic Policy (such as eradication of poverty annd 14 modernizationoffieruralcomnunities). TteSecondandThird Malaysia Plane clearly stated fie importance of reduction of regional inequalities. Tie Fonrth Malaysia Plan (1981, p.185) reiteratedfinat "reducirgregional disparities . . . remainsan important issue in national developnent. " Spatial disparities in development between fie varions regions in Malaysia is, fierefore, an importannt development problem annd issue. As stated earlier, wifin fie inauguration of fie New Economic Policy fie finrust of Malaysia's development strategy has been on equitywifingrowfinrafierfinannonmere (economic) growth. Between 1970 end 1980 Malaysia had an impressive record of economic growth (Mahangas, 1982; Aris Ofiman, 1982). Tre Gross Domestic Product (GDP), afterhavinggrownatGpercentperannumduringfie1960s recorded a rate of growfin of 7.8 percent per annum during 1971-1980, reenlting in a rising per capita income (Fonrth Malaysia Plan, 1981). In 1971, fie GDP per capita for Malaysia was (Malaysian Ringgit) $1,172/=, while for Peninsular Malaysia it was $1,190/=. In 1980 the GDP per capita increased to $1,836/= for Malaysia annd $1.886/= for Peninnsular Malaysia (Table 1.1). Tne high growth rate was also acconpanniedbychangeinfiestmctureoffieecoomy. Table 1.2 inndicates fie substantial sectoral change of fie Malaysian economy during fie period. Despite Malaysia' 3 impressive aggregate rates of economic growth, most of fie studies on development performance on Malaysia fonnd little improvement in distributive equity (Griffin annd Khan, 1979; Cnnan, 1979; C'neo'g, 1979; Mahangas, 1982; Aris Ofinman, 1984). Creong 15 (1979, p.200), for example, fonnd finat "in spite of fie relatively rapid growth of fie Malaysiann economy, fie varions states in Malaysia donotseemtohanneanequalshareoffieeconomicprogress." Rather fie internal spatial configuration of deveth inndicated finat fie benefits of development were still largely distributed in areas in annd aromd fie mnajcr urban centers or regions relatively more developed in Malaysia. In fieir analysis of regional ecoonmic differeces in fie state of Selangor, ann annd WanAbdul Rahim (1979, p. 34) fonnd finat "signnificant differeces in economic conditions exist betweenfieurbannanndruraldistrictsof Selangcr." Theyobserved finatalfioughfiestateofSelangorhasfiehigl‘estGNPpercapita anongfiestatesinPeninsularMalaysia, suchinndicatorwasnot representative of fie less developed districts in finat state (such as Kuala Selangor annd Sabak Bernem). Tie analyses of fie temporal- spatial impress of ‘modernization' (viewed as development) in Peninnsular Malaysia from 1895 to 1969 by Leinbach (1972) revealed that fie development surface was not only clustered aromd certain urbanized areas, such as Kuala Limpur, Georgetown, Ipoh, Seremban, MalaccaanndJotoreBahru, butalsofinatfieseareasstill continuedto dominate fie spatial developnent scene. Osborn (1974a, p.361), wno 16 Table 1.1 Per Capita Gross Domestic Product, 1971 end 1980 (in Malaysian Ringgit at 1970 Prices) States 1971 1980 1) Jchore $1,083.? $1,726 2) Kedah $ 728.3 * $1,101 3) Kelantan $ 564.1 s 842 4) Malacca $ 877.0 $1,469 5) Negri Sembilan $1,144.5 $1,817 6) Pahang $1,169.8 $1,486 7) Penang $1.035.2 $2,357 8) Panel: $1,166.7 $1,583 9) Perlis $ 728.3 * $1,094 10) Selangcr $2,152.9 + $2,655 11) Terenggannu $ 614.8 $1,316 12) Federal Territory - $3,991 of Kuala Lunpnr 13) Sabah $1,302.9 $1,847 14) Sarawak $ 915.2 $1,382 Malaysia $1,172.2 $1,836 Peninnsular Malaysia @ $1,189.9 $1,886 Note: * Due to concination of data, fie states of Kedah annd Perlis had fie sane per capita GDP in 1971. + This includes fie Federal Territory of Kuala Lumpur, formed in 1971. @ This conprises fie states of Johore, Kedah, Kelantan, Malacca, Negri Senbilan, Pahang, Penang, Perak, Perlis, Selangor, Tereggannu, annd fie Federal Territory of Kuala Lumpur. Sonrce: Fonrfin Malaysia Plan, 1981-1985, pp.100—101. Table 1.2 Percent Share of fie Gross Domestic Product By Industry of Origin, Malaysia, 1970 and 1980 17 Industry of Origin 1970 1980 1) Agriculture, Forestry, annd Fishing 31 22 2) Mining annd Quarrying 6 5 3) Manufacturing 13 21 4) Construction 4 5 5) Electricity, Water, annd Sanitary Services 2 2 6) Transport, Storage, annd Oomunication 5 7 7) Molesale annd Retail Trade 13 13 8) Bannking, Innsurance, annd Real Estate 4 ) 8 9) Ownership of Dwellings 4 ) 10) Public Administration and Defece 11 13 GDP at factor Cost 100 100 Soirces: Fonrth Malaysia Plan, 1981-1985, p.11; Aris Ofinman (1982). "Growth, Equality annd Poverty in Malaysia, 1957-80." Ph.D. Dissertation, Boston University, Massachussetts. 18 mndertodn a spatial annalysis of developnent in Peninnsular Malaysia (1974b), annd contended finat alfinongh "development policy has . . . beenbeenconsistentovertinme, yet [it] has . . . certaindiscernible trends in its areal content. Development policy has been [mnainnly] city-centred in fie post-war period." This observation was endorsed by Jchari Mat (1983, pp.1-2): 'ne . . . practice of cocentrating developnent on established coreareaswasable, inalimnitedwaytostregfiennational growth: but, in a more substantive way, fie practice failed to develop peripteral and local populations economically and socially. Develcpnent studies from non-spatial perspective, such as ficse undertaken by Aris Ofimen (1984), Annannd (1983) and Geog (1979), also testify to fie fact finat, overall, fie problem of socio-economic inequality in Peninsular Malaysia has not been significantly reduced, rafier it has exacerbated over time. In his annalysis of developnent in Pmineular Malaysia from 1957 to 1980, Aris Ofirnan (1984, p.v) has made fie following coclusion: "In termns of the Kuznets' hypofiesis, at Malaysia's per capita income, income distribution snould have remained unchanged during 1957-70 but in fact inequality rose sharply. During 1971-80, inequality should have fallen slightly; instead, it increased furfier." Tle annalysis of fie time-path of Peninsular Malaysia' 3 regional growth rates in per capita gross domestic product (GDP) from 1963 to 1970 by Geog (1979) inndicated finat not only did inter-regional differeces in income exist, but also fie reduction of regional inequalities was not signnificannt over time. According to 19 Geog's study (1979, p.201), "fiere appears to have been an increasing trend towards income inequality." Annand (1983, p.273), wl'ounndertodnanalysesof inequalityandpovertyforPeninsular Malaysia from data geerated by fie 1970 Post-Enumeration Survey (PIES), needed a simnilar coclusion regarding fie inequality problem: "A detailed examination of PBS and income cocepts shows overall inequality in Malaysia to be high." 'ne persistece of spatial annd socioeconomic disparities in fie distribution of deveth is not only dysfunctional for articulating effective diffusion of development beefits, but also is clearly incommensurate wifin as well as detrimental and prejudicial to fie objective of pronoting equitable development. In fact, fie agglomeration of development in annd aromd fie mnajor urban centers or regions relatively more developed tends to perpetuate, as well as accentuate, polarized develcpnent. In turn, this gives rise to such problems as overurbanization (Hoselitz, 1957; Sovani, 1964; .Gugler, 1982), unndermrbanization (Rondinelli, Ionbardo annd Yeh, 1979), pseudo-urbanization (McGee, 1971; Rondirelli, Ionnbardo and Yeh, 1979 ) , inter-regional disparities in developnent and excessive rural- urbann mnigration. Tie cocentrated spatial pattern of development not only reflects, but also induces an unnbalanced allocation arnd distribution of national developnent resolrces. This means finat fine effects annd benefits of development efforts are, inevitably, limited and constrained by fie physical extent of fieir distribution. Since "fie spatial dimension is an important basic framework wifinin 20 which coeiderations of developnental problem must be set" (Manshard, in Boyle's, 1974, Foreword), annd fie fact finat fie spatial pattern of development inLMalaysia.appears to be focused in.areas in.and around the major'urban.centers or regions relatively more developed, this suggests that a change in.the areal configuration of development is essential and.desirab1e in.crder to promote and foster the accomplishment of a more equitable distribution.of benefits of development. Indeed, change is imperative when it is remembered that discrepancy.tn the spatial distribution.cf development also bears significant implications for the New Economic Policy's twinrprcng objectives of (1) eradication of poverty, irrespective of ethnic background, and (2) restructuring of society, so as to erase identification.of economic functions with ethnicity and.geographical location. The phenomena of regional inequalities become crucial when it is further recalled that the bulk.of the Malays, the major ethnic group ianalaysia, are prominently to be found in the peripheral rural areas, and.the Chinese, the next major ethnic group, are mostly to be fonnd in fie urban eclaves (Ariff, 1973; Jim-Bee, 1976: Bussink, 1980) . Thus, geographical polarization of developnent also has efinnnic polarization overtoes in Malaysia. As Ariff (1973, p.378) aptly remarked: "dualistic development has serious racial [ethnic], regional and political implications in fie Malaysian content." Geog (1979, p.200) endorsed this view: Regional . . . disparity in Peninsular Malaysia is conponnnded by fiefactfinatitbearsstrogracialovertoe. . . . Itis obvions, fierefore, that reducing fie gap between fie Malays annd non-Malays . . . wouldrequireareductioninthedisparities among regions. 21 In view of fie Malaysian auficrities' recognnition that peace and stability were prerequisites for development (Fonrth Malaysia Plan, 1981-1985, 1981, p.vii), a balanced socio-economic developnent among regioe is fierefcre an essential element of fie developnent planning strategy in Malaysia for achieving. Purfinermore, prior to 1970 "area" was largely a residual cocern in fie development planning strategy of Malaysia. Thisdoesnot, however, meanthatfiearealcoceptsof development were totally absent in fie plannnning deliberations annd dooments. Rafier, fie primary focus of Malaysia's development strategy finen was heavily bent towards fie pronotion of rapid national aggregatesorfinatenphasiswasmoretowardsfiesectoralperspective of developnent (Osborn, 1974b) . Problem Statement Spatial disparities in development can be reduced by appropriate plannnning innterventions (Dangschat annd Zirwes, 1982). Availability of adequate relevannt innformation is funndamental inn efforts to address the problem. However, lack of adequate, corprehensive data pertaining to fie problem of spatial development in Malaysia, especially at fie district level, cann constrain efforts to deal with fie problem. A large body of known information pertainning to fie problem of spatial disparities in development in Malaysia is available at a scale of annalysisfinatiseifiertcolargeastosuppressorcocealsomeof fie important details relating to fie problem, or too specific so that finey have limited relevance annd usefulness other finan 22 for fie specific area which such information econpass. Most of the available information on development performance, sucln as fiose contained in fie five-year plan dooments, indicates fie sectoral performance of developnent at fie state-scale of analysis. While such information is useful, it , however, lacks important details. On fine ofier hand, information mnade available by different regional studies, unndertaken eifier by private developnent consultants or by certain government departments annd agecies, also has limnited utility. This isbecausefiesestudiesarerotonlybasedondifferenttermeof referece annd objectives, but also cover only certain specific areas during different time periods. According to Abdul Hamid (l979,p. 137): "Tie varions plannnning ‘region' . . . have been . . . studied annd planned in isolation at different times by different grops of consultants. " Such innfornmation, fierefore, has limitations wifinrespecttofieconprehensivenessofarealcoverage. Conparability of such information is also constrained not only by fie different tinnne frannne during which such studies are covered, but also because different variables annd technniques enter into fiese isolated studies. 'ITe lack of adequate, conprehensive information relating to fie problem of spatial disparities in development in Malaysia colld, fierefore, pose an important constraint in fie efforts to deal effectively wifin fie problem. As aptly stated by Elyas Omar (1980, p.48): "Tre problem of policy plannnning in Malaysia is related to fie availability of information in terms of its sufficiecy, relevance annd accuracy. " 23 In fie absece of adequate data base, articulation of planning strategies would finen be characterized by irrational "muddlinng finrongh" (Lindblom, 1978): resulting fierefore in 1d _h9c_ and unnintegrated action. Actions based on limited innformnation colld, in fact, furfier exacerbate rafier finan contain fie problem. Tterefore, adequate data annd information on fie problem of fie extent of spatial disparities in development in Malaysia, especially at a geographical scale finat can provide sufficient details, is essential to facilitate efforts to address fie problem effectively. Objective of fie Study 'Iherearetwobasicrationales forpursuingfinisstudy. First, it relates to fie fieoretical proposition finat evaluation of fie reduction of inequality constitutes an important conpoent of fie study of developnent (Sears, 1972; Adelman and Morris, 1972) as well asfiefieoreticalcontentionfinatinequaliwisanecessaryand cocomitannt process of developnent, annd finat it will dimninish over time (Kuznets, 1955: Williamson, 1965: Hirschman, 1958). Tre second funndenentalreasonfornmdertakingthisstudyismotivatedbyfine Malaysiann Government's cocern annd interest over the problem posed by inter-regional development inequalities as well as its recognition finat reduction of regional inequality is also crucial to fie developnent process. As fie Fonrth Malaysia Plan, 1981-1985 (p.185) states: "Reducing regional disparities . . . remains an important issue in national developnent. " 41f 24 As mentioed earlier, fie eye-opener on fie importance of addressing fie problem of inter-regional inequality was fie oltburst of fie 1969 efinnnic violece in Peninsular Malaysia. That tragic incidentwasrotonlyonlyseenassymptommaticoffieprevaleceof unbalanced distribution of fie beefits of socio-economic development, but also clearly denoetrated fie negative consequeces as well as the danger posed by fie persistece of lop-sided developnent. Furfiermore, ensuing policy pronouncements clearly reflected fie Malaysiann Government's cocern over fie problem of unnbalanced regional development. Despite fie importance of addressing annd unnderstannding the problem of regional development disparity, fiere was little research onfieproblem. Tremedtofccusonfieproblemofregional disparities in development was fierefore imperative. Further, this was also spurred by fie Malaysian Government's disclosure that acconplistment in regional developnent was mnarginnal . Tie marginal success in reducing inter-state [regional] imbalancesrequiresareviewandaredirectionoffiestrategy for regional . . . developnent (Mid-Term Review of fie Fonrth Malaysia Plann, 1984, p.21). In light of fie importance of fie issue of inter-regional inequalities of developnent in Malaysia, it is appropriate to examine fie extent of regional inequalities fiereby facilitating a better understanding annd appreciation of fie problem. Only fien can appropriate annd sonnd remedial measures be articulated annd 25 implenented. As Ooates, Johnston annd Knox (1977, p.2) aptly stated: "Solution [to any problem] requires understanding." With fieseconsiderationsinmind, fiemnainobjectiveoffinisstudyisto enaminefiepatternofsgtialdevelopmentandassessfieextentof inter-regional ineqnalities in development in Peninsular Malaysia between 1970 annd 1980. Specifically, fie research will: (1) (2) (3) (4) ascertain wtefier fie problem of inter-regional disparities in development in Peninnsular Malaysia was reduced or accentuated between 1970 annd 1980, identify regions which are relatively more developed as well as regions which are relatively less developed, determine fie effect of using different geograhical unnits of analysis upon measnnres of spatial inequality, and, suggest policy measures to enhance efforts for acconplishing reduction of inter-regional inequality of developnent as well as in fie promotion of a more equitable distribution of fie beefits of developnnent in Peninsular Malaysia, in particular, annd in Malaysia, in general. Hypofineses Tnehypofinesesadvancedforthisstudyare: (1) (2) Spatial inequalities in development exist at bofin state and district scales of analysis; Regional inequalities in development are relatively moreacuteatfiedistrictrafinerthanfiestatescaleof 26 analysis: (3) Regional inequalities in development are relatively moreacuteforfieruralrafierfinanfieurbandistricts; (4) Regions which are relatively more urbanized tend to have relatively higler level of developnent than regions which are relatively less urbanized: (5) Regional inequalities in developnent are relatively moreacutefordistrictsinfieeastcoaststatesfinan districts in fie west coast states of Peninnsular Malaysia: (6) ‘Ite pattern of inter-regional inequalities in development has not changed significantly over time. S_ignificance of fie Sbndy Despite fie importance of unnderstanding fie problem of regional inequalities in Malaysia, research on fie subject is small-- fie exannples docunented are fie 1982 Ibuseold Well-Being Survey by fine Socio-Economic Research Unit of fie Prime Minister's Department, Malaysia; Annannd, 1983; Snodgrass, 1980; Geog, 1979; Abdul Hamid, 1979; Gnn and Wan Abdul Rahim, 1979; Ahmad Idris Mcnhd. Noor, 1983; Leinbach, 1972. Inmostcases, rowever, fiemnainfccushasbeenon inter-state variations in ecoomic annd social development mainly from aspatial perspective. While it is not disputed finat such studies can annd do contribute to some conpreension of inter-state regional variations in economic annd social developnent, finey do not, however, sled much light on intra-state regional disparities in development. 27 Efforts to reduce inter-state regional inequalities in development requiresandtninngesnponappropriateremedial actionswifinineach state, as well as in cooperation wifin fie ofier states. Understanding of intra-state regional variations in development is, fierefore, also importantinfietasktoaddressandredressfieproblemofregional inequalities. This study, fierefore, looks at fie problem of regional disparities using fie administrative districts as its unnit of annalysis. An examination of intra-state regional disparities in development is desirable to indicate fie extent of fie inequality problem wifinin fiestateaswellasinconparisontofieofierregionsinofier states. Tte information and enpirical knowledge geerated from finis studywouldnctonly fill someof fiegaps in information pertaining to fie problem of inter-regional developnent in Peninnsular Malaysia but would also modestly provide sonne innsights regarding fie extent of spatial disparities. Thus, this study also examnines regional disparities using ofier geographical units; urban and rural districtsannddistrictsinfiewestcoaststatesandeastcoast states. Understanding of fie magnitude of fie problem of spatial inequalities in development is useful, especially to planners and policy-makers, since, as Streeten (1981, p.137) aptly states: "what we know . . . enters into or models and policies." Thus, Cant (1975, p.76) has stated that "fie kncwledge that can be gained from enpirical research is invaluable." Itisalsotcpedthatthisstudywill stimulate furfier enpirical studies as well as geerate relevant inquiries and 28 discussions pertainning to fie problem of spatial inequality of developnent. Plannnning on fie basis of regions has been stated as an important strategy for articulating a more effective reduction of inter-regional inequalities of developnent in Malaysia (Mid-Term Review, Folrfin Malaysia Plan, 1981-1985, 1984, p.21). Undoubtedly, an important precondition to facilitating fie implenentation of this strategy is fie determination of fie developnent status of fie regions inquestion. Byexaminingfiepatternof spatial developmentand magnitude of regional inequalities finis study conld contribute towards making ‘plannnning on fie basis of regions' a workable developnent planning frannnework. According to Oolm and Geiger (1962, p.66): "it is never ecugh to judge fieories, decision models , etc., only by fieir logical validity; fiey must also be submitted to such enpirical verification as may be possible." In this regard, finis study will modestly enhance fie fieoretical propositions finat evaluation of reduction inequalityisanimportantconpoentoffiestudyofdevelopnentas well as finat developnent is multi-dimensional in character and fierefore requires fie use of multiple indicators of developnent. Perhapsamoreimportantcontributionofthisstudyistotestfie theoretical proposition that inequality in developnent will be reduced over time. 29 KeyOoceptsOf'neStudy Developnent, inequality, and spatial inequality constitute fine finreekeycoceptsofthisstudy. Admittedly, fieyareconcepts not easily defined. Noefieless, to facilitate their understanding as usedbyfinisstudy, itisfinerefcreexpedientanndusefulto coceptualize fiese terms -- ficugh it should be clearly cautioed finat itwillnotbepossibletoarrive atanexactand precise definition of finese terms. Developnent is an intriguing and conplex cocept. Being abstract and ambiguous, development means different finings to different people. Its coceptual meanning also varies in different contexts, culturally (Khan, 1981; Horn, 1984) as well as tenporally (Mabcgunje, 1981). chwever, alargeconsensusseesdevelopmentasanormative cocept, almost a synonym for improvement (Okunn and Richardson, 1961; Seers, 1972; Oolmnan and Nixson, 1978; Abdul Hamid, 1979: Honjo, 1980; Etocshan and Misra, 1980; Bryant and White, 1982). Tnere is also growingcoeensnnsthatecoomicgrowfinperseisrotdevelopment (Pafinaik, 1982; Vyasulu, 1977; Weinannd, 1973; Colm and Geiger, 1967; Bryannt and White, 1982). This is not only because growth involves "merely a set of increases in quantities produced" (Boudeville, 1966, p.168-9), but because fie cocept of developnent, in reality, means fillCh more finann finat: it econpasses all tangible and intangible aspects of life. Wifin fiese qualifications, developnent in this study is taken to 30 mean a continnuous, multidimensional process innvolving positive change or improvement in bofin fie quantitative (material) and qualitative (nun-material) aspects of life for fie ultimate well-being of fie people (Okun and Richardson, 1961; Harbinson, 1967). Thus, it has been contended finat "developnent means fie developnent of people" (Nyerere, 1968, p.123; Weinand, 1973; Leupolt, 1977). Tl'e "qualitative transformations must occur cocurrently wifin quantitative increases" (Oolm and Geiger, 1967, p.272). Also, in this study, developnent is operationally evaluated in terms of certain, selective tangible indicators of development, primarily based upon fieirfieoreticalpertinnecetodevelopnentandalsoupon consideration of availability of data. It should be clear finat finese indicators are only pointers of developnent. ‘I‘hronghont fie study, fie term "developmenn " is often used interchangeably wifin such terms as improved socio-economic well-being and improved level of living. According to fie Webster‘s Third New International Dictionary, 1981), inequality means ‘fie quality of being unequal or uneven'. In fie content of developnent, inequality in developnent therefore connotes fie unequal or uneven distribution of developnent beefits. Since it is geerally agreed finat fie principal and ultinmate beeficiary of developnent is fie people, as well as considering fie fact finat fie population is not unniformly distributed in a geographical space, inequality in developnent in fie content of this study fierefore means fie uneven distribution of development relative to the proportional geographical distribution of people (Adarkwa, 1982). 31 This implies finat, ideally, fie beefits of development should be distributed in accordance wifin fie spatial distribution of fie population. Anny deviation from this ideal pattern will, for fie purpose of finis study, be considered inequality. Also, it should be made clear finat statenents cocerning fie undesirabiliiy of fie persistece of inequality in developnent in Malaysia does not imply thatwhatissonghtinredressalmeasuresisactiontowardscreation of equality in developnent. Equality as an objective of development is realistically an impossible task to acconplish, let aloe to pursue; since fie concept of equality as applied to mannkind implies finatallpeopleareequalandfinatfieyshouldhaveequalstatus, wealth, and influence (Jumper, Bell and Ralston, 1980). Tlerefore, fie cocept of equality is, at best, an idealistic goal. Spatial inequality relates to fie differeces existing between different areas based upon spatially-defined and conparable variables. Spatial inequaliiymnayrefertobothadesiredandtoanundesired difference. Variation in fie landscape is geerally considered desirable, while large or growing differeces in coditions and level of living are deemed undesirable (Folmer and Oosterhaven, 1979). In fie content of developnent, spatial inequality connotes an undesired differece. It implies not only fie unbalanced distribution of developnent beefits, but also fie maldistribution of resonrces for development. An undesired spatial differece is, fierefore, also viewed as spatial inequity. "Spatial inequality associated with . . . an undesired differece . . . [is] defined as spatial inequity" 32 (Folmer and Oosterhaven, 1979, p.1). Equitable developnent, fierefore, implies actions toward mitigating fie problem posed by spatial inequity. This is pertinent as "development . . . nnow stresses equity as well as growfin" (Edwards and Todaro, 1974, p.25). Throngl'outthisstudy, fietermspatial inequalitywill alsobeused interchangeably wifin terms such as spatial variation and regional disparity of development. Qggannization Of Tte Study This dissertationis organnized into six chapters. Gnapter One introduces spatial inequality of developnent as an important issue of development -- bofin in developing countries in geeral and in Peninsular Malaysia in particular. Statement of fie problem pronptingfinisstndyiscontainedinthisinitial chapter. This chapter also includes fie objective, significance, and definition of keycoceptsoffiestudy. Researchquestionsandhypothesesare presented inn finis chapter. Gnapter T'wo provides a review of fie literature and fieoretical constructs pertaining to development and regional inequality in developnent. Tte review provides fie fieoretical rationale for addressing fie problem posed by fie problem of spatial inequality of developnent. GnapterThreediscussesfiemefiodologyoffiestudy. It ontlines fie focus of fie study, fine variables utilized, fie unit of annalysisused, andfieproceduresandresearchtechnniquesusedfor annalysis of data. 33 GnapterFonrpresentsfiefindingsoffiestudy,baseduponfine annalysesofdatausedinfiestudy. Italsoattenptstoanswerfine researchquestionspcsedaswelltotestfiehypofiesesthathavebeen advancedforfiestudy. Gnapter Five examines policy and planning implications of fie findings of fie study for Malaysia and directions for policy guidelines. GnapterSixsummarizesfieoverall findingsofthestudy, annd providessuggestionsforfutureresearch. mm REVIEW OF THE LITERATURE ON DEVELOPMENT AND REGIONAL INEQRLITY This chapter reviews some of fie relevant propositions and studies from fie literature which bear pertinence to developnent and to regional inequality in development. Trere are five main objectives: (1) to highlight problens associated wifin definition and measurenent of developnent, (2) to emphasize inequality as an importannt issue of developnent, (3) to rationalize why fie spatial aspect of developnent deserves appropriate consideration in deliberatios on developnent, (4) to highlight relevant fieoretical propositions and related enpirical studies that are pertinent to developnent and to inequalities in development, and (5) to derive research questions and hypofieses for this study. Meaning, Dimensions and Goals of Deveth A precise definnition of developnent is difficult (Bruton, 1985; DeKadt, 1985). Tte literature is abomd with different definitions of developnent, reflecting lack of unanimity among scrclars as well as practitioers on what developnent means. The difficulty in arriving at a unniform definition of development can be attributed to its complexity as a cocept. This, inevitably creates ambiguity on what 34 35 development entails. Thus, fiere are varying and subjective notions of developnent (Furtado, 1977). The result is fie emergece of a myriad of perceptios of developnent (McGranahan, 1972; Streeten, 1981; Bryannt and finite, 1982). Tne varions interpretations of developnent are actually manifestations of different perceptions of developnent. As DeKadt (1985, p.551) aptly points wt: "based on or value judgements, we can give it [development] a meaning." Earlier, Goulet (1969) has enpressedasimilarview, assertingfinatstandardsofgoodlifeand good society are based on what fie society perceives. Sears (1969) hasalsoopinedfinatdevelopnentisancrmativecoceptbasedonvalue judgements. Oolmnann and Nixson (1978, p.2) have reinforced finis notion of developnent when fiey state: "developnent cann be considered . . . asaprocessofimprovenentwifinrespecttoasetofvalues . . . [which] relates to a desired coditions in society." Essentially, fie varions perceptions of developnent can be viewed into five major coceptualization of development: (1) developnent as growth (2) developnent as change (3) development as human well-being (4) developnent as reduction of inequalities, and (5) development as liberation from depedecy. All fiese cocepts identify development with positive rafier finan negative characteristics. Trere is also geeral agreement among finese different perceptions finat developnent is a process that produces beefits rafier finan disbeefits. 36 Treperceptionofdevelopnentasgrowfinisoeinwhich developnent is perceived to be rapid and sustained "linear increments inasetofvariablesinasocietywhicharecharacterizedbybeing measurable, especially in moetary terms" (Mabogunnje, 1981, p.334). Tieresultisseenascontributingtoriseinrealontpnntperleadas well as attedannt shifts in technological, economic and denographic characteristics (Dos Santos, 1977). This notion of developnent gives priority in fie developnent process to increased comnodity oltput rafier finann human-beings innvolved in fineir production. This perception of developnent is usually identified wifin fiose wno View developnent mainly in terms of economic coditions. Hece finis View of development is often referred to as economic transformation (Fortes, 1976). Dadzie (1980, p.59), for example, states: "developnent implies profonnd change in fie economic arrangenents within as well as among societies." Tie coceptualization of development as change is oe in which developnent is seen as innvolving shift or adjustment from a less desirable to a more desirable codition or state (Okunn and Richardson, 1961: Seers, 1972; Colman and Nixson, 1978; Higgins, 1980; Bhooshan and Misra, 1980). As Streeten (1972, p.30) states: "developnent as an objectiveanddevelopmenntasaprocessbothembraceachangein fundamental attitudes . . . and in social, cultural and political institutions" (also see Schramm and Lerner, 1976: Inayatullah, 1967). Infinisrespect, twomajortypesofchangehavebeenconceived: (1) change wifinin fie national system or society; such as from 37 tradition to modernity, agricultural to industrial societies, or less urbanized to more urbanized societies: (2) change wifinin fie individual in values and attitudes. Tte first type of change sees developnent as transformation which societies undergo. Human societies are envisioed as growing organierswhichpassedfinroghaseriesoforderedandinevitable stages. Developnent is coceptualized as gradual, qualitative passage from less to more differentiated social forms. This occurs finrongh processes of ever more conplex specialization and functional interdependece. Throlgh finem, social roles are transformed to approach modern standards of unniversalism, specificity, and achievenent (Portes, 1976, p.63). Drawing on Boeke's (1953) cocept of social and ecocmic dualism, fiemodernsectorisrationalizedasbeingresponsivetochange, while fietraditionalsectorisseenasunresponsivetochange. Lewis (1954) calls fiese industrial and agricultural societies, while Hoselitz (1960) views finem as developed and underdeveloped societies. Lerner (1965) and Levy (1966) refers to finem as modernity and tradition. This perception is usually identified with fie coceptualization of developnent as social or socio-economic differentiation. Smelser (1966, pp.110-111) provides a clearer picture of what finis perception of development means. Whenn we enploy fie term [development] we usually have at least fonr distinct but interrelated processes in mind: 1) In fie realnm of technology, a developing society is changing from simple and traditionalized technniques toward fie application of scientific knowledge. 2) In agriculture, fie developing society evolves from subsistece farmning toward fie commercial production ofagriculturalgoods. . . . 3) Inindustry, fiedeveloping 38 societyundergoesafiansitionfromfieuseofhumanandanimal powertowardindustrializationproper . . . . 4) lnecological arrangenents, fie developing society moves from fie farm and village toward urban cocentration. This perception of developnent represents a shift from a comnodity- oriented emphasis to a lumen-oriented enphasis. Alfinough a less excessively narrow ecoonic interpretation of developnent, Mabogunj e (1981, p.38) asserts finat finis view of developnent is closer to growth-oriented cocept of developnent, since it "innvolves principally rowtomakefiepopulationofaconntryunderstandandacceptfienew rules of fie ecoomic growth genre." Tlesecodtypeofchangeiswnereindividualsareseenas acquiringvaluesandattitudes finataredeemeddesirabletofie developnent process as well as for nation-building. This notion of development is represented by Weiner (1966), wlo feels finat fie starting point of any definition of development should deal wifin fine character of individuals. T‘re sanme aufior observes finat "alfiough fierearedifferencesemogsocial scientistsastohowvaluesand attitudes cann be changed, it is possible to speak of oe sctool of fiought finat believes that attitudinnal and value changes are prerequisites to creating a modern sociew, ecoomic, and political system" (Weiner, 1966, p.9). McClelland (1963, p.17) has even ventured to state: "it is values, motives, or psyclological forces finat determine fie rate of ecoomic and social development." According to Innkeles (1966, p.138): "fie ideal of development requires fie transformation of fie nature of man -- a transformation finat is bothameannstofieedofyetgreatergrowfinandatfiesametime 39 oe of fie great eds itself of fie developnent process" (also see Lerner, 1965: Fortes, 1974). Tie value-enactment view of developnent is controversial. Critics have noted, for enenple, its neglect of international and political linnkages. As Fortes (1976, p.71) points ont: "individual actionishighlycoditioedbyexternalsocial arrangenents . . . . [Furfiermore] societies are not simple ‘additive' sum of individual members." Tie finird major perception of developnent is oe which views developnent as improvement in human well-being. Nyerere's (1968, p.123) view finat "developnent means fie developnent of people" and Honnjo's (1980) perception finat developnent stould be geared to fie bettermentofhumanbeingsasawtole are, perhaps, indicativeand representativeoffiegrcwingconsensusthatfiefnmdamentaland ultimate objective underlying developnent is fie people. Hece, Misra (1980, p.21) has asserted finat "all developnent processes aim at human welfare" and Mabcgunje (1981, p.236) has conteded that "improvement in fie ‘quality' of fie population is, of conrse, what development is all abont" (also see Weissmannnn, 1968; Webster, 1980). Even Schumacher (1973) has opined finat development does not start wifin goods but, rafier, with people (also see DeKadt, 1985). Such assertions are also in line wifin Seers' (1972) view that development means creating coditions for fie realization of fie human personality as well as with Ellis' (1980) contention finat fie ultimate purpose of development is to secure a better quality of life for fie people (also see 4O Lenfiek, 1980). As Goulet (1968, p.387) has eloquently noted: Although deveth implies econonic, political annd cultural transformations, fiese are net ends in finenselves but [finey are] indispeeable neans for enriching fie quality of human life. Developnent as reduction of inequalities is fie fourth major peroeption of developnent. Sears (1972) annd Adelman annd Morris (1972) viev reduction of inequalities as an important criterion of developnent. Such an assertion is logical sinnce fie overriding purpose of developnent is, normatively, to bring about desired, positive changes (Gore, 1984; Cant, 1975). Implicit in this nnotion of developnent is an equitable distribution of developnent beefits and opportunities. Tle fifth major perception of developnent is related to its conceptualization as liberation from internal annd external dependency. As Fortes (1976, p.77) notes: "developnent consist . . . of liberation front ecternal control annd fron fie internal structures of inequality which it prototee" (also see Frank, 1967). Dependency can be defined asasituationinwhichfieeconuyofcertainconntries (orparts fiereof) is conditioed or dictated by fie developnent and expansion of anofier econony to which fie former is subjected (Dos Santos, 1970). This is based on fie notion that contemporary developnent is not a matter of autonomous change but are composed largely of exchange and confrontation in an integrated world or national system (Wallerstein, 1974; Fortes, 1974; Sunkel, 1974). Involvement in fine Nev International Economic Order annd pursuit of fie principles of national reeilience and self-reliance are manifestations of fie 41 effortstodealwifinfiedwendencyissue. 'ne edstenoe of various conceptualizations of developnent testifiestofiefactfinatdevelopmentconpriseeanfleoonpassesa mmberofdimensionsoraspects. ThisisclearlyindicatedbyTodaro (1977, p.62) who aptly states that developnent is a "multidimensional process involving major changes in social structures, popular attitudes annd national institutions as well as fie accumulation of economic grcwfin, fie reduction of inequality, annd fie eradication of poverty." A similar view has also been expressed by fie Brandt's Commission (Brandt, 1980, p.40): Develqmentismorefinanfiepassagefrompoortorich, from traditionalruraleconomytoasophisticatedurbanoe. It carries wifin it not only fie idea of economic betterment, but also of greater hmnan dignity, security, justice, and equity. It should be noted, rowever, that fie various dimensions of developnent are not mutually exclusive of each ofier, rafier they are interdependent. Tie fact finat developnent has been conceived in many different wayssuggeetsfinatfiereexistsnocongruityamogfieorists, aswell aspractitioers, astowhat development exactlymeansandrowit should occur (see Gore, 1984). As such, it is not surprising that fierehaveemergedanndensuedallkindsofdebatesregarding developnent -- such as growfin versus distribution, growfin versus growfin wifin equity, top-down versus bottom-up approach to development, growth-pole versus agropolitan developnent, centralization versus decentralization, people prosperity versus place prosperity. 