Document de treball de l IEB 2015/21

Transcription

1 Document de treball de l IEB 2015/21 FISCAL EQUALIZATION UNDER POLITICAL PRESSURES Alejandro Esteller-Moré, Umberto Galmarn, Leonzo Rzzo Fscal Federalsm

2 Documents de Treball de l IEB 2015/21 FISCAL EQUALIZATION UNDER POLITICAL PRESSURES Alejandro Esteller-Moré, Umberto Galmarn, Leonzo Rzzo The IEB research program n Fscal Federalsm ams at promotng research n the publc fnance ssues that arse n decentralzed countres. Specal emphass s put on appled research and on work that tres to shed lght on polcy-desgn ssues. Research that s partcularly polcy-relevant from a Spansh perspectve s gven specal consderaton. Dssemnatng research fndngs to a broader audence s also an am of the program. The program enjoys the support from the IEB-Foundaton and the IEB- UB Char n Fscal Federalsm funded by Fundacón ICO, Insttuto de Estudos Fscales and Insttut d Estuds Autonòmcs. The Barcelona Insttute of Economcs (IEB) s a research centre at the Unversty of Barcelona (UB) whch specalzes n the feld of appled economcs. The IEB s a foundaton funded by the followng nsttutons: Applus, Aberts, Ajuntament de Barcelona, Dputacó de Barcelona, Gas Natural, La Caxa and Unverstat de Barcelona. Postal Address: Insttut d Economa de Barcelona Facultat d Economa Empresa Unverstat de Barcelona C/ John M. Keynes, 1-11 (08034) Barcelona, Span Tel.: eb@ub.edu The IEB workng papers represent ongong research that s crculated to encourage dscusson and has not undergone a peer revew process. Any opnons expressed here are those of the author(s) and not those of IEB.

3 Documents de Treball de l IEB 2015/21 FISCAL EQUALIZATION UNDER POLITICAL PRESSURES * Alejandro Esteller-Moré, Umberto Galmarn, Leonzo Rzzo ABSTRACT: We examne the desgn of fscal equalzaton transfers amed at nterregonal redstrbuton n a settng n whch specal nterest groups dstort the fscal polces of local governments. Equty always calls for tax-base equalzaton whle effcency calls for tax-base equalzaton of fscal capactes backed by strong lobby groups and for taxrevenue equalzaton of those backed by weak lobby groups. Hence, t s optmal to rely only on tax-base equalzaton f the specal nterest groups are smlar n terms of lobbyng power, whereas a mxed system s optmal f they are hghly heterogeneous. Tax competton renforces the role of tax-base, whle tax exportng that of tax-revenue, fscal equalzaton. JEL Codes: H77, D72, H21 Keywords: Fscal-capacty equalzaton-grants, nter-regonal redstrbuton, tax competton, equty-effcency tradeoff, specal nterest groups, lobbyng Alejandro Esteller-Moré Unverstat de Barcelona & IEB Avda. Dagonal Barcelona, Span E-mal: aesteller@ub.edu Umberto Galmarn Dpartmento d Drtto, Economa e Culture Unverstà dell Insubra & IEB Va S. Abbondo, Como, Italy E-mal: umberto.galmarn@unnsubra.t Leonzo Rzzo Unversty of Ferrara & IEB Dpt. of Economcs and Management Va Voltapaletto Ferrara, Italy E-mal: leonzo.rzzo@unfe.t * Prevous versons of ths paper crculated under the ttle Fscal-capacty equalzaton-grants wth taxpayers lobbyng and were presented at the 68th IIPF Congress, Dresden, August 16-19, 2012, at the V Workshop on Fscal Federalsm, IEB, Barcelona, June 13-14, 2013, at the PET Conference, Lsbon, July 5-7, We thank Lsa Grazzn and semnar partcpants for useful comments. The authors gratefully acknowledge fundng from the Spansh Mnstry of Economy and Compettveness (ECO ).

4 1 Introducton Fscal equalzaton programs represent an mportant feature of publc sector nance n many states wth multple levels of government (Blöchlger and Charbt, 2008, Blöchlger, 2013). Snce, generally, the local tax bases and the local publc expendture needs are not unformly dstrbuted over the terrtores of the state, scal equalzaton ams at reducng such gaps, thereby lmtng the dspartes n terms of net scal bene ts for smlar ndvduals lvng n d erent regons. However, the gans n terms of equty stemmng from scal equalzaton must be confronted wth ts costs n terms of e cency, snce the transfer mechansm can dstort the local governments tax and expendture decsons, as well as the locatonal choces of ndvduals and rms. The lterature on scal equalzaton can be dvded nto two man strands. The rst examnes, at a general level, under ether a postve or a normatve perspectve, the assgnment of the allocatve and dstrbutve functons between the central and the sub-natonal governments, and the role of nter-governmental transfer programs. Ths lterature can be traced back to the works of Buchanan and Goetz (1972), Flatters et al. (1974), Boadway and Flatters (1982), where emphass s gven to ne cences n the allocaton of the populaton as a consequence of scal externaltes n the presence of perfect moblty of the populaton. In ths settng, the role of transfers s to nternalze the scal externaltes. More recent contrbutons, e.g. Dahlby (1996), Sato (2000) and Albouy (2012), extend the prevous models by assumng mperfect moblty of the populaton and dstortonary taxaton. The analyss of the optmal desgn of grant schemes based on the equalzaton of the socal margnal cost of rasng tax revenue across jursdctons s due to Dahlby and Wlson (1994), whle Gordon and Cullen (2011) examne the nterplay between redstrbuton polces at the central and at the local level. The second strand of the lterature s more polcy-orented and examnes the functonng of partcular forms of scal equalzaton schemes that are used n practce. Among these, the grants amed at scal-capacty equalzaton receve a great deal of attenton, snce programs of ths knd are mplemented both n federal (e.g., Canada and Germany) and untary (e.g., Norway and Italy) states. Fscal capacty equalzaton can be based ether on standard tax revenues (tax-base equalzaton) or on e ectve tax revenues (tax-revenue equalzaton). Focusng on the representatve tax system (RTS) equalzaton scheme for the Canadan Provnces, Smart (1998) o ers a clear theoretcal analyss of the ncentve e ects of tax-base equalzaton, by whch a recevng local jursdcton s enttled to a grant that depends on the d erence between the natonalaverage and ts own per capta standard tax revenues. Snce the latter, whch depend on 2

5 the e ectve tax bases, are a decreasng functon of the e ectve tax rates, the equalzaton program gves the local governments an ncentve to obtan more grants by rasng ther tax rates. Of the opposte sgn are the ncentves provded by tax-revenue equalzaton, snce when a local government reduces ts own tax rates, by reducng ts own tax revenues t ncreases the equalzaton grant, whch s based on the gap between average and own e ectve tax revenues. The emprcal works by Esteller-Moré and Solé-Ollé (2002) and Smart (2007) for Canada, and Buettner (2006) for Germany, generally con rm the relevance of the ncentves that tax-base equalzaton gves to set excessvely hgh tax rates at the local level, whle Barett et al. (2002) show that the German nterstate transfer system, based on tax revenues, dscourages scal e ort. Followng ths lne of research, some authors (see, e.g., Koethenbuerger, 2002; Bucovetsky and Smart, 2006; Rzzo, 2008), have observed that the dstortng ncentves of scal equalzaton should be confronted wth those arsng from horzontal tax competton among local governments. Hence, the focus of these studes s to characterze the condtons under whch scal equalzaton programs amed at nter-regonal redstrbuton can brng about also some e cency gans, by mtgatng the adverse e ects of tax competton. A common hypothess of ths lterature s that polcy makers are benevolent socal welfare maxmzers. Sub-optmalty of local scal polces, and hence the correctve role, f any, of scal equalzaton, s due to the falure by local governments to nternalze varous types of scal externaltes. In the present work, nstead, we contrbute to the second strand of the lterature by takng a poltcal economy perspectve. In partcular, our premse s that scal polces at the local level can be n uenced and dstorted by specal nterest groups competng for preferental treatment, and accordngly examne how these dstortons a ect the desgn of scal capacty equalzaton programs, based ether on tax bases or on tax revenues, that pursue equty and e cency objectves. We set up a smple publc nance model, n whch a large number of small local authortes nance ther local publc expendture by taxng ncomes accrung to two types of producton factors that, beng the expresson of d erent soco-economc groups, also engage n lobbyng actvtes n the attempt to bend the polcy choces to ther advantage. We examne rst the case n whch producton factors are mmoble and then extend the analyss to factors moblty. We abstract from d erences n scal needs or servce-cost provson at the local level, whle we allow for d erental scal capactes. In de nng the transfer program, we assume that the central authorty s guded by the max-mn crtera of maxmzng the per capta socal welfare of the less well-o local jursdctons. 1 1 The lterature on scal capacty equalzaton (see, e.g., the revew by Boadway, 2006) usually 3

6 We frame the actvtes of the specal nterest groups by means of the common agency approach developed by Bernhem and Whnston (1986a, 1986b). However, we stress that the scope of our results extends to other forms of ctzens n uence on scal choces (e.g., Brusco et al., 2014), as well as to stuatons n whch scal polcy s drven by electoral concerns, lke n probablstc votng settngs (e.g., Hettch and Wner, 1988, Dxt and Londregan, 1998). We nd that t s mportant to account for poltcally nduced dstortons n the desgn of scal equalzaton transfers. In partcular, f the two lobby groups are not hghly heterogeneous n terms of lobbyng power, then t s optmal to rely exclusvely on tax-base equalzaton. Ceters parbus, the degree of equalzaton on the tax base backed by the stronger lobby group s hgher than that on the tax base backed by the weaker group. If, nstead, the specal nterest groups are hghly heterogeneous n terms of lobbyng power, then t s optmal to employ a mxed system: tax-base equalzaton on the tax base backed by the strong lobby group and tax-revenue on that backed by the weak one. The explanaton s smple. From the equty perspectve, tax-base equalzaton s superor to tax-revenue equalzaton. If local governments are on the ncreasng sde of ther La er curve, tax-base equalzaton, by fosterng tax rates and local tax revenues, ncreases nter-regonal redstrbuton. On the contrary, tax-revenue equalzaton, by depressng tax rates and revenues, reduces the amount of nter-regonal redstrbuton. From the e cency perspectve, each one of the scal equalzaton mechansms works on a d erent sde of the same con. Snce lobbyng dstorts taxaton downward on the more powerful group and upward on the less powerful one, tax-base equalzaton on the former tax base, by stmulatng scal e ort, and tax-revenue equalzaton on the latter, by hnderng scal e ort, can help to redrect taxaton toward ts e cent structure. Takng together equty and e cency consderatons, we obtan the normatve prescrptons descrbed above. Inter-regonal moblty of producton factors determnes two contrastng ncentves on local governments tax settng. One s tax competton, puttng a downward pressure on taxaton. The other one s tax exportng, gvng ncentves of the opposte sgn. Whle tax competton renforces the role of tax-base equalzaton, tax exportng renforces that of tax-revenue equalzaton. Ceters parbus, factors moblty reduces the gap between the scal equalzaton rates of the strong and the weak lobby groups. de nes the equty objectve n terms of horzontal equty, by whch otherwse dentcal ndvduals should bear equal tax burdens, rrespectve of ther place of resdence. Snce our man goal s to analyze the equty-e cency tradeo n the desgn of scal equalzaton schemes under lobbyng and tax competton dstortons, we de ne the objectve functon n terms of a socal welfare functonal. 4

