Explaining the freedom of information in local government: what influences information disclosure?

www.cya.unam.mx/index.php/cya Con adu ía y Adminis ación 65 (4), 2020, 1-29 1 Accoun ing & Managemen *Co esponding au ho . E-mail add ess: F ancisca.T[email p o ec ed] (F. Tejedo Rome o). Pee Re iew unde he esponsibili y o Uni e sidad Nacional Au ónoma de México. h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 0186- 1042/©2019 Uni e sidad Nacional Au ónoma de México, Facul ad de Con adu ía y Adminis ación. This is an open access a icle unde he CC BY-NC-SA (h ps://c ea i ecommons.o g/licenses/by-nc-sa/4.0/) Explaining he eedom o in o ma ion in local go e nmen : Wha in luences in o ma ion disclosu e? Explicando el lib e acceso a la in o mación en el gobie no local: ¿qué in luye en la di ulgación de in o mación? F ancisca Tejedo-Rome o 1*, Joaquim Filipe Fe az Es e es A aujo2 Abs ac In esponse o he g owing demand om socie y o access public in o ma ion, go e nmen s ha e been o ced o be mo e anspa en in he conduc o hei ac i i ies. Based on he li e a u e o in o ma ion disclosu e, his pape analyses ac o s ha in luenced he eedom o in o ma ion in Spanish municipa- li ies, and which explain he le els o in o ma ion disclosu e. Fo he pu pose o his s udy, a dynamic panel Sys em Gene al Me hod o Momen s (Sys em-GMM) es ima ion was used o analyses da a om 100 Spanish la ges municipali ies o he pe iod o 2008-2014. Addi ionally, he sample was pa i ioned in wo ca ego ies acco ding o he p o ince capi al o no p o ince capi al c i e ia in o de o ind i he e a e in e es ing di e ences. The analysis indica es ha disclosu e o in o ma ion abou municipal Recei ed No embe 17, 2018; accep ed Oc obe 17, 2019 A ailable online Oc obe 21, 2019 1Depa amen o de Adminis ación de Emp esas, Uni e sidad de Cas illa-La Mancha, España 2Uni e sidade do Minho-Escola de Economia e Ges ao, Po ugal F. Tejedo-Rome o, e al. / Con adu ía y Adminis ación 65(4) 2020, 1-29 h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 2 managemen is associa ed wi h inancial dependence, in e ne access, unemploymen , elec o al u nou and poli ical i al y. Di e ences ound in he wo ca ego ies o municipali ies sugges ha he e may be ins i u ional cha ac e is ics ha could in luence he dissemina ion o in o ma ion. This s udy is a con ibu ion o he g owing body li e a u e o in o ma ion disclosu e in o de o unde s and wha explains he a ia ions o in o ma ion disclosu e among he municipali ies. JEL codes: H70, H79, H80 Keywo ds: In o ma ion disclosu e; T anspa ency; Local go e nmen ; F eedom o in o ma ion; Municipali ies Resumen En espues a a la c ecien e demanda de la sociedad pa a accede a la in o mación pública, los gobie nos se han is o o zados a se más anspa en es en el desa ollo de sus ac i idades. Basado en la li e a u a de di ulgación de in o mación, es e a ículo analiza los ac o es que in luye on en el lib e acceso a la in o mación en los municipios españoles y que explican los ni eles de di ulgación de in o mación. A los e ec os de es e es udio, se u ilizó la es imación del Mé odo Gene alizado de Momen os (GMM) de Sis ema (Sys em-GMM) de da os de panel en un modelo dinámico pa a analiza los da os de 100 municipios españoles más g andes pa a el pe íodo 2008-2014. Además, la mues a se ha di idido en dos ca ego ías según el c i e io de si los municipios son capi al de p o incia o no pa a de e mina si exis en di e encias in e esan es. El análisis indica que la di ulgación de in o mación sob e la ges ión municipal es á asociada con la dependencia inancie a, el acceso a in e ne , el desempleo, la pa icipación elec o al y la i alidad polí ica. Las di e encias encon adas en las dos ca ego ías de municipios sugie en que puede habe ca ac e ís icas ins i ucionales que pod ían in lui en la di ulgación de in o mación. Es e es udio es una con ibución a la c ecien e li e a u a sob e di ulgación de in o mación con la inalidad de comp ende qué explica las a iaciones de di ulgación de in o mación en e los municipios. Código JEL: H8, H79 Palab as cla e: Di ulgación de in o mación; T anspa encia; Gobie no local; Lib e acceso a la in o mación; Municipios In oduc ion In ecen decades, he p essu e o make public o ganiza ions mo e anspa en and he F ee- dom o In o ma ion laws1 (FOI), al e ed he adi ional cul u e o Public Adminis a ion and inc eased in o ma ion disclosu e, ein o cing he abili y o he public o access go e nmen in o ma ion (Pio owski & Van Ryzin, 2007). 1 The F eedom o In o ma ion laws a e egula ions which gi es e e y ci izen he igh o eques ce ain in o - ma ion held by he public bodies and o ganisa ions, including he go e nmen , egula o s and any public se ice companies. F. Tejedo-Rome o, e al. / Con adu ía y Adminis ación 65(4) 2020, 1-29 h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 3 In o ma ion disclosu e is a mechanism o imp o e he public us , and hold poli icians and o icials accoun able o hei ac ions and ou comes (Pa k & Blenkinsopp, 2011). I has inc eased wo ldwide in he public sec o unde he designa ion o anspa ency and i is an indi ec e ec p omo ed by New Public Managemen and New Public Go e nance (Gonzalez, Gasco & Llopis, 2007; Osbo ne, 2006), by opening public o ganiza ions, and inc easing he mechanisms o communica ion, accoun abili y and anspa ency. Conce ning he disclosu e o municipal in o ma ion, i is wo hwhile o build up i s unde s anding (G eco, Sciulli & D’onza, 2012). E icien use o sca ce esou ces equi es e ec i e mechanisms o dissemina e in o ma ion abou municipali ies’ decision-making, o imp o e public sec o communica ion and con idence. The issues in luencing in o ma ion disclosu e a e complex and been he esul o in e nal and ex e nal ac o s o he municipali ies. P e ious esea ch has indica ed ha in o ma ion disclosu e in local go e nmen s depend on poli ical and socio-economic ac o s (Albala e, 2013; Tejedo-Rome o & A aujo, 2018). O he s udies analyse in e nal ac o s like cul u e (Lopes & Se asquei o, 2017). Howe e , he e is s ill ambigui y in esul s and esea che s canno make consis en conclusions (Cucciniello, Po umbescu & G immelikhuijsen 2017). Maybe he inconsis ency o hese esul s lies in he design o he s udies (Alcaide-Muñoz, Rod íguez-Bolí a , & López-He nández, 2017), o in he implemen a ion o in o ma ion dis- closu e policies in coun ies wi h di e en