WHY ARE CIVIL LIBERTIES MORE IMPORTANT THAN EXECUTIVE CONSTRAINTS IN ECONOMIC DEVELOPMENT? A PROPERTY RIGHTS APPROACH 1

Transcription

1 Society nd Economy (), pp. DOI:./SocEc... WHY ARE IVIL LIBERTIES MORE IMPORTANT THAN EXETIVE ONSTRAINTS IN EONOMI DEVELOPMENT? A PROPERTY RIGHTS APPROAH PÁL ZEGLÉDI Associte professor, Fculty of Economics nd Business Administrtion, niversity of Debrecen E-mil: pl.czegledi@econ.unideb.hu This pper develops simple model tht helps understnd n importnt fct concerning cross-country pttern of growth nd institutions shown by BenYishy nd Betncourt (). They show tht civil freedoms, especilly one of their components clled Autonomy nd Individul Rights, re more importnt determinnts of economic development thn constrints on executives, widely used mesure in the literture on institutions nd growth. The pper provides n interprettion of this fct through the lense of n rgument tht puts emphsis on three insights. The first is tht civil freedoms cn be seen s property rights brodly understood. The second is tht with higher scope of property rights enforced, the government must be ble to commit to lower level of exproprition of income. Third, institutions of freedom re sticky: they must be in line with the culture of the country so tht they cn be enforced with resonble cost. By ddressing this specific question of constrints on executives versus civil freedoms the pper joins the literture which emphsizes the importnce of culture in economic development. Keywords: civil liberties, property rights, development JEL codes: P, P, O The pper ws supported by the János Bolyi Reserch Scholrship of the Hungrin Acdemy of Sciences. The uthor wishes to thnk two nonymous referees for their genuinely helpful comments on numerous points of the pper nd for their suggestions tht substntilly improved the pper. Any remining errors re those of the uthor. -/$. Akdémii Kidó, Budpest

2 PÁL ZEGLÉDI. INTRODTION: ONSTRAINT ON EXETIVES OR IVIL LIBERTIES? Economists re long wy from greement on the role of democrcy in economic development. There re some widely ccepted clims, however. One of them, t lest in wht I cll the minstrem of institutionl reserch, is tht the checks nd blnces which the democrtic institutions re ble to crete re the fundmentl determinnts of economic development. Tht is becuse these checks nd blnces constrin politicl power, nd these constrints re ble to contin the risk of exproprition, nd, so the rgument goes, low risk of exproprition is necessry for investment in physicl nd humn cpitl, nd most of ll, for innovtion. Indeed, Acemoglu nd Johnson () found tht property rights institutions, mesured s the constrint on executive, re more importnt determinnts of economic development thn contrcting institutions which in the long run do not even relly mtter. Since then, constrints on executives s documented in the Polity IV dtbse hve become the most importnt mesure of the security of property rights, nd of institutions in generl, especilly in the literture serching for institutionl cuses of economic development. This is why the empiricl result derived by BenYishy nd Betncourt () is fundmentlly importnt for n understnding of the question s to how politicl institutions, mrket institutions, nd economic development relte to ech other. They show in cross-sectionl nlysis of countries tht civil freedoms, nd especilly the prt of it clled by Freedom House () Autonomy nd Individul Rights, perform better in terms of explntory power nd sttisticl significnce thn the widely used constrints on executives mesure does. This is true even if control vribles re included nd reverse custion is controlled by SLS method. The interprettion of this importnt fct my not be cler-cut, though. As I will show, in section, there re t lest three pproches to understnding the cusl mechnisms between democrtic institutions nd economic development. My own interprettion which is intended to be the contribution of this pper is inspired by the insight tht, s emphsized by Sobel nd oyne (: ), civil liberties cpture mny elements of embedded informl institutions, including trditions, religion, nd customs. Bsed on this insight this pper is imed t building n rgument tht is helpful in understnding BenYishy nd Betncourt s () result concerning the im- See the next section for more detiled elbortion of wht is ment by this. Kozenkow () provides brief but very good overview of New Institutionl Economics including the min conclusions of the empiricl reserch cited here nd the property rights pproch pplied in the pper. Society nd Economy ()

3 WHY ARE IVIL LIBERTIES MORE IMPORTANT THAN EXETIVE ONSTRAINTS? portnce of civil freedoms in economic development s compred to tht of executive constrints. The rgument is tht this fct cn be trced bck to three chrcteristics of institutions of individul freedom. First, higher level of freedom mens tht the government must be ble to commit itself to lower level of exproprition. Second, institutions of freedom (civil s well s economic) re sticky, tht is, they hve to be in line with culture. Third, civil freedoms cn be seen s the enforcement of reltively brod scope of property rights. On n bstrct level, economic freedoms nd civil freedoms re not relly different. The pper shows tht when these three ssumptions re put together in simple model they re ble to explin why civil freedom is better predictor of development thn executive constrints re. This rgument will be presented in the following steps. The next section briefly reviews the literture on the three different explntions concerning the reltionship between economic development nd civil liberties or democrcy, nd describes the brod pproch pplied in the pper. Section, though not intended to be literture review, mkes n extensive use of the literture on the similr nture of economic nd civil freedoms with the help of which I provide n interprettion of civil liberties in the frmework of property rights. sing this frmework, in section I develop very simple model of how constrints on executives, culture, nd technology ffect the provision of civil liberties. Section uses this model to understnd the different effect of executive constrints nd civil liberties on development. Section drws some conclusions.. POLITIAL RIGHTS, IVIL FREEDOM AND EONOMI DEVELOPMENT A BRIEF LITERATRE REVIEW Seeing economic development nd democrcy s two vribles which re correlted in time s well s cross countries, there re logiclly three possibilities to model their reltionship. The first is to sy tht it is democrcy tht cuses development. The second is to sy tht it is just the other wy round nd development cuses democrcy. The third possibility is to sy tht there re third fctors tht The work concentrting on this question sees democrcy s mix of different institutions providing politicl rights nd civil liberties. These rights nd liberties re theoreticlly different, but they re very closely ssocited empiriclly. This my be the reson why most of the studies cited do not relly emphsize the difference between politicl rights, civil liberties, nd the institutions tht mke them possible. In the studies reviewed below it is often democrcy tht is used s vrible, which includes civil liberties. Tht is the reson I will not limit this overview only to those works tht discuss civil liberties in nrrow sense. Society nd Economy ()

