Growth, Public Investment and Corruption with Failing Institutions - PowerPoint PPT Presentation

Introduction Growth Model Empirical Analysis Results Conclusion Growth, Public Investment and Corruption with Failing Institutions David de la Croix 1 Clara Delavallade 2 1 dept. of economics & CORE, Univ. cath. Louvain www.de-la-croix.be 2 Panth´ eon Sorbonne Economie October 2006 1 / 26

Introduction Growth Model Empirical Analysis Results Conclusion Channels through which corruption hampers growth Corruption... 1. acts as a tax on entrepreneurship and productive activity, discouraging investment; 2. pushes firms and activities into the informal sector; 3. diverts talent into rent-seeking; 4. decreases tax revenue and affects state budget equilibrium; 5. lowers the quality of public infrastructure and reduces public spending efficiency; 6. distorts the public spending structure. 2 / 26

Introduction Growth Model Empirical Analysis Results Conclusion Research question • Key question : How does corruption distort the structure of public spending, therefore growth? • Antecedents : • Mauro (1997) • Empirical results: Education expenditure are scaled down in countries with widespread corruption. • Model: Corruption acts as an additional tax on budget surplus and does not distort the composition of public spending. • Gupta et al. (2000) • Empirical results: Corruption increases military spending • Delavallade (2006) • Empirical results: Corruption modifies the composition of public spending • ... reduces the share of education and health spending 3 / 26

Introduction Growth Model Empirical Analysis Results Conclusion Our set-up A model with endogenous corruption • Two types of public investment with different exposition to rent-seeking. • Households choose between productive activity and rent-seeking: the incentive constraint. • Voters choose the allocation and level of public investment subject to the incentive constraint. • Rent-seekers may have more political influence than others in the voting process. 4 / 26

Introduction Growth Model Empirical Analysis Results Conclusion Methodology 1. The dynamic growth model • How is standard growth theory modified by rent-seeking. • What are the key parameters. 2. A joint empirical estimation of a system of simultaneous equations with original instruments. • How are the intensity of corruption, the structure of public investment and GDP growth affected by the predatory technology and the distribution of political power? 5 / 26

Introduction Growth Model Empirical Analysis Results Conclusion Our contribution • Bridge the gap between the static theory of corruption and growth. • The object of corruption is explicit: Corruption technology allows rent-seekers to capture part of public spending. • Estimation of the effect of failing institutions on the allocation of spending. 6 / 26

Introduction Growth Model Empirical Analysis Results Conclusion Technology Two types of productive public capital: h t and k t . (1 + n ) h t +1 = (1 − δ H ) h t + g t (1) (1 + n ) k t +1 = (1 − δ K ) k t + (1 − ν x t ) i t (2) ν : corruption technology Production function: q t = b [1 − x t ] f [ h t , k t ] . 1 − x t : share of the population in the production sector 7 / 26

Introduction Growth Model Empirical Analysis Results Conclusion Technology Budget of the government: T t = g t + i t Firms are owned by the workers: y t = b [1 − x t ] f [ h t , k t ] − T t 1 − x t Utility in the production sector: U t = u [ y t ] Utility in the rent-seeking sector: V t = u [ ν i t − g t − i t ] 8 / 26

Introduction Growth Model Empirical Analysis Results Conclusion Possible Regimes U t U t U t V t V t V t x t x ⋆ x t x ⋆ x t t = 0 1 1 1 1 t = 1 1 x ⋆ t ν ν ν 9 / 26

Introduction Growth Model Empirical Analysis Results Conclusion Incentive Constraint Proposition If corruption at equilibrium satisfies x t ∈ (0 , 1 /ν ) , then the following constraint holds: U t = V t Γ[1 − x t ] f [ h t , k t ] = ν i t . (3) ⇒ 10 / 26

Introduction Growth Model Empirical Analysis Results Conclusion Probabilistic Voting Political parties maximize their vote share. Equivalent to maximize a social welfare function: ∞ � ρ t W t subject to (1) , (2) , (3) , and h 0 , k 0 given . max t =0 with W t = (1 − x t ) U t + (1 + θ ) x t V t (4) θ : political power of the rent-seekers. Resolution: Kuhn-Tucker 11 / 26

Introduction Growth Model Empirical Analysis Results Conclusion Four possible regimes • Regime with no corruption, no distortion (benchmark). • x t = 0, • λ t = µ t , • needs ν and θ low enough. • Regime with no corruption, but with distortion • x t = 0, • λ t < µ t • Interior regime with some corruption. • x t ∈ (0 , 1 /ν ), • Incentive constraint holds, • Public spending are distorted. • Regime with maximum corruption. Only temporary. 12 / 26

