Reading notes LT - Summary Development Economics PDF

Preview

Summary

WEEK 1 LT The Regulation of Entry by Djankov, S.; La Porta, R.; Lopez-de-Silanes, F.; Shleifer, A.Can Labor Regulation Hinder Economic Performance? Evidence from India by Besley, T.; Burgess, R.Motivation To evaluate economic theories of regulation and examine which governments regulate entry (more ...

Description

WEEK 1 LT Motivation

`The Regulation of Entry` by Djankov, S.; La Porta, R.; Lopez-de-Silanes, F.; Shleifer, A. To evaluate economic theories of regulation and examine which governments regulate entry (more representative or more autocratic)

`Can Labor Regulation Hinder Economic Performance? Evidence from India` by Besley, T.; Burgess, R. To investigate whether industrial relations climate in Indian states has affected the pattern of manufacturing growth in 1958-1992

Special settings

Data on the regulation of entry of start-up firms in 85 countries in 1999, Although amendments to the industrial disputes act were passed on central look at the official requirements, cost and time to understand the structure of level, each state then had a choice which to accept; there were also states official regulation with no amendments (control states)

Theory

The public interest theory predicts that heavier regulation is associated with socially superior outcomes, but the public choice theory sees regulation as socially inefficient leading to less competition and higher competition.

Labor regulation affects economic performance via 1) a relative price effect (raising MC of employee) and 2) an expropriation effect (discouraging investment)

Empirical design

Use three measures of entry regulation: 1) the number of procedures that firms must go through, 2) the official time required to complete the process, and 3) its official cost. These measures are uniform and easily comparable across countries. To keep track of all the procedures required by law to start a business include only those that require the entrepreneur to interact with outside entities: state and local government offices etc. To measure time, we collect information on the sequence in which procedures are to be completed and rely on official figures as to how many business days it takes to complete each procedure. Estimate the cost of entry regulation based on all identifiable official expenses: fees, costs of procedures and forms, photocopies, fiscal stamps, legal and notary charges, etc.

Utilize both time series and cross-section variation in labor regulation. Overcome OVB from cross-sectional variation across countries: withincountry panel data allows to control for state and time fixed effects.

Data (frequency, questions)

Construct country rankings by the total number of entry procedures, then by the time it takes to complete them, and finally by the cost of entry (Table III). Table IV: OLS regression of seven measures of consequences of regulation on # of procedures. Table VII: regressions of # of procedures on political variables controlling for income. Official sources used for the number of procedures, time and cost; for each country at least one independent report from a local law firm; focus on `standardized` firm with set of characteristics, yet data is very likely to underestimate complexity of entry; costs are only official not including potential bribes Data on a variety of characteristics of political systems (how good government is).

Panel data regressions:

Data and context allow them to alleviate reverse causality concern. Reverse causality solutions: Solution 1: Matching - match states by level of unionization (create pairs: treatment and control states with similar levels of unionization), then regress difference in output on difference in labor regulations for each pair (p.113). Solution 2: IV - two instruments: level of unionization interacted with Post1977 Dummy and land revenue collection regimes interacted with Post1977. Exploit 1975-77 Gandhi’s declaration of state of emergency, which led to decrease of power for the congress party and changes in political power in several States (p.116). Data on labor regulation come from looking at each state amendments to the industrial disputes act; the amendments are classified as pro-worker (coded as 1)/employer (as -1) or neutral (as 0) and then cumulated (so the higher the score, the more pro-worker) shown on fig. II.