7 The equty-e cency trade-o facng nter-regonal redstrbuton programs, as well as some aspects of the poltcal determnants of the transfer polces, have already been examned n the lterature on scal federalsm, but never jontly, as we do n ths paper, and however wth a focus on ssues that d erent from ours. Lockwood (1999) and Bordgnon et al. (2001) focus on redstrbuton among regons n a settng of asymmetrc nformaton n whch, because of moral hazard and adverse selecton, the optmal equalzaton transfers are second best. Kotsoganns and Schwager (2008) examne how scal capacty equalzaton mpacts on the accountablty of local poltcans. The mportance of poltcal ncentves on the allocaton of grants s stressed by a lne of emprcal research: Dahlberg and Johansson (2002) focus on electoral ncentves, whle Levtt and Snyder (1995), Larcnese et al. (2006), and Solé-Ollé and Sorrbas-Navarro (2008), focus on the partsan algnment hypothess. The rest of the paper s organzed as follows. Secton 2 presents the theoretcal framework. Secton 3 examnes the normatve aspects of scal equalzaton n a settng wth mmoble tax bases. Secton 4 extends the analyss to the case of moble tax bases. Secton 5 concludes and an Appendx contans some of the techncal detals. 2 The model Consder a federaton composed of two types of local jursdctons, or regons, ndexed by j = 1; 2, and assume that there s an equal, and large, number, J > 1, of jursdctons of each type. In both types of regons, a prvate good s produced by means of a lnear technology that uses two types of nputs, labelled = a; b, each one suppled by a d erent type of agent. All agents of a gven type have dentcal preferences and endowments; nstead, ther productvty can be d erent n the two types of regons. We start our analyss by assumng that the regons are nhabted by mmoble agents,.e., by ndvduals supplyng ther own producton factor only n ther regon of resdence. We then consder the mplcatons of factors moblty n Secton 4. In both types of regons, the local government provdes a publc good (namely, a publcly provded prvate good) usng a technology that transforms one unt of the prvate good nto one unt of the publc good. The prvate good serves also the role of consumpton good and that of numerare good. All markets are perfectly compettve. 2.1 Factors supples All agents are endowed wth an exogenous amount, normalzed to unty, of a good that can be ether drectly consumed or suppled as a producton factor. For a type- agent 5

8 resdent n a type-j regon, the endowment constrant s h j + x j = 1, where h j 0 s the quantty drectly consumed and x j 0 s the quantty suppled for producton. Let p j be the gross market prce of producton factor n a type-j regon. Snce markets are perfectly compettve, by lnearty of the producton functon, p j s constant and equal to the (exogenously gven) margnal productvty of producton factor n a type-j regon. Gross ncome, y j = p j x j, s taxed at source at a proportonal rate t j 2 [0; 1]. The ndvdual budget constrant s then equal to c j = (1 t j )y j, where c j s the consumpton good purchased n the market (the output, and numerare, good, de ned above). Preferences of a type- agent for consumpton bundles (c j ; h j ) are represented by the utlty functon: u (c j ; h j ) = c j + (h j ), (1) where the strctly concave functon (:) represents the utlty of drect consumpton of the endowment. Quas lnearty of the utlty functon (1) mples that all ncome e ects fall on the demand for market consumpton. 2 Takng t j and p j as gven, the representatve type- ndvdual solves: max x j (1 t j )p j x j + (1 x j ). (2) Denote wth h and hh the rst- and the second-order dervatves of, respectvely. The equlbrum factor supply, ~x j (p j ; t j ), that solves problem (2) s de ned by the rst order condton: (1 t j )p j = h (1 x j). (3) To ensure an nteror soluton, ~x j 2 (0; 1), assume that lm xj!1 h (1 x j) = +1, h (1) = 0. Factor s supply s ncreasng n p j =@p j = (1 t j )= hh (~ h j ) > 0, and decreasng n t j =@t j = p j = hh (~ h j ) < 0. To smplfy the analyss, we assume that the elastcty of factors supply, " > 0, s constant. j = ~x j " 1 t j < 0. (4) By nsertng the equlbrum quanttes nto Eq. (1), the ndrect utlty functon of a type- agent resdent n a type-j regon s: v j (t j ) = (1 t j )p j ~x j + (1 ~x j ), (5) 2 The quas-lnearty assumpton s made for analytcal convenence and can be relaxed. For some types of producton factors, the emprcal estmates of the mpact of ncome taxes on ther supply nd weak ncome e ects (see, e.g., Gruber and Saez, 2002, for the case of labor supply). 6

9 where, by applyng the envelope theorem, t j = p j ~x j < 0. (6) The supply of producton factors modelled above can be nterpreted n famlar ways. If the endowment s expressed n unts of tme, and x j s labor tme, h j s lesure tme, and p j s the wage rate, then one obtans the standard neoclasscal model of labor supply. As another example, consder a two-perod framework wth an endowment that n the rst perod can be allocated to consumpton, h j, and nvestment, x j. Let p j be the second-perod return of nvestment (ncludng the prncpal). Then, n the absence of second-perod bequests, c j s second-perod consumpton, and the model can be nterpreted as one of captal supply. Note that, strctly speakng, the use of dstortonary ncome taxes s unjust ed n the present setup. Snce taxpayers are dvded nto two dstnct, and observable, groups of dentcal agents, the e cent nstrument would be a unform lump sum tax f the only objectve s publc good nancng, or a par of group-spec c lump sum taxes f also some degree of between-groups redstrbuton s deemed as desrable. However, the am of ths paper s not that of justfyng the use of second-best dstortonary taxes n place of rst-best lump sum nstruments, whch s usually the focus of optmal tax models. Instead, the objectve s the analyss of the e cency and equty propertes of transfer programs that are based on the equalzaton of scal capactes of local governments nancng ther own publc servces wth dstortonary taxes; to ths end, and to avod unnecessary analytcal complcatons, we use a smple two-group, dentcalagents, setup. 3 Note, nally, that n the economy descrbed above, n whch there are two producton factors and a sngle prvate good produced, the use of group-spec c ncome taxes makes redundant the use of any other tax nstrument. In fact, a proportonal sales tax leved on producers, or a proportonal consumpton tax, s equvalent to a unform ncome tax (.e., t a = t b ) on the supplers of producton factors. Thus, output and consumpton taxes are embedded as specal cases of producton factors ncome taxaton. 4 3 Also Smart (1998) uses dstortonary taxes n a settng n whch lump sum taxes are feasble. 4 The equvalence between consumpton and ncome taxaton requres somewhat strong condtons f there s tax competton among local governments. Income taxes collected at source are equvalent commodty taxes under the orgn prncple f factors moblty nduced by d erentals n regonal ncome tax rates bear the same mpact on local tax revenues as consumers cross border shoppng n response to d erentals n regonal commodty tax rates. 7

10 2.2 Tax revenues and equalzaton grants Let n j > 0 be the number of type- agents supplyng ther own producton factor n a type-j regon. Let t j = (t aj ; t bj ). Let z kj = n kjp kj ~x kj n aj + n bj, (7) be the per capta tax base of producton factor k n a type-j jursdcton, and let z k = P 2 r=1 n krp kr ~x kr P 2 r=1 (n ar + n br ), (8) be the natonal per capta tax base of producton factor k. In a type-j jursdcton, the per capta tax revenues are equal to: T j (t j ) = X k2fa;bg t kj z kj, (9) and the per capta equalzaton grant takes the followng form: where E j (t j ) = X k2fa;bg B k t k (z k z kj ) + X k2fa;bg R k ( t k z k t kj z kj ), (10) t k = P 2 r=1 t krn kr p kr ~x kr P 2 r=1 n krp kr ~x kr, (11) s the natonal average tax rate on producton factor k s ncome. The transfer mechansm de ned n Eq. (10) s based on the equalzaton of scal capactes. The parameters S = ( S a; S b ), S k 2 [0; 1], B k R k = 0, S = B; R, k = a; b, express the degree of equalzaton of the scal capacty system. In partcular, for each tax source, ether one of two scal capacty measures are used: tax-base (B) or taxrevenue (R). Note that whle we do not allow for a mxed mechansm for any sngle tax source, we do allow for a d erent mechansm on the two tax sources. The grant formula shows that, for producton factor k, gven the natonal per capta tax base, z k, and the natonal average tax rate, t k, under tax-base equalzaton (TBE) the grant covers the share B k of the gap between the standardzed natonal per capta tax revenue, t k z k, and the standardzed per capta tax revenue, t k z kj, of the gven typej jursdcton. Alternatvely, under tax-revenue equalzaton (TRE), the grant covers the share R k of the gap between the standardzed natonal per capta tax revenue, t k z k, and the e ectve per capta tax revenue, t kj z kj, of the gven type-j jursdcton. Both equalzaton systems are of the net type, snce the transfers cashed by the recpent 8

11 jursdctons (those wth below-the-average scal capactes) are nanced by the donor jursdctons (those wth above-the-average scal capactes), and n the aggregate total transfers sum up to zero (n contrast, the Canadan RTS system s of the gross type, by whch only postve transfers are pad to some Provnces). The two types of transfer mechansm have opposte e ects on the tax-settng ncentves of local governments. Under TBE, from the pont of vew of a sngle jursdcton, an ncrease n ts own tax rate t kj determnes, coeters parbus, an ncrease n the grant, snce the per capta tax base z kj s decreasng n t kj. That s, the equalzaton program gves ncentves to local governments to expand taxaton, an ssue clearly exposed by Smart (1998). On the contrary, under TRE, an ncrease n ts own tax rate t kj determnes, coeters parbus, a reducton n the grant, snce the per capta e ectve tax revenues t kj z kj are ncreasng n t kj, provded that, of course, taxaton s on the ncreasng sde of the La er curve. Note also that, n prncple, the transfers depend not only on the tax rates set by the recevng jursdcton, but also on those set by all other jursdctons, snce the latter a ect the natonal per capta tax bases, z k, and the natonal average tax rates, t k, k = a; b. However, recall that we are focusng the analyss on a large number of local jursdctons of each type. Ths means that any sngle jursdcton s small wth respect to the federaton to whch t belongs, and thus t takes the average values (z k ; t k ) as exogenously gven when settng ts own scal polcy. That s, any gven type-j jursdcton perceves the mpact of a change n ts own tax rate t j on the equalzaton transfer as beng equal j = z j B t j 1 t j z j R 1 t j " 1 t j. (12) In a type-j jursdcton, per capta publc good supply s equal to ts per capta publc resources, T j (:) + E j (:). By d erentatng Eq. (9), and combnng wth Eq. (12), we j j = z j 1 R 2.3 Socal welfare 1 R tj B t! ". (13) 1 t j The welfare of a type- agent supplyng ts own producton factor n a type-j regon s: w j (t j ) = v j (t j ) + [T j (t j ) + E j (t j )], (14) where v j (:) s the utlty of prvate consumpton, de ned n Eq. (5), and > 1 s the margnal utlty of publc expendture, whch s assumed to be constant and exogenously 9