4 PÁL ZEGLÉDI mke both vribles go in step with ech other. All the three views re represented in the economics literture (see Tble ). The first explntion the one tht cn be viewed s the institutionl minstrem clims tht s democrtic institutions re constrints on the politicl elite they reduce the risk of exproprition (North Weingst ; Olson ; Acemoglu et l. ). This view hs been given much support in the pst decde by showing tht the most importnt institutions in economic development re those tht constrin politicl power (Acemoglu et l. ; Acemoglu Johnson Tble Three pproches to the reltionship between economic development nd democrcy Questions Interprettion of cuslity Role of civil liberties Role of executive constrints Role of culture Leding uthors nd works Institutionl minstrem Democrtic institutions cuse development ivil liberties constrin executives onstrints on executives lower the risk of exproprition ulture hs no role Acemoglu Robinson () North Weingst () Lypset hypothesis (moderniztion thesis) Development leds to democrcy ivil liberties re luxury goods onstrints re demnded s result of better eduction nd higher income ulture my impede the process in which development brings bout democrcy Lipset () Brro () Pldm Gundlch () Gleser et l. () Note: This tble summrises the literture review of section. ulture nd technology view Development nd democrcy go step by step s result of third fctors ivil liberties re importnt for constitutionl exchnge Executive constrints cn be seen s result of constitutionl exchnge fuelled by new elite interest n ideology Ideology is needed for liberl economic nd politicl reforms ongleton () Mcloskey () Although Mcloskey () does not directly ddress the democrcy nd development issue, she emphsizes tht modern economic growth ws cused by new ideology giving dignity to bourgeois life. The ltter, I think, is first step towrds developing democrtic institutions. Society nd Economy ()

5 WHY ARE IVIL LIBERTIES MORE IMPORTANT THAN EXETIVE ONSTRAINTS? ); tht it is democrcy tht comes first, not development or humn cpitl ccumultion (Acemoglu et l., b); nd tht in non-democrtic societies the elite cn block the introduction of new technologies tht would led to economic development becuse they fer the politicl uncertinties tht it might bring (Acemoglu Robinson ). The view tht emphsizes the cuslity running from development to democrcy is usully clled Lipset s hypothesis (Lipset ; Brro ; Pldm ). According to this view (Pldm Gundlch ) economic development nd humn cpitl ccumultion closely ssocited with it will cuse n incresing demnd for democrtic institutions. Pldm s () nlysis gives support to this view, lso referred to s the grnd trnsition, nd to some of the ccompnying culturl pproches, lthough he does not find it relevnt tht democrcy cuses growth. This is brodly the sme conclusion reched in severl other ppers (e.g. Pldm Gundlch, ). Put differently, it is bsolutely not cler tht there is ny cuslity running from politicl or civil freedom towrds development. Pldm nd Gundlch () use Grnger regressions nd different instrumentl vribles to compre short- nd medium-run effects with long-run effects nd find tht in the long run the cuslity tht runs from income to democrcy is lmost overwhelming, while the reverse effect cn only be detected in the short run ( yers). They lso use vribles of the culturl kind nd conclude tht some of them such s the shre of Protestnts nd Muslims within the popultion mtter. sing historicl dt on schooling Murtin nd Wczirg () lso reffirm the Lipset hypothesis nd find no evidence of cuslity running from democrcy to development. This grnd trnsition thesis is lso echoed by those theories nd empiricl studies of economic development (Gleser et l. ; Esterly Levine ) tht show tht long-run economic development ws determined to lrge extent by humn cpitl nd probbly culture nd not by the strtegies of coloniztion (or the initil nturl conditions by which it ws ffected) nd by the different politicl institutions they resulted in, s is suggested by the first view. It is possible to formulte third view emphsizing different third fctors when explining the co-movement of democrtic institutions nd development. Some reserchers hve done so. As we hve just seen, Pldm () cnnot reject entirely the ide tht culture cn ccount for the development of democrcy t lest prtilly, nd on the other hnd some economists (e.g. Tbellini ) lso emphsize the role of culture in economic development. Indeed, Sobel nd oyne () show, s we sw in section, tht different institutions re closely relted nd civil liberties most probbly reflect some deep-seted culturl element. In his book on constitutionl development in the West, ongleton () explins tht liberl ideology ws one of the driving forces of democrtiztion, to- Society nd Economy ()

