Methodological approach to the quality of the regional productive structure: The case of Andalusia

1 METHODOLOGICAL APPROACH TO THE QUALITY OF THE REGIONAL PRODUCTIVE STRUCTURE: THE CASE OF ANDALUSIA J. Guzmán Cue as, F. Liñán Alcalde, I. Rome o Luna, F.J. San os Cumplido1 Dp o. Economía Aplicada I. A . Ramón y Cajal, 1. 41018-Se illa (Spain). Uni e si y o Se ille 1. - In oduc ion: economic de elopmen and en ep eneu ial ac i i y. Un il he Second Wo ld Wa , he economis s had always wo ied abou he p oblems o economic g ow h, elabo a ing di e se models o explain i . Fo he Classical economis s, excep o Adam Smi h, he eally impo an hing was capi al accumula ion, dis ega ding echnical change (Hagen 1971). Neoclassical economis s, on hei pa , wo ied, in p inciple, only o he alloca ion o esou ces and he dis ibu ion o income. Howe e , hey ealised la e , due o he pe iodic c ises ha hi he plane , ha he economies we e no always in equilib ium and hey began o elabo a e heo ies o a dynamic cha ac e , as ha o J.E. Meade, o ins ance (Meade 1976), o heo ies ha did ake in o accoun echnical change, like i was he case o p o esso R. Solow's heo y (Solow 1976). The new de elopmen economy ha a ises a e he Second Wo ld Wa , begins o in oduce o he aspec s in i s heo ies o he de elopmen p ocesses, among which e e ences appea o he en ep eneu and he ype o en ep eneu ial ac i i y. In ac , i s a s o be hough ha one o he ac o s ha in luences he economic backwa dness o less de eloped egions is he absence o en ep eneu ial spi i , leading o a weak and disjoin ed en ep eneu ial ne wo k ha implies he con igu a ion o a dependen economy. This new ocus will be used in many occasions o jus i y he ac ha coun ies wi h simila endowmen s in basic ac o s o p oduc ion in some momen o hei his o y ha e no e ol ed in a same way. Thus, I is necessa y o men ion, in his sense, he heo y o “The S ages o Economic G ow h” (Ros ow, 1960), he app oaches ha pu he emphasis in 1 We wish o hank J. E. Rome o Ga cía and E. M. He nández Salme ón o hei help ul collabo a ion. 2 indus ialisa ion, ei he om he pe spec i e o an “Unbalanced de elopmen ” (Hi schman, 1961), o om a “balanced de elopmen ” (Nu kse 1953) and he es uc u alis app oaches (P ebisch 1971), as well. Ne e heless, all he con ibu ions o his new app oaches, in spi e o conside ing o a ce ain ex en he en ep eneu 's ole and he ype o en ep eneu ial ac i i y, ollowed a model ha we could denomina e “ op-down” in which he absence o capi al and quali ied labou is conside ed as he basic p oblem ha hinde s he de elopmen o backwa d a eas, and hey p opose as solu ion bo h ex e nal inancial help o la ge en e p ises and he mobili y o he labou o ce. This app oach s a ed o change in he eigh ies, as he measu es i p oposes do no gi e an answe o economic backwa dness in an inc easingly compe i i e en i onmen . I is hen when models o endogenous o “bo om-up” de elopmen begin o be elabo a ed. Theses models conside as he main obs acle o de elopmen he unde -u ilisa ion o he au och honous esou ces, specially he local en ep eneu s, because hey a e he economic agen s ha ca y ou he inno a ions and hey can con ibu e o he c ea ion o an en ep eneu ial ne wo k ha is no dependen on o he economic a eas. The models o endogenous de elopmen poin ou -as solu ion o he economic backwa dness- o he incen i a ion o economic e iciency, ollowing a less in e en ionis a i ude, o e ing se ices o he local en ep eneu s and s imula ing hei coope a ion, so ha hey become he ue ac o s o he de elopmen p ocess, and no he ex e nal economic help exclusi ely. They a e models ha ake a mo e ac i e s a egy, as hei pu pose is o educe he echnological and o ganisa ional gap be ween he s onges and weakes egions, emphasising he concep o dynamic compe i i eness, based on he capaci y o adop inno a ions and o p omo e he in e na ionalisa ion o en e p ises (Cappellin 1991). F om his poin o iew, he models o endogenous de elopmen g an special impo ance o he dynamism o he local en e p ises, gene ally o small and medium size, ma ked by inno a i e en ep eneu s wi h ini ia i e capabili y. In his sense, 3 endogenous de elopmen is also in ima