12 gven. 5 Recall also that we have assumed that local publc servces take the form of a publcly provded prvate good; hence, ndvdual welfare depends on per capta local publc expendture, T j + E j. Usng a Utltaran crtera to aggregate the ndvdual utltes, socal welfare n a type-j jursdcton s de ned as: W j (t j ) = X n kj w kj (t j ). (15) k2fa;bg 2.4 Taxpayers lobbyng At the local level, we frame taxpayers lobbyng actvtes wthn the conventonal buyng n uence approach, orgnally conceved by Bernhem and Whnston (1986a, 1986b) and then further developed by Grossman and Helpman (1994, 2001) and Dxt et al. (1997). We setup a perfect-nformaton common-agency game wth three players: two prncpals (the lobbysts, each one representng the nterests of one group of taxpayers) and one agent (the local polcy maker). Interest groups compete to sway tax polcy to ther advantage by openly o erng legal monetary rewards (n the form, for nstance, of campagn contrbutons) to the polcy maker. Formally, the game presents two stages. In the rst, the taxpayers lobbysts announce to the polcy maker a menu of monetary o ers whch are contngent on scal polces. Followng acceptance of the o ers, n the second stage the polcy maker takes her preferred choce under the n uence of the ncentves provded by lobbysts. The game can be solved backward usng the noton of subgame perfecton. Followng Dxt et al. (1997), we assume that the contrbuton functon presented by a lobbyst to the polcy maker s of the truthful, or compensatng, type. A truthful contrbuton s a contnuous functon of the polcy nstruments that s de ned along an nd erence curve of the lobbyst; hence, a change n a polcy nstrument determnes a change n the contrbuton o ered to the polcy maker that re ects one-to-one the mpact of the polcy change on the lobbyst s welfare. Truthful contrbutons are a helpful devce for solvng ths class of common agency games, snce the set of the best responses of each prncpal to the contrbuton functons (not necessarly truthful) of the other prncpals always contans a truthful contrbuton schedule (Dxt et al., 1997, Proposton 2). Hence, there s no loss of generalty n usng ths type of contrbuton functons. Moreover, wth truthful contrbutons the equlbrum of the game s often unque, whle n general common agency games admt multple equlbra. 5 The analyss can be easly generalzed, albet at the cost of some analytcal complexty, to the case n whch the margnal bene ts of the publc good are a decreasng functon of publc expendture. 10

13 In a type-j jursdcton, the truthful contrbuton functon o ered by group lobbyst s, n per capta terms: c j (t j ; j ; ^ j ) = j max f0; w j (t j ) ^ j g, (16) where ^ j 0 s a scalar representng the per capta net payo of the lobby group. The parameter j 2 [0; 1] represents a measure of the e ectveness of group n ts lobbyng actvty n a type-j regon: the group s fully powerful f j = 1; t s completely ne ectve f j = 0. Followng the lterature quoted above, we do not model how a group organzes ts lobbyng actvty n order to provde ts members wth the ncentves to pursue the common nterest and to overcome free rdng behavor. We smply assume the exstence of a lobbyst for each group of taxpayers, wth an exogenously gven capacty to n uence polcy makng, as represented by the parameter j. The local polcy maker cares both for socal welfare, de ned n Eq. (15), and for poltcal contrbutons, de ned n Eq. (16). Her objectve functon s assumed to be lnearly addtve, wth unform weghts, n the two components. Ignorng the nonnegatvty constrant on the contrbuton functons, the objectve functon of the local polcy maker n the second stage of the lobby game can be wrtten as: 6 j (t j ) = X n kj (1 + kj )w kj (t j ). (17) k2fa;bg By lookng at Eq. (17), we mmedately note that poltcal pressures mpact on scal polcy only f groups a and b are not equally e ectve n lobbyng (.e., f aj 6= bj ), so that the polcy maker ends up maxmzng a dstorted socal welfare functon. Instead, f the two groups are equally e ectve n lobbyng (.e., f aj = bj ), then lobbyng does not mpact on scal polcy. Note also that the objectve functon (17) s compatble wth a poltcal settng n whch tax polcy s drven by electoral competton nstead of by specal nterest groups. In partcular, an objectve functon smlar to that shown n Eq. (17) can be obtaned wthn a standard probablstc votng model n whch two partes run for electon wth the am of maxmzng ther vote share, wth voters of groups a and b deologcally dspersed n favor of one of the two partes. In ths settng, the weght kj represents a 6 The lobby game can be solved by rst computng the equlbrum n the polcy varables at the second stage by maxmzng Eq. (17), whch gnores the non-negatvty constrants on poltcal contrbutons. The latter can then be checked ex-post, after havng computed the equlbrum net payo s of the lobby groups, (^ aj; ^ bj), at the rst stage of the game. However, n the present work we do not solve for the rst stage of the game, snce we are not nterested n the dstrbutonal e ects of lobbyng due to the monetary transfers from the lobby groups to the polcy maker. We focus only on the second stage of the game, where lobbyng mpacts on scal polcy. 11

14 measure of the poltcal clout of group k n regon j, as a functon of the mass of swng voters (see Persson and Tabelln, 2000, for a textbook exposton of the probablstc votng model, and Dxt and Londregan, 1996, for an applcaton to redstrbutve tax polcy). 3 Fscal equalzaton wth mmoble factors In ths secton, we characterze the scal polces set by local governments under the n- uence of taxpayers lobbyng, and then examne nter-regonal redstrbuton by means of scal equalzaton. 3.1 Equlbrum local scal polces Consder a type-j jursdcton. The local polcy maker maxmzes the objectve functon de ned n Eq. (17) wth respect to the own tax rates t j, takng as gven the other jursdctons tax rates and the average tax rates, and tax bases, at the natonal level. The rst order condtons are: j = n j (1+ j )p j ~x j + De nng: j = n aj aj + n bj bj n aj + n j X k2fa;bg n kj (1+ kj ) = 0, = a; b. (18) and usng Eq. (13), Eq. (18) can be manpulated to obtan: 1 R tj B t = 1 R (1 + j )=(1 + j ), j = 1; 2, = a; b. (19) 1 t j " In prncple, the equlbrum local scal polces of the federaton are de ned by a system of 4J equatons n the 4J tax rates set by the 2J jursdctons. However, snce the J jursdctons of type j, j = 1; 2, are dentcal, we can focus on the symmetrc equlbra n whch all jursdctons of the same type set the same scal polcy. Ths reduces the 4J-equaton system to a four-equaton system. However, note that snce the average tax rate t s a (non-lnear) functon of the tax rates (t 1 ; t 2 ), the equlbrum tax rates (t 1 ; t 2 ), for gven, are mplctly de ned by a two-equaton system. Note also that the equlbrum tax rate t j depends on the entre vector,, of the lobby 7 We assume that the second order condton for a maxmum holds at the equlbrum tax rates de ned below n Eq. (19). Second order condtons are sats ed n the numercal smulatons presented n Table 2 Secton 3.3, below. 12

15 weghts, = ( a1 ; b1 ; a2 ; b2 ), but t depends only on the equalzaton coe cents, B and R, of the correspondng tax base. Eqs. (19) hghlght, on a coeters parbus bass, the factors drvng tax choces. Frst, taxaton on factor s lower the hgher s ts elastcty of (compensated) supply ". Ths s a Ramsey-type e cency argument, by whch the mnmzaton of the excess burden of taxaton requres a taxaton level whch s nversely related to the elastcty of the tax base. Second, the hgher the margnal bene ts of publc expendture, the hgher the tax rates are. Note that, absent lobbyng actvtes and scal equalzaton, local taxaton s second-best e cent, wth tax rates set at a level below the blss pont of the La er curve: t j 1 t j = 1 ". (20) The other two factors bearng on taxaton are lobbyng and scal equalzaton. If, say, group b s more powerful than group a n lobbyng,.e., f bj > j > aj, then lobbyng dstorts taxaton downward on group b and upward on group a; the opposte otherwse, f t s group a the more capable n lobbyng. Fnally, the hgher the degree B of scal equalzaton on tax base, the hgher ts equlbrum tax rate s, snce TBE gves ncentves to local governments to expand taxaton. On the contrary, the hgher the degree R of scal equalzaton on tax revenues, the lower ts equlbrum tax rate s, snce TRE gves ncentves to local governments to reduce taxaton. Note that, n the absence of lobbyng, type-1 and type-2 regons apply n equlbrum the same tax rate on tax base, snce j = 0, = a; b, j = 1; 2, mples that the r.h.s. of Eq. (19) s ndependent of jursdctons type j. Instead, n the presence of lobbyng, type-1 and type-2 jursdctons apply n general d erent tax rates on the same type of producton factor, unless the lobbyng power s unformly dstrbuted n the federaton between groups a and b. The latter condton s formally de ned n the followng assumpton. Assumpton 1 (Unform dstrbuton of lobbyng power across regons) The absolute lobbyng power, and the relatve sze, of groups a and b are the same n type-1 and type-2 jursdctons. That s: aj = a, bj = b, n aj n bj =, j = 1; 2. (21) The followng proposton derves the equlbrum tax rates under Assumpton 1. 13

16 Proposton 1 If lobbyng power s unformly dstrbuted across regons, then the equlbrum tax rate on type- producton factor s the same n type-1 and type-2 jursdctons. That s, t j = t, j = 1; 2, where: t 1 t = 1 R (1 + )=(1 + ) (1 B R, = a; b, a + b = )" + 1. (22) Proof. By the condtons gven n Eq. (21), the r.h.s. of Eq. (19) s ndependent of j and therefore also t j on the l.h.s. s ndependent of j. Thus t 1 = t 2 = t = t and Eq. (19) reduces to Eq. (22). A stuaton wth unform lobbyng greatly smpl es the normatve analyss on optmal equalzaton, snce the equlbrum tax rates de ned n Eq. (22) are explctly expressed n terms of the model s parameters. Moreover, t s not a too restrctve assumpton, snce t requres only that groups a and b are equally e ectve n lobbyng across regons, and that ther relatve sze s the same, n both types of regons. Assumpton 1 admts, nstead, stuatons n whch type-1 and type-2 regons are heterogeneous n terms of total populaton and producton factors productvty. Hence, we mantan the unform lobbyng assumpton throughout the analyss. We denote wth t = (t a; t b ) the vector of the equlbrum tax rates de ned by Eq. (22), and we use starred varables to denote ther correspondng values after substtuton for the equlbrum tax rates. Hence, for nstance, x j = ~x j(p j ; t ) denotes the equlbrum supply of type- factor n type-j regons, and Tj = T j(t ) denotes the equlbrum tax revenues n type-j regons. A nal restrcton we ntroduce nto the analyss concerns the sze of the margnal bene ts of publc expendture. Assumpton 2 1 R > (1 + )=(1 + ), = a; b. Recall that, n Secton 2.3, we assumed > 1; otherwse, t s ne cent to provde publc goods, snce the margnal utlty of prvate consumpton s equal to unty by assumpton. In the absence of lobbyng dstortons and of TRE, > 1 ensures that both tax rates are postve, see Eq. (22). However, ether one or both tax rates can be negatve f R > 0 or f a 6= b. Snce equalzng subsdes means redstrbuton from low to hgh scal capacty regons, we appeal to Assumpton 2 to ensure that taxaton s postve on both tax bases. In fact, the condton s not too restrctve, snce, as we show below, t can be optmal to use TRE,.e., to set R > 0, only on the less powerful lobby group,.e., that for whch (1 + )=(1 + ) < 1. 14