6 PÁL ZEGLÉDI gether with the new economic interests nd politicl pressures creted by economic development which were mde possible by the liberl economic reforms. Beside detiled historicl cse studies of the countries in question he pplies Grnger cuslity tests of the time series dt of these countries (ongleton : ) nd his conclusion of the bootstrpping nture of politicl nd economic processes re supported by the mutul determintion of development nd democrcy. The three ssumptions mde in the introduction re echoed in these pproches lthough not ll of them re embrced. The first one is included in ll three pproches: no one denies tht for innovtion to hppen low risk of exproprition is needed. The role of culture or ideology is considered in the third pproch s condition of constitutionl exchnge the result of which is higher level of freedom. The third ssumption is, however, not relly emphsized by ny of these pproches. This is why I consider the ppliction of the property rights view to be the contribution mde by this pper. The next section elbortes on this third ssumption.. FREEDOM AS THE SOPE AND ENFOREMENT OF PROPERTY RIGHTS My interprettion of the fct described in the introduction rests on three ssumptions. Nmely I ssume tht () higher level of freedom mens tht the government must be ble to commit itself to lower level of exproprition, but () the institutions of freedom (civil s well s economic) hve to be in line with culture, nd () civil freedoms cn be seen s the enforcement of reltively brod scope of property rights. In this section I will try to show how these ssumptions cn be integrted by pplying the pproch of the so clled property rights school. Assumption () cn be derived by recognizing tht lbelling some rights property nd others civil is n rbitrry wy of thinking in this pproch. Hence n ppliction of this wy of thinking mkes it possible to see economic freedoms nd civil freedoms in unified frmework. In this clssicl liberl interprettion there is no fundmentl difference between economic freedom nd civil freedom. In ccordnce with my ssumption () freedom is defined s n bsence of coercion (Hyek : ). More precisely, some level of coercion is lwys present, becuse some coercion is needed to enforce the rules ensuring freedom. Freedom is t its mximum, when coercion This section develops further the rgument I present in zeglédi () where I propose tht the difference between economic nd civil freedom re vgue nd rbitrry nd this hs implictions when we look t these freedoms s determinnts of the welth of countries. Society nd Economy ()

7 WHY ARE IVIL LIBERTIES MORE IMPORTANT THAN EXETIVE ONSTRAINTS? including privte coercion is t its minimum. While, for exmple F. A. Hyek is generlly seen s chmpion of economic freedom, in one of his mgnum opuses he cme to the conclusion tht [t]he importnce of freedom [ ] does not depend on the elevted chrcter of the ctivities it mkes possible. Freedom of ction, even in humble things, is s importnt s freedom of thought. It hs become common prctice to disprge freedom of ction by clling it economic liberty. But the concept of freedom of ction is much wider thn tht of economic liberty, which it includes; nd wht is more importnt, it is very questionble whether there re ny ctions which cn be clled merely economic nd whether ny restrictions on liberty cn be confined to wht re clled merely economic spects. Economic considertions re merely those by which we reconcile nd djust our different purposes, none of which, in the lst resort, re economic (excepting those of the miser or the mn for whom mking money hs become n end itself) (Hyek : ). The sme view is formulted by Knight (: ) when he sys tht [t]he ide of distinction between economic wnts nd other wnts must be bndoned. There is no definble objective, whether subsistence, grtifiction of fundmentl impulses or plesure, which will serve to seprte ny of our ctivities from the body of conduct s whole. Tht is, there is no wy to decide whether certin kind of humn ction is economic or not. However, if this decision is impossible, it is lso impossible to decide whether certin kind of freedom is economic or not, since this lbel should describe the ction tht it mkes possible. Insted of tlking bout economic nd civil freedom, it mkes more sense to tlk bout two dimensions of freedom, the enforcement of rights nd the scope of rights. The difference between these two dimensions of property rights cn be derived from the property rights pproch (Alchin nd Demsetz ). Alchin nd Demsetz (: ) give very concise description of wht is ment by property rights in this trdition when they sy tht [w]ht re owned re socilly recognized rights of ction, tht is, [w]ht is owned re rights to use resources, including one s body nd mind, nd these rights re lwys circumscribed, often by the prohibition of certin ctions (emphsis in originl). It is cler tht from this perspective civil rights re property rights. As Brzel (:, footnote ) mkes it explicit, [h]umn rights re simply prt of people s property rights. Humn rights my be difficult to protect or to exchnge, but so re rights to mny other ssets. BenYishy nd Betncourt (: ) mention s well tht the protection of humn rights [ ] follows the sme logic of property rights defined s the right to consume services of, the right to generte income from nd the rights to liente n sset. Society nd Economy ()

8 PÁL ZEGLÉDI From this point of view it would require us to define those rights tht re economic to get mesure of economic freedom, nd, similrly one should be ble to define civil use of n sset to get mesure of civil liberties. It is not cler whether there is generl criterion to mke this ctegoriztion. Gwrtney nd Lwson (: ) rgue tht economic freedom reflects different sphere of humn interction, lthough the foundtion of politicl nd civil liberty is identicl to tht of economic freedom. But they do not provide the reder with generl concept to estblish difference. Such criterion might come from Vnberg (: ). He proposes tht the definition of property rights given bove confuses two different dimensions of property rights which he clls the issue of ssigning rights which nswers the question who owns wht? nd the issue of defining rights which nswers the question of wht does it men to own something?. By this txonomy Vnberg () is clrifying the difference between the view of lissez fire liberlism nd tht of constitutionl liberlism. Lissez fire liberls put gret emphsis on the ssigning issue, most importntly, tht of ssigning rights between the stte nd privte plyers. This is roughly the wy Gwrtney nd Lwson (: ) rgue, too, writing tht [i]nstitutions nd policies re consistent with economic freedom when they provide n infrstructure for voluntry exchnge, nd protect individuls nd their property from ggressors seeking to use violence, coercion, nd frud to seize things tht do not belong to them. lerly this definition is bsed on the question who owns wht? nd not on the mening of ownership, since one is supposed to know wht it mens tht something belongs to someone. This llows us to interpret economic freedom s mesure of the extent to which the government in country is rellocting property rights from its originl owner to someone else. As shown by the construction of the Economic Freedom of the World index (Gwrtney et l. ) or the Index of Economic Freedom (Miller et l. ) this could hppen through direct redistribution, by legl system nd regultion tht discrimintes or tht works with high trnsction costs, by infltion, by restrictions on interntionl trde, nd on business inside the borders. The two Vnbergin dimensions re, however, implicitly included in the clssicl rticle by Alchin ([]: ) which climed tht (emphsis dded) The Frser Institute s Economic Freedom of the World Index hs five res: () size of government, () legl structure nd the security of property rights, () ccess to sound money, () freedom to trde interntionlly, nd () regultion of credit, lbour nd business. See Gwrtney et l. () for more detiled exposition of these res. The Index of Economic Freedom of the Heritge Foundtion ggregtes ten freedoms: property rights, freedom from corruption, fiscl freedom, government spending, business freedom, lbour freedom, monetry freedom, trde freedom, investment freedom, finncil freedom. See Miller et l. () for more detiled exposition of these res. Society nd Economy ()