42 Notwifinstanding fiese polemics, scholars geerally agree that no single factor can adequately account for a process as complex as developnent. Ttereisalsobroadandgrowingagreementinfieliterature behindfienotimfietgrowfinisnotsynonymunswifindeveloplent (see Fatnaik, 1981: Mabcgunje, 1981; Vyasulu, 1977; de Souza and Porter, 1974; Weinand, 1973). Sears (1972) pioeered fie notion finat economic growth could not be equated wifin developnent. His question "Why do we confuse developnent with economic growth ?" is reflective of fie. confusion that exists between growth and developnent (Seers, 1972, p.21). As Datoo annd Gray (1978, p.252) observe: Tte significannt distinction between ‘growfin' and ‘development' is becomingmorewidelyacceptedinfiefaceofmomtingevidence finatrapideconomicgrowfinisnotnecessarilyaccompaniedbyfie structuralohangaeessential forsocialandecoomicprogressto beefit society as a whole. Distinction between development annd growfin is crucial sinnce fie way fineee phenomena-are perceived bears significant influence not only uponfiewayinwhichdevelopmentisexecuted, but alsoinfiepattern annd nature of spatial development that could emerge. As Mabcgunje (1981, p.334) points out: "growth has tended to be concerned with linear increments in a set of variables in a society which are characterized by being measurable, especially in monetary terms." On fie ofierhand, "development, while it embraces . . . growthgoes beyond it to involve changes in fie relations between various classes insocietyendbetweenfinemendfieenvironentalresourcesonwhich 43 fiey depend" (also see Bondeville, 1966). Earlier, Bonlding (1956, cited in Mabcgunje, 1981, p.334) has remarked: "development involves rotonlychangesinfieoverall sizeoffiesystembutalsoinits conplexiiy." 'Itereisalsobroadconsensusinfieliteratureondevelopnent finatnoabsolutestandardenistsforspecifyingwhattypeofchange constitutes developnent: since fie criteria will be based on different preferences, problems or needs of fie society or culture. Tte criteria of desirable change also varies for different time periods (Ooates, Johnston, annd Knox, 1977). As Gore (1984, p. 241) aptly pointsont: "development . . . intermsofbothrietoricandactual policy measures, varies considerably between states" (see Frankel, 1952). Since problems annd needs of different nations also varies in nature annd mnagnnitude, it would not, fierefore, be appropriate to gauge developnent performannce of any conntry, which are guided by different developnent goals annd objectives, in terms of developnent performance criteria as used by ofier countries. This, McGee (1974, p.32) contends, is because developnent is not a simple unilinear change. McGee (1974) end Khan, M. R. (1981) have, fierefore, asserted that developnentin'lhirdWcrldconntries shonldnotbevienedintermsof development as happeed or prevailing in fie more advannced countries, such as Weetern Europe annd fie United States; where, conventionally, developnent has been viewed in terms of change from tradition to modernity: a shift from agricultural to industrial societies; from lowly urbanized societies to highly urbanized societies . According to Bruton (1985, p.1103): "in both fie literature annd in practice, developmentcametomeanareplicationoffieWest . . . [Infiese comtries] fie underlying fineme of deveth has been to imitate fie Westasquicklyaspossibleintermsoffieformandcontentoffieir economic [development] performnannce. " Sinnce fie process of developnent snould be indigenous and unnique to each society, it has, fierefore, beensuggestedfinatdevelopingcomtrieemustfindandpursueaconrse of developnent which reflects fieir own peculiarities, capacities and style (Forbes, 1976). As Soedjatmoko (1971, cited in Hunter, 1972, p.12-23) has stated: "each nation will have to develop its own vision of fie future, ont of fie materials of its own history, its on problems, its on natural make-up" (also see Friedmann, 1980; Bruton, 1985). Illich (1969) has goe to fie extent of calling on developing nations to abandon fie model offered by already more advanced oonntries. From fie varions conceptualization of developnent, it can be seen thatfiegoalsofdevelopnentcanbesumnnedupinternmsof fostering and achieving (1) improved quality of life, (2) increased (material) productivity, and (3) desirable values and attitudes. 'I‘Iese are pursued with a viev to achieve ‘independence', progress and stability (which includes national unnity and defence). These broad goals of development capture bofin fie tangible (or quantitative) and intangible (or qualitative) aspects of life. Ttese goals are not independent of each ofier, refiner fiey are inter-related to oe annofier. These five major perceptions of deveth are evident in Malaysia's developnent policy. Tre pursuit of economic growth 45 programs in Malaysia's developnent strategy clearly manifest fie incorporation of fie gronnfin perception of developnent. As fie Fonrth Malaysia Plan (1981, p.3) states: "Rapid growfin . . . carries wifin it fiepromiseofstnucturalchange, fiecreationofamoderneconomy, and fie geeration of enployment opportunities in productive activities." Programs to achieve modernization, pronote urbanization and transform agriculture from subsistence to comnercialization reflect fie perception of deveth as change (wifinin fie national system). Tre statement finat "furfier progress wifin stability in development will require fie adoption of values which are progressive andcoeistentwifinfieneedsofamodernizingandindustrializing plural socie " indicates fie inclusion of fie perception of developnent as onenge in values and attitudes (Mid-Term Review of the Forth Malaysia Plan, 1984, p.13). Wifinin fie Malaysian context, fie social well-being notion of developnent also exists. 'Iun Haji Abdul Razak (1973, p.7), fie secod Prime Minnister of Malaysia from 1971-1975 and also popularly hailed as ‘Malaysia's Fafier of Developnent', has stated: while economic development is a very basic and significant step in national developnent, we must not forget that it is only a means towards a higter objective, i.e. of creating a better socialorderinwhichonrpeoplecanenjoyahigher standardof living, peace and happiness. This vie: of developnent has been reiterated by fie Minister of Agrionlfinre, AnwarIbrahim, duringaspeechtofieFarmers' Organnizationin'I‘regganuinMarch, 1986. 'ITeMinistermadefiethe remark finat ‘fie [Malaysian] Government is committed to bringing 46 developnenttofiepeopleandtonpgradingfieirlivingstandards' ('I‘re New Straite Times, Malaysia, 1986). Tie notion of developnent as reduction of inequalities is clearly evident from fie New Economic Policy's objectives of eradication of poverty and restrncturing of society. Varions policy statenents in fie official five-year plan regarding programs of narrowing inter— regional inequalities also indicate finis notion of development. Malaysia's involvement in fie drive for fie New International Economic Order as well as ter adoption of national resiliece and self-reliance concepts are clear manifestations of fie incorporation of fie perception of developnent as liberation from depedecy in Malaysia's deveth policy. For major characteristics are apparent regarding development: (1) developnent is nnornmative -- it is cocerned with progress towards desired goals: (2) developnent is multidimensional -- it is cocerned wifin all aspects of life and fie varions needs of people and society; (3) developnentisunique foreachconntry-- itisconcerned with a partionlar set of goals formulated at a particular point inn time, or, in ofier words, different nations have different styles of developnent; and, (4) developnent must be a coherent process -- if a variety of goalsaretobesoghtsinmnltaneouslyfieymustfirstbe recociled with each ofier. 47 Infinisstudydevelopmentisvieredasamultidimensionalprocess involving all fie major perceptions of developnent as mentioed. Growth is a necessary but not a sufficient condition of development by itself. So, too, is social development. Meaning and Dimensions of Ineqnality Like development, inequality is also a cocept that is difficult to define. 'I'here also exist different perceptions of inequality (Seers, 1973; Atkinson, 1978). Essentially, fie different perceptions of inequality revolve aromd two major viewpoints: (1) inequality as a natural phenomenon (2) inequality as an artificial phenonenon. That inequality implies fie quality of being different seems to beacomunfineneofmostliteraturesthatdealwithfiequestionof inequality. Geerally, fiere is consensus in fie literature that inequality due to interent characteristics, as between man's reight and weight or between different environmental features (deserts, plains, mountains, etc) are accepted as natural and inevitable. What generates controversy is inequality finat is fie ontcone of differences in (1) omertunity, (2) treatment, or (3) benefits (Cans, 1972; Atkinson, 1978; Hoe, 1982). Inequality of finis type is regarded as artificial and undesirable. Also, inequality that is geerated finrongh differeces in opportunity, treatment, or benefits raises problems and controversies because of its derogative implications . 48 Since developnent undertaken by nations are conscions efforts aimed at acconplishing desirable ends, inequality finat prevail within such developmental content implies not only stortfalls in develogment strategies, but also lack of developnent opportunnities and beefits. Wilson (1966) views inequality as ‘fie fact of occupying a more or less advantageons position'. To Waldman (1977, p.229) "inequality is . . . fieententofdisproportionbetweeneachshareoffiningsheld andfieproportioneachcategoryorrolderconstitutesoffietotal nnumber of categories or folders." In fie content of developnent, inequality fierefore connotes fie unequal or uneven distribution of developnent opportunities and beefits. In fie context of developnent, inequality can be discussed from different perspectives: (1) income distribution (Kuznets, 1955; Williamson, 1965; Atnlnmwalia, 1974; Atkinson, 1975; Ahluwalia, Carter and Onenery, 1979); (2) economic growth (Rodan-Rosenstein, 1943; Lewis, 1954 and 1955); or (3) socioeconomic developnent, level of living or social well-being (McGranahan, 1970; Drenowski, 1970 and 1974; Knox, 1974; Smith, 1974; UNRISD, 1976). Foreachoffineseaspectsfierearealsovarionsdimensionsof spatial configuration of inequality upon which discussion and analysis cann pursue -- global, supra-national [groping of conntries such as fie Association of Sonth-East Asian Nations (ASEAN)] , inter-national , intra-national (i.e. between sub-national areas, such as provinces or 49 states), inter-regional , inter-urban, intra-urban, inter-rural , intra— rural, inter-city, or intra-city (Slater, 1977; Wood, 1977; Colman and Nixson, 1978; Cole, 1981). Thus, fie way in which regional inequality indevelcpmentisperceiveddependsalsoonfiespatial frameworkused to assess it. As Snmifin (1982, p.9) states: "Each nation, region and city has its own distinctive pattern of inequality." Whatever fie wates finat ensue on developnent -- wtefier it is growfinversusdistribution, top-downversusbottom-up strategyof developnent or centralization versus decentralization -- fie fundamental questioe finat arise essentially hinge on fie status of fie people. were specifically, what is fie status of fie people vis- a-vis fiese issnes? As, fieoretically, development implies fie geeration of beefits and greater opportunities for advancement, any discrepancy in fie delivery or receipt of beefits and opportunities for (furfier) advancement connotes and creates a situation where inequality is said to prevail. Such a situation is deened to be undesirable for fie acconplishnent of development goals, not only because it imposes artificial constraints upon fiose wto are affected butalsobecausefiecontinmedpersisteoeofsuchaptenonenm provides a ready-made ammunition for provoking potential social tension and conflict (Chatterji, 1976; Smith, 1982). Social disharmony and political innstability not only conld slow down fie pace of developnent, but also conld, if it deteriorates, stifle and even negate fie developnent process. No less important, disparity in developnent also implies an allocation or distribution of developnent 50 resonrcesfinatignorefieimportanceofaddressingfieinequality issue. Hece, fieexistence, andmostofallfiepersistence, of inequality as well as its attedant implications to create polarization between people and places not only challeges or moral sense, butalsoraisesfiequestionoffiefairnessandjustnessof fie political and economic order which prevail (see Harvey, in Smith, 1982, p.1). Indeed, "inequality in its various forms is oe of fie most serions problems facing fie contenporery world" (Snmifin, 1982, p.7). 'netermsinequaliwandinequityhavealsoreceivedattentionin fie literature relating to development. Tre use of finese terms have oftenbeenconfusing. ‘I'rereareviews finat spatial inequalitycan referbofintoadesiredandtoannndesireddifference. Forexample, differece in fie landscape is geerally regarded as desirable, while spatial differentiation in living conditions are geerally considered as undesirable. In making such distinction, Folmer and Oosterhaven (1979, p.1) have clarified: "Spatial inequality . . . describing an undesired differece, will be defined as spatial inequity." It has alsobeenstatedfinatinequalitycanbeevaluatedintermsoffie deviation measured from a hypothetical state of perfect equality, applying such statistical technniques as Gini coefficient and Lorenz onrve. In contrast, as Hoe (1982, p.68) points ont: "‘equity' can onlybeassessed, notbysuchstatisticalmefiodsbutbyappealingto subjective values or efinical judgements." Apart from its unnefinical connotations as well as its implications to create polarization between people and places, fie undesirability 51 of inequalityindevelcpmentinMalaysiaisconpomdedbyfiethreat of comunist insurgecy. The persistece of inequality in developnent conldbeenploitedbyfiecomunisttoprovokedissesionandstir instability. Since national nmnity and security are important fomdations for stability and progress, reduction of inequality is fierefcre an importannt developnent issue and objective in Malaysia. In finis study, inter-regional inequalities in development is viewed as fie uneven distribution of developnent opportunities and beefits relative to fie proportional spatial distribution of fie population. Space and Development Friednennn and Aloso (1964, p.1) have conteded finat "regios and space . . . arenecessarydinnensiosoffiefieoryandpracticeof economic developmen " (also see Richardson, 1973; Wood, 1977). Society, it has been fieorized, is spatially organized -- in fie sese finat human activities and social interactios are ‘space-forming' as well as ‘space-contingent'. As Soja and Tobin (1979, p.158) explains: "fieyareqaace—forminginfinatfieymrktoshapeandstructurehmman interactioninspace . . . . [and] fieyarespace-contingent. . . [in finat] fieir space organnizing influence is itself shaped by fie enisting spatial frenework" (also see Friedmann and Sullivan, 1972; Coates, Jdnnston and Knox, 1977, p.3; Mabcgunje, 1981, p.51). This point has also been coerently stated by Hilrorst (1968, p.21): 52 Humanbeingsforfieexecutionoffieiractivitiesrequire space. 'nese activities are of differing nature and include at least fiose of a public administrative, economic, political, recreational and social character. Tre relationships resulting from fiese activities will necessarily have spatial dimensions. Since society is spatially organized, Friedmann (1972, cited by Soja and Tobin, 1979, p.157) has conteded finat "fie developnent process will also be influenced by fie existing pattern of spatial relation and fie dynamic tensions finat will result from finem." Friedmann and Alonso (1964, p.1) have furfier reinforced fie importance of fie spatial aspect of developnent wren fiey state: "not only must decisiosbemadeonrowmuchascarceresonrcesshallbeallocated to a given purme, but also on wtere investments [will] take place." In claiming that each activity requires a locational decision, J.P. Lewis (1964), acknowledged fie importance of fie spatial aspect of development. Space has been viewed as an important and an implicit factor in any study of developnent (see Wood, 1977). In terms of developnent (as distribution of beefits), two fieories can be identified as being pertinent to spatial developnent. Tlese relates to: (l) Hagerstrand's Innovative Diffusion Tleory (1967), and (2) Friedmann's Tleory of Polarized Growth (1966). Hagerstrand'sworkdescribesthespatial stagesinfiespreadofa numberofnewi$asandtechniques. Itprovidesausefulunnderstanding ofhcwdevelcpmentchangescanoccurinspaceovertime. Asa geographical cocept, Hagerstrannd's diffusion process occur dowm fie hierarchy of cities (Bradford and Kent, 1978). On fie ofier hand, Friedmann's fieory of polarized developnent is relevant as it 53 demonstratestowandwhyspatial inmbalancesindevelopmentoccurand row eventually fie regios of a nation are gradually integrated over time. Ofier fieories which indicate fie importance of fie relationship between space and human activities include: (1) Christaller's annd Losch's Central Place Tteory (Bradford and Kent, 1978) (2) Von 'Ihunen's Agricultural Land Use Tleory (Hall, 1966) (3) Weber's Industrial Location Tneory (Snmifin, 1971) (4) Burgess' Oocentric Zoe 'I'neory (Park, 1925) (5) Hoyt's Radial Sector Tleory (Mayer and Kohn, 1959) (6) Harris-Ullmann's Miltiple Nuclei 'I'teory (Mayer and Kohnn, 1959). Ttefirstfinreefiecriescosiderfieorganizationofeconomic activities wifinin an hypofietical plain, while fie latter finree fieories deal wifin fie organization of space for relevant land-use activities wifininn an urban setting. Despite fie recognition finat fiere exist close inter-relations between spatial forms and social processes (see Mabcgunnje, 1981; Smifin, 1982), fie spatial aspect of developnent has not received as much attention and cosideration in treatment of developnent (Soja, 1976). This relates to fie fact that fie adoption of fie growth model of developnent which focus deveth in sectoral rafier finan spatial terms. In reference to finis point, Higgins (1980, p.v) has remarked finat "sectoral plannnning had been tried in ‘fie fifties and sixties, along wifin fie application of growfin models, . . . [but] it has not worked very well." Katchanmat (1978, p.144-45) also states: Tiemajcrweaknessofmostofmostnational econonmic development planning lies in fie fact finat fie developnent plann is carried on only in global and sectoral orientatios. Tie . . . allocationofresonrcesismadewifinlittleorno cosideration of fieir location. Annmderstandingoffierelationshipbetweenspaceand development is fierefore essential for fie annalysis of fie spatial aspect of developnent, as well as for fie formulation of appropriate spatial policies finat can effectively influece fie development process. Heoe, Ononkgil (1981, p.57) has stated: "If spatial policies are to be effective, it is essential finat fie relationship between space and development be articulated properly." Since "societies occupy territories" and in reality fie level of living vary accordingtowlerepeople live, fienngeographical spaceandfie organnizationof lifeinfinisspacehaveabearingonwl'ogetswhat (Ooates et a1, 1977, p.3; Smifin, 1982, p.18). Tre importance of fie spatial aspect of developnent is furfiner endorsed by Coates et al. 1977, p.5) wto remarked: "If we do not expect to discriminate against people on fie basis of race, religion, color, or social class, fienn neifiershouldwediscriminate againstpeopleonfiebasisof location." Manshard (in Hoyle, 1974, in Foreword) has, fierefore, asserted finat "fie spatial dimension is an important, basic framework within which developmental problem must be set" (see Rhoda, 1982, p.4). 'I're terms space and region deserves amplification, since fiey can have different connotatios depending upon fie context in which 55 fiey are Wind. According to Boudeville (1966), space can mean eifier geographic, econonic or mafiematical space. Geographic space is usually identified with geograplers. This dimension of space isoewhichviewsfierelationshipandactivities ofmaninfie natural environment. Econnic space, an entity of fie econonnists, relates to fie environment in which economic variables or activities takeplace. Bothgeographicspaceanndecononicspacecanbe regionalized. However, mafinenatical space, derived on fie basis of logic, is entirely abstract and is geographically non-existent. In referece to fie latter, it has fierefore been stated finat "space is not necessarily a region" (Patnaik, 1982, p.27). Regionisusedtomeanterritorial frameworkwhichcanbevieved atdifferent scales -- dependinguponwhatfiespatial frameworkisin referece to -- global, supra-national , national , sub-national , rural , or urban (Chatterji and Nijkanp, 1983; Pafinaik, 1982; Cole, 1981; Colman annd Nixson, 1978; Wood, 1977; Slater, 1977). Such a numerous spatial notion of region has led Perloff (1971), at oe point, to express scepticism regarding fie practicality of regional planning. Since space annd region can have different connotations, sore conceptual clarity is needed. In referece to developnent, space as usedinthisstudy, referstogeographical space. However, "for . . . Plannninngcrpolicy (purposes) . . . fiecoceptofregionisnore useful finan fine cocept of space" (Patnaik, 1982, p.27). This is because fie concept of region tends to be associated not only with mlatively more defined territorial framework, but also with territorial contiguiw. This has been endorsed by Bandnan (1975) who 56 states finat Soviet scientific literature on regional research have usedfiewcrdterritorynoreoftenfinanspacetoelphasizereferece toaspecificterritory. Exceptasuseinageographiccontext, space is abstract (as mafinenatical space) and have no defined territorial boundary (as econonic space). Fron fie perspective of developnent policy and planning, fie coceptualization of region as territorial framewok at fie sub- national scale is more realistic and pragmatic; since region at such scaleiswithinfiede jurecontnol andplrviewoffierespective government. Also, according to Misra annd Prantilla (1983), regional planning at fie sub-national level provides fie necessary linnkage for fie attainment of integrated national developnent . I-bwever, region coceptualized as territorial frenework at fie global or supra- national scale may not be as useful fron developnent policy and planning perspective unless comnitment by all fie governmental parties cocerned is possible to deal with fieir ‘regional' problers. As observed by Perloff (1971) annd Patnaik (1982), a consensus is evolving vinidnsnggeetsregionasspacewhichislargerfinanenysingleurban area or group of villages annd finat its spatial framework is sub- national (also see Bendavid-Val, 1983). Plannningatfienational levelisahallmarkoffiegrowth approach to developnent. As alternative to national planning, disaggregationinspacehasbeenadvocatedasaneansofbringing about a acre effective distribution of developnent benefits (see Higgins, 1980). In this regard, two major approaches are evident in 57 fie literature: (1) through hierarchy of centers (2) through regional plannning. Rondinelli’ (1979/1980; 1982: 1983; 1985) is most noted for advocatingfiedevelopnentofahierarchyofcentersasaneansto achieveamorebalancedspatial systemaswell asfcrbringingabout nore equitable distribution of development beefits (also see Rondinelli annd Evans, 1983; Rondirelli annd Huddle, 1977). Friednann is noted for regional plannnning, especially with his agropolitan nodel proposal (1979). Tl‘e ofier approacles to regional planning are in terns of fie application of Perroux's econonic model of growth poles annd growth centers (Darkoh, 1977). Stuckey ( 1975) has criticised Friednann's agropolitan model of development in terms of failure to incorporate fie influece of external factors in developnent of regions. Darkch (1977) has argued that growth poles annd growth centers are essentially still growth-oriented approach to developnent. In studies on developnent as well as on regional inequalities in developnent, many studies have focussed on fie largest or fairly extensive adninistrative units as fieir analytical base. Slater (1975) has attributed finis to data constraints at fie smaller administrative units. Tre application of fie growth model of development also accounted for fine utilization of macro- rafier finann micro-unit of observations. In finis regard, studies on regional inequality in Malaysia is also no exception. Difficulties in obtaining appropriate data below fie state-level have produced more studies on inter-regional inequalities in developnent being unndertaken 58 at fie state-scale of analysis (Geog, 1979; Aris Ofirnan, 1983) finan at fie sub-state territorial framework. For that matter, Malaysia's five-year developnent plan dooments also have used fie state-scale of annalysis to evaluate Malaysia's developnent performance. When intro-state differentiation in developnent are considered veryoftenfieytendtodealonlywithconsiderationsofaregionof fiecomtryorfieyusedgecgraphical points, suchastownnsannd villages, as fineir unnits of analysis. In this regard, nest of fie analysis on developnent performance in Peninsular Malaysia have been conductedeifierusingfiecorponentstatesoffinepeninsulaorusing fie urban centers annd local goverments units; fie former exerplified by studies such as C’neog's (1979), Anand's (1983), Aris Ofiman's (1983) end Malaysia's five-year developnent plann docunents, while fie latter are as undertaken by studies such as Osborn's (1974) annd Kon's (1978). Leinbach's study (1972) can be considered as signnificant departure fron fie ofier studies since it utilized unnits of analysis that are enaller finan fie state-level. However, fie stortconings of Leinbach's study is its failure to produce unnits of observations finat areinncogruecewifinanyecistingadninistrativebonndaryu state ordistrict. Infact, Leinbach'shexagonalconstructnotonlydonot fit wifin even fine smallest adninnistrative unit boundary (i.e. fie district), but also in sore instances, they overlapped district and state bounndaries. Since studies on deveth annd inter—regional inequality at smaller scale of analysis would provide more details for understanding 59 of the nature and problems of developnent, this study therefore focusses on fie district adninistrative unnits for examination of fie pattern of spatial developnent annd fie magnitude of spatial inequality in Peninsular Malaysia. As stated by Cnetwymi, Jr. (1983, p.15): Breaking [up] national developnent into specific regional contexts, it was fond, reduced conplexity, lelped to identify . . . innbalances annd pinpointed specific regional problems and cmortunities . . . all ultimately to facilitate achieving developnent goals. Examination of spatial developnent annd spatial inequality in developnent at finis scale, especially wifin extensive coverage, is relatively few. Also, fie district constitutes fie smallest adninnistrative units in fie conntry. Approach to Development In practice, fie different perceptions of developnent, especially wifininn a capitalist or capitalist-oriented politico-ecoomic system, canbeviewedintermsoftwomajorfienesof developnent strategy: (1) developnent finat emphasizes on fie growth appraoch (2) developnent finat emphasizes on growth with equity or distribution. 'negrowfinamroachtodevelopnenthasbeennostpervasively applied both by developed and developing countries. Tre growth wifin distribution or equity trodel of developnent, on fie ofier hand, isanalternativeapproachtofinegrowfinuodelemergingonlyasa 6O consequence of disillusionment with fie gronfin peformance. Mnchhasbeenwrittenaboltfiegrwfinnodel of developnent (Iewis, 1955 end 1961; Ford, 1966; Okunn and Richardson, 1961). Suffice it is to state rere fie funndamental features of this model. Basically, fie growth approach to developnent is one which places strogemphasisonincreaseinproductionandsizeoffieproductive sectors based upon fine principles of (1) efficiency, (2) economies of scale, (3) aggloneration economies, (4) conparative advantage end (5) (limited) public intervention to regulate fie market. High end rapid growthratesoffieGrossNationalProduct (GNP) offieeconomyis viewed as essential for initiating annd stimulating developnent. This istobeattainedmainlyintermecf accumulationof capital anndits investment inn industrialization annd urbanzation. 'I‘re focus on econonic growth comes with it an aggregate bias abolt developnent; that ithastobeplannedanndexecutedbynational govermentsand finrough fie use of capital-intensive technology. By virtue of its focus on certain productive annd efficient sectors, as well as its tendecy to cocentrate investments in certain locations of conparative advantage, proponents of economic growth have geerally asserted finat "inequality . . . is an inevitable coconitant annd condition of growth." (Hirschnan, 1958, p.184). It is geerally believed that fie expansion of fie productive ecoomic sectors will give rise to multiplier and spread effects to fie ofier sectors of fine economy. Hence, through fie filtering-down process economic growth will result in a broad permeation of beefits finroughout the society. Inequality is considered as an essential part of gmfin 61 (Hirscl‘man, 1958). In fie interest of efficiency, developnent unnder fie growth model cocentrates investments only in certain localities crregions. However, thismodel ofdevelcpnentpresumesfinatovera period of time, through fie trickle-down effects, beefits of developnent will filter to fie peripneral areas or regions. Such process will finen reduce inequality. mstoffiedevelopingconntries, especiallyfinoseinfie‘l‘hird World, hereadcptedfiegrowfinapproachtodevelopnent. 'neywere reavily influeced by this model of developnent for a number of reasons. First, it relates to fie fact finat most of fie developnent planners annd bureaucrats in newly independent Third World countries receivedfieireducationanndprofessional traininginfieWestand were inspired by fieir school of fiought on developnent (Patnaik, 1982). Most of fie western countries strogly subscribed to fie viewfinatecoomicgrodfinwasfieanswertofieproblensof underdevelopment of Third World countries . Anofierreasonrelatestoconditionsimposedbyaiddoor countries and international development funding organizations, suchasfieUnitedNationsandfieWorldBank, whichrequirefinat recipient nations apply fie growth model to developnent (Rogers, 1976; Rondinelli, 1985). Experiences of technological annd economic advancenentbywesterncountries applyingfieecononicgrovthapproach to developnent have led to fie belief among fiese nations, including international developnent-funding organizations, that fie economic growfin model would be fie appropriate developnent strategy to be 62 applied by fie underdeveloped counties of fie Third World. Professor W. Arthur Lewis (1955), a major proponent of ecoomic growth, has assertedfinatecoemicgrowfinmsmeasuredbyGNP) increasesfierange ofhumandoice. Suchviewhavehadaprofonndinnflueceuponfie developnent process of most of fie Third World countries. This point has been endorsed by Bruton (1985, p.1099) as follows: Development economics originated in fie late 19403 in response to a real world question, fie existence of extremely rich countries alogside extelnely poorer oes. That fact set in motion . . . and resulted a great oltpolrings of literature anndagreatarrayofcoceptsanndmodels . . . --dualeconomy, labor surplus, low level equilibrium trap, balanced growfin, viciols circles of poverty, critical minimum effort, big push, dependency,centeranndperiphery-- . . . . Virfilallyevery pocrercountyhadbeconeinterestedinits economic developnent. Policymakersweredirectlycocernedwifinfiefinndingsoffie economists, andeconomistsweremuchindemandasconsultants. Boonomistswerelisteedto, and[finey] didinfactinfluece [development] policy. Tie growth model to development has been advocated for developing counties in view of fie limited capital resolrces which finey have. Ithasbeenrationalizedbypropoentsoffiegrowthmodel thatwifin linmited resolrces it would be inefficient and ineffective to attempt to sprinnkle developnent investments fininly over fie national territory. Rafier, key urban centers or regions will be selected for cocentated investment programs finat would benefit from economies of scale annd external economies of agglomeration. This model is also advocated because in fie log-nun, finrongh fie tickle-down effects, beeficial spread effects will flow annd permeate to fie lagging regions. Despite fie force with which the hardline ecoomists have expressedfieirviewsandfieextenttowhichfieirrecomendations havebeenfollowedinfie'rhirdWorld, fiereisscantevidecefinata narrowfcolsoneconomicgrowfinhasresultedinabroadpermeationof beefitsfinroughortsociety. Indeed, fietendecyappearstobe towardsgreaterrafierfinanlessincotedisparityasgrorfincccurs (Ahluwalia, 1974; Drakakis-Smith, 1980). Steeten, 1981). As Ahluwalia, Carter, annd Clenery (1979, p.299) state: "despite fie developingcounties' impressivegrorfinoffiepast25years, its beefitshaveonlyreacl‘edfiepoortoaverylimiteddegree." Logan (1972) end Mehretu (1986) have attibuted finis to fie stuctural biases finatwereputinplaceinnmostoffie'mirdWorldcounties during colonial times. As Mehretu (1986, p.30) points olt: "Socio- spatial polarization in levels of living was accelerated by post- independencemodelsofdeveloptentasfieseservedtoe-ntendfie colonial extactivepatternorweretcoweaktopreventfielagged effect of colonial policies." Failure to dismantle fie extractive structurefrustratedattemptstoincreasefierateofgrowfinand developnent in Africa, for example. Mehretu also stated finat spatial biases in production annd distibution were among fie intactable problems finat frustated broad-based developnent in Africa. Sore of fie major disillusionment wifin fie growth approach to developnent were: (1) that alfiough overall fie rates of growth increased, the beefits of growth have not been equitably distibuted between social grolps annd regions (Streeten, 1981; Ranis, 64 1977b; Lakshamanan, 1982); (2) finat growth model to developnent tends to deepen differeces between ‘growfin' points annd lagging areas -- fiereby giving rise to spatial stuctures reflecting different patterns and status of development. According to Mabcgunje (1981), favored localities annd regions are fortified and sustained byeverincreasinginternalandecternal economiesanndgainn at fie expense of ofier localities and regions. Myrdal (1957) has rationalised this in his ‘circular annd cumulative causation' principle: (3) finat spread effects from growfin centers have been mininmal finan expected (Stohr amd 'I‘cdtling, 1979: Appalraju annd Safier, 1976: Waller, 1974; Gilbert, 1975); (4) finatalfioughsoneinnovationshavebeenobservedtoflow down from developnent centers after fie mannner postulated by adlerents of diffusion view, fie nature or content of such spreadeffectsmaynotnecessarilybeinfieinterestcffie lagging areas. As Smith (1982, p.325) contends: "what diffuses may simply reflect fie values annd preferences of fie metropolitan elite which may have little bearing on the real needs of fie people in the countyside." 'I're contention by fie growth propoents finat inequality is essential annd inevitable for growfin was also challenged. Myrdal (1968 end 1971) argued finat fie opposite was in fact a necessary pre- condition for more rapid developnent . 65 Inequalityandfietrendtowardsrisinginequalitystandasa complex of inhibitions annd obstacles to developnent annd that, consequently, fiereisanurgentneedfcrreversingfietend andcreatinggreaterequalityasaconditionforspeedingnp developnent (Myrdal, 1971, pp.63-64). Forall itssuccessinraisinggrowthratesofGNP, fiedominnant post-war stater of economic growth came under heavy criticism for its failures: continued unemployment, growfin in incone inequality wifininandacrossnationsandanincreaseinpoverty. Inresponse, several alternative approacles to achieve more effective developnent infie'lhirdWorldcountieshaveemerged. 'I‘researeternmdasgrowth wifin equity or distribution. In finis regard six major approacnes may be mentioed: (1) Employment Geeration (2) Redirecting Investnent (3) Meeting Basic Needs (4) Human Resonrce Developnent (5) Agriculture First Developnent (6) Integrated Rural Developnent All spring from a conviction that taditional reliance on growfin of (NP aloe will not beefit the poor or would not beefit then quickly enough. Ttesealternativeapproacrestogrowfinalsoshareacomon feature of recognizing fie importance of social, political, annd cultural factors in developnent. They agree finat oe of fie crucial limitationsofpastapproacl'eshavebeenfieirnarrowfocuson economicfactors. 66 Tleneedtoincreaseemploymentisoeoffieapproacnesinfie growfin wifin equity or distibution model to developnent. This came abontasaresultofwidespreadandgrowingnunemploymentinsoreof fineThirdWorldcounties despitefiegrowthoffieGNP. 'Ihis perspectivecalls for focusofattentiononfieinfornmal sectorofthe variols Third World counties: wtere a great deal of entepreeurial talentexistsbutfiemainnbarriertogreatercontributiononfieir partwasaccesstocapitalontermscompetitivewifinfieformal sector. In developing finis statgey, fie International Labor Organization (11.0) (1970 end 1972) places prinmary emphasis on increasing fie availability of capital in finis sector and cocentating expenditures on enployment creating activities . Partiollar attention was also given to fie rural area, especially to fie use of labor-intensive production techniques in agricultural growfin. The second approach deals with fie reorientation of capital formation, away from large-scale centralized projects, to investments which will relate directly to fie poor: education, realth, credit. Invesbnen‘Ginfiosesectorswill leadtoincreaseinproductivityof fie poor annd hece increase fieir incone. Clenery, Ahluwalia, Bell, Duloy, annd Jolly (1974) are major proponents of this approach to developnent. ‘Ihisapproachisbasedonfiepremisefinatfiepoormust havegreatercapitaltogeerateincorenecessarytomeetfieirneeds. While in fie short-run such reorientation of capital formation may affect growth, in fie log-run increased productivity annd income of 67 finepoorwill raiseincolesof allmembersoffiescciety. Perhaps, fie most notable of fiese alternative approaches to growfin is fie basic needs approach. This approach was officially lauded by fie World Ehployment Oonferece of fie 11.0 in May, 1975 (110, 1977). Mahbub ul Haq (1973) end Steeten annd Burki (1977) also favor fie basic needs approach to developnent. Essentially, fie basic needs approach include 2 elements. First, it advocates certain minimum requirenents of a fanmily for private consumption: adequate food, stelter, clofining, annd certain nouseold equipment and furniture. Second, itrequiresessential servicestobeprovi®dby end for fie community at large, such as safe drinnking water, sannitation, tannsport, realth, education annd cultural facilities (ILO, 1977: Sinngh, 1979). Adelmann(l975) isfineproponentoffiehumanresolrces developnent route to achieving growth wifin equity. A precondition for success of finis approach is redistibution of productive asse -- lannd annd physical capital -- as occurred in Japan, Taiwan and Solth Korea. Alsoprovisionmustbemadetoensurecontinuedaccessto assets for fie poor oce fie redistibution has taken place. This is followed by a massive program to develop human resources. TIe next step is a human resource-intensive inndustialization annd growth strategy. Tie high rate of employment to be geerated by inndustialization will provide incone which will lead to a demand for fie goods produced annd fiereby ensure a wide distibution of beefits. Thisapproachrequiresstoggoverment, forfieprocessofhuman resolrcesdeveloptentcanbeacconpaniedbyslowgrowfinoffieGNP, 68 resulting in social tension annd political innstability. Tneenphasisonagriculturefirstapproachwasadvccatedby Mellcr (1976). This approach requires lannd reform first. In this approach agriculture is envisaged to play two major roles: (1) Agriculturemustsupplyfiewagegoodswhicharenecessary for employment creation. Increases in agricultural production are essential to achieve this. (2) Agricnlltnue must supply enployment, through technnical change in agriculture, primarily biological research: new seeds, new fertiliser practises, and irrigation. Though fieresultantincreasedontputwouldnotdirectlyraise employment, fie increased spending of farmers will; finrough fie multiplier effects. Theinntegr‘atedruraldevelopnent approach, asadvocatedby Waterston (1974/1975), is based on fie premise that top-down approadestodeveloprenthavenotbeensuccessfulinmeetingfie social needs of fie rural poor. Waterston (1974/1975) contends finat stategies that focus on agriculture aloe can result in enrichment of fiealreadyrichfarmers; sinceonlyfiosefarmerswhoconldafford fie necessary inputs colld take advantage of fie new high-yielding varieties wifin fieir concomitant needs for water, fertilizer, pesticides, annd insecticides. He also argues finat social service provision by government leads to "welfare mentality" as he saw in Tannzania annd Sri Lannka. Thus, this approach advocates agricultural developnent alog wifin social infrastucture annd services. Six 69 elementsarenecessaryforsuccessofthisapproach, wifinfie precondition finat lannd first be equitably distributed. (1) labor inntensive farming by small farmers: (2) use of off-season labor surplus in building minor developnent works annd infrastucture: ( 3) labor-using light inndustry for processing of agricultural products, production of intermediate goods for agricultural production, anndproductionof lightconsnmergcodsbasedon local rev materials; (4) self-kelp or self-reliant; (5) implementation by a government organization with power cutting across ordinnary ministy jurisdictions: (6) "regional planning" wifin a hierarchy of develoment centers bricging the gap between villages end the capital city. Tlere are at least finree major features characterizing fie various alternatives to fie growth approach to developnent: (1) an equity orientation, emphasizing a direct attack on poverty and setting of mininmnm of consumption finrestolds; (2) a recognition finat more production annd better distibution ' must be geerated togefier to define develoment; annd, (3) a preferece for self-reliance, ‘bottom-up' planning styles. Itisapparentfinatevenscmecffiealternativeapprcactesto growth may be difficult to implement, especially where fiey involve land reforms. It is also clear finat emphasis on any oe of fiese alternative approacl'es to growfin will not necessarily lead to more To. 70 effective distribution of develoment beefits. Since developnent is a multidimensional process, a combination of fiese approacl‘es togefier withfiegrowthstategyisnee®dtobringabontdesirablechangee As mentioed earlier, fie conventional approach to development (fie growfin model) innvolves fie national spatial framework for planning, implementation or evaluation of develoment. waever, fie new approactes to development advocate fie use of smaller territorial framework for more effective distibution of develcment beefits. As Higgins (1980, p.vi) aptly states: All point to plannnning develoment in terms of smaller geograplnical nunits . . . . Smaller political annd geographic units [are] . . .moreresponsivetofieaspirationsand needscffiepeoplefinanacentalgovernmentcanbe. Thus, regional planning annd decentalized form of urbanzation are new spatial frameworks for more equitable distbution of developnent beefits. Like ofier Third World counties, Malaysia, too, had adopted the growfinapproachtodevelomnent. Infect, fiegrowfinapproachto developnent dominated Malaysia' 3 develoment stategy unntil fie inception of the New Economic Policy in 1971. Since finen Malaysia has embarked upon fie growfin with equity or distibution approacles to developnent. 