WHAT DO MEMBERS EXPECT FROM CLUSTER MEMBERSHIP? THE CASE OF THE CROATIAN WOOD CLUSTER

59 2, XXV, 2022 Business Adminis a ion and Managemen 10.15240/ ul/001/2022-2-004 WHAT DO MEMBERS EXPECT FROM CLUSTER MEMBERSHIP? THE CASE OF THE CROATIAN WOOD CLUSTER I an-Dami Anić1, I ana Rašić2, Zo an A alica3 1 The Ins i u e o Economics, Depa men o Inno a ion, Business Economics and Economic Sec o s, C oa ia; ORCID: 0000-0003-0179-6410, dAnić@eizg.h ; 2 The Ins i u e o Economics, Depa men o Regional De elopmen , C oa ia, ORCID: 0000-0002-5157-9134, [email p o ec ed]; 3 The Ins i u e o Economics, Depa men o Regional De elopmen , Depa men o Inno a ion, Business Economics and Economic Sec o s, C oa ia, ORCID: 0000-0001-6290-5866, [email p o ec ed]. Abs ac : The pu pose o his pape is o in es iga e he composi ion o se ices and i ms in a wood indus ial clus e based on membe s’ in e es s and examine he di e ences among he g oups. Since i ms ha e e y di e en in e es s when joining he clus e , his pape de elops he axonomy o clus e membe s based on hei expec a ions ega ding se ices and in es iga es he di e ences and simila i ies among g oups in i ms’ cha ac e is ics, hei esou ces, inno a i e ac i i ies, he pe o mance o i ms, and pe cei ed clus e pe o mance. The da a collec ed om membe s o he C oa ian wood clus e we e explo ed using ac o and clus e analyses. The esul s e ealed h ee ca ego ies o se ices ha he clus e should p o ide: 1) lobbying; 2) acili a ion o ne wo king and coope a ion; and 3) suppo ing join logis ics and join access o he ma ke . Fou g oups o membe s we e iden i ied, named as 1) coope a ion and lobbying-o ien ed; 2) ma ke - o ien ed; 3) pu ely lobbying-o ien ed; 4) pu ely ne wo king and coope a ion-o ien ed. G oup 1 (44% o he sample) sco es high on all ca ego ies o se ices. I con ains mo e i ms ha inno a e and mo e high-g ow h i ms. Membe s in his g oup e alua e he e ec s o he clus e be e han he o he wo g oups. The s udy o e s new e idence on he alignmen o clus e se ices wi h he in e es s o clus e membe s, indica ing di e en le els o pe o mance among g oups in a small clus e in adi ional o es -based indus ies. The indings can help manage s no only in he C oa ian wood clus e bu also in o he simila small clus e o ganisa ions in Cen al Eas e n and Sou h- Eas e n Eu ope o use hei limi ed esou ces mo e e icien ly and enhance he compe i i eness and pe o mance o he i ms/sec o . Keywo ds: Wood indus ial clus e , indus ial clus e , clus e o ganisa ion, clus e se ices. JEL Classi ica ion: L14, L21, L23, L26, L68, L73. APA S yle Ci a ion: Anić, I.-D., Rašić, I., & A alica, Z. (2022). Wha do Membe s Expec om Clus e Membe ship? The Case o he C oa ian Wood Clus e . E&M Economics and Managemen , 25(2), 59–74. h ps://doi.o g/10.15240/ ul/001/2022-2-004 In oduc ion Indus ial clus e s ha e been conside ed a success ul o m o i ms’ o ganisa ion, and an ins umen o inc easing he compe i i eness and g ow h o he egion and i ms’ pe o mance (A angu en e al., 2013; Abdesslem & Chiappini, 2016; S ojčić e al., 2019; Pecze, 2019). The idea behind clus e o ma ion and i s p omo ion is ha i ms bene i om agglome a ion e ec s due o knowledge and echnology spillo e s, inpu and ou pu sha ing, and loca ion ex e nali ies (Ma ioli e al., 2016). Clus e o ganisa ions should acili a e collabo a ion among i ms, scien i ic and public ac o s, ne wo king, in e - i m linkages, sha ing o in o ma ion, esou ces, and knowledge o EM_2_2022.indd 59 1.6.2022 17:52:11 60 2022, XXV, 2 Business Adminis a ion and Managemen build new g ow h oppo uni ies, and o e come coo dina ion ailu es (Ma ioli e al., 2016). Pas esea ch has examined he impac o clus e s on i m pe o mance in a ious indus ies and he wood indus y as well, indica ing ha clus e membe ship has a posi i e impac on i ms’ su i al, inno a ion ac i i ies and pe o mance – expo s, p oduc i i y, and sales (e.g., S øje e al., 2003; Töd ling & T ippl, 2004; Wennbe g & Lindq is , 2008; Eisinge ich e al., 2010; Abdesslem & Chiappini, 2016; S ojčić e al., 2019). On he con a y, se e al esea che s sugges ha clus e s migh also p oduce disappoin ing esul s and ha he de elopmen o clus e s is challenging in many egions and coun ies (e.g., Ande sson e al., 2004; Hsieh e al., 2012; Rašić Baka ić, 2017; Anić e al., 2019). Se e al pape s u he a gue ha he success o clus e s la gely depends on clus e managemen , go e nance and he se ices p o ided by clus e o ganisa ions (Söl ell