9 WHY ARE IVIL LIBERTIES MORE IMPORTANT THAN EXETIVE ONSTRAINTS? [b]y system of property rights I men method of ssigning to prticulr individuls the uthority to select for specific goods, ny use from nonprohibited clss of uses. the concepts of uthority nd non-prohibited rely on some concept of enforcement or inducement to respect the ssignment nd scope of prohibited choice. Vnberg s () two dimensionl theoreticl structure my be useful to differentite civil liberties from economic freedom. I suggest pplying simple structure ccording to which economic freedom refers to the lloction of rights between the stte nd the privte plyers while civil liberties re minly bout the scope of rights tht re socilly recognized. For exmple, by putting hevy restriction on business, the government is re-llocting income from consumers to some privileged group. But if it is the freedom of expression tht is restricted then no privte plyer cn use their own sset to express ides which government officils do not like. This is in line with the generl expression of humn rights ccording to which humn rights s we know them tody re minly politicl norms deling with how people should be treted by their governments nd institutions (Nickel ). I do not clim tht every kind of humn right fits the property rights pproch. It is only individul nd negtive rights tht do. This is why I confine the discussion to civil liberties nd not humn rights. Freedom of expression seems to be good exmple to explin the ide tht the enforcement of freedom (ssignment of rights) is seprte decision from the definition of rights. As it is expressed in the nited Ntions niversl Declrtion of Humn Rights (), [e]veryone hs the right to freedom of opinion nd expression; this right includes freedom to hold opinions without interference nd to seek, receive nd imprt informtion nd ides through ny medi nd regrdless of frontiers. Regrding tht some persusion is included in every sector of the economy, some persusion is needed for every mrket trnsction (Mcloskey Klmer ). The mrket cn be seen s socil spce with perpetul converstion between the plyers (Storr ). As Storr (: ) concludes erecting trde brriers lso locks off potentil converstions nd ccess to potentil dilogicl prtners. This mens tht you cnnot weken the enforcement of economic freedom lone, becuse the mrket is socil spce where both commercil nd non-commercil ctivities tke plce. By preventing some trding In ddition, civil liberties must be seen s different from politicl rights. The fct tht these two sets of rights re conceptully different is expressed, for exmple, in the views held by some gret clssicl liberl thinkers who supported civic freedom fully but were not democrts t ll (Director : -). Note lso, tht Sen (:.) uses similr exmple to illustrte the point of freedom-centred normtive rgument when he writes tht [t]he freedom to exchnge words, or goods, or gifts does not need defensive justifiction in terms of their fvourble but distnt effects; they re prt of the wy humn beings in society live nd interct with ech other (unless stopped by regultion or fit). Society nd Economy ()

10 PÁL ZEGLÉDI you will prevent some tlking. Once it is decided wht scope of rights re non-prohibited, it becomes very hrd to enforce vrious groups of property rights vriously. The socilly recognized nture of property rights s required by the definition bove llows us to incorporte the role of those fctors tht re usully described vguely s informl, or generlly, s culturl. My ssumption () rests on the ide tht the greter the extent to which right is ccepted culturlly, the less costly it is to enforce tht right. The notion tht constitutionl rules re formed by informl rules hs been formulted by some of the economists trying to understnd constitutions s they re brodly understood. For exmple, Hyek ( []: ) ws gretly worried by the decline of the virtues of independence nd self-relince, individul inititive nd locl responsibility, the successful relince on voluntry ctivity, non-interference with one s neighbour nd tolernce of the different nd queer, respect for custom nd trdition, nd helthy suspicion of power nd uthority. He sw these vlues s the foundtions of the freedom of the West. Indeed, in Hyek (: ) he explins tht the recognition of individul spheres is necessry so tht individul freedom cn be enforced, nd this hppens through the cceptnce of generl rules. This mens, s he (Hyek : ) concludes, tht the procedurl sfegurds of liberty presuppose for their effectiveness the cceptnce of the rule of lw, politicl creed which people will defend even when they do not fully understnd its significnce. Buchnn ( []: ) lso points out the reltionship between enforcement costs nd the willingness to follow the rules, when sying tht there is reltionship between potentil voluntry complince independent of enforcement-punishment nd the resource investment tht will be required to ttin specified behviourl limits. The rgument of this section breks down to wht cn be summrized s Property Rights Possibility Frontier s illustrted in Figure : there is trde-off between the scope of nonprohibited clss of uses nd the level of exproprition which is provided by the government. The rte of this trde-off is set by informl fctors. Enforcing lrger scope of the nonprohibited clss of uses will require An interesting exmple of how ethicl rules my ffect constitutionl choice cn be found in Buchnn nd Tullock ([ ]: ). They rgue tht when ethicl rules prevent vote trding, the verge citizen behind the veil of uncertinty will choose constitution tht puts more constrints on the mjority. The reson is tht s possible member of some minority she will find it more difficult to mke brgin over her individul rights with the mjority tht tries to violte them if ethicl rules preventing tht brgin re in plce. I thnk n nonymous referee for suggesting the development of Figure s well s Tble. Society nd Economy ()