71 Measurement of Develoment Exteeive discussions on measurement and measures of develoment existinfieliterature. Metisclearisthatjustasitis difficult to clearly define develoment, it is also equally as difficult to measure develoment. "This multiplicity of possible indicators for any given . . . dimensions of develoment simply corpoundsfieproblemsarisingfromfieexistececfseveral . . . dimensions [of develomentJ" (Colman annd Nixson, 1978, p.6). This hasbeenborneoltbyfiefactthatfieliteraturehaveindicated finat develoment performance can be measured by a number of different inndicators. Pnnr‘fiermore, fiere is also fie question of dealing with qualitative or objective aspects of develoment. Although it has been annd continues to be debated wtefiner qualitative or subjective aspects of develoment can satisfactorily be measured by inndirectly using inndicators finat are directly measurable, fie majority of soolarsseemtoagreewifinfiecoclusionofsuchwritersas Easterlin (1974), Calpbell et al. (1976) annd Knox (1976) "finat subjective indicators serve as a useful annd necessary supplement to fie ‘hard' objective measures" (Dale, 1980, p.504). Knox annd MacLaran (1977) have found a positive annd statistically siginificant correla- tion between objective annd subjective measures in fieir study of level of living for Dnundee, Scotlannd. Thus, fiey have cocluded: "in view of fie positive correlation we have found between objectively measured circumstances annd both values annd perceptions of most life- to. D. .3 72 domains, we . . . cocludefinat, forfinepurposeofgeerally describing or evaluating . . . disparities in well-being, conventional ‘hard' data are as good a surrogate as any." Itisapparentfromfieliteraturethatfiereisnostaight- forward, widely accepted way of~ measuring development (Cole, 1977). Tie choice of mefiods for measuring annd evaluating develoment performancedependslargelyonwhatisimpliedbyfieconceptof develoment. Tre different perceptions of develoment calls for the use of different inndicators for measuring develoment. Thereareat leasttwomajorwaysinwhichdevelomenthavebeen annd can be measrred: (l) in terms of growth (sectoral approach) (2) in terms of distibution of develoment beefits (mite amach) The most widely used indicators for measuring develoment performancehasbeenfieuseoffiemacro-economic accounts, suchas fie gross national product (GNP), gross donestic product (GDP), income per capita, etc. (de Solza annd Porter, 1974; Hicks annd Steeten, 1979; Friedmann, 1980). The use of such measures indicate and reflect fie amnesis placed upon fie economic—growfin model of develoment. It is becoming increasingly apparent in fie more recent literature finat fie use of economic criteria aloe to measure development is nnow geerally recognized to be insufficient annd far from satisfactory; on grounds finat itdcesnotmeasurefiewiderormulti-dimensional aspectsof develoment, annd also because it neglects fie distibutional perspective of develoment. "GNP aloe cannot satisfy variois 73 structural changes and.distribution.prcblems whidh.constitute the central and essential issues of developmen " (Takamori and.YamaShita, 1973, p.111; also see Knox, 1974; Hay, Jr., 1979). Another linmitation of the use of the GNP, or other macro—economic accounts, is that they fail to take into account economic activities conducted.outside the national monetized accounting system. Such activities, known as the informal economy play an important, if not dominant, part in the lives of manny people in fie Third World counties (801‘, 1982). Anofier severe limitationuof the macro~economic accounts is that they do not potray'inequities in their internal distribution.of development or the social costs imposed on fie community (Hay, Jr., 1979; Friedmann, 1980; Ehcel et al., 1975; dearo» 1977; Rhoda, 1982; Carley, 1983; Bruton, 1985). This is because aggregate statistics as average income or total value of goods and services produced.may hide the fact that great inequalities still exists (see Smith, 1982). The use of macro-economic indicators per se to gauge development is, conventionally, related to the conceptualization of development as economic growth. This approach to development has been criticised on grounds that growth is not synonymous with development. This vieW'has been aptly sounded.by Easter (1972, p.1): "However important the economic dimension of development, it is dangerous to use it as a proxy of develoment. " While it is geerally accepted finat economic growfin is oe of fie multi-dimensional aspects of development, it has been argued that economic growth can take place without development (Streeten, 1972). Weissman (1968, p.97) has finus stated: "Human 74 progressisnotanautomaticresultofeconomicgrowfin. Toobtainit, fienationmust . . . allocateasuitableshareoffiewealfinit produces to social develoment." Indeed, as Weisskopf (1983) has argued, cocentation of attention on economic growth performance as well as attempts to explain development merely in terms of a limited set of economic variables tends to reinforce fie false notion of an nunderlying similarity in develoment experiece annd develoment potential across all societies annd systems. Tie idea finat economic growfinwasfieanswertofienunderdevelomentoffiedeveloping conntrieshavegeerallybeenprorotedbywesterneconomistsbasedon fie economic annd technological success of such approach to develoment in fie Western nations. Since it is geerally accepted finat develoment does not merely involves fie maximization of economic growth, but also includes ofier cocernns, such as quality of life annd equitable distibution of develoment beefits, ofier alternative mefiods for gauging develomentperformancehavebeenandarestillbeingdevisedinorder to arrive at a more meaningful measurement of develoment. Sussman (1972), a sociologist, has argued that fie economic growfin model tends to result in preoccupation with materialistic objects and hece divert attention from sore of fie more important intangible aspects of life. Sussman (1972, p.133) has stated: "Mental well-being has been an increasingly recognized needs of society." In this regard, oe of the most common approaches emerging as an acceptable alternnative to replacefieuseoffiemacro-economicaccountsperseisfie application of a mefiod finat combines several inndicatcrs of diverse (‘I a: la 75 dimensions of developnent, economic and non-economic, to form a conpositeindexwhichisteatedasaproxymeasure fordeveloment (see United Nations Organization, 1954; Drewnowski, 1974; OECD, 1976: Smnifin, 1973) . Tne selection of indicators of develoment depeds very much upon how deveth is perceived. Tie indicators of develoment used must not only be properly quantified, especially if fiey represent sone aspects of develoment that are intangible or qualitative in nature, but also fiey must be standardizedtoenablecomparisontobemade; sincefiedifferent variables colld have been measured in different nunits. It snould be noted finat fie variols indicators of developnent do not led finemselvestoaggregationinfiesamewayasdoesfieuseoffie macro—economicaccounts; sincefiedirectionoffieindicatorsmustbe cosidered in containing fie different indicators' values (see Knox, 1974) . For certain indicators negative instead of positive values would be desirable for develoment -- for example, fiose for mortality rates. Tie use of fie corposite index of develoment requires selection of appropriate measures to reflect fie different conpoents and aspects of develoment. (In this question of fie measurement of development, fiere is increasing evidece inn fie literature calling for fie usage of bofin economic and non-economic indicators (see Ram, 1982; Meeler, 1980; Hicks and Steeten, 1979; Hicks, 1979). Hnnwever, lack of reliable, comparable, readily obtained data comonly linnit the use or doice of indicators finat may be more relevannt finan ofiers. 76 Noefieless, fie use of finis mefiod to gauge developnent performance isgeerallyconsideredtobeimprovementoverfieuseofmacro- economic accounts aloe, since fie corposite develoment index also attempts to include, and hece in doing so reflect, fie ofier facets of developnent finat would ofierwise be excluded. In any case, no ideal setofindicatorscaneverbederivedorproducedforfie purposeofmeaslringdevelomentaslogasitisrecognizedfinat developnent is a complex process wifin limitless possibilities in terms of its attributes. This is also related to fie fact finat develoment isnortastaticbutrafieradynamicprocess. Furfiermore, fie variols aspects of develoment are sonehow inticately interrelated to oe anofier. ‘ Measlrerent of development, as well as inequalities in developnent, can be nudertaken in terms of a social (including economic) or spatial categorization of people and activities (Soja, 1976; Seers, 1972). However, since "fine social and spatial stuctures of inequality are sesitively and dialectically interactive." (Soja, 1976, p.1), it is necessary that fie cotplex dynamics of fie develoment process considers fie spatial and non-spatial dimensions simultaneolsly. SuchaviewhasalsobeenexpressedbyTakamoriand Yamashita (1973, p.111): Interdisciplinary approacl'es skould be taken for purposes of . . .measurerent. . . . First, becauseeconomicdeveloment cannot be teated separately from fie interlocking links with cultural, social, ecological, and political factors . . . Secodly, fiere is increasing criticism against fie overenphasis on economic growth. 77 In referece to spatial inequalities, Gore (1984, p.25) fierefore has aptly pointed ont finat "regional disparity in developnent . . . may be measured using varions indicators." Performancecf develomentcanbeanalysedusingfiesectoral amroadn or fie spatial approach (Rroda, 1982). Sectoral annalysis of develoment is oe in which a county's economy is divided into sectoral components, such as agriculture, education, lealth, etc., and fien investigation on fie characteristics and interaction within fine sectorsarenudertaken. Incontasts, fiespatialapproachstartsby dividing a county into spatial unnits, such as states, disticts, cities; and finen examination on fie activities, processes and dynamics wifininandbetweenfieselectednunitsof analysisarenudertaken. Sectoral annalysis of development, which is aspatial in nature, has been fie (ruminant mode of evaluating develoment performance. This is due to fine fact that develoment stategies of develOped and mostTWCshavebeenbasedonsectoral plannnningorientationasa consequece of plusuing fie growth-oriented model of develoment (Prantilla, 1981; Chatterji, 1977). T‘re abstaction of space in such mode of planning teded to give rise to spatial polarization of develoment finrough fie ‘deviation—amplifying' principle (Soja and Tobin, 1979, p.158) and ‘circular and cumulative causation' principle (Myrdal, 1957). Such a plannnning and developnent orientation, it is conteded, creates inequities in fie beefits of develoment among regions wifinin a county (see Katchamat, 1978). Sears (1972) and Adelman and Morris (1972) have suggested that oewaytoassessdevelomentperformancewonldbetoexaminethe 78 problem of inequality in develoment, i.e. to investigate wrefier inequalityhasincreasedordecreasedinmagnitude. Suchanassertion seems logical and appropriate since, after all, fie overriding purpose of develoment is, at least nornmatively, to bring about desired, positive changes. Since inequality in develoment implies a relative deprivation of develoment beefits and opportunities, it cannot, fierefore, costitute a desirable change. The existece of inequality in development also reflects irregularities in fie developnent process. Treirpersisteceposesmoreofaproblemfinananincentive to develoment. Also, fieir persistece will inevitably, exert stain upon resouoes needed to redress fiem. DeKadt (1985, p.552) aptly points olt finat "information can influece what policies may be considered in fie future." This is geerallytue, since "whatwekncw . . . enters intoolrmodelsand policies" (Steeten, 1981, p.137). According to fie planning- progranming-implenentation process, information derived from fie evaluation of developnent performance also provide important inputs for furfier and future plannnning; which ,in turn, would determine develoment programs to be formulated and projects to be implemented in future (Robbins, 1976). This is based on fie fact finat ‘knnowledge ofandluowledgeaboitfiedevelomentprocessinaparticular enviromentataparticulartimeisfieprcductoffiedeveloment process itself' (Tedler, 1975, cited by Bruton, 1985, p.1109). Tlerefore, fie way in which developnent performance is measured has importannt implications and bearing upon future development planning, 79 formulation of developnent progrens and implementation of develoment projects. As stated by Bruton (1985, p.1109): "Ge must also recognnizefinat ‘formal l-cnowledge' --[such as] models . . . -- that are misleadirgmnayactuallyimpede. . .effortstobringtobearoe's knowledge. . .onanissue." Inequality “in Development: Empirical Evidence Two major characteristics are evident from fie literature dealing with the issue of inequality in develoment. First, inequality in develoment is assessed in economic (aspatial) or spatial terms. Betweenfietwo, fiereseemtobemoreliteratureonfienon—spatial perspective of inequality inn develoment. Secod, fie Kuznet's divergence-convergence hypofiesis (also known as "U" or "Inverted- " hypofiesis) isoftenreferredtoinorderdetermineorestablishfie pattern of fie empirical observation. Wifinin fie context of develoment, two major fieoretical views are pertinent to discussions on fie ptenomena of spatial inequalities in developnent. These, incidentally, coincide with fie two major approacles to developnent: (1) fie growth model, and (2) fie growth with equity or distibution model . Literatureonbothfieecoomicgrowthapproachandfiegrowfin with equity or distribution have indicated, implicitly or explicitly, finat finey are cocerned with resolving fie phenonena of (spatial) inequalities in developnent. However, fiey differ distinctively infiemanunerinwhichreductionininequalitiesindevelomentis 80 tobeaccomplisted. lTegrmfinapproachtodevelopnentregards fieinequalityissuennotonlyasanaturalanndconcomitantprocess finrough which fie developnent path will occur but also consider finat inequality will, finrough fie trickle-down effect, he reduced inmagnitudeovertinme. Qnfieofierhannd, fievariousalternnatives to fie grovfin approach to developnent consider inequalities in developnent as unnecessary annd undesirable to developnent. ‘ITe growth wifin equity or distribution model to developnent, fierefore, calls foradirectapproachtodealwifinfieinequalityproblem. Figure 2.1 illustrates fie differennt hypothetical developnent pafins which fiese two major approacres to developnent travel annd how, in turn, fiey affect fie question of inequality in development. In reference to Figure 2.1, an ideal development path is oe which travels along fie 45 degree trajectory. This developnent path inndicates finat fiere is both growfin arnd equality or less inequality. In reality, l'ersver, this is difficult, if not impossible, to accomplish. underfieeconomicgrovfinapproach, itishypofinesizedfinatfine pafin of developnent will be oe which moves fron ‘A' to ‘B' annd, finough fie spread or trickle-down process, fie developnent pafin is expected to continue along ‘C'. I-bwever, if that fails, due probably to faulty operations or irregularities of fie growth approach, then fie developnent path ends up travellinng along ‘D'; implying, fierefore, that fiere will be growth but with greater inequality or inequity. 81 mfieofierhand, underfiegronfinwithdistributionorequity approachtodevelopmentappropriatemeasureswolldhavetobe undertaken to bring about effective and equitable distribution of developnent beefits. According to this fieoretical postulate, fie development path will first travel from ‘A' to ‘E', and, finen, as fie develoment process picks up, it is hypofiesized finat finis will occur along ‘F'. Hovever, if fie system fails, fien fie developnent path underfiegrowfinwifinequityordistributionapproachwillproceed along ‘F'; implying finat fiere will be less inequality/inequity but little growth. According to Streeten (1981) annd Bryant and White (1982), different perceptions of developnent will not only result in fie application or adaption of different approach to fie implerentation of develoment, but also, in turn, will have different consequences upon fie pattern of spatial development as well as spatial inequality in development. Tle Structure-Conduct-Performance (SCP) model (Shaffer, 1980) snppor'ts such contention. In essence, fie SCP model postulates finat fie performance of any firm (or organization) can be rationalized in terms of fie organization's ‘structure' and ‘conduct'. In fie context of development, developnent policies and strategies can be viewed as making-up fie ‘structure' of development, while actual developnent programs annd projects can be said to ooetitute fie ‘conduct' of developnent. Regarding fie different patterns of spatial developnent finat can emerge as a result of fie application or adoption of fie different 82 GROWTH 4\ D C The ideal development path \4/ an EQUALITY/EQUITY \/ Figure 2.1: A Hypofietical Wbdel of fie Development Pafin. Source: Paul Chen. (1979). "Tie Third Malaysia Plan and Social Economics," in C.K. Check, K.S. Mun and R. Thillainafinan. eds. Malaysia: Scme Contemporary Issues in Socio-Eoononic Development. Kuala Lumpur, Malaysia, University of Malaya Press. approach to developnent, Knox (1982) has fieorized finat finree major patterns are likely to emerge, namely (1) convergence, (2) divergence, annd (3) status quo. Convergence refers to a spatial pattern which is fie outcome of a relative reduction in fie level of differentiation between regime over a period of time. Divergence relates to a spatial patternwhichisfieoutcone ofarelative increaseinfine level of differentiation between regime over a period of time. Status quo refers to a spatial pattern of developnent which exhibits no significant change from fie previous spatial configuration. 'ne contention that inequality in development is an inevitable conditionofgrowthanndfinatasgrowthproceedinequalitywillbe graduallyreducedhasbeenbackedbytwomajorempirical studies -- finose by Kuznets' (1955 and 1963) and Williamson's (1965). Bofin fiese studies, based upon comparisons of macro-economic accounts, found finat fiedevelqrrentprocessfolloedfieconrseofaninwertedu direction; whereby inequality would first increase during fie early stagesofdevelopmentandfienovertimelag, finroughfiespreador trickle-down effects, inequality would gradually decline. A number of studies -- Adelman and Morris (1973), Paukert (1973), Cheery annd Syrquin (1975), Ahluwalia (1976) and Fields (1980) -- using different and progressively more reliable sets of cross-country data, have reported confirmation of Kuzret's hypofiesis to some degree. For example, Ahluwalia's study has indicated finat fie poor hasgainedinincomeasfieGNPrises. Field'sstudy, basedontime series data for finirteen countries, have found finat economic growfin 84 hasbeensuccessfulinupgradingfieeconomicconditionsofa significant number of fie population in 10 of fie 13 counties re studied -- nenely in Bangladesh, Brazil, Costa Rica, Pakistan, Puerto Rico, Singapore, Sri Lanka, Taiwan, Thailand and Mexico. Only three counties, nanely Argentina, Inndia and Phillipines, stowed finat absolute inequality worsens inspite of significant growfin in fieir economy. While acknowledging finat fie conventional aggregate measures of economic performance are, for fie various groups of less developed counties for fie 1950-80 period, geerally acceptable, Bruton (1985, p.1100) admits finat "fie variannce around fie average is high. " There are frequent arguments and assertions, finerefore, finat ‘development' finat has, in fact, occurred has brought less satisfaction finan anticipated. Hence, fiere were also studies finat produced results finat were contary to Kuznets' or Williamson's studies. Studies by fie International Labor Organization (1969, 1979a, 1979b) and Griffin anthan(l978) havefounndfinatgrowthhasbeenaccompaniedbyan absolute as well as relative decline in fie average inncole of fie poor. For example, fie study by Griffin and Khan (1978) have indicated finat in several Asian counties -- such as Malaysia, Bangladesh and Sri Lannka and Phillipines -- the problem of poverty anndinequalitieshaveremainedunchangeddespitefieirgrowfinin production. "While sore developing nations showed impressive aggregate rates of economic growth in fie 19603 and 19703 mnany did not; and almost all experienced mounting problens of inequality." (Weisskopf, 1983, p.896). With specific reference to counties in 85 Soufieast Asia, particularly fincse in fie Association of Soufieast Asiann Nations (ASEAN) -- finen comprising Malaysia, Indoesia, Thailand, Phillipines and Singapore -- Mahangas (1982, p.270) has observedfinat "fieASEANregionhasenperiencedtelendouseconotdc growfin but wifin little improvement in distibutive equity. " That fine application of fie economic growth approach to developnent has not proventoclosefieinequalitygapinmostThirdWorldcounties, as well fie realization finat fie phenonena of inequalities in development continneto'persistinevenfiemostdevelopedcountiesoffiewest, inncludingfieUnitedStates, havefiereforebeggedahostofquestions among developnent fieorists and practitioers regarding fie usefulness andpracticalityoffieeconomicgrowfinmodel asanappropriate develoment stategy for Third World counties to emulate. Tie economic growth emphasis to developnent also creates problems of spatial inequality in developnent. In pursuannce of efficiency in production, finis approach to developnent focussed on concentation of investments in certain stategic locations of comparative advantage . Two major arguments have typically been invoked to justify policies of ooncentating economic growth wifinin a few selected places or ‘growth centers' . The first is finat concentation engenders certain economies of aggloreration. Tie secod argument is finat spatial concentation is an efficient means of indirectly promoting higter levels of developnent over a much wider area. According to Berry (1969, p.288): "Growth impulses and economic developnent . . . tickle-down to smaller places and ultimately infuse dynamism into even fie most 86 tadition bound peripteries." Bofin empirical evidence and fieoretical explanations have pointed to fie tendencies of developnent toward concentation or spatial polarization, during fie early stages, in a few core areas wifinin the developing national systems (see El-Shakhs, 1972 and 1977: Alonso, 1968; Friedmann, 1973). As El-Shakhs (1977, p.127) pointed olt: "Such tendenciesbecomestogerfiemorefieyreinnforcefiefavorable position and comparative advantage of fieee cores." Alfinogh sore degree of spatial inequality in developnent is inevitable, its persistence is viewed as undesirable for fie development process. 'I‘tereisgeneralconsensusfinatfineproblemof inequalitiesin developnentcam‘otbeindirectlytackledfinroghfiedynamicsoffie growth or market forces. Rafier, inequalities in developnent must be directly addressed through public intervention as well as adoption and implementation of appropriate policies. As Bruton (1985, p.1100) aptly asserts: Wemust firdwaystodesignadevelopmentprocessthat includes fiese [economic] variables as part of fie [development] process itself . . . [Also] wemustseekadevelopnentprocessthat inncludes, as an interent part, acceptable rates of growth . . . alog wifin an acceptable distribution of [developnent] oltput. Trereisalsogrowingcontentionthatbalancerafierfinan inequality is more desirable for developnent. Friedmann (1966) contends finat inter-regional balance are essential coditions for national developnent. Amongst some of fie major reasons which Friedmann has given for fie undesirability of spatial inequalities are: 87 (l) fiepowerfulregion ‘reducesfierestoffiespaceeconomy to fie role of a tibutary area' (p.99); (2) fieperipteryis ‘drainedof itsresources, mannpowerannd capital (due to migratim)‘ (p.99); (3) finat once fie dualistic center-periplery structure is establisred, ‘fie unnrestained forces of a dynamic market economyappearstobeworkingagainstacowergenceoffine center and periphery (ID-18); and, (4) a lasting center-periplery relation leads to ectene inequities . . . and tends to restict developnent (p.99). SincefieexperiencesofmostThirdWorldcomtieshaveshownfinat inequalitytendstoincneaserafierfinandecreaseovertime, Hicks and Steeten (1979, p.568) have finerefore remarked: "Inequaliw and povertywerefonndnottobeanecessarycoditionofgrowfinand indeed were often an obstacle to it." Trere are also growing views in fie literature that blind acbption of western-oriented concept of developnent should be rejected. 'megeneralargument forwardedhasbeenfinatwhatworks for fie western developed counties may not now be suitable for fie Third World connties' needs. Seers' (1981) echoes such view. He has asserted that developing connties, as late-comers in fie field of developnent, essentially face problems which are of different nature fromfieearlystarters. Seersalsodisagreedwithfietotaladcption of fie concept of stages of development finat is implicit in fie economic growth concept of development. Sears (1978) argued finat the 88 linear view of developnent ruled out options for different styles of developnent as well as fie fact finat development problems and needs arenmniquetoeachsocietyandnation. Studies on inequality in developnent in Malaysia are relatively of recent origin (see Mahangas, 1982). This can, perhaps, be attributed to fie fact that concern for inequality in deveth was first implicitly recognized as an important developnent policy issue wifin fie incorporation of poverty eradication policy and programs in fie First Malaysia Plan, 1966-1970 (see Abu Asmara, 1982). It was only with fie enunnciation of fie New Economic Policy in 1971, as incorporated in fie Secod Malaysia Plan, 1971-1975, finat concern for inequality in developnent became more explicit. Studies on inequalities in developnent were often from fie perspective of development finat emphasized performance in terms of economic growfin (Geog, 1979: (Inn and Wan Abdul Halinm, 1979; Anannd, 1983; Aris Ofinman, 1984). Studies finat attempt to assess inequality based upon fie molt-dimensional character of development is relatively few. Tie prevalence of studies on inequality in developnent in terms of economic growth were, in fact, self- explanatory of fie prominent part which fie economic growth model play in fie national development stategy of Malaysia before fie enunciation of fie NEP (see Muniappan, 1982: Aris Ofirnan, 1984). that studies that have examined Malaysia's developnent performance have fond finat alfinogh Malaysia has experienced increasedeconomicgrowthfierehavebeenlittle improvementin distibutive equity (Linn, 1983; Cnnan, 1979; Cheong, 1979; Mahangas, 89 1982: Aris Ofirman, 1984). 'Ihisstudyisundertakennotonlytoexaminefiepatternof spatial developnent between 1970 and 1980 but also to establish wlefierregional inequalitieshavebeenreducedoraccentuatedover time. Sincethisstudyusesconposite indicatorstomeasure developnent, fiefinndingscanalsobeusedtoconparewifinecononic growth figures to indicate whefier development has beefitted or not from growth. GIAPTERTHREE mm This chapter elucidates fie mefinodological approaches and proceduresenployedtonundertakefinisstudyonpatternsofspatial developnent and magnitude of regional inequalities in Peninnsular Malaysiafrom1970tol980. Tlefirstpartoffinischapterisdevoted to explanations of fie focus of fie study, fie time period involved, fievariables selected, fieunit of analysisusedandfiefieoretical justification of fie variables chosen for this study. T'Ie secod part covers fie procedures enployed for deriving weights of fie variables aswellaspresentsfieresearchtechniquesusedtodetermninefie pattern of spatial developnent and fie extent of regional inequalities in Peninsular Malaysia. Focus of fie Study Thisstudyisconcernedwifinfiequestionoffiespatial distibution of developnent wifinin a developing country. Specifically, this study examines fie problem of spatial inequalities in development as fiey were found in Peninsular Malaysia between 1970 and 1980. Bofin synchronic and diachronic analyses are used to investigate fie pternomenon. Tte choice of Peninsular Malaysia, instead of fie whole country, Malaysia, is based on consideration of 90 91 fie availability of relevant and cotparable data. Region, as used inn this study, refers to two different types of administative units: (1) administative distict, a constituent unit of fie state, and, (2) state, a component unit of fie county. Between fie two, fie administative distict constitutes fie main focus of finis study. Time Period of fie Study Ideally, a corparative investigation and assessment of spatial developnent and regional inequalities in Peninnsular Malaysia should comenceanndhaveitsbaseyearfromatimeperiodimmediately following fie country's attainment of political independence, i.e. in 1957 for Peninsnlar Malaysia (fien lanown as fie Federation of Malaya). Ibwever, forfiepurposeofthisstudy, finisisnotpossiblesinncefie 1957 Census, as well as fie earlier censuses, did not contain sufficient details of data at fie distict level. Also, it is not possible to undertake an up-to-date examination of fie spatial variations in regional developnent in Peninsular Malaysia because of fielackofup—to-datedata. Duetofieseconstaints, thisstudy colld only examine fie problem of spatial development and regional inequalities in Peninsular Malaysia for fie years 1970 and 1980. Tne selection of 1970 and 1980 as fie base years for analysis is primarily dictated by fie availability of data. Even finough most of fie data at fie distict level have not been documented and publisred, eifier in fie 1970 or 1980 Censuses, relatively adequate comparable 92 dataatfinisregionalscaleconldbeobtainedforl970andl980from unpublisted records at fie Department of Statistics, Malaysia. Apart from finat, 1970 and 1980 also constitute important tenporel points for evaluating fie spatial aspects of development inn Peninsular Malaysia sinnce fiey marked two different eras of developnent. 1970 was fie terminal point of Malaysia's developnent stategy finat Ieavily emphasized on rapid national economic growth, while 1980, as reflected byfieadoptionofandcommitmenttofieNechonomic Policy (NEP), witnessed a decade of relatively more consciols efforts on fie part of fie Malaysian Government, to address and redress regional disparities in developnent, as well as to pronote and foster a more equitable distibution of fie beefits of development . Tre use of two time dimensions, 1970 and 1980, differentiates finisstudyfromsoneoffieearlierstudiesdealingwifinfie inequality issue in developnent in Peninsular Malaysia. Tte earlier researcneswereeifierbasedcnonlyoetimeperiod (examples, Abdullah, 1979; Annand, 1983) or had fieir year of annalysis ending in or before 1970 (examples, Leinbach, 1972; Geog, 1979). "Since fine pattern [of developnent] is always changing, corparisons which ignore Ere may be of a limited value" (De Solza and Porter, 1974, p.26) (emphasis added). 93 Data for the Study Beingacomplexconceptandmulti-dimensional incharactermakes developnent a difficult phenomenon to define and measure. However, if it were accepted finat develoment broadly connotes a complex societal tansformation involving relatively improved and sustained social , economic and physical well-being, finen it would, at least, be possible tovisualizeandoperationalizefiephenomenonofdevelcpnentonfie basisoffinosedimensionsandattributesthatareperceivedtobe salient constituents of developnent. From such a conceptual frame- work,- developnent is also recognized to have multi—dimensional attibutes. Since "development . . . is multidimensional, involving changes in stucture, capacity, as well as ontput" (Easter, 1972, p.1), finen, as Horn (1984, p.178) aptly states: "Developnent . . . therefore requires indicators from many fields of human activity. " (Also see Forde, 1968; Takamori and Yamashita, 1973; Knox, 1974; Core, 1984; Smith, 1982). In view of fie abstact and subjective nature of developnent, the variables used to measure it are, fierefore, at best, merely indicators or surrogates for developnent. Tie choice of indicators of developnent for this study were based on considerations of (l) fineir fieoretical pertinence to developnent, (2) fieir use or application in ofier enpirical studies on development, (3) availability of conparable data at both fie distict and state level, and (4) availability of data for both 1970 and 1980. 94 While it is tue finat, fieoretically, fiere can be as manny dinmensionsandcriteriaofdevelopnentasfierearewaystoperceive developnent, in practice only a limited number of fie many dimensions of developnent cann be operationalized for fie purpose of measuring developnent performance. As Colman and Nixson (1978, p.6) states: "This multiplicity of possible indicators for any given . . . dimension of developnent simply corpouds fie problems arising from fieenistenceofssveral . . . dimensions." Tteabseceorlackof documented data, published or nunpublisned, for certain dimensions of development, fie absece or lack of documented data at fie distict scale of observation, and also fie difficulty of obtaining conparable data 1970 and 1980 limnit fie scope of fie developnent dimensions finat can be operationalized and used in finis study. As Scott (1979, p.453) aptlystates: "fiekinndsofdatacollected . . . dependtoscmeextent on fie available sonroes." Inn view of such oonstaints, aspects of development such as politics, rnutition, environment, attitudes, etc . havenotformedpartoffievariablesusedinfinisstudy. Smifin (1973) and Oyebanji (1982), for example, clearly recognized finat paucityofdatamakesit impossible forannysingle study, eveninthe advanced counties, to include every necessary variable. Bothinputanndoutputtypeof indicatorsofdeveloglentareused infinissnndy. Thisisbecausebofinareimportanttodevelopment. Most ontput indicators are also inputs. As stated by Hicks and Steeten, 1979, p.572): "For purposes of assessing policies and monitoring performance [of developnent], bofin sets of [input and ontput] indicatorsarenecessary." Incaseswheredataforcertain 95 indicators of development are not readily available or are difficult tcaoquire, estimateshavebeenocmputedtofillsuchdatagalis (see Appendix 1 for relevant notation). Thirty variables, referred to as indicators of developnent are used in fie analysis. Data for fie selected variables were derived fmmseveralsolrces; finemainoesbeirgfronfie1970and1980 Population and Housing Censuses of Malaysia and related documents, unpublisled records at fie Department of Statistics, Malaysia and Malaysia's Five-Year Plan documents. Also, part of fie data relating toweights forfievariableswerederived fromasurveyconducted mug a selected parel of Malaysian professionals and scholars who werestudyingorworkinginfieUnitedStatesdurirgfietimefine survey was undertaken. 'I'leoretical Justification For Variables Selected Variableschosenforfinisstudyarebaseduponfieirfieoretical relevance to fie development process, especially as finey are pertinent to Malaysia, as well as fieir general application by ofier empirical studies on developnent. However, lack of suitable and comparable measures of development, especially at fie district level , inevitably limitsfiesocpeofvariablesfinatcanbeselectedandusedinfinis study. As a normative concept, developnent has been perceived arnd oonceivedinmanydifferentways, varyinginemphasisandsubstance fronnationtonationandfrontimetotl'me. Fiscaloonstraintsare 96 also considered a major determinant in fie articulation annd shaping of matters of priority inn development, especially in times of global economic recession. Recognizing finat fie conceptualization arnd prioritizationofdeveloprentismiquetofiecircunstaneesandneeds prevailing in a given country, it is, fierefore, only logical to assunefinat forannyconmntry, atanygivenpointintime, thesetof criteria that are important for gauging fie performance of developnent will nnot necessarily be similar. It should also be nnoted finat even fincugh certain set of criteria for measuring developnent performance have coeistontly been utilized byhmcst countries, that does nnot, however, necessarily mean that fie process of developnent is unniform for all countries. Rafier, it means finat finose criteria that have been repeatedly utilized are considered by many countries as being fundamental to fie developnent process. Bearing in minnd finat developnent is a multi-dimensional process innvolving numerous dimensions or aspects (Easter, 1972; Seers, 1972; Todaro, 1977a; Colman annd Nixson, 1978; Horn, 1984) performance of regional developnent in Peninsular Malaysia will, fierefore, be ganged by a set of variables that reflect or manifest some of finese salient aspects of developnent. Based on data availability annd also fieir relevance to fine developnent process for Malaysia, fie following indicators have been used to develop criteria of progress: (1) Education (2) Health 97 (3) I'D-15119 (4) Amenities (5) Public Security (6) Oommmnications (7) Manpower (8) Economic Develognent (9) Rural Develognent (10) Urban Development (11) Women's Participation (12) Eradication of Poverty (13) Restructuring of Society Table 3. 