e al., 2003; Ra inho & Hen iques, 2010; Hsieh e al., 2012; Lindq is e al., 2013; Albaha i e al., 2019). Fi ms ha e di e en mo i a ions o join clus e o ganisa ions (Sch e len e al., 2011; Obadić & Tijanić, 2014), which may p o oke con lic s and ensions among membe s (Eu opean Commission, 2013), and impac clus e pe o mance nega i ely. Di e en in e es s equi e di e en app oaches ela ed o he deli e y o clus e se ices, managemen , and go e nance. The e o e, i is impo an o unde s and he in e es s o a ious ac o s in clus e s and adjus clus e se ices o membe s’ expec a ions, which is belie ed o be he bes way o imp o e hei pe o mance in he clus e s. Su p isingly, he e is li le esea ch on his issue and no consensus exis s in he li e a u e on he deli e y o he igh se ices, al hough i is well-known ha clus e s ha e limi ed esou ces and a e o en alued by he se ices hey p o ide (Söl ell e al., 2003; Albaha i e al., 2019). Ou s udy ills he gap in he li e a u e on indus ial clus e s by in es iga ing he composi ion o se ices and i ms based on membe s’ in e es s and examining he di e ences among he g oups o i ms in a wood indus ial clus e in C oa ia. I aims a iden i ying he ypes o se ices clus e o ganisa ions should p o ide o membe s o ul il hei expec a ions. Those a e he se ices ha a e he mos alued by membe s and ep esen he incen i e o joining he clus e , and as such a e impo an o he success o he clus e . Fu he mo e, he pape de elops he axonomy o clus e membe s based on hei expec a ions abou clus e se ices and in es iga es he di e ences and simila i ies among g oups in i ms’ cha ac e is ics (size and age), hei esou ces (ne wo k openness, he access o clus e esou ces), inno a ion ac i i ies and ac ual i ms’ pe o mance, and pe cei ed clus e s’ pe o mance. The C oa ian wood clus e (he ea e CWC) was aken as he example o empi ical analysis because i is he mos signi ican clus e in he wood-p ocessing and u ni u e indus ies in C oa ia, and hus can be conside ed as he ep esen a i e case o he analysis. The con ibu ion o ou pape is he applied me hodology o examining he expec a ions o membe s o he clus e se ices in ela ion o hei business and inno a i e pe o mance. The pape o e s new e idence ela ed o he ype o se ices clus e o ganisa ions in adi ional o es -based indus ies should deli e o mee membe s’ expec a ions. We also de elop he ypology o membe s based on he pe cei ed impo ance o clus e se ices. As he heo y sugges s, i ms ace a ious challenges, do no bene i he same om clus e s and ha e di e en mo i a ions o join hem (Pe y, 2007; Ba be o e al., 2012), we expec o ind dis inc g oups o membe s wi h he same in e es s and equi emen s. Su p isingly, such esea ch is e y limi ed, al hough membe s’ expec a ions can la gely in luence he pe o mance o he clus e . Mo eo e , ou analy ical amewo k examines he di e ences and simila i ies among he g oup o membe s wi h espec o se e al a iables ha can explain he success and ailu e o indus ial clus e s, which has no been esea ched in such away. Ou inal con ibu ion is he se ing. This pape analyses he case o he C oa ian wood clus e , which is a e y small clus e in he wood-p ocessing and u ni u e adi ional indus ies ha ope a es in a e y challenging en i onmen . Namely, wood p ocessing indus ies we e hi badly by he 2008 inancial c isis, and a e a pe iod o slow eco e y, he 2020 COVID-19 c isis had a nega i e e ec on sec o s’ pe o mance (C oa ian wood clus e , 2020a). We a gue ha wood-based indus ies need o ind new g ow h oppo uni ies and enhance he alue-added o hei p oduc s using he indus ial clus e s as an e ec i e pla o m EM_2_2022.indd 60 1.6.2022 17:52:11 61 2, XXV, 2022 Business Adminis a ion and Managemen o achie ing hese goals (e.g., Ke san-Škabić, 2014; S ojčić e al., 2019). The indings o ou s udy migh help small clus e s no only in C oa ia bu also clus e s in Cen al Eas e n and Sou h-Eas e n Eu ope o gain new knowledge on how o design clus e se ices in line wi h membe s’ expec a ions mo e e ec i ely, conside ing di e en in e es s and pe o mance o he e ogeneous popula ion in clus e s. The emainde o he pape is s uc u ed as ollows. The li e a u e e iew is p esen ed in Sec ion 1, ollowed by he con ex o he s udy in Sec ion 2. Sec ion 3 desc ibes he me hodology, ollowed by he esul s in Sec ion 4, and conclusions in Sec ion 5. 