11 WHY ARE IVIL LIBERTIES MORE IMPORTANT THAN EXETIVE ONSTRAINTS? low exproprition Trde-off with worse culture Trde-off with better culture high nrrow scope of nonprohibited clss of uses wide more resources which will, of course, require some kind of tx brodly understood nd will increse the level of exproprition. This increse in enforcement costs will be lower if the socil recognition of the rules to be enforced is lrger.. A SIMPLE THEORY OF THE ENFOREMENT AND THE SOPE OF PROPERTY RIGHTS In the following three subsections I will present model tht is built on the insights cquired in previous sections in the spirit of the property rights school. The rgument builds on wht ws sid in the previous two sections. In this spirit the government is modelled s mking decision concerning the scope of nonprohibited clss of uses to be enforced nd the level of enforcement. It will be demonstrted tht whether or not wide scope of rights re enforced depends on the cost of limiting rent-seeking. This cost is deemed to result from culturl fctors. Figure. Property Rights Possibility Frontier: Trde-off between exproprition (enforcement of rights) nd the scope of rights The cost of enforcement plys n importnt role in Brzel s (: ) insights on the bolition of slvery when he sys tht [b]y the end of the nineteenth century, slvery hd lrgely been bolished, perhps becuse the costs of policing the institutions exceeded its gins. Tht is, the cost of enforcing the property rights of the slve owners becme so high tht it did not py off to enforce them. This suggests tht the cost of enforcement is the most importnt determinnt of whether or not n ctivity is protected by force. However, this quote does not specify whose gin nd whose cost mtter. In the model discussed in the following three subsections it depends on whether the government is constrined or not. In somewht similr rgument exmining religious liberty Gill () lso emphsizes mong other things on the enforcement costs of religious hegemonies nd regultions of entry to the religious mrket. Society nd Economy ()

12 PÁL ZEGLÉDI.. Rent seeking nd the enforcement of rights In wht follows I will use different extensions of Murphy et l. s () model (MSV). The min modifiction of the MSV model is n introduction of government which must incur some costs to enforce rights. There re two types of ctivities: production nd rent seeking. The return on production is but only n g shre of it is exproprible by the government. Rent seeking yields return of b nd reduces the net return on production by the sme mount. In the originl MSV model there re three different kinds of equilibri depending on the size of b: one with unique good equilibrium, one with stble bd nd stble good, nd n unstble intermedite equilibrium, nd one with unique bd equilibrium. For the ske of simplicity I drop the second cse nd ssume tht g = nd continue with only two possible cses: strong enforcement of property rights ( >> b s ) nd wek enforcement of rights (b w > > ). Further I suppose tht wek enforcement requires the stte to incur cost c, while strong enforcement requires it to incur cost of c. This cost is the key prmeter of the model nd lso the one I interpret s mesure of culture. Boettke (: ) writes for exmple tht, we must lso ddress the question of enforcement of the rules of the gme. In world where the informl rules (norms) legitimte the forml rules, the costs of enforcement will be lower, nd in world where the informl rules re in conflict, the costs of enforcing the forml rules my often be prohibitive. Stringhm () explins crefully why internl morl constrints re importnt to mintin mrket economy giving this proposition corrobortion tht is bsed on theory, empiricl reserch, nd common sense. Externl constrints, he rgues, re not enough to mintin well functioning mrket which mens tht the rules of the mrket will be more costly to enforce if they re not bcked by morls. Tht is, forml constrints imed t enforcing mrket rules will hve n esier tsk if these forml constrints re in line with the morl ones. The prmeter c cn thus be seen s mesure of the extent to which socil norms legitimte the rules of privte property by ttributing low pyoff to rent seeking. This reltionship between enforcement cost nd the ttitude towrd mrket exchnge cn lso be derived from Buchnn (: ) whose min point is tht [t]he ordering over goods cnnot be seprted from the mens through which goods re expected to be secured. As result, once rent seeking s mens is condemned to lrger extent thn mrket exchnge s mens, enforcing mrket rules will be less costly. Society nd Economy ()

13 WHY ARE IVIL LIBERTIES MORE IMPORTANT THAN EXETIVE ONSTRAINTS? The government cn be of two types. It my be net revenue (rent) mximizing (unconstrined) government tht seeks to mximize net totl tx nd rent, or it my be of constrined type, in which cse it will seek to mximize net income (production minus enforcement cost) of the medin voter. ivil liberties re incorported through two key ssumptions: providing nd enforcing civil liberties () increses income nd the cost of enforcement in the sme proportion nd () reduces the expropribility of income. These ssumptions re of course simplifying but not unresonble. Wht underpins the first is the fct tht stronger enforcement of property rights will increse income by providing incentives to pply better technology nd increse productivity. A rise in productivity will increse the mrginl product of fctors of production, too, by the sme proportion if the technologicl improvement is imgined s n increse in totl fctor productivity. The second ssumption is simply description of wht stronger incentives men. A simple wy to formulte this is to ssume tht when providing nd enforcing civil liberties, the return on productive ctivities grows by fctor of d >, nd so does enforcement cost: it becomes cd in the cse of wek enforcement nd cd in the cse of strong enforcement. However, s formultion of nrrowing expropribility, I ssume tht when civil liberties re introduced, the non-exproprible level of income increses from to d. The ide behind this ssumption is tht providing civil liberties mens higher level of commitment to secure property rights by the government. The enforcement of civil liberties wide scope of rights will only be credible if the government cn promise not to exproprite such lrge shre of income s it could with nrrow scope of rights. Note tht ll these specifics of the model include the core ssumptions I mde in section. The fct tht d > reflects ssumption () by formulting the ide tht stronger enforcement, which mens lower level of exproprition, provides incentives to produce more. Assumption (), regrding the stickiness of forml institutions, is cptured by the fct tht the cost of enforcement is prmeter which cn be lrger or smller. Finlly, ssumption () is incorported by giving the government the possibility to choose between wide nd nrrow scope of rights... nconstrined government nd civil rights In this model the government hs four options to choose from: strong nd wek enforcement of rights which do, or do not, include civil liberties. The provision of For the bsic model in which civil liberties re not modelled see the Appendix. Society nd Economy ()