adminis a i e cul u e whose in e nal and ex e nal ac o s o he municipali ies in luence in o ma ion disclosu e (Rod íguez, Ga cía & Gallego, 2011). Howe e , he e is s ill insu icien empi ical e idence o clea ly es ablish he in luence o such ac o s in in o ma ion disclosu e (Alcaide-Muñoz e al., 2017). Lowa cha in and Meni ield (2015) poin ou he need o u he de elopmen conce ning he ac o s in luencing he le els o anspa ency. In his ega d, Cucciniello e al. (2017) p opose he way o wa d sugges ing o academics o do simila s udies in o de o sys ema ically iden i y con ex ual condi ions ha a ec anspa ency ou comes, and conduc analyses using di e en app oaches. Resea ch shows he e a e impo an ac o s ha explain he le el o in o ma ion disclosu e in municipali ies (Alca az-Quiles, Na a o-Gale a & O iz-Rod íguez, 2015; Pio owski & Van Ryzin, 2007). In he case o Spanish municipali ies, i is wo h o analyse in o ma ion disclosu e h ough in e ne communica ion o acili a e ac i e in o ma ion, since in e ne has been an impo an ool o in o ma ion disclosu e. This s udy uses he anspa ency index as a p oxy o in o ma ion disclosu e, and iden i y he ac o s ha a ec in o ma ion disclosu e, in pa icula he egional e ec on anspa ency. The choice o conduc ing his analysis in Spanish municipali ies has o do wi h hei p oblems wi h in o ma ion disclosu e, which a e mo e equen a local le el (Cuad ado-Balles e os, 2014). Fi s ly, because social p essu e exe ed by ci izen mo emen s, as he case o he “los indignados”, compels municipali ies F. Tejedo-Rome o, e al. / Con adu ía y Adminis ación 65(4) 2020, 1-29 h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 4 o imp o e managemen e iciency, and be mo e anspa en in he managemen o public esou ces. Secondly, he inancial c isis p essu ed o mo e anspa ency in he managemen o public esou ces, pa icula ly hose unds come om cen al o egional ans e s. In bo h cases, in e ne access, as a oll o in o ma ion disclosu e, played an impo an ole. Hence, in o ma ion disclosu e and anspa ency may ha e an impo an ole wi hin he communica ion p ocess by opening o ganiza ions o esponsi e communica i e p ac ices owa ds ci izens. This pape con ibu es o bo h he academic and p ac ical ield on in o ma ion disclosu e analysing in e nal and ex e nal ac o s ha in luence municipali ies o disclosu e in o ma ion. Wi h espec o he academic ield, i empi ically es s a heo e ical amewo k explaining, wi hin he communica ion p ocess, wha in luenced in o ma ion disclosu e in Spanish muni- cipali ies be ween 2008 and 2014. Addi ionally, he sample was pa i ioned in wo ca ego ies acco ding o he p o ince capi al o no p o ince capi al c i e ia in o de o ind i he e a e in e es ing di e ences. I in ends o con ibu e o he ein o esea ch ha s udies he ac o s in luencing municipali ies’ in o ma ion disclosu e wi h impac on communica ion p ac ices. Fo his pu pose, his s udy does an econome ic analysis conside ing spa ial and empo al e ec s (municipali ies-yea s, panel da a me hodology) imp o ing he e iciency o econo- me ic es ima es by cap u ing unobse able he e ogenei y (Bal agi, 2014) and is especially ele an in iew o he empi ical obus ness o his s udy. A apis and Rei ano (2018) poin ou ha panel da a analysis is necessa y o imp o e he unde s anding o anspa ency and i s e olu ion o e ime. I p o ides he analysis wi h a ce ain deg ee o dynamism, achie ing bo h g ea e consis ency and be e explana o y powe . Howe e , o he s udies which use panel da a me hodology (Albala e, 2013), do no ake in o accoun a se ies o p oblems ha a ise when using panel da a me hodology, namely he c oss-sec ional dependence, g oupwise he e oskedas ici y and au oco ela ion, as well as he possible endogenei y p oblem. This s udy mi iga es hese p oblems by using he Sys em- GMM es ima o o a dynamic model o panel da a. Pe aining o p ac i ione s, he ou comes o his s udy b ing insigh s in o local go e nmen s, pa icula ly by answe ing he ollowing cen al esea ch ques ion: wha ac o s in luenced municipali ies’ in o ma ion? The indings could help iden i y he condi ions which ensu e ha municipali ies a e mo e anspa en , since he le el o anspa ency is he consequence o he e ec i e communica ion p ac ices wi hin municipali ies. In ac , he use o anspa ency has been ad oca ed in public ela ions o decades. The es o he pape is o ganized as ollows. The nex sec ion discusses he heo e ical amewo k o in o ma ion disclosu e ha will guide he analysis, and he ac o s ha in luen- ce such disclosu e. Then, he esea ch me hodology and he model will be p esen ed, while p o iding da a desc ip ion, ollowed by he empi ical analysis o he esul s and discussion o he main indings. Finally, he pape will ocus on he main conclusions. F. Tejedo-Rome o, e al. / Con adu ía y Adminis ación 65(4) 2020, 1-29 h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 5 In o ma ion Disclosu e as a mechanism o Good Go e nance Cu en schola ship and p ac ices emphasize he bene i s o openness in he disclosu e o in o ma ion and he p ocess o communica ion. Demand o in o ma ion disclosu e g ew apidly, wi h o ganiza ions, in bo h he p i a e and public sec o s, being encou aged o be mo e anspa en (Ball, 2009). In o ma ion disclosu e became an impo an i em in he agenda o nea ly e e y o ganiza ion, no only o open he go e nmen and make i mo e accoun able, bu also as a means o add essing he loss o public us in go e nmen and imp o ing he p ocess o communica ion. The li e a u e on anspa ency p esen s a di e si y o concep s, s essing h ee main issues: a) make public o ganiza ions’ decisions and ac ions mo e accoun able o ci izens, b) e- duce go e nmen sec ecy and, c) p o ide in o ma ion o he public by se e al mechanisms in o de o ha e accoun able o icials and decisions (Ball, 2009). Acco ding o Ellis and Fende (2006), anspa ency allows indi iduals o ind ou wha is happening inside he go e nmen h ough a ious in o ma ion disclosu e mechanisms. I is a means o imp o e go e nmen ou comes, adminis a i e and manage ial e ec i eness, as well as policy e ec i eness, by conduc ing public a ai s in he open o be subjec o public sc u iny. I includes he legal, poli ical, and ins i u ional s uc u es ha make in o ma ion on public o ganiza ions a ailable o ac o s, bo h inside and ou side he domes ic poli ical sys em. Al hough such anspa ency does no ensu e accoun abili