Introduction Growth Model Empirical Analysis Results Conclusion A glimpse of the results Proposition Along a balanced growth path with h and k given by (5) and (6), if Γ[1] f ′ H [ h , k ] ν < ( n + δ K ) k u [ y ] + u ′ [ y ] ( ν ( n + δ K ) k + Γ ′ [1] f [ h , k ]) 1 + θ < . u [( ν − 1)( n + δ K ) k − ( n + δ H ) h ] 1. There is no corruption: x = 0 . 2. The marginal productivity of both types of capital is equalized: 1 − δ H + Γ[1] f ′ H [ h , k ] = 1 − δ K + Γ[1] f ′ K [ h , k ] . (5) 3. The Modified Golden Rule holds: H [ h , k ] = 1 + n 1 − δ H + Γ[1] f ′ . (6) ρ 13 / 26

Introduction Growth Model Empirical Analysis Results Conclusion Possible regimes θ ν θ φ x g i g / i h / k y A 4 1/4 0 0 0.37 0.37 1 1 2.13 B 9 1/4 0.015 0 0.36 0.30 1.19 1.19 2.04 C 4 3/4 0.262 0.20 0.24 0.54 0.45 2.02 1.38 φ t > 0, x t ∈ (0 , 1 /ν ), ω t = 0 C C’ 1 φ t = 0, x t = 0, ω t > 0 A B φ t > 0, x t = 0, ω t > 0 0 10 ν 14 / 26

Introduction Growth Model Empirical Analysis Results Conclusion System of simultaneous equations three equations where each endogenous variable is a function of the measured parameters and initial conditions Ratio it = α 1 + α 2 Techcor it + α 3 Techcor it + α 4 Polbias it + α 5 Polbias  ln Y 0 ln Y 0 it    + α 6 Patience it + α 7 Pop it + α 8 Tropic it + α 9 Ldlock it + α 10 ln Y 0 it + ε it      Corrup it = β 1 + β 2 Techcor it + β 3 Techcor it + β 4 Polbias it + β 5 Polbias    ln Y 0 ln Y 0 it  + β 6 Patience it + β 7 Pop it + β 8 Tropic it + β 9 Ldlock it + β 10 ln Y 0 it + ρ it     Growth it = γ 1 + γ 2 Techcor it + γ 3 Techcor it + γ 4 Polbias it + γ 5 Polbias    ln Y 0 ln Y 0 it     + γ 6 Patience it + γ 7 Pop it + γ 8 Tropic it + γ 9 Ldlock it + γ 10 ln Y 0 it + ς it  15 / 26

Introduction Growth Model Empirical Analysis Results Conclusion Endogenous variables • Ratio g / i : Government Finance Statistics Yearbook • Share of spending for each sector in total spending • g = education, health i = housing, fuel and energy, agriculture, mining and manufacture, transport (and other economic activities) • Level of corruption: 1 of the 6 World Bank governance indicators (Kaufmann, Kraay, Mastruzzi, 2003), transformed into a 0 to 5 index • Aggregated perceptions indicator of grand and petty corruption • Growth rate: World Development Indicators (World Bank) 16 / 26

Introduction Growth Model Empirical Analysis Results Conclusion ν and θ • Techcor : Rule of Law index • aggregate of perceptions of the incidence of crime, • of the effectiveness and predictability of the judiciary, • and of the enforceability of contracts. • Polbias : indicator of the lack of political rights taken from Freedom House . Few political rights for the population indicate a strong concentration of power in the hands of a very few. 17 / 26

Introduction Growth Model Empirical Analysis Results Conclusion Non linear terms We have included non-linear terms: α 2 Techcor it + α 3 Techcor ln Y 0 it In the benchmark regime, the endogenous variables are not affected by small variations in ν and θ . 1 10−year Growth Rate .5 0 −.5 0 1 2 3 4 Technology of Corruption 18 / 26

Introduction Growth Model Empirical Analysis Results Conclusion 63 Countries’ Legal and Political Institutions 7 Theta CHN 6 CMR SYR IRN PAK BDI 5 TUN EGY FJI KEN UGA 4 SGP MYS ETH ZMB 3 TUR LSO COL PER VEN BRA 2 IDN LKA NPL BGD BWA THA JAM SLV MEX 1 CHL IND MDG KOR PHL NZL SWE DOM PNG AUS DNK NOR NLD CAN GBR BEL ESP MUS ARG PAN BOL CYP GRC FIN IRL ISR 0 CRI URY AUT DEU 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 ISL USA Nu 19 / 26

Introduction Growth Model Empirical Analysis Results Conclusion Estimation method • Three-stage least squares: • Enables to take into account the correlation between residuals. Budgetary choices, corruption level and growth rate are interdependent: some of their omitted factors are common (institutional and political context). • Reduces endogeneity • Instruments: • state antiquity • number of years of independence of the state • latitude of the country • origin of the legal system (french, british, socialist) • ten-year lagged index of political rights • ten-year lagged growth rate of the population • percentage of natural resources exports in GDP 20 / 26