14 PÁL ZEGLÉDI civil liberties is equivlent with the provision of wide scope of rights. The pyoffs of these four options open to the revenue mximizing government re the following: () civil liberties with strong enforcement: () civil liberties with wek enforcement: S d d dc; () L W L d ( d) () no civil liberties with strong enforcement: () no civil liberties with wek enforcement: dc; () S c; () W c. () The key conclusion of the exmintion of these options is tht the unconstrined government will never choose the option of wek enforcement with civil liberties, but the other three options re possible depending on the cost of enforcement (c) nd productivity (). Proposition The unconstrined government will choose () civil liberties with strong enforcement if c nd d +<; () no civil liberties with strong enforcement if c S nd d +<or c S nd< d+or c nd ; Remember tht the pyoff of the government is net tx nd rent in this cse. See the Appendix for the detils of the clcultion. Society nd Economy ()

15 WHY ARE IVIL LIBERTIES MORE IMPORTANT THAN EXETIVE ONSTRAINTS? () no civil liberties with wek enforcement if c W nd, S W where ( d ) ( ),,,. Otherwise the government will not incur ny cost to enforce the rules of privte property. Proof See the Appendix. Wht mkes understnding proposition somewht difficult is tht the threshold enforcement cost itself is function of productivity. Figure, n illustrtion of the proposition, clrifies this. As is cler from Figure, below certin income the unconstrined government will not choose the option of strong enforcement with civil liberties even if the cost of enforcement is low certin level of productivity is needed, too. But the lower the cost of enforcement, the lower is the threshold productivity. If productivity is higher thn d +,it is only the cost of enforcement tht mkes the government provide civil rights enforcement. The min conclusion is tht informl institutions (the cost of enforcement) nd productivity together determine whether civil liberties will be provided even when the protection of civil liberties is ssumed to led to higher income. The key ssumption is the decresing expropribility of this higher income. ä (productivity) civil liberties with strong enforcement no civil liberties with strong enforcement no civil liberties with wek enforcement W / S c (enforcement cost) Figure. Illustrtion of proposition Note: The figure incorportes the ssumptions tht > nd d >(d + > ). The curves re the W S grphs of,,, s defined in proposition. Society nd Economy ()

16 PÁL ZEGLÉDI.. onstrined government In the cse of constrined government the pyoffs re the following: () civil liberties with strong enforcement: () civil liberties with wek enforcement: () no civil liberties with strong enforcement: S d dc; () L W d dc; () L S c; () () no civil liberties with wek enforcement: c. () W Proposition The constrined government will choose () civil liberties with strong enforcement if d nd c K SL or d nd c K ; () civil liberties with wek enforcement if d nd K cd; d SL where K K K d d,,, K. d d Proof See the Appendix. Remember tht the constrined government is ssumed to be s democrtic s possible nd it is ssumed to mximize the net income of the medin voter which, in this simple model, is equl to the verge net income. Net here mens net of enforcement costs. Society nd Economy ()

17 WHY ARE IVIL LIBERTIES MORE IMPORTANT THAN EXETIVE ONSTRAINTS? The cse of the constrined government is simpler thn tht of the unconstrined one but gin the thresholds re not independent of the income of productive ctivities, s illustrted in Figure. The constrined government will lwys provide civil rights but will not lwys enforce them strongly. Whether it does or not depends on the cost of enforcement nd productive income t the sme time. Below certin level of income (d) the government will not choose the option of strong enforcement, while bove tht the informl fctors represented by the cost of enforcement re the key. ä ä (productivity) civil liberties with strong enforcement civil liberties with strong enforcement civil liberties with wek enforcement K ä K SL c (enforcement cost) Figure. Illustrtion of proposition Note: The figure incorportes the ssumptions tht > nd d >(d + > ). The curves re the grphs of K, K SL s defined in proposition.. PREDITIONS ONERNING INOME DIFFERENES The model predicts tht income levels will depend on the scope of rights nd the enforcement of rights s shown in Tble. Tble Predictions of the model concerning per cpit income with different definition nd ssignment of rights Enforcement of non-prohibited clss of uses wek strong Scope of non-prohibited nrrow clss of uses wide d d Society nd Economy ()