y, o he end o co up ion, i does make i mo e likely. Mo eo e , se e al communica o s ecognize he posi i e e ec s o democ acy esul ing om anspa en communica ion p ac ices and he exis ence o a public in o med abou go e nmen ac ions. In o ma ion disclosu e goes beyond he me e access o in o ma ion, demanding in o ma ion o be unde s andable o ex e nal s akeholde s (A aujo & Tejedo-Rome o, 2016). The li e a u e on in o ma ion disclosu e e e s ou main p inciples. Fi s , in o ma ion abou public o ganiza ions’ ac i i ies a ec ing ci izen in e es s should be a ailable o he public. Second, such in o ma ion should be adjus ed o he in luence ha hose o ganisa ions ha e in he in e es s o ci izens. Thi d, he in o ma ion should be accessible o hose who use i , and, inally, social, poli ical, and economic s uc u es o socie y should be o ganized in ways ha allow ci izens o ake ac ion (Ball, 2009; Fung, 2013). Disclosing in o ma ion is he decision o make public he ac s o he go e nmen and i s ep esen a i es, and p o ide he ci il socie y wi h ele an in o ma ion in a comple e, eliable, imely, easily comp ehensible and eachable way. Disclosu e in o ma ion decision depends om ex e nal and in e nal ac o s o he municipali ies. The ole played by hose ac o s a e c i ical in in o ma ion disclosu e. The ex e nal ac o s can impose o p essu e o in o ma ion disclosu e, o ins ances p essu es om ci izens and go e nmen s, while in e nal ac o s, like cul u e and o he o ganiza ional cha ac e is ics, in luence openness and willingness o F. Tejedo-Rome o, e al. / Con adu ía y Adminis ación 65(4) 2020, 1-29 h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 6 disclose in o ma ion. Ne e heless, in o ma ion disclosu e abou go e nmen ac i i ies and public policies is a co e issue in anspa ency (Alca az-Quiles e al. 2015; A ms ong, 2011). This con ibu es owa ds mo e accu a e public in o ma ion, and a educ ion o he in o ma- ion de ici and asymme y be ween ci izens and go e nmen . Consequen ly, i imp o es he pe manen dialogue be ween policy make s and ci izens du ing he decision-making p ocess. I boos s he le el o con idence o ci izens in hei go e nmen and p omo es g ea e ci ic pa icipa ion. I is essen ial o good go e nance and o he p omo ion o hones and e hical go e nmen s (Bauh & G imes, 2013; Pa k & Blenkinsopp, 2011). In o he wo ds, he mos impo an pa o a public communica ion p ocess is ha go e nmen s communica e in an open, hones and imely way wi h hei ci izens, wi hou manipula ing he in o ma ion hey sha e. Imp o ing he quali y o good go e nance equi es imp o emen s o in o ma ion disclosu e le els. How does one measu e in o ma ion disclosu e? The e a e wo main app oaches o quan i y and measu e he in o ma ion p o ided by o ga- niza ions. The i s app oach uses con en analysis (K ippendo , 2004), which is a me hod o coding ex con en in o ca ego ies based on selec ed c i e ia. The second is he cons uc ion o a disclosu e index ha compa es and explains he di e ences in he scope and b ead h o in o ma ion disclosu e. The c ea ion o a disclosu e index is based on he gene al p inciples o con en (o hema ic) analysis as a me hod es ablished in social sciences. Fo he pu pose o his s udy, he anspa ency index is used as a p oxy o in o ma ion disclosu e, meaning he abili y o he public o access go e nmen in o ma ion (Pio owski & Van Ryzin, 2007; Robe s, 2006). Acco ding o K ippendo (2004), an index is a o m o in e ence ha can be pe o med on con en analysis, and i is de ined as a a iable whose impo ance in a gi en in es iga ion depends on he deg ee o which a co ela e o o he phenomena may be consi- de ed. I is an ins umen designed o measu e a numbe o i ems which, when he alues a e added, p o ides an indica i e alue o he le el o disclosu e in a speci ic con ex . The e o e, he Spanish municipali ies’ anspa ency index (TI), published by T anspa ency In e na ional Spain, is used o explo e he d i e s o Municipal public in o ma ion disclosu e. Theo e ical backg ound o T anspa ency Since he 1970’s, Local Go e nmen anspa ency has been a conce n o schola s, pa icu- la ly ocused on ac o s p omo ing mo e and be e disclosu e o public inancial in o ma ion (Rod íguez-Bolí a , Alcaide-Muñoz & López-He nández, 2013). Li e a u e abou anspa- ency is based on he explana ions p o ided by he agency heo y and he legi imacy heo y (Pa en, 1992; Suchman, 1995). Agency heo y ega ds elec ed o icials as agen s who ac in F. Tejedo-Rome o, e al. / Con adu ía y Adminis ación 65(4) 2020, 1-29 h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 7 he in e es o ci izens, he p incipal (Laswad, Fishe , & Oyele e, 2005). In such ela ionship, some p oblems can a ise when he e is in o ma ion asymme y (Eisenha d , 1989). Elec ed o icials do no ha e he same in e es s as ci izens, and he e o e, mus be held accoun able o hei ac ions, namely by p o ing hey ha e ac ed acco ding o hei esponsibili ies. In he same way, anspa en communica ion p ac ices and in o ma ion disclosu e, by olun a y o compulso y p o isions, has been a way o moni o he ac ions o elec ed o icials (Laswad e al., 2005), and educe in o ma ion asymme ies. I expands he le el o anspa ency and con idence in he go e nmen (Bauh & G imes, 2013). Hence, agency ela ionships be ween public sec o o ganiza ions and ci izens gene a e an incen i e o poli icians and public man- age s o disclose and inc ease he in o ma ion o e ed by hese en i ies, hus allowing hei ac i i ies o be con olled and sc u inized by ci izens and socie y. On he o he hand, Legi imacy Theo y claims ha in o ma ion disclosu e is a way o o ga- niza ions o legi imize hei ac ions owa ds ci izens. Legi imacy is “a gene alised pe cep ion o assump ion ha he ac ions o any en i y a e desi able, p ope , o app op ia e wi hin some socially cons uc ed sys em o no ms, alues, belie s and de ini ions” (Suchman, 1995, 574). In o ma ion disclosu e abou o ganiza ion’s ac i i ies is a way o legi imize hei ac ions be o e hei s akeholde s, and awa d legi imacy o o ganiza ions (Pa en, 1992). The heo y claims ha , i o ganiza ional legi imacy is h ea ened, o ganiza ions will disclose in o ma ion in an e o o p omo e i . The dissemina ion o in o ma ion may be used o win ci izens us back, and ep esen an a emp o change hei pe cep ions abou he unc ioning o he o ganiza ion. In addi ion, o e ing an image o good go e nance in