1 shows fie list of development criteria annd associated indicatorsasusedinfinisstudyforfieanalysisoffiepatternand of spatial developnent annd regional inequalities in Peninsular Malaysia between 1970 and 1980. It should be pointed out finat these setofcriteriabyncmeaneenconpassallfiepossibleaspectsof deveth for Malaysia. As already stated, availability of suitable indicators constrained fie generation of ofier relevant facets of development, such as enviromental, nutritional, and attitudinal aspects. Noefieless, fie set of developnent criteria finat have been developed for finis study constitute some of fie prinncipal significant concerns finat have consistently been embodied and included in Malaysia's developnent plan documents, i.e. fie Five-Year Developnent Plan. Education, formalorinformal, isinndeed, ashasalwaysbeen, one offiemajorconventionalinnstnnrentsinfieprocessofhlmnanresource 98 development. 'I‘ne educational system is not only designed to develop human resources for economic reasons, but also, as equally important, tcimpartandshapedesiredvalues, ideas, attitudesandaspirations which are conducive for fie developnent process and nation-building. 'neformerisnctonlyintermsofincreasingfieopportunityof inndividuals in fie labor market to earn higher incores (Chiswick, 1968; Bowles, 1972; Bondon, 1974; Buchanan, 1975; Simmons, 1979; Taylor annd Williams, 1982), but also in terms of improving fie capabilitiesorskillsoffielabor forceanndhennceenharncefieir productivity and ontput (Schultz, 1967; Bcgle, 1977; Streeten, 1979). According to Chandrasekhar and Hultman (1967, p.xiii), "fie poor qualityoffielabcrforoestemningfromlackcfeducation . . . retard worker mobility, annd limit [fieir] economic opportunities." Harbinson (1967, p.189) has expressed fie view finat "developnent of human resources -- i.e., fie building annd effective utilisation of fie skills of people -- is an essential element of any modern developnent strategy." an fie orfier hannd, fie innculcation of desirable national vale and social norms, which trannscennds religious, cultural, annd regional barriers, are also critical for fostering and forging national integration annd unity -- a vital element in national developnent annd nation-building of a plural society such as Malaysia . "Wifincut educational advancement, " according to Jumper, Bell annd Ralston (1980, p.27), "fie receptivity of a population to change and 99 Table 3.1 Develognent Criteria annd Inndicators Used In Analysis of Pattern Of Spatial Developnent annd Regional Inequalities in Peninsular Malaysia, 1970 and 1980. Criteria Indicators Direction* Education 1) Population with schooling + 2) Population literacy + 3) Teacl'er-pcpulation ratio + Health 4) Infant mortality rate - 5) Maternal mortality rate - I-bnsing 6) Housing stock + 7) Availability of flush toilets + Amenities 8) Piped-water supply + 9) Public-supply electricity + Public 10) Access to public law and order personnel + Security 11) Access to fire service + Communications 12) Road availability + 13) Access to postal service + Manpower 14) Availability of professional and technical manpower + 15) Availability of administrative annd al level personnel + 16) Availability of public servants + Economic 17) Availability of manufacturing activities + 18) Availability of comeroial sector + 19) Gross dorestic product + Rural 20)Accesstotelephoeinruralareas + Development 21) Access to agricultural extension services + Urban 22) Urban population + Development 23) Urban places + Women's 24) Female literacy + Participation 25) Wonen' s participation in secondary and tertiary sectors + Poverty 26) Population above poverty line + Eradication 27) Access of poor to housing + 28) Malays inn agricultural vocations - Restructuring 29) Malay participation in manufacturing of sector 4- Society 30) Malay participation in comercial sector + Note:* + meanns high positive scores are ‘good' and low scores ‘bad' - means high negative scores are ‘good' and low scores ‘bad' 100 its capacity to innnnovate are likely to stagnnate at fie lowest possible levels. " 'ne signnificarnce of education for fie developnent process has also been clearly stressed by fie United Nations (1951, p.13), which states that "fie greatest progress will occur in fiese countries wlere education is widespread." This is due to fie fact finat "fie spreadof educationmakeebcthmenandworenmorereceptivetonew ideas" (Drakakis—Smith, 1980, p.8). Thus, education is an important inetrunent of development (see Taylor annd Williams, 1982; Ahluwalia, 1976). Brimer annd Pauli (1971, p.127) have even ventured to state finat education "is fie most profitable of all social investment." Schumacner' (1973) acknowledged finat fie gift of knowledge is infinitely preferable to fie gift of material things. Tne inndicators used to reflect this aspect of developnent are (1) literacy of fie population, (2) population with formal schooling, annd (3) teacher-population ratio. 'I‘le literacy inndicatcr is a direct measureoffieliteracystatusoffiearea, whichinturnis fundamental in fie efforts to impart desired values, attitudes, ideas and aspirations conducive to fie developnent process annd overall nation-building. Also, "literacy measures fie effectiveess of the educational system" (Hicks and Streeten, 1979, p.572; also see Rac, 1984; Cole, 1981; Weeler, 1980). According to Schumannn, Inkeles and Smnifin (1967, p.11), "literacy specifically . . . open a man's mninnd to newideas [and] fieycanchangefinoseofhisattitudes." Irncreasein literacy also reflects "a distributional improvenent because fie proportion of beneficiaries has risen" (Hicks and Streeten, 1979, 101 p.571). Thus, "literacy is itself a valuable skill in fie developnent process" (Dixon, 1984, p.768). A high percentage of population literacy is desirable. Tie schooling experience indicator not only reflectsfiegeeraleducational leveloffieregioninquestion (see McGranahan, 1972, p.91), but also indicates fie potential socio- economic status of fie people in fie region; sinnce higier educational level improves fie opportunity and access of fie people to relatively better-paid economic positions, which in turn has important innfluennce upon fieir socie-economic status and well—being. Hence, of fie thirty Articles of fie United Nations ' Universal Declaration of Humnan Rights, acbpted in 1948 by fie Geerel Assembly of fie United Nations, Article 26 declares: "Everyoe has fie right to education," (cited in United Nations Social Developnent Division, 1969, p.404). A high percentage of population wifin formal schooling is desirable. Teacter-pcpulation ratio indicates fie availability of fie teaching resources, which in turnmayhavebearingnponfiegeneral educational performance. High teacrer-pcpulation ratio is desirable. Goodrealfinisavitalprerequisiteandanessentialaspectof fie quality of life finat have signnificant bearing for high levels of productivity (Hicks and Streeten, 1979; Streeten, 1979). Improvenent of lealfin coditions and fie availability of fie requisite healfin and medical facilities and services, constitute an important part of socio-economic developnent (Rac, 1984; Taylor and Williams, 1982). Tl'e state of health is an important criterion for fie prolongation of life as well as for stimulating active and productive activities of fie society. 102 Tie indicators used to reflect fie tealth criteria of developnent are (1) innfannt mortality rate, and (2) maternal mortality rate. Infant mortality rate is a sensitive indicator of the availability, utilization.and.effectiveness of the health services (Armstrong, 1966; Ntfhtruflerh 1972). "Deaths to infants during the last 11 months of fieirfinstyearof lifereflect . . . fieimpactofecononicand social coditions" (Hauser, 1959, p.101). Maternal mortality rate indicates the riSk tonmcthers during pregnancy and ohildrbirth. It is influenced by the general socio-econcmic conditions, including health and.nutrition, as well as maternal health care. In.sum, mortality rates capture differences in tealfin care, fie inncidennce of diseases and reflects fie nutritional status of fie community. Also, according to‘Hidks and Streeten (1979, pp.571-72), "measures suCh as infant mortality. . . indicatefiedegreetcwhichbasicreedshavebeen fulfilled." Hence, health indicators represent presumed causes or instruments of good.health (see McGranahan, 1972, p.93). Low mortality rates are more conducive to development. It should be noted that fie realfin indicators used only reflect coditions of health and not availability or use of health services; mainlybecauselackofdata forfieseofieraspectscfhealthatfie district level. anetheless, the indicators on conditions of health indirectly also reflect the availability and effectiveness of the health services. The nutritional aspect of health is also not included.due to the unavailability of data, especially at the district level. Hoever, fie nutritional aspect of developnent is indirectly 103 covered by fie mortality rate measures. Sl'elter, in fie form of housing, has universally and always been recognizedandcitedasoeoffiefnmdamentalbasicreedsformankind (Taylor and Williams, 1982). Tie type and quality of housing in which people are sleltered and fie facilities available in fie environnent inwhichitis locatedhaveimportantbearinguponfiegeeralquality of life and well-being of fie inhabitants (Rnoda, 1982). Access to housing as well as improvenent in availability of housing constitutes a pertinent conpoent of social and economic developnent (Rao, 1984; Higgins, 1980). Tie indicators used to mnannifest finis aspect of developnent are (1) ratio of occupied housing units to population, and (2) percentage occupied rousing unnits having flush and pour-flush toilet system. T‘te indicator of lensing-population ratio reflects fie adequacy crinadequacyofhcusingstockwithinaregioninquestion. Housing stortages reflected by this indicator imply housing needs. Alfinough thisindicatcrdoesrotindicatethequalityofhousing, fieseood indicator relating to fie availability of flush and pour-flush toilet system indirectly reflects finat aspect of rousing. As Webber (1978, p.98) states: "decent, sanitary . . . rousing is itself oe of fie salient attributes of good life." Geerally, availability of flush and pour-flush toilet system are more commonly found in fie relatively youngerandmoremodernlivingquarters. 'I‘tepresenceof flushand pour-flush toilets in fie rural areas, in a way, also reflect fie socioeconomic status of fie innhabitants, sinnce only those wto could 104 afford will have such sanitation system installed. Furfiermore, fie United Nations recognnized finat sanitation is important in contolling fie spread of diseases such as diarrhoeas, typhoid, fevers, etc. A high housing-population ratio and high percentage of occupied housing unnits with flush and pour-flush toilet system would be positive for fie developnent process. Water and electicity are not only basic needs for fie people, fiey are also essential infrastructural requirenents for developnent (Meerman, 1979). Almost every economic activity requires fie use of electricity and/or water to geerate production. The importannce of such infrastructure has been clearly stated by Busterud (Jumper et al., 1980, p.90): "infrastucture . . . innfluences fie concentation ofgrowfinwhichmnight ofierwisehaveoccurredelsewl‘ere infie region. " Also, water and electricity are vital to social advanncenent and fence improvenent in human capital investment (Chandrasekhar and Hultnnan, 1967; Taylor and Williams, 1982). As Meerman (1979, p.613) states: "As a housel'old consumption good, it [electricity] provides numerous opportunnities for activities that would cfierwise be difficult or impossible. As a meanns of production available for many possible uses, it expands fie opportunity horizon of fie rousetold or firm. " Much of the interest in water stone from its importance to public Iealfin. In mnany developing counnties, waterborne and water- related diseases are among fie major causes of deafin. Tie World Healfin Organization considers fine provision of a safe and convenient water supply to be ‘fie single most important activity finat could be undertaken to improve fie healfin of people living in rural areas' (World Bannk, 1976, p.5). 105 T‘Ieindicatorsusedtoreflectfieinfrastructuralaspectof development are (l) occupied housing unnits that have piped-water supply, and (2) occnnpied housing units that have public-supply electricity. Bofin indicators also reflect fine extent of public invesnmnent for social and economic advanncenent. A high percentage for both indicators would be desirable for developnent. Publicorderanndsecurityareimportantmattersofsocialconcern (Rao, 1984; Taylor and Williams, 1982). Public safety is considered a necessary prerequisite for fie survival and developnent of societies. In fact, safety needs is considered by Maslcw (1970) as oe fie five levels of needs for humnan developnent. Potential economic investors arealsoconcernedcversecurityoffieenvironnent. Theindicators used to denote public order and security are (1) ratio of public security enforcement personnel to population, and (2) ratio of fire service personnel to population. A higln ratio for both indicators would be desirable for developnent. Connnunnications and telecommunications, in terms of connectivity and interaction, alfiough not a direct measure of fie level of living, forms part of fie necessary infrastucture for national developnent. Thisisrotonlyintermsoffieertenttowhichcommnnications facilitate fie flow of people, goods, services and information between places (Cole, 1981, p.48), but also of fie interaction between people in different places. Coates, Johnston and Knox (1977) View communications not only in terms of reduction in ‘distance friction', but also in terms of ‘cost reductionn'. It was implied by Losch (1954) 106 finat residents in location wtere tansportation network is relatively poorwould, fieoretically, have lowernet innconesincefieymustpaya higterproportionoffieirincomeintanspcrt, goodsandservices (Nbrril annd Wohlenberg, 1971). This has been edorsed by Ccates et al. (1977, pp.3-4) who state: "T'te cost of movenent are an elenent in real incones: fie lesstavel involvedinmovingyoutcwhatyouwant and/orinmovingwhatyouwanttcyou, fiemorecfyourinconeyoucan sped on goods and services ofier finan tavel". Conmmnnnicatios also costitutes an important media for ‘tansfer of information and ideas', which Zimmerman (1951, p.10) regards as "the mother of all cfier resources." According to Pryor (1973, p.53), "fie flow of innfornmation, in its broadest sese, is directly related to fie degree of elaboration and linkages of fie . . . conmnunications systens." Chmmicatios also facilitates diffusion of information and innovation through interaction. As Pedersen (1978, p.310) states: "fie speed of diffusion inncreases when fie fie interaction grms." Tie availability and provision of an efficient linnkage or conmmnication system, fierefore, is an important integral part of the development process. he existence, as well as improvement or expansion of fie conmnunication networks could, potentially, serve to facilitate and accelerate fie developnent process. Tle indicators used in finis study to reflect fie communications aspect of development are (1) road density, and (2) ratio of postal service worker to population. Tie indicator road de'sity reflects not only fie potential flows of people, goods, services and information but also fie extent of 107 accessibility of the population to other areas. Road density also bears significance for the exploitation.and.marketing of resources. "Road.transport infrastructure is one of the crucial factors of development; its spatial pattern is closely related to that of developnent as a wrole" (Smith, J.A., 1974, p.308). Mabcgunje (1981, p.293) is also of fie view finat "tansport is an infrastuctural element with profound.implications for overall development." The indicatoru O/.'¢: \"(l‘ Dun." \.. F r.- K. JIIIMUI II “"‘Ru' ‘1 ' . F4. 00' 0°. 3' ) " .3' co .. \ : ..... 0° YE);N‘?5::..M/\b.o:{f lulu ‘bllon :‘ '0‘!” a E_GER1 SEMBLN/ 2{:¥./ 1' n In “\u‘ "K’ ”I )0 I I V I. u \ I}. Wilayah PorquluIn [‘3‘ k: J/OHOR I: IIIIIII. 2" (clung --.. ..—- ,2. 00 I’m. hm) F" 00' LN -'A_l\ Km Haul u“ JIM: \Ichu 100° 00' 101' 00‘ :01‘00' ' nor'oo' My figure 3 4 Ammustrative Districts in Peninsular Malaysia 1980 125 Malaysia, 1970 and 1980 -- For Oanparative Use', Appendix 4 on ‘Districts in Peninsular Malaysia in 1970 and 1980', and Appendix 5 on ‘Districts in Peninsular Malaysia, 1970 and 1980 -- For Comparative Use'). Weighting of Variables Bearinginmindthattheprocessof developnent "isnotauniform linear progressim of a set of variables but rather a changing canplex of factors which nova at different rates in relation to each other . . . at different levels of develoment and in different types of country" (UNRISD, 1972, p.4), the variables chosen for this study have been assigned ‘weights' to indicate the different degree of influence or oontributionwhicheachofthesevariables exertsintheprocessof developnent (Colman and Nixson, 1978). As aptly remarked by Islam and Khan (1985, p.150), "weights should be attached to . . . indicators [of developnent] rather than treating all of them equivalently." Undoubtedly, thisprooessofderivingweightswill notbeaneasyand uncontroversial task. Generally, scholars and practitioners share the view that the process of assigning weights to variables is very much a subjective or judgmental process, and as such it is unlikely that consensus can easily be achieved. In any case, every decision-making act inevitably involves a subjective or judgemental procedure (Davidoff and Reiner, 1978). For the purpose of this study, weights for the variables have been derived through a ‘Value Judgement Survey' from a panel of 126 Malaysimprofessimalsarrisdnlarsmrkingorstudyinginthetklited States (seeAppendixGonsxmreydocumentsused). mthebasisofa paired-conparison technique, every respondent was asked to indicate how each of the (selected) variable conpares to every other variable usedinthisstudyinternsoftheirrelativeinportanoetothe process of development. This was done by distributing a total of ten (10) pointstoeachpairof indicatorsbeingoanpared. Thismeans that foreveryoellinthe30x30matrix (seeAppendix6) therewould betmvaluesgivenbyttfirespondentthatsmtstoten. Avalue distribution of ten for a particular variable and the autanatic 0 (zero) fortheothervariableinthepairbeingocmparedueansflmat the former variable is regarded by the respondent as being absolutely more desirable to developnent than the other variable; 0 (zero) means absolutely uninportant to developnent. m the other hand, a value distributimofSandeoreachofthevariablesinthepairbeing oanpared indicates that the respondent considered these two variables to be equally important or desirable to developnent. The different carbination of values assigned to each pair of variables being ompared reflects the relative degree of importance of the variables to developnent. The relative weight for each variable was derived using the following procedures: (1) values given to every variable by each respondent were smuedandthenexpressedasapercentagepointofthe total points for all the variables; 127 (2) the percentage point for every variable [as derived from (1) above] foralltherespondentswerethensmmed: (3) the total points for each variable by all respondents [asderivedfmn(2) above] werethenccnputedasapercent of the total points for all the variables by all mm. ‘Ihevaluesobtainedinstepnumber(3) abovewouldbeusedasweights for the variables. Research Techniques Inordertoexamineandassessthepatternof spatial developnent as well as the magnitude of regional inequalities in Peninsular Malaysiaitisneceasarytouseprocedureswhichcanreflectand measure the pattern of spatial developnent as well as indicate the extent or magnitude of regional inequalities in developnent. While it is true that developnent is a multidimensional phenanena that is not always possible to quantify, some process of quantification is needed inordertobeabletodeterminethepatternof spatial deveth and the magnitude of regional inequalities in developnent. Allthe30variableswereusedinexaminingandassessingthe pattern of spatial developnent as well as the magnitude of spatial inequalities in developnent in Peninsular Malaysia between 1970 and 1980. Two types of analytical approaches are employed: (1) examination of spatial developnent and inter-regional inequalities in development in 1970 and 1980, respectively: 128 (2) canparisms of spatial developnent and inter-regional inequalities in developnent between 1970 and 1980. The first approach, called synchronic analysis, involves "comparisons at a given point of time . . . of the situation in different regions of the same camtry." (Horn, 1984, p.179). The second approach, called diachrcnic analysis, is based on "canparisons of change over time" (Horn, 1984, p.179). Tbenablebcththesynohrauicarridiachronicanalysesofregimal developmt and inter-regional inequalities in developnent in PeninsularMalaysiatobemdertaken, sane formofivaluemustbe producedasayardstick formakingtheassessmentandconparison. For this purpose, the study will generate a composite value or index of development. Using the 30 variables as indicators of developnent, four major steps were taken to produce the canposite value of developnent. Firstly, all the data, for all variables and regions, were controlled for their population. Secondly, since the selected variables are in different Wpes of measures the data are transformed into standardized scoresinordertocbtainadistributionnoreamenabletoanalysisand conparability. In this regard, Johnson (1978, p.12) indicates that: "The value of z-scores . . . allow comparison between observations of unlike things in terms of their relative position in the set of observations" (also see Bracken, 1981). Standardized-scores, for each variable are calculated with the following formula: 129 5131 where: Zi= standardizedscorevariablei Xi=observedscoreoftheithvariable 21: meanofvariableXi SDi = standard deviation of variable Xi TheabovefornulaproducasresultssuduthatwhenXi=§thenZi=O and warm Xi = SDi then Zi = 1.0, irrespective of the variables' units of measurement. The z-scores produced indicate how far any observaticnisfrcmitsmean. AZ-scoreofO (zero) canbe interpreted as being equivalent to average condition of developnent performance, a negative Z-score wcmld indicate below average condition of developnent performance, and a positive z-score connotes above average cmdition of developnent performance (Smith, 1982, p.71). Ihetiairdsteptowardsproducingtheccnpositevalueorindexof developnent involves the derivation of (relative) weights for all the variables used in the study. As mentioned earlier, this was done through a ‘Value Judgement Survey'. 01 the basis of the values given to the variables by all the respondents, computations are made to obtain weights of the selected variables used for this study. The weights obtained for each variable are then would multiplied to their correspaading z-scores. 130 Itefanfluandfinalmajorstepingeneratingthecaupositeindex of development is the aggregation of the performances of all the indicators of developnent. 'Ihis approach is well-established in spatial analysis and is called the Standard—Score Additive Model (Smith, 1977; Oyebanji, 1982; Knox, 1982). The umeighted and weighted index of developnent for each district or state is obtained using the following formulas: (a) Umeighted canposite index of development: n DIj =ZZij j=l...m i=1 (b) Weighted canposite index of development: n DWj =2 WiZij j=1...m i=1 DI j = unweighted canposite deveth index for region j DWJ = weighted canposite developnent index for region j 21:] = the standardized score for region j of variable Xi Wi = the weight of the i th variable The index of developnent can be interpreted as follows: 0 : average developnent performance < O : below average developnent performance > O : above average developnent performance 131 The index of development indicates not only the level of. developnent for each region, but also reveals the degree of variations in the level of developneut between regions. Synchronic analyses of theindicesofdevelopnentareusedto:(l)assessthe1evelof developnent for every regim in Peninsular Malaysia for 1970 and 1980, (2) exanineandevaluatehowtheregimscanparetocneanotherin terms of level of developnent for each point in time, (3) establish the differences of developnent between urban and rural regions, (4) identify the categories of urban regions with relatively higher level of developnent, (5) establish differences between western and eastern districts, and (6) determine how different geographical units affect variatims in level of developnent. Diachrmic analysis of the indices of developnent establishes the charge of inter-regional inequalities over time. This is determined by examinirg the spatial pattern of developnent between 1970 and 1980 usirg indices of developnent. Also the gap between most developed and 1aggirgregionsisexaninedforboth1970arri1980. Thechargeinthe patternofspatial developnentisexpressedbythechangeinthe number and type of districts having above average developnent performance. The analysis tries to answer questions such as whether more rural districts have registered development performance that is above average, whether more districts from the eastern coastal states of Peninsular Malaysia have shown above average developnent performance, and whether districts that did not fare well in developnent performance in 1970 have remained in the same position 132 vis-a-vis other districts . Examination of the gap between regions with high index of developnent and the rest of the regime, for 1970 and 1980, reveals the differences between these regions. A t-test of significance is undertaken to establish the significance of the charges over time. Synchronic exanination of the level of developnent tests the hypotheses that spatial disparities in developnent exist at both the state and district scale of analysis. Diachronic analysis, on the otherhand, teststhedegreeofchargecvertime. It should be pointed out that the Standard Score Additive Model usedinthisstudyisnottheonlytechniquethatcouldbeusedto derive the standard-scores and the canposite index of developnent. Othernoresophisticatedtechniques exist, suchasPrincipal Canponent Analysis and Factor Analysis. These techniques are particularly useful when the number of variables used are relatively large because they help to collapse variables into a few canponents or factors and also deal with the problen of multi-collinearity amorg variables. However, the application of Principal Cauponent Analysis or Factor Analysis techniques would pose problems when applied to (simultaneous) camarisaswertire. 'Ihisisduetothefactthatthevariablesto becanpared, forinstameovertwotimeperiods, wouldnotnecessarily collapse into the sane canponents or factors, therefore makirg the canparisa'x impossible. 0n the other hand, the Standard-Score Additive Model allowssinultaneouscanparisontobemdefcrtwotineperiods. Itsmajorshortcanirg, however, isthatitdoesnotaddressthe problem of multi-collinearity among the variables used. The Standard- 133 ScoreAdditiveModelisalsousefultpexpressquestionsof develoment or disberefits finrough the use of positive or negative signs. For instance, if fie nature of the variable indicates positive directian for developnent, its z-score is preceded by a positive sign. If fie variable indicates negative direction, its z-score is preceded by a negative sign. Exaninirg how fie variables are distributed spatially also provides a useful understanding of fine nature of regional disparities in developnent. Such information is useful for developirg policies to redress regional irequalities. Tre Gini coefficient technique is enplcyedinthisstudytoirriicatehowaparticularindicatorof developnent is distributed spatially. As Smith (1982, p.28) states: "Gini coefficient is fie most cannon gereral measure of inequality in distribution." 'ne Gini coefficient is defimd as fie ratio of fie areabetweenfielorenzCurveandfielireofequalitytofieareaof fie triargle below finis lire. It varies between 0 and 1. Zero means perfect equality and 1 means perfect irequality. 'I're greater fie departureoffielorenzcurvefranfiediagonalfielargerfievalue of fie Gini coefficient (Anand, 1983: Alker, 1970). Tle Lorenz curve provides a useful graphical illustratian of fie distributian of a particular variable over space and time. It is used l'ere to exemplify tle distribution of selected indicators of developnent. Carparisans of Gini coefficients for fine various regions indicate whefier particular develoment variable has narrowed or widened over 134 time. Infinisstudy, theGinicoefficientsareusedtocanparefie distribution of variables for fine follavirg cases: (a) (b) (c) (d) (e) (f) all states, 1970 and 1980 all districts, 1970 and 1980 all urban disticts, 1970 and 1980 all rural districts, 1970 and 1980 all western disticts, 1970 and 1980 all eastern districts, 1970 and 1980. W, ratios of Gini coefficients are calculated to canpare differences over regicns and tine. Gini coefficients are derived by fie followirg steps: (1) (2) (3) Variables to be used in Gini coefficients are selected. Infinisstudy, all30variablesareused; Reference variables are selected to be used in fie canputatian of Iocatian Quotients (see Appendix 11 for details); Locatian Quotients are calculated for all regime usirg fie followirg formula: V3 )2 VJ LQJ' = Bi 2 Bi wtere: LQJ = Location quotient for region j VJ = Value of a developnent variable in region j Z Vj = Sum of fie developnent variable for all regions 135 8:] = Reference variable value for regianj Z 83 = Sum of fie reference variable for all regions (4) Regiansarerarkedinascendingorderaccordirgtofieir Locatianouotientvalues; (5) Gini Coefficients are obtained usirg fie following formula: Gx=z (YJX -Y Xi) 9? Gini coefficient for variable x «25. ll fie cunnlative share of variable X correspondirg to regionj Yj fie cumlative share of fie referees variable correspondingtoregianj 0e type of Gini coefficient ratio is calculated to determine wtefier fie gap of spatial irequality has been narrowed, widered, or remaired fie sane. Tre followirg pairwise canparisans are perforned: (1) states and districts (2) urban districts and rural districts (3) western districts and eastern districts. When fie Gini coefficient ratio is close to unity fie two sets of regime are almost similar in fie more of fieir spatial irequality. To fie extent that finis ratio is different fran 1.0 regional differeees exist. 136 A second type of Gini coefficient ratio is calculated to slew chargesof irequalityovertine. ThisisdorebydividingfieGini coefficients of 1980 over 1970. This type of Gini coefficient ratio indicatesrechargeintinewhenfieratioisequaltolfl. Aratio ofnerefinanlisregardedasaneasureofdivergence, whilearatio of less finan l is treated as an indication of convergeee. Analyses of Gini coefficients and Gini coefficient ratios will attempt to test fie followirg hypofieses: (1) That spatial irequalities in develoment exist at both state and district scales of analysis; (2) That regional irequalities in developnent are relatively Here acute at the district rafier than the state scale level; (3) That regional inequalities in developnent are relatively more acute for rural fien for urban districts; (4) That regional inequalities in developnent are relatively moreacute fordistrictsinfieeastcoaststatesthan fieseinfiewestcoaststates; CHAPTERPUJR PATTERN OF SPATIAL DEVEIDPMENT AND WITUDE OF REX-31m INEUJALITIES IN PENINSULAR MAlAYSIA, 1970-1980 Introductian Thischapterpresentsfieanalysisofdataperforrredtoidentify patterns of spatial developnent and to determire fie magnitude of regional irequalities in developnent in Peninsular Malaysia for 1970 and 1980. Bofin synchronic and diachronic analyses are undertaken. Resultsoffieanalysesarepresentedatbothfiedistrictandfie state scales, wifinfiefonrercaetitutirgfieneinfocusoffinis study. Tlefirstpartcffinisclepterisdevotedtofieexaninaticn and discussian of fie pattern of spatial develoment in Perinsular Malaysia, while fie second part is concerred with fie extent of regicnal irequalities. Appropriate answers to fie research questions posedinthisstudyaremadeinbothparts, but specificallythis chapterattenptstoaddressappropriateresponsestofiehypofieses advareed for this study (see Gnapter (he). Tie extent of inter-regional irequalities in developnent in Peninsular Malaysia will be analysed in two fundamental ways: (1) By examinirg fie level of development of fie various districts and states, and, (2) By examining few fie various developnent variables are distributed spatially. 137 138 Pattern of Spatial Developnent Trepirposecferaminirgfiepatternof spatial developentin Peninsular Malaysia between 1970 and 1980 is to find out how developnent is spatially distributed amag fie various administrative districts. This is to reveal fie performeree of developnent, spatially, and. to facilitate identificatian of regions, within states, whicharerelativelynoredevelopedornot. Findirgs fransuch examination can offer useful guidaree in addressing solutions to problems of irequality. As Coates, Johnston, and Krex (1977, p.2) have aptly stated: "Solution [to any problem] requires understandirg." Tre practicality of investigatirg fie pattern of spatial developnent hasalsobeenendorsedbyPeet (1977), wtehasremarkedthat understandirg of fie pattern of spatial developnent canstitutes a useful initialstepinfiesearchforprocessesthathavegeerated fie pattern and problem. In a similar vein, Rreda (1982), in alludirg to fie regional develoment problem, has also aptly pointed out that understandirg fie pattern of spatial developnent forms an important step towards fie formulatian of appropriate regional develoment policies and strategies. Index of Developrent An index of developnent is developed for every district and state, and is used to exanire (1) fie extent of spatial developnent for every district and state in Peninsular Malaysia in 1970 and 1980, respectively, (2) fieir relative position or rankirg vis-a-vis one 139 arefier in fie hierarchy of spatial developnent, in 1970 and 1980, (3) fie relative developnent gap between between districts or states, and(4)wrefierfiechangeofdistrictsistavardhigrerdeveloprent levels. While it is geerally agreed that developnent embodies a number of aspects, problers of caeeptualizatian and operationalizatian limit fietypeandscopeofvariablesthatcanactuallybeutilized. In this study, 30 variables, referred to as indicators of developnent, areused to determire fie level of development for every district and state in Peninsular Malaysia for 1970 and 1980. Table 3.1 and Appendix 1 describe fie selection, fie sources and fie definitim of fie indicators of develoment. As mentianed in Gnapter Three, since different variables are treasured in different units standardized z-scores are canputed for every developnent indicator. All z-scores were assenbled in matrix format as follows: (1) a 64 x 30 matrix for the canparable districts in 1970 and 1980 (2) a 70 x 30 matrix for fie districts in 1970 (3) a 78 x 30 matrix for fie districts in 1980. Trelasttwonetriceswereforfiedistrictsasfieyactually prevailed in 1970 and 1980, respectively. Tress matrices were used to canstruct regional develoment indices as per fie Standard-Score Additive Model. Specifically, fie model applied has fie followirg structure: 140 DIszoplit+Scteol+Teacrer-Infanert—Mafinort+fisgstock + Toilet + water + Electrcy + Pubsec + Firesv + Roadden + Postalsv + Proftech + Adminmag + Govtoff + Manuftg + Ctnneroe + GDP + Ruraltel + Extsnsv + urbanpop + urbanpl + Femalit + Famecona + Abovepov + Plcwchsg - Nayagric + Nuymrtg + Mlyaann: D13 2 Composite development index for region j Poplit = Percent population 10 years and above who are literate School = Percent population 5 years and above with formal schooling Teadher = Teachers per 1,000 population 5 years and above Infamort = Infant (under 1 year old) deaths per 1,000 live births (by occurrence) Ntnncrt: = Maternal deaths (due to pueperal causes) per 1,000 live births (by occurrence) HSgstodk = Occupied housing units per 1,000 population Toilet = Percent occupied housing units having flush and pour-flush toilets water = Percent occupied housing units having pipedswater SUPPIY Electrcy = Percent cccupied.housing units having public-supply electricity Pubsec = Public security personnel per 10,000 population Firesv = Fire service personnel per 10,000 population Roadden = Paved road.mileage (in.miles) per 10 square miles Postalsv = Postal service personnel per 10,000 population 141 Proftech Percent labor force 10 years and older in professional and technical group occupations Adnirmag = Percent labor force lOyearsandolderin adninistrative and managerial group occupations Govtoff = Government officials (ireludes only fiese in legislative, administrative occupatians and fiese in executive positions) per 10,000 population Manuftg = Percent laborforce lOyearsandolderin manufacturirgsector Camerce = PercentlaborforcelOyearsandolderincamercial sector GDP = Grossdcnesticproductpercapita(estimated) Ruraltel = Public rural telephore booths per 10,000 rural population Extsnsv = Agricultural extension station acreage (in acres) per 1,000 labor force in agriculture Urbanpop = Peroent population in places 10,000 and above Urbanpl = Populatian in places 20,000 and above as percent of total urban population Fenalit = Percent fenale population 10 years and older who are literate Fetecona = Percentwatenlabor force lOyearsandolderin rein-agricultural econanic activities Abovepov Percent labor force 10 years and older with nean monfinly ireare (Malaysian) $300/= and above (estimated) Plowehsg = Public low-cost reusirg units per 1,000 population below poverty-lire, i.e. having nean monfinly incare (Malaysian) 8300/: and less (estimated) Mlyagric = Peroent total Malay labor force 10 years and older in agricultural vocations Mlymftg = Malay labor force 10 years and older in manufacturing sector as percent of total labor force in manufacturirg Mlyoalmc = Malay labor force 10 years and older in camercial sector as percent of total labor force in camerce 142 Tle above nodel provides fie unweighted canposite developnent index for each regian. T're weighted canposite deveth index is obtaired by nultiplyirg fie value of each of fie variables by its correspandirgweightderivedfranfieparel survey. Itshouldbenotedfinatfiez-scoresareaggregatedusirgfie appropriate signs dependirg an whefier or ret fie variables indicate positive or regative oantributions to developnent. For instaree, in fie case of infant nortality, siree larger rates of nortality indicate less developnent, fieir signs were reversed before aggregatian. In finiswayfiedesiredorundesireddirectionoffiedevelopnentprocess for all variables were properly accounted for. Asexplairedincnapter'rhree, fieadninistrativedistrictsin PeninsularMalaysiacanstitutefiemainunitof analysisinfinis study. Tte utilizatian of districts as fie unit of analysis seen nore practical siree it deals with fie smallest spatial framework within which developnent projects are planted and implerented in Malaysia. Figure 4.1 illustrates fie plannirg process fran fie natianal to fie district level. . Treindexofdeveloprenthasbeenproducedforbothfie carparative districts (and states) as well as fine non-canparative districts (and states). Canparative districts refer to those disticts (or states) for which canparisan of deveth performance between 1970 and 1980 were possible. Appendices 3 and 5 list fie nane of districts and states that have been collapsed so that canparisons are possible. T'rere were a total of 64 canparable districts and 11 canparable states 143 for both 1970 and 1980. Consolidation was needed because boundary and areal charges occurred siree 1970. As indicated earlier, in 1970 fierewere70districtswhilein1980fieirnumbersincreasedto78 (seeAppendicesZannd4). Moreimportantstill, fieareaofsaneof fie districts also charged: fierefore, it would not be possible to make fair oanparisans between such districts. It is precisely because of this diffianlty finat merger of certain districts was necessary. Tteindexofdevelognentwasalsoproducedforall districtsand states in Peninsular Malaysia for 1970 and 1980. However, direct canparisons of fiese indices of developnent is not possible; ret only becauseofdifferereesinfienumberofdistrictsorstates, but also becausefieindicesofdevelopnentforfiesetwoperiodswerecanputed fran different data bases. Unweighted arnd weighted indices of developnent were geerated for this study. The arnalysis of unweighted indices of development is fine central focusofthisresearoh. Weightedindicesareusedtotest wlefier or not final results charge wten different degrees of importanee are attacked to the developnent variables. Tie geographical distribution of deveth is shown by mapping fie various indices of deveth for fie districts in Peninsular Malaysia, for 1970 arnd 1980. This is done by partitioning fie 144 CABIET {------- NATIONAL ECONOMIC COUNCIL (NEG) .T (NDPC) NATIONAL DEVELOPMENT PLANNIm CGI‘IITTEE ECOIDHIC PLANNI m UNIT (EPU) T" NATIONAL ACTION COUNCIL (NAG) EXECUTIVE CWITTEE STATE DEVELOPMENT PLANNIN} CWITTEES T__. STATE ACTION CWITTEES DISTRICT DEVELOPMENT PLANNI m CWITTEES DISTRICT ACTION COMITTEES PIANNDC PRCXIBS HTLB-BH‘ATIW PRCI‘EL‘SS Figure 4.1: Process of Policy/Program Planning and Implementation, Malaysia. Source: Elyas Cmar.