ely ela ed wi h he local economic po en ial ha a ises om local socio-economic ins i u ions and ac o s, because he public and p i a e local ac o s a e hose esponsible o he in es men ac ions and o he con ol o p ocesses (Vázquez 1997). Ne e heless, o de e mine he deg ee o endogenous de elopmen o a egion and o implemen he ecipes o hese new models, i is necessa y o ha e a p e ious diagnosis o allow o an accu a e knowledge o he p o ile o en ep eneu s and ac i i y sec o s ha con ibu e mos o he de elopmen p ocess om an endogenous pe spec i e. This ypology has o be elabo a ed based on ce ain a iables ha ga he essen ial aspec s o endogenous de elopmen , such as he o wa d and backwa d in e sec o al linkages, he deg ee o in e na ionalisa ion, he p oduc i i y le el o he di e en ac i i ies, he a e age size o en e p ises in each ac i i y, e c. In his sense, he ques ion we speci ically ou line in his wo k is he explana ion o a me hodology ha may be alid o iden i y he quali y o he egional p oduc i e s uc u e om he endogenous de elopmen pe spec i e and ha may be used o achie e a g ea e e ec i eness om local and egional economic policy. This ask, in a heo e ical sense, will be ca ied ou aking he inpu -ou pu analysis as he basis, and hen, as an empi ical demons a ion, using speci ic da a om Andalusia, one o he mos backwa d Spanish egions. 2. - Me hodological p oposal o he analysis o he quali y o he egional p oduc i e s uc u e As i has been poin ed ou , he me hodology we p opose in his pape consis s essen ially on he cons uc ion o a ypology o p oduc ion sec o s acco ding o hei impo ance o he endogenous de elopmen o a egion, his will help us o know which is he quali y o he egional p oduc i e s uc u e. To do ha , we will s a om a ce ain disagg ega ion o he egional p oduc i e ac i i y in 4 sec o s and om he in o ma ion p o ided by six a iables ha , in ou opinion, a e ele an o such a pu pose. The conside ed a iables a e desc ibed below, classi ying hem in h ee g oups, and hei inco po a ion o he analysis is jus i ied. The in o ma ion ega ding all he included a iables can be ob ained om he egional inpu -ou pu ables, equi ing in some cases he elemen a y applica ion o ce ain de elopmen s o he inpu -ou pu model. 2.1. - Desc ip ion o he a iables A. - P oduc i e a icula ion F om an endogenous de elopmen pe spec i e, i is o special in e es he deg ee o p oduc i e a icula ion o a egional economy, as he solidi y o he g ow h pa e n is condi ioned by he densi y o ha “ issue” o in e sec o linkages ha con o ms he egional p oduc i e s uc u e. In a mo e pa icula sense, he s a egic cha ac e o a sec o c ucially depends on he ex en o which i is in eg a ed in he egional p oduc i e s uc u e. To app oach he in ensi y o hose in e sec o ela ionships we use as indica o s wo mul iplie s o gene alised use in he ield o he inpu -ou pu analysis. 1. Regional Ou pu Mul iplie s. The Ou pu Mul iplie s es ima e he o al e ec , di ec and indi ec , ha he inc ease in a mone a y uni o he inal demand o a speci ic sec o has on he p oduc ion o all he o he sec o s join ly. We a e speci ically conside ing he Regional Ou pu Mul iplie s, ob ained by summing up he columns o he in e se ma ix o egional echnical coe icien s2. 2 The in e se ma ix o egional echnical coe icien s ga he s he pa o he in e media e consump ions supplied wi hin he egion, he e o e, excluding he impo s. In he whole pape , we will unde s and as impo s o expo s he pu chases o sales wi h o igin o des ina ion ex e nal o he egion, including ade wi h o he egions wi hin he same s a e. 5 This a iable allows he quan i ica ion o he pulling capaci y o each sec o on he whole o he egional economy. Ob iously, hose sec o s ha p esen a highes alue o he mul iplie will be s a egic om an endogenous de elopmen pe spec i e, due o hei capaci y o inc ease he dynamism o he egional p oduc i e appa a us. 2. Uni o m Demand Expansion Mul iplie s This a iable, ob ained summing up he ows o he egional in e se ma ix o echnical coe icien s, shows he inc ease in he p oduc ion o a speci ic sec o due o an uni a y inc emen in he inal demand o all sec o s. Fo he pu pose o ou analyses, his indica o e lec