tv.pS 3b6 POLICY RESEARCH WORKING PAPER 3 098 Governance and Economic Growth Mark Gradstein The World Bank Development Research Group Public Services July 2003 I 2678POLIcY RESEARCH WORKING PAPER 3098 Abstract Because protection of property rights cannot be framework. Drawing on North (1990), he presents a appropriated by any individual, it is widely recognized as model where economic performance and enforcement of being the state's responsibility. Moreover, recent property rights may reinforce each other. Initial empirical evidence .gests that protection of property conditions determine the economy's convergence to a rights leads to higher investment levels and faster high-income or a low-income steady state. Existing growth. The extent of r.-onerty -ghts protection differs empirical evidence offers tentative support for this significantly across countries. Gradstein integrates the theory. emergence of property rights within a simple growth This paper-a product of Public Services, Development Research Group-is part of a larger effort in the group to understand the role of governance for economic development. Copies of the paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contact Hedy Sladovich, room MC3-607, telephone 202-473-7698, fax 202-522-1154, email address hsladovich@worldbank.org. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The author may be contacted at mgradstein@worldbank.org. July 2003. (14 pages) The Policy Researcb Working Paper Seoes disseminates the fndings of Buork in progress to encourage the exchange of ideas about development issues. An objective of the series is toget the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the autbors. They do not necessarily represent the viewv of the World Bank, its E^xecutive Directors, or the countries they represent. Produced by Partnerships, Capacity Building, and Outreach Governance and Economic Growth Mark Gradstein mgradstein@worldbank-org JEL classification: D23, D72, 011, 04 1. Keywords: rent seeking, property rights, govemance, institutions, growth. 1. Introduction Maintaining law and order, in particular, securing property rights, is probably the most acceptable rationale for government intervention. Theoretically, it is argued that enforcement of property rights being a public good, its provision can only be materialized through collective action. Empirically, several studies, discussed more in detail below, have reported robust correlations between the enforceability of property rights and measures of economic performance. Yet, economies differ greatly in the extent to which property rights are enforced. Several pieces of empirical evidence suggest, in particular, a strong positive association between the level of a country's development and the enforcement of property rights. Bardhan (1997) for example, cites the experience of Singapore, where recent economic growth has induced a drastic reduction in corruption, so that Singapore is now one of the world's least corrupt countries. It is also interesting to compare the recent experience of some transition economies in East Europe in this regard. While countries like Estonia and Hungary have attained moderate scores on the quality of government and robust growth rates in the post-communist era, the relatively more backwards countries such as Moldova and the Ukraine have achieved little on both counts. Moreover, in their authoritative account, two experts on the transition experience in Russia write: "In developed market economies, a conventional system of property rights enforcement and contract implementation is provided by the government and the judiciary and paid for by taxes. However, if this can be considered to represent the first-best solution, the immediate implementation of such a solution in Russian case is hopeless... Widespread tax evasion has left the government without enough revenues to pay even those meager salaries it offers to its law- enforcement officers (including the police force, prosecutors, and judges)." (Braguinsky and Yavlinsky 2000) Taken together, these examples suggest a double feedback relationship between economic development and enforcement of property rights, in particular, indicating that affluent economies are likely to more affectively enforce property rights than poor economies.1 This paper is an attempt to capture this relationship in a simple growth model, augmented with political economy features. It is assumed that a part of productive investment in the economy is subject to rent-seeking redistributive activity. The fraction of resources available for such redistribution is endogenously determined through collective decisions on the extent of property rights enforcement. Specifically, property rights can be fully secured by incurring a cost. In line with the public good nature of property rights, we assume indivisibility in the production of their enforcement. This ensures that enforcement of property rights will only take place in rich economies, where the individuals are affluent enough to be willing to meet the enforcement cost. But a better enforcement of property rights causes economic growth, thus perpetuating the willingness to secure property rights. As a result, it is shown that two steady states are likely to be realized: one, with a full protection of property rights and a high income level, and another, with only a minimal protection of property rights and a low income level. One implication of this analysis emphasizes the importance of commitment mechanisms to ensure enforcement of property rights; the absence of such commitment may induce lower investment and higher rent seeking thus lowering growth. Another implication indicates a role of international lending institutions in providing resources to implement governance reforms so as to allow a poor economy to take off. The paper is organized as follows. The next section presents the basic setup, which is followed by the presentation of the results in section 3. Section 4 discusses empirical findings, policy implications, and relation to the literature, and section 5 concludes with brief remarks. 2. Basic Framework The economy is populated by a continuum of households indexed by i and represented by the unit interval, each consisting of a parent and child; it operates in discrete time t. The initial level of household income is yo, and y denotes the income 'As North (1990) puts it, "economic history is overwhelmingly a story of economies that failed to produce a set of economic rules of the game (with enforcement) that induce sustained economic growth." 