development economics, condensed

Dollar and Kraay (2002): Effect of Growth on the Poor

A typical paper by World Bank economists. Takes a readily available cross-country data set (Summers-Heston, can be found at the Penn World Tables). This uses real per-capita GDP income figures to proxy for household income (and thus welfare). This is not very accurate on a large number of counts. Anyway, their results purport to show that the growth of the share of the income of the poor grows one-for-one with growth in incomes, in general. Thus, policies which help countries get rich help the poor, and there is no need for additional pro-poor policy. They argue that this is not due to a “trickle down” effect, but that effects which encourage growth such as institutions and rule of law simultaneously help everyone. I am, as usual suspicious of cross-country data simply done on a wider range of countries, as the cross-country variation in itself masks so many determinants, so one learns little about specific policies.

Mankiw, Romer and Weil (1992): Empirics of Economic Growth

Countries do not grow at the same rate, but conditional convergence is possible in the long run. This means that increasing savings rates, reducing population growth will increase short-run growth, and thus influence the level of GDP per capita. Population growth, especially is seen as having an especially large drag on growth as available human capital and physical capital must be spread more thinly.

Aghion et al. (2005): Effects of financial development on convergence

Financing is critical to the ability to absorb new technology. In less-developed countries, financing is constrained because of asymmetric information. This is because there is a temptation for voluntary default, where entrepreneurs hide the results of their innovations, or take the money and run. Thus, there is credit rationing at below-market interest rates, to reduce the incentive to default. Thus improving financial infrastructure and enforcement technology through a better banking system, credit-rating system will encourage growth through financing of technology

Hall and Jones (1999): Social infrastructure and Economic Performance

Finds tht differences in productivity matter more than differences in capital or human capital accumulation in determining incomes. Argues that this is determined by social infrastructure, the different policies and institutions which encourage the accumulation of capital instead of predatory behaviour.

Isham and Kaufmann (1999): Productivity of Investment Projects

Examines the economic rate of return of 1276 World Bank funded projects. Finds that countries which do not distort their exchange rate, trade and pricing policies have the highest rate of return on investments.

Galor and Zeira (1993): Income Distribution and Macroeconomics

Shows that in imperfect credit markets, and fixed costs of investing in human capital, the initial distribution of wealth affects aggregate ouput, and explains the persistent differences in performance between countries. This is because the poor, with less assets/bequests are unable to access educational opportunities.

Easterly (2007), Engermann and Sokoloff (1997): Inequality

Show that agricultural endowments predict structural inequality. Abundance of cash crops encourages extractive institutions and plantation labour. Creates structural inequality, which leads to bad institutions and underdevelopment intent on perpetuating the power structure, and the elite. In contrast to family faming systems in North America, which leads to a more equal distribution fo income and development of market inequality, with growth through incentives. Shows that bad institutions affect income.

Bloom and Sachs, Sachs and Warner (1997)

Suggest that tropical location and geography, being landlocked are greater factors influencing development, through delibilating disease, and thus that intervention to ameliorate these factors can lead to growth, especially in Sub-Saharan Africa

Acemoglu, Johnson and Robinson (2001): Colonial Origins of Development

Argues that settler mortality caused by the local disease environments affected whether Europeans created extractive colonies, or decided to settle and create good institutions. Thus uses settler mortality from the 17th-19th century to instrument for institutions and their effect on growth. Their conclusion: bad institutions suck

Murphy et al (1989)

Simultaneous industrialization and be profitable when each sector alone may not be if they did so alone. Encourages a big-push in industrialization through targeting leading sectors with many linkages, or in sectors where the private sector is unlikely to lead due to high fixed costs but which are essential and have many linkages, such as railways.

Jones and Olken (2004): Do Leaders Matter?

Tries to determine if the “great person” theories of history as advanced by historians such as Thomas Carlyle are more appropriate as compared the Marxist view that we are just the product of circumstances. Use the assassination of 57 leaders as events to see if there are significant changes in a country’s growth upon their death. This impact is generally through monetary policy.

Bandiera and Rasul (2006) : Social Networks and Technological Adoption

Study in Northern Mozambique showing an inverse-U-shaped relationship between adoption of a new cash crop (sunflower) and the number of people in his network who had already adopted. The incentive to do so is initially increasing as there are more people to share information with to determine the optimal target input levels. However, there is also an incentive to free-ride and delay adoption to see how the trials turn out. Example of complementarities, show importance of targeting and subsidizing people with many links.