Key findings

Interpretation/ policy implications

WEEK 2 LT Motivation

The world average of number of procedures is 10, of official time is 47 business days, of official cost of 47% of annual per capita income; # of procedures is highly correlated with time and cost (Table IV); Table III Panel B: the higher GDP per capita, the lower the cost of entry and shorter delays The stricter regulation is not associated with higher quality products, better pollution etc., but is associated with sharply higher levels of corruption and greater unofficial economy. Even holding constant various measures of freedom of press and per capita income, # of procedures is still associated with inferior social outcomes; Table VII: countries with more representative governments have lighter regulation (fig. IV) The fact that corruption levels and intensity of entry regulation are positively correlated (fig. III, Table V.A) suggests that heavy regulation promotes corruption Problems: with Table VII: multicollinearity and direction of causation. The public choice theory then predicts that rent extraction from regulation should be moderated by better governments. Context matters: need to consider local policy/ institutional environment.

Table II: labor regulation is strongly pos. correlated with workdays lost to strikes and lockouts per worker Table III: non-agricultural output is neg. correlated with labor regulation (col 3), specifically registered manufacturing Table V: 1) regulation has no sig effect on workers` payments (col 3); 2) increasing worker bargaining power reduces capital formation (col 4); Table VII: pos correlation between pro-worker regulation and urban poverty (col 2) Regulation in pro-worker direction brought lowered output, employment, investment and productivity in registered or formal manufacturing, is also associated with increases in urban poverty and increase of informal sector, reduction in overall manufacturing output Problem of endogeneity (larger interests in manufacturing to have greater pressure in pro-worker direction) is ruled out in Table VI Attempts to redress the balance of power between capital and labor can end up hurting the poor, workers do not appear to be gaining from pro-worker amendments Welfare results: pro-employer regulation decreases urban poverty Tighter labor regulation has a positive impact on output in unregistered manufacturing, deterring formal registration

`The Value of Democracy: Evidence from Road Building in Kenya` by Burgess R., Jedwab R., Miguel E., Morjaria A To determine to what extent ethnic favouritism is prevalent and whether the emergence of democracy has helped to mitigate it

`The political economy of government responsiveness: theory and evidence from India` by Besley T., Burgess R. To understand what makes government responsive to citizens` needs; what institutions and mechanisms enable vulnerable citizens to have their preferences represented in policy

Special settings Road building across 41 Kenyan districts with dramatic ethnic segregation (each dominated by a single ethnic group); road expenditure can be directly measured and is centrally allocated; switches both in the ethnicity of the president and switches into and out of democracy under the same president. Major ethnic groups: Kikuyu (20%), Luo (13%), Luhya (13%), Kalenjin (11%). Theory

The authors constructed a counterfactual road network based on the goal of maximizing market potential (i.e. the utopian model of how the network should have looked like). Market potential is defined as the sum of populations in the settlement pair divided by distance between them. Construct a counterfactual road network that has the same length of paved roads as the real network and shows how the network should have looked like if a social planner was maximizing market potential

India is home to a large vulnerable population, which regularly faces natural shocks and measures have been developed to deal with them; India has elected state governments, there is a relatively free and independent press, so the authors are able to demonstrate the link between mass media, political factors and government responsiveness.

A two-period model: at 1 a politician has been voted into office; politicians can be altruistic / selfish / opportunistic; citizens are vulnerable and nonvulnerable; a fraction  of the vulnerable is hit by a shock that can be mitigated by public action; vulnerable citizens can learn about politician type from the media or if they experience a shock; mass media allow citizens to monitor politicians, this creates incentives for politicians to build reputations for being responsive. So are states with high newspaper circulation and better functioning political systems also more responsive to droughts and floods?

Empirical design

Panel data regressions of this form: First method is graphical inspection of: 1) Districts of the president; 2) Kikuyu vs Kalenjin vs others. The main outcome variable in empirical analysis is the share of road expenditure received by a district divided by the population share of the district in the national population, so a value of 1 implies road spending exactly proportional to population. Second method is regression:

Data (frequency, questions)

District-level panel data on expenditure for new roads for all Kenyan districts for each year from 1963 to 2011. Panel of road presence for all the districts from 1963-2011 from Michelin maps: independent from government interference → help addressing the concern that road spending might not be accurately recorded. Counterfactual paved road network for 1964-2002: where and when paved roads would have been built if a social planner wanted to maximize market potential. Ethnic composition for each district; ethnic affiliation of serving president. Timing of periods of autocracy and multiparty democracy in Kenya.