esponse o gene al p essu es is a way o secu e legi imacy, and enhance local go e nmen epu a ion (Pina, To es & Royo, 2010). These heo ies help us unde s and in o ma ion asymme y in poli ical con ex s, and he need o in o ma ion disclosu e and anspa ency. Howe e , hey do no explain which ac o s ac ually in luence in o ma ion disclosu e. Since social and economic de elopmen s in Spain p essu ed poli icians o g ea e openness o make public o ganiza ions mo e accoun able and anspa en i is wo h o empi ically es he in luence o socioeconomic and poli ical ac o s on he le el o anspa ency. T anspa ency in Spanish Municipali ies Spanish municipali ies a e adap ing o socie y’s demand o mo e anspa en communi- ca ion p ac ices and anspa ency. Mo eo e , he e a e in e na ional o ganiza ions e alua ing he le el o in o ma ion disclosu e in di e en coun ies, including Spain. The cons uc ion o a anspa ency index by T anspa ency In e na ional Spain o e ed he oppo uni y and a oused he in e es o such analysis. The T anspa ency Index ocuses p ima ily on p oac- i e communica ion o in o ma ion by municipali ies. Access o public in o ma ion a local F. Tejedo-Rome o, e al. / Con adu ía y Adminis ación 65(4) 2020, 1-29 h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 8 go e nmen le el plays an essen ial ole in he o e all pe cep ion o he exis ence, o no , o an open go e nmen . The wo kings o a go e nmen will be mo e anspa en i he e is mo e in o ma ion a ailable and unde s andable, and he decision- making p ocesses a e mo e accessible (A ms ong, 2011). Acco ding o he Agency Theo y (Laswad, e al., 2005), dis- closu e o in o ma ion esponds o in e nal and ex e nal p essu es o anspa ency, educing in o ma ion asymme y. Spanish social mo emen s and he p essu e om cen al go e nmen o iscal con ol o e municipali ies we e ex e nal sou ces o p essu e o mo e anspa ency. In he same way, he esponse om municipali ies could be in luence municipal execu i e ideology and i s inancial dependence om cen al go e nmen . The inancial dependence om ex e nal unds equi e ha municipali ies should disclosu e mo e in o ma ion abou inancial managemen o hold elec ed o icials accoun able o hei ac ions. In o ma ion disclosu e, ei he as olun a y o compulso y, has been a way o educe in o ma ion asymme ies, and o inc ease he deg ee o con idence and public us in poli ical playe s (Fung, 2013) enhancing municipali ies’ epu a ion. I enhances legi imacy o elec ed public o icials (Pa en, 1992), p oducing a posi i e e ec on public pe cep ions abou poli ical decisions, p ojec ing an image o good go e nance. Ano he ac o ha a ec s communica ion and in o ma ion access, a e he In o ma ion and communica ion echnologies (ICTs), such as he In e ne , which ha e e olu ionized he way socie y communica es and in e ac s. This is also ue in he ela ionship be ween ins i u ions, ci izens, and he a ious le els o go e nmen . The p o ision o in o ma ion and documen s in he municipali ies’ websi es enables hem - ega dless o hei size and/o esou ces - o p o ide in o ma ion o ci izens abou hei o ganiza ion, ac i i ies, and decisions. The use o he In e ne by local go e nmen s o p oac i e communica ion and dissemina ion o in o - ma ion has se e al ad an ages. I c ea es moni o ing and con ol mechanisms, which is a way o enhance ci izen’s powe is-à- is local go e nmen (Laswad e al., 2005). I also allows ci izens o pa icipa e in public managemen in a mo e in o med manne , hus leading o an inc ease in ep esen a ion, and s eng hening he democ a ic sys em. To explo e wha d i es municipali ies’ in o ma ion disclosu e, his s udy used he Spanish municipali ies’ anspa ency index (TI) published by T anspa ency In e na ional Spain (TI- Spain). This index has been used in some p e ious s udies (Albala e, 2013), and is based on he collec ion o in o ma ion and ques ionnai es sen ou o municipali ies. The s udy used da a om T anspa ency In e na ional Spain om 2008 o 2014, as a p oxy o in o ma ion disclosu e. Fac o s A ec ing Municipali ies’ In o ma ion Disclosu e Some s udies ocus on in e nal ac o s o explain anspa ency in public o ganiza ions (Ruije , 2016). Howe e , he e idence is no unequi ocal because o he limi a ions o da a F. Tejedo-Rome o, e al. / Con adu ía y Adminis ación 65(4) 2020, 1-29 h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 9 ga he ed and econome ic analysis. O he s udies abou inancial anspa ency ocus on poli ical and socioeconomic ac o s (Guillamón, Bas ida & Beni o, 2011; Caba-Pé ez, Rod íguez-Bo- li a , & López-He nández, 2014). P oac i e in o ma ion disclosu e by municipali ies may be condi ioned by he exis ence o ce ain ac o s which may in luence and encou age he disse- mina ion o in o ma ion. In he pe iod analysed he e was p essu es om social mo emen s claiming o mo e anspa ency abou public se ice managemen and accoun abili y om poli icians. The in e ne and social ne wo ks we e impo an ool in in o ma ion disclosu e. D awing om and expanding upon p io wo ks he model build ook as i s e e ence he p e- ious li e a u e and a gues ha municipali ies’ p edisposi ion o anspa ency was in luenced by he ollowing in e nal and ex e nal ac o s o he municipali ies: Financial Dependence, In e ne access, Unemploymen , Elec o al Tu nou , Poli ical Ri al y, and Poli ical Ideology. I examines hese p e iously men ioned a iables because hey cons i u e he common basis o he exis ing models in academic li e a u e. The objec i e o his pape is o build up an explana o y model, suppo ed by he agency and he legi imacy heo ies ha apply alid ope a ionaliza ion o hese ac o s and es hei ela i e impo ance in Spanish municipal in o ma ion disclosu e. Addi ionally, i was con olled con ex ual Spanish municipal cha ac- e is ics such as, popula ion size, ci izens le el o educa ions and egional e ec , which we e including as con ol a iables (see con ol a iables in Resea ch Me hodology sec ion). The hypo heses a e de eloped and p oposed below. Financial Dependence. Much o he local e enue comes om cen al o sup ana ional go e nmen ans e s. Financial dependence plays an impo an ole in in o ma ion disclosu e by local go e nmen s. The g ea e he inancial dependence, he mo e accoun ing in o ma ion municipali ies ha e o disclose, and mo e anspa ency is equi ed, o accoun o he use o unds ecei ed. Accoun abili y and in o ma ion disclosu e by