17 3.2 Regonal heterogenety and socal welfare By nsertng the equlbrum tax rates, t j = t, nto Eq. (15), socal welfare n a type-j regon, as a functon of the equalzaton parameters, s equal to: Wj () = X n kj fv kj (t k ) + [T j(t ) + E j (t )]g, j = 1; 2, (23) k2fa;bg where = B a ; B b ; R a ; R b s the vector of the equalzaton parameters. Recall that all agents of a gven type have dentcal preferences and endowments, no matter ther regon of resdence. However, ther productvty can be d erent n type- 1 and type-2 regons. Hence, n per capta terms, any gap n the per capta regonal socal welfare levels can be due only to d erences n producton factors productvty. And snce there are two types of producton factors and two types of regons, there are only two possble cases. Ether both types of producton factors are more productve n one type of regons than n the other one, or each type of producton factor s more productve n one type of regons and less productve n the other, and vce versa. Our choce s to focus the analyss on the rst case only, snce t mples a non-ambguous rank between type-1 and type-2 regons n terms of per capta socal welfare, hence delverng clear-cut results, whle there s no value added n examnng also the second case. To smplfy the exposton, we assume, wthout loss of generalty, that the more productve regons are type-2. Assumpton 3 Both types of producton factors are more productve n type-2 than n type-1 regons. That s, p 1 < p 2, = a; b. An mplcaton of the hypothess ntroduced so far s that n the absence of central government nterventon the per capta socal welfare n type-2 regons s hgher than that n type-1 regons. Fscal equalzaton redstrbutes from rch to poor regons. However, t can never determne re-rankng between hgh and low productvty regons, n terms of ether per capta scal capacty, ndvdual welfare and per capta regonal welfare. These propertes are stated formally n the followng lemma (see Appendx A.1 for the proof). Lemma 1 Assume A.1, A.2 and A.3. Then, for any degree, S 2 [0; 1), of scal equalzaton, S = B; R, the per capta tax base on producton factor, the welfare of type- agents, and the per capta regonal socal welfare, are strctly hgher n type-2 than n type-1 regons. That s: z 1 < z 2, w 1 < w 2, = a; b; W 1 n a1 + n b1 < W 2 n a2 + n b2. 15

18 By aggregatng over type-1 and type-2 regons (recall that there s an equal number of regons of each type), and takng nto account that the equalzaton transfers sum up to zero n the aggregate,.e., 2X (n aj + n bj )E j (t ) = 0, j=1 the natonal socal welfare, W = P 2 j=1 W j, s expressed as: W () = 2X X j=1 k2fa;bg n kj vkj (t kj ) + T j(t ). (24) Gven that () the margnal utlty of prvate consumpton s constant and unform for all types of agents n all regons, and () the margnal utlty of local publc expendture s also constant and unform n all regons, the natonal socal welfare functon (24) shows no preference for redstrbuton, nether at the ndvdual nor at the regonal level; ndeed, t s nvarant to any knd of balanced-budget lump sum transfers among ndvduals or regons. It s, nstead, senstve to tax dstortons. Hence, Eq. (24) represents an approprate measure of the mpact of scal equalzaton on e cency. 3.3 Fscal equalzaton: e cency and equty We are now ready to examne the choce of the scal equalzaton system, and the determnaton of ts parameters, by the central government. We assume that, when settng the equalzaton program, the central polcy maker accounts for the n uence (f any) of the lobby groups on local scal polces. On the other hand, taxpayers do not attempt to n uence the polcy choces at the central level. In partcular, we assume that the central authorty takes a Rawlsan vew n terms of per capta socal welfare at the regonal level. Wthn our framework, the max-mn crteron amounts to maxmzng socal welfare of type-1 regons, snce scal equalzaton can never determne, by Lemma 1, re-rankng between type-1 and type-2 regons n terms of per capta socal welfare. To ease the presentaton of the results, we also assume, wthout loss of generalty, that one of the two groups s always (weakly) more powerful than the other one n lobbyng. Assumpton 4 Group a s always less powerful, at most equally powerful, than group b n lobbyng. That s, a b. 16

19 To compute the mpact on a type-j regon s socal welfare of an ncrease n scal equalzaton, we d erentate Eq. (23) wth respect to S = n j p j S + (n aj + n bj S S, S = B; R. (25) Recall that, under the condtons of Proposton 1, t j = t = t, j = 1; 2. Hence, n equlbrum, tax revenues and equalzaton grants are equal to: Tj (t j) = X t k z kj, E j (t j) = X B k + R k t k z k z kj, k2fa;bg k2fa;bg S = z j t 1 1 S, S = B; R, = t z z j + B + R z z j t 1 1 t, S = B; R. S Usng the fact that, by Eq. (22), t " 1 t = 1 R 1 B R, = , and usng Eqs. (26)-(27), Eq. (25) can be nally wrtten =@S = t z z j + B n aj + n bj + R z z j B 1 S + +z j 1 ( 1) B + 1 B S, S = B; R. (28) Recall that, for each tax base, only one type of equalzaton system s employed, ether tax-base (B) or tax-revenue (R). Hence, Eq. (28) can be spec ed for each equalzaton system as =@B = z z n aj + n bj j t + z z B j R {z } 1 B B + mechancal e ect {z } (equty) behavoral e ect (equty) + z j 1 ( 1) 1 B, (29) {z } behavoral e ect (e =@R = z z n aj + n bj j t + z z j B {z } 1 R R + mechancal e ect {z } (equty) behavoral e ect (equty) + z j R. (30) {z } behavoral e ect (e cency) 17

20 tax base = a tax base = b TBE TRE TBE TRE regons regons regons regons j = 1 j = 2 j = 1 j = 2 j = 1 j = 2 j = 1 j = 2 mechancal e ect: equty behavoral e ect: equty + (*) (*) + + (*) (*) + behavoral e ect: e cency ( ) + f R a < R + f B b < R f R a > R f B b > R (*) f B < =. ( ) R:E: a and B:E b are de ned n Eq. (33) Table 1: Fscal equalzaton: equty and e cency. Provded that the second order condtons hold, 8 and that the soluton s unque, the optmal equalzaton parameters, B, R, are obtaned by =@B = 0 ) =@R = 0 ) R. n aj + n bj n aj + n bj R =0 B =0 For each tax base, ether B or R s chosen, dependng on whch equalzaton system gves hgher socal welfare for type-1 regons. 9 Dervatves (29) and (30) hghlght the key factors drvng the choce of the scal equalzaton system, by showng that an ncrease n the degree of scal equalzaton, of ether types, bears three e ects on socal welfare, whch are summarzed n Table 1. The rst e ect s mechancal, expressng the margnal transfer of resources from rch to poor regons of a margnal ncrease n scal equalzaton, for gven levels of local tax revenues. Under Assumpton 3, these terms are postve for type-1, and negatve for type-2, regons, snce z > z 1, z < z 2. The other two e ects are behavoral, for they are due to the mpact of a margnal change n the degree of scal equalzaton on local tax rates and tax bases. One e ect mpacts on equty, the other one on e cency. For 8 We do not state formally the condtons for the optmal equalzaton problem to be concave. The numercal model presented n Table 2, for nstance, s well-behaved. 9 Note that, gven the structure of our model, by whch taxaton on type-a agents mpacts on the welfare of type-b agents only through the supply of publc goods (and vceversa), the choce of the best equalzaton system can be done ndependently on each tax base. 18

21 both, the sgn of ther mpact on socal welfare crucally depends on the sgn of the mpact of scal equalzaton on the equlbrum tax whch s postve for TBE and negatve for TRE. We examne the roles of scal equalzaton by consderng cases of ncreasng order of complexty. We start by focusng on e cency, that s by abstractng from the redstrbutve mpact of the equalzaton transfers. Ths amounts to maxmzng the natonal socal welfare functon, de ned n Eq. (24), or, equvalently, the aggregate of Eqs. (29)-(30) over j = 1; 2, gvng rst order condtons that depend only the behavoral e B = z 1 ( 1) R =0 1 B = R = z 1 + B =0 1 R = 0. (32) From these equatons, t s mmedate to see that, f lobbyng does not dstort local tax rates (snce groups a and b are equally powerful n lobbyng, so that a = b = 1), then the soluton s B = R = 0. The reason s obvous. Absent lobbyng dstortons, local taxes are optmally set, and scal equalzaton would ntroduce a dstorton that s costly n terms of e cency. The ncentves provded by scal equalzaton on tax settng are nstead useful to reap e cency gans when taxaton s dstorted by lobbyng. Snce, by Assumpton 4, group b s a more powerful lobbyst than a, so that a < 1 < b, then taxaton s dstorted downward on tax base b and upward on a. Fscal equalzaton can then be used to provde adequate ncentves for brngng tax rates to ther e cent level, by employng TBE on tax base b and TRE on a. By solvng Eq. (31) for = b and Eq. (32) for = a, the optmal equalzaton parameters that brng tax rates at ther e cent levels de ned n Eq. (20) are equal to: 10 B:E: b = b 1 > 0, R:E: a = 1 a > 0. (33) 1 Now consder nter-regonal redstrbuton. In the absence of lobbyng dstortons, t s clear that TBE s superor to TRE n terms of equty. Indeed, under TBE both the mechancal and the behavoral e ect are postve for type-1 regons, whereas under TRE the former s postve but the latter s negatve. The explanaton s smple. Local taxes are on the ncreasng sde of ther La er curve. Hence, whle TBE, by stmulatng 10 Clearly, e cency can be obtaned also wth negatve TBE on tax base a, at rate ( a 1)=( 1) < 0, and wth negatve TRE on tax base b, at rate 1 they redstrbute n the wrong drecton, from poor to rch regons. b < 0. However, we dscard these solutons snce 19

22 an ncrease n tax rates and local tax revenues, ncreases nter-regonal redstrbuton, TRE, by reducng tax rates and revenues, reduces t. 11 Snce, as noted above, both equalzaton systems are costly n term of e cency (absent lobbyng dstortons), overall TBE s a better redstrbutve devce than TRE. Note also that, n ths stuaton n whch the ntal allocaton wthout central government nterventon s e cent, scal equalzaton faces a standard equty-e cency tradeo, snce the gans n terms of transfers to the poor regons must be balanced aganst the e cency costs n terms of tax dstortons. Fnally, we rentroduce lobbyng nto the pcture. As llustrated above, we know that e cency calls for TBE on the tax base backed by the more powerful group b and for TRE on that backed by the less powerful group a. We also know that TBE s preferable on equty grounds. By combnng these elements, we obtan the followng optmalty condtons. If lobbyng dstortons are not very mportant, because group b s not much more powerful than a, then TBE should be used on both tax bases, snce the man drver of scal equalzaton s equty. For gven level of productvty gap, p 2 p 1, between the two types of regons, the optmal degree of scal equalzaton depends on the measure, b a, of the dstance between groups n terms of lobbyng power: f the latter ncreases, t s optmal to ncrease the degree of TBE on tax base b and reduce t on tax base a, snce t s necessary to counteract a stronger pressure for low tax rates on tax base b, that n turn nduces local governments to ncrease taxaton on tax base a. Note also that equty and e cency do not con ct when TBE s ncreased on tax base b, whle they do when TBE s reduced on a. Ths means that there exsts a threshold of the lobby-power-gap, b a, above whch t becomes optmal, for tax base a, to shft from TBE to TRE, snce the latter serves well both e cency and equty objectves. 12 Note nally that, when lobbyng dstortons are mportant, scal equalzaton does not face an equty-e cency tradeo, snce the optmal polcy can mprove both equty (.e., per capta socal welfare n the poor regons) and e cency (.e., natonal socal welfare), or even obtan a Pareto mprovement (.e., an ncrease n 11 That local tax revenues are on the ncreasng sde of ther La er curve s clearly a condton for optmalty, both at the local and at the central level of government. In partcular, taxaton s always on the ncreasng sde of the La er curve under TRE, snce the latter lowers taxaton. Instead, under TBE, taxaton s on the ncreasng sde of the La er curve provded that the degree of equalzaton s below a gven threshold, snce TBE ncreases taxaton. Namely, the condton s B > 0, ensurng that the behavoral e ect (equty) n Eq. (29) s postve for type-1 regons and negatve for type The condtons for optmalty can be stated also as follows: for gven level, b a, of lobby-power gap, there exsts a threshold level of the productvty gap, p 2 p 1, such that TBE should be used on tax base b and TRE on a f the productvty gap s below the threshold level, whle TBE should be used on both tax bases f t s above t. 20