1. Li e a u e Re iew Indus ial clus e s can be de ined as he geog aphic concen a ion o in e connec ed companies and ins i u ions in a pa icula ield (Po e , 1998), ha coope a e and de elop linkages wi h o he ac o s o imp o e compe i i eness (Ande sson e al., 2004; M osek e al., 2010). They a e a o m o o ganising alue chain (Hsieh e al., 2012), ocusing on c ea ing alue, which a ec s he decision o ac o s o en e and s ay in he clus e and business ne wo ks (Hazley, 2000; Hsieh e al., 2012; Albaha i e al., 2019). Acco ding o agglome a ion heo y, in clus e s economies o specializa ion, labou ma ke economies, and knowledge spillo e s d i e i ms o g oup geog aphically, educe cos s, inc ease in es men s and business pe o mance (Ma shall, 1920; Sha e & Flye , 2000; Andada i e al., 2012; Lazza e i e al., 2014; Abdesslem & Chiappini, 2016; Pecze, 2019). In clus e s, i ms ely on coope a ion, sha e he cos s, and seek join access o o eign ma ke s (A angu en e al., 2013). A special ype o clus e is a clus e o ganisa ion wi h es ablished clus e managemen and o mal membe ship. Such clus e o ganisa ions we e es ablished in se e al indus ies and ha e been suppo ed wi h he aim o sol e coo dina ion ailu es and inc ease he compe i i eness o i ms/ egions/sec o s by acili a ing public-p i a e collabo a ion, inno a ions, knowledge, and echnology ans e (e.g., Ma ioli e al., 2016). Such clus e o ganisa ions encompass a ious ac o s, including i ms and suppo ing ins i u ions, educa ion and esea ch o ganiza ions, capi al p o ide s, go e nmen , and public ins i u ions (Lindq is e al., 2013). This o m o he clus e is he mos alued by he p o ision o se ices, which help membe s o o e come he obs acles hey encoun e and inc ease he compe i i eness in such a way (Lindq is e al., 2013). Common se ices p o ided by clus e o ganisa ions a e lobbying and c ea ing a dialogue be ween ac o s, p o iding suppo o ne wo king comme cial coope a ion, inno a ion and echnology ans e , and human esou ces upg ading (Söl ell e al., 2003; Lindq is e al., 2013). Clus e s help i ms o access clus e esou ces mo e e ec i ely, which in luences business pe o mance (He ás-Oli e & Albo s- Ga igós, 2007; P im e al., 2016). Clus e s a e also o en seen as poli ical o ganisa ions ha engage in lobbying wi h local, na ional and EU go e nmen s o a icula e clus e needs, ob ain a be e posi ion on he ma ke and egula o y p o ec ion o he sec o (Pe y, 2007; Sch e len e al., 2011). One s eam o li e a u e has examined he e ec s o clus e de elopmen p og ams and clus e o ganisa ions on i ms in he clus e , measu ing i s e ec s on coo dina ion and linkages, esou ce alloca ion and in es men decisions, business p ac ices and echnologies, and business pe o mance. Clus e e ec s can be sho - and long- e m (Ma ioli e al., 2016). Mos pape s sugges ha he e ec s o clus e membe ship on i ms a e posi i e (e.g., S øje e al., 2003; Töd ling & T ippl, 2004; Wennbe g & Lindq is , 2008; Eisinge ich e al., 2010; Abdesslem & Chiappini, 2016; S ojčić e al., 2019), al hough he e a e s udies indica ing ha clus e s migh also ail (e.g., Ande sson e al., 2004; Rašić Baka ić, 2017). One eason o his migh be he con lic ing in e es s o membe s ha a e no conside ed e icien enough by clus e managemen and go e nance. Li e a u e also sugges s ha i ms ha e a ious in e es s o join he clus e s. To succeed clus e o ganisa ions mus ul il hei expec a ions. Clus e se ices we e shown o be pa icula ly in e es ing o small- and medium-sized companies ha ha e limi ed esou ces because h ough clus e s hey migh educe ansac ion cos s, gain new knowledge and echnology, and suppo hei businesses h ough lobbying (S ojčić e al., 2019). Fi ms migh be also mo i a ed o pa icipa e in clus e s o imp o e hei access o inpu s, skilled wo ke s, public in as uc u e, and esea ch, and inc ease he abili y o inno a e and dec ease cos s (Po e , 2000; Pecze, 2019). EM_2_2022.indd 61 1.6.2022 17:52:11 62 2022, XXV, 2 Business Adminis a ion and Managemen Small companies migh seek o imp o e hei ma ke posi ion h ough coope a ion wi h mo e es ablished i ms, while i ms om high-g ow h eme ging indus ies a e mos ly inno a ion-o ien ed and migh look o hese se ices when joining he clus e s (Anić e al., 2019). On he o he hand, la ge i ms, as compa ed o small ones and i ms ope a ing in declining, old, and adi ional indus ies, we e shown o be he mos in e es ed in lobbying (Fon agné e al., 2010; Ma in e al., 2011; Abdesslem & Chiappini, 2016), and migh join clus e s o achie e his (Anić e al., 2019). Fo i ms ope a ing in unce ain en i onmen s (Eisinge ich e al., 2010; M osek e al., 2010; Andada i e al., 2012), os e ing ne wo king and coope a ion, especially wi h o eign i ms, is c ucial. S ong ne wo ks acili a e o ganisa ional lea ning and enhance inno a ion po en ial by p o iding he access o esou ces h ough in e ac ions, which maximizes he syne gies, inc eases us and knowledge ans e (Pikul- Biniek, 2009; Eisinge ich e al., 2010). Al hough wood-p ocessing and u ni u e indus ies belong o low- ech indus ies (O ganisa ion o Economic Coope a ion and De elopmen , 2011) wi h a low le el o inno a i eness, se e al pape s epo ha some i ms ope a ing in he wood sec o a e also inno a i e and use mode n echnologies, whe eas clus e membe ship has a posi i e e ec on inno a i eness (Hazley, 2000; Uk ainski & Va blane, 2005; Ng e al., 2012; G zego zewska e al., 2014). One s eam o he li e a u e sugges s ha he membe s’ common and p i a e in e es s impac he alliance’s pe o mance. Common bene i s a e gene a ed by collec i e p ocesses sha ed by all membe s, while p i a e bene i s e e o he dis ibu ion o common bene i s o indi idual pa ne s. I he e is a highe a io o common in e es s o p i a e in e es s in an alliance, beha iou end o be coope a i e and esou ce alloca ion will con e ge o op imal (e.g., Khanna e al., 1998; La ie, 2007). This concep can be also applied o clus e s. F om he li e a u e, i is, howe e , no clea how clus e se ices a e aligned wi h he in e es s o a ious membe s in he clus e , and how is his ela ed o di e ences among membe s in pe o mance. 2. S udy Con ex The o es sec o consis s o e ically linked indus ies and ac i i ies, anging om o es y, and wood supply o indus y, o es -based indus ial p oduc ion, and ade (Hazley, 2000; M osek e al., 2010). This sec o , and in pa icula wood-p ocessing (C16) and u ni u e indus ies (C31) a e e y impo an o he C oa ian economy, as hey accoun ed in 2017 o 62.7% o he g oss alue added o he wood sec o ; 13.3% o i ms and 10.7% o employmen in manu ac u ing indus y (Eu os a , 2020). These indus ies consis mos ly o mic o, small and medium-sized companies. The e we e only 15 la ge companies wi h mo e han 250 employees in he sec o . C oa ian wood-p ocessing and u ni u e indus ies we e ha dly hi by he 2008 inancial c isis. In he pos -c isis 2011–2017 pe iod, he numbe o companies dec eased by 11.8%, bu employmen g ew by 5.9%. The wood sec o in C oa ia is expo -o ien ed and la gely depends on access o in e na ional ma ke s. The da a show ha du ing ha pe iod p oduc ion alue inc eased by 64.4%, and expo s by 65%. Howe e , he expo s o he wood sec o consis p edomina ely o aw wood ma e ial and semi- inished goods, low alue- added p oduc s. The sha e o wood-p ocessing and u ni u e indus ies in g oss alue added was in C oa ia only 0.9%, as compa ed o 3.4% o he EU a e age (Eu os a , 2020; C oa ian Bu eau o S a is ics, 2012–2020). COVID-19 pandemic p o oundly hi he C oa ian wood- p ocessing indus y. O icial da a o he C oa ian Bu eau o S a is ics show ha in 2020 in his sec o indus ial p oduc ion dec eased by 6.7% and expo s by 7.9%. Clus e s a e an impo an ool o he C oa ian de elopmen policy (Obadić & Tijanić, 2014; S ojčić e al., 2019), which objec i es a e o s eng hen he de elopmen o clus e s, suppo expo s, and ensu e e icien use o EU unds (C oa ian go e nmen , 2011). The main goal o he o ma ion o indus ial clus e s is o acili a e inno a ions, ans e o echnologies, compe i i eness, and g ow h. Exis ing clus e s in C oa ia a e e y small and wo k as legal en i ies in ol ing clus e o ganisa ions o p omo e join p i a e-public in e es s (C oa ian go e nmen , 2011). The C oa ian wood clus e is one o he mos impo an indus ial clus e s in C oa ia. I ope a es in he wood sec o and was o mally es ablished in 2013 as an associa ion wi h olun a y membe ship. The clus e expanded om 61 membe s in 2017 o 77 membe s in 2020. Mos i ms (78%) a e small and EM_2_2022.indd 62 1.6.2022 17:52:11 63 2, XXV, 2022 Business Adminis a ion and Managemen medium-sized companies ha ope a e in he wood-p ocessing and u ni u e indus ies. O he membe s a e p o ide s o specialized inpu s, machine y and associa ed se ices, and he e a e se en esea ch and educa ional o ganisa ions, and echnology cen es (C oa ian wood clus e , 2020b). The main objec i es o CWC a e s eng hening he compe i i eness o he whole wood-p ocessing sec o by encou aging ac i i ies in he ield o esea ch and de elopmen , echnology de elopmen , comme cializa ion o inno a ion and in as uc u e in es men s, and o he ela ed ac i i ies. The clus e p o ides se e al se ices o i s membe s, including access o public suppo and lobbying, ad iso y se ices, acili a ion o collabo a ion be ween membe s and c oss-sec o ial coope a ion, p omo ion o clus e and ma ke ing ac i i ies (Eu opean clus e collabo a ion pla o m, 2020). CWC is e y ac i e in ma ke ing, o ganizing sec o al e en s and in e na ional con e ences, s udy isi s, aining, and p o iding help ela ed o EU p ojec s and e icien use o EU unds. CWC is inanced h ough membe ship ees, p ojec s, and he o ganisa ion o di e en e en s (C oa ian wood clus e , 2020b). 