Key findings Kenyan districts that share the ethnicity of the president receive twice as much expenditure on roads (Table 1) and almost five times the length of paved roads built relative to what would be predicted by their population share (Table 2). However, these biases are prevalent only in periods of autocracy, and disappear during democracy (Table 4). The president has been the dominant force in allocating road spending, the vice president is also able to a more limited extent to skew allocation, but both these forms disappear during democracy (Table 5). Fig. 2: democratization in Kenya co-moves with the rest of Sub-Saharan Africa. Fig. 4: ratios are above 1 all throughout autocracy for coethnic districts, but near 1 on average for both types during democracy Fig. 5: rise in spending on Kalenjin districts is meteoric (from 0.5 to close to 3 in a few years); the `other` districts line has a U-shaped pattern (close to 1 during democracy, falling during autocracy and rising back during democracy). Fi 7: i th i K nd Af i hol highe t duri

Data from 16 major Indian states from 1958-1992; exploit cross-state and cross-time variation; India is linguistically diverse: exploit this to see whether regional papers have different effects than papers published in English or Hindi; India has been prone to droughts and floods (Fig. 3 & 4): exploit this variation; relief system to deal with these shocks are well developed, but requires triggering and vigorous implementation

Economic factors (e.g. state income, revenue from centre) have a limited influence on government responsiveness (Table III); however, columns 2 and 5 (Table III) reveal a positive correlation between newspaper circulation levels and two measures of government responses. Local press published in regional languages play a key role in forcing state governments to respond to shocks (columns 3 and 6 Table IV). Greater electoral turnout is associated with greater responsiveness (columns 1 and 4 Table VI); greater political competition is associated with increased government responsiveness (columns 2 and 5 Table VI); states are more responsive to droughts closer to elections (columns 3 and 6 Table VI).

Interpretation/ policy implications

OVB problem: regime shifts are not exogenous. There could be other factors correlated with the regime changes that have an impact on road building. Kenya is representative of sub-Saharan Africa: results might thus be applicable to a wider range of African countries ( external validity) Even `imperfect` democratic institutions (like Kenyan) have value by imposing constraints on executive action, so even moving in prodemocratic direction may bring the positive results. Free flows of information, a vocal civil society and an independent parliament may all curtail the ability of the executive to discriminate between districts. Removing severe ethnic biases in public spending may be a factor in explaining why economic growth and multiparty democracy have coincided in Africa. `Economic shocks and Civil Conflict: An Instrumental Variable` by Miguel, E., Shanker S., Ernest S.

Mass media and open political institutions can affect government activism and responsiveness, while economic development appears to be relatively unimportant. Elections provide an incentive for politicians to perform which can be enhanced by development of the media. Shared vulnerability and mass media allow minority to have big effect on policy choices of politicians. The importance of local language newspapers in transmitting information. Freedom of information and democracy are central in determining whether preferences of citizens are reflected in government policy. Elections alone are not sufficient to deliver a responsive government, so additional effective institutions for information transmission are required. Development of these key institutions may be more important than economic development. `Commodity Price Shocks and Civil Conflict: Evidence from Colombia` by Oeindrilla D., Juan V.

Motivation

To estimate the impact of economic conditions on the likelihood of civil conflict using rainfall variation as an instrumental variable for economic growth.

To assess how different income shocks affect conflict exploiting exogenous price shocks in international commodity markets and whether different types of shocks have differential effects.

Special settings

Cross-country regression for 41 African countries during 1981-99, which are very undiversified, thus a great dependence on rain-fed agriculture and variations in rainfall may credibly drive variations in economic growth.

Within country variation. Municipalities in Colombia between 1988-2005. Rich data; high incidence of conflicts (get both geographic and time variation). Some municipalities produce labour-intensive commodities (coffee) more intensively, while others specialize in the extraction of natural resources (oil).