anspa en communica ion a e a c i ical pa o inancial con ol. Acco ding o he Agency Theo y (Laswad e al., 2005), i is expec ed ha when go e nmen unds a e ecei ed, municipali ies disclose in o ma ion accoun ing o he use o such unds, whils seeking o demons a e ha public manage s ha e ac ed acco ding o hei esponsibili ies (Ing am & DeJong, 1987). Empi ical s udies, such as hose conduc ed by Ing am and DeJong (1987), ha e concluded ha , when a municipali y ecei es a highe p opo ion o i s o al income as ans e s o unds om s a e and egional go e nmen s, his is an incen i e o g ea e disclosu e o inancial in o ma ion, in o de o demons a e he anspa en managemen o esou ces ecei ed. Al- caide-Muñoz, Rod íguez-Bolí a and López-He nández (2013) and Guillamón e al. (2011) ound ha municipali ies ha ecei e la ge ans e s epo mo e inancial in o ma ion, and a e mo e anspa en . Howe e , s udies by Caba-Pé ez, Rod íguez Bolí a and López He nández (2008) and Laswad e al., (2005) ound no signi ican ela ionship be ween ans e s ecei ed by he municipali y and i s disclosu e o in o ma ion. Fu he mo e, O iz-Rod íguez, Na a - F. Tejedo-Rome o, e al. / Con adu ía y Adminis ación 65(4) 2020, 1-29 h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 16 Panel B: Dummies Va iables 1 0 1 0 1 0 1 0 1 0 1 0 Poli ical Ideology Pe cen age (%) 54 46 54 46 54 46 25 75 25 75 42.4 57.6 Regional Pa y Pe cen age (%) 56 44 56 44 56 44 57 43 57 43 56.4 43.6 N (Obse a ions) 100 100 100 100 100 500 Sou ce: Own elabo a ion Panel A shows ha , o e he yea s, he e we e inc eases in he alue o he anspa ency index in he municipali ies analysed, wi h he index aking an a e age alue o 69.1 du ing he pe iod 2008-2014, sugges ing a mo e anspa en communica ion p ocess. This inc ease may be due o he ac ha hose who go e n and manage he municipali y became awa e ha hei s akeholde s, especially ci izens, a e demanding mo e accoun able ac ions om poli icians as a esul o mis us . Acco ding o he legi imacy heo y in o ma- ion disclosu e and a be e communica ion p ocess is a way o eco e ha los us . This sugges s ha he le el o anspa ency in Spanish municipali ies is inc easing, in o de o legi imize hei ac ions o ci izens and educe he p oblem o agency. These esul s a e in line wi h hose ob ained in he wo k o Guillamón, e al. (2011), who obse ed an inc ease in he anspa ency index du ing 2008-2010. Also o ha pe iod, Cuad ado-Balles e os (2014) ob ained an a e age alue o he anspa ency index o 62.45 o he sample o he 110 la ges Spanish municipali ies. A alue simila o ha was ob ained in his s udy – 69.1 – which is somewha highe due o he inclusion in he s udy o he 2012 and 2014 alues. The esul s a e also consis en wi h hose ob ained by Albala e (2013) o 2012. Conce ning inancial dependence, he e was a s ong inc ease in 2009, which may ha e been mo i a ed by an a emp o boos he local economy h ough an inc ease in inancial esou ces om he cen al go e nmen , a ge ing lowe unemploymen . Howe e , wi h he wo sening o he c isis and he p essu es om he Eu opean Union and he cen al go e nmen o educe public deb , ans e s ecei ed we e hea ily in luenced, and he e was a d ama ic educ ion o inancial dependence in 2012. In e ne access has inc eased in Spanish households, om an a e age alue o 77.5 in 2008, o 93.8 in 2014. The e o e, mos Spanish households ha e access o he In e ne . In o - ma ion and communica ion echnologies ha e e olu ionized he way socie y communica es and in e ac . The unemploymen a e has inc eased o e he yea s. The mos signi ican inc ease oc- cu ed be ween 2008 and 2009, when he economic and inancial c isis s a ed in Spain. The F. Tejedo-Rome o, e al. / Con adu ía y Adminis ación 65(4) 2020, 1-29 h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 17 (18) OECD o ecas o he coming yea s es ima es ha he e will be a weak and slow eco e y o unemploymen (OECD, 2014). On he o he hand, elec o al u nou inc eased om 59% o 61.8% in he las elec ions. Pe haps he economic c isis and he social mo emen s mobilized ci izens o be mo e pa - icipa i e. Conce ning poli ical i al y, he numbe o poli ical pa ies inc eased in he las elec ions - om an a e age o nine poli ical pa ies by municipali y o ele en - maybe be- cause o he exis ing mis us among ci izens, esul ing om co up ion issues in some local go e nmen s, and a mo e agmen ed pa y sys em. The amoun o popula ion has no changed much o e he yea s, wi h an a e age alue o 216,210 inhabi an s o he pe iod. The a e age educa ional le el has emained s able o e ime, wi h a alue o 31.5 o he pe iod unde analysis. Panel B o Table 1 shows ha , in he 2007 elec ions, mos municipali ies in he sample we e uled by le -wing pa ies - 54% e sus 46%. Howe e , a e he las elec ions in 2011, poli ical ideology o he uling pa ies in mos municipali ies changed adically, mainly o igh -wing go e nmen s in 75% o he cases. The esul s o he 2011 local elec ions we e a punishmen o he le -wing uling pa y a cen al le el, esul ing om he bad economic ou comes ha ollowed he inancial c isis. The ex en o punishmen did no ake in o ac- coun he good pe o mance o le -wing pa ies a egional and local le el, no he cases o co up ion o bad pe o mance o some local go e nmen s uled by igh -wing pa ies. I was an exp ession o dissa is ac ion wi h espec o le -wing pa ies. The Regional Pa y a iable shows a link be ween he uling pa ies in each le el. In he egional-municipal case, 56.4% o municipali ies ha e he same uling pa y in each le el. The esul s om es ima ing Eq. (1) a e epo ed in Table 2, using dynamic panel Sys em- GMM es ima ion wi h obus s anda d e o s. Columns 2, 3 and 4 p esen one-s ep Sys em- GMM eg essions and columns 5, 6 and 7 p esen wo-s ep Sys em-GMM eg essions. In bo h models, he eg ession was es ima ed o each o he sample: he ull sample, he sample o he p o ince capi al municipali ies and he sample o no p o ince capi al municipali ies. Howe e , since he es ima ion esul s a e simila o he wo models (one-s ep and wo-s eps), he analysis will concen a e on he wo-s eps Sys em-GMM es ima ion. This es ima ion is mo e obus han he one-s ep. The pu pose o his s udy is o es ima e he e ec o he selec ed a iables on in o ma ion disclosu e, in o de o iden i y wha ac o s d i e anspa ency. To educe he in luence o ex eme alues, unemploymen a iable has been winso ized a he 5 h and 95 h pe cen iles. In addi ion, he