3. Resea ch Me hodology The ques ionnai e was ca ied ou in 2017 among CWC membe s o collec he da a. I was de eloped based on li e a u e e iew and ace- o- ace in e iews wi h clus e managemen and p o essionals. P e- es ing included i e esponden s. The da a was collec ed using Compu e -Assis ed Web In e iewing me hod (CAWI) om May o Sep embe 2017. The ques ionnai e was sen o 61 membe s o CWC (53 i ms we e e alua ed as ac i e membe s o clus e ). The a ge g oup we e owne s o he companies and p o essionals in manage ial posi ions, who we e able o espond o he ques ionnai e. Addi ionally, a p o essional in e iewe was hi ed o emind he esponden s o comple e he ques ionnai e. A e sending 3 eminde s, 39 ques ionnai es we e collec ed, which makes a esponse a e o 63.9%. A e elimina ing 2 duplica es, 37 ques ionnai es we e usable o he analysis. The cha ac e is ics o ou sample a e p esen ed in Tab. 1. The sample consis s mos ly o i ms ope a ing in C16 and C31 indus ies (78.4%). Suppo ing indus ies in he sample a e he manu ac u ing o me al p oduc s indus ies, ene gy supply, ade sec o s, consul ing ac i i ies, and p o essional o ganisa ions. Membe s we e p edominan ly small companies (51.4%), p i a e companies (97.3%), and domes ically owned companies (91.9%). On a e age, hey we e ela i ely young (22 yea s) and had sales e enues o HRK 52.0 million. The da a was analysed using explo a o y ac o analysis (EFA), clus e analysis, one- way analysis o a iance (ANOVA) and chi- squa e es . Since he numbe and names o ac o s we e no known p io o esea ch, he use o ac o analysis in his s udy was pu ely explo a o y and i was pe o med o iden i y a smalle se o a iables wi hin he da ase . As clus e ing a iables, we used membe s’ pe cep ions o he impo ance o clus e se ices. In he ques ionnai e, we Cha ac e is ics n % Main company ac i i y, % 16 Manu ac u e o wood and o p oduc s o wood and co k 23 62.2 17 P oduc ion o pape and pape p oduc s 1 2.7 31 Manu ac u e o u ni u e 6 16.2 Suppo ing ac i i ies 7 18.9 Company size Mic o companies 4 10.8 Small-sized companies 19 51.4 Medium-sized companies 14 37.8 Sou ce: own; Poslo na H a ska Tab. 1: Cha ac e is ics o he sample, n = 37 EM_2_2022.indd 63 1.6.2022 17:52:11 64 2022, XXV, 2 Business Adminis a ion and Managemen asked he esponden s o a e he impo ance o 25 ac i i ies ha CWC should pu sue and coo dina e o hei membe s on a Like scale, anging om 1 ( o ally unimpo an se ice) o 5 (ex emely impo an se ice). These ac i i ies ep esen he se ices and p io i y objec i es CWC should ca y o achie e i s goals ela ed o compe i i eness and sec o g ow h. As such, hey also ep esen he d i e s o clus e de elopmen . These i ems we e aken om he Annual epo o he C oa ian wood clus e (2014), Söl ell e al. (2003), and Lindq is e al. (2013), while some i ems we e de eloped by au ho s a e consul a ion wi h p o essionals in he CWC clus e . We analysed wo ca ego ies o pe o mance as esponse a iables – ac ual i ms’ sales g ow h ( i ms’ g ow h) and pe cei ed impac o CWC on membe s’ pe o mance (clus e impac ). Fo measu ing membe s’ ac ual pe o mance, we used he da a on sales e enues o he pe iod o 2015–2017 om he business po al Poslo na H a ska. Fi ms’ sales g ow h was coded as 0 ( he i m had a decline in sales o a smalle , insigni ican sales g ow h), and 1 ( he i m had sales g ow h mo e han he median). To cap u e he inno a ion ac i i y o membe s, esponden s we e asked o indica e how many new p oduc s and se ices hey in oduced o he ma ke in he las h ee yea s, and hen companies we e g ouped and coded as 0 (ze o inno a ions) and 1 ( he company in oduced a leas one new p oduc ). Rega ding pe cei ed clus e e ec s (clus e impac ), esponden s indica ed how much CWC a ec ed he pe o mance o i ms conce ning in e nal business p ocesses, ma ke pe