(1980) p.23 or me mama-union ooonnnmrou qun (ICU) I I nnznsuar , l I 1 I MINISTRIES/STATE 1 PUBLIC AGENCIES ' "‘ CORPORATIOIS I l l 1 145 developnent irndices into four categories of developnent level: (1) Very High (nore finan plus 1 starndard deviation) (2') High (between 0 arnd plus 1 standard deviation) (3) Low (between 0 and minus 1 standard deviation) (4) Very Low (less finan minus 1 standard deviatian). Results of Survey Infinisstudyweightswerederivedfranasurveyannongstagroup of selected Malaysian professionals arnd scholars wre were working or studying in fie United States in 1986. Ideally, it might, perhaps, havebeennorerealistictoundertakefiesurveyanorgacross-section of fie society in Malaysia itself. However, this could not be done because of bofin finareial and tine constraints. Norefieless, fie surveyfinatwascarriedoutretonlyservesasfienextbest alternative urnder fie constraints canfronted, but also it cannpensates for any arbitrariness or bias finat might be introduced if weights were attenprted fran fie writer's personal valuation alore. That fie survey was conducted amang a group of acadennically qualified Malaysians, We understand fie sigrnificance of fie developnent process adds strength to fie weights derived. Tie thirty-two respandents represent different efinnic backgrounds and have different professional and acadenic experiences in Malaysia. Responsesobtainedfranfiesurveywereprocessedasoutliredin ChapterThree. A30x30matrixwas assenbledandusedtoderive weights for all variables. Table 4.1 stews fie 30 weights arranged in 146 orderoffieimportareeattackedtofinenbyfierespandents. Trerespansesobtairedindicateclearlythatfiereisawide variability in peroeptions of what is important for developnent. This variability is consistent with fie view finat perceptions of developnent vary anorgst irndividuals within fie society as well as in different natians or cultures. Twoquestiansthatwererelevanttothisstudywereaskedto (1)1-Iavwmldyouratefieissueofreductionofregional irequalities (i.e. between adninistrative districts) in Malaysia ? (2) Irmwmldywratefieroleoffiegovermentinreducing regional irequalities (i.e. between administrative districts) in Malaysia ? Respanseswereratedinanordinal scale: veryinportant, fairly important, less important, end not important. Tre survey establishes fie importance of regianal disparities in developnent as an inportant national issue. It reveals fie inportance finatmustbeattacledtofieroleofthegovermentinaddressirgfie problen of regianal disparities in developnent. In fie survey, 69 percent of fie resporndents considered fie question of reducing inter- regional irequalities as being very important. Tre ofier 31 percent viewed finat issue as being fairly important. In fie second question, 56peroentoffierespandentsfioughtfineroleoffiegovermentin reducing inter-regional irequalities as been very important. 40 147 Table 4. l Weights of 30 Selected Indicators of Development (Based an Value Judgenent Survey) * Rarnk Indicators of Developnent Weight Std. Deviatian 1. Labor Foroe Above Poverty 4.05 0.53 2. Malays in Cameroial Activities 4.00 0.30 2. Malays in Manufacturing Activities 4.00 0.29 4. Infant Nbrtality 3.93 0.87 5. Maternnal Nbrtality 3.89 0.91 6. Malays in Agricultural Vocations 3.88 0.64 7. Population With Formal Schooling 3.86 0.58 8. Teacrer-Populatian Ratio 3.79 0.45 9. Population Literacy 3.74 0.60 10. Gross Danestic Product 3.70 0.49 11. Flush arnd Pour-Flush Toilets 3.57 0.46 12 . Piped-Water Supply 3 . 54 0 . 43 13. Public Low-Cost Housing 3.51 0.54 14. Marnufacturirg Activities 3 .47 0.49 15. Government Officials 3.42 0.39 15. Public-Supply Electricity 3 . 42 0.39 17. Camercial Activities 3.36 0.50 18. Professional and Technical Manpower 3.24 0.50 19. Agrienltural Extensian Service 3.13 0.41 20. Postal Services 2.99 0.56 21. Rural Telephore Boofins 2.98 0.56 22. Road Network 2.97 0.53 22. Female Literacy 2.97 0.67 24. Female in Nan-Agricultural 2.95 0.61 Econanic Activities 25. I-busirg Stock 2.92 0.66 26. Administrative and Managerial 2.89 0.68 Manpower 27. Fire Services 2.65 0.74 27 . Public Security 2.65 0.66 29. Large Urban Places 2.51 0.76 30. Urban Population 2.02 0 88 Note: *Variableewifinsanerankirgmeansthatfineyhaveweightsof equal value. For furfier details regarding variables refer to Table 3.1 andAppendixl. 148 percent of fie respondents regarded fie government's role in reducirg inter-regional irequalities as being fairly important and fie ortler 3 percent viewed it as being less important. 1970 Development Irndex Umeighted developnent indices were calculated for fie canparative districts according to fie formula previously presented. Table 4.2 shows fie index of developnent for fie various districts in PeninsularMalaysiainl970. Itcanbeobservedthatfierearewide variatians arnd gaps in fie perfornmaree of developnent amorg fie districts. Such differences amorg fie districts indicate fie existeee of spatial irequalities. Figure 4.2 shows fie level of developnent for fie canparative districts in 1970, and Figure 4.3 for districts as fiey actually prevailed in 1970. Of fie 64 camparative districts in 1970 only 28 districts recorded positive indices of developnent. In ofier words, 44 percent .of fineee districts received more finan average share of fie beefits of developnent when cansiderirg all 30 variables. Correspondingly, 36 districts (56 percent) registered regative indices of developnent: that is, fiey had below average share of fie beefits of developnent. Table 4.3 shows that 21 of fie 28 districts with positive developnent indices (75 percent) are districts in fie west coast states of Peninsular Malaysia, namely Perlis, Kedah, Penang, Perak, 149 Table 4.2 1970 Canparative Districts, Peninsular Malaysia: Index of Developnent (Unweighted) and Ranking Rarnk District Development Rank District Develognent Index Index ( Positive ) (Negative ) l) Klang/Ulu Selargor/ 29) Dindings/Kuala Kargsar/ Kuala Lunnrnur * 52.87 Hilir Perak * -0.03 2) Pulau Pinarg Timur 30) Rembau -0.61 Laut 50.40 31) Segamat -0.89 3) Kuantan 30.60 32) Melaka Utara —0.91 4) Joker Bahru 30.02 33) Seberang Perai 5) Melaka Tegah 29.35 Selatan -1.35 6) Kinta 29.10 34) Kennaman -l.43 7) Serenban 24.64 35) Kulim -1.84 8) Seberang Perai 36) Tenerloh -2.26 Utara 12.76 37) Kuala Muda -2.64 9) Seberang Perai 38) Raub -3.11 Tegah 12.54 39) Perlis -3.54 10) Mersirg 11.36 40) Melaka Selatan -3.71 11) Tanpin 10.29 41) Pmtian -3.95 12) Kota Bahru 9.02 42) Kota Tinggi -4.36 13) Cameron Highland 8.87 43) Kuala Selargor/ l4) Pulau Pinarg Sabak Bernnam * -4.90 Barat Daya 7.72 44) Hulu Perak -6.32 15) Port Dicksan 7.30 45) Jerantut -8.55 16) Bentag 6.74 46) Lipis -8.58 17) Pekan 6.19 47) Mararg -9.08 18) Keluang 6.18 48) Largkawi -ll.83 l9) Durgun 5.52 49) Yan -12.50 20) Kuala Langat/ 50) Tumpat -l3.35 Ulu Largat * 4.89 51) Bachok -14.38 21) Jelebu 3.97 52) Besut -l4.54 22) Larut & Matarg 3.83 53) Barndar Bahru -14.61 23) Batarg Padang 3.37 54) Pasir Mas -15.41 24) Kuala Pilah 3.08 55) Kerian -15.94 25) Mnar 1.60 56) Macharg -16.69 26) Kuala Teregganu 1.04 57) Pasir Puteh ~17.10 27) Batu Pahat 0.50 58) Balirg -20.80 28) Kota Setar 0.07 59) Ulu Tereggarnu -21.43 60) Kubarg Pasu -21.89 61) Sik -22.29 62) Ulu Kelantan -23.62 63) Tanah Merah -23.69 64) Padang Terap -25.67 Standard Deviatian = 16.18 Note: * Denotes districts that have been canbined to facilitate canparisan wifin 1980 canparative districts. PENINSULAR MALAYSIA = DISTRICTS . 1970 70 0 70 L0 60 30 km Md ' . I I mulllh '7 w ' Tr i. fiiiiégti .:._—_—. ~33“? 7%;; f. , ‘v- S m . — -_ . I; _- W?E;2AEZAJ :~ ‘ -fi ‘ --IL- ;; ufl========= Development Level ~ very High ==== B§§— High 3?... , Ti ‘5‘ Ari," - .- Low “:§Ef 5'" 7"I- Very Low Figure 4.2: 1970 Comparative Districts, Peninsular Malaysia: Level of Development. 151 PENINSULAR MALAYSIA: DISTRICTS . 1970 a 20 o 20 L0 so no km W Very High High Low Very Low Figure 4.3: 1970 Districts, Peninsular Malaysia: Level of Develognafin 152 Selangor, Negri Sanbilan, Malacca, and Johore. Administrative districtsinthesestatesarehereafterreferredtoaswestern districts. Since there are a total of 42 canparative western districts, thismeansthatSOpercentof allwesterndistrictshad averageormorethanaverageshareofthebenefitsofdevelopnentin 1970. The other 7 (25 percent) are fmn districts in the east coast states of Kelantan, Terengganu, and Pahang. These districts are hereafterreferredtoaseasterndistricts. Outofthetotalof22 canparative eastern districts, only 32 percent had average or above average share of the benefits of developnent in 1970. Table 4.3 1970 Oarlparative Districts, Peninsular Malaysia: Distrihntion of Developnent Index (Unweighted) by Geographical Location and Urban and Rural Districts Index of Western Eastern Urban Rural Total Developnent District District District District Positive Development 21 7 18 10 28 Index Negative Development 21 15 2 34 36 Index Total 42 22 20 44 64 18 of the 28 districts (64 percent) with positive deve1opnent indices are urban districts. In this study, an urban district is defined as one which has at least an urban center with a population 153 concentration of 20,000 and more. Appendix 8 lists the (canparative) urban and rural districts in Peninsular Malaysia for 1970 and 1980. The 18 urban districts represent 90 percent of all urban districts in Peninsular Malaysia in 1970. Of these 18 urban districts with positive developnent indices, 14 are western districts and 4 are eastern districts. Only 10 of the 28 districts with positive developnent indices, or36percmt, areruraldistricts. Inoontrasttoanurbandistrict, aruraldistrict, inthisstudy, isonewhichdoesnothavea significant urban center (population concentration of 20,000 and more). Based on this definition, 44 of the 1970 canparative districts are rural districts. In terms of developnent performance of districts by states, 4 statesarefoundtohavemrefimanSOpercentoftheiroanparative districts registering positive developnent indices. These are the states of Negri Senbilan (83 percent), Penang (80 percent), Selangor (67 percent), and Johore (62 percent). The states of Pahang and Perak have exactly 50 percent of their canparative districts with positive developnent indices (see Table 4.4). In examining the developnent index for the districts as they actually prevailed in 1970, it was alsofoundthatfliestateofselangorhasnnrethanSOpercentof its districts (4 out of 7) with positive developnent indices, while the state of Perak still registers 50 percent of its districts (4 out of 8) with positive developnent indices. Overall, 36 of the 64 canparative districts, or 56 percent, recorded negative developnent indices (see Table 4.3) . In absolute 154 mmberstherearenorewestemdistrictsthaneastenndistrictswith negative developnent indices, that is 21 to 15. It should be pointed antthattheleesterndistrictsmakeuponlySOpercentofallthe western districts. However, the 15 eastern districts with negative developlmtindioesmakeup68peroentofalltheeastemdistricts. Only 2 urban districts register negative developnent indices . Thisocnstitutelepercentofalltheurbandistrictsand6percent of all districts with negative developnent performance. The reminder of the 36 districts with negative developnent indices are rural districts. The 34 rural districts with negative develoment indices constitute 77 percent of all rural districts, and 94 percent of all districts with negative developnent performance . Table 4.4 1970 Oarparative Districts, Peninsular Malaysia Breakdown of Districts with Positive Developnent Index by States Number of Conparative As As State Districts with Positive Percentage Permntage of Development Index of Total State Total 1) Johore 5 (out of 8) 17.86 62.5 2) Kedah 1 (out of 10) 3.57 10.0 3) Kelantan 1 (out of 8) 3.57 12.5 4) Melaka 1 (out of 3) 3.57 33.33 5) N. Serbilan 5 (out of 6) 17.86 83.33 6) Pahang 4 (out of 8) 14.29 50.0 7) Penarg 4 (out of 5) 14.29 80.0 8) Perak 3 (out of 6) 10.71 50.0 9) Perlis 0 (out of 1) 0.0 0.0 10) Selangor 2 (out of 3) 7.14 66.67 11) Terengganu 2 (out of 6) 7.14 33.33 Total 28 ( 64 ) 43.75 155 The states of Perlis, Kedah, Kelantan, and Terengganu have more than 50 percent of their districts with negative developnent indices. The whole state of Perlis, which is also treated 'as a district, shows a negative developnent performance. In Kedah, 9 out 10 districts register negative developnent indices, in Kelantan 7 out of 8 districtshavenegativeindices, andinTerengganuitis4outof 6. Incaseoftheweighteddevelopnentindex, apartfrcnnanincrease inthetotalmmberoftheomnparativedistrictswithpositive developnent indices, fmn 28 to 29, as well as the slight shift in the relative position of sore of the districts, the weighted developnent index, displays almost similar distributional patterns as those generated by the unweighted developnent index (see Table 4.5). The weighted developnent index also indicates that western districts ocntimnedtosharealargeproportionofthegroupofdislz‘ictswith positive developnent indices. It also reveals that two-thirds of theeasterndistrictsoantinmedtoranaininthegroupofdistricts with below average developnent. Just as the table of unweighted developnent indices, Table 4. 6 of weighted indices shows that there are more urban than rural districts with above average developnent. To measure the degree of agreement between results from weighted and unweighted indices, a Spearman rank correlation coefficient was calculated. The coefficient is 0.99, indicating a very high correlation (significant at 0.05 level). In other words, analysis of developnent based on unweighted and weighted developnent indices produce similar outcomes. 156 Table 4.5 1970 Oanparative Districts, Peninsular Malaysia Ranking of Districts According to Development Index Devt. Index Devt. Index District District JOHORE STATE PAHAM; STATE 1) Batu Pahat 27+ 26+ 36) Bentorg 16+ 19+ 2) Johor Bahru 4+ 3+ 37) Cameron 3) Keluang 18+ 17+ Highland 13+ 12+ 4) Kota Tinggi 42 41 38) Jerantut 45 46 5) Mersing 10+ 10+ 39) Kuantan 3+ 4+ 6) Muar 25+ 25+ 40) Lipis 46 47 7) Pontian 41 40 41) Pekan 17+ 16+ 8) Seganat 31 32 42) Raub 38 42 KEDAH STATE 43) Tennerloh 36 35 9) Baling 58 58 PENANG STATE 10) Bandar Bahru 53 53 44) S.P. Utara 8+ 8+ 11) Kota Setar 28+ 30 45) S.P. Tengah 9+ 9+ 12) Kuala Mida 37 37 46) S.P. Selatan 33 33 13) Kubang Pasu 60 59 47) P.P. T. Laut 2+ 2+ 14) Kulim 35 34 48) P.P. B. Daya 14+ 15+ 15) Langkawi 48 49 PERAK STATE 16) Padang Terap 64 64 49) Batang Padang 23+ 23+ 17) 8:11: 61 61 50) Dindings/Kuala 18) Yan 49 48 Kangsar/Perak KELANI‘AN STATE Hilir * 29 29+ 19) Bachok 51 51 51) Kinta 6+ 6 20) Kate Bahru 12+ 13+ 52) Kerian 55 55 21) Macharg 56 57 53) Larut & Mtg. 22+ 22+ 22) Pasir Mas 54 54 54) Ulu Perak 44 44 23) Pasir Puteh 57 56 PERLIS STATE 24) Tanah Merah 63 62 55) Perlis 39 38 25) Tunpat 50 50 SELANGOR STATE 26) Ulu Kelantan 62 63 56) Kuala Selangor/ MELAKA STATE S. Bernnam * 43 43 27) Melaka Utara 32 31 57) Kuala Langat/ 28) Melaka Selatan 40 39 U.Langat * 20+ 20+ 29) Melaka Tergah 5+ 5+ 58) Klang/Ulu S'gor/ K. Lumper * 1+ 1+ NEGIU SEVIBILAN STATE 30) Jelebu 31) Kuala Pilah 32) Port Dickson 33) Rembau 34) Seremban 35) Tamphn 157 Table 4.5 (continued) 21+ 24+ 15+ 30 7+ 11+ 21+ 24+ 14+ 28+ 7+ 11+ TEREI‘GGANU STATE 59) Besut 60) Dungun 61) Kennaman 62) K.Terengganu 63) Marang 64) U.Terenggannu 52 19+ 34 26+ 47 59 52 18+ 36 27+ 45 Note: * denotes districts that have been canbinned to facilitate conparison + denotes ranking of districts with positive index of developnent 158 Table 4.6 1970 Outparative Districts, Peninsnlar Malaysia: Index of Developnent (Weighted) and Ranking Rank District Devt. Rank District Devt. Index Index (Positive) (Negative) 1) Klang/Ulu Selangcr/ 30) Kota Setar -0.38 . Kuala Lumpur * 171.93 31) Melaka Utara -0.49 2) Pulau Pinnang Tintur 32) Segamat -1.89 Laut 164.90 33) Seberang Perai 3) Johor Bahru 99.74 Selatan -3.29 4) Kuantan 99.66 34) Kulim -6.38 5) Melaka Tegah 96.35 35) Temerloh -7.86 6) Kinta 94.47 36) Kemaman -9.47 7) Seremban 80.58 37) Kuala Muda -9.61 8) Seberang Perai 38) Perlis -10.73 Utara 41.51 39) Melaka Selatan -10.91 9) Seberang Perai 40) Pontian -11.42 Tengah 41.47 41) Kota Tinggi -14.35 10) Mersing 36.59 42) Raub -14.55 11) Tampin 35.15 43) Kuala Selangor/ 12) Canerrn Highland 31.60 Sabak Bemam * -15.38 13) Kota Bahru 30.29 44) Hulu Perak -22.49 14) Port Dickson 28.64 45) Marang -29.19 15) Pulau Pinanng 46) Jerantut -29.31 Barat Daya 26.69 47) Lipis -30.08 16) Pekan 20.80 48) Yan -38.45 17) Keluang 20.30 49) Largkawi -39.84 18) Dungun 19.59 50) Tunpat -44.16 19) Bentcrg 19.25 51) Bachok -44.52 20) Kuala Langat/ 52) Besut -46.23 Ulu Langat * 18.43 53) Bandar Bahru -47.22 21) Jelebu 13.14 54) Pasir Mas -49.91 22) Larut & Matan'g 12.48 55) Kerian -53.72 23) Batang Padang 11.46 56) Pasir Puteh -54.45 24) Kuala Pilah 11.11 57) Macharg -54.79 25) Muar 6.20 58) Baling -68.93 26) Batu Pahat 3.79 59) Kubanng Pasu -73.41 27) Kuala Terengganu 2.57 60) Ulu Terengganu -73.63 28) Renbau 1.51 61) Sik -73.83 29) Dinding/K.Kangsar/ 62) Tanah Merah -80.02 Hilir Perak 0.86 63) Ulu Kelantan -82.60 64) Padang Terap -87.56 Standard Deviation = 53.39 Note: * denotes districts that have been oonbined to facilitate conparison with 1980 canparative districts. 159 As shown in Table 4.7, 22 out of the 29 districts with positive developnent indices are western districts. This cauprise 46 percentofallwesterndistricts. Qntheortherhand, only7ofthe29 districts with positive developnent indices are eastern districts. Still, most of the eastern districts continued to have below average developnent (15 out of 22). Table 4.7 1970 Camparative Districts, Peninsular Malaysia: Distribution of Development Index (Weighted) by Geographical Location and Urban and Rural Districts of Western Eastern Urbann Rural Total Devt. District District District District Positive Devt. 22 7 18 11 29 Negative Devt. 20 15 2 33 35 Internnsofurbanandrural districts, 18ofthe29 districtswith positive deveth indices are urban districts. 0f the 35 districts with below average developnent performance, 33 are rural districts . The developnent index for the 1970 districts as they actually prevailedalsoproducedresultsthatsrnwthatmostofthewestemand urban districts have positive developnent indices, while most of the eastern and rural districts register negative developnent indices (see Tables 4.8 and 4.9). 160 Table 4.8 1970 Non—Carparative Districts, Peninsular Malaysia: Distribution.of Development Index (unweighted) by Geographical Location.and.Urban.and.Rura1 Districts of western Eastern. urban Rural Total Devt. District District District District Positive Devt. 22 7 18 11 29 Index Negative Devt. 26 15 3 38 41 Index Total 48 22 21 49 70 Table 4 . 9 1970 Nan—Canparative Districts, Peninsular Malaysia: Distribution of Development Index (Weighted) by Geographical LocationandUrbanandRuralDistricts Index of Western Eastern Urban Rural Total Devt. District District District District Positive Devt. 23 7 19 11 30 Index Negative Devt. 25 15 3 38 40 Index 161 1980 Develoguent Index Asfor1970, districtsinl980wereexaminnedintermsoftheir pattern of spatial developnent. Of the 64 canparative districts in 1980, it is found that 30 (or 47 percent) register positive developnent indices. The other 34 districts (or 53 percent) have negative developnent indices (see Table 4.10). Figure 4.4 shows the level of developnent for the canparative districts and Figure 4. 5 shows the level of developnent for all districts as they actually prevailed in 1980. 23 of these 30 districts with positive developnent indices (or 77 percent) consist of western districts. This figure represents 55 percent of all western districts. Still, only 7 eastern districts (or 23 percent) have positive developnent indices. This means that only 32 percent of all eastern districts have average or more than average share of the benefits of developnent. Inl970therewere20urbanndistrictsand44rural districts. However, inl980themmnberofurbandistrictsincreasedby6t026; 4beingnewurbancenters. Correspondingly, thenumbercfrural districts declinned from 44 to 38. As shown in Table 4.11, 19 of the 26 urban districts (or 73 percent) have positive developnent indices . This constitutes 63 percent of all districts with positive developnent indices The 1980 unweighted development index also indicates that 11 cf the 38 rural districts have average or above average developnent 162 Table 4.10 1980 (Imperative Districts, Peninnsular Malaysia: Index of Developnent (Unnweighted) and Rannking Rank District Devt. Rank District Devt. Index Index (Positive) (Negative) 1) Ganbak/Klang/Petaling 31) Muar -0.03 Ulu Selangor/Wilayah 32) Batu Pahat -0.16 Persekutuan (Kuala 33) Pontian -0.39 Lumpur) * 47.16 34) K.Setar/Pedang * -1.19 2) Pulau Pinnang Tinur 35) Pekan/Ranpin * -l.24 Laut 38.34 36) Keluang -2.49 3) Kuantan 27.40 37) Batang Padang -2.72 4) Serenban 26.27 38) Ulu Perak -3.28 5) Melaka Tegah 26.10 ' 39) Manjung/K.Kangsar 6) Johor Bahru 26.07 Hilir Perak/Perak 7) Kinta 17.69 Tegah * -4.50 8) Kcta Bahru 15.47 40) Segamat -5.28 9) S.Perai Tegah 15.43 41) Marang -5.31 10) S.Perai Utara 14.68 42) Jerantut -5.35 11) Pulau Pinang Barat 43) Kota Tinggi -6.79 Daya 13.81 44) Largkawi -6.99 12) Kuala Terengganu 13.48 45) Tennerloh -7.36 13) Dungun 13.17 46) Cameron Highland -7.39 14) Kuala Langat/Ulu 47) Bandar Bahru -8.65 Langat/Sepang * 12.48 48) Kuala Selangor/ 15) Mersirg 11.45 Sabak Bernam * -10.10 16) Terrain 7.55 49) Lipis -10.39 17) Remau 7.54 50) Macharg -13.53 18) Port Dickson 4.90 51) Yan -15.21 19) Larut & Matang 4.89 52) Besut —15.26 20) Jelebu 3.81 53) Tumpat -15.54 21) Benteg 3.28 54) Ulu Terengganu -15.66 22) Kuala Pilah/Jenpol * 2.98 55) Pasir Puteh -15.70 23) Kenaman 1.71 56) Kubang Pasu -16.18 24) Melaka Utara 1.57 57) Pasir Mas -16.63 25) Perlis 1.36 58) Kerian -16.69 26) Kuala Muda 1.34 59) Tanah Merah -19.16 27) Melaka Selatann 1.26 60) Bachck -19.87 28) Raub 0.61 61) Baling -22.18 29) Seberang Perai 62) Padang Terap -22.29 Selatann 0.17 63) Ulu Kelantan/ 30) Kulim 0.02 Kuala Krai * -22.54 64) Sik -25.96 Standard Deviatien = 14.94 Note: * denotes districts that have been combined to facilitate conparison with 1970 canparative districts . 163 PBNINSULAR MALAYSIA : DISTRICTS . \l 1980 '/ . "Him: in a. a i In 70 U 2” L0 60 8!) km EH=k=i=bd _ — k -_ ' .-L - ') / ..- ,= l, . - A. /: i l 5225 \ Ll___ is. ll. '~ 5 ._." A‘, - In K \ _..=_...__,_£_,__ — Ill-::::._;— ‘15-..- -—==1 33:1 i geese .fiiixsmiilsfrf::rré‘ — w = a» ~‘~._ I ‘1—‘1: __ High Low Very Low Figure 4.4: 1980 Cbmparative Districts, Peninsular Malaysia: Level of Development. 164 PENINSULAR MALAYSIA : DISTRICTS. 1980 20 0 'zn 1.0 60 «0 km Hates Development Level Very High High Low Very Low Figure 4.5: 1980 Districts, Peninsular Malaysia: Level of Develcrnrxfl; 165 Table 4.11 1980 Omnparative Districts, Peninsular Malaysia: Distribution of Developnent Index (Unweighted) by Geographical Location and Urban and Rural Districts of Western Eastern Urban Rural Total Devt. District District District District Positive Devt. 23 7 19 ll 30 Negative Devt. 19 15 7 27 34 performance. This represents 29 percent of all rural districts. However, fiebulkcftherureldistrictsstillhavebelowaverage developnent performance as evidenced by their negative indices. FivestateshavencrethanSOpercentoftheircomparative districts that register positive deveth indices. Of these, Malacca, Negri Senbilan, Penang, and Perlis have all their districts (100 percent) with positive developnent indices. Se1angor (including the Federal Territory of Kuala Lumpur) has 67 percent of its canparative districts with above average developnent performannce . In the state of Terengganu 50 percent of its districts have positive developnent performannce (see Table 4.12). Except for some changes in the figures, results are similar when the non-canparative districts are used: these states still show 50 percent or more of their districts have positive developnent indices. 166 34 of the 64 canparative districts (or 53 percent) register negative developnent indices in 1980. Of these, 19 (or 56 percent) arewesterndistrictsand15(or4percent)areeasterndistricts. Eventhoughwesterndistrictsantnmnberedeasterndistrictswith negative developnent index, the 19 western districts represent only 45 percentofallwesterndistricts. Ontheotherhand, thelSeastern districtscanprise68percentof alltheeasterndistricts. Table 4.12 1980 Comparative Districts, Peninnsular Malaysia: Breakdown of Districts with Positive Develcpnent Index by States Number of Cauparative As As Districts with Percentage Percentage of State Positive Devt . Index of Total State Total 1) Johorre 2 (ant of 8) 3.125 25.0 2) Kedah 2 (ant of 10) 3.125 20.0 3) Kelantan 1 (ant of 8) 1.56 12.5 4) Melaka 3 (ant of 3) 4.69 100.00 5) N. Senbilan 6 (ant of 6) 9.375 100.00 6) Pahang 3 (ant of 8) 4.69 37.5 7) Penanng 5 (ant of 5) 7.81 100.00 8) Perak 2 (ant of 6) 3.125 33.33 9) Perlis 1 (ant of 1) 1.56 100.00 10) Selangor 2 (ant of 3) 3.125 66.67 11) Terengganu 3 (ant of 6) 4.69 50.00 Total 30 ( 64 ) 46.875 Of these 34 districts with negative developnent indices, 7 (or 21 percent) are urban districts and 27 (or 79 percent) are rural districts. 167 In1980, 5stateshavemorefinan50percentoffineirdistricts wifin negative developnent indices, namely Kelantan (87 percent), Kedah (80 percent), Johore (75 percent), Perak (67 percent) and Pahang (62 percent) (see Table 4.12). According to fine weighted developnent indices (in Table 4.13) the number of districts with positive developnent indices are 31 innstead of 30. Wifin fine exception of Kulim and Raub, all other canparative districtscantinnnedtohaveaboveaveragedeveloptentbyfineweighted developnent indices. Batu Pahat, Muar, and Pantian (all in the state ofJohore) arefineofinerfinreedistrictswithpositiveindices. The application of weights in aggregating the z-soores have resulted in finese 3 districts having positive indices, while Kulim and Raub have nnegative indices. The nunber of eastern districts with positive developnentindexdecreasedfmn7to6. Also, thenumberofurban districts with positive developnent index increased fmn 19 to 20. Apart fran finese slight changes, the 1980 weighted developnent index, basically reflects almost similar pattern of spatial deveth as that produced by fine unweighted developnent indices (see Table 4.14). As shown in Table 4.15, fine majority of western and urban districts continued to register positive developnent performannce, while most of fine eastern and rural districts are characterized by negative developnent indices . 168 Table 4.13 1980 Omparative Districts, Peninsular Malaysia: Index of Develognent (Weiglnted) and Ranking Rank District Devt. Rank District Devt. Index Index (Positive) (Negative) 1) Ganbak/Klang/Petaling 32) Kulim -0.74 Ulu Selangor/Wilayah 33) Raub -1.70 Persekutuan (Kuala 34) Pekan/Ra'npin* -2.62 Lunpur) * 151.57 35) K.Setar/Pendang * -5.01 2) Pulau Pinnang Tinnur 36) Keluang -8.46 Lannt ‘ 125.13 37 ) Batang Padang -10.55 3) mlaka Tegah 89.82 38) Marang -13.63 4) Kuanntan 86.92 39) Ulu Perak -14.23 5) Johor Bahru 86.74 40) Manjung/K.Kangsar 6) Serenban 86.45 H.Perak/Perak 7) Kinta 55.89 Tegah * -15.24 8) S.Perai Tegah 54.42 41) Jeranhnt -17.25 9) P.P. Barat Daya 49.32 42) Segamat -17.76 10) Kcta Bahru 48.72 43) Kota Tinggi -20.29 11) S.Perai Utara 48.10 44) Tenerloh -24.79 12) Kuala Langat/Ulu 45) Langkawi -25.26 Langat/Sepang" 46.65 46) Bandar Bahru -26.60 13) Kuala Teregganu 42.47 47) C.Highland -29.16 14) Dungm 42.04 48) Kuala Selangor/ 15) Mersing 35.32 Sabak Bernam* -31.77 16) Rembau 28.59 49) Lipis -38.15 17) Tanpin 27.37 50) Machang -43.34 18) Port Dickson 18.06 51) Besut -48.63 19) Jelebu 13.74 52) Tumpat -50.22 20) Larut & Matang 12.84 53) Pasir Puteh -50.37 21) Kuala Pilah/Jenpol" 10.99 54) Yan -50.92 22) Melaka Utara 8.27 55) Ulu Terenggannu -51.14 23) Melaka Selatann 7.67 56) Kubang Pasu -53.58 24) Kennaman 7.40 57) Pasir Mas -55.21 25) Benteg 6.31 58) Kerian -56.56 26) Perlis 4.90 59) Bachok ~64.90 27) Kuala Mnda 4.72 60) Tanah Merah -65.12 28) Batu Pahat 1.89 61) Padang Terap -74.15 29) Mnar 1.82 62) Baling -74.84 30) Seberang Perai 63) Ulu Kelantan/ Selatann 1.77 Kuala Krai* -77.77 31) Pontiann 1.61 64) Sik -87.53 Standard Deviation = 49.39 Note: * Denotes districts that have been combined to facilitate conparisan with 1970 canparative districts. Rannking of Districts According to Developnent Index 169 Table 4.14 1980 Couparative Districts, Peninsular Malaysia Devt.1ndex Devt. Index District District Unwtd. Wtd. Unwtd. Wtd. JGDRE STATE PAHANG STATE 1) Batu Pahat 32 28+ 36) Bentong 21+ 25+ 2) Johor Bahru 6+ 5+ 37) Cameron 3) Keluang 36 36 Highland 46 47 4) Kota Tinggi 43 43 38) Jerantut 42 41 5) Mersing 15+ 15+ 39) Kuanntann 3+ 4+ 6) Mnar 31 29+ 40) Lipis 49 49 7) Pontian 33 31+ 41) Pekan/Ronpin * 35 34 8) Sega-mat 40 42 42) Raub 28+ 33 KEDAH STATE 43) Tenerloh 45 44 9) Baling 61 46 PENANG STATE 10) Bandar Bahru 47 62 44) S.P. Utara 10+ 11+ 11) Kota Setar/ 45) S.P. Tegah 9+ 8+ Pedang * 34 35 46) S.P. Selatan 29+ 30+ 12) Kuala Muda 26+ 27+ 47) P.P.T.Laut 2+ 2+ 13) Kubang Pasu 56 56 48) P.P.B.Daya 11+ 9+ 14) Kulim 30+ 32 PERAK STATE 15) Lanngkawi 44 45 49) Batang Padang 37 37 16) Padang Terap 62 61 50) Manjung/K.K./ 17) Sik 64 64 H.Pk/P.Tegah* 39 40 18) Yan 51 54 51) Kinta 7+ 7+ KEIANTAN STATE 52) Keriann 58 58 19) Bachok 60 59 53) Larut & Mtg. 19+ 20+ 20) Kata Bahru 8+ 10+ 54) Ulu Perak 38 39 21) Machang 50 50 PERLIS STATE 22) Pasir Mas 57 57 55) Perlis 25+ 26+ 23) Pasir Puteh 55 53 SELANGOR STATE 24) Tannah Merah 59 60 56) Kuala Selangor/ 25) Turpat 53 52 S.Bernam * 48 48 26) Ulu Kelanntan/ 57) Kuala Langat/Ulu Kuala Krai * 63 63 Lgt/Sepang * 14+ 12+ MELAKA STATE 58) Gonbak/Klang/ 27) Melaka Utara 24+ 22+ Petaling/U.S'gor 28) Melaka Selatann 27+ 23+ /W.P.* 1+ 1+ 29) Melaka Tegah 5+ 3+ TERENGGANU STATE NEBRI SEIVBILAN STATE 59) Besut 52 52 30) Jelebu 20+ 19+ 60) Dungunn 13+ 14+ 31) Kuala Pilah/ 61) Kemaman 23+ 24+ Jenpol * 22+ 21+ 62) K.Terengganu 12+ 13+ 32) Port Dickson 18+ 18+ 63) Marang 41 38 33) Renbau 17+ 16+ 64) U.Terengganu 54 55 34) Seremban 4+ 6+ 35) Tenpin 16+ 17+ Note: * conbinned districts to facilitate couparisons + districts with positive developnent index . 170 Table 4.15 1980 Couparative Districts, Peninsular Malaysia: Distribution of Development Index (Weighted) by Geographical location and Urban and Rural Districts of Western Eastern Urban Rural Total Devt . District District District District Positive Devt. 25 6 20 11 31 Negative Devt. 17 16 6 27 33 25 of fine 31 districts (or 81 percent) with positive developnent indicesarewesterndistricts. Thisconstitutes60peroentof all westerndistricts. Only6easterndistricts (or 19 percent) have positive developnent indices. This made up 27.27 percent of all eastern districts. 20 of fine 31 districts (or 65 percent) registering average or aboveaveragedevelopnentperformanceareurbandistricts. This constitutes 77 percent of all urban districts. only 11 of fine 38 rural districts have above average developnent performance. The ofiner 27 rural districts (or 71 percent) have negative developnent indices. Exceptforfinefactthattherearesoneslightchangesinfine unweighted and weighted development indices for fine 1980 non- conparative districts, it was observed that overall the results produced are not significantly different fron finose produced by using indices based on conparative districts. As in the 1980 developnent 171 index for the comparative districts, the development index.fbr the nonrcomparative districts continue to display the spatial pattern.in whichnwestern.and.urban.districts dominate the groupnof districts with positive development. This also indicates that tn 1980 most of the eastern.and.rural districts continued to record.negative development indices (see Tables 4.16 and 4.17). The Spearman.rank correlation.between unweighted and weighted development indices for 1980 produces a 0.99 coefficient, indicating a high.correlation.(significant at 0.05 level). This indicates that analysis of development fbr 1980 based.on unweighted and.weighted development indices produces similar pattern.of spatial development. Table 4.16 1980 Non-Conparative Districts, Peninnsular Malaysia: Distribution.of Development Index (unweighted) by Geographical Location and.Urban.and Rural Districts of western. Eastern. urban. Rural Total Devt. District District District District Positive Devt. 26 8 22 12 34 Negative Devt. 28 16 9 35 44 172 Table 4.17 1980 Non-Conparative Districts, Peninsular Malaysia: Distribution of Developnent Index (Weighted) by Geographical LocationandUrbanandRuralDistricts of Western Eastern Urban Rural Total Devt. District District District District Positive . Devt. 31 7 25 13 38 Negative Devt. 23 17 6 34 40 Oolnparison of 1970 and 1980 Development Indices The examination of fine levels of developnent of the canparative districts for 1970 and 1980 shows innmediately one striking characteristic. Thishastodowithfineprepoderannceoffine canparative districts with negative indices of development . Even finough the nunber of canparative districts that register average and above average developnent performance recorded a slight increase over time, overall for both 1970 and 1980, the total number of conparative districts with positive deveth indices conprise less than 50 percent of all fine conparative districts. As can be seen fron Table 4.18, in 1970 only 28 ant of fine 64 conparative districts (or 44 percent) have positive indices, while in 1980 there are 30 ant of 64 (or 47 percent). There are more comparative districts with belav average developnent performance; as evidenced by fine negative indices 173 for 1970 and 1980. Table 4.19 stnows finat 36 of fine 64 canparative districts (or 56 percent) have negative indices in 1970, while in 1980 thereare36antof64(or53percent). Table 4.18 0onparative Districts: Number of Districts Wifin Positive Index Of Development Unmeighted Weighted Year Develcpnent Index % Developnent Index % 1970 28 (ant of 64) 43.75 29 (ant of 64) 45.31 1980 30 (ant of 64) 46.85 31 (ant of 64) 48.44 Table 4.19 Corparative Districts: Number of Districts With Negative Index of Developnent ‘ unweighted Weighted Year Developnent Index % Deveth Index % 1970 36 (ant of 64) 56.25 35 (ant of 64) 54.69 1980 34 (ant of 64) 53.13 33 (ant of 64) 51.56 It is interesting to note that fine weighted index of developnent for both 1970 and 1980 also indicate that for these two tine periods fiere are more canparative districts wifin negative developnent indices (see Tables 4.18 and 4.19). In sunnmary, it can be stated finat over fine period fine majority of fine canparative districts remained below fie average developnent performannce. Even when developnent performance is examined for fie districts without 174 combiningcertainoffinedistrictsconcerned, fienumberofdistricts wifin negative developnent indices still exceed those wifin positive developnent indices. Tte indices of developnent for districts in 1970 and 1980 (in Tables 4.20 and 4.21) and fie summary in Table 4.22 illustrate finis point. Table 4.2.3 stews tnow fie various oonpoent states in Peninsular Malaysia performed in developnent as ganged by fie composite index of developnent. In conparing finis table with Tables 4.2 and 4.10, which stmfiedevelopmentindices forfinevarianscomparativedistrictsin 1970 and 1980, respectively, it can be seen that fine annalysis of development perfornmannce at fie district scale nnot only provides more information regarding fie pattern of spatial development but also shows that fiere is relatively greater variation and disparity of spatial developnent compared to fine state scale of analysis. fibreimportentisfinatfiecomparisonoffiesetablesdenuetratesfiefact finatfieuseoffiestatescaleofanalysisdoesnottrulyrevealfine range of variations and fie distribution of spatial inequality within fienm. For example, alfinangh the state of Selangor stuns high positive developnent indices in 1970, only 3 districts in that state actually have positive developnent indices. Similarly, alfinangh fine states of KelantanandKedahhavefielavestdevelopnent indicesanmogfine statesinPeninsularMalaysia, itcanbeseenfinatfinedistrictsof Kottharu(inKe1antan)andfinecombineddistrictsofKotaSetarand Pedang (in Kedah) have relatively better developnent indices than fiese districts which register mgative developnent indices in Selangor. Hence, examinnation of fie level of developnent at fine 1970 Districts, Peninsular Malaysia: 175 Table 4. 20 Indexof Development (Unweighted) and Ranking Devt. Devt. Rannk District Index Rank District Index (Positive) (Negative) 1) Kuala Lumpur 62.17 30) Batu Pahat -0.23 2) Pulau Pinnang 31) Kota Setar -0.50 Timur Laut 49.50 32) Rembau -1.28 3) Kuantan 28.73 33) S. Perai Selatan -1.39 4) Johor Bahru 28.41 34) Hilir Perak -1.46 5) Melaka Tegah 28.30 35) Melaka Utara -1.53 6) Kinta 28.15 36) Segannat -1.66 7) Seremban- 23.30 37) Kemaman -1.88 8) Klang 22.42 38) Kulim -2.24 9) Ulu Langat 12.98 39) Sabak Bernanm -2.94 10) S. Perai Tegah 11.93 40) Tenerloh -3.04 11) S. Perai Utara 11.90 41) Kuala Muda -3.12 12) Mersing 10.47 42) Kuala Langat -3.39 13) Tampin 9.38 43) Raub -3.87 14) Kota Bahru 8.73 44) Melaka Selatan -4.04 15) Cameron Highland 7.61 45) Perlis -4.10 16) Pulau Pinnang 46) Dinding -4.l9 Barat Daya 7.24 47) Pontian -4.44 17) Port Dickson 6.27 48) Kota Tinggi -5.09 18) Bentog 6.23 49) Hulu Perak -7.11 19) Ulu Selangor 5.85 50) Kuala Selangor -8.92 20) Keluang 5.17 51) Marang -9.00 21) Dungunn 5.06 52) Lipis -9.17 22) Pekan 4.91 53) Jerantut -9.30 23) Jelebu 3.21 54) Langkawi —11.89 24) Larut & Matang 3.03 55) Yan -12.92 25) Batang Padang 2.51 56) Tumpat -13.08 26) Kuala Pilah 2.24 57) Bachok -l4.08 27) Kuala Kangsar 1.70 58) Besut -14.42 28) Mnar 1.00 59) Bandar Bahru -l4.94 29) Kuala Teregganu 0.83 60) Pasir Mas -15.03 61) Kerian -16.31 62) Macharg -16.42 63) Pasir Puteh -16.85 64) Baling -20.96 65) Ulu Teregganu -21.52 66) Kubang Pasu —22.12 67) Sik -22.48 68) Tanah Merah -23.40 69) Ulu Kelantan -23.53 70) Padang Terap -25.34 Standard Deviation = 16.03 176 Table 4 . 21 1980 Districts, Peninsular Malaysia: Index of Developnent (Unweighted) and Ranking Devt. Devt. Index Index Rannk District (Positive) Rank District (Negative) 1) Petaling 47.55 35) Kuala Langat -0.13 2) Kuala Lunpur 43.41 36) Muar -0.17 3) Pulau Pinnang 37) S. Perai Selatan -0.27 Timur Lannt 33.97 38) Pontian -0.37 4) Gonbak 27.43 39) Kulim -0.42 5) Kuanntan 24.71 40) Batu Pahat -0.55 6) Johor Bahru 23.57 41) Kuala Kangsar -1.21 7) Serenban 23.47 42) Keluang -2.34 8) Melaka Tegah 23.02 43) Batang Padang -3.05 9) Klang 22.01 44) Hulu Perak -3.42 10) Ulu Langat 19.04 45) Ulu Selangor -3.54 11) Kinta 16.08 46) Sepang -4.59 12) S. Perai Tegah 13.20 47) Hilir Perak -4.83 13) Dungunn 13.13 47) Marang -4.83 14) Kota Bahru 13.08 49) Segamat -5.23 15) S. Perai Utara 12.92 50) Jerantut -5.34 16) Kuala Pilah 12.52 51) Kate Tinggi -6.57 17) Kuala Terengganu 12.29 52) Kuala Selangor -7.02 18) Pulau Pinang 53) Langkawi -7.09 Barat Daya 12.12 54) C. Highland -7.26 19 ) Mersing 10 . 