s he sensibili y o he di e en sec o s o he gene al e olu ion o he egional economy, so ha a highe alue o he mul iplie e lec s a g ea e in eg a ion o ha sec o in he egional economy. Ne e heless, o a co ec in e p e a ion o he alue o his mul iplie i will be necessa y o keep in mind he si ua ion o he sec o wi hin he alue chain, dis inguishing he basic sec o s om hose o he s whose p oduc ion is basically add essed o he inal demand. While he o me , gi en hei pape as inpu supplie s o o he ac i i ies, gene ally p esen a high alue o he mul iplie , he la e will p obably p esen a low alue. B. - P oduc i i y The inclusion in he analysis o p oduc i i y indica o s is equally necessa y, as he sec o s wi h highe capaci y o es imula e economic g ow h a e p ecisely hose wi h highe le els o p oduc i i y. 3. Appa en labou p oduc i i y We ake as basic indica o o labou o ce p oduc i i y he appa en p oduc i i y, de ined as he quo ien be ween he sec o al G oss Value Added a ma ke p ices (GVAmp) and he numbe o employees in he sec o . 6 4. Ra io GVA / E ec i e P oduc ion. Ano he indica o ha we include in he analysis is he ela ionship be ween he GVA a ma ke p ices and he E ec i e P oduc ion (EP), indica o ha ep esen s he alue added gene a ion capabili y o each sec o 3. This a iable can be equally in e p e ed as an indica o o he p oduc i i y o he a iable capi al4. C. – Ex e nal T ade 5. Expo Ra io We inco po a e o he analysis, as an indica o o expo dynamism o each sec o , he quo ien be ween he alue o he expo s and he E ec i e P oduc ion, which e lec s he p opo ion o he sec o al p oduc ion de o ed o expo a ion. In gene al e ms, i will be necessa y o alue posi i ely a high Expo Ra io as e ealing a s ong compe i i e capabili y o he sec o . Besides, we ha e o keep in mind ha a sec o ca ying ou an in ense expo ac i i y a oids he limi a ion o i s expansion possibili ies by he g ow h o he egional ma ke . Ne e heless, one has also o be conscious ha ce ain sec o s ha e a esiden ial cha ac e , and hey de o e hei p oduc ion almos exclusi ely o he local ma ke , by hei own na u e; his would be he case, basically, o ce ain ac i i ies o he se ice sec o , such as Pe sonal Se ices, Educa ion, Heal h, e c. 3 The alue o he E ec i e P oduc ion (EP) co esponding o any ac i i y is ob ained, based on he in o ma ion p o ided by he inpu -ou pu amewo k, h ough he addi ion o g oss alue added a ma ke p ices in each sec o and he alue o he necessa y in e media e consump ions o ob ain he sec o al p oduc ion. 4 An app oxima ion o he a iable capi al inco po a ed o he sec o al p oduc ion p ocesses could be he alue o he in e media e consump ions om o he ac i i ies; so he quo ien be ween he g oss alue added a ma ke p ices (GVAmp) and he alue o he in e media e consump ions (IC) could be used as an indica o o he a iable capi al p oduc i i y. And i can be easily shown ha he e exis s a simple a i hme ic ela ionship be ween his indica o and he p oposed ela ionship GVA/EP: GVA GVA/EP -------- = ----------------- IC 1 – GVA/EP 7 6. - Need o indi ec impo s We also include as a ele an a iable he alue o he in e media e impo s o uni o inal demand o each sec o 5. A high need o impo s is nega i e om a poin o iew o endogenous de elopmen because i implies ex e nal dependence and i condi ions a sca ce p oduc i e a icula ion o he egional economy. Impo s ac , he e o e, as an escape channel wi hin he whole o he sec o al in e ela ions ha con o m he egional p oduc i e s uc u e. 2.2. - Clus e analysis Based on he in o ma ion p o ided by he chosen a iables, we y o classi y he di e en ac i i y sec o s ha con o m he egional p oduc i e appa a us in di e en g oups. Fo his aim, he applica ion o some echnique o clus e analysis is equi ed, and he me hod o he k-a e ages has been adop ed. This p ocedu e assigns each elemen (in ou case each sec o ) o a clus e so ha he dis ance wi h ega d o he cen e o he same one is minimum. The cen es o he clus e s, in u n, can be known o unknown a p io i; in ou case we s a om unknown cen es (Visau a, B. ;1998). P e iously o he applica ion o he clus e analysis, i will be necessa y o check o he absence o co ela ion among he