23 per capta socal welfare for both types of regons). 13 The rules governng the desgn of the scal equalzaton scheme n the presence of lobbyng are summarzed n Proposton 2 (ts proof s omtted, snce the arguments have been gven above). An llustraton of the results by means of a numercal example s gven n Table 2. Proposton 2 Assume A.1, A.2, A.3, A.4, and that scal equalzaton ams at maxmzng the per capta socal welfare of type-1 regons. Then, for a gven level of productvty gap between type-1 and type-2 regons, the optmal scal equalzaton polcy s governed by the followng rules. If groups a and b heterogenety n terms of lobbyng power, as measured by b a, s below a gven threshold, then TBE should be employed on both tax bases. The degree of equalzaton of tax base b s ncreasng, whle that of tax base a s decreasng, n b a. If groups a and b heterogenety n terms of lobbyng power s above a gven threshold, then TBE should be employed on tax base b whle TRE should be employed on tax base a. Both equalzaton coe cents are ncreasng n b a. Row 1 n Table 2 shows the equlbrum tax rates and socal welfare, n the absence of scal equalzaton, n a stuaton n whch lobbyng does not dstort scal polcy, snce the two groups are equally powerful ( b = a = :2). In order to hghlght the mpact of lobbyng, the smulaton assumes the same elastcty of supply for factors a and b (" b = " a = :2). Row 2 shows that TBE maxmzes socal welfare of type-1 regons wth a 51% equalzaton rate on both tax bases. Ths, however, dstorts taxaton upward from 37 to 54%, hence there s a reducton n natonal socal welfare. Row 3 shows that TBE at 51% on tax base b and TRE at 6% on tax base a s not as good, n equty terms, as TBE on both tax bases. The reason s that TRE dstorts taxes downward and also redstrbutes less than TBE. Rows 4-6 show what happens f group b s a more powerful lobbyst than a. Wth no equalzaton (row 4), taxaton s dstorted by lobbyng upward on group a, from 37 to 44%, and downward, from 37 to 27%, on b. Row 5 shows that TBE at 39% on tax base a and TBE at 66% on b reduces the gap between tax rates from 17 to 5 percentage ponts, thus partally counteractng the lobbyng dstorton; however, at the same tme both tax rates ncrease because of TBE. Row 6 shows that, n terms of type-1 regons 13 A Pareto mprovement at a regonal level does not mply, n general, a Pareto mprovement at the ndvdual level. Typcally, scal equalzaton, by lowerng taxaton on the less powerful lobby group a and by ncreasng t on the more powerful one b, favors the former and penalzes the latter. 21

24 a b B a B b Immoble producton factors R a t a1 t a2 t b1 t b2 W 1 W 2 W T 1 T 2 je j a b B a B b Moble producton factors R a t a1 t a2 t b1 t b2 W 1 W 2 W T 1 T 2 je j Model s parameters: n a1 = n b1 = n a2 = n b2 = 1, p a1 = p b1 = 1, p a2 = p b2 = 1:25, = 1:3, " a = " b = :4, F a (0) = F b (0) = :1, f a (0) = f b (0) = :5, R 0 1 df a = R 0 1 df b = :02. Model s spec caton: = 1 5 " 1+" x (1+")=" j, ~x j = 1 5 ((1 t j)p j ) " Table 2: Optmal scal equalzaton: some numercal examples. 22

25 socal welfare, a mxed system TBE-TRE s equvalent to one based only on TBE. Indeed, the mxed system s superor to a pure TBE system n terms of e cency, snce natonal socal welfare s hgher. Note, however, that the mxed system overturn the structure of taxaton, wth the more powerful lobby group beng taxed more heavly than the less powerful one. Fnally, rows 7-9 show that when there s hgh heterogenety n terms of lobbyng power, the best equalzaton system s the mxed one, both on equty and e cency grounds. TRE at 22% on tax base a and TBE at 80% on base b performs better than TBE at 31% on base a and at 80% on b. The remanng part of Table 2 refers to the model wth factors moblty, to whch we turn now. 4 Tax competton and tax exportng In ths secton, the analyss s extended by allowng for moblty of producton factors. In partcular, whle we contnue to assume that the place of resdence s xed for all agents, we allow them to choose the place where to earn ther ncome. 4.1 Moblty of producton factors We model factors moblty n a smple way, by assumng that each regon s connected wth a regon of the same type, so that the agents who are resdent n a gven type-j regon have the opportunty to supply ther own producton factor ether at home or abroad, n another, and only one, regon of the same knd. An agent supplyng her own producton factor n the foregn regon connected to her place of resdence sustans a lump sum cost, wth dstrbuted wthn the group of type- agents accordng to the densty functon f () 0 on the support = ( 1; +1), wth cumulatve dstrbuton denoted by F (). On the one hand, a postve can be nterpreted as the moblty cost sustaned for supplyng abroad nstead of at home. On the other hand, a negatve can be nterpreted as a net opportunty cost for not supplyng abroad; an nstance of a negatve s when the agent can enjoy a better workplace envronment abroad than at home, by an amount that more than compensates for the costs of commutng. Obvously, a reasonable restrcton to mpose on the dstrbuton of s F (0) < :5. That s, moblty costs are postve for more than 50% of the resdents n the regon; otherwse, t s hard to sustan the hypothess of xed resdence (.e., no mgraton). Note, however, that none of the results hnges on spec c restrctons on the dstrbuton of moblty costs. 23

26 Consder a type-(; ) agent who s resdent n a type-j regon. To llustrate, and wth a slght abuse of notaton, j denotes the home regon and the connected foregn regon. The agent chooses to supply at home f v j (t j ) v (t ), and abroad otherwse, where v r, r = j;, s the ndrect utlty functon de ned n Eq. (5). The number of type- agents supplyng at home s thus equal to: ~n jj (t j ; t ) = n j [1 F (v v j )], whle that of those supplyng abroad s equal to ~n j (t j ; t ) = n j ~n jj (t j ; t ). Specularly, the number of type- agents resdent n regon that supply n regon j s equal to: ~n j (t ; t j ) = n F (v j v ). Hence, the total number of type- agents supplyng n regon j s equal to: ~n j (t j ; t ) = ~n jj (t j ; t ) + ~n j (t ; t j ). (34) An ncrease n the home tax rate t j causes an out ow of producton factors from regon j to regon that s equal j = p j ~x j [n j f (v v j ) + n f (v j v )] 0. (35) Per capta tax revenues and equalzaton grants are de ned as n Eqs. (9)-(10), respectvely, wth ~n j, de ned n Eq. (34), that substtutes n j n the Eqs. (7), (8) and (11), de nng the average tax bases and tax rates, z kj, z k and t k, respectvely. For a type-j regon, an ncrease n ts own tax rate t j mpacts on tax tax revenues and equalzaton grants as follows: 1 j j = z j 1 R tj B t " 1 t j =@t j ~n j. (36) The type- agents that are resdent n a type-j regon are dvded nto two groups: that of those supplyng at home, n number ~n jj, wth welfare v j + (T j + E j ), and that of those supplyng abroad, n number ~n j, wth welfare v + (T j + E j ). Note that, whle ncome can be earned n the foregn regon, publc goods are always enjoyed n the home regon. Hence, although ~n jj agents supply at home and ~n j abroad, the number those bene ttng from publc goods s always equal to the number of resdents, n j. Aggregate socal welfare s thus equal to: ~W j (t j ) = X k2fa;bg ~n kjj v kj + ~n kj v k n kj Z vk v kj 1 df + n kj (T j + E j ). 24

27 4.2 Fscal equalzaton under tax competton and lobbyng We assume that only the resdent agents lobby ther own local polcy maker. That s, agents supplyng abroad lobby at home but not abroad. Hence, followng the same steps descrbed n Secton 2.4, the objectve functon of a type-j local polcy maker s: ~ j (t j ) = X k2fa;bg Z vk v kj (1 + k ) ~n kjj v kj + ~n kj v k n kj 1 df + n kj (T j + E j ). By maxmzng Eq. (37) wth respect to the own tax rates, takng as gven the tax rates set n the other jursdctons, the equlbrum local scal polces are obtaned. Note that, snce each regon s connected wth an dentcal regon of the same type, the equlbrum s always a symmetrc one, wth t j = t k, v j = v k, ~n jj = n j [1 F (0)], ~n j = n j F (0), ~n j = n j. In general, however, regons of d erent types set d erent tax polces. (37) Formally, the equlbrum tax rates on tax base are de ned by the followng two-equaton system n (t 1 ; t 2 ) (the dervaton s n Appendx A.2): 1 R tj B t = 1 R [1 F (0)] (1 + )=(1 + ), j = 1; 2. (38) 1 t j [" + 2(1 t j )p j ~x j f (0)] Eq. (38) represents a generalzaton of Eq. (19) to the case of moble producton factors. It shows that moblty produces two contrastng ncentves on tax settng by local governments. On the one hand, t determnes ncentves to lower taxaton, n the attempt to prevent the resdent producton factors to supply abroad and to attract factors from abroad, thereby augmentng the tax base. Ths e ect, whch s present n the denomnator of the r.h.s. of Eq. (38), s proportonal to the after-tax ncome, (1 t j )p j ~x j, multpled by the densty, f (0), of the agents that n equlbrum are nd erent between supplyng at home and abroad. Clearly, the hgher the densty f (0), the hgher s the elastcty of the tax base dependng on moblty, and hence the lower s the tax rate n equlbrum. On the other hand, moblty determnes ncentves to hgher taxaton, snce local governments can export part of ther tax burden by taxng non-resdent agents. Ths e ect, whch s present n the numerator of the r.h.s. of Eq. (38), s proportonal to the mass, F (0), of agents supplyng abroad n equlbrum. The hgher the mass of agents supplyng abroad, the hgher s taxaton n the attempt to export taxaton on non-resdents. Note that, although Eq. (38) s derved under Assumpton 1 of unform lobbyng power across the federaton, n equlbrum, under factors moblty type-1 and type-2 regons set d erent tax rates on the same tax base, snce, n general, p 1 ~x 1 < p 2 ~x 2 under Assumpton 3. Ths means that scal equalzaton, d erently from the case 25