18 PÁL ZEGLÉDI Becuse of the ssumptions of the model d nd > (see section.), nd it is resonble to ssume, tht d d ( > d). These three ssumptions simply men tht gross income (which includes the cost of enforcement) increses if wider scope of rights re enforced with the sme level of enforcement (level of exproprition), or if the sme scope of rights re enforced with stronger enforcement. To put it differently, when costs of enforcement re not considered, wider scope of rights or higher level of their enforcement will increse income. Propositions nd in section give some guidnce regrding the probbilities with which country will provide wide or nrrow scope of rights nd whether it will enforce it wekly or strongly. The min messges of the two propositions re included in Tbles nd. Tble shows these probbilities for the unconstrined government while Tble shows the sme for the constrined one. Tble Probbilities of providing nrrow/wide scope of rights with wek/strong enforcement for the unconstrined government Enforcement of non-prohibited clss of uses wek strong Scope of non-prohibited nrrow pns p p NS clss of uses wide p Tble Probbilities of providing nrrow/wide scope of rights with wek/strong enforcement for the constrined government Enforcement of non-prohibited clss of uses wek strong Scope of non-prohibited nrrow clss of uses wide p p Tble includes the result of proposition ccording to which n unconstrined government will not provide wide scope of rights with wek enforcement, while Tble includes the result of proposition ccording to which constrined government will lwys provide wide scope of rights. sing the probbilities in Tbles nd it is possible to derive the condition under which the provision of civil liberties hve lrger effect on income thn the constrint on executives does. onsider the expected income in country with constrined nd n unconstrined government: Society nd Economy ()

19 WHY ARE IVIL LIBERTIES MORE IMPORTANT THAN EXETIVE ONSTRAINTS? E(y government is constrined) = p d ( p ) d, () E(y government is unconstrined) = p d p ( p p ). () The difference between the two expected incomes bove gives the prediction concerning the effect of constrint on executives: D E = E(y government is constrined) E(y government is unconstrined) = = ( p p ) d ( p ) d p ( ) ( p ). NS NS NS () Similrly, it is possible to derive the expected income difference tht is ssocited with the presence of civil liberties s opposed to the lck of it by considering first the conditionl expecttions of income: E(y civil liberties re provided) = p p / W[ d d ] ( p )[ p d ( p ) d ], () / W E(y civil liberties re not provided) = p/ W[ pns ( pns p ] ( p N )[ ], () where p / W is the probbility tht the government is unconstrined if civil liberties re provided, while p / N is the probbility tht the government is unconstrined if civil liberties re not provided. The Byes rule llows us to derive tht p / W pp p p ( p ) / nd p / N, () where p is the probbility tht country hs n unconstrined government. sing (), (), nd () gives the prediction concerning the effect of civil liberties: D L = E(y civil liberties re provided) E(y civil liberties re not provided) = [ p p ( p ) p ] d ( p )( p ) d p W / W / W / ompring () nd () leds to the following proposition. NS ( ) ( p ). () Society nd Economy ()

20 PÁL ZEGLÉDI Proposition if then D L D. E Proof See the Appendix. p ( d ( d) p p ) () Proposition sys tht civil liberties provision will hve lrger effect on expected income thn executive constrints if the probbility tht country hs n unconstrined government is below threshold level. The threshold is set by the prmeter of the model. Most importntly, bering in mind tht > d, the threshold is decresing in p nd incresing in p. Tht is, the threshold is lower if the constrined government is more probble to provide wide scope of rights with strong enforcement nd it is higher if the unconstrined government is more probble to provide wide scope of rights with strong enforcement. It is lso interesting to note tht the provision of wide scope of rights (civil liberties) is better predictor of strong enforcement of rights thn constrints on executives re. To see this, consider tht the probbility tht constrined government will pply strong enforcement is p, while the probbility tht government tht provides wide scope of rights will pply strong enforcement is ( p ) p p p p ( p ). lerly, this ltter probbility is lrger thn the former until p. This conclusion is in line with BesYishy nd Betncourt s () suggestion ccording to which the provision of civil liberties is better indictor of the government s commitment to mintin the rule of lw thn is the constrint on executives.. ONLSION Individul freedom is the most importnt determinnt of economic development. It is not only libertrin conviction but fct supported by lrge number of empiricl results. However, s I rgued in this pper individul freedom includes two constitutionl decisions. Property rights need to be defined first, nd then they need to be enforced. A strong enforcement of wide scope of property rights is wht possibly leds to the highest development s opposed to wek enforcement or nrrow scope of rights. Society nd Economy ()

21 WHY ARE IVIL LIBERTIES MORE IMPORTANT THAN EXETIVE ONSTRAINTS? There re two resons why constrints on executives re n imperfect indictors of providing nd strongly enforcing wide scope of rights. One reson is tht even n unconstrined government cn decide to provide nd to strongly enforce wide scope of property rights. The second is tht even if the government is constrined, it my not strongly enforce wide scope of property rights. The imperfect ssocition between constrined government nd strong enforcement of wide scope of rights cn be explined by, s I hve rgued in the pper, culture nd technology. ulture determines the cost of enforcing, nd technology determines the income tht cn be generted by strong enforcement of property rights. As result, culture nd technology determines the pyoffs of strongly enforcing wide scope of rights for the rent seeking government s well s for the medin voter. Putting these pieces together we hve n explntion tht helps us understnd the fct tht the level of civil freedoms is better predictor of development thn is the level of the constrint on executives. This pper hs thus given n explntion for the cross-country pttern identified by BenYishy nd Betncourt (). No very specific, if ny, policy conclusions cn be derived from such result. I would rther emphsize three brod ones. First, the rgument suggests tht the (politicl) rgument for economic nd civil freedoms should not be seprted in the sense tht economic freedom is importnt for mteril welfre while civil freedoms re importnt from n immteril (morl) point of view. Rther the sme morl or instrumentl rgument cn be used in both cses. Second, culture hs been defined here s those informl fctors tht determine the costs of enforcement of the rules of freedom. This does not imply tht such fctors cn only chnge in the very long run, lthough culture is usully understood in this wy. Whether the informl fctors behind civil freedoms nd growth cn chnge reltively quickly is very interesting question for possible future reserch which seems to be crucil for nswering the big question s to how economic development cn be fostered. Third, this pper csts some doubt on those theories nd especilly those policy proposls (Freedom House ) tht emphsize the cuslity running from the constrints on executives to good outcomes, primrily economic development, nd tht suggest ccordingly tht these outcomes cn be reched by politicl pressure from outside. An nonymous referee rised the importnt issue s to why the cost of enforcement cn chnge within country, too. My model does not imply nything tht would nswer tht question if dmitting the possibility of their vribility is left side. The possibility of reltively quick chnge in informl fctors cn be n nswer to tht nd very importnt nd difficult issue for further reserch. Society nd Economy ()