conside ed a iables and o p oceed o hei no malisa ion ( ypi yca ion). In case co ela ion exis s among any o he a iables, i will be necessa y o exclude some o hem om he analysis, un il ha ing a g oup o unco ela ed ones. This ope a ion is necessa y, as he inclusion 5 This indica o is ob ained om he in o ma ion p o ided by he inpu -ou pu able in he ollowing way (Delgado Cabeza, M. ;1995): s a building a ma ix C o coe icien s o impo needs, i s gene ic elemen is gi en by he exp ession: Cij = Xij / Xj whe e Xij ep esen s he alue o he impo s o good i needed o ob ain he p oduc ion o he sec o j (Xj). S a ing om he ma ix C, he ma ix M o o al impo needs (di ec and indi ec ) is ob ained as: M = C(I-A)-1 whe e A is he ma ix o echnical coe icien s o p oduc ion. The sum by columns o he elemen s o he ma ix M p o ides he alue o he in e media e impo s o uni o inal demand o each sec o . 8 o co ela ed a iables would ha e he e ec o g an ing a la ge impo ance o he in o ma ion hey p o ide o he de e mina ion o he g oups. This echnique o mul i a ia e analysis lea es he decision on he numbe o clus e s o he esea che ’s will, ponde ing he cla i y in he in e p e a ion o he esul ing g oups as well as he homogenei y o he ac i i ies assigned o each one o hem. In his sense, once ca ied ou he classi ica ion and he assignmen o each sec o o a g oup, i is necessa y o check ha he g oups a e su icien ly sepa a e o each o he . I would also be con enien o apply he disc iminan analysis o con i m ha he di e en sec o s a e assigned co ec ly o he g oups o med by i ue o he clus e analysis. Then, we will p oceed o he in e p e a ion o each one o he g oups, e alua ing hei impo ance om he iewpoin o endogenous de elopmen acco ding o he cha ac e is ics ha iden i y hem. 2.3. - Cha ac e is ic o he en ep eneu ial ne wo k F om an endogenous de elopmen s andpoin , i is basic in he p oposed me hodology o conside he cha ac e is ics o he en ep eneu ial ne wo k. Al hough he s eng h o a egional economy is condi ioned by he exis ence o la ge en e p ises, hese may in many cases ac independen ly om he es o he economy wi hou ca ying ou , he e o e, an impo an ole as a icula ing agen s o he egional p oduc i e s uc u e. On he con a y, an en ep eneu ial s uc u e cha ac e ised by he p eponde ance o mic oen e p ises implies an a omised en ep eneu ial ne wo k, insu icien o ini ia e a con inuous p ocess o p oduc i e in es men s ha p omo e egional de elopmen . Mic oen e p ises gene ally main ain a dependen ela ionship wi h espec o la ge en e p ises o , al e na i ely, hey a e guided o in e nal demand, supplying local ma ke s p o ec ed om na ional and in e na ional compe i ion. In any case, whe he hey a e a ype o mic oen e p ises o he o he , he esul will be he same: a weak and ulne able en ep eneu ial ne wo k (Guzmán J., San os, J. and Cáce es, R.; 1998). 9 The impo ance o he en ep eneu ial s uc u e o endogenous de elopmen , jus i ied abo e, akes us o conside his aspec as ele an in o de o comple e he analysis o he quali y o he egional p oduc i e s uc u e. Thus, he las phase o he me hodology would consis on app oaching he s udy o he en ep eneu ial composi ion o he sec o s o each clus e , wi h he pu pose o cap u ing possible dis inc i e ea u es among hem. In his sense, i will be necessa y o conside a iables as he a e age size o en e p ises in he sec o s o each g oup, o he ela i e p esence o mic o-, small, medium and la ge en e p ises (measu ed by he pe cen ages o en e p ises included in each ca ego y). Many au ho s ag ee in poin ing ou how i is pa icula ly a ou able, om an endogenous de elopmen pe spec i e, he p esence o small and medium-sized en e p ises (SMEs), due o hei lexibili y and adap a ion capabili y in dynamic en i onmen s, hei capaci y o employmen gene a ion, hei in eg a ion in he egional p oduc i e s uc u e and hei na ow linkage o he e i o y. This ype o en e p ises will be called o assume a main ole encou aging egional de elopmen (Wadley, D.; 1988). Finally, he analysis o he cha ac e is ics o he p oduc i e s uc u e would be comple ed wi h he numbe o en e p ises ha comp ise each clus e and i s con ibu ion o he egional GVA. In his way, he quali y o he egional p oduc i e s uc u e would be de e mined om he endogenous de elopmen pe spec i e. 3.