28 of mmoble producton factors (see Eq. 33), cannot restore e cency f taxaton s dstorted by tax competton, tax exportng, or lobbyng. 14 We conclude by lookng at the results of the numercal example n Table 2. Rows 10 to 18 ntroduce factors moblty nto the correspondng rows 1 to 9 n whch factors are mmoble. By comparng the two parts of the Table, we see that n the absence of scal equalzaton taxaton s n general lower under moblty (rows 10, 13, 16) than under no-moblty (rows 1, 4, 7). Of course, ths s not a general result, snce t depends on the chosen parametrzaton of the example, n whch tax competton domnates over tax exportng. Under scal equalzaton, however, tax rates under moblty and under no-moblty are more smlar. Because of tax competton, TBE on both tax bases s the best equalzaton polcy even when groups a and b are hghly heterogeneous n terms of lobbyng power (rows 17-18). Intutvely, snce tax competton tends to lower taxaton, TBE s the best nstrument to reap e cency gans. Obvously, by gvng more weght to tax exportng, t s possble to construct examples n whch a mxed system TBE- TRE, or even a pure TRE system, s the best opton, snce n ths case e cency calls for gvng ncentves to lower taxaton, whch s n ated by tax exportng. 5 Concludng remarks In ths paper, we set up a smple publc nance model to examne whether and how poltcal pressures by specal nterest groups on local governments a ect ther scal choces and the desgn of optmal scal equalzaton schemes by the central authorty. Our man results show that t s optmal to rely exclusvely on tax-base equalzaton f the lobby groups are smlar n terms of lobbyng power. A mxed system s nstead optmal f the specal nterest groups are hghly heterogeneous n terms of lobbyng power, wth tax-base equalzaton on the tax base backed by the strong lobby group and tax-revenue on that backed by the weak one. The analyss was conducted under several assumptons, some of whch could be relaxed n future work to test the theoretcal robustness of the results. The hypothess of perfect substtutablty between producton factors can be relaxed by allowng for mperfect substtutablty, or to analyze the opposte polar case of perfect complementarty. In both cases, taxaton would bear general equlbrum e ects on the before-tax market returns of producton factors, thus determnng tax shftng between sectors. Fscal equalzaton focused on scal capacty, wth regons that are homogeneous 14 Of course, scal equalzaton can restore e cency f all regons are dentcal. However, ths s not an nterestng case, snce scal equalzaton s motvated n the rst place by the fact that regons are heterogeneous. 26

29 wth respect to expendture needs and servce cost provson. The analyss can be extended to account for these relevant factors of regonal heterogenety. Fnally, lobbyng was con ned only at the local level. For some producton factors, especally those that are hghly moble, t s probably a better strategy to lobby at the central level of government than at the local one. Moreover, t s also possble that local authortes lobby the central one to bend the equalzaton scheme towards ther nterests. Appendx A.1 Proof of Lemma 1 Recall that, by Assumptons 1 and 2, type-1 and type-2 regons set dentcal and postve tax rates t. Hence, by Assumpton 3, (1 t )p 1 < (1 t )p 2. Snce factors supples are upward slopng, the latter nequalty mples y1 < y 2, v 1 < v 2. Assumpton 1 (namely, type-a and type-b agents are n the same proporton n both types of regons) then mples z 1 < z 2. For < 1, per capta publc expendture, (T j + E j )=(n aj + n bj ), s hgher n type-2 than n type-1 regons, snce tax revenues are hgher whle the equalzaton mechansm does not fully equalzes total resources. Hgher welfare from prvate consumpton (v1 < v 2 ) and hgher bene ts from publc goods mply that all agents of a gven type resdent n type-2 regons are better o than those of the same type resdent n type-1 regons. That s, w1 < w 2. Ths also mples that the per capta socal welfare s hgher n type-2 than n type-1 regons. A.2 Dervaton of Eq. (38) By d erentatng Eq. (37) wth respect to t j, we get the rst order ~ j = (1 + ) ~n jj p j ~x j + v j + j + X k2fa;bg + n j (v v j )f(v v j )p j ~x (1 + k )n kj j +. (A.1) Snce the Nash equlbrum between the two dentcal connected regons, j and, s symmetrc, t s: v = v j as: (1 + )n j [1 F (0)] p j ~x j + X j, ~n jj = n j [1 F (0)]. Hence, Eq. (A.1) can be wrtten (1 + k )n kj j = 0. j Dvdng by n aj + n bj, usng Eq. (36), and dvdng by z j, we get: h 1 R 1 R t j B j =@t j 1 + t 1 t j ~n j 1 + [1 F (0)] = 0. (A.3) In a symmetrc equlbrum, t s: (@~n j =@t j )=n j = expresson nto Eq. (A.3), and rearrangng, we nally obtan Eq. (38). 2p j ~x j f (0). Substtutng the latter 27

30 References Albouy, D., Evaluatng the e cency and equty of federal scal equalzaton. Journal of Publc Economcs 96, Barett, C., Huber, B., Lchtblau, K., A tax on tax revenue: The ncentve e ects of equalzng transfers: Evdence from Germany. Internatonal Tax and Publc Fnance 9, Bernhem, B.D., Whnston, M.D., 1986a. Common agency. Econometrca 54, Bernhem, B.D., Whnston, M.D., 1986b. Menu actons, resource allocaton, and economc n uence. Quarterly Journal of Economcs 101, Blöchlger, H., Charbt, C., Fscal equalsaton. OECD Economc Studes No. 44, 2008/1. Blöchlger, H., Fscal equalsaton: A key to decentralsed publc nances. In: OECD, Fscal Federalsm 2014: Makng decentralsaton work. OECD Publshng. Boadway, R., Intergovernmental redstrbutve transfers: e cency and equty. In Ahmad, E., Broso, G., Handbook of Fscal Federalsm. Edward Elgar. Boadway, R., Flatters, F., E cency and equalzaton payments n a central system of government: a synthess and extenson of recent results. Canadan Journal of Economcs 15, Bordgnon, M., Manasse, P., Tabelln, G., Optmal regonal redstrbuton under asymmetrc nformaton. Amercan Economc Revew 91, Brusco, S., Colombo, L., Galmarn, U., Tax d erentaton, lobbyng, and welfare. Socal Choce and Welfare 42, Buchanan, J., Goetz, C., E cency lmts of scal moblty: an assessment of the Tebout model. Journal of Publc Economcs 1, Bucovetsky, S., Smart, M., The e cency consequences of local revenue equalzaton: tax competton and tax dstortons. Journal of Publc Economc Theory 8, Buettner T., The ncentve e ect of scal equalzaton transfers on tax polcy. Journal of Publc Economcs 90, Dahlberg M., Johansson, E., 2002, On the vote-purchasng behavor of ncumbent governments. Amercan Poltcal Scence Revew 96, Dahlby, B., Fscal externaltes and the desgn of ntergovernmental grants. Internatonal Tax and Publc Fnance 3, Dahlby, B., Wlson, L.S., Fscal capacty, tax e ort, and optmal equalzaton grants. Canadan Journal of Economcs 27, Dxt, A., Grossman, G.M., Helpman, E., Common agency and coordnaton: general theory and applcaton to government polcy makng. Journal of Poltcal Economy 105, Dxt, A., Londregan, J., The determnants of success of specal nterests n redstrbutve poltcs. The Journal of Poltcs 58, Dxt, A., Londregan, J., Ideology, tactcs, and e cency n redstrbutve poltcs. Quarterly Journal of Economcs 113, Esteller-Moré, A., Solé-Ollé, A., Tax settng n a federal system: The case of personal ncome taxaton n Canada. Internatonal Tax and Publc Fnance 9, Flatters, F., Henderson, V., Meszkowsk, P., Publc goods, e cency, and regonal scal equalzaton. Journal of Publc Economcs 3,

31 Gordon, R.H., Cullen, J.B., Income redstrbuton n a Federal system of governments. Journal of Publc Economcs 96, Grossman, G.M., Helpman, E., Protecton for sale. Amercan Economc Revew, 84, Grossman, G.M., Helpman, E., Specal nterest poltcs. The MIT Press, Cambrdge, MA. Gruber, J., Saez, E., The elastcty of taxable ncome: evdence and mplcatons. Journal of Publc Economcs 84, Hettch, W., Wner, S.L., Economc and poltcal foundatons of tax structure. Amercan Economc Revew 78, Koethenbuerger, M., Tax competton and scal equalzaton. Internatonal Tax and Publc Fnance 9, Kotsoganns, C., Schwager, R., Accountablty and scal equalzaton. Journal of Publc Economcs 92, Larcnese, V., Rzzo, L., Testa, C., Allocatng the U.S. Federal budget to the States: The mpact of the Presdent. The Journal of Poltcs 68, Levtt S., Snyder, J., 1995, Poltcal partes and the dstrbuton of federal outlays. Amercan Journal of Poltcal Scence 39, Lockwood, B., Inter-regonal nsurance. Journal of Publc Economcs 72, Persson, T., Tabelln, G., Poltcal economcs: explanng economc polcy. MIT Press. Rzzo, L., Local government responsveness to federal transfers: theory and evdence. Internatonal Tax and Publc Fnance 15, Sato, M., Fscal externaltes and e cent transfers n a federaton. Internatonal Tax and Publc Fnance 7, Smart, M., Taxaton and deadweght loss n a system of ntergovernmental transfers. Canadan Journal of Economcs 31, Smart, M., Rasng taxes through equalzaton. Canadan Journal of Economcs 40, Solé-Ollé, A., Sorrbas-Navarro, P., The e ects of partsan algnment on the allocaton of ntergovernmental transfers. D erences-n-d erences estmates for Span. Journal of Publc Economcs 92,