na u al loga i hm o popula ion size a iable has also been used. F. Tejedo-Rome o, e al. / Con adu ía y Adminis ación 65(4) 2020, 1-29 h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 18 Table 2 De e minan s o Municipal T anspa ency in Spain in he pe iod o 2008-2014 Independen a iables Model One-s ep Model Two-s ep Full Sample Coe . (Rob. S d. E .) p- alue P o ince Cap. Sample Coe . (Rob. S d. E .) p- alue No P o ince Cap. Sample Coe . (Rob. S d. E .) p- alue Full Sample Coe . (Rob. S d. E .) p- alue P o ince Cap. Sam- ple Coe . (Rob. S d. E .) p- alue No P o - ince Cap. Sample Coe . (Rob. S d. E .) p- alue T anspa ency Index -1 0.3092*** 0.1488 0.3287*** 0.2923*** 0.1641 0.3220*** (0.0791) (0.106) (0.0915) (0.0895) (0.1374) (0.0945) 0.0001 0.1603 0.0003 0.0011 0.2322 0.0007 Financial Dependence 0.8012*** 0.7930*** 0.6532*** 0.7189*** 0.3636 0.8541*** (0.1986) (0.297) (0.2205) (0.1849) (0.3248) (0.1999) 0.0001 0.0076 0.0031 0.0001 0.2629 0.0000 In e ne access 1.4743*** 1.9619*** 0.9476* 1.3459*** 2.6770*** 0.8734** (0.3794) (0.6479) (0.5474) (0.4572) (0.9406) (0.4138) 0.0001 0.0025 0.0835 0.0032 0.0044 0.0348 Unemploy- men -1.9251*** -2.0111 -1.6120** -2.5881** -7.8638* -2.5264* (0.6194) (1.3606) (0.7937) (1.1368) (4.3641) (1.3443) 0.0019 0.1394 0.0423 0.0228 0.0716 0.0602 Elec o al u nou -0.7743*** -1.3138*** 0.1682 -0.7071** -1.2691** 0.0385 (0.272) (0.4587) (0.3721) (0.353) (0.5499) (0.5735) 0.0044 0.0042 0.6513 0.0452 0.021 0.9464 Poli ical Ri al y 1.2979*** 1.1453** 1.1129* 1.3752*** 1.5043 1.7407** (0.3548) (0.5626) (0.6394) (0.4932) (0.9644) (0.7237) 0.0003 0.0418 0.0818 0.0053 0.1188 0.0162 Poli ical Ideology 2.8777 -3.7608 12.2889** 0.5337 -4.4517 7.3358 (4.1581) (6.0089) (5.8102) (4.4402) (6.142) (6.7511) 0.4889 0.5314 0.0344 0.9043 0.4686 0.2772 Popula ion Size -7.2795*** -3.9546 -9.6656*** -6.8757*** -3.4706 -9.6189*** (1.9708) (2.8476) (3.3882) (2.3841) (3.8264) (3.431) 0.0002 0.1649 0.0043 0.0039 0.3644 0.0051 F. Tejedo-Rome o, e al. / Con adu ía y Adminis ación 65(4) 2020, 1-29 h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 19 Independen a iables Model One-s ep Model Two-s ep Full Sample Coe . (Rob. S d. E .) p- alue P o ince Cap. Sample Coe . (Rob. S d. E .) p- alue No P o ince Cap. Sample Coe . (Rob. S d. E .) p- alue Full Sample Coe . (Rob. S d. E .) p- alue P o ince Cap. Sam- ple Coe . (Rob. S d. E .) p- alue No P o - ince Cap. Sample Coe . (Rob. S d. E .) p- alue Educa ion Le el 1.7437** 0.1479 2.5825** 2.4537** 2.0829 3.2609*** (0.8352) (1.4861) (1.0062) (1.0335) (1.9909) (1.2339) 0.0368 0.9207 0.0103 0.0176 0.2955 0.0082 Regional Pa y -0.6841 0.9076 0.0208 0.6301 6.5180 -0.7800 (1.8599) (3.3731) (2.7121) (2.4591) (5.4888) (3.4753) 0.713 0.7879 0.9939 0.7978 0.235 0.8224 Region: A agon -25.3256* -4.6088 0 -34.1757* -72.7873 0 (13.1177) (17.8972) 0 (18.7406) (50.9259) 0 0.0535 0.7968 . 0.0682 0.1529 . As u ias -7.3803 18.6647 -34.4558*** -17.0186 -41.5862 -63.2681** (8.373) (13.9394) (11.6101) (11.7454) (41.2382) (29.9172) 0.3781 0.1806 0.003 0.1473 0.3132 0.0344 Balea ic Islands 0.2645 -3.6243 0 -35.9422 -84.6557 0 (6.6514) (13.062) 0 (40.395) (58.5158) 0 0.9683 0.7814 . 0.3736 0.148 . Cana y Is- lands -13.4056* -5.4752 -22.6149** -14.7390** -5.1564 -27.4710*** (6.8993) (9.8062) (9.1927) (6.8709) (11.2436) (10.2383) 0.052 0.5766 0.0139 0.0319 0.6465 0.0073 Can ab ia -5.5566 16.7647 0 -19.3609 -18.9557 0 (8.5709) (16.3687) 0 (58.8387) (34.9095) 0 0.5168 0.3057 . 0.7421 0.5871 . Cas illa La Mancha 10.4199* 24.0728*** -14.7145* 6.0189 -16.1104 9.8389 (6.0058) (8.7792) (8.231) (7.9555) (27.4016) (44.3539) 0.0827 0.0061 0.0738 0.4493 0.5566 0.8244 F. Tejedo-Rome o, e al. / Con adu ía y Adminis ación 65(4) 2020, 1-29 h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 20 Independen a iables Model One-s ep Model Two-s ep Full Sample Coe . (Rob. S d. E .) p- alue P o ince Cap. Sample Coe . (Rob. S d. E .) p- alue No P o ince Cap. Sample Coe . (Rob. S d. E .) p- alue Full Sample Coe . (Rob. S d. E .) p- alue P o ince Cap. Sam- ple Coe . (Rob. S d. E .) p- alue No P o - ince Cap. Sample Coe . (Rob. S d. E .) p- alue Cas illa Leon -6.0186 15.9380 0 -13.8340 -32.1572 0 (7.5529) (13.3016) 0 (10.4215) (34.6809) 0 0.4255 0.2308 . 0.1844 0.3538 . Ca alonia -28.3625*** -19.4364 -29.8969*** -35.5136*** -91.8768* -37.3135** (9.2322) (16.1833) (9.9364) (13.2798) (54.4679) (14.6375) 0.0021 0.2297 0.0026 0.0075 0.0916 0.0108 Ex emadu a -2.1578 0.7552 0 -6.0082 -8.0170 0 (8.1312) (10.9834) 0 (8.1854) (12.522) 0 0.7907 0.9452 . 0.4629 0.522 . Galicia -5.0009 18.4750 -20.2937** -11.4734 -23.4584 -24.8194** (7.0887) (11.3477) (8.0548) (8.4332) (28.4867) (9.9235) 0.4805 0.1035 0.0118 0.1737 0.4102 0.0124 Rioja -3.6812 23.2549 0 -36.9443 -79.0726 0 (8.7217 (15.3403) 0 (23.4242) (51.6805) 0 0.673 0.1295 . 0.1148 0.126 . Mad id -28.3484* 6.6320 -53.6193*** -45.9573** -105.1658 -64.1958*** (14.6581) (24.1057) (16.2301) (19.5924) (79.7565) (21.6535) 0.0531 0.7832 0.001 0.019 0.1873 0.003 Mu cia 10.7369 35.8136*** -4.1623 6.1970 -41.7531 -3.1848 (9.3686) (7.3976) (10.094) (13.4085) (48.8588) (11.0285) 0.2518 0 0.6801 0.644 0.3928 0.7727 Na a e -26.0260** 4.3969 0 -40.0361 -77.6562 0 (11.4126) (20.8769) 0 (56.8888) (72.4412) 0 0.0226 0.8332 . 0.4816 0.2837 . Basque Coun y -53.2956*** -12.8451 -78.3443*** -67.6400*** -96.9007 -96.3094*** (18.0147) (29.9309) (21.0283) (23.4845) (65.434) (26.5652) 0.0031 0.6678 0.0002 0.004 0.1386 0.0003 F. Tejedo-Rome o, e al. / Con adu ía y Adminis ación 65(4) 2020, 1-29 h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 21 Independen a iables Model One-s ep Model Two-s ep Full Sample Coe . (Rob. S d. E .) p- alue P o ince Cap. Sample Coe . (Rob. S d. E .) p- alue No P o ince Cap. Sample Coe . (Rob. S d. E .) p- alue Full Sample Coe . (Rob. S d. E .) p- alue P o ince Cap. Sam- ple Coe . (Rob. S d. E .) p- alue No P o - ince Cap. Sample Coe . (Rob. S d. E .) p- alue Valencia 4.4688 5.2705 -8.4144 1.8065 -23.6939 -8.4129 (5.8911) (10.8178) (7.1791) (8.1325) (25.4518) (10.523) 0.4481 0.6261 0.2412 0.8242 0.3519 0.424 Obse a ions 400 204 196 400 204 196 Nº g oups 100 51 49 100 51 49 Nº ins u- men s 50 50 43 50 50 43 A ellano– Bond es o AR(1)/ (p- alue) -5.43/(0.000) -3.84/(0.000) -3.30/(0.001) -4.80/(0.000) -3.30/ (0.001) -3.04/ (0.002) A ellano– Bond es o AR(2)/ (p- alue) 1.37/(0.170) 