2 level in period t. Initially, the amount of law and order as reflected by the protected fraction of individual income is Lo, 0 < Lo < 1, and L, denotes the protected fraction of income in period t. The level Lo is interpreted as a minimal protection of property being guaranteed by the prevailing social norms or "natural law" and, therefore, not requiring any explicit costs of enforcement.2 The role of informal, cultural factors in maintaining property rights has been recently empirically documented, see, for example, Knack and Keefer (1997b). More specifically, Mauro (1995), and Easterly and Levine (1997), in their cross-country analyses find that measures of ethnolinguistic fractionalization are directly related to corruption and rent seeking; Alesina and others (1999), detect their negative impact on policies within U.S. communities. Glaeser and others (2000), find relationship between ethnicity and trust, which affects the efficiency of economic transactions in an experimental setting. These factors can provide an empirical underpinning for the initial minimal level of property rights enforcement. To obtain interesting results, Lo is assumed to be small. In contrast, to secure a full protection of property rights, L, = 1, requires a costly investment, T, which is funded through taxes.3 Because all individuals within a cohort possess identical incomes, their burden in financing the cost of enforcement is also identical, so that Tcan also be interpreted as an individual cost in protecting property rights. The interesting case will be the one where Tis not very small. In each period, the individuals divide their income between paying taxes to meet the cost of law and order, current consumption cit and investment hit so as to satisfy the budget constraint: yit= cit + kit + T6t 2 See Konrad and Skaperdas (1998), for a simple model of how social norms shape initial property rights. 3Apart from direct costs of designing the legal systemn, this consists of the cost of collecting information, monitoring behaviour, and building reputation for law enforcement. 3 where , is an indicator function, assuming the values of 1 if property rights are fully protected and 0 if no investment is made to guarantee property rights protection. Each household is endowed with one unit of time, which is allocated between productive activity, wit and unproductive activity, uit according to the constraint: 1= Wjj + ui, (2) Next-period gross income of household i, zt+j is then produced using both capital and labor as inputs, according to the following technology: zit+l = A kita wit (3) where A > 0 is a exogenously given parameter of technology to which all individuals have access, and 0 < a < 1, which implies diminishing returns to scale. While a fraction L, of income is fully protected, the remaining 1- L, is available for redistribution through unproductive or rent-seeking activity. Letting Z,+, = f ziI+ di denote the aggregate income, (1- L,)Z,+, is, therefore, the amount of income available for rent seeking. It is assumed that a spending of ui, on rent seeking secures individual i the fraction of r(ujt)/ f r( uit)di of aggregate income, where r is increasing, concave, with r(O) = 0, r(l) = 1, and r(O)/ f r( 0) di = 0. This specification is very common in the rent-seeking literature-see Nitzan's survey (1994). The important difference is that in the present formulation, because every agent is atomistically small, the individuals presume that their rent-seeking efforts do not have aggregate consequences; see, however, an extension below which modifies this assumption. The net next-period income, yit+1, is the sum total of work-generated income and income that accrues through rent seeking, Yit+1 = LI zi, + (1- LI) Z,+, r(u1:)/ f d( Ui) di (4) 4 The income is bequeathed to one's child. Each parent's preferences derive from consumption as well as from the amount of income transferred to the child. Assuming logarithmic preferences, we write: ci(t, yit+I) = (1-fllog(c,t) + /llog(yi,+A), 0 < p < 1 (5) In each period, the adult individuals first detennine the extent of property rights protection by collectively setting Lt. Then each parent makes his consumption- investment decision; thereafter, the work-appropriation decisions follow. The equilibrium consists of such mutually consistent decisions. 3. Equilibrium The above assumptions guarantee that, despite the dynamic setting, the decisionmaking problem is essentially a static one and consists of equilibrium allocations made by the parents. The analysis proceeds backwards starting with the determination of time allocation between work and rent seeking given the investment decisions; its details are presented in the appendix available from the author. Note that, given that the productivity parameter A is large enough, under each regime the economy converges to a steady state level of income denoted yo when Lt = Lo, andy' when L, = L,.4 Provided that Lo is sufficiently small as we have assumed is the case, y2 >y°, Figure 1 illustrates this by presenting the intertemporal income evolution in both cases. Moreover, when current income level is small enough, next- period income is lower under full protection of property rights than under minimal protection. The reason for this is that the economic performance under full protection of property rights is adversely affected in the short run by the tax burden, but recovers afterwards. These properties are summarized in Proposition 1 below and illustrated in Figure 1. 4Specifical1y,)y=(A(1-u))l't-)[a8fLoy0l(a/6Lo+ I -A]r-a)andyl=A (a8y' - 7)! (af8+ 1 fl)]a 5 : Yt+I Figure 1: Intertemporal income evolution 450 y t+i yO * yl Yt Proposition 1. Current consumption is higher, but steady state income level is lower under minimal protection of property rights than under full protection. Next- period income is higher under the former if present income is low, but is higher under the latter when present income is high. The above results have direct implications for welfare comparisons between the two regimes. When present income is low, both consumption and next-period income are higher without full protection of property rights, so that this regime attains a higher level of welfare. In contrast, when present income is high enough, the resulting increase in next-period income under full enforcement of property rights more than compensates the lower level of consumption, thus causing welfare to be higher. The intuition here is straightforward: while the tax burden associated with full enforcement is significant in a poor economy it ceases being so, as taxes contribute a smaller share of income. To sum up, 6 Proposition 2. When the economy is poor enough, the regime of minimal protection of property rights leads to a higher welfare level; however, in a rich economy, full property rights protection is a preferable regime. It can then be shown that welfare increases in income faster under full protection of property rights. Along with the above proposition this implies that there exists a unique threshold level of income that leads to indifference between the two regimes, y**: when present income is higher than the threshold (and only then) is the regime of full protection of property rights superior to the regime of minimal protection. Also note thaty** must be higher than the level of income which makes next-period income equal under the two, y*. The reason for this is that consumption is lower under full protection. Moreover, if Lo is sufficiently small then y** > yO (becauseyo is arbitrarily small)-see Figure 1. We are now in a position to trace the intertemporal evolution of the economy. If the initial income level is below y**, then minimal protection of property rights is welfare superior. If the economy's growth rate is low in this case, then minimal property rights protection will continue to dominate throughout the convergence to the steady state; specifically, if Lo is small enough, the steady-state income level is small, yo