o mance, collabo a ion, us , ans e o knowledge and new echnologies on a Like scale anging om 1 (s ongly disag ee) o 5 (s ongly ag ee). Those i ems we e de eloped by he au ho s. Explo a o y ac o analysis (EFA) was used o dec ease he numbe o pe o mance i ems. Th ee ac o s we e gene a ed, labelled as clus e impac on in e nal business pe o mance and inno a ions; clus e impac on ma ke pe o mance; and clus e impac on coope a ion, which we e used in u he analysis (see Tab. A3 in Appendix). To u he examine he cha ac e is ics o he i ms in he clus e , we used se e al addi ional a iables, including age and size o he i ms, he deg ee o membe s’ ne wo king, access o clus e esou ces, and he in oduc ion o new p oduc s and se ices. Ques ions ela ed o he deg ee o ne wo king we e aken om Eisinge ich e al. (2010). The esponden s e alua ed he deg ee o hei coope a ion wi h o he ac o s in he clus e (ne wo k s eng h) and he deg ee o ex e nal linkages exis ing ou side he clus e (ne wo k openness) on he Like scale, anging om 1 (s ongly disag ee) o 5 (s ongly ag ee). Responden s we e also asked o indica e how much local ins i u ional and ne wo k clus e esou ces con ibu e o hei compe i i eness. These i ems we e aken om P im e al. (2016), we e measu ed by a Like scale anging om 1 (no a all) o 5 (comple ely), and we e analysed sepa a ely as he access o ins i u ional, local, and business esou ces. Those i ems show how much clus e esou ces a e use ul o hem. I ems a e p esen ed in Tab. A1 and Tab. A2 in Appendix. 4. Empi ical Analysis Explo a o y ac o analysis was pe o med on 25 i ems indica ing he pe cei ed impo ance o clus e se ices. Due o an ex emely small sample, we emo ed ques ions ha had signi ican c oss-loadings and/o low loadings (Hai e al., 2009). Fu he analysis con inued wi h en i ems. Kaise Meye Olkin’s measu e con i med he sampling adequacy, whe eas Ba le ’s es o sphe ici y sugges s ha co ela ions be ween i ems we e adequa e o conduc ing ac o analysis (KMO = 0.749; chi- squa ed = 254.724; deg ees o eedom = 45; p < 0.000) (Hu cheson & So oniou, 1999; Field, 2009). The p incipal componen analysis and Va imax o a ion we e used, conside ing only ac o s wi h Eigen alue g ea e han 1 in cases when he ob ained ac o s a e used as he basis o calcula ing ac o sco es as inpu a iables o u he analysis (Johnson & Wiche n, 1992). Th ee ac o s eme ged om he analysis labelled acco ding o he dominan i ems as ollows: 1) join logis ics and ma ke access; 2) lobbying; and 3) ne wo king and coope a ion. A h ee- ac o solu ion wi h ac o loadings anging om 0.64 o 0.95 explained 80.5% o he o al a iance. C onbach’s alpha coe icien s we e used o assess in e nal consis ency and sugges ha he i ems p oduced adequa e and high sco es o hose scales (C onbach’s alpha o Fac o 1 = 0.896; C onbach’s alpha o Fac o 2 = 0.858; C onbach’s alpha o Fac o 3 = 0.843). Tab. 2 shows ac o loadings a e o a ion. EM_2_2022.indd 64 1.6.2022 17:52:12 65 2, XXV, 2022 Business Adminis a ion and Managemen The join logis ics and ma ke access ac o accoun s o 30.4% o he o al a iance and is explained by ou i ems ela ed o join wa ehousing and join anspo a ion, bu also by he coo dina ion o public-p i a e in es men s and join selling ac i i ies. Lobbying, he second ac o , explains 27.6% o he a iance and includes i ems ela ed o he p omo ion o he indus y h ough coope a ion wi h supplie s o inpu s (i.e., lobbying a supplie s o inpu s) and lobbying o laws and indus y in e es s a he na ional go e nmen . The ne wo king and coope a ion ac o explains 22.5% o he a iance and includes ne wo king and p omo ion o clus e s h ough con e ences, ai ades, clus e ’s webpage, join ad e ising in media, and membe ship in leading associa ions. The mos impo an se ice appea ed o be lobbying, ollowed by ne wo king and coope a ion, whe eas join logis ics and ma ke access seem o be less impo an . We pe o med hie a chical and non- hie a chical clus e analyses o iden i y he g oups o i ms wi h simila goal o ien a ions. The ac o sco es o each o he h ee ac o s we e calcula ed and hen submi ed o hie a chical clus e analysis o a oid mul icollinea i y (Hai e al., 2009). The heo y sugges s he applica ion o o hogonal ac o o a ion – a imax o a ion in cases when he ob ained ac o s a e used as he basis o calcula ing ac o sco es as inpu a iables o u he analysis (Johnson & Wiche n, 1992). Acco ding o he heo y, a ac o sco e indica es he ex en o which each i m has a high sco e on