- Applica ion o he Andalusian Economy 3.1.- Typology o p oduc ion ac i i ies In his sec ion we will ca y ou a p ac ical applica ion o he p oposed me hodology, in which i s possible u ili y o iden i y he ac i i ies ha con ibu e mos o he endogenous de elopmen , as well as he cha ac e is ics o hose ac i i ies, will be shown. 16 egion. This lis ing has been pu i ied elimina ing hose es ablishmen s ha do no ha e he conside a ion o en e p ises, such as public bodies, neighbou s' associa ions, poli ical and eligious o ganisa ions, ade unions, public schools and hospi als, e c.. On he o he hand, he es ablishmen s belonging o he same company ha e been added up, o a oid duplici ies. Likewise, om ha lis ing we ha e excluded en e p ises wi hou any sala ied employee (sel -employed wo ke s, basically), because we conside necessa y o s a om a minimum le el o consis ency and dimension o he p oduc i e uni . The small and medium-sized en e p ises (SMEs), as we ha e exposed in he second sec ion, play a e y impo an ole in he p ocess o endogenous egional de elopmen . In con as , when la ge en e p ises -al hough hey undoub edly also con ibu e in a posi i e way o economic g ow h- a e no au och honous, hei ac i i y may con ibu e o inc ease he dependence on o he a eas o a egional economy. The e o e, he s udy o he en ep eneu ial composi ion o he i e clus e s conside ed allows us o analyse he p esence o di e en ca ego ies o en e p ises in each one o hem. Table 3 Dis ibu ion o en e p ises by clus e s G oup No. en e p ises A e age size Size dis ibu ion (No. wo ke s)1 Mic oen (1-9) Small (10-99) Medium (100-499) La ge (+ 500) 1 2 3 4 5 2 20371 31819 3418 6059 817.00 8.37 11.54 17.26 17.56 0.00% 81.56% 75.62% 66.35% 66.78% 0.00% 17.67% 23.27% 31.25% 31.28% 0.00% 0.69% 0.96% 2.14% 1.58% 100.00% 0.08% 0.15% 0.26% 0.36% TOTAL 61669 11.43 76.20% 22.65% 1.00% 0.16% 1 Acco ding o Eu opean Obse a o y o SMEs (1995) G oup 1 is made up o only wo oil e ine ies wi h mo e han 500 employees each one. I is a clea example o how la ge en e p ises can main ain sca ce ela ionships wi h he p oduc i e s uc u e o he egion whe e hey a e loca ed. In spi e o he impo ance o his sec o as an ene gy sou ce, i does no con ibu e in any ema kable way o he egional de elopmen , o hei li le in eg a ion wi hin he local economy. 17 G oup 2, on he con a y, p esen s a e y high pe cen age o mic oen e p ises, abo e 80% (see Table 3), and he lowes in all he o he sizes. They a e e y a omised ac i i ies, wi h en e p ises o low dimension - he a e age size is only 8.4 wo ke s-. In logical co espondence, he Regional Employmen Mul iplie is he lowes in he ou g oups, wi h a alue o 0.176 (wi hou conside ing g oup 1)6. This g oup con ains one- hi d o en e p ises, bu i ep esen s only 25% o o al p oduc ion. I shows, he e o e, ha hese ac i i ies ha a e basic supplie s o he es o he economy, p esen an impo an weakness in hei en ep eneu ial composi ion. The dis ibu ion by sizes in g oup 3 nea ly coincides wi h he global a e age. We ha e al eady conside ed, in he p e ious sec ion, he weaknesses o he p oduc i e ac i i ies o his clus e . The ac ha he global dis ibu ion o en e p ises coincides wi h ha o his g oup should make us medi a e abou he gene al weakness o he whole o he Andalusian economy, specially when his g oup 3 ep esen s mo e han hal o he egional p oduc ion and o he conside ed en e p ises. Howe e , wi hin his g oup he e exis ema kable di e ences wi h espec o he ac i i y ype. In he indus y, he medium and la ge en e p ises p esen a g ea e ela i e abundance. This is cohe en wi h ha exposed in sec ion 3.1, whe e we indica ed ha hese a e ac i i ies a om he inal consump ion (in e media e o capi al-good p oduc ions) ha end o equi e a la ge plan size. The se ice ac i i ies in his g oup a e he mos a omised ones and whe e he e is, wi h di e ence, smalle p esence o medium and la ge en e p ises. In gene al e ms, hese ac i i ies a e cha ac e ised by hei in ensi e use o manpowe , bu in e y small p oduc ion uni s, which educes hei compe i i eness. The