22 PÁL ZEGLÉDI APPENDIX The government s decision in the bsic model of section nconstrined government In the cse of strong enforcement ( b s) the rtio of rent-seekers to producers will be in equilibrium nd the government cn tx wy the income bove. onsequently its net revenue is c. () In the cse of wek enforcement (b w ) rent-seeking hs return of b w until the return of production ( b w n) reches its minimum,, tht is until n. onsequently, in equilibrium the rtio of the number of rent-seekers to bw tht of producers will stisfy the eqution tht is, n * n*. () If the number of the lbour force (producers nd rent seekers) is normlized to one, then the number of rent seekers is this is why totl rent is nd the government s pyoff is n *, n * ( ), c. () Society nd Economy ()

23 WHY ARE IVIL LIBERTIES MORE IMPORTANT THAN EXETIVE ONSTRAINTS? onstrined government The cse of the constrined government is strightforwrd. The pyoffs re shown in n orgnized wy in Tble. The pyoffs cn be clculted in the sme wy in the more complicted cse when civil liberties re incorported. Tble The pyoff of the government under different regimes nd enforcement in the bsic model onstrined government nconstrined government Strong enforcement c c Wek enforcement c c strong enforcement strong enforcement c wek enforcement ompring the pyoffs mkes it cler tht revenue mximizing government will choose strong enforcement if ( ) c nd c, otherwise it will choose wek enforcement provided tht c c. c Figure. Strong nd wek enforcement of rights under unconstrined government in the bsic model Note: The figure incorportes the ssumptions tht >. c Society nd Economy ()

24 PÁL ZEGLÉDI strong enforcement A constitutionl government will choose strong enforcement if c <. Otherwise it will choose wek enforcement until c <. These results mke it cler tht the choice between strong nd wek enforcement will depend on the type of government, the cost of enforcement, nd the income of the productive ctivities. These solutions re summrized in Figures nd. Proof of Proposition onsider the four possible cses of property rights enforcement. () The government chooses civil liberties with strong enforcement (S-L) if the conditions S W S S S W S,,, L wek enforcement ll hold. These conditions cn be summrized s L L c Figure. Strong nd wek enforcement of rights under constrined government in the bsic model Note: The figure incorportes the ssumptions tht >. L L c min (,,, SL ), () c Society nd Economy ()

25 WHY ARE IVIL LIBERTIES MORE IMPORTANT THAN EXETIVE ONSTRAINTS? where ( d ) ( d) d( d) ( ),,, ( d ) With some clcultion it cn be shown tht S, SL ( ) [ ( d d ) ], [ ( d ) d ]. L d d. None of the three differences cn be positive becuse of the ssumptions regrding nd d. This mens tht c min(,,, ). SL Obviously it is only possible to stisfy this condition if is positive, tht is, if > d +. () The government will never choose civil liberties with wek enforcement. To show this, suppose tht the government chooses this option. In tht cse notice first tht four conditions would hve to be stisfied: W S W S W W W,,, L L L L L. These would require the cost of enforcement to stisfy the condition where mx(, ) c min(, WL ), () ( ) ( ) d ( d),, ( d ) d ( d) ( ), ( d ) WL d ( d). Society nd Economy ()

26 PÁL ZEGLÉDI But this condition cnnot be stisfied becuse one cn show tht: WL {( d)[ d( d ) ( )]}. ( d ) Tht is, the condition () bove will never hold. : WL () The government will choose no civil liberties with strong enforcement if ll of the following conditions hold t the sme time: These re true if where S S S W S W S,,,. L L min(, ) c min(,, S ), () ( d ) ( ) ( ) d ( d),,, ( d ) S It is possible to show tht min(,, ) the difference nd the difference S S ( )( ), S ( d)[ ( ) d d d ], re positive under the condition >. S However, when,min(,, ), becuse nd the difference S ( )( ), S. if y >. onsider tht both Society nd Economy ()

27 WHY ARE IVIL LIBERTIES MORE IMPORTANT THAN EXETIVE ONSTRAINTS? S ( d)[ ( d d ) d ] y is positive. This implies tht S. In ddition, if, then, which implies tht the condition () will tke the form of c. () The government will choose no civil liberties with wek enforcement if W S W W W S W,,,. L L To show tht the conditions formulted in the proposition below must hold so tht these four inequlites must be stisfied, note tht if mx(,, ) W, choosing this option will not py off under ny condition. But with some lgebric mnipultion it is possible to derive tht W ( )( ) which is positive if <; W dd ( ) ( d )which is positive if d ; W d [ ( d ) ]which is positive if ( d ) d ( d ) nd d d ( ). All three conditions re stisfied if < <. On the other hnd, it is possible to show tht mx(,, ). First, consider tht ( ) d ( d) ( ) ( ) ( d )( ) d, ( d ) tht is ( d) d( ) ( d )( ) ( d) ( d )( ) which is positive under the ssumptions we mde. Society nd Economy ()