a g oup o cha ac e is ics ha ha e a high loading on a ele an ac o . Wa d’s me hod wi h he Squa ed Euclidean dis ance was applied o iden i y he numbe o clus e s (g oups) wi hin he sample o membe s. A e we explo ed he agglome a ion ma ix and he dend og am, a ou -clus e solu ion was iden i ied o be app op ia e. A e ha , he non-hie a chical K-means clus e algo i hm con i med ou g oups o membe s wi h dis inc and signi ican di e ences be ween g oups on he pe cei ed impo ance o clus e se ices. G oups we e named acco ding o he dominan ea u es as ollows: G oup 1) Coope a ion- and lobbying- o ien ed; G oup 2) Pu ely ma ke -o ien ed; G oup 3) Pu ely lobbying-o ien ed; G oup 4) Pu ely ne wo king- and coope a ion-o ien ed. Tab. 3 and Tab. 4 show he esul s. G oup 1 (coope a ion- and lobbying- o ien ed) is he la ges segmen and has high ac o sco es on ne wo king, coope a ion, and lobbying ac o s. Membe s in his g oup a e he mos in e es ed in ne wo king and coope a ion and pe cei e lobbying ac i i ies wi h he go e nmen and supplie s as e y impo an . G oup 2 (ma ke -o ien ed) is he second- Sha ed clus e se ices Fac o 1: Join logis ics and ma ke access Fac o 2: Lobbying Fac o 3: Ne wo king and coope a ion Join wa ehousing 0.9487 0.0496 0.0570 Join anspo a ion 0.9390 0.1263 0.0292 Co-o dina ing public-p i a e in es men s 0.7477 0.1844 0.4178 Join selling ac i i ies 0.7305 0.3681 0.0435 Coope a ion o he indus y wi h supplie s o inpu s 0.0420 0.8560 0.2732 Lobbying o he sec o a he go e nmen 0.1460 0.8422 0.2024 Making p oposals and lobbying o laws a he go e nmen 0.3116 0.8377 0.1580 Ne wo king and p omo ion o clus e 0.1220 0.0647 0.8941 Coope a ion and exchange o expe ience wi h pa ne s 0.0382 0.3865 0.8522 Sou ce: own Tab. 2: Fac o analysis on he pe cei ed impo ance o sha ed clus e se ices and ac o loadings, n = 36 EM_2_2022.indd 65 1.6.2022 17:52:12 66 2022, XXV, 2 Business Adminis a ion and Managemen la ges g oup ha includes membe s who ha e ela i ely high sco es on join ma ke -based ac i i ies, including join logis ics and selling p oduc s, and equi e om CWC o acili a e hese ac i i ies. The hi d and he ou h g oups con ain a e y small numbe o membe s. While membe s in he hi d g oup (pu ely lobbying- o ien ed) a e in e es ed in lobbying signi ican ly mo e han in o he ac i i ies, he ou h g oup (pu ely ne wo king and coope a ion-o ien ed) a e he mos in e es ed in he acili a ion o ne wo king and coope a ion ac i i ies. To examine he di e ences and simila i ies among g oups we used ac ual i ms’ g ow h pe o mance, inno a ion ac i i ies and clus e impac as dependen a iables, while ou g oups o CWC membe s we e aken as independen a iables in Pea son’s chi-squa ed es χ2 analysis and one-way analysis o a iance (ANOVA). Findings a e p esen ed in Tab. 5 and Tab. 6. The di e ences among g oups we e signi ican in sales g ow h and inno a ion ac i i ies. Membe s o G oup 1 a e mo e o ien ed owa ds inno a ion ac i i ies and mos o hem a e high-g ow h i ms. G oup 2 ollows in e ms o he equency o high-g ow h i ms and i ms ha pu sue inno a ion ac i i ies in he o es sec o . G oup 3 and G oup 4 a e he wo s ega ding he p esence o i ms ha pu sued inno a ion ac i i ies, and he e a e no i ms ha expe ienced as g ow h in he p e ious pe iod. Cons uc s G oup 1: (n = 16) G oup 2: (n = 11) G oup 3: (n = 5) G oup 4: (n = 4) Join logis ics and ma ke access −0.18 0.86 −0.87 −0.57 Lobbying 0.54 −0.30 0.42 −1.88 Ne wo king and coope a ion 0.71 −0.50 −1.77 0.74 Sou ce: own One-way ANOVA G oup 1: (n = 16) G oup 2: (n = 11) G oup 3: (n = 5) G oup 4: (n = 4) Sample (n = 36) F P Join logis ics and ma ke access 3.16 (1.01) 3.86 (0.7) 2.05 (1.05) 2.25 (0.20) 3.12 (1.06) 6.45 0.002 Lobbying 4.50 (0.46) 3.64 (0.74) 3.67 (1.18) 2.17 (0.84) 3.86 (0.99) 12.34 0.000 Ne wo king and coope a ion 4.58 (0.38) 3.39 (0.49) 2.33 (0.78) 3.75 (0.56) 3.75 (0.93) 30.11 0.000 Sou ce: own Chi-squa ed es G oup 1: (n = 16) G oup 2: (n = 11) G oup 3: (n = 5) G oup 4: (n = 4) Sample (n = 36) χ2 d. . P Sales g ow h* 56.25 36.36 0 0 38.24 6.548 30.088 New p oduc s in oduced** 85.71 90.0 33.33 0 73.33 13.222 30.004 Sou ce: own No e: * signi ican a 0.1 le el; **signi ican a 0.05 le el. Tab. 3: Cha ac e is ics o he g oups o CWC membe s, a e age ac o sco es Tab. 4: P o ile o he ou g oups o CWC membe s, mean alues, M (SD) Tab. 5: Pe o mance o g oup membe s in CWC, he pe cen age o companies EM_2_2022.indd 66 1.6.2022 17:52:13 67 2, XXV, 2022 Business