0.36/(0.717) 1.80/(0.282) 1.33/(0.185) 0.50/ (0.615) 1.25/(0.212) Hansen es / (p- alue) 28.48/(0.240) 17.08/ (0.845) 20.54/ (0.247) 28.48/(0.240) 17.08/ (0.845) 20.54/ (0.247) Χ2 26 440.32*** 13850.14*** 107.17*** 8210.95*** 355.27*** 2791.14*** Robus s anda d e o s wi h Windemeije ’s ini e sample co ec ion a e in pa en heses. *p<0.1; **p<0.05; ***p<0.01. “.” omi ed a iables. Sou ce: Own elabo a ion The speci ica ion es s sugges ed by Roodman (2006) ha e been used o alida ion o he model by he wo-s ep Sys em-GMM eg ession. In his sense, he A ellano-Bond es s gen- e ally iden i ied high i s -o de au oco ela ion. The es shows ha he comple e es ima ion has an AR(1)=-4.80 and p- alue=0.000, he p o ince capi al municipali ies sample an AR(1)=- 4.80 and p- alue=0.000 and he no p o ince capi al municipali ies samples an AR(1)=-3.04 F. Tejedo-Rome o, e al. / Con adu ía y Adminis ación 65(4) 2020, 1-29 h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 22 and p- alue=0.0002. The esul s did no p o ide e idence o second-o de au oco ela ion wi h alues o AR(2)=1.33, p- alue=0.185; AR(2)=0.50, p- alue=0.615; AR(2)=1.25, p- alue=0.212; espec i ely. The Hansen es s o o e - iden i ying es ic ions con i m ha he se o ins umen s can be conside ed alid (Hansen es =28.48, p- alue=0.240; Hansen es =17.08, p- alue=0.845; Hansen es =20.54, p- alue=0.247; espec i ely). Mo eo e , he numbe o ins umen s (50, 50 and 43; espec i ely) does no exceed he numbe o g oups (100, 51 and 49; espec i ely), which suppo s he model’s co ec speci ica ion and is alid. And, he χ2 shows ha he o e all eg essions a e signi ican (χ226=8210.95, p- alue=0.000; χ226=355275, p- alue=0.000; χ226=2791.14, p- alue=0.000; espec i ely). The lagged alue o he dependen a iable (T anspa ency Index -1) was s a is ically sig- ni ican a 1% le el, in he ull and no p o ince capi al municipali ies samples (β0=0.2923, p- alue=0.0011; β0=0.3220, p- alue=0.0007; espec i ely). In o he wo ds, he amoun o in o ma ion disclosed depends on he amoun ha was p o ided in p e ious yea s. How- e e , hese esul s could no be con i med in he case o no p o ince capi al municipali ies (β0=0.1641, p- alue=0.2322). I was obse ed ha he inancial dependence a iable has a posi i e and signi ican e - ec o a signi icance le el o 1, in he ull and no p o ince capi al municipali ies samples (β1=0.7189, p- alue=0.0001; β1=0.8541, p- alue=0.0000; espec i ely). G ea e inancial dependence in municipali ies means hey will be mo e inclined o disclose in o ma ion as a way o legi imizing he ac ions o poli icians. Acco ding o he agency and legi imacy heo ies, his educes he agency p oblem and imp o es he con idence on municipali ies’ managemen . Alcaide-Muñoz e al., (2013) and Guillamón e al. (2011) ound ha munici- pali ies ha ecei e la ge ans e s om he s a e and egional go e nmen s disclose mo e in o ma ion. Howe e , hese esul s could no be con i med in he case o no p o ince capi al municipali ies (β1=0.3636, p- alue=0.2629). The e o e, hypo hesis H1 o mula ed in he heo e ical amewo k is accep ed. In e ne Access is signi ican a 1% le el, in he ull and no p o ince capi al municipa- li ies samples (β2=1.3459, p- alue=0.0032; β2=2.6770, p- alue=0.0044; espec i ely). and 5% le el o no p o ince capi al municipali ies sample (β2=0.8734, p- alue=0.0348). The g owing use o new in o ma ion and communica ion echnologies imp o es he communica ion p ocess and allows ci izens o inc ease hei use o he In e ne . I con ibu es o in o ma- ion dissemina ion, educing in o ma ion asymme y. A g ea e access o he In e ne could be a way o ci izens o pu p essu e on municipali ies o be mo e anspa en and disclose in o ma ion. The esul s con i m hose ob ained by Caba-Pé ez e al., (2014) and Pina e al., (2010). Hence, hypo hesis H2 is accep ed. Rega ding he unemploymen a iable, he e is a signi ican nega i e ela ionship o a signi icance le el o 5% in he ull sample (β3 =-2.5881, p- alue=0.0228) and o a signi i- F. Tejedo-Rome o, e al. / Con adu ía y Adminis ación 65(4) 2020, 1-29 h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 23 cance le el o 10% in he p o ince capi al and no p o ince capi al municipali ies samples (β3 =-7.8638, p- alue=0.0716; β3 =-2.5264, p- alue=0.0602; espec i ely). As discussed in he heo e ical amewo k, unemploymen is a p oxy o he economic s a us o municipali ies. In pe iods o c isis, an inc ease in his a iable is expec ed, and, consequen ly, he e is a s ong a emp by poli icians o hide he si ua ion, and he e o e, no disclose so much in o ma ion. Acco ding o he heo y o legi imacy, his is an a emp no o lose he con idence o he ci izens. A nega i e ela ionship has been obse ed, as concluded in he esea ch o Albala e (2013) and Caamaño-Aleg e e al. (2013). So, hypo hesis H3 is accep ed. Rega ding he a ia- ble elec o al u nou , i is signi ican a he 5% le el, and i is nega i ely associa ed wi h he dependen a iable, in he ull and no p o ince capi al municipali ies samples (β4=-0.7071, p- alue=0.0452; β4=-1.2691, p- alue=0.021; espec i ely) Hypo hesis H4 is con i med. I is easonable o assume ha he lowe he elec o al u nou , he lowe he us in municipal go e nmen . The e o e, as sugges ed by legi imacy and agency heo ies, in o de o inc ease ci izens’ in ol emen , municipali ies should p o ide ele an in o ma ion on budge s, plans, and ac i i ies con ibu ing o municipal anspa ency. Howe e , hese esul s could no be con i med o municipali ies ha a e no he p o ince capi al (β4=0.0385, p- alue=0.9464). The poli ical i al y a iable is posi i ely co ela ed wi h he anspa ency index a he 1% and 5% le el in he ull and no p o ince capi al municipali ies samples, espec i ely (β5 =1.3752, p- alue=0.0053; β5 = 1.7407, p- alue=0.0162; espec i ely). Hence, hypo hesis H5 can be con i med. This con i ms ha poli ical i al y c ea es incen i es o educe agency p oblems and in o ma ion asymme y. In he Spanish con ex , Ga cía-Sánchez e al. (2013) ound no signi ican in luence o poli ical i al y in in o ma ion disclosu e. This esul could no be con i med o he no p o ince capi al municipali ies (β5=1.5043, p- alue=0.1188). The e is no signi ican ela ionship wi h poli ical ideology. This esul con i ms o he s udies which a gue ha le -wing pa ies and igh -wing pa ies a e bo h in e es ed in anspa ency. The e o e, hypo hesis H6 is no con i med. O he ac o s, such as