32 Documents de Treball de l IEB /1, Oppedsano, V; Turat, G.: "What are the causes of educatonal nequaltes and of ther evoluton over tme n Europe? Evdence from PISA" 2011/2, Dahlberg, M; Edmark, K; Lundqvst, H.: "Ethnc dversty and preferences for redstrbuton" 2011/3, Canova, L.; Vaglo, A.: "Why do educated mothers matter? A model of parental help 2011/4, Delgado, F.J.; Lago-Peñas, S.; Mayor, M.: On the determnants of local tax rates: new evdence from Span 2011/5, Polatto, A.; Schuett, F.: A model of musc pracy wth popularty-dependent copyng costs 2011/6, Duch, N.; García-Estévez, J.; Parellada, M.: Unverstes and regonal economc growth n Spansh regons 2011/7, Duch, N.; García-Estévez, J.: Do unverstes affect frms locaton decsons? Evdence from Span 2011/8, Dahlberg, M.; Mörk, E.: Is there an electon cycle n publc employment? Separatng tme effects from electon year effects 2011/9, Costas-Pérez, E.; Solé-Ollé, A.; Sorrbas-Navarro, P.: Corrupton scandals, press reportng, and accountablty. Evdence from Spansh mayors 2011/10, Cho, A.; Calero, J.; Escardíbul, J.O.: Hell to touch the sky? Prvate tutorng and academc achevement n Korea 2011/11, Mra Godnho, M.; Cartaxo, R.: Unversty patentng, lcensng and technology transfer: how organzatonal context and avalable resources determne performance 2011/12, Duch-Brown, N.; García-Quevedo, J.; Montolo, D.: The lnk between publc support and prvate R&D effort: What s the optmal subsdy? 2011/13, Breullé, M.L.; Duran-Vgneron, P.; Samson, A.L.: To assemble to resemble? A study of tax dspartes among French muncpaltes 2011/14, McCann, P.; Ortega-Arglés, R.: Smart specalsaton, regonal growth and applcatons to EU coheson polcy 2011/15, Montolo, D.; Trllas, F.: Regulatory federalsm and ndustral polcy n broadband telecommuncatons 2011/16, Pelegrín, A.; Bolancé, C.: Offshorng and company characterstcs: some evdence from the analyss of Spansh frm data 2011/17, Ln, C.: Gve me your wred and your hghly sklled: measurng the mpact of mmgraton polcy on employers and shareholders 2011/18, Banchn, L.; Revell, F.: Green poltes: urban envronmental performance and government popularty 2011/19, López Real, J.: Famly reunfcaton or pont-based mmgraton system? The case of the U.S. and Mexco 2011/20, Boglacno, F.; Pva, M.; Vvarell, M.: The mpact of R&D on employment n Europe: a frm-level analyss 2011/21, Tonello, M.: Mechansms of peer nteractons between natve and non-natve students: rejecton or ntegraton? 2011/22, García-Quevedo, J.; Mas-Verdú, F.; Montolo, D.: What type of nnovatve frms acqure knowledge ntensve servces and from whch supplers? 2011/23, Banal-Estañol, A.; Macho-Stadler, I.; Pérez-Castrllo, D.: Research output from unversty-ndustry collaboratve projects 2011/24, Lgthart, J.E.; Van Oudheusden, P.: In government we trust: the role of fscal decentralzaton 2011/25, Mongran, S.; Wlson, J.D.: Tax competton wth heterogeneous captal moblty 2011/26, Caruso, R.; Costa, J.; Rccut, R.: The probablty of mltary rule n Afrca, /27, Solé-Ollé, A.; Vladecans-Marsal, E.: Local spendng and the housng boom 2011/28, Smón, H.; Ramos, R.; Sanromá, E.: Occupatonal moblty of mmgrants n a low sklled economy. The Spansh case 2011/29, Polatto, A.; Trotn, G.: Optmal tax enforcement under prospect theory 2011/30, Montolo, D; Polatto, A.: Fnancng publc educaton when altrustc agents have retrement concerns 2011/31, García-Quevedo, J.; Pellegrno, G.; Vvarell, M.: The determnants of YICs R&D actvty 2011/32, Goodspeed, T.J.: Corrupton, accountablty, and decentralzaton: theory and evdence from Mexco 2011/33, Pedraja, F.; Cordero, J.M.: Analyss of alternatve proposals to reform the Spansh ntergovernmental transfer system for muncpaltes 2011/34, Jofre-Monseny, J.; Sorrbas-Navarro, P.; Vázquez-Grenno, J.: Welfare spendng and ethnc heterogenety: evdence from a massve mmgraton wave 2011/35, Lyytkänen, T.: Tax competton among local governments: evdence from a property tax reform n Fnland 2011/36, Brülhart, M.; Schmdheny, K.: Estmatng the Rvalness of State-Level Inward FDI 2011/37, García-Pérez, J.I.; Hdalgo-Hdalgo, M.; Robles-Zurta, J.A.: Does grade retenton affect achevement? Some evdence from Psa 2011/38, Boffa, f.; Panzar. J.: Bottleneck co-ownershp as a regulatory alternatve

33 Documents de Treball de l IEB 2011/39, González-Val, R.; Olmo, J.: Growth n a cross-secton of ctes: locaton, ncreasng returns or random growth? 2011/40, Anes, V.; De Donder, P.: Votng under the threat of secesson: accommodaton vs. represson 2011/41, D Petro, G.; Mora, T.: The effect of the l Aqula earthquake on labour market outcomes 2011/42, Brueckner, J.K.; Neumark, D.: Beaches, sunshne, and publc-sector pay: theory and evdence on amentes and rent extracton by government workers 2011/43, Cortés, D.: Decentralzaton of government and contractng wth the prvate sector 2011/44, Turat, G.; Montolo, D.; Pacenza, M.: Fscal decentralsaton, prvate school fundng, and students achevements. A tale from two Roman catholc countres /1, Montolo, D.; Trujllo, E.: "What drves nvestment n telecommuncatons? The role of regulaton, frms nternatonalzaton and market knowledge" 2012/2, Gesen, K.; Suedekum, J.: "The sze dstrbuton across all ctes : a unfyng approach" 2012/3, Foremny, D.; Redel, N.: "Busness taxes and the electoral cycle" 2012/4, García-Estévez, J.; Duch-Brown, N.: "Student graduaton: to what extent does unversty expendture matter?" 2012/5, Durán-Cabré, J.M.; Esteller-Moré, A.; Salvador, L.: "Emprcal evdence on horzontal competton n tax enforcement" 2012/6, Pckerng, A.C.; Rockey, J.: "Ideology and the growth of US state government" 2012/7, Vergoln, L.; Zann, N.: "How does ad matter? The effect of fnancal ad on unversty enrolment decsons" 2012/8, Backus, P.: "Gbrat s law and legacy for non-proft organsatons: a non-parametrc analyss" 2012/9, Jofre-Monseny, J.; Marín-López, R.; Vladecans-Marsal, E.: "What underles localzaton and urbanzaton economes? Evdence from the locaton of new frms" 2012/10, Mantovan, A.; Vandekerckhove, J.: "The strategc nterplay between bundlng and mergng n complementary markets" 2012/11, Garca-López, M.A.: "Urban spatal structure, suburbanzaton and transportaton n Barcelona" 2012/12, Revell, F.: "Busness taxaton and economc performance n herarchcal government structures" 2012/13, Arqué-Castells, P.; Mohnen, P.: "Sunk costs, extensve R&D subsdes and permanent nducement effects" 2012/14, Boffa, F.; Polatto, A.; Ponzetto, G.: "Centralzaton and accountablty: theory and evdence from the Clean Ar Act" 2012/15, Cheshre, P.C.; Hlber, C.A.L.; Kaplans, I.: "Land use regulaton and productvty land matters: evdence from a UK supermarket chan" 2012/16, Cho, A.; Calero, J.: "The contrbuton of the dsabled to the attanment of the Europe 2020 strategy headlne targets" 2012/17, Slva, J.I.; Vázquez-Grenno, J.: "The ns and outs of unemployment n a two-ter labor market" 2012/18, González-Val, R.; Lanaspa, L.; Sanz, F.: "New evdence on Gbrat s law for ctes" 2012/19, Vázquez-Grenno, J.: "Job search methods n tmes of crss: natve and mmgrant strateges n Span" 2012/20, Lessmann, C.: "Regonal nequalty and decentralzaton an emprcal analyss" 2012/21, Nuevo-Chquero, A.: "Trends n shotgun marrages: the pll, the wll or the cost?" 2012/22, Pl Damm, A.: "Neghborhood qualty and labor market outcomes: evdence from quas-random neghborhood assgnment of mmgrants" 2012/23, Ploeckl, F.: "Space, settlements, towns: the nfluence of geography and market access on settlement dstrbuton and urbanzaton" 2012/24, Algan, Y.; Hémet, C.; Latn, D.: "Dversty and local publc goods: a natural experment wth exogenous resdental allocaton" 2012/25, Martnez, D.; Sjögren, T.: "Vertcal externaltes wth lump-sum taxes: how much dfference does unemployment make?" 2012/26, Cubel, M.; Sanchez-Pages, S.: "The effect of wthn-group nequalty n a conflct aganst a untary threat" 2012/27, Andn, M.; De Blaso, G.; Duranton, G.; Strange, W.C.: "Marshallan labor market poolng: evdence from Italy" 2012/28, Solé-Ollé, A.; Vladecans-Marsal, E.: "Do poltcal partes matter for local land use polces?" 2012/29, Buonanno, P.; Durante, R.; Prarolo, G.; Vann, P.: "Poor nsttutons, rch mnes: resource curse and the orgns of the Sclan mafa" 2012/30, Anghel, B.; Cabrales, A.; Carro, J.M.: "Evaluatng a blngual educaton program n Span: the mpact beyond foregn language learnng" 2012/31, Curto-Grau, M.; Solé-Ollé, A.; Sorrbas-Navarro, P.: "Partsan targetng of nter-governmental transfers & state nterference n local electons: evdence from Span"

34 Documents de Treball de l IEB 2012/32, Kappeler, A.; Solé-Ollé, A.; Stephan, A.; Vällä, T.: "Does fscal decentralzaton foster regonal nvestment n productve nfrastructure?" 2012/33, Rzzo, L.; Zanard, A.: "Sngle vs double ballot and party coaltons: the mpact on fscal polcy. Evdence from Italy" 2012/34, Ramachandran, R.: "Language use n educaton and prmary schoolng attanment: evdence from a natural experment n Ethopa" 2012/35, Rothsten, J.: "Teacher qualty polcy when supply matters" 2012/36, Ahlfeldt, G.M.: "The hdden dmensons of urbanty" 2012/37, Mora, T.; Gl, J.; Scras-Manar, A.: "The nfluence of BMI, obesty and overweght on medcal costs: a panel data approach" 2012/38, Pelegrín, A.; García-Quevedo, J.: "Whch frms are nvolved n foregn vertcal ntegraton?" 2012/39, Agasst, T.; Longobard, S.: "Inequalty n educaton: can Italan dsadvantaged students close the gap? A focus on reslence n the Italan school system" /1, Sánchez-Vdal, M.; González-Val, R.; Vladecans-Marsal, E.: "Sequental cty growth n the US: does age matter?" 2013/2, Hortas Rco, M.: "Sprawl, blght and the role of urban contanment polces. Evdence from US ctes" 2013/3, Lampón, J.F.; Cabanelas-Lorenzo, P-; Lago-Peñas, S.: "Why frms relocate ther producton overseas? The answer les nsde: corporate, logstc and technologcal determnants" 2013/4, Montolo, D.; Planells, S.: "Does toursm boost crmnal actvty? Evdence from a top tourstc country" 2013/5, Garca-López, M.A.; Holl, A.; Vladecans-Marsal, E.: "Suburbanzaton and hghways: when the Romans, the Bourbons and the frst cars stll shape Spansh ctes" 2013/6, Bosch, N.; Espasa, M.; Montolo, D.: "Should large Spansh muncpaltes be fnancally compensated? Costs and benefts of beng a captal/central muncpalty" 2013/7, Escardíbul, J.O.; Mora, T.: "Teacher gender and student performance n mathematcs. Evdence from Catalona" 2013/8, Arqué-Castells, P.; Vladecans-Marsal, E.: "Bankng towards development: evdence from the Spansh bankng expanson plan" 2013/9, Asenso, J.; Gómez-Lobo, A.; Matas, A.: "How effectve are polces to reduce gasolne consumpton? Evaluatng a quas-natural experment n Span" 2013/10, Jofre-Monseny, J.: "The effects of unemployment benefts on mgraton n laggng regons" 2013/11, Segarra, A.; García-Quevedo, J.; Teruel, M.: "Fnancal constrants and the falure of nnovaton projects" 2013/12, Jerrm, J.; Cho, A.: "The mathematcs sklls of school chldren: How does England compare to the hgh performng East Asan jursdctons?" 2013/13, González-Val, R.; Trado-Fabregat, D.A.; Vladecans-Marsal, E.: "Market potental and cty growth: Span " 2013/14, Lundqvst, H.: "Is t worth t? On the returns to holdng poltcal offce" 2013/15, Ahlfeldt, G.M.; Maenng, W.: "Homevoters vs. leasevoters: a spatal analyss of arport effects" 2013/16, Lampón, J.F.; Lago-Peñas, S.: "Factors behnd nternatonal relocaton and changes n producton geography n the European automoble components ndustry" 2013/17, Guío, J.M.; Cho, A.: "Evoluton of the school falure rsk durng the 2000 decade n Span: analyss of Psa results wth a two-level logstc mode" 2013/18, Dahlby, B.; Rodden, J.: "A poltcal economy model of the vertcal fscal gap and vertcal fscal mbalances n a federaton" 2013/19, Acaca, F.; Cubel, M.: "Strategc votng and happness" 2013/20, Hellersten, J.K.; Kutzbach, M.J.; Neumark, D.: "Do labor market networks have an mportant spatal dmenson?" 2013/21, Pellegrno, G.; Savona, M.: "Is money all? Fnancng versus knowledge and demand constrants to nnovaton" 2013/22, Ln, J.: "Regonal reslence" 2013/23, Costa-Camp, M.T.; Duch-Brown, N.; García-Quevedo, J.: "R&D drvers and obstacles to nnovaton n the energy ndustry" 2013/24, Husman, R.; Stradnc, V.; Westgaard, S.: "Renewable energy and electrcty prces: ndrect emprcal evdence from hydro power" 2013/25, Dargaud, E.; Mantovan, A.; Reggan, C.: "The fght aganst cartels: a transatlantc perspectve" 2013/26, Lambertn, L.; Mantovan, A.: "Feedback equlbra n a dynamc renewable resource olgopoly: preempton, voracty and exhauston"