demand 6 The employmen mul iplie measu es, using he in e se ma ix, and he ec o o employmen o p oduc ion a ios, he o al employmen ha is gene a ed in he economy due o an inc ease in he sales o he ac i i y sec o being conside ed. In his case, i means ha an inc ease o Pese as 1 million in he sales o any sec o o he g oup 2 causes a g ow h in employmen o 0.176 wo k posi ions. O , o say i in a di e en way, ha he 18 o hese se ice sec o s depends on he egional en , as hey a e e y much linked o inal consump ion. The e o e, hei demand g ow h is de i ed om ha o o he sec o s, h ough he e ec ha hese o he ac i i ies gene a e in he Andalusian en and weal h. The e o e, hese se ice ac i i ies canno be cons i u ed in locomo i es o he egional de elopmen . In any e en , he high Employmen Mul iplie o his g oup 3 (0.252, he highes in he i e g oups), in spi e o he educed alue o he Ou pu Mul iplie , con i ms he high in ensi y o employmen , especially in he se ice ac i i ies ha comp ise i . The G oup 4, ha we ha e iden i ied as he one ha p esen s a g ea e endogenous de elopmen po en ial, also p esen s a mo e a ou able en ep eneu ial composi ion. The a e age size o en e p ises is no ably highe , wi h a pe cen age o small, medium and la ge en e p ises also e y supe io o he gene al a e age. Thus, hese ac i i ies a e no only mo e in eg a ed wi hin he es o he egional ac i i y, bu hey a e also made up o la ge en e p ises ha can compe e unde be e condi ions in he ex e nal ma ke s. Howe e , ha be e ela i e posi ion is somewha coun e ac ed by he educed size o he g oup. These ac i i ies ep esen only 5.5% o he en e p ises and ha dly each 6.3% o he egional p oduc ion. We should conclude, he e o e ha he po en ially mos in e es ing ac i i ies in Andalusia ep esen a e y small pa o he p oduc i e ne wo k, wha undoub edly cons i u es an obs acle o i s o de in he p ocess o egional de elopmen . Las ly, in he G oup 5 we mee wi h an en ep eneu ial s uc u e e y simila o ha o g oup 4, wi h a sligh ly lowe pe cen age o medium-sized en e p ises, and a p opo ion o mic o-, small and la ge en e p ises sligh ly highe . The high a e age size o he en e p ises o his g oup may ha e ela ionship wi h he lowe in eg a ion in he egional p oduc i e s uc u e and wi h he si ua ion o dependence o hese sales o he ac i i ies o his g oup has o g ow in Pese as 5.682 million o gene a e an addi ional employmen in Andalusia. 19 en e p ises wi h espec o o he s loca ed ou side he egion, as hey would ha e o ope a e guided owa ds much wide ma ke s ha hose s ic ly Andalusian. I seems ha he en e p ises o up o 100 employees could be wo king as subo dina es o o he la ge en e p ises om ou side o he egion, while he la ge ones would be loca ed in Andalusia, o a conside able ex en , o ake ad an age o subsidies and/o cheape manpowe . Howe e , he da a on en ep eneu ial dis ibu ion a e no su icien ly conclusi e in his espec . The ac i i ies o he las wo g oups a e no only e y simila in hei dis ibu ion o en e p ises, bu also in he Regional Employmen Mul iplie (lowe o ha o g oup 3, bu highe o ha o g oups 1 and 2), al hough his is highe in g oup 4 (0.199 jobs pe Pese as 1 million in on o 0.188 in g oup 5). This would be explained by he g ea e egional in eg a ion in ha g oup 4. In spi e o he a e age size o he en e p ises being app oxima ely he same in he wo g oups, en e p ises in g oup 4 no only gene a e employmen in i s own sec o , bu hey also ha e a g ea e pulling e ec on o he egional ac i i ies, c ea ing employmen in hem. 4.- Conclusions This pape has s a ed ou lining he impo ance ha achie ing de elopmen om an endogenous pe spec i e ep esen s wi hin he cu en economic con ex . This means making be e use o he own p oduc i e esou ces, among which he local en ep eneu s a e included as a comp ising pa o he human esou ces. The local en ep eneu s and hei inno a ion capaci y can encou age economic de elopmen , p e en ing an excessi e dependence o he egion whe e hey ac wi h espec o o he mo e de eloped egions, and also con ibu ing o a mo e sus ained and balanced de elopmen . In his sense, an indica o o he con ibu ion o local en ep eneu s o he endogenous de elopmen