75 55 ) Tennerloh -7 . 45 20) Remnbann 6.85 56) Bandar Bahru —8.22 21) Tanpin 6 67 57) Manjung ~9.05 22) Pekan 5.83 58) Lipis -10.44 23) Kcta Setar 5.37 59) Jenpol -10.51 24) Larut & Matang 4.48 60) Rompin -12.08 25) Port Dickson 4.13 61) Sabak Bernam -12.47 26) Jelebu 3.60 62) Machang -12.86 27) Bentog 2.80 63) Yarn -14.57 28) Melaka Utara 1.24 64) Besut -14.68 29) Kemerenn 1.17 65) Pasir Puteh -14.97 30) Melaka Selatann 1.06 66) Tumpat -15.07 31) Kuala Mda 0.67 67) Kubang Pasu —15.37 32) Perak Tegah 0.55 68) Ulu Terengganu -15.72 33) Perlis 0. 36 69 ) Kerian -15 . 87 34) Raub 0.02 70) Pasir Mas -16.28 177 Table 4.21 (continued) Standard Deviation = 15.16 71) 72) 73) 74) 75) 76) 77) 78) Tannah Merah Bachok Kuala Krai Padang Terap Baling Sik Pendarg Ulu Kelantan .-18.07 -18.77 -19.72 -20.68 -21.02 -24.28 -25.28 -27.43 178 Table 4.22 Non-Congarative Districts: Number of Districts With Positive And Negative Indices of Developnent Year Positive Devt. Index Negative Devt. Index Districts Unnweighted weighted Umeighted Weighted Total 1970 1980 29 (41.43) 30 (42.86) 41 (58.57) 40 (57.14) 70 34 (43.59) 38 (48.72) 44 (56.41) 40 (51.28) 78 179 Table 4.23 Conparative States: Composite Index of Development (Unweighted), 1970 and 1980 1970 1980 Rank State Devt.]:ndex Rank State Devt.In‘ndex 1. Selangor 36.23 1) Selangor & W.P.* 32.02 2. Pulau Pinang 27.10 2) Pulau Pinnang 19.90 3. Malacca 6.82 3) Malacca 10.56 4. Negri Sembilan 5.91 4) Negri Sembilan 8.23 5. Johore 1.26 5) Johore -1.64 6. Perak 1.14 6) Perak -4.53 7. Patnang -1.24 7) Terengganu -5.95 8. Perlis -13.44 8) Pahang -6.67 9. Terengganu -l6.66 9) Perlis -7.40 10. Kedah -22.29 10) Kedah —22.18 11. Kelantan -22.82 11) Kelantan -22.35 Std.Deviation 18.35 Std. Deviation 15.81 Note: *TteFederal Territorychuala Lumpurhasbeencombinedwifin fie state of Selangor for comparative purpose. dist: diffn m m asf stem of of entree 180 district scale provides more realistic picture of fie spatial differences in development performance finan fie state scale of analysis. It should, fierefore, provide better guidannce for developnent of policy and planning measures to redress inequality. Differences in fie index of developnent for fie districts as well asforfiestatesusingoomparativeandnun-comparativedistrictsand states indicate finat variations inn spatial developnent exist amnog fine districtsandstates. Thissupportsfinehypofinesisfinatdisparitiesin spatial developnentexistatbofinfinedistrictaswellasfiestate scales of observatioe. Annofier distinnctive feature derived from comparing 1970 and 1980 development indices of fie comparative districts is fine fact finat most of fie districts with positive development indices are western districts. Inn 1970, western districts comprised 75 percent (or 21 out of 28 districts) of all fie canparative districts with positive developnent indices. In1980, fineymakeup77percent (or23antof 30 districts). For both 1970 and 1980, fiere are only 7 eastern districts wifin positive developnent indices. In 1970, eastern districts aocannnted for only 25 percent of all the comparative districts wifin positive developnent indices, while in 1980 fine percentage declined to 23 percent (see Table 4.24). In comparing western and eastern districts with negative indices, it is appropriate to use fieir proportions; since in absolute numbers fiere are more western than eastern districts in 1970 and 1980. In finis regard, it is fond finat a large proportion of fine eastern districts have belav average developnent performannce as compared to 181 fie western districts. It can be seen in Table 4.25, for both 1970 and 1980, finat 68 percent of all fie eastern districts register regative development indices. W, a decline is observed for western districts as fine percentage of districts wifin negative developnent indices decreased from 50 to 45 percent. Tte weighted developnent indices for 1970 and 1980 also stnow that fine western districts have relatively better developnent performance finannfieeasternndistricts. Whereasfineproportionoffinewestern districtswifinpositiveindicesincreasefmm52t060percentfine proportionoffineeasterndistricts decline from32t027 percent. Table4.258tnowsfiemmberandproportionofwesternandeastern districts with below average developnent performance. In sum, data for umeighted and weighted developnent indices for 1970 and 1980 show consistently finat western districts fare better finan eastern districts. This, fierefore, supports fine hypofinesis finat districts in fine west coast states of Peninsular Malaysia have ahcwn relatively betterdevelopnentperformanceovertimefinandistrictsinfieeast coast states. Annofier pattern emerging from fine comparison of 1970 and 1980 developnent indices of fie comparative districts is fine difference of developnent performance between urban and fine rural districts. The indices of deveth produced for 1970 and 1980 clearly indicate finat fie m:bann districts fare better finan fine rural districts. This can be seen in Table 4.26, wtnere in 1970 fine urban districts comprise 182 Table 4.24 1970 and.1980 comparative Districts: western.and Eastern Districts With.Positive Development Index 1970 1980 Devt.Index Devt. Index Geographical Location. thtd. Wtd. thtd. Wtd. utenxurntustricts (42) Number Wifin Positive 21 22‘ 23 25 Development Index As Percentagel.0 1) Population Literacy 0 38 3 23 2) Formal Schooling 0 36 4 24 3) Teaclers 0 44 1 19 4) Infant Mortality 0 31 1 32 5) Maternal Mortality 6 27 0 31 6) Housing Stock 0 20 3 41 7) Flush Toilets 0 52 0 12 8) Piped-Water 0 40 2 22 9) Electricity 0 47 0 17 10) Public Security 1 30 O 33 11) Fire Service 4 36 0 24 12) Road Density 0 21 0 43 13) Postal Service 0 50 0 l4 14) Professional and Technical Manpower 0 51 0 13 15) Administrative and Managerial Manpower 0 55 1 8 16) Goverment Officials 0 52 0 12 17) Manufacturing Sector 0 52 0 12 18) Comercial Sector 0 48 2 14 19) Gross Donestic Product 0 50 1 13 20) Rural Telephoe Bccfins 0 32 0 32 21) Agricultural Extension Service 41 8 0 15 22) Urban Population 34 16 0 14 23) Large Urban Centers 44 9 0 ll 24) Fenale Literacy 0 38 0 26 25) Fernale In Non-Agric. Econonic Activities 0 47 0 17 26) Above Poverty Lire 0 52 0 12 27) Public Low-Cost Housing 10 43 0 11 28) Malays in Agriculture 0 16 0 48 29) Malays in Manufacturing 0 29 0 35 30) Malays in Comerce 0 22 l 41 200 Table 4.32 Location Quotients for Comparative Districts: Performance of Variables By Districts, 1980 Number of Districts With location Quotient Variables 0 (1.0 1.0 >l.0 1) Population Literacy 0 40 4 20 2) Formal Schooling 0 37 7 20 3) Teacrers 0 39 1 24 4) Infant Nbrtality 0 33 l 30 5) Maternal Mortality 8 24 0 32 6) Boeing Stock 0 27 2 35 7) Flush Toilets 0 37 l 26 8) Piped-Water 0 32 0 32 9) Electriciw 0 40 0 24 10) Public Security 0 37 1 26 11) Fire Service 4 34 0 26 12) Road Density 0 24 0 40 13) Postal Service 0 51 0 l3 14) Professional and Technical Manpower 0 53 0 ll 15) Adninistrative and a1 0 57 0 7 16) Goverment Officials 0 53 1 10 17) Manufacturing Sector 0 50 0 14 18) Comercial Sector 0 50 3 11 19) Gross Donestic Product 0 50 0 14 20) Rural Telephoe Bcofins 0 29 2 33 21) Agricultural Extension Service 40 10 0 14 22) Urban Population 28 24 0 12 23) Large Urban Centers 38 13 0 13 24) Fenale Literacy 0 46 2 16 25) Fenale In Non-Agric. Econonic Activities 0 48 0 16 26) Above Poverty Line 0 54 0 10 27) Public Low—00st Boeing 3 43 0 18 28) Malays in Agriculture 0 15 1 48 29) Malays in Manufacturing 0 l9 0 45 30) Malays in Oomnerce 0 26 0 38 201 Table 4 . 33 location Quotients for Comparative Districts: Performance of Districts By Variables, 1970 Number of Variables Wifin location Quotient Districts 0 <1.0 1.0 >l.0 1) Batu Pahat 0 22 0 8 2) Johor Bahru 1 3 0 26 3) Keluang l 19 1 9 4) Kota Tinggi 3 18 0 9 5) Mersing 3 12 0 15 6) Muar 0 20 0 10 7) Pontian 3 22 0 5 8) Segamat 2 20 l 7 9) Baling 3 20 1 6 10) Bandar Bahru 4 20 0 6 11) Kate Setar/Pendang * 0 21 l 8 12) Kuala Mlda l 23 l 8 13) Kubang Pasu 2 23 l 5 14) Kulim 2 22 l 5 15) Langkawi 4 19 0 7 16) Padang Terap 5 18 0 7 l7) Sik 4 19 0 7 18) Yan' 3 20 0 7 19) Bachck 4 19 0 7 20) Kata Bahru 0 l4 0 16 21) Machang 4 18 0 8 22) Pasir Mas 3 21 0 6 23) Pasir Puteh 4 20 0 6 24) Tanah Merah 3 20 0 7 25) Tunpat 3 18 0 9 26) Ulu Kelantan/K.Krai * 4 19 0 7 27) Melaka Utara 2 15 2 ll 28) Melaka Selatan 3 18 0 9 29) Melaka Tengah 0 5 0 25 30) Jelebu 3 17 0 10 31) Kuala Pilah/Jenpol * 2 l4 0 14 32) Port Dickson 3 l8 0 9 33) Renbau 4 16 0 10 34) Seretban l 7 0 22 35) Tampin 3 11 0 16 36) Bentong l 18 0 11 37) Cereron Highland 4 15 0 ll 38) Jerantut 3 l8 0 9 202 Number of Variables Wifin location Quotient Districts 0 <1.0 1.0 >l.0 39) Kuantan 0 6 0 24 40) Lipis 3 20 3 7 41) Pekan/Ronpin * 1 17 0 12 42) Raub 3 17 3 7 43) Tenerlch l 20 l 9 44) Seberang Perai Utara 0 13 l 16 45) Seberang Perai Tengah 0 l7 0 13 46) Seberang Perai Selatan 3 20 l 6 47) Pulau Pinang Timlr Laut l 7 1 21 48) Pulau Pinang Barat Daya 4 12 1 13 49) Batang Padang 2 l7 1 10 50) Manjung/Kuala Kangsar/ P.Hilir/P.Tengah * 0 23 1 6 51) Kinta 1 10 0 19 52) Krian 3 20 1 6 53) Larut & Matang l 16 0 13 54) Ulu Perak 3 l7 0 10 55) Perlis 2 18 0 10 56) Kuala Selangcr/ Sabak Bernan * 2 19 0 9 57) Kuala Langat/Ulu Largat/Sepang * 0 22 1 7 58) Gorbak/Klang/Petaling/ Ulu Selangor/ Kuala Lunpur * 0 7 0 23 59) Besut 3 21 0 6 60) Dungun 2 l7 0 ll 61) Kenenan 2 18 0 10 62) Kuala Terengganu: l 16 0 13 63) Marang 5 16 0 9 64) Ulu Terengganu 2 21 0 7 Note: * Districts which have been conbired to facilitate conparison. 203 Table 4.34 location Qntient for (Imperative Districts: Performance of Districts By Variables, 1980 Nmber of Variables Wifin location Quotient Districts 0 <1.0 1.0 >l.0 l) Batu Pahat 0 23 1 6 2) Jchcr Bahru 1 9 1 19 3) Keluang l 19 l 9 4) Kota Tinggi 2 20 0 8 5) Mersing 2 16 0 12 6) Muar 0 22 1 8 7) Pontian 2 22 0 6 8) Segamat 1 19 3 7 9) Baling 3 20 0 7 10) Bandar Bahru 5 18 l 6 ll) Kota Setar/Pendang * 0 22 0 8 12) Kuala Mlda 1 20 0 9 l3) Kubang Pasu l 24 0 5 14) Kulim 1 20 1 2 15) Largkawi 3 17 l 9 16) Padang Terap 4 20 0 6 17) Sik 3 20 0 7 18) Yan 3 20 0 7 l9) Badnck 3 22 0 5 20) Kate Bahru 0 13 0 17 21) Machang 3 20 0 7 22) Pasir Mas 3 18 0 9 23) Pasir Puteh 3 20 0 7 24) Tanah Merah 2 20 0 8 25) Turpat 4 19 0 7 26) Ulu Kelenten/K.Krai * 3 21 0 6 27) Melaka Utara 2 15 l 12 28) Melaka Selatan 4 14 l 11 29) Melaka Tengah 0 8 0 22 30) Jelebu 3 13 1 13 31) Kuala Pilah/Jenpol * 2 18 0 10 32) Port Dickson 2 19 1 8 33) Rembau 4 12 0 14 34) Sererban l 6 0 23 35) Tannin 3 l4 0 13 36) Bentong 1 l6 2 11 37) Cameron Highland 3 19 0 8 38) Jerantut 3 16 0 ll 39) Kuantan 0 6 0 24 40) Lipis 2 18 0 10 41) Pekan/Romain * 2 20 0 8 42) Raub 1 l8 1 10 _ 4 4 4 4 ‘L‘ [A II III \ 204 Table 4.34 (continued) Nunber of Variables Wifin Location Quotient Districts 0 (1.0 1.0 >l.0 43) Tenerlch l 17 0 12 44) Seberang Perai Utara 0 14 l 15 45) Seberang Perai Tengaln 0 l7 0 13 46) Seberang Perai Selatan 3 18 2 7 47) Pulau Pinang’Tinur Laut l 7 0 22 48) Pulau Pinang Barat Daya 3 14 0 13 49) Batang Padang 2 22 0 6 50) Manjnmg/Kuala Kangsar/ P.Hilir/P.Tengah * 0 22 1 7 51) Kinta l 10 0 19 52) Krian 3 19 1 7 53) Larut & Matang l 13 2 14 54) Ulu Perak 3 19 0 8 55) Perlis l 18 0 ll 56) Kuala Selangor/ Sabak Bernan * 2 21 0 7 57 ) Kuala Langat/Ulu Langat/Sepang * 0 14 0 16 58) Gorbak/Klang/Petaling/ Ulu Selangcr/ Kuala Lunpur * 0 9 0 21 59) Besut 3 19 0 8 60) Dungun 0 17 0 13 61) Kenenan 2 20 l 7 62) Kuala Terengganu 1 12 0 17 63) Marang 5 17 0 8 64) Ulu Terengganu 2 19 0 9 Note: * Districts which have been conbired to facilitate conparison. 205 Table 4.35 Gini Coefficients At State, District, Urban District, Rural District, Western District, and Eastern District Scales of Analysis, 1970. AllUrbanRuralWest.East. Variables * States Dists. Dists. Dists. Dists. Dists. 1. Population Literacy 5.33 6.55 4.87 7.89 4.47 9.32 2. Formal Schooling 4.98 6.18 4.18 6.74 4.17‘ 7.33 3. Teachers 4.48 10.25 8.49 9.25 8.85 14.61 4. Infant Mortality 10.52 13.43 12.03 14.27 10.80 9.93 5. Maternal Mortality 22.43 36.51 33.56 32.99 35.42 31.50 6. lbusing Stock 5.84 7.27 5.68 6.32 5.93 4.49 7. Flush Tciltes 31.23 41.77 32.54 38.75 37.31 46.92 8. Piped-Water Supply 22.49 28.09 18.94 35.10 19.79 43.49 9. Electricity 24.16 33.40 21.56 35.66 29.33 32.77 10. Public Security 9.53 18.26 15.38 22.09 17.78 19.56 11. Fire Service 16.20 30.87 24.95 39.21 28.17 39.59 12. Road Density 33.07 47.48 30.92 49.41 39.73 44.61 13. Postal Service 19.97 29.12 24.39 22.45 27.82 26.98 14. Professional end Technical Manpower 13.54 20.96 13.75 14.86 19.49 20.01 15. Adninistrative and Managerial Manpower 25.82 38.14 28.79 24.89 36.88 32.61 16. Goverment Officials 23.31 37.00 33.92 22.99 39.20 18.79 17. Gross Donestic Product 22.28 26.06 25.26 17.51 24.66 20.66 18. Manufacturing Sector 22.77 36.50 24.33 32.18 36.24 34.14 19. Comercial Sector 15.66 24.29 17.91 14.15 23.69 19.70 20. Rural Telephoe Booths 13.11 23.35 15.56 29.13 18.12 39.12 21. Agricultural Extension Service 41 . 81 76 . 12 57 . 85 89 . 16 68 . 55 88 . 57 22. Urban Population 27.33 50.45 28.96 79.52 48.15 52.74 23. Large Urban Centers 32.93 55.62 30.21 100.00 51.13 70.19 24. Female Literacy 6.01 7.47 4.56 11.37 5.34 10.11 25. Fenale in Non-Agric. Econonic Activities 21.05 29.99 22.60 21.26 29.06 28.02 26. Above Poverty Lire 13.69 20.97 15.11 13.04 20.38 17.50 27. Public low-Cost Housing 38.32 50.44 40.54 53.21 44.07 53.08 28. Malays in Agriculture 16.15 22.45 31.80 8.20 26.12 12.58 29. Malays in Manufacturing 33.58 39.08 34.10 32.71 24.31 18.62 30. Malays in Comerce 34.08 40.74 36.35 31.35 29.58 19.15 Average 20.38 30.29 23.30 30.52 27.15 29.56 * Details regarding cperationalization and sources of variables are ascutlired inTable 3.1 andAppendix 1. 206 Table 4.36 Gund.0befficients At State, District, Utban.District, Rural District, Western District, and Eastern District Scales of Analysis, 1980. Variables * States Dists. Dists. Dists. Dists. Dists. AllUrbanRuralWest.East. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. Population Literacy Formal Schooling Teachers Infant Mortality Maternal Mortality Housing Stock Flush Toilets Piped-Water Supply Electricity Public Security Fire Service Road Density Postal Service Professional and Technical Manpower Adninistrative and anagerial Manpower Government Officials Gross Donestic product Manufacturing Sector Comercial Sector Rural Telephoe Booths Agricultural Extension Service Urban Population Large Urban Centers Fenale Literacy Fennale in Non-Agric . Econonnic Activities Above Poverty Lire Public Low-Cost Housing Malays inAgriculture Malays inflhnufacturing Malays inOomnerce Average 4.26 5.20 3.75 6.20 4.03 7.03 4.04 4.69 3.13 5.53 3.10 5.96 2.66 9.44 8.62 8.62 8.98 11.05 8.56 13.97 14.58 12.11 12.53 13.51 30.65 39.22 36.21 34.70 35.66 25.32 3.75 4.43 3.15 4.20 3.84 3.14 11.89 14.30 11.07 17.26 12.79 16.50 15.81 18.49 12.04 27.87 11.37 34.04 11.91 16.73 9.35 22.64 13.92 19.63 20.91 25.98 22.85 30.07 23.13 30.33 12.02 20.38 15.45 30.58 19.12 24.45 33.68 47.37 36.17 48.40 41.17 41.31 16.32 25.18 21.25 20.39 24.18 26.44 10.73 16.87 12.42 10.72 15.76 18.44 33.69 45.26 35.80 22.22 42.98 37.82 22.93 33.37 32.03 19.36 34.70 21.71 20.50 22.85 20.89 16.88 21.76 20.69 23.40 30.88 21.19 33.40 28.39 25.18 13.91 21.47 16.10 11.34 20.16 21.30 15.31 24.99 20.67 29.47 20.97 36.53 54.76 82.14 76.29 86.00 68.74 92.20 21.68 45.44 27.45 73.96 42.20 55.61 23.86 49.12 28.02 100.00 44.92 64.09 5.08 6.26 4.73 6.54 5.27 7.55 18.67 25.74 17.94 22.44 23.61 26.32 12.25 17.92 14.74 9.09 17.34 30.23 27.11 41.48 36.03 48.89 35.85 43.27 19.57 29.29 35.77 9.58 33.67 17.04 16.23 20.76 18.08 16.17 16.13 12.99 24.59 29.30 25.11 23.34 20.51 16.61 18.02 26.28 21.36 26.93 23.56 26.88 * Details regarding cperationalization and sources of variables are as outlired in Table 3.1 and Appendix 1. 207 Table 4.37 Frequency Distribution.of Gini Coefficients for State and District Scale of Analysis, 1970 and 1980 STATE DISTRICT Gini values (%) 1970 1980 1970 1980 0 to 9.99 6 6 4 5 10 to 19.99 8 12 3 6 20 to 29.99 9 8 9 10 30 to 39.99 6 3 7 3 40 to 49.99 1 0 3 5 50 to 59.99 0 1 3 0 60 to 69.99 0 O 0 0 70 to 79.99 0 0 1 0 80 to 89.99 0 0 0 l 90 to 100 0 0 0 0 Total 30 30 30 30 Average 20.38 18.02 30.29 26.28 In.comparing with the other regional units, the Gini coefficients at the state scale are found to be relatively smaller fcr most of the development variables. This means that the magnitude of spatial inequality is relatively less acute at the state scale as compared to other regional units. Table 4.38 indicates this. Figures 4.6 to 4.9 are graphical illustrations of the distribution of pipedrwater supply and agricultural extension service for states and districts in 1970 and 1980. Spatial Inequality: urban and Rural Districts. In.comparing the distribution of the 30 development variables between urban and rural districts, it is found that the rural districts have more development variables with relatively higher Gini coefficients than those for the Piped Water Supply Agricultural Stations Acreage 208 100 90— 1970(Gini - 22.49) 1980(Gini - 15.31) .l Oi/rn 'finfirfirrffirrroiv O 10 20 3O 4O 50 60 7O 80 90 100 Tctcl Occupied Housing Units Figure 4.6: Distribution of Piped Woter Supply in Stotes, Pemnsulcr Moloysrc, 197D and 1980 100 r /;1 j —— 1970mm; - 41.81) / n 90 —- 1980(Ginl -54.76) /’///" /‘ I l 80 ‘ . I , / , 7o / j, 1" / 60-1 // 504 /’ 4c~ .1 .30d 20--l J 10-4 -l 0‘4 r I r r T r r 1 fl 0 10 20 .30 4O 50 60 7O 80 90 100 Total Labor Force in Agriculturcl Sector F'i urc_ 4.7: Distribution of A riculturcl Ex enSIcn Sen/Ice in Stctes, emnsulor Mclcysio, 1970 and 1980 Piped Water Supply Agricultural Stations Acreage 209 100 a a —- 1970mm: — 28.09) 90~ -- 1980(Gini - 18.49) ao-l 70-l .l 604 so-l 40-1 30~ 20-l 10.. c To 20 so 20 so so 70 so 90 100 Totcl Occupied Housing Units Figure 4.8: Distribution of Piped Wcter Supply in Districts. Peninsulcr Moloysnc, 197D cnd 1980 100 —— 1970mm; - 76.12) /’/i 90 —- 1980(Gini - 82.14) /’ i so / / /: /’ ,l l 0101\1 OOO NU OO .5 O 04,..1..1_1.L_L._L.1_L_L 1._l_ L_L_L_L. L.L_L.L_.n 00 / Y \ \ T f 1' fi' , . , 10 2‘0 30 4c Totol Lobcr Force in Agriculture! Sector Figure 4.9: Distribution of Agriculturcl ExtenSIon Service in Districts. Peninsulcr Mclcysic, 197D and 1980 210 Table 4.38 Conparison of Gini Coefficients at State Scale of Analysis Wifin Ofier Regional Units, 1970 and 1980 All Urban Rural Western Eastern Districts Districts Districts Districts Districts Nunber of Devt. Variables Which fie State-Scale 30 22 22 24 23 Has Smaller Gini Values, 1970 Nunber of Devt. Variables Which fie State-Scale 30 21 21 25 24 Has Smaller Gini Values, 1980 urban districts. For iretance, In 1970 nireteen variables have higier Gini coefficients finan corresponding variables for fie urban districts. This figure slightly decline in 1980, fron 19 to 17. Table 4.39 shows fiefrequencydistributioncffieGiniccefficients fcrfienrrbanand rural districts in 1970 and 1980. Similarly, fie mean Gini coefficient of all 30 variables for rural districts is again higrer than for fie mban districts, both in 1970 end 1980 (Table 4.40). In conparing wifin Gini coefficients for districts as a whole, it is found finat urban districts have smaller Gini coefficients for almost all fie developnent variables; 28 in 1970 and 29 in 1980. Rural districts, leoaever, did not fare as well as fie urban districts. In 1970 end 1980 rural districts have 15 developnent variables wifin Gini coefficients finat are higher finan corresponding variables for fie districts taken togefier. This furfier supports fie contention that disaggregation of data is more useful for analysis regional inequalities. 211 Table 4 . 39 Frequency Distribution of Gini Coefficients for Urban District and Rural Districts, 1970 and 1980 U R B A N R U R A L Gini Values (%) 1970 1980 1970 1980 0 to 9.99 5 6 5 7 10 to 19.99 7 9 6 7 20 to 29.99 8 8 6 7 30 to 39.99 8 6 8 4 40 to 49.99 1 0 l 2 50 to 59.99 1 0 1 0 60 to 69.99 0 0 0 0 70 to 79.99 0 1 1 1 80 to 89.99 0 l 1 1 90 to 100 0 0 l 1 Total 30 30 30 30 Average 23.30 21.36 30.52 26.93 Table 4.40 Mean for Tie 30 Gini Coefficients, 1970 and 1980 Scale of Analysis Mean Gini, 1970 Mean Gini, 1980 States 20.38 18.02 All Districts 30.29 26.28 Urban Districts 23.30 21.36 Rural Districts 30.52 26.93 Western Districts 27.15 23.56 Eastern Districts 29.56 26.88 Figures 4. 10 to 4. 13 show fie distribution of piped-water supply and agricultural extension service for urban and rural districts in 1970 and 1980. Piped Water Supply Agricultural Stations Acreage 212 100 -l — 1970(Gini-18.94) // 90-l -- 1930 c ' - 12.04 ,9 J ( Inn ) / // /r // / /:// / /,/ n//// // //’ x’/ l n r I ' Tf l on] 60 7D 80 90 10 Totol Occupied Housing Units Figure 4.10: Distribution of. Piped Wctcr . Supply in Urbon Districts. Peninsulor Mclcysm, 197D cnd 1980 100 T — 1970(Gini - 57.85) / 90 —— 1980(Gini - 76.29) i} /' / / I I c 8 /’/,/ / ll 70 // / i // / I 60 // / / /' // / 50 // /// 4o 1b 20 :50 4c 50 so 7b ab 90 100 Totcl chcr Force in Agricultural Sector Figure 4.11: Distribution of A riculturol extension Service in Urbcn istricts. Peninsulcr Mclcysic, 197D and 1980 O 213 100 —- 1970mm: - 35.10) 90-1 —- 1980(Gini - 27.87) acJ 7o- 604 50- 404 Piped Water Supply :50-4 l Totcl Occupied Housing Units urc .12 Distribution of Piped Woter Supr?y in Rurol Districts, Peninsulcr lvlclcysio, 197D and 1980 100 4 --- 1970(Gini - 89.16) an 90a —- 1980(Gini - 86.00) 31 n /. a: 80-4 .1 L. 4 /,x U / < 70— /./ an 4 ,r" C . o 60 i/' 123 / 3 50 U) _. 40 E / :3 :1 302i // ,I/ 3 l .’ .3 20—1 /./ /,’/ /,' //‘ 0" <1; 10 ”4V // __=,/.,,:'/’/ I ' T T 1 r r ' n ‘5‘ 1 r r * Tf 20 30 40 5O 60 7D 80 90 100 Tctcl Labor Force in Agriculture! Sector Fi ure 4.13: Distribution of A riculturol xtension Service in Rurol istricts, Peninsular Mclcysic,1970 and 1980 214 In exenining fie location Quotient it is found finat large proportioncfurbandistrictshavenerethanfieirprcportionalshare of developnent beefits for nest of fie developnent variables (Tables 4.33and4.34). Ornfieofierhand, uestoffieruraldistrictshave LQs less finan 1.0 on nest of fie developnent variables. Spatial Irequality: Western and Eastern Districts . 'I're categorization of Gini coefficients between western arnd eastern districts indicate fieexisbencecfawidevariationinspatial inequalitywhenfietwc regional groups are corpared. It is observed finat for 1970 and 1980 fie eastern districts have larger number of development variables with higherGinivaluesthancorrespondingvariables forfiewestern districts. In 1970 seventeen developnent variables for fie eastern districts have higler Gini coefficients finan for western districts. In 1980 finis nunber increased to 21 (Tables 4.35 and 4.36). Tie frequency distribution of fie Gini coefficients for western and easterndistrictsisshcwninTable4.41. AscanbeseeninTable 4.40, fie mean Gini coefficient for fie all fie 30 Gini coefficients also shows finat fie eastern districts have higher mean Gini values finan finat for fie western districts, both for 1970 and 1980. As can be seen fronn Tables 4.35 and 4.36, nest of fie developnent variables for fie western districts registered lower Gini coefficients finan fiese recorded for all districts taken togefier. In 1970 and 1980, 28 variables for western districts have Gini values snaller than for all districts togefier. Tie eastern districts did not fare as well as fie western districts. In 1970, 13 variables for eastern districts 215 have greater spatial inequality than that indicated by the districts as a.whcle. The magnitude of spatial inequality for eastern.districts becones worse in 1980 when 18 cf fie 30 developnent variables have higher Gini coefficients finan fiese for districts as a wtele. Table 4.41 Frequency Distribution.of Gini Coefficients fbr'Western.District and Eastern Districts 1970 and 1980 WESTERN DISTRICTS EASTERN DISTRICTS Gini values (%) 1970 1980 1970 1980 0 to 9.99 5 5 4 4 10 to 19.99 5 8 9 8 20 to 29.99 10 8 4 8 30 to 39.99 6 4 6 5 40 to 49.99 2 4 3 2 50 to 59.99 1 0 2 l 60 to 69.99 1 l 0 l 70 to 79.99 0 0 1 0 80 to 89.99 0 0 1 0 90 to 100 0 0 0 1 Total 30 30 30 30 Average 27.15 23.56 29.56 26.88 Figures 4.14 to 4.17 show the distribution.of pipedrwater supply and agricultural extension service for western.and eastern.districts in 1970 and 1980. The Location Quotients in.Tab1es 4.33 and 4.34 indicate that the a large proportion.cf western districts have more than their proportional benefits of development as compared.tc those in the east coast states. Piped Water Supply Agricultural Stations Acreage 216 100 -— 1970(cini - 19.79) I/ 902 —- 1980(Gini - 11.37) éj’ ‘ // ao-j / 703 60‘ 504 4ch :5 »’/ 0 // //’// 20 / 7” .’(,/ CT I n v o v I r a ' e r I “r jfi re I er 0 10 20 3O 40 50 60 7O 80 90 100 Total Occupied Housing Units Figure 4.14: Distribution of .Piped Water . Supply In Western Districts. Peninsular Molaysnc, 197D and 1980 1004 , — 1970(Gini - 68.55) 90-i —- 1980(Gini - 68.74) /.//// 80-4 /‘ 1' 701 / V -i x -l / / so: ,/ / 50~ / / 40- / 30~ 20i ” 'i / i / ,/ 10"1 / // -l / / 0 fl I I I r I e I r I e I I I r j fi— I f O 10 20 30 4O 5O 60 7O 80 90 100 Total Labor Force in Agricultural Sector Figure 4.15: Distribution of Agricultural Extenszon Service in Western Districts, Peninsular Malaysia, 1970 and 1980 Piped Water Supply Agricultural Stations Acreage 217 100 90—4 80d 70-4 60-4 50-4 40-4 .l $50--l .l 20- 10 —— 1970(Gini - 43.49) 1980(Ginl - 34.04) Tfii 510 40 50 Toto! Occupied Housing Units Figure 4.16: Distribution of Piped Water Supply in Eastern Districts, Peninsular Malaysia. 1970 and 1980 70.1 so; 50~ 40-1 .30"-l .1 20" H 10- Ill 1970(Gini - 88.57) ,/ .l 1980(Gini - 92.20) /" ff. 0 0 fi T fl 1' 60 r l 20 fl T 30 ’ T T 1 I o 10 4D 50 70 Total Labor Force in Agricultural Sector Figure 4.17: Distribution of Agricultural Extension. Service in Eastern Districts. Peninsular Malaysia, 1970 and 1980 218 Gaps in Spatial W113! Differences that prevailed between the different regional units (statesanddistricts, urbanandruraldistricts, andwesternand eastern districts) also provide useful insights regarding fie extent of spatial inequality. Gaps between fie different regional units will be ascertaired by dividing their Gini coefficients in fie following manner: (1) Districts' by State's (2) Rural Districts' by Urban Districts' (3) Eastern Districts' by Western Districts' . A Gini coefficient ratio close to unity (1) indicates finat fie regionalmitsbeingcorparedaresimilarwithregardtofieirspatial irequality. Tte larger fie Gini Coefficient ratio, fie greater is fie departure in similarity of spatial irequality between fie regional units in question. Tables 4.42 and 4.43 show fie relative gap between fie different regional units for 1970 and 1980, respectively. Conparisonsoffiesetwctableswill indicatefinechangeinfieGini coefficient ratios between 1970 and 1980. In corparing between fie Gini coefficient ratios for state and district scales of analysis, it is found that for nest of fie variables fie difference in fieir spatial inequality have been reduced (Table 4.44). This means finat spatial inequality between state and district scale of analysis are alnest similar on nest variables in 1980 as conpared to 1970. 219 Table 4. 42 Gini Coefficient Ratios, 1970: Spatial Inequality Gaps Between State and District Scales of Analysis, Urban Districts and Rural Districts, and Western Districts and Eastern Districts Gap Between Gap Between Gap Between Districts Rural and Eastern and and States* Urban Dists.+ Western Dists@. Variables 1) Population Literacy 1.23 1.62 2.08 2) Formal Schooling 1.24 1.61 1.76 3) Teacrers 2.29 1.09 1.65 4) Infant Mortality 1.28 1.19 0.92 5) Maternal brutality 1.63 0.98 0.89 6) Housing Stock 1.24 1.11 0.76 7) Flush Toilets 1.34 1.19 1.26 8) Piped-Water Supply 1.25 1. 85 2. 20 9) Electricity 1.38 1.65 1.12 10) Fire service 1.91 1.57 1.10 11) Public Security 1.92 1.44 1.41 12) Road Density 1.44 1.60 1.12 13) Postal Service 1.46 0.92 0.97 14) Professional and Technical Manpower 1.55 1.08 1.03 15) Administrative arnd Managerial Manpower 1.48 0.86 0.88 16) Govermnent Officials 1.59 0.68 0.48 17) Gross Donnestic Product 1.17 0.69 0.84 18) Manufacturing Sector 1.60 1.33 0.94 19) Oomnercial Sector 1.55 0.79 0.83 20) Rural Telephone Booths 1.78 1.87 2.16 21) Agricultural Extension Service 1.82 1.54 1.29 22) Urban Population 1.85 2.75 1.10 23) Large Urban Place 1.69 3.31 1.37 24) Fenale Literacy 1.24 2.49 1.89 25) Fenale In Man-Agricultural Econonic Activities 1.42 0.94 0.96 26) Labor Force Above Poverty 1.53 0.86 0.86 27) Public low-Cost Housing 1.32 1.31 1.20 28) Malays in Agriculture 1.39 0.26 0.48 29) Malays in Mamnfacturing 1.16 0.96 0.77 30) Malays in Comerce 1.20 0.86 0.65 Average 1.49 1.31 1.09 Note: * States' Gini Coefficient is fie denominator + Urban Districts' Gini Coefficient is fie denoninator @ Western Districts' Gini Coefficient is fie denoninatcr 220 Table 4.43 Gini Coefficient Ratios, 1970: Spatial Irequality Gaps Between State end District Scales of Analysis, Urban Districts and Rural Districts, and Western Districts and Eastern Districts Gap Between Gap Between Gap Between Districts Rural and Eastern and and States* Urban Dists.+ Western Dists@. Variables 1) Population Literacy 1.22 1.65 1.74 2) Formal Schooling 1.16 1.77 1.92 3) Teacters 3.55 1.00 1.23 4) Infant Mortality 1.63 0.83 1.08 5) Maternal Nbrtality 1.28 0.96 0.71 6) I-busing Stock 1.18 1.33 0.82 7) Flush Toilets 1.20 1.56 1.29 8) Piped-Water Supply 1.17 2.31 2.99 9) Electricity 1.40 2.42 1.41 10) Fire Service 1.70 1.98 1.38 11) Public Security 1.24 1.32 1.28 12) Road Density 1.41 1.34 1.00 13) Postal Service 1.54 1.06 1.09 14) Professional and Technical Manpower 1.57 0.86 1.17 15) Administrative and Managerial Manpower 1.34 0.62 0.88 16) Governnent Officials 1.46 0.60 0.63 17) Gross Donestic Product 1.11 0.81 0.95 18) Manufacturing Sector 1.32 1.58 0.89 19) Comercial Sector 1.54 0.70 1.06 20) Rural Telephone Bccfins 1.63 1.43 1.74 21) Agricultural Extension Service 1.50 1.13 1.34 22) Urban Population 2.10 2.69 1.32 23) Large Urban Place 2.06 3.57 1.43 24) Fenale Literacy 1.23 1.38 1.43 25) Fenale In Non-Agricultural Econonic Activities 1.38 1.25 1.11 26) Labor Force Above Poverty 1.41 0.62 1.74 27) Public low-Cost Housirg 1.53 1.36 1.21 28) Malays in Agriculture 1.50 0.27 0.51 29) Malays in Manufacturing 1.28 0.89 0.81 30) Malays in Comerce 1.19 0.93 0.81 Average 1.46 1.26 1.14 Note: * States' Gini Coefficient is fie denoninator + Urban Districts' Gini Coefficient is fie denoninatcr @ Western Districts' Gini Coefficient is fie denominator 221 muse“ Gini Coefficient Ratios -- Direction arnd Rate of Change ( Tables 4.34 and 4.35 Relevant) Dist/ State Rural/Urban East/West 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 21. 22. 23. 24. 25. 26. 27. 28. 29. mmmgfink flmhmnas meM%eSmmy exumuy mmmemfiy nmSeuw manage mmaSewm meeemmam mmMleWme mmmemueam mmmfiaMmmmr Gwemmtmfidas amsmmaummd Mmhammg%®m Cmmmhl$®m mmlmemmemmm mnmnuaemmem &mme mmnmdem MmeWMnmme meeummq mmhianmno rammcmuwue wwemwnyum Pmucmwme Emmy Mmqsmfigmmmm mesmMmMmmnm mhwinmmem 8 8888885888888 mew L NOHL ' ° Sum» memmwe st OHHN HNl-‘N ssss ss ssss ssssss s sssssssssssss O 8888388888888 testes e ssss ssssss s sssssssssssss Lube HGOH 5% ss sea see -0.07 8885888888888 L. e NI-‘Ull-‘UIS o eLOHH 01 ssss ssssss s sssssssssssss s 90 00) A as ssss é asse Note: * Details regarding cperationalization and sources of variables are outlined in Table 3.1 and.Appendix 1. Mmuwvmmsmmsa®ammrfiechmemtm Gmimdfidmtmucmmmnmemucfwmwamm wmemd 222 Between urban arnd rural districts, fie Gini Coefficent ratios indicateanincreaseinfieirrelativegap. Thisconnotesfinatfie Gini values for fie urban districts have decreased at a relatively faster rate finan finat for fie rural districts. Tie relatively smaller Gini Coefficients which urban districts register (Tables 4.35 and 4.36) as well as better location Quotient performance which fiey have (Tables 4.33 and 4.34) clearly indicate that developnent variables are relatively better distributed wifinin urban districts rafier finan rural districts. This enhances fie hypcfiesis that spatial inequality in developnent are relatively Inere acute for rural rafier finan urban districts. Corparirg western and eastern districts, fie Gini coefficient ratios indicate finat for half of fie developnent variables (15) fieir relativegapshaveincreasedin1980asconparedtol970. Ttemeanof fie Gini coefficient ratios for 1970 end 1980 also indicate finat fie relativegapbetweenwesternandeasterndistrictshaveincrease rafier finan decreased (Tables 4.42 and 4.43). CnangesOverTime Table 4.45 contains fie Gini coefficient ratios finat are derived by dividing fie Gini coefficients recorded for 1980 by fie corresponding Gini values for 1970. A ratio of urnity (1) indicates re charge fron 1970 (status quo). On fie ofier hand, a ratio of less finanoecanbetreatedasameasureofclnangeinspatial irequality, towards fie direction of reduction its magnitude. Conversely, a ratio 223 Table 4. 45 Ratios Between Gini Coefficients of 1970 and 1980 States, Districts, Urban Districts, Rural Districts, Western Districts, and Eastern Districts. Urban Rural Western Eastern States Dists. Dists. Dists. Dists. All Dists. Variables* 5166 058 25 8 66 15 25 7 0001000000100 0 IIIMIO. 1100 01.. 0100 41615 778 1 15 1969 m7oloeus46mM7 a nwmmwm mew as 8256 anmlllnwmnmnmnwllnu. 0 100001.. 1000 00 0100 92 55659 8681 6 6 2794 Qnmnwmlmmnwnwnmlnmnw 0 000101.. 0010 10 0100 521854 2977 5 98 m7nssssmefinla % umammn age 79 .se 0011100000110 0 100001 1001 00 0100 962 6 62 9 85 8 65 20 2 oomlmnmmnmnwnmllnw 0 100001 1000 00 0100 1917 8094922 9 1925 99 11 2 OOOOInwnmnwnwnmn/ulnw 0 1001.01 1000 00 0100 Technical Manpower 15. Administrative and Managerial Manpower 16. Government Officials 1?. Gross Donestic Product 18. Manufacturing Sector Service 22. Urban Population 23. Large Urban Place 24. Fenale Literacy 25. Wonen in Non-Agric. Economic Sectors 26. Above Poverty-Lire 27. Public Low-Cost l9. Ootmercial Sector 20. anal Telephoe Booths 21. Agricultural Extension 28. Malays in Agriculture 29. Malays in Manuftg. 30. Malays in 1. Population Literacy 2. Schooling 3. Teaclers 4. Infant Mortality 5. Maternal Mortality 6. Housing Stock 7. Flush Toilets 8 . Piped-Water Supply 9. Electricity 10. Fire Service 11. Public Security 12. Road Density 14. Professional and 13. Postal Service 0.87 0.92 0.88 0.87 0.91 0.88 Average Table 4. 46 224 Frequency Distribution of Gini Coefficient Ratios Showing Convergence, Divergence, arnd Status Quo in Rate of Gnange Number of Developnent Variables Indicating Regional Units ‘Convergence' ‘Divergence' ‘Status Quo' States 22 8 O Districts 22 7 1 Urban Districts 20 10 0 Rural Districts 23 6 1 Western Districts 21 8 1 Eastern Districts 20 9 l ofnmrethanoecanbeviewedasanindicationof afurfierincrease in fie extent spatial inequality in 1980 as compared to 1970. As can be seen in Tables 4.45 arnd 4.46, convergence has been more frequentfinandivergenceorstatusquoformostoffiedeveloprent variables between 1970 and 1980. Anong fie 3O developnent variables, ‘Housing Stock', ‘Toilets', ‘Piped-Water', ‘Electricity' and ‘Malays InManufacturing' havepositiverateofchange finataremorefinanfine mean rate of charnge for all fie Gini coefficient ratios. ‘Malays In Comerce', ‘Population Literacy' and ‘Fire Service' conprise fie next group of variables whose Gini coefficient ratio were more finan fie mean rate of change for all fie 3O Gini coefficient ratios. Conversely, fie developnent variables ‘Public Security' and ‘Malays In Agriculture' show an irncrease in fie Gini coefficient ratio for all regional units as indicated in Table 4.45. The next group of developnent variables for which there are divergence on most of the different urnits of observation are ‘Maternal Mortality' , ‘Aoninistrative and Managerial Manpower', ‘Rural Telephones', and ‘Agricultural Extension Service' . 225 Even though fie rate of charge of fie Gini values between 1970 and 1980 indicate that for most of the developnent variables fie directionofchargeismovingtowardsconvergence, fieirGini coefficients are still quite high (Tables 4.35 arnd 4.36) to denote significant charge in spatial inequality. Tables 4.47 arnd 4.48 show fie rankirg of fie 3O developnent variables acoordirg to fieir Gini Coefficient ratios of charge in performance. Tre highest rankirg corresponds to variables which relatively have fie best rate of reduction in spatial irequality. Sumnary of Finding Analysis of spatial pattern of developnent in Peninsular Malaysia for 1970 arnd 1980 reveals that wide variations in developnent performance exist at all scales of analysis (states, districts, urban districts, rural districts, western districts, and rural districts). Indices of deveth disclose that smaller regional units disclose greater variations finan larger regional units. Bofin 197C and 1980 indices of developnent, nmweighted and weighted, indicate that developnent performance are more pronounced at fie district rather finan fie state scale of analysis. Between urban and rural districts, fie study finds finat urban districts have relatively better developnent performance finan rural districts. Among urban districts, it is found finat finose that contain large urban centers consistently occupy top rankings for districts wifin high positive developnent irndices. Similarly, districts in fie west coast states fared better finanfinoseinfieeastooaststates. 226 Table 4. 47 Ranking of Gini Coefficient Ratios of 1970 end 1980 (Conparisons of Variable Performance Wifinin Regional Unit) .All Urban RuralWestern Eastern Variables* States Dists . Dists. Dists. Dists . Dists . 1. Population Literacy 13 9 9 10 20 7 2. Schoolirg l4 8 8 13 8 12 3. Teacrers 4 22 21 19 23 9 4. Infant Martality 14 24 26 15 26 28 5. Paternal Nbrtality 29 25 23 27 23 11 6. Housirg Stock 5 4 4 5 4 4 7. Plush Toilets 1 l l l l l 8 . Piped-Water Supply 6 5 6 10 3 10 9. Electricity 3 2 2 4 2 2 10. Fire Service 10 5 5 9 6 3 11. Public Security 30 29 30 30 3O 29 12. Road Deeity 23 23 25 23 25 17 13. Postal Service 16 15 12 17 15 20 14. Professional and Technical Manponer ll 10 15 7 10 16 15. Adninistrative arnd Managerial Manpower 27 28 27 16 28 25 16. Government Officials 22 20 18 14 19 25 17. Gross Donestic Product 21 17 11 21 16 21 18. Marmfacturing Sector 24 13 12 26 9 6 l9. Comercial Sector 18 17 15 12 13 24 20. Rural Telephoe Boofins 25 25 29 25 26 17 21 . Extension Service 28 27 28 21 22 22 22. Urban Population 11 20 19 19 16 23 23. Large Urban Place 8 17 17 24 16 15 24. Female Literacy 17 12 22 3 21 7 25. Wonen in Non-Agric. Eoononic Sectors 18 15 10 28 10 19 26. Above Poverty Lire 18 13 20 6 13 3O 27. Public low-Cost Housirg 7 ll 14 18 10 13 28. Malays in Agriculture 26 3O 24 29 29 27 29. Malays in Manufacturirg 2 3 3 2 5 4 30. Malays in Comeroe 8 7 7 8 7 l4 Note: Sane rankirg for a variable means that Gini Coefficient ratios are fie same. 227 Table 4.48 Ranking of tie Gini Coefficient Ratios of 197C and 1980 (Conparisons of Variable Performance Aneng Regional Units) All Urban Rural Western Eastern Variables* States Dists . Dists . Dists . Dists . Dists. 1. Population Literacy 2. Schooling 3. Teacrers 4. Infant Nbrtality 5. Maternal Nbrtality 6. Housing Stock 7. Flush Toilets 8 . Piped-Water Supply 9. Electricity 10. Fire Service 11. Public Security 12. Road Density 13. Postal Service 14. Professional and Technical Manpower 15. Administrative and Managerial Manpower 16. Government Officials 17. Gross Donestic Product 18. Manufacturing Sector 19. Comercial Sector 20. Rural Telephae Booths 21. Agricultural Extereion Service 22. Urban Popilation 23. Large Urban Place 24. Female Literacy 25. Wonen in Nan-Agric. Eoononic Activities 26. Above Poverty Line 27. Public Low-Cost Housing 28. Malays in Agriculture 29. Malays in Mamnfacturing 30. Malays in Comerce HubO‘U'lthIQOI-‘I-‘nbm unwwonoxononunwweonw H Honr-aononN on onHmHunmnbonI—Ioswnhw 0010103 O‘UIIhHIbU'I U'I NOIhI-‘HNHHUIUIO‘NN NHO‘U’IHH Haunt-J bub bf-‘l-‘Ln U1I§UII>UIO N N900.) 