35 Documents de Treball de l IEB 2013/27, Feld, L.P.; Kalb, A.; Moessnger, M.D.; Osterloh, S.: "Soveregn bond market reactons to fscal rules and no-balout clauses the Swss experence" 2013/28, Hlber, C.A.L.; Vermeulen, W.: "The mpact of supply constrants on house prces n England" 2013/29, Revell, F.: "Tax lmts and local democracy" 2013/30, Wang, R.; Wang, W.: "Dress-up contest: a dark sde of fscal decentralzaton" 2013/31, Dargaud, E.; Mantovan, A.; Reggan, C.: "The fght aganst cartels: a transatlantc perspectve" 2013/32, Saarmaa, T.; Tukanen, J.: "Local representaton and strategc votng: evdence from electoral boundary reforms" 2013/33, Agasst, T.; Murtnu, S.: "Are we wastng publc money? No! The effects of grants on Italan unversty students performances" 2013/34, Flacher, D.; Harar-Kermadec, H.; Mouln, L.: "Fnancng hgher educaton: a contrbutory scheme" 2013/35, Carozz, F.; Repetto, L.: "Sendng the pork home: brth town bas n transfers to Italan muncpaltes" 2013/36, Coad, A.; Franksh, J.S.; Roberts, R.G.; Storey, D.J.: "New venture survval and growth: Does the fog lft?" 2013/37, Gulett, M.; Gross, L.; Waterson, M.: "Revenues from storage n a compettve electrcty market: Emprcal evdence from Great Brtan" /1, Montolo, D.; Planells-Struse, S.: "When polce patrols matter. The effect of polce proxmty on ctzens crme rsk percepton" 2014/2, Garca-López, M.A.; Solé-Ollé, A.; Vladecans-Marsal, E.: "Do land use polces follow road constructon?" 2014/3, Polatto, A.; Rablen, M.D.: "Prospect theory and tax evason: a reconsderaton of the Ytzhak puzzle" 2014/4, Cuberes, D.; González-Val, R.: "The effect of the Spansh Reconquest on Iberan Ctes" 2014/5, Durán-Cabré, J.M.; Esteller-Moré, E.: "Tax professonals' vew of the Spansh tax system: effcency, equty and tax plannng" 2014/6, Cubel, M.; Sanchez-Pages, S.: "Dfference-form group contests" 2014/7, Del Rey, E.; Raconero, M.: "Choosng the type of ncome-contngent loan: rsk-sharng versus rskpoolng" 2014/8, Torregrosa Hetland, S.: "A fscal revoluton? Progressvty n the Spansh tax system, " 2014/9, Polatto, A.: "Itemsed deductons: a devce to reduce tax evason" 2014/10, Costa, M.T.; García-Quevedo, J.; Segarra, A.: "Energy effcency determnants: an emprcal analyss of Spansh nnovatve frms" 2014/11, García-Quevedo, J.; Pellegrno, G.; Savona, M.: "Revvng demand-pull perspectves: the effect of demand uncertanty and stagnancy on R&D strategy" 2014/12, Calero, J.; Escardíbul, J.O.: "Barrers to non-formal professonal tranng n Span n perods of economc growth and crss. An analyss wth specal attenton to the effect of the prevous human captal of workers" 2014/13, Cubel, M.; Sanchez-Pages, S.: "Gender dfferences and stereotypes n the beauty" 2014/14, Polatto, A.; Schuett, F.: "Meda competton and electoral poltcs" 2014/15, Montolo, D.; Trllas, F.; Trujllo-Baute, E.: "Regulatory envronment and frm performance n EU telecommuncatons servces" 2014/16, Lopez-Rodrguez, J.; Martnez, D.: "Beyond the R&D effects on nnovaton: the contrbuton of non- R&D actvtes to TFP growth n the EU" 2014/17, González-Val, R.: "Cross-sectonal growth n US ctes from 1990 to 2000" 2014/18, Vona, F.; Ncoll, F.: "Energy market lberalzaton and renewable energy polces n OECD countres" 2014/19, Curto-Grau, M.: "Voters responsveness to publc employment polces" 2014/20, Duro, J.A.; Texdó-Fgueras, J.; Padlla, E.: "The causal factors of nternatonal nequalty n co2 emssons per capta: a regresson-based nequalty decomposton analyss" 2014/21, Fleten, S.E.; Husman, R.; Klc, M.; Pennngs, E.; Westgaard, S.: "Electrcty futures prces: tme varyng senstvty to fundamentals" 2014/22, Afcha, S.; García-Quevedo, J,: "The mpact of R&D subsdes on R&D employment composton" 2014/23, Mr-Artgues, P.; del Río, P.: "Combnng tarffs, nvestment subsdes and soft loans n a renewable electrcty deployment polcy" 2014/24, Romero-Jordán, D.; del Río, P.; Peñasco, C.: "Household electrcty demand n Spansh regons. Publc polcy mplcatons" 2014/25, Salnas, P.: "The effect of decentralzaton on educatonal outcomes: real autonomy matters!" 2014/26, Solé-Ollé, A.; Sorrbas-Navarro, P.: "Does corrupton erode trust n government? Evdence from a recent surge of local scandals n Span" 2014/27, Costas-Pérez, E.: "Poltcal corrupton and voter turnout: moblzaton or dsaffecton?"

36 Documents de Treball de l IEB 2014/28, Cubel, M.; Nuevo-Chquero, A.; Sanchez-Pages, S.; Vdal-Fernandez, M.: "Do personalty trats affect productvty? Evdence from the LAB" 2014/29, Teresa Costa, M.T.; Trujllo-Baute, E.: "Retal prce effects of feed-n tarff regulaton" 2014/30, Klc, M.; Trujllo-Baute, E.: "The stablzng effect of hydro reservor levels on ntraday power prces under wnd forecast errors" 2014/31, Costa-Camp, M.T.; Duch-Brown, N.: "The dffuson of patented ol and gas technology wth envronmental uses: a forward patent ctaton analyss" 2014/32, Ramos, R.; Sanromá, E.; Smón, H.: "Publc-prvate sector wage dfferentals by type of contract: evdence from Span" 2014/33, Backus, P.; Esteller-Moré, A.: "Is ncome redstrbuton a form of nsurance, a publc good or both?" 2014/34, Husman, R.; Trujllo-Baute, E.: "Costs of power supply flexblty: the ndrect mpact of a Spansh polcy change" 2014/35, Jerrm, J.; Cho, A.; Smancas Rodríguez, R.: "Two-sample two-stage least squares (TSTSLS) estmates of earnngs moblty: how consstent are they?" 2014/36, Mantovan, A.; Tarola, O.; Vergar, C.: "Hedonc qualty, socal norms, and envronmental campagns" 2014/37, Ferrares, M.; Galmarn, U.; Rzzo, L.: "Local nfrastructures and externaltes: Does the sze matter?" 2014/38, Ferrares, M.; Rzzo, L.; Zanard, A.: "Polcy outcomes of sngle and double-ballot electons" /1, Foremny, D.; Freer, R.; Moessnger, M-D.; Yeter, M.: "Overlappng poltcal budget cycles n the legslatve and the executve" 2015/2, Colombo, L.; Galmarn, U.: "Optmalty and dstortonary lobbyng: regulatng tobacco consumpton" 2015/3, Pellegrno, G.: "Barrers to nnovaton: Can frm age help lower them?" 2015/4, Hémet, C.: "Dversty and employment prospects: neghbors matter!" 2015/5, Cubel, M.; Sanchez-Pages, S.: "An axomatzaton of dfference-form contest success functons" 2015/6, Cho, A.; Jerrm, J.: "The use (and msuse) of Psa n gudng polcy reform: the case of Span" 2015/7, Durán-Cabré, J.M.; Esteller-Moré, A.; Salvador, L.: "Emprcal evdence on tax cooperaton between sub-central admnstratons" 2015/8, Batalla-Bejerano, J.; Trujllo-Baute, E.: "Analysng the senstvty of electrcty system operatonal costs to devatons n supply and demand" 2015/9, Salvador, L.: "Does tax enforcement counteract the negatve effects of terrorsm? A case study of the Basque Country" 2015/10, Montolo, D.; Planells-Struse, S.: "How tme shapes crme: the temporal mpacts of football matches on crme" 2015/11, Polatto, A.: "Onlne bookng and nformaton: competton and welfare consequences of revew aggregators" 2015/12, Boffa, F.; Pngal, V.; Sala, F.: "Strategc nvestment n merchant transmsson: the mpact of capacty utlzaton rules" 2015/13, Slemrod, J.: "Tax admnstraton and tax systems" 2015/14, Arqué-Castells, P.; Cartaxo, R.M.; García-Quevedo, J.; Mra Godnho, M.: "How nventor royalty shares affect patentng and ncome n Portugal and Span" 2015/15, Montolo, D.; Planells-Struse, S.: "Measurng the negatve externaltes of a prvate lesure actvty: hoolgans and pckpockets around the stadum" 2015/16, Batalla-Bejerano, J.; Costa-Camp, M.T.; Trujllo-Baute, E.: "Unexpected consequences of lberalsaton: meterng, losses, load profles and cost settlement n Span s electrcty system" 2015/17, Batalla-Bejerano, J.; Trujllo-Baute, E.: "Impacts of ntermttent renewable generaton on electrcty system costs" 2015/18, Costa-Camp, M.T.; Panagua, J.; Trujllo-Baute, E.: "Are energy market ntegratons a green lght for FDI?" 2015/19, Jofre-Monseny, J.; Sánchez-Vdal, M.; Vladecans-Marsal, E.: "Bg plant closures and agglomeraton economes" 2015/20, Garca-López, M.A.; Hémet, C.; Vladecans-Marsal, E.: "How does transportaton shape ntrametropoltan growth? An answer from the regonal express ral"

37 Fscal Federalsm