is he exis ence o a high quali y en ep eneu ial ne wo k, ha is o say, an en ep eneu ial ne wo k comp ised o a la ge ex en o compe i i e en e p ises o au och honous capi al. Ne e heless, keeping in mind ha mos o he la ge en e p ises loca ed in he backwa d a eas usually a e o o eign capi al, he quali y o he en ep eneu ial ne wo k om he endogenous pe spec i e would be basically condi ioned by he kind o exis ing small 20 and medium-sized en e p ises. The e o e, based on his p emise, we ha e elabo a ed a me hodology ha allows he assessmen o he deg ee o endogenous de elopmen o a ce ain egion s a ing om he analysis o i s p oduc i e s uc u e, ha assessmen has also been conside ed a undamen al ins umen o p o ide in o ma ion o hose esponsible o applying egional and local economic policy measu es. The p oposed me hodology consis s o h ee clea ly di e en ia ed phases. In i s place, he a iables ha we conside as bes explaining he quali ies he egional p oduc i e s uc u e should possess o con ibu e o an endogenous-like de elopmen ha e been desc ibed. Some o hese a iables a e ob ained di ec ly om he inpu -ou pu analysis, as hose de ining he deg ee o p oduc i e a icula ion: he Regional Ou pu Mul iplie and he Uni o m Demand Expansion Mul iplie . These wo a iables a e basic o de e mine he in e sec o linkages among he p oduc i e ac i i ies. In his sense, he g ea e he in e sec o linkages o a ce ain ac i i y a e, he mo e i con ibu es o he endogenous de elopmen . Wi h espec o he es o he a iables, bo h hose measu ing p oduc i i y (Appa en Labou P oduc i i y o GVA / E ec i e P oduc ion Ra io) o he one de ining he expo dynamism (Expo Ra io), hey can also be ob ained om he in o ma ion p o ided by he inpu -ou pu ables. P oduc i i y as well as expo dynamism a e wo elemen s ha e lec he compe i i eness o a p oduc i e ac i i y. In his sense, i has been poin ed ou ha he g ea e p oduc i i y and expo dynamism o an ac i i y a e, he mo e hey con ibu e o he endogenous de elopmen . The second phase o he p oposed me hodology has consis ed on elabo a ing a ypology o p oduc i e ac i i ies based on he whole se o a iables de ining endogenous de elopmen . In his sense, he clus e analysis has been p oposed, using he me hod o he k-a e ages, o alloca e each p oduc i e ac i i y o a ce ain clus e , so ha he dominan ype o ac i i y can be de e mined in each case. Ob iously, i he clus e con ibu ing o a g ea e ex en o he endogenous de elopmen is domina ed by p oduc i e ac i i ies wi h low demand and low echnological con en , hen he quali y o he p oduc i e s uc u e will no seem o be he mos app op ia e. 21 Las ly, o de e mine wi h g ea e p ecision he quali y o he p oduc i e s uc u e, a hi d phase has been p oposed in he me hodology o analyse some o he cha ac e is ics o he en e p ises in eg a ing each clus e , as o example, hei a e age size, he numbe o en e p ises comp ising he clus e , o hei inal con ibu ion o he egional GVA. Rega ding he size o he en e p ises, we s a om he assump ion ha he clus e whe e small and medium-sized en e p ises (SMEs) clea ly domina e will include ac i i ies ha may p esen a g ea e quali y in ela ion o he endogenous de elopmen . Ne e heless, he p esence o a high numbe o la ge en e p ises in a ce ain clus e does no necessa ily imply sca ce connec ion wi h he endogenous de elopmen , p o ided a s ong p esence o SMEs exis s, wha would make he si ua ion e en mo e desi able. The s udy o he numbe o en e p ises ha comp ise each clus e and hei inal con ibu ion o he egional GVA comple es he analysis, since –once he cha ac e is ics o each g oup a e known- i shows he quali y le el o he egional p oduc i e s uc u e. To conclude his pape , he p oposed me hodology has been applied o he case o he Andalusian egion, in Spain, one o he mos backwa d a eas in he Eu opean Union. In his sense, s a ing om he da a p o ided by he las inpu - ou pu able elabo a ed o he egion, in he yea 1990, he alues o he ou a iables ha ha e been conside ed as de ining he endogenous de elopmen (Regional Ou pu Mul iplie , UDE Mul iplie , Appa en Labou P oduc i i y and Expo Ra io) ha e been calcula ed o each ac i i y and, hen, he ypology