0300010.) N0) 0.)wa wwwwwp (A) wwwwpwpwpwwww HO‘ HU'IIbN NN UINNN #NNNNQ) lb LOUII-‘Ile-‘anNnhUlf-‘m con—awh- i-‘wi-‘UI U'IH U1NNO3 @0000 0‘01 Note: Sane ranking for a variable means that their Gini coefficient ratios are tte same. 228 Urban districts which contain relatively large urban centers are also inportant adninistrative and comercial centers. 'I'te more urbanizeddistrictsalsocontainstatecapitals. Mostoffielarge urbancentersinfiedistrictsarealsoinportantpoints formed, railway, air, and shipping transportation. Large financial, busiress, arflacadenic institutionsarealso locatedwifininnestcffieseurban districts. Distictsinfiewestcoaststateshavebeenexposedboelenents ofnedernizationearlierthanfinoseinfineeastcoaststatesof PenireularMalaysia. “newestcoaststateswerefinefirsttobe colonized by fie Enmcpeoe. Under British colonial rule, fie first roadandrailwayretworkwereestablisredinfiewestcoaststatesof Penireular Malaysia to facilitate fie exploitation of tin and rubber industries. Until today, fiewestcoaststateenotcnlyhave unportantstretchcftinandrubberbelt along itscoastalplainbut alsodenseretworkofcomunicationlinesasconparedtoeastcoast states. Largetractscf landsinfineintericrpartoffieeastcoast statee of Pahang, Kelantan, and Terengganu are still covered by virgin jungles. To develop finese areas, regional developnent aufinorities have been established, nanely fie Pahang Tenggara Developnent Authority (or DARA in Pahang), fie South Kelantan Regional developnent Aufinority (or KESEDAR in Kelantan), and fie Terengganu Tengah Regional Development Aufinority (or W in Terengganu). Also, highways havebeenandarebeingbuilttolinkfineeastcoaststateswifinwest coast states. This study demonstrates finat fine pattern of spatial developnent 229 in 1980 is not significantly different fron finat in 1970. Nbst of fie districts (or states) finat have high developnent indices in 1970 continue to have relatively high level of developnent in 1980. Also, corparisoe of districts' rankings between 1970 and 1980 reveal finat fieir hierarchy of developnent are not significantly different. Analysis of spatial irequalities indicate that developnent beefits are not proportionately distributed among fie different scales of analysis. Gini coefficients show finat district scale of analysis reveal greater magnitude of spatial inequalities finan at fie - statescale. Inconparingbetweenurbanandrural districts, it is observed finat fie magnitude of spatial inequalities is less acute for urban rafier than mal districts. Similarly, distribution of developnentisfoundtobemreirequitableenongdistrictsinfie eastcoaststatesofPeninsularMalaysiawhencouparedwifinfinosein fie west coast states. Location Quotients confirm finat developnent beefits are disproportionately distributed among districts . Analyses of spatial pattern of developnent and magnitude of spatial irequalities demonstrates and confirm that regional units below fie state level provide more information regarding fie existence of wide variations in developnent performance and fie intensity of spatial inequalities. Ttefindingsoffinisstudyconfirmprevious studiesbyLeinbach (1972) and Osborn (1974b) which found districts wifin large urban 230 centers to be peaks of developnent surface in Peninsular Malaysia. In finding that spatial inequalities are more acute for fie rural refiner thanfineurbandistricts, finisstudyisalsoinaccordwifinAnand's (1983) study. CHAPTERFIVE POLICY IMPLICRTIONS FOR EUJITABLE DEVEIDPMENT Basedonfieanalysisofdevelopnent indicesandGini coefficients as well as knowledge gained while preparing this research, finis chapter outlires sane of fie salient implications for developnent policy and strategies finat could be used to address fie problene rendered by fie persistence of inter-regional disparities and to promote fie accouplistment of more equitable distribution of developnent beefits in Malaysia. Tre first part of this chapter discusses fie specific implications related to fie results of fie analysis of data. ‘ne second part outlires geeral implications for articulation of more coterent and effective developnent policies and strategies for mounting equitable developnent. It should be pointed, however, finat fie implications indicated here do not cover all fie possible ramifications, but rafier provides fie fundamental framework for furfier deliberations. It is in fact imperative finat further discussions should be undertaken before appropriate developnent policies and strategies are formulated and implerented. Indeed, suggestionsinthischapterareintendedmoretoevokefinoughts for action rafier finan to prescribe solutions . 231 232 Develcpnnent Policy Implications From Results Of Tl'e Study 'I‘hescale atwhichitmlldbeappropriatetoundertake evaluation of developnent performance, by and large, depends upon fieobjectivebeingpursued. Iffiepurposeforundertakingfie evaluation of deveth performance is to obtain information to retires problens posed by fie persistence of inter-regional disparities in developnent, fien, it is apparent frmn finis study that fieuseoffieunitofanalysisatfiestatescaleisnotasuseful asfinatofusingfiedistrictscale. AsshomninCrnapter4, fie aggregationofdatatendstosuppressorconcealsoneoffieimportant details of irequality. 'Ile aggregation process also have fie potential of misclassifyirng areas finat are relatively less developed and vice-versa for finose finat are relatively more developed. This, forexanple, wasborreoutbyfiefactfinatalfinoughfieindexof developnent forfiestateofSelangor, asawhole, wasfoundtobe highly positive, such information aloe, however, cannot reveal fie factfinatcertaindistriotsinfiestatesuchasSabakBernam, Kuala Selarngor, Ulu Selangor, Separng and Kuala Langat were not only relatively less developed wifinin fie state, but also in conparison with fie ofier districts in fie ofier states in Peninsular Malaysia. On fie contrary, alfinough fie state of Kelantan was found to register relatively low, regative index of developnent as conpared to fie ofiner states, information derived at this scale of analysis again could not revealfiefactfinatfiedistrictofKottharunotonlyhasa developnent index finat was relatively high and positive amongst fine 233 districtsinthatstatebut, also, wtenitisconparedwifinfieofier districts in fie ofier states of Peninsular Malaysia. As evidenced froa fie five-year developnent plan documents, evaluation of developnent performance for Malaysia, in geeral, tends to be mndertakenusingfiesectoralapprcachatfiestatescaleofanalysis. Trelackofdetailsbyfieuseoffinestatescaleof analysismay prevent fie proper identification of areas which are critically afflicted wifin developnent problems. Heavy reliance on information derived fron relatively larger units of cservation may furfier contributetofieperpetuationandpersistenceoffi'cseareaswhich are relatively less developed or acre developed. It is apparent finat if acre details and greater accuracy of information are required, study at a relatively smaller scale of observations is needed. The United Natioe (1971) recomnends fie use of districts for data collection. Srivinas (1977 ) also points out that micro-studies give greater accuracy of figures as compared with ficse fron macro-studies. 'ITe practicality of using fie districts as pragmatic territorial framework for planning and implementation of developnent has also been supportedbyFriednann (1980) whocontends finatfiedistrictsare largeenoughtomeetacstoffiebasicneedsofthepopulationand fiey are also relatively small enough so that fie entire population of fieareacanhavereasonableaccesstodevelopnent innovations (see Higgins, 1980). In sum, it is recoanended finat monitoring and evaluation of developnent in Malaysia be undertaken at fie district level for fie following reasons: (1) It is smaller than fie state-scale 234 (2) It is fournd in all states in Malaysia (3) It constitutes fie lowest level at which developnent planning arnd implementation has been undertaken (4) It constitutes fie snallest adninistrative level at which public agencies are represented. Thisrecomendationisinconsonancewithofficialproposaltoupdate fie district adninistrative system so that it can be used effectively for developnent planning and adninistration (Third Malaysia Plan, 1976, p.265). Allocation of resources for deveth slnould rct aerely be based upon fie criterion of fie overall developnent performarnce of fie state, but rafier it should consider fie actual developnent performanceoffiecorponentpartsoffiestates. Infinisway, finose areas finat are relatively less developed would rct be overlooked and henceslmldreceiveacrepricritythanareas thatarealready relatively acre developed. Consideration of fie allocation of resources based upon fie developnent performance of fie state might result in fie neglect of fie relatively less developed areas within fie state fron getting appropriate attention and action. For example, if fie state of Selangor continues to show promising level of developnentandfieallocationofresourcesweremadeonsuch criterion aloe finen fie relatively less developed districts in Selangor, such as Kuala Selangor, Sabak Bernam, Sepang, Kuala Largat andUluSelargormaysuffer; sincefineywouldbeconsideredaspartof fie relatively acre developed state. Findings based at fie district 235 scalecanbeauseful tool forfieallccationofresourcesand implementation of programs in areas where small increments of developnent may provide larger overall returns. It can be argued that fiefindirgsderivedfronfiestatescaleofanalysiscanbefurfier treated by micro-level studies. However, if fie micro-level studies areundertakeninpiecemealmanner, fienfiefindingsproducedmay also have limitations; since fie consequent actions world be ad—lnoc in nature and wificut integration with fie other regions. Ttecoapositeindexofdevelopnentproducedbyfinisstudyfoand finat, in geeral, acst of fie districts in Peninsular Malaysia continued to have below average developnent performance. Comparisons of developnent indices and Gini coefficients between urban and rural districtsstcwfieturbandistrictstendtohavebetterdevelopnent performance and less irequalities. Similarly, conparison of developnent performance and regional irequalities between western and eastern districts reveal finat western districts have greater developnent beefits and relatively less regional irequalties than easterndistricts. Sinceacstoffieeasternandruraldistricts continued to register below average developnent performance, this fien suggests finat fiese districts ought to be given priority in developnent plannirg and implementation. 'I‘Ie fact that fie study also found finat acst of fie districts wifin relatively large urban centers tend to register highly positive developnent indices, finis, too, suggests finat developnent policies and planning ought to give greater focus on deveth in fie less urbanized districts. 236 This study has establisted that regional inequalities have nct charged significantly between 1970 and 1980. However, during fie same periodecononicgrowfinhasincreased. TreGrossDonesticProduct (GDP) per capita in 1971 was (Malaysian Ringgit) $1,172/=, while in 1980 it inncreased to $1,836/=. This inaplies finat fiere is discrepancy between growfin performannce annd distribution of developnent beefits . Tneexistenceofsuchagapreflectspoorlyonfiegrowthwifinequity or distribution strategy to developnent. Corrective masures are necessarytoensurefinatminimaldiscrepancyprevailbetweengrowfin performance annd distribution of developnent beefits. More workable approades to bring about effective distribution of developnent beefits must be devised and pursued. 'I'leuseofdevelcpnentindexasconstructedhereshowsfinatitis possible to gauge fie developnent performance of fie various regions in Malaysia. What is required is fie availability of fie relevant datapertainingtofievariablesfinathavebeendeterminedas constituting important indicators of developnent. Since fie Standard- ScoreAdditivelVbdelusedhereisnotassophisticatedasofier technniques (such as Prinncipal Component Analysis and Factor Analysis ) , it can fierefore be easily applied by developnent administrators at fie district level. Furfiernncre, fine meficd is acre useful because it enables coaparisons over time. Tne Standard-Score Additive Model sl'culd fierefore be considered for application in evaluating developnent performance of fie various districts in Malaysia. Tne Gini coefficient and location Quotient technniques have been founnd to produce results finat are helpful in determining how fie 237 different developnent inndicators are distributed spatially. Findings fron fie use of Gini coefficients annd Location Quotients offer useful inputs for developnent policy annd strategies to prorcte equitable distribution of developnent beefits . Resources for developnent can be targeted or allocated to finose developnent variables that are still nnot effectively distributed. Tne use of Gini coefficient and location Quotient teclnniques provide useful analytical tools for determining fie extent of spatial inequalities and in relping to improve decision- makirng for fie efficient and effective allocation of resources to developnent variables arnd regions in critical need of finenn. 'ITeanalysismadewifinregardtofieperformanceofspatial developnent in Peninsular Malaysia inndicates finat fie relative developnent gap between fie various adninistrative districts has nnot changed significantly between 1970 end 1980. The persistence of disparities in developnent between fie various regions in Peninsular Malaysiarequiresfinateffectiveaeansannstbefomdtoaddressfie problen. Unabated regional disparities are undesirable annd may affect future developnent initiatives . Fundamental to fie identification and pursuanceof appropriateaneasurestodealwifinfieproblemofregional disparities is fie existence of a consistent public policy ccamitment. While it is true that anere policy comnibnent to reducing inter-regional inequalities in developnent does not guarantee fie acconplishment of such objective, it will, as Fields (1980) ncted, be a useful step in facilitating fie efforts to address fie problem; since in its absence fie flow of resources for developnent to areas in need of attention 238 may nnot be properly chanelled. Evenfinoughfieuseofweights forfiedevelopnnentindexdidnct produce signnificannt difference in fie relative position of fie districtsanndstates, finisdoesnotmeanthatfinisapproachwouldnct be useful. Since weights change fie relative importance of fie variablesusedtogauge developnent, theycanfiereforebeusedby decision-makers to test fie effect of resource allocation schemes upon various societal groups. when resources are limited annd determinnation of priorities are necessary, fien fie use of local residents' value judgments regarding developnent objectives annd reeds facilitates plannnning annd decision-making annd also stimulates active participation of fie local innhabitants in planning annd pronction of developnent from fie ‘bottonn—up' . Developnnentasanormativeconceptandprocessisnctattainable in and by itself. Developnnent has to be purposefully organized and activelypursuedinordertoacconnplishwhathavebeendeterminedto be desirable by fie society. As Portes (1976, p.59) aptly states: Innstannces of accelerated econonic growth or social transformation anog acdernn nations have geerally been preceded by deliberate policies initiated and sustained by national governments. Givenfieurgencyoffiedevelopment agendaaswell asfie factthat fierecognizedpurposeof developnent istoproncteandachieve desirable changes, fien, it is important that fine developnent process beguidedbyfieGovernmenttowardsgcalsfinatproducefieacst beefits for society. Sinnce fie private forces or fie market aechannismarennotconcernedwithdistributionalanndequityissuesin 239 developnent, fieycannnctbeexpectedtoguidefienationtowardssuch equity goals (Adelman, 1978; United Nations, 1951). It would nnot be realistic .to expect finat fie spontaneous generation of conditions, on: fie natural forces of adjustment annd equalization, would guide towards desirable developnent (Soja annd Tobin, 1977). Rnrfiernncre, fie conflicting interests finat exist in fie private sector poses acute problenns for reconciliation towards a cormon interest or good. Tne only alternative left is fie government; fie legitimate representative offiepeopleinadeaccraticstatebasedonfiemandateitpossesses byvirtueofpopularelections. Ofcourse, fiereisnoassurancefinat fie developnent process unnder public intervention will automatically steer itself towards fie desirable equity goals (Chan, 1979). However, fie government by virtue of its control over allocation of resourcesandfiepowerfinat ithastoestablishpriorities for implenentation of projects annd also because of its concernn for its own survival annd continuity, it represents a acre realistic sector to proncte equitable developnent (also see Ford, 1966). As Gore (1984, p.241) points out, "fie government possesses the executive annd legislative power to regulate fie developnent process so as to ensure finatfailuresinnfiemarket forcesareavoi®dandtoprioritizefie execution of fie developnent process so as to provide developnental actions on finose areas that need finem acre finan ofiers." Similarly, Spodek, 1976, p.106) has aptly stated that "fie key variable in developnent is . . . fie political system which underlies it." In view of fieir urgency and connplexity, problems annd issues associated with 240 development require considerable governmental action (see Taylor annd Williams, 1982). It is fien geerally agreed finat governmental intervention is desirable annd feasible in order to reduce regional inequalities in developnent (Jumper et al., 1980; Rhoda, 1982, p.10). As Rlnoda (1982, p.10) has aptly stated: "Tie action of all governments influece patterns of developennnt [since] government decisioe annd actions have impacts on regions." Tlesurve‘ycarriedoutforfinisstudysupportsfinispublic approach. 97 percent of fie respondents inndicate finat fie role of fie government is eifier "fairly important" or "very important" in reducing regional inequalities in developnent. Implications Fronn Personal Experience Problenns which fie aufinor experienced while doing fie research also indicate matters finat could be considered in efforts to redress regional inequalities and proacte acre equitable developnent benefits. (Xe of fie problens confronting developnent planning in acst Third World countries is fie lack of adequate data (Cubukgil, 1981; Elyas Onnar, 1980). This is also true for Malaysia, wl'ere relevant data below fie national level are eifier insufficient or difficult to gafier. Inadequaciesofdataareduetoanumberofreasons: (1) lack of guidannce as to fie types of data to be collected, (2) innadequate staffing to collect fie pertinent data, (3) budgetary limitations, or (4) problens of reliability of data. Sore useful infornnation pertaining to developnent performance is currently available at fie 241 stateanddistrict levels. However, improperstorageandpoorrecord keeping makes accessibility of fie data very difficult. The availability of data at fie national level aloe is not adequate to properly gauge fie performance of developnent of smaller units. Most offiedatafrcadocnmentedandpublisredsoircesfronfieneparbnent of Statistics, Malaysia do nct contain sufficient details at fie district level. Since availability of information pertaining to fie performance of key developnent variables is necessary annd vital for fine articulation of effective equity goals annd strategies, it is fierefore essential to develop sounnd data collection systen by incorporating at fie district level responsibilities wtereby basic data are originated fronn fie areas in question. As Manngahas (1982, p.270) puts it, "Tnere is need . . . . to collect fie relevant data sofinat allcocernedcanbeguidedbyfieobjective facts." A resolution passed at fie National Seninnar on District Officers, Peninsular Malaysia in February, 1983 calling for fie establislment of basic data at fie district level is a clear manifestation of such need. T‘te difficulty of retrieving data wifinout unnecessary delay is also equally important. A systen which can facilitate fie efficient storageandretrieval ofdata, suchasfieuseofconputers, could perhaps be given consideration to accomplish such purpose. Alfinough it is true that " statistical systenns cannct be revolutionized overnight" (Scott, 1979, p.451), systenmatic integration of variols resolroesisneededtoaddressfieproblenoffiepaucityofdata vital for gauging fie perforanannce of developnent. A prerequisite in fie efforts to deal wifin fie problems of 242 regional disparities in deveth is not only fie ability to identify areas or regions which are afflicted with developnent problens, but also to have a acdus operanndi finat ensures conparability offiefindingsinspaceandtime. Infinisregard, itisimportanntto rennember finat developnent being multi—dimennsional in character requiresapracticalmechanismtosnmmarizeorsynfiesizeonfiebasis ofcriteriafinatarecomcntoall regions. Tobeabletodevelopa uniform specification of inndicators, it would be required finat fie working definition of developnent is clearly understood. Constant exchange of views annog relevannt government agecies are needed to arrive at a feasible annd workable definition of development. Thisstudycouldnotcoverallaspectsordimensionsof developnent. Refinenennt of the variables to be used for measuring developnentperfcrmanceissuggestedtoensurethatfieimportant aspects of developnent are included. It is important to note finat eachoffievarioisaspectsofdevelopment isaplencmecninitself. Since deveth is a connplex multi-dimensional process, fie different aspects of developnent are interrelated. While oe or acre of fiese aspects can gain prominence independently, ultimately all austoccurifasocietyistoundergoachange. Suchtransitionis facilitated when all fie salient aspects of developnent evolves acre or less sinultaneously (Abdul Hamid, 1979). To designn developnent strategies finat will effectively address fie problen of inter—regional inequalities, it is desirable finat fiere exists proper coordination of various policies. It is clear that 243 regionalinequalitieswifininacounntrycannctbecorrectedby implenenting uncoordinnated annd unnintegrated regional plan: since fie developnent of oe region affects fie developnent of ofier regions. Alfinough fiere are variols "regional" plans in Peninsular Malaysia, fiese plane are studied annd plannned in isolation and at different tines by different consultants. "Ncst plans fail to forsee fie inter- regional effects of fieir proposals" (Abdul Hamid, 1979, p.137). T'te sneer number of different, isolated regional planns contribute to fie problen of coordination and, lence, add to fie problen of efficient allocation of scarce resources. As Higgins (1975, p.395) nctes: "One cannot plann effectively for any oe region in isolation. T'te plans austalwaysbeputintofiecontextoffienational econonny, withfull cognizance of inter-regional . . . flows." Since fie various adninistrative districts in Malaysia are unnder fie jurisdiction of states, it is necessary finat fie varions state governments cooperate togefier to ensure finat regional planning annd developnent will not create conflicts that could ultimately have negative innpacts on fieir developnent. As Abdul Hamid (1979, p.146) correctly observes: "If developnent plannnning at fie regional level is to be acconplisled, cooperation between states is ananndatory . . ." (also see Honjo, 1980). It is imperative finerefore that fie Federal Governmentprovidefienecessarysupportandguidelines. Sincencst of fie funnding for developnent projects comes from fie Federal Government it could annd sl'culd ensure effective coordinnation of regional developnent by exercising its power annd control over finance. As Deenond (1975, p. 173) states: "One of fie acst important steps in 244 policy implenentation is fie budget. In fact throlgh fie budget process policies are often changed annd priorities altered in fie interests of oe sector or ancfier." As mentioed before, in dealing with developnent it stculd be recognizedfinisisachangeinseveraldinnensionsfinatamstbe interrelated annd whose ultimate end is fie enhancennent of societal well-being. Since developnent is multidimensional inn nature, fie approach to developnent plannning ainst fierefore reflect such aultidiaensionality. Since problens of development are area specific, as Prantilla (1981) aptly puts it, policies directed towards deveth of fie region(s) has to be also area specific (also Misra. 1980). A number of enpirical studies have dencnstrated finat reliance on inndirect regional strategies, such as dependence on trickle-down effects, may nnot only prove to be ineffective but also requires loger tine-frame for its achievenent. Evaluation of developnent performance is a basic prerequisite inn fie process annd efforts to prorcte developnent. It is just as important as fie plannnning, implennentation and coordinnation aspects of developnent. It has been slnownn bofin in fie literature as well as in fie practice of developnent finat evaluation of developnent is an important conponent of fie developnent cycle (Drewe, 1979). Information derived from evaluation of developnent will not only indicate fie conditions annd status of developnent for fie unnit of analysis unnder examination, but also fiey can becone useful inputs for 245 future developnent plannnning. Figure 5 . 1 depicts an evaluation process acdel for developnent, describing fie flow from objectives annd stanndardsfinroughaeasurenent, connparisonandcorrection, anndback again to fie objectives. As Robbin (1976, p.414) states: "not only does plannnning influece evaluating, fie process also works fie ofier waysinceeffectiveevaluatingprovides feedback . . . . Trerefcrewe should view planning annd evaluating as being closely linnked, each influencing fie ofier." Despite fie fieoretical annd practical importance of evaluating deveth performance, fiese activities do not seen to be given as auch eaphasis as fie planning, implenentation annd coordinnation functions of fie development cycle. This is clearly borneortbyfieexisteceofanunberofplamingandimplenentation coordination agencies at fie federal annd state levels annd fie noticeableabsencecfanyspecificbodytohandlefieacnitoringand evaluation responsibilities. It appears finat fie acnitoring annd evaluation activities are left to fie discretion of fie various plannnning annd operating agecies to undertake fie task in addition to fieir existing responsibilities. More often than not fie task of acnitoring annd evaluation of developnent performance do not receive as much attention as fie plannnning, implennenntation, annd coordination functions. Whilenctspeculatingonfiereasonsforfieresidual attention given to fie monitoring annd evaluation activities, it is 246 _ _ __ DEVELOPMENT 1.1. I. OBJECTIVES I I I I I l l ' STANDARDS OF DESIRED ' PERFORMANCE l I 8 ...4 I u I .3 > ' 8 1 ___SL H 2 DEVELOPMENT 3 DEVELOPMENT PERFORMANCE ‘ .5 PERFORMANCE <2 7T; g c C I c 0 a) In ' -H H L L. I g 8 e E I 8 8 L CORRECTIVE r ACTIONS (If Necessary) ’7ZK' l l I : Deviation ——2uér \ Exists L --- - - -— - MEASUREMENT OP DEVELOPMENT PERFORMANCE Figure 5.1: A Model Of The Evaluation Process Of Development 247 appropriate to ennphasize that equal attention should also be given to fie amitoring annd evaluation activities of developnent performance . Perhaps, fie task of monitoring and evaluating developnent could be better accouplisl'ed if a specific public agency is assigned to unndertake finenI. Inter-regional disparities in developnent is only a facet of fie inequality problen in Malaysia. Since regions contain population and econouic activities, regional disparities fierefore also have social annd economic implications. Inequality as a developnent issue assures greateriaportancewhenitisrenenberedfinat socialanndecononic dualities in Malaysia also have efinnnic connotations. In view of fie above implications, it is essential finat a acre connprenensive approach he adopted by fie Malaysian auficrities to solve fie inequality problen as a national issue. It has been stated in fie Mid-Term Review of fie Fourfin Malaysia Plan (1984, para. 56, p.21) finat ‘planning on fie basis of regions' constitutes an important aneanns to articulate acre effective reduction of inter-regional inequalities of developnent in Malaysia. To facilitate finis, finree important considerations must be addressed. First, a definition of ‘region' is essential. Its existence will provideacommworkingguidelineandpurpose fcrficseegagedin fie plannnning, implenentation, coordination, annd evaluation of regional development. Second, existence of fie appropriate institutions to unndertake finis function at fie ‘regional' level is required. There should be adequate personnel. Third, in-service course on fieoretical 248 annd practical aspects of regional plannnning annd developnent pertinent to Malaysia must be establisted. Fulfillment of fiese considerations would not only enhance efforts to pronncte regional planning and developnent but also would facilitate prorction of gronfin wifin equity approach to development. Political Will To achieve a substantial solution to fie problem of regional inequalities in Malaysia will require a cocerted effort to truly upgrade annd accelerate fie developnent process of fie lagging regions. Inequalityissueswonldrenainmere academic rnetoric iffineyarenct seriously considered in fie actual fornmllation of deveth policy and planning. In finis respect, it would be difficult to dienniss fie importance of fie government's political will annd wisdom to stimulate annd induce fie realization of desirable deveth goals. As fie United Nations (1976, p.61) sees it: "Governments must have fie political will to evolve annd implement innncvative annd adequate urban anndrural . . . policies, [anndact] asacornerstoeoffieirefforts to improve fie quality of life in human settlenents." Tle importance of fie government's political will has also been enphasized by Waterston (1967, p.14), wrc states finat "sustained governmental connnnibnent is a sine qua non for developnent." Furfiernncre, "if fie belief of fie people in fie government is to be maintainned, [fien] the rnetoric must be backed by explicit actions" (Gore, 1984, p.249). Efforts by fie Malaysian auficrities to establish a national 249 urbanization policy and to implenent urbanization of traditional villages are steps in fie right direction to deal with problems of inter-regional inequalities . However, such innitiatives must be carefully deliberated annd planned so finat polarization reversal can effectively occur annd equitable deveth be haracnionsly acconplisled. T're cocept of urbanization in developnent must be clearly understood, so finat fie proposed urbanization policy annd strategies will not be based on narrow annd faulty interpretations. "Progressdependstoalarge extentuponfineadoptionbygovernments of appropriate administrative annd legislative actions" (Unnited Nations, 1951, p.17; also see Higgins, 1980; Chan, 1979). T‘Iere are at least finree major cocepts of urbannization that should be considered: fie behavioral, fie structural, annd fie dencgraphic (Lenpard, 1965). Tie first of fiese is cocerned wifin fie experiece of individuals over tinnne annd wifin patternsof behavior (Wirth, 1938); fie second is related to fie activities of the wncle population end is primarily cocerned wifin changes inn economic structure (Lannpard, 1965); fiefinirdisfiedencgraphiccoceptwherefieprocessisseen primarily as oe of population concentration (Lampard, 1965). Also, in fie proposed urbannization policy lagging regions must be given priority. Development of lagging regions must take place in fieir own right, withcognizanceanndunnderstandingoffieirneedsand contributions in fie national developnent process. Advancement of lagging regions should not be made to rely upon the trickle down process. Strategies which still give undue prominece to developed 250 regions will not only perpetuate fie dualistic pattern of spatial development, but also, finrongh polarization effects, hinder efforts to foster integrated annd haracnnions national developnent . CHAPTERSIX SLM’IARYANDMUSION Wifinin fie anefinodological annd data constraints, finis study has examined fie develoment performance of fie variols administrative districts in Peninnsular Malaysia. In doing this, fie magnitude annd fie distribution of development beefits has been establisled. Tie rennainnderoffinischapterpresentsfiesunmaryoffiefindingsand makessnggestionsforfutureresearch. Contextual Overview Tl'efactthatfieexistececfsomedegreeofinequaliwin developnent beefits is inevitable does not justify fie neglect in dealing wifin finis problem. In fact, inequality in deveth requires appropriate attention annd redress, since its continnued persistece, by polarizing effects, could adversely affect on—going annd future development efforts. While fie pursuance of perfect equality or even distribution of development beefits is unnrealistic, efforts to reduce fie intensity of inequality in development are, We, within practical reach. 251 252 Inmanyways, finisstudyhasbeenactivatedbyfieconcern expressed by Malaysian auficrities over fie problem of inter-regional inequalities in developnent in Malaysia. Tle auficrities recognized fie potential problems finat conld be geerated from fie persistece of inter-regional inequalities in development in Malaysia, especially as finey bear implications Lpon fie innitiatives to proncte, consolidate annd sustain positive progressfor overall national development. Alsofinisstudyhasbeenpronptedbyfieneedfcradequateand conpreheeive innforanation at a scale of annalysis appropriate for a acrecoterentunderstandingoffienatureandmagnitudeofof inter- regional inequalities in developnent in Malaysia. Much of current development information and planning is doe at fie state level, as evideced in fie official five-year plan documents annd Treasury annnual reports. Information on regional developnent for different geographical scales is only available for sonne specific regional studies, mainly doe by private consultants or specialised government agencies (such as fie Federal Town annd Comtry Plannnning Department, Ministry of Land and Regional Developnent, Regional Develcpnent Auficrities) . Information from fiese isolated regional development studies have limited applications because, being specific to certain geographical areas and performed unnder different terms of referece, fineymayoranaynctfit intofiebroadernational plan. T‘rese regional studies are often marked by pursuance of objectives finat are quite different from oe ancfier annd valid for different time periods. This study focussed attention on fie examination of spatial developnent annd spatial inequalities by conparing all regions at the 253 sametime. SummaryOf Findings Annalysis of 1970 annd 1980 data reveals finat fiere exists a wide spatial variation of development perfornnannce among fie varions states annd administrative distriCts in Peninsular Malaysia. The relative differece of composite indices of development inndicates finat fiere arenctonlygapsinnfieperformanceofdevelopnentaacngfiese regions, but also finat fie benefits of development were not well distributed. This is admist fie impressive aggregate rates of economic growth inn Malaysia between 1970 annd 1980; wifin a rate of grmfinof7.8percentperanmnmasconparedto6percentperannumin fie 1960's. Also, fie GDP per capita for Malaysia witnessed an increase from (Malaysian Ringgit) $1,172/= in 1971 to $1,836/= in 1980. Thisfinndingisfierefcreinconsonancewifinfindingsofofier studies which contend that fie distributional aspect of development in Malaysia has not matcled the innpressive perfornmance of ecocmic growth (Griffin annd Khan, 1979; Geog, 1979; Chan, 1979; Mahangas, 1982; Aris Ofimnan, 1984). By use of Gini coefficients annd Location Quotients for fie varions inndicators of developnent it is clearly sncwn that fieir geographical distribution is not proportionally uniform at fie state ncr at fie administrative district level. This study reveals that disparities increase as fie geographical unit of analysis decrease. Amog administrative districts, finis analysis indicates that fine 254 urbannandfiewesterndistrictsinPeninsularMalaysia, asawtcle, register conparatively better developnent performance than fie rural andfieeasterndistricts. Tnesefindingssupportfiehypofiesesfinat (1) urban districts have better development performance and less inequalityfinanruraldistricts, and(2) districtsinfiewestcoast states have better developnent perforanance and less inequality finan districtsinfieeastcoaststates. Anand's (1983) studyonincome inequalityinPenineularMalaysiahasalso fondthatrural areashad greaterproblenswithinconeinequalityfinanurbanareas. Infinding finat fie acre urbanized districts have higer levels of development, finisstudyisalsoinconsonancewithfindingsestablistedby Leinbach's (1972) and Osborn's (1974) studies. Conparisoe of composite indices of development for 1970 and l980revealfinatfiechangeindevelopnentperformancehasbeenemall as differeces recorded failed to be statistically significant. This verifies fie hypofietical contention finat fiere was no significant change in deveth in Peninsular Malaysia in 1980. Similarly, examination of ratios between fie Ginni coefficients of 1970 and 1980 stcwsfinatfiechangeinspatial inequalityhasbeenmnininnal. Infinis regard, finis finding is not able to support Kuznet's (1955) and Williamson's (1965) hypofietical assertions fie process of economic development will reduce the magnitude of inequalities over time. 255 gggestionsForFUrfierResearch This study focusses only on inter-district disparities in developnent. This represents only oe scale of analysis finat colld be amropriate. I-lrrfier innsiglnts and understanding of inequalities in developnent conldperhapscolldbeenhancedif futureresearchwould consider intra-district differeces, as it is quite possible that wifinin districts pockets of inequality may exist. T'le availability of fiese studies would considerably contribute to a much better identification of severe social and economic deviations. These coild enhance fie process of targetting appropriate measures for acne deprived grams. Thisstudyisncttobemisconstruedasdealingonlywifinspatial differentiation in fie beefits of developnent. While fie focus is clearly spatial, fie variables used and fie findings are social in nature. As log as people inhabit fie variols geographical regions, wecannotignorefiequestioncfwtcisgettingwhat. Thisstudyhas indirectly attenpted to annswer this question. Knowledge abont wtc gets what provides useful information for fie pursuance of development policy and planning towards national unity and social equity. T'te fact finat Malaysia's New Economic Policy has national unnity as its overriding goal suggests finat research finat investigates fine social aspects of fie distribution of development beefits is a contribution tofinisobjective. Futureresearchisreconnnendedtogobeyond spatial differentiation, towards the unnderstanding of social differentiation inn fie distribution of development benefits. It will 256 be useful to explore wrefier changes in spatial dispariw affect social differentiation, annog different efinnnic and socio-economic groups. T‘l‘eindexofdevelcpnenthasbeenusedasageeralyardstickof relative level of developnent. Being based on enpirical observations itissubjecttoanannyidicsynncraciesof local systems, andfieresults mnaybeaffectedbyfinem. Noattemptwasmadeinfinisstudytotake into accomt fie variations in attitudes, perceptions and priorities which exists from region to region wifinin Peninnsular Malaysia. But fie introduction of variable weights suggests finat fie inequality resultsmaynnotbealteredsubstantially. Whatmaybemcreimportant isfieuseofatechniquetorecognizefiecomplenentaritiesamogfie various developnent variables. It shouldbepointedolt finatncstoffievariablesusedinfinis study coild, in fact, be furfier diaggregated to provide acre details, but finen fiere is fie risk of "overloading" fie significance of sone. Future research might determining what fie appropriate mixture of geographical scale and variable breakdown is. Inn this study, fie administrative districts have been classified intowesternandeasterndistrictsaswellasintourbanandrural districts forfiepurposeofdeternnninningfieirperfcrnmancein developnent. T‘Ie categorization of western and eastern districts has beenusedtoconparetwogeographicareasfinatseemtobequite different from fie developnent point of view. Similarly, fie categorizationofurbananndruraldistrictshasbeenusedtoconpare 257 fieirdevelcpmentperformance. Ofiercontrastingareasmaybe conparedinfiefutureusingcriteriadifferentfromfieoesadopted here. Most of fie data for fie selected indicators of development used infinisstudywerederivedfrnomonlytwopointsintime,1970and 1980. Fromfinisstudyitisnctpossibletosayveryanuchabontfie variabilityoverfieperiod,fieupsanddownsofregional developnent. For finis, time-series data are required. Since developnentisadynamicprocessperhapsthisstudyconldshedmcre lightwhenvariablesaretakenoveraperiodoftimnerafierthanona singleyear. Tiereisalwaysfiedangerthatdataconldhavebeen prodlced from developnent activity that occurred only in finat partiollar year. Cocluding Note Regional development sl'culd nct aim only at increased welfare in aggregate terms, but also srculd strive towards acre more equitable distribution of developnent among areas and groups of people. Whatever fie present sl'crtfalls, fie ainn of regional and national developnentmustbetoensure finatoverareasonableperiodoftiae all regions in fie conntry sl'culd realize fieir potential for development and attain levels of living not far rencved from ficse of fie nation as a wl'cle. Cotesive and harmonious developnent can be effectively prorcted and sustained wlen regional disparities are within acceptable linmits. 