Theory

Poor economic performance can cause civil conflict through labor markets (pool of underemployed youths), income inequality etc.

Income shocks may have two opposite effects: 1) the opportunity cost: higher wages may lower conflict by reducing labor supply to appropriation; 2) rapacity effect: higher contestible income may increase violence from raising gains from appropriation.

Empirical design

Weather variation is used to instrument for per capita economic growth in the first stage controlling for other country characteristics, then instrument growth on conflict in the second stage. Table 2: the first-stage relationship between rainfall and income growth is strongly positive (col.1). Benefits: 1)address the problem of measurement error in African national income; 2)address the attenuation bias from mismeasured explanatory variables.

Difference in difference estimator by assessing whether changes in commodity prices affect violence disproportionately in municipalities that produce more of these commodities. The main specification uses one step via 2SLS estimation. Post in the regression means period after the price shock, treat is being heavily engaged in a commodity production. Possible endogeneity concerns: conflict affecting commodity prices. So, use IV for coffee prices: coffee export volume as an instrument. However, potential violation of exclusion restriction: other exporting nations adjust their exports according to what may be happening in the Colombian coffee market (e.g. prices or quantities).

WEEK 3 LT

Data (frequency, questions)

The 19 sample years (1981-99); using definition of conflict with threshold of 25 battle deaths per year, but no information at the sub-national level or at a frequency below annual. Rainfall dataset relies on both gauge and satellite data on monthly rainfall. The principal measure of a rainfall shock is the proportional change in rainfall from the previous year.

The dataset on the Colombian civil war includes more than 21,000 warrelated episodes in over 950 municipalities during 1988-2005. Incidents are aggregated to the municipality-year level. Oil is measured in the average barrels of crude oil produced per day in each municipality in 1988. All agricultural commodities are measured in the hectares of land used for cultivating that crop in a given year.

GDP growth is significantly negatively related to the incidence of civil conflict in sub-Saharan Africa: a five-percentage-point drop in annual economic growth increases the likelihood of a civil conflict (at least 25 deaths per year) in the following year by over 12 percentage points. Table 4: IV-2SLS columns 5 (country controls) and 6 (fixed-effects).

Table 2: col.1-4 show that coffee price shocks have a negative relationship with conflict and that the oil price shock exerts the opposite effect on conflict. Table 3: col. 1-2 show that coffee shock exerts substantially larger effects on both wages and work hours of rural workers in areas with more coffee cultivation, while oil interaction terms display insignificant effects on both labour outcomes; col.3 indicate that the oil price shock significantly increases capital revenue at the disposal of municipal government.

Key findings

Interpretation/ policy implications

The impact of income shocks on civil conflict is not significantly different in richer, more democratic, more ethnically diverse, or more mountainous African countries or in countries with a range of different political iEconomic tit ti factors l h dominate t i ti all others in determining the incidence of civil conflict, while institutional and social characteristics have minimal impact on averting the effect of economic shocks. Policy solutions: design of better income insurance for unemployed young men during hard economic times (e.g. public work projects funded by international donors, in particular focused on irrigation or road building). However, the identification strategy is inappropriate for other regions of the world, since weather is not sufficiently linked to income growth.

Special settings

Theory

2. Since natural resource revenue is found to promote rapacity, improved monitoring may prevent these funds from fuelling conflict. 3. Since funds leak through local governments, fiscal structure may interact with price shocks in affecting conflict outcomes

WEEK 4 LT Motivation

1. Price stabilization schemes which place a floor on the price of labourintensive commodities can help mitigate violence in the wake of price shocks.

“Do Rural Banks Matter? Evidence from the Indian Social Banking To evaluate how a large state-led bank branch expansion program in India affected rural poverty The social banking program between 1969-90; to open a branch in a banked location must open four branches in unbanked ones; roughly 30,000 branches were opened in rural locations with no pri...