o ac i i ies has been elabo a ed h ough he clus e analysis, ob aining i e ypes o sec o s as a esul . The ac i i ies o he g oup 4, hose ha u n ou o con ibu e mos o he endogenous de elopmen , a e basically indus ial ac i i ies o low demand and low echnological con en . Ob iously, his esul is in acco dance wi h he low deg ee o bo h indus ial and economic de elopmen o he egion. Finally, he hi d phase o he me hodology has been ca ied ou using he da a on en ep eneu ial size and numbe o en e p ises o each ac i i y p o ided by a da abase o Andalusian en e p ises, o he yea 1990 as well. The esul has been ha –wi h espec o en ep eneu ial size- he g oup 4, again, is comp ised o 22 he ac i i ies ha con ibu e mos o he endogenous de elopmen , as i con ains a high p opo ion o SMEs and also a good numbe o la ge en e p ises. Ne e heless, he weakness o his g oup o ac i i ies is clea , as shown by he educed numbe o en e p ises wi hin i and i s small con ibu ion o he egional GVA. In conclusion, i we sum up his las in o ma ion o he p e iously indica ed ac ha his g oup 4 ga he s ac i i ies o low demand and low echnological con en , we come o he conclusion ha he quali y o he Andalusian p oduc i e s uc u e is no he mos sui ed o encou age endogenous de elopmen . Re e ences Cappellin, R (1991) "The New G a i y Cen e s o Regional De elopmen in he Eu ope o he 90's". Pape p esen ed o he XVII Reunión de Es udios Regionales, Ba celona Delgado Cabeza, M. (1995): “Las elaciones con el ex e io de la es uc u a p oduc i a andaluza. Análisis de las necesidades de impo ación. 1980-1990”; in Ins i u o de Es adís ica de Andalucía: Con abilidad Regional y Tablas Inpu -Ou pu de Andalucía. Análisis de Resul ados. Se illa. Diez de Cas o, E.P. (Coo ) (1995): La Emp esa en Andalucía. Ed. Ci i as. Mad id. Eu opean Obse a o y o SMEs (1995): Thi d Annual Repo . Eu opean Ne wo k o SME Resea ch. Zoe e mee (Ne he lands). Guzmán Cue as, J., San os Cumplido, J., Cáce es Ca asco, R. (1998): “An Analysis o Regional De elopmen Agencies in Spain om Endogenuos De elopmen Pe spec i e”, in Halkie , H., Danson, M. and Dambo g, C. (eds): Regional De elopmen Agencies in Eu ope. “Regional Policy and De elopmen Se ies”, nº 21. Jessica Kingsley Publishe s L d. and Regional S udies Associa ion. London and Philadelphia. Hagen, E e e E. (1962) On he Theo y o Social Change: How Economic G ow h Begins. Massachuse s Ins i u e o Technology. The Do sey P ess Inc, Homewood, Illinois Hi schman, Albe . O (1958) The S a egy o Economic De elopmen . Yale Uni e si y P ess, New Ha en. Ins i u o de Es adís ica de Andalucía (1993): Con abilidad Regional y Tablas inpu -ou pu de Andalucía 1990. P esen ación de esul ados. Se illa. Meade, J.E. (1976) Una Teo ía Neoclásica del C ecimien o Económico. (1s ed. 1961: A Neo- Classical Theo y o Economic G ow h) Fondo de Cul u a Económica, México. Mille , R.E. y Blai , P.D. (1985) Inpu -Ou pu analysis. Founda ions and ex ensions. P en ice-Hall, Inc. Englewood Cli s, New Je sey. Nu kse, R. (1953) P oblems o Capi al Fo ma ion in Unde eloped Coun ies. Ox o d Uni e si y P ess. O e o Mo eno, J.M. (1995): “Mul iplicado es de la economía andaluza: concep os, medida y guía de aplicación”; in Ins i u o de Es adís ica de Andalucía: Con abilidad Regional y Tablas Inpu -Ou pu de Andalucía. Análisis de Resul ados. Se illa. 23 P ebisch R. (1971) Hacia una Dinámica del Desa ollo La inoame icano. Fondo de Cul u a Económica, México. Pulido, A. and Fon ela, E. (1993): Análisis Inpu -ou pu . Modelos, Da os y Aplicaciones, Ed. Pi ámide, Mad id. Ros ow, W.W. (1960) The S ages o Economic G ow h: A Non-Comunis Mani es o. Camb idge Uni e si y P ess, Camb idge San os Cumplido, F. J. (1997) “El Papel del Emp esa io y la Polí ica de Apoyo a las PYMEs”. Temas Labo ales, nº.42, pp. 235-258. Solow, R. (1976) La Teo ía del C ecimien o. (1s edi ion: G ow h Theo y 1970) Fondo de Cul u a Económica, México. Visau a Vinacua, B. (1998): Análisis es adís ico con SPSS PARA WINDOWS. Es adís ica Mul i a ian e, Ed. Mcg aw-HILL/In e ame icana de España, Mad id. Wadley, D. (1988): “Es a egias de desa ollo egional”, Papeles de Economía Española, nº 35.