258 Strog political commitment is fundamental for fie pursuance of an effective policy to acconplish growfin wifin equity (Myrdal, 1957; Hirscrmnann, 1958; Williamson, 1965; Waterston, 1967; Rlcda, 1982). Equally important, deveth plannners must equip themselves with fie requisite skills and have adequate information to respod effectively to fie dynamics of development and guide finem towards acre efficient execution of plannnning, implementation, coordination, and evaluation of development. In finis regard, it is befitting to quote fie message by Malaysia's ‘Fafier of Development', Tun Haji Abdul Razak (1973), wlc states: T'teconrseofrapiddevelopnentshouldnotonlybe steered by able and skillful managenent but also [should] be activated by clear and sonnd ideals distilled throgh extensive and intensive inquiries into fie deeper meaning of developnent and progress. Finnally, as resolution 2626 (XXV) of fie United Nations Geeral Assembly, passed on 24fin October, 1970, aptly states: Tie ultimate objective of development must be to bring a sustained improvenennt of fie well-being of fie individual and bestow beefits on all. If undue priveleges, extrenes of wealfin and social injustices persists, fienn developnent fails in fie essential purpose. LIST OF APPENDICES APPENDIX 1 259 APPENDIX 1: INDICATORS OF DEVELOPMENT -- OPERATIG‘IALIZATIGNI, SQIRCES OF DATA, AND MATION *a 1) Percent population 10 years and older wtc are literate. (Source: Department of Statistics, Malaysia) 2) PercentpopulationSyearsandolderwhohave/had fornmal sclcoling. *b (Soiree: Departnent of Statistics, Malaysia) *c 3) Number of teacl'ers per 1,000 population 5 years and older. (Source: Department of Statistics, Malaysia) *d 4) Number of infant deaths (under 1 year old) per 1,000 live birfins (by occurrence). (Solrce: Department of Statistics (1972) Vital Statistics, West Malaysia, 1970. Kuala Lumpur, Government Printer. 5) Numberofmaternnal deafinsduetopueperal causesperl,OOO live births (by occurrence). *e (Sonrce: Department of Statistics (1972) Vital Statistics, West Malaysia, 1970. Kuala Lunnpur, Government Printer. 6) Number of occupied housing units per 1,000 population. *f (Sonrce: Departnent of Statistics, Malaysia) 7) Percent occupied housing unnits having flush and polr-flush toilets. *g (Sonrce: Departnnennt of Statistics, Malaysia) 8) Percent occupied housing unnits having piped-water supply. *h (Sonrce: Department of Statistics, Malaysia) 9) Percent occupied housing units having public-supply e1ectricity.*i (Soirce: Departnent of Statistics, Malaysia) 10) Number of public law and order enforcenent personnel per 10,000 population. 11) Number of fire (protection) service personnel per 10,000 population. (Sonrce: Fire Service Departnent, Malaysia) 12) Road mileage (in miles) per 10 square miles. *j (Sonrce: Ministry of Public Works, Malaysia & Economic Planning Unit, Prime Minister's Departnent, Malaysia) 13) Miner of postal service personnel per 10,000 population. *k (Sonrce: Postal Services Department, Malaysia) 14) Percent labor force 10 years and older in professional and technical group occupations. *1 (Sonrce: Department of Statistics, Malaysia) 15) Percent labor force 10 years and older in administrative and managerial group occupations. *m (Soiree: Department of Statistics, Malaysia) 16) Number of government legislative, administrative and executive officials per 10,000 population. *n (Soiree: Departnnent of Statistics, Malaysia) 17) Percent of labor force 10 years and older in manufacturing sector. *0 and *p (Department of Statistics, Malaysia) 260 APPENDIX 1 (cont'd.). 18) 19) 20) 21) 22) 23) 24) 25) 26) 27) 28) 29) 30) Percent labor force 10 years and older in commercial sector. *q (Solrce: Department of Statistics, Malaysia) Gross domestic product per capita. *r (Sonrce: Departnennt of Statistics, Malaysia & Five-Year Plan documents) MmberofpublictelephoeboofinsinruralareasperlOfiOO rural population. *3 (Solrce: Telecoms Department, Malaysia & Department of Statistics, Malaysia) Agricultural extension stations acreage per 1,000 labor force in agriculture. *t (Sonrce: Deaprtment of Agriculture, Malaysia & Department of Statistics, Malaysia) Percent population in urban places. *u (Sonrce: Department of Statistics, Malaysia & Federal Town and Conntry Planning Departnnent) Population of places 20,000 and more as percent of total urban population. *v (Sonrce: Department of Statistics, Malaysia) Peroent female population 10 years and older who are literate. (Source: Departnent of Statistics, Malaysia) Peroent fennale labor force 10 years and older in non-agricultural economic activities. *w (Soiree: Department of Statistics, Malaysia) Peroent labor force 10 years and older receiving mean acnfinly incone earnings acre finan Malaysian RS300/=. *x (Source: Department of Statistics & Anand, S. (1983). Inequality and Poverty in Malaysia. New York, Oxford University Press) Number of p.iblic low-cost reusing units per 1,000 labor force receiving mean acnfinly incone earnnings less than Malaysia R$300/=. *y (Source: State Secretariats (Housing Section) in Peninsular Malaysia; Anand, S. (1983). op.cit., Department of Statistics, Malaysia) Malay labor force 10 years and older in agricultural vocations as percent of total Malay labor force. (Sonrce: Department of Statistics, Malaysia) Malay labor force 10 years and older as percent of labor force in manufacturing sector. (Source: Departnnent of Statistics, Malaysia) Malay labor force 10 years and older as percent of labor force in commercial sector. (Source: Department of Statistics, Malaysia) Notes: *a) Refers to ficse able to read and write a simple letter in any language. Personswnccolldonlyreadandwriteafewwordsare not considered to be literate but were categorised as illiterate. 261 APPENDIX 1 (cont'd. ). *b) *c) *d) *f) *9) *h) *1) *J ) *k) *1) *m) *n) *0) *p) *q) Doesnnotincludennurseryandkindergarten, adulteducationand religions education. Tne percentage wifin formal schooling is derived from fie percentage of population with no sclcoling (obtained from fie Departnent of Statistics, Malaysia). By substracting fie percentage of population with no sclcoling by 100 percent, fie percentage of population wifin sdcoling is obtained. Basedonficselistedas ‘teacters' underfieocopationalgroup tables for unpublisled districts files. and *e) ‘By residece' data are preferred. Such data are not available for 1970. Thus, conparison wifin 1980 ‘by residece' data is not possible. I-kosing units are classified as occupied if fiere were people livingand/orsleepinginfiehousingunits atfietimeoffie Census. Among fie residential sannitation system, flush and pour-flush toilets could be considered as acdern system. Includes all piped-water, regardless wtefier it is treated or from public-SLpply or private tank. Refers to electricity supplied by fie National Electricity Board, local auficrities and/or ofier electricity conpanies. For major, paved roads. Data prepared by fie Departnent of Public Works, Malaysia. Includes only uniformed postman and fie clerical staff. Excluded are ofier ancillary staff, such as typists, stecgraplers, office boys, ad drivers. and*m) Tleoccupationofapersonreferstofiekindofwork performed by fie person at his/her place of enployaent. Occupatioe have been classified according to fie Dictionary of Occupational Classification, 1980 p.1le by fie Ministry of LabourandManpower (which isbasedonfie International Standard Classification of Occupations, 1968 prepared by fie ILO). ‘Government officials' nere include only (1) legislative officials and government administrators, and (2) government executive officials. Thisreferstopersonsageleandolderwtcduringfiereference week were eifier enployed or unnenployed. Manufacturing industries are based on classification according to fie Malaysian Industrial Classification, 1972 (updated) (which is based on fie U.N. International Standard Industrial Classification Of All Economic Activities, 1968). Includes large and small mnannufacturing activities. ‘Connnneroe' l‘ere includes (1) wl'clesale trade, (2) retail trade, (3) banks and ofier financial institutions, (4) insurance, (5) real estate, and (6) total and restaurant. Alficugh in 1970 ‘business services' were tabulated under ‘Services', breakdown for all districts, except Kuala Lumpur, were obtained. State (DPS are available from Five-Year Plan documents. However, GDPs for districts are not available. So estimates are derived based on fie following formula: labor force in relevant 262 APPENDIX 1 (cont'd. ). *r) *S) *t) *u) *V) industries in fie district divided by similar labor force in fie state and fien multiply by fie state GDP. Based on records obtained from fie Telecoms Department, Malaysia. Sole adjustments are made to fie data provided, especially wten it is not indicated in which district fie rural teleplcne boofin is located. T‘le total nnunmber of rural telephone booths for 1970 and 1980 are based on cumulative figures from 1959 till end of 1970 for fie 1970 and from 1971 till end of 1980 for 1980 data. Total for districts and states may not tally because for ficse telephone boothswhichcoildnctbelocatedinfiedistrictsfieywerenct accountedforinfietotalunderdistricts; butfineyare considered in fie total for fie state. Based mainly on I.F.T. Wog's book on land use in Peninnsular Malaysia. For 1970, used fie 1974 statistics for size of agricultural station(s)' acreage; since no data available for 1970. For 1980, used fie estimates made by the Departrent of Agriculture (Soil Scie'ce Section). Some adjustments are made to fit fie data wifin fie canparative districts used. Wog's book indicate that in 1974 Perak Tegah had 146 acres of agricultural stations. Since in 1970 Perak Tegah did not exist yet, finis 146 acres were placed under fie district of Kuala Kangsar; since it is fond finat fiese stations were largely found at Titi Gantung (near Parit). The 525 acres listed in 1974 under Petaling are placed under fie district of Ulu Langat in 1970; since acst of fiese 525 acres are fond in Serdang. According to Wog's book, small agricultural stations are usually lessfinan2acres; servingasoutpostsandservicingcenters for chanelling of technnical publications and subsidies to farmers in fie surrounding areas. On fie ofier hand, large agricultural stations have wider activities which include: dencnstration plots, field verification trials, research into newly introduced crops, production of planting anaterials and stock seeds, multiplication of stock materials, hybridization and large scale production. Urban areas refer to all gazetted areas wifin a population 10,000 or acre at fie time of fie Census. Places wifin population 20,000 and more is used tere as an indicator of urban development for fie following reasons: (1) fieoretically, 20,000 and acre population is considered a minimum dencgraphic finresl'cld for efficient existece and delivery of basic urban services and facilities, and (2) most of fie active large towns inn Peninsular Malaysia have a population at least 20,000 and above. Include all ficse activities finat are net listed under ‘Agriculture, Forestry, and Fishing. 'Tnese would then include fine following sectors: ‘Mining, Manufacturing, Construction, Utilities, Transportation, Storage and Communications, Wtclesale and Retail, Hotel and Restaurant, Finance, Banking, Insurance, Real Estate, Business, Government Services, and Other Services'. 263 APPENDIX 1 (cont'd. ). *w)UsefiebenclmnarkofMalaysianR$300/=andlessperaensenasfie *x) *Y) poveriyline (basedonreportsbyfieSocio-Economic ResearchUnit, Prime Minister's Department, 1976/77 Agricultural Census, and Five- Year Plan documents). Using Anand's (1983) Table 6-11, which shows fiemeaninconeforfievarionsoccupationsinMalaysia (bytwo- digit occupational category), fie mean incone of fie labor force fcrfiedistrictwasconputed. Thenfienumberof laborforcewith aeaninconelessorabovefiepovertylinewasderived. Tl'ese are public low-cost rousing built by fie public agencies -- state governments and fieir statutory bodies. The figures for 1970 and 1980 are based on cumulative statistics; for 1970 all low-cost unnits before Dec.,197O and for 1980 from 1971 till end of 1980. 1970 figures for Manjung/Kuala Kangsar/Perak Tegah has been apportioed according to fieir percentage share of low-cost housing. T'l'ennumberoflaborforcebelowpovertylinearederivedby estimatesasin*w. APPENDIX 2 264 APPENDIX 2(A): STATES IN PENINSULAR MALAYSIA, 1970 AND 1980 States in 1970 States in 1980 1) Johore l) Jolcre 2) Kedah 2) Kedah 3) Kelantan 3) Kelantan 4) Malacca 4) Malacca (Melaka) (Melaka) 5) Negri Senbilan 5) Negri Sembilan 6) Pahang 6) Pahang 7’ fin“: pm) 7’ m pm) 8) Perak 8) Perak 9) Perlis 9) Perlis 10) Selangor lO) Selangor ll) Treggannu 11) Treggannu 12) Federal Territory of Kuala Lumpur (Wilayah Persekutuann Kuala Luapur) * Note: Tre Federal Territory of Kuala Lunpur is treated as a state in finis study. 265 APPENDIX 2(8): AREA OF STATES IN PENINSULAR MALAYSIA, 1970 AND 1980 States in 1970 Area States inn 1980 Area (sq- m1) (sq. m1) 1) JChore 7330 1) JOhore 7330 2) Kedah. 3639 2) Kedah, 3639 3) Kelantan 5765 3) Kelantan. 5765 4) Malacca 640 4) Malacca 640 (Melaka) (Nelaka) 5) Negri Sembilan. 2566 5) Negri Sembilan 2566 6) Pahang 13884 6) Pahang 13884 7) Penang 398 7) Penang 398 (Pulau Pinang) (Pulau Pinang) 8) Perak 8110 8) Perak 8110 9) Perlis 307 9) Perlis 307 10) Selangor 3166 10) Selangor 3072 ll) Terengganu 5002 11) Terengganu 5002 12) Federal Territory 94 of Kuala Lumpur (Wilayah.Persekutuan Kuala Lumpur) * the: The Federal Territory of Kuala Lumpur is treated as a state in this study. 266 APPENDIX 2(C): POPULATION OF STATES IN PENINSULAR MALAYSIA, 1970 AND 1980 States in 1970 Population States in 1980 Population 1) Jotcre 1271794 1) Joncre 1580423 2) Kedah 952421 2) Kedah 1077815 3) Kelantan 684312 3) Kelantan 859270 4) Malacca 403061 4) Malacca 446769 (Melaka) (Melaka) 5) Negri Sennbilan 480053 5) Negri Sembilan 551442 6) Pahang 503030 6) Pahang 768801 7) Penang 773327 7) Penang 900772 (Pulau Pinang) (Pulau Pinang) 8) Perak 1561184 8) Perak 1743655 9) Perlis 120996 9) Perlis 144782 10) Selangor 1625625 10) Selangor 1426250 11) Treggannu 404924 11) Treggannu 525255 \ 12) Federal Territory 919610 of .Kuala Lunpur (Wilayah Persekutuan Kuala Lunpur) * Note: Tle Federal Territory of Kuala Lumpur is treated asastateinfinisstudy. APPENDIX 3 267 APPENDIX 3: STATES IN PENINSULAR MALAYSIA, 1970 AND 1980 (FOR (INPARATIVE USE) Conparative States in 1970 Conparative States in 1980 1) Johore 1) Jotcre 2) Kedah 2) Kedah 3) Kelantan 3) Kelantan 4) Malacca 4) Malacca (Melaka) (Melaka) 5) Negri Sembilan 5) Negri Sembilan 6) Palnang 6) Pahang 7) Penang 7) Penang (Pulau Pinang) (Pulau Pinang) 8) Perak 8) Perak 9) Perlis 9) Perlis 10) Selangor 10) Selangor and fie Federal Territory of Kuala Lunpur 11) Treggannu (Wilayah Persekutuan Kuala Lumpur) * 11) Treggannu Note: Tre Federal Territory of Kuala Lunpur is combined wifin fie state of Selangor to facilitate conparison wifin fie sane area in 1970. In 1970, fie Federal TerritoryofKualaLunpurwaspartoffiedistrict of Kuala Lunpur in fie state of Selangor. APPENDIX 4 268 APPENDIX 4 (A): ADVIINISTRATIVE DISTRICTS IN PENINSULAR MALAYSIA, 1970 AND 980 Districts in 1970 District in 1980 JGIDRE STATE JOHORE STATE 1) Batu Pahat l) Batu Pahat 2) Jolcr Bahru 2) Jotcr Bahru 3) Kluang 3) Kluang 4) Kota Tinggi 4) Kota Tinggi 5) Mersing 5) Mersing 6) War 6) Muar 7) Pontian 7) Pontian 8) Segamat 8) Segamat KEDAH STATE KEDAH STATE 9) Baling 9) Baling 10) Bandar Bahru 10) Bandar Bahru 11) Kota Setar 11) Kota Setar 12) Kuala Muda 12) Kuala Muda 13) Kubang Pasu 13) Kubang Pasu 14) Kulim l4) Kulim 15) Largkawi 15) Langkawi 16) Padang Terap 16) Padang Terap 17) Sik 17) Sik 18) Yan 18) Yan 19) Pedang KELANTAN STATE KELAN'I'AN STATE 19) Bachok 20) Baclnok 20) Kota Bahru 21) Kota Bahru 21) Machang 22) Machang 22) Pasir Mas 23) Pasir Mas 23) Pasir Puteh 24) Pasir Puteh 24) Tanah Merah 25) Tannah Merah 25) Tunpat 26) Tunpat 26) Ulu Kelantan 27) Ulu Kelantan 28) Kuala Krai MALALm STATE MALACI‘A STATE 27 ) Melaka Utara 29) Melaka Utara 28) Melaka Selatan 30) Melaka Selatan 29) Melaka Tegah 31) Melaka Tegah NEERI SEMBILAN STATE NEERI SEMBILAN STATE 30) Jelebu 32) Jelebu 31) Kuala Pilah 33) Kuala Pilah 32) Port Dickson 34) Port Dickson 33) Rembau 35) Rembau APPENDIX 4(A) (cont'd. ). Districts in 1970 Districts in 1980 34) Serenban 35) Tanpin PAHAI‘G STATE 36 ) Bentog 37) Cameron Highland 38 ) Jerantut 39 ) Kuantan 40) Lipis 41 ) Pekan 42 ) Raub 43 ) Tenerloh PENAING STATE 44) Seberang Perai Utara 45) Seberang Perai Tegah 46) Seberang Perai Selatan 47) Pulau Pinnang Timur Laut 48) Pulau Pinang Barat Daya PERAK STATE 49) Batang Padang 50) Dindings 51) Kinta 52) Kriann 53) Kuala Kangsar 54) Larut & Matang 55) Hilir Perak 56) Ulu Perak PERLIS STATE 57) Perlis SELAme STATE 58) Klang 59) Kuala Langat 60) Kuala Lunpur 61) Kuala Selangor 62) Sabak Bernnam 63) Ulu Langat 64) Ulu Selangor 36) Seremban 37) Tanpin PAT-1AM? STATE 39) 40) 41) 42) 47) Bentog Cameron Highland Jerantut Kuantan Lipis Pekan Raub Tenerloh Ronpin PENAM} STATE 48) 49) 50) 51) 52) Seberag Perai Utara Seberang Perai Tegah Seberang Perai Selatan Pulau Pinang Timur Laut Pulau Pinnang Barat Daya PERAK STATE 53) 54) 55) 56) 57) 58) 59) 60) 61) Batag Padang Manjung Kinta Kriann Kuala Kangsar Larut & Matang Hilir Perak Ulu Perak Perak Tegah PERLIS STATE 62) Perlis SELAMDR STATE 63) 64) 65) 66) 67) 68) 69) 70) 71) Gonbak Klang Kuala Langat Kuala Selangor Petaling Sabak Bernam Sepang Ulu Langat Ulu Selangor 270 APPENDIX 4 (Cant'd. ). TEREMBANU STATE W STATE 65) Besut 72) Besut 55) Dugun 73) Dungunn 67) Kennaman 74) Kemamnan 68) Kuala Teregganu 75) Kuala Tereggannu 69) Marang 76) Marang 70) Ulu Tereggannu 77) Ulu Terenggannu 271 APPENDIX 4 (B): AREA OF ADVIINISTRATIVE DISTRICTS IN PENII‘BULAR MALAYSIA, 1970 AND 1980 Area Area Districts in 1970 (sq.ml) District in 1980 (sq.ml) JOE-DRE STATE JOPDRE STATE 1) Batu Pahat 725 l) Batu Pahat 725 2) Joker Bahru 702 2) Jolcr Bahru 702 3) Huang 1101 3) Kluang 1101 4) Kota Tinggi 1345 4) Kota Tinggi 1345 5) Mersing 1095 5) Mersing 1095 6) Muar 906 6) Muar 906 7) Pontian 355 7) Pontian 355 8) Segamat 1101 8) Segamat 1101 KEDAH STATE KEDAH STATE 9) Baling 586 9) Baling 586 10) Bandar Bahru 104 10) Bandar Bahru 104 ll) Kota Setar 499 11) Kota Setar 256 12) Kuala Mlda 356 12) Kuala Muda 356 13) Kubang Pasu 366 13) Kubang Pasu 366 14) Kulim 296 14) Kulim 296 15) Langkawi 180 15) Largkawi 180 16) Padang Terap 524 16) Padang Terap 524 17) Sik 635 17) Sik 635 18) Yan 93 18) Yan 93 19) Pedang 243 KELANTAN STATE KELANTAN STATE 19) Bachok 102 20) Bachok 102 20) Kota Bahru 158 21) Kota Bahru 158 21) Machang 210 22) Machang 210 22) Pasir Mas 223 23) Pasir Mas 223 23) Pasir Puteh 168 24) Pasir Puteh 168 24) Tanah Merah 574 25) Tanah Merah 574 25) Tunpat 65 26) Tunpat 65 26) Ulu Kelantan 4265 27) Ulu Kelantan 3129 28) Kuala Krai 1136 mm STATE W STATE 27) Melaka Utara 256 29) Melaka Utara 256 28) Melaka Selatan 269 30) Melaka Selatan 269 29) Melaka Tegah 115 31) Melaka Tegah 115 NERI SEVIBILAN STATE NEGRI SEMBILAN STATE 30) Jelebu 528 32) Jelebu 528 31) Kuala Pilah 956 33) Kuala Pilah 421 32) Port Dickson 216 34) Port Dickson 216 33) Rerbau 160 35) Rembau 160 272 APPENDIX 4(B) (cont'd.). Area Area Districts in 1970 (sq.ml) Districts in 1980 (sq.ml) 34) Serenban 367 36) Seremban 367 35) Tanpin 339 37) Tanpin 339 PARADE STATE PARADE STATE 36) Bentog 707 39) Bentog 707 37) Cameron Highland 275 40) Cameron Highland 275 38) Jerantut 2918 41) Jerantut 2918 39) Kuantan 1143 42) Kuanntan 1143 40) Lipis 2007 43) Lipis 2007 41) Pekann 3721 44) Pekan 1714 42) Raub 876 45) Raub 876 43) Tenerlch 2237 46) Tenerloh 2237 47) Ronpin 2007 PENAM; STATE PENAMS STATE 44) S.Perai Utara 101 48) S.Perai Utara 101 45) S.Perai Tegah 91 49) S.Perai Tegah 91 46) S.Perai Selatan 93 50) S.Perai Selatan 93 47) P.Pinang Timur Laut 46 51) P.Pinnang Timur Laut 46 48) P.Pinang Barat Daya 67 52) P.Pinang Barat Daya 67 PERAK STATE PERAK STATE 49) Batang Padang 1047 53) Batang Padang 1047 50) Dindings 513 54) Manjung 451 51) Kinta 756 55) Kinta 756 52) Krian 370 56) Krian 370 53) Kuala Kangsar 1194 57) Kuala Kangsar 981 54) Larut & Matang 809 58) Larut & Matang 809 55) Hilir Perak 887 59) Hilir Perak 667 56) Ulu Perak 2534 60) Ulu Perak 2534 61) Perak Tegah 495 PERLIS STATE PERLIS STATE 57) Perlis 307 62) Perlis 307 SELAINUOR STATE SELAMDR STATE 58) Klang 391 63) Gombak 252 59) Kuala Langat 385 64) Klang 243 60) Kuala Lumpur 355 65) Kuala Langat 324 61) Kuala Selangor 570 66) Kuala Selangor 456 62) Sabak Bernnam 275 67) Petaling 188 63) Ulu Langat 417 68) Sabak Bernnam 386 64) Ulu Selangor 776 69) Sepang 239 70) Ulu Langat 310 71) Ulu Selangor 674 TEREMISANU STATE TERENIEANU STATE 65) Besut 764 72) Besut 764 66) Dungun 1056 73) Dungun 1056 APPENDIX 4 (B) (cont'd.). 67) Kemaman 979 68) Kuala Terengganu 506 69) Marang 201 70) Ulu Terengganu 1496 273 74) Kemaman 75) Kuala Terengganu 76) Marang 77) Ulu Terengganu FEDERAL TERRITORY 78) Kuala Lunpur 979 506 201 1496 94 APPENDIX 4 (C): POPUIATICN OF ADVIINISTRATIVE DISTRICTS IN 274 PENINSULAR MALAYSIA, 1970 AND 1980 Districts in 1970 Population District in 1980 Population JG-DRE STATE (JO-DRE STATE 1) Batu Pahat 249596 1) Batu Pahat 274625 2) Johor Bahru 267913 2) Johor Bahru 406871 3) Kluang 134510 3) Kluang 179791 4) Kota Tinggi 60942 4) Kota Tinggi 114267 5) Mersing 34644 5) Mersing 42208 6) mar 278903 6) Muar 291129 7) Pontian 117686 7) Pmtian 121031 8) Segamat 127600 8) Segamat 150501 KEDAH STATE KEDAH STATE 9) Baling 104142 9) Baling 104858 10) Bandar Bahru 33257 10) Bandar Bahru 31724 11) Kota Setar 301213 11) Kota Setar 279567 12) Kuala Mxia 159831 12) Kuala Muda 192308 13) Kubang Pasu 117451 13) Kubang Pasu 129808 14) Kulim 88346 14) Kulim 92525 15) Laugkawi 23788 15) Langkawi 28340 16) Padang Terap 30363 16) Padang Terap 40428 17) Sik 39027 17) Sik 43366 18) Yan 55003 18) Yan 59030 19) Pendang 75861 KELAN'I'AN STATE KELANTAN STATE 19) Bachok 62162 20) Bachok 73953 20) Kota Bahru 207394 21) Kata Bahru 275986 21) Machang 51570 22) Macharg 58040 22) Pasir Mas 100604 23) Pasir Mas 118153 23) Pasir Puteh 70155 24) Pasir Puteh 80959 24) Tanah Merah 57591 25) Tanah Merah 78097 25) Tunpat 73056 26) Tunpat 85952 26) Ulu Kelantan 61780 27) Ulu Kelantan 18578 28) Kuala Krai 69552 MALAEITA STATE MALACDA STATE 27) Melaka Utara 114550 29) Melaka Utara 113083 28) Melaka Selatan 84816 29) Melaka Tengah 203695 30) Melaka Selatan 87523 NEERI SEMBILAN STATE 31) Melaka Tengah 246163 NmRI SEMBILAN STATE 30) Jelebu 32897 32) Jelebu 36730 31) Kuala Pilah 115240 33) Kuala Pilah 67296 32) Port Dickscn 76032 34) Port Dickson 83561 33) Ranbau 39842 35) Rembau 36350 275 APPENDIX 4(C) (cont'd.). Districts in 1970 Population Districts in 1980 Population 34) Seranban 168175 36) Seremban 202790 35) Tatpin 47867 37) Tanpin 57507 38) Jempol 67208 PAHAM; STATE PAHAI‘G STATE 36) Bentog 56744 39) Bentcng 72865 37) C. Highland 15365 40) C. Highland 21502 38) Jerantut 36739 41) Jerantut 67875 39) Kuantan 96327 42) Kuantan 170573 40) Lipis 50645 43) Lipis 56996 41) Pekan 70345 44) Pekan 62246 42) Raub 57300 45) Raub 64414 43) Tanarloh 119565 46) Temerloh 213355 47) Ronpin 38975 ' PENAM; STATE PENAN} STATE 44) S.Perai Utara 161524 48) S.Perai Utara 199449 45) S.Perai Tengah 117475 49) S.Perai Tengah 161975 46) S.Perai Selatan 63626 50) S.Perai Selatan 71558 47) P.P.Tinur Laut 369991 51) P.P.Tinur Laut 391400 48) P.P.Barat Daya 60711 52) P.P.Barat Daya 76390 PERAK STATE PERAK STATE 49) Batang Padang 121121 53) Batang Padang 136473 50) Dindings 127861 54) Manjung 143610 51) Kinta 477758 55) Kinta 564500 52) Krian 153692 56) Krian 155765 53) Kuala Kangsar 188601 57) Kuala Kangsar 146292 54) Larut & Matang 220169 58) Larut & Matang 249550 55) Hilir Perak 210843 59) Hilir Perak 203028 56) Ulu Perak 61139 60) Ulu Perak 71372 61) Perak Tengah 73065 PERLIS STATE PERLIS STATE 57) Perlis 120996 62) Perlis 144782 SEIAMIDR STATE SWR STATE 58) Klang 233524 63) Ganbak 166059 59) Kuala Langat 107121 64) Klang 279349 60) Kuala Lunpur 871909 65) Kuala Langat 101578 61) Kuala Selangor 135646 66) Kuala Selangor 110366 62) Sabak Bernam 77968 67) Petaling 360056 63) Ulu Langat 95865 68) Sabak Bernam 103261 64) Ulu Selangor 103592 69) Sepang 46025 70) Ulu Langat 177877 71) Ulu Selangor 81679 TERENISANU STATE TEREJNGGANU STATE 65) Besut 79197 72) Besut 100830 66) Dungun 54347 73) Dungun 58360 67) Kenanan 44724 74) Kemaman 64899 APPENDIX 4 (C) (Cmt'd.). 68) Kuala Terenggarm 69) Marang 70) Ulu Termgganu 173304 19692 33660 276 75) K.Terengganu 232730 76) Marang 24977 77) Ulu Terengganu 43459 FEDERAL TERRITORY 78) Kuala Lumpur 919610 APPENDIX 5 277 APPENDIX 5: AININISTRATIVE DISTRICTS IN PENINSULAR MALAYSIA, 1970 AND 1980 -- FUR CD’IPARATIVE USE (Imparative Districts in 1970 Conparative Districts in 1980 JGDRE STATE 1) Bat; Pahat 2) Jahar Bahru 3) Kluang 4) Kata Tinggi 5) Mersing 6) Muar 7) Pmtian 8) Segamat KEDAH STATE 9) Baling 10) Bandar Bahru 11) Kata Setar 12) Kuala Mlda 13) Kubang Pasu 14) Kulim 15) Langkawi 16) Padang Terap 17) Sik 18) Yan KEIANI'AN STATE 19) Badm 20) Kata Bahru 21) Maduang 22) Pasir Mas 23) Pasir Puteh 24) Tanah Merah 25) Tunpat 26) Ulu Kelantan mm STATE 27) Melaka Utara 28) Melaka Selatan 29) Melaka Tengah NEERI SEIVBILAN STATE 30) Jelebu 31) Kuala Pilah 32) Part Dickson 33) Rembau 34) Serenban 35) Tampin PARADE STATE 36) Bentawg 37) Cameron Highland 38) Jerantut 39) Kuantan JUI-DRE STATE 1) Batu Pahat 2) Johar Bahru 3) Kluang 4) Kata Tinggi 5) Naming 6) Muar 7) Pontian 8) Seganat KEDAH STATE 9) Baling 10) Bandar Bahru 11) Kata Setar/Padang * 12) Kuala Muda 13) Kubang Pasu 14) Kulim 15) Largkawi 16) Padang Terap 17) Sik 18) Yan KELAN'I'AN STATE 19) Bachak 20) Kata Bahru 21) Macharg 22) Pasir Mas 23) Pasir Puteh 24) Tanah Merah 25) Tumpat 26) Ulu Kelantan/Kuala Krai ** MALAcx‘A STATE 27) Melaka Utara 28) Melaka Selatan 29) Melaka Tengah NEERI SEMBILAN STATE 30) Jelebu 31) Kuala Pilah/Jenpol *** 32) Part Dickson 33) Rembau 34) Seremban 35) Tampin PAHAm STATE 36) Bentang 37) Cameron Highland 38) Jerantut 39) Kuantan APPENDIX 5 (cont'd.). 278 Oanparative Districts in 1970 Canparative Districts in 1980 40) Lipis 41) Pekan 42) Raub 43) Temerlah PENAAB STATE 46) Seberang Perai Selatan 47) Pulau Pinang Tinur Laut 48) Pulau Pinang Barat Daya PERAK STATE 49) Batang Padang 50) Dindings/Kuala Kangsar/ Hilir Perak *+ 51) Kinta 52) Krian 53) Larut & Matang 54) Ulu Perak PERLIS STATE 55) Perlis SELAMIJR STATE 56) Kuala Selangar/ Sabak Bernan ++* 57) Kuala Langat/Ulu Langat **+ 58) Klang/Ulu Selangar/ Kuala Lumpur @ TREbmAW STATE 59) Besut 60) Dungun 61) Kemaman 62) Kuala Trengganu 63) Marang 64) Ulu Trengganu 40) Lipis 41) Pekan/Ronpin **** 42) Raub 43) Tanarlah PENAM; STATE 44) Seberang Perai Utara 45) Seberang Perai Tengah 46) Seberang Perai Selatan 47) Pulau Pinang Timur Laut 48) Pulau Pinang Barat Daya PERAK STATE 49) Batang Padang 50) Manjung/Kuala Kangsar/ Hilir Perak/Perak Tengah +* 51) Kinta 52) Krian 53) Larut & Matang 54) Ulu Perak PERLIS STATE 55) Perlis SELAMEOR STATE 56) Kuala Selangar/ Sabak Bernam ++* 57) Kuala Langat/Ulu Langat/ Sepam +++* 58) Ganbak/Klang/Petaling/Ulu Selangar/ F.T. of K.Lumpur # TREAGGANU STATE 59) Besut 60) Dungun 61) Kemaman 62) Kuala Trengganu 63) Marang 64) Ulu Trengganu Note° * For 1980, the districts of KOTA SEPAR and PENDAm (in Kedah State) have been canbined as one district to facilitate comparison between these two districts in 1970 and 1980. PENDAAG is wholly created from part of KOTA SETAR. **Far1980, thedistrictsanLUKELANI‘ANandKUALAKRAIhavebeen canbined as one district to facilitate conparison of these two districts in 1970 and 1980. KLMA KRAI is wholly created frcm part of ULU KELANI‘AN. 279 APPENDIX 5 (cont'd. ). *** For1980, thedistrictsofKUAIAPIIAH andJEMPOLhavebeen ocmbined as one district to facilitate ccmpariscm of these two districts in 1970 and 1980. JEMPOL is wholly created frun part of KLRLA PILAH. **** For 1980, the districts of PEKAN and RQ/IPIN have been oanbired as one district to facilitate conparison of these two districts in 1970 and 1980. RCMPIN is wholly created from part of PEKAN. APPENDIX 5(oont) *+ For 1970, the districts of DINDIABS, KUALA KADGSAR and PERAK HILIR havebeenombinedasaxedistricttofacilitateoarparisauofthese districts for 1970 and 1980. Such combination is necessitated by the fact that after 1970 PERAK TEMEAH, a new adninistrative district, was createdfrompar'tsofthesetheedistricts. Alsothis oanbinationis necessary because of the difficulty of determining the distribution of 1970 data for PERAK TEN3AH. +* For 1980, the districts of MANJUM; (formerly DINDII‘GS), KLmLA KAMSSAR, PERAKHILIRandPERAKTEAGAHhavebeenomlbinedasone district to facilitate conparison of similar areas between 1970 and 1980. Explanation given for combination of districts of DINDINGS, KLRLA KAbGSAR and PERAK HILIR FOR 1970 is relevant (refer). ++* For 1970 and 1980, the districts of mm SELAm‘OR and SABAK BERMMhavebeaaombinedasmedistr-icttofacilitateomparismof theseareasbetweml970and1980. Themainreasmforoanbiningthese two districts together instead of other districts is because SABAK BERNAM gained additional area fron KUALA SELAMER following the restructuring of district boundaries in SELANGOR STATE afetr the FEDERAL TERRITORY 0F KUALA LUVIPUR came into being in 1971. **+ For 1970, the districts of KUALA LANGAT and ULU LAME-AT have been ocmbined as one district to facilitate comparison of these areas between 1970 and 1980. This has been necessitated by the fact that the newdistrictof SEPAM-Ehasbeencreatedfrompartsofthesetwo districts. +++* For 1980, the districts of KUALA LAMEAT, ULU LANGAT and SEPAM; have been combined as one district to facilitate conparison of these areas between 1970 and 1980. Explanantions for oanbination of the districts of W LAMSAT and ULU LANGAT in 1970 is relevant (refer). @ For 1970, the districts of W, ULU SELAMIDR and KUALA LLMPUR have been oanbined as one district to facilitate comparison of these areas between 1970 and 1980. This decision has been made because after 1971 severalchangesoccurredtotheseareasthatthereforerequirethis adjustment —- (1) the district of KUALA LUMPUR was abolished, (2) the new district of PETALIM; has been created from parts of MAN; and the former district of KUALA LUMPUR, (3) the new district of GGVIBAK has 280 APPENDIX 5 (cont'd.). beencreatedfranpartsofULUSEIANGORandthefomerdistrictof KUAIALIMPUR, (4)thenewFEDERALTERRI'IORYofKUALALwIPURhasbeen createdfrcmpartofthe former district ofKIEIALLMPUR, i.e. fron theMJI-zimofmalammmr. # For 1980, the districts of GOVIBAK, KLAm, PEI'ALIm, ULU SELAMEOR and theFEDERALTERRITORYOFIQIALALLMPURhavebeenccnbinedasale district to facilitate conparison of these areas between 1970 and 1980. Explanations for the oanbination of the districts of KLAm, ULU SW and KUALA LIMPUR in 1970 above is relevant (refer). APPENDIX 6 281 VALUEJUDGEMENTSURVEY Please fill in the following information: NADEIASTPIACEOFD’IPIOYI‘ENI‘INMALAYSIA: ....................... TITLE/DESIGNIATICN HELD: ....... . .................................. TWIN/CITY OF RESIDENCE IN MALAYSIA: ........... . .................. Please respond as objectively as you can to the following questia‘maires' 1) INDICATE, BY MARKDG TIE APPROPRIATE PLACE, mw MILD YO.) RATE THE ISSUE OF REIIJCI‘IGT OF REBIONAL WITIES (LE. BETWEEN AD’IINIST'RATIVE DISTRICTS) IN MALAYSIA ? Very Important: .......... Fairly Important: .......... Less Important: .......... Not Important: .......... 2) INDICATE, IN THE APPROPRIATE PLACE, HOW MDULD Ya! RATE THE ROLE OF TIE CIJVERNVIENI‘ IN REDUCIM; RHHONAL INEQIALITIES (LE. BETWEEN ADVIINISTRATIVE DISTRICTS) IN MALAYSIA ? Very Important: .......... Fairly Important: .......... Less Important: .......... Not Important: .......... 3) PLEASE RATE EVERY PAIR OF INDICATORS OF DEA/mm IN T'I-E 30 X 30 MATRIX OF SELECTED INDICATORS. PLEASE REFER TO THE ATTACHEDSHEETSGT ~INSTRUCTIONONI-ICJAITOFILLINT‘IEMATRIX CELLS' AND ‘LIST OF SELECTED INDICATORS OF DEVELOPMENT" . 282 INST'RUCI‘IGle-UNTOFILLINTTEMATRIXCELLS Thisszrveyattemptstodeterminetherelativeimpartanceof a list of selected ‘development objectives', hereafter called ‘indicators'. Thirty(30)indicatorsareusedinthissurvey(please seeattached sheetm ‘ListOfSelected IrfiicatorsOfDevelanent. Theseindicatarsaretabeoarparedtoeachothermapaired basis.Tadathisatataloften(10)paintsaredistributed toeachindicatorinapairaooordingtotheirrelative importance.Thismeansthatforeveryoellinthe30x30matrix (attached)therearetwo(2) values thatmustsumtatenuO). For exalple, in making comparison between indicator ‘X' and indicator ‘Y', if indicator ‘X' is more desirable to ‘development' than indicator ‘Y' then cne might want to give indicator ‘X' a rating value of 7 and 3 to indicator ‘Y'. However, if one feels that indicator ‘X' is absolutely more desirable than indicator ‘Y' then a distributicn of 10 for ‘X' and 0 for ‘Y' will be appropriate. mtheotherhand, iffimeyareequallydesirabletrmaoell distributim 5 and 5 will reflect this equality. In filling the matrix cells you can use any combination of valuesaslcngasthesumistenuO). 283 LIST OF SELECTED INDICATORS OF DEVEII'JOPIVIEJN'II CRITERIA CHDE INDIOK'IORS Q: DEVELOPIVIEil‘fII CRITERIA E1 ..... Increase populatim with formal schooling EZ ..... Increase population literacy E3 ..... Increase number of teachers Hl ..... Reducenurberofinfantdeathsflessthanlyearold) 1-12.....Reduce number of maternal deaths R2 ..... Improve residential sanitation system Al ..... Increase beneficiaries of piped-water supply A2 ..... Increase beneficiaries of public-supply electricity Sl. . . . .Increase accessibility of population to public law and order enforcement personnel $2 ..... Increase accessibility of population to fire service 01 ..... Improve transportation network (2.....Increase accessibility of population to postal service Ml.....Increase professional and technical marpower m ..... Increase administrative- and managerial-level marpower m ..... Improve distribution of government officials E01....Increase manufacturing activities 1302....Enaourage commercial and business activities Emwlncrease gross domestic production RDl. . . .Increase accessibility of rural population to telephone RD2....Increase agricultural extension services UDl....Increase urban population UD2....Increase size of cities with higher-order services, amenities and facilities WPl....Increase female literacy WP2....Increase participation of female population in non-agricultural economic activities PEl....Reduce number of people in poverty PE2....Increase access of poor people to housing PE3..Reduce number of Malays in agricultural vocations RSl....Increase Malay participation in commercial and business sectors R52....Increase Malay participation in manufacturing sector 30 X 30 MATRIX 0P SELECTED INDICATORS OP DEVELOPMENT Plaaaa till your judge-ant only in calla above the darkened diagonal. §§AMPLE FOR FILLING IHE HAIEIX QELLS Com rim El and £2 The following distribution of 10 points shown that :1 is absolutely more important than :2. 283A APPENDIX 7 284 APPENDIX 7: URBAN & MIRAL DIST'RICI'SH, 1970 AND 1980. UrbanorRural UrbanorRural District 1970 1980 District 1970 1980 JGDRE STATE 42) Kuantan U U l) Batu Pahat U U 43) Lipis R R 2) Johar Bahru U U 44) Pekan R R 3) Kluang U U 45) Raub R U 4) Kata Tinggi R R 46) Tamerloh R R 5) Mersing R R 47) Rmpin * R 6) Muar U U PENAM; STATE 7) Pontian R U 48) S.P. Utara U U 8) Segamat R U 49) S.P. Tengah U U KEDAH STATE 50) S.P. Selatan R R 9) Baling R R 51) P.P. Timur Laut U U 10) Bandar Bahru R R 52) P.P. Barat Daya R R 11) Kata Setar U U PERAK STATE 12) Kuala Mnda U U 53) Batang Padang R R 13) Kubang Pasu R R 54) Dinding/Manjung R R 14) Kulim R U 55) Kinta U U 15) Langkawi R R 56) Krian R R 16) Padang Terap R R 57) Kuala Kangsar U U 17) Sik R R 58) Larut & Matang U U 18) Yan R R 59) Perak Hilir U U 19) Pendang * R 60) Ulu Perak R R KELANTAN STATE 61) Perak Tengah R R 20) Bachok R R PERLIS STATE 21) Kata Bahru U U 62) Perlis R R 22) Machang R R SELAbGOR STATE 23) Pasir Mas R R 63) Gombak * R 24) Pasir Puteh R R 64) Klang U U 25) Tanah Merah R R 65) Kuala Langat R R 26) Tunpat R R 66) Kuala Selangor R R 27) Ulu Kelantan R R 67) Petaling * U 28) Kuala Krai * R 68) Sabak Bernnam R R m STATE 69) Sepang * R 29) Melaka Utara R R 70) Ulu Langat U U 30) Melaka Selatan R R 71) Ulu Selangor R R 31) Melaka Tengah U U TEREMSGANU STATE N. SMILAN STATE 72) Besut R R 32) Jelebu R R 73) Dungun R U 33) Kuala Pilah R R 74) Kamaman R R 34) Port Dickson R U 75) Kuala Terengganu U U 35) Rambau R R 76) Marang R R 36) Seremban U U 77) Ulu Terengganu R R 37) Tanpin R R KUALA LIMPUR 285 APPENDIX 7 ((Int'd. ). 38) Jenpol * R 78) Kuala Lumpur Dist. U ** PAHAM; STATE 78) F.T. of K.Lumpur * U 39) Bentog U U 40)CameronHighland R R *meansdidnatexistyet. 41) Jerantut R R ** means no longer exist. U=URBAN R=RURAL +UrbanDistrict = hasurbanplace with population 20,000 andmore. APPENDIX 8 286 APPENDIX 8: URBAN & RURAL DIST'RICI'SH, 1970 AND 1980 Urban or Rural Urban or Rural District 1970 1980 District 1970 1980 PAHANG STATE JGDRE STATE 36) Bentong U U l) Batu Pahat U U 37) Cameron Highland R R 2) Johar Bahru U U 38) Jerantut R R 3) Kluang U U 39) Kuantan U U 4) Kate Tinggi R R 40) Lipis R R 5) Mersing R R 41) Pekan/Rompin R R 6) Muar U U 42) Raub R U 7) Pontian R U 43) Temerloh R R 8) Segamat R U PENAAG STATE KEDAH STATE 44) S.P. Utara 9) Baling 45) S.P. Tengah 10) Bandar Bahru ll) Kata Setar/ 46) S.P. Selatan 47) P.P. Timur Laut 48) P.P. Barat Daya wcwcc wcwcc: Pendang 12) Kuala Muda wwwwwmcc 2150 wwwwcwcc E12! 13) Kubang Pasu PERAK STATE 14) Kulim 49) Batang Padang R R 15) Langkawi 50) Manjung(Dinding)/ 16) Padang Terap K.Kangsar/P.Hilir/ l7) Sik Perak Tengah U U 18) Yan 51) Kinta U U 52) Kerian R R KELANTAN STATE 53) Larut & Matang U U 19) Bachak R R 54) Ulu Perak R R 20) Kata Bahru U U 21) Machang R R PERLIS STATE 22) Pasir Mas R R 55) Perlis R R 23) Pasir Puteh R R 24) Tanah Merah R R SELADIDR STATE 25) Tmpat R R 56) Kuala Selangor/ 26) Ulu Kelantan/K.Krai R R Sabak Bernnam R R 57) Kuala Langat/Ulu MALACIR STATE Langat/Sepang U U 27) Melaka Utara R R 58) Gombak/Klang/Ulu 28) Melaka Selatan R R Selangor/Petaling/ 29) Melaka Tengah U U W.Perselmtuan(KL) U U N. SEMBILAN STATE TEREDGGANU STATE 30) Jelebu R R 59) Besut R R 31) K.Pilah/Jempol R R 60) Dungun R U 32) Port Dickson R U 61) Kemaman R R 33) Rembau R R 62) Kuala Terengganu U U 287 APPENDIX 8 (QIIt'd. ). 34) Seremban U U 63) Marang R 35) Tanpin R R 64) Ulu Terengganu R U = URBAN DISTRICTS R = RURAL DISTRICTS ++UrbanDistrict=hasurbanplaoewithpap. 20000andmore. APPENDIX 9 ND. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 288 APPENDIX 9:PLACES IN PENINSULAR MALAYSIA WITH A POPUIATION OF 10,000 AND IVORE IN 1970 NAME g PLACE POPULATIGV DISTRICT STATE BATU PAHAT 53291 BAT‘U PAHAT Ja-DRE (JG-DR BAHRU 136229 (IQ-DR BAHRU JCT-DRE KUIAI 11841 JGDR BAHRU MERE W 43272 W JGDRE D’UAR 61218 MJAR JCT-DRE TADG