Adminis a ion and Managemen Rega ding clus e impac , signi ican di e ences exis only in he impac o he clus e on ma ke pe o mance, whe eas g oups do no di e in e ms o clus e impac on in e nal business pe o mance, inno a ion, and coope a ion ac o . In gene al, membe s pe cei e ha he con ibu ion o CWC o pe o mance was e y low and ha he clus e did no do enough o help hem imp o e hei pe o mance. Membe s in G oup 1 pe cei e clus e e ec s on a ma ke pe o mance be e han he o he g oups, whe eas he membe s in G oup 3 – pu ely lobbying-o ien ed, had he wo s opinion abou CWC. To examine o he cha ac e is ics o g oups, se e al a iables impo an o clus e success we e conside ed in he analysis, including age and size o he i m, he deg ee o ne wo king and access o clus e esou ces. Resul s a e shown in Tab. 7 and Tab. 8. Clus e impac G oup 1: (n = 16) G oup 2: (n = 11) G oup 3: (n = 5) G oup 4: (n = 4) Sample (n = 36) F P In e nal business pe o mance and inno a ions 2.47 2.36 1.74 2.15 2.29 0.72 0.547 Ma ke pe o mance** 2.40 1.92 1.13 2.29 2.07 2.98 0.047 Coope a ion 2.91 2.29 2.33 2.83 2.64 1.10 0.36 Sou ce: own No e: ** signi ican a 0.05 le el. Chi-squa ed es G oup 1: (n = 16) G oup 2: (n = 11) G oup 3: (n = 5) G oup 4: (n = 4) Sample (n = 36) χ2 d. . P Younge i ms 50.00 54.55 60.00 25.00 50.00 1.291 30.731 Small i ms 56.25 72.73 60.00 75.00 63.89 4.191 6 0.651 Sou ce: own No e: The age o he company was spli in o wo hal es ep esen ing below he a e age alue and abo e-a e age alue, ca ego ized as 1 (younge i ms up o 22 yea s, below-a e age alue) and 2 (olde i ms wi h mo e han 22 yea s, abo e-a e age alue). Size includes he pe cen age o small and mic o i ms. One-way ANOVA G oup 1: (n = 16) G oup 2: (n = 11) G oup 3: (n = 5) G oup 4: (n = 4) Sample (n = 36) F P Ne wo k s eng h** 3.35 2.83 2.73 3.08 3.09 2.77 0.050 Ne wo k openness** 3.48 3.03 2.55 3.50 3.22 4.23 0.013 Ins i u ional esou ces* 3.31 2.87 2.00 2.92 2.94 2.67 0.065 Local esou ces** 3.19 2.86 1.48 2.75 2.76 4.46 0.011 Business esou ces** 3.21 2.83 1.83 2.75 2.84 3.86 0.019 Sou ce: own No e: * signi ican a 0.1 le el, ** signi ican a 0.05 le el. Tab. 6: Pe cei ed impac o CWC on he pe o mance o companies by g oups o clus e s wi h mean alues, ANOVA Tab. 7: Cha ac e is ics o g oup o CWC membe s in e ms o i m size and age, he pe cen age o i ms Tab. 8: Cha ac e is ics o g oups o CWC membe s in e ms o he deg ee o ne wo king and clus e esou ces, mean alues EM_2_2022.indd 67 1.6.2022 17:52:13 74 2022, XXV, 2 Business Adminis a ion and Managemen I ems desc ip ion Fac o 1: Impac on in e nal business pe o mance and inno a ions Fac o 2: Impac on ma ke pe o mance Fac o 3: Impac on coope a ion Sho e ime o esponse o clien s’ equi emen s. 0.9241 0.1783 0.1228 Imp o ed quali y o p oduc s and se ices. 0.8897 0.1318 0.2092 Mee ing echnological s anda ds. 0.8892 0.0459 0.3293 Imp o ed access o dis ibu ion channels and supply. 0.8799 0.1893 0.1042 Sha ing cos s and esou ces. 0.8676 0.1472 0.0557 In oduc ion o new me hods o p oduc s in oduc ion and dis ibu ion channels. 0.8578 0.0975 0.2754 De elopmen o new p oduc s and se ices. 0.8487 0.1873 0.3517 Dec easing he isk o di e si ica ion 0.8308 0.2373 0.0924 In oduc ion o new me hods o se ing p ices o goods and se ices. 0.8223 0.2807 0.2521 Imp o ed access o new echnology. 0.8184 0.0612 0.4140 Clus e has helped he company inc ease e enues. 0.1521 0.9181 0.1934 Clus e has led o inc eased employmen . 0.2147 0.9110 0.1154 Clus e has imp o ed in e na ional compe i i eness o company. 0.0136 0.8997 0.0449 Clus e has p omo ed expo o highe alue-added p oduc s. 0.1065 0.8552 0.2012 Clus e has inc eased FDI. 0.3653 0.8276 0.2375 The clus e has led o inc eased collabo a ion wi h in e na ional companies wi hin global alue chains. 0.1203 0.8198 0.2875 Clus e has led o close indus y- academia ies. 0.2013 0.3019 0.8443 The coope a ion wi h o he clus e membe s has led o highe le el o us . 0.3846 0.2780 0.7904 Clus e has led o close ies wi h o he companies wi hin he clus e . 0.3841 0.3375 0.7422 Sou ce: ques ionnai e No e: Th ee ac o s explain 83.7% o a iance. C onbach’s alpha o clus e impac on in e nal business pe o mance = 0.978; C onbach’s alpha o clus e impac on ma ke pe o mance = 0.958; C onbach’s alpha o impac on coope a ion = 0.892; KMO = 0.783; chi-squa ed = 763.433; deg ees o eedom = 171; p- alue = 0.000. Tab. A3: Explo a o y ac o analysis on clus e impac i ems EM_2_2022.indd 74 1.6.2022 17:52:15