popula ion size, ci izens’ educa ional le el, and egion ha e a sig- ni ican in luence in he le el o municipal anspa ency in he ull and no p o ince capi al samples. These ac o s a e no signi ican o he p o ince capi al municipali ies’ sample. Howe e , o all cases, he e is no ela ionship ega ding he e ec o he egional pa y in he le el o municipal anspa ency. Conclusions This pape aimed o analyse he d i e s o in o ma ion disclosu e in he la ges Spanish municipali ies using panel da a analysis. Since he la ge municipali ies a e ele an , he sample was pa i ioned in o wo ca ego ies acco ding o he c i e ia: p o ince capi al municipali ies F. Tejedo-Rome o, e al. / Con adu ía y Adminis ación 65(4) 2020, 1-29 h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 24 and no p o ince capi al municipali ies. The indings o his pape a e a con ibu ion o he lou ishing body o e idence on his impo an ein o knowledge. In he pe iod unde analysis, esul s sugges ed an inc ease in he le el o in o ma ion disclosu e in Spanish municipali ies, helping o legi imize hei ac ions o ci izens and educe he p oblem o he agency. The g ea es a ia ion ook place in 2012, which co esponds o a change in he elec o al cycle due o he 2011 elec ions. Pe haps his was he consequence o social mo emen s, such as “los indignados”, whose o igin was mo i a ed by he pligh o ci izens. These indings a e consis en wi h he exis ence o p essu es om ci izens ega ding decision-making p ocess and how esou ces a e used, leading municipali ies o educe he asymme y o in o ma ion by disclosing in o ma ion. I shows ha municipal managemen changed owa ds mo e anspa ency, ying o espond o ci izens and o he s akeholde s’ p essu es. Fu he mo e, he e is empi ical e idence ha ce ain in e nal and ex e nal ac o s o he municipali ies in luence in o ma ion disclosu e. The ac o s ha ha e a signi ican e ec on in o ma ion disclosu e and d i e a mo e anspa en communica ion p ocess a e: inancial dependence, in e ne access, unemploymen , elec o al u nou , and poli ical i al y. Despi e he ini ial expec a ions, poli ical ideology had no signi ican e ec on he le el o in o ma ion disclosu e. Conside ing he p o ince capi al municipali ies and hose who a e no p o ince capi al, he esul s show he e a e di e ences in he ac o s ha in luence in o ma ion disclosu e. Hence, o p o ince capi al municipali ies, da a sugges ed ha ex e nal ac o s (unemploymen , elec o al u nou and in e ne access) ha e a signi ican e ec on in o ma ion disclosu e. Fo he no p o ince capi al municipali ies, bo h in e nal and ex e nal ac o s ( inancial depen- dence, poli ical i al y, unemploymen and in e ne access) ha e a signi ican e ec . On he o he hand, a iables such size o popula ion, educa ion le el and egion e ec in luenced in o ma ion disclosu e in he ull sample. Despi e he ini ial expec a ions, he egional pa y had no signi ican e ec in he le el o in o ma ion disclosu e in he ull sample; hese esul s we e simila o hose municipali ies ha a e no p o ince capi al. Howe e , hese a iables had no signi ican in luence in he p o ince capi al municipali ies. Ne e heless, he di e ence ound on municipali ies sugges s ha ins i u ional cha ac- e is ics could in luence in o ma ion disclosu e. In e nal ea u es o he municipali ies, such as o ganiza ional complexi y and cul u e and he s yle o public managemen , among o he s, could in luence he le el o in o ma ion disclosu e. This is wo h o s udy in u he s udies. The indings a e in line wi h he Agency Theo y, which pos ula e he exis ence o p essu e om ci izens on how decisions and ac ions a e aken, leading municipali ies o educe asym- me y o in o ma ion by disclosing in o ma ion. In he same way, acco ding o he Theo y o Legi imacy, municipali ies inc eased he le els o in o ma ion disclosu e as a way o legi imize he ac ions o poli ical leade s owa d ci izens. This sugges s ha ci izens a e demanding mo e F. Tejedo-Rome o, e al. / Con adu ía y Adminis ación 65(4) 2020, 1-29 h p://dx.doi.o g/10.22201/ ca.24488410e.2020.2299 25 openness and in o ma ion om poli icians due o he loss o con idence in hem. One o he main implica ions o p ac i ione s is ha in o ma ion disclosu e is a key aspec o educe in o ma ion asymme y be ween ci izens and poli icians, imp o e he communica- ion p ocess and inc ease accoun abili y. The highe he social engagemen and anspa ency, he mo e legi ima e is he decision-making p ocess o municipali ies. Go e nmen s should enhance he mechanisms o in o ma ion disclosu e, wo king side by side wi h in e na ional o ganiza ions o lea n om he in e na ional expe ience. On he o he hand, in e na ional o ganiza ions should encou age cen al go e nmen s o os e in o ma ion disclosu e, ac- coun abili y and anspa ency. As o he limi a ions o his esea ch, one has o poin ou da a a ailabili y. In o ma ion disclosu e was measu ed wi h a p oxy o he anspa ency index. The anspa ency index ocuses on he supply side, and i does no conside whe he he in o ma ion a ailable o ci izens is comple e, eliable, imely, easily comp ehensible and accessible. Howe e , he s udy does no analyse how ci izens pe cei e he in o ma ion. I would be in e es ing o ha e ex ensi e ques ionnai es om ci izens in each o he 100 municipali ies included in he da ase , in o de o assess hei le el o knowledge on hei municipali y’s ac i i ies and decisions. These opics would help expand he unde s anding on municipali ies’ disclosu e managemen . Since in o ma ion disclosu e is a complex issue, u u e esea ch should analyse he in luence o o he municipal in e nal ac o s, like cul u al aspec s, s a egic managemen , decision-making p ocess, which could in luence in o ma ion disclosu e and anspa ency. Acknowledgmen This s udy was conduc ed a he Resea ch Cen e in Poli ical Science (UIDB/ CPO/00758/2020), Uni e si y o Minho/Uni e si y o É o a and suppo ed by he Po uguese Founda ion o Science and Technology (FCT) and he Po uguese Minis y o Educa ion and Science h ough na ional unds Re e ences Albala e, D. (2013), The ins i u ional, economic and social de e minan s o local go e n- men anspa ency. Jou nal o Economic Policy Re o m, 16(1), 90-107. h ps://doi.o g/1 0.1080/17487870.2012.759422 Alcaide-Muñoz, L., Rod íguez-Bolí a , M. P., and López-He nández, A. M. (2013), Finan- cial de e minan s o imp o ing public sec o accoun abili y: a me a-analy ic s udy. T an-