Bandiera (2005): Contract Duration in Northern Sicily

Shows that contract duration and whether or not output is shared between the landlord and tenant depends on the type of crop and the characteristics of the borrower and lender. A poorer tenant is more risk-averse and thus would like to sharecrop. Trees are more risky and require more fixed investment, and thus would attract longer contracts. Richer tenants prefer retaining the marginal output. In addition, poorer tenants have shorter contracts in order for the landlord to maintain a threat of eviction.

Khwaja and Mian (2005): Do lenders favour politically connected firms?

Shows that lenders in Pakistan favour politically connected firms, or those who have a politician on their board, and the strength of this favouritism depends on the their electoral performance. Shows a mechanism that inequality persists in Pakistan, that electoral success depends on wealth, and also generates greater rents through corruption

Eswaran and Kotwal (1985): 2-tier markets in agrarian economies.

Attempts to explain the prevalence of permanent contracts in agrarian markets although casual workers are paid less and also for a shorter duration. There are certain unobservable tasks which require monitoring, such as pruning and irrigation. Landlords need people to carry out these tasks, but face the possibility of their employees shirking. They thus offer greater security and wages to provide an incentive for the workers to carry out their tasks, otherwise they will be fired and return to the casual labour market facing insecurity of employment. Explains unemployment and wages above the market rate. (Similar to Shapiro and Stiglitz’s efficiency wage theory, but theirs is applied to a more general setting).

Jayachandran (2005): Selling Labour Low

Shows that wage volatility is more acute in places where workers are poorer, less able to migrate and more credit-constrained, due to a more inelastic labour supply. Thus, they are at greater risk to productivity shocks such as drought, but is beneficial to landowners which see it as a form of insurance. Argues for alleviating credit constraints to the poor and greater linkages between rural markets, and reducing the amount of landless through land redistribution.

Munshi and Rosenzweig (2006): Caste, Gender and Schooling Choice

Study in Bombay to see how the caste system shapes career choices in the presence of globalization. Working-class males in the lower caste continue to be channelled into local language schools leading to traditional occupations, despite the increasing returns to nontraditional white-collar jobs, as it was much easy to be directed to these occupations through existing referrals. In contrast, lower caste girls who had lower labour participation rates did not benefit from such networks, and thus took advantage of the opportunities that became available from English schools.

Putnam (1993), (2000) Social Capital

Shows that in the US, associational activity (e.g. bowling clubs) have been decling since the 1950’s while the economy has grown at a rapid pace. So, is social capital essential for development? Otherwise, what substitues for it? And why are village economies poor when they have so much of it?

Yet, in Italian regions where associational activity is higher, government is more efficient and growth is faster

Arnott and Stiglitz (1991): Welfare Economics of Moral Hazard

Shows disadvantage of social capital. Insurance firms only offer partial insurance because of moral hazard, as with full insurance the insured will not take proper care to avoid whatever he is insured for. But with a dense network of family and friends, agents can top up the formal policy. Thus the insurer’s incentive mechanisms will fail as they are fully insured, and thus discourages pooling of risk.

Banerjee and Newman (1998): Information and the dual Economy

2 sectors: 1 modern, more productive, but less is known. The other is traditional but secure. It is harder to obtain credit to move to the first sector, and thus not everyone will move to the more productive sector, implying inefficiency. The move only takes place once sufficient people have gone over and passed down information. Similar to immigration studies showing that those with a sibling already emigrated more likely to migrate (Palloni et al 2001).

Lee and Brinton (1996)

Study of graduates of elite colleges in South Korea. Shows that institutional factors (which university you came from, social ties and introduction to firms provied by the university) more important than private social capital (family/friendship ties)

Besley (1995): Property Rights and Investment Incentives: Evidence from Ghana

Property rights theoretically should increase investment, by providing a source of collateral, by allowing one to sell the fruits of one’s labour and reduce the risk of it being taken away. The study shows that certain rights are more important, and that the major channel is the ability to claim land and the reduced risk of expropriation. However, notes the low marginal effect of increasing property rights, a marginal propensity to invest of 2-5%, and it may not even increase welfare.

Duflo (2001): Returns to Education, Evidence from Indonesia

Wants to resolve an age-old question: What are the effects to education? Examines effects of surge of school building during 1970’s oil boom in Indonesia. An additional school built per 1000 children raises education from 0.12 - 0.2 years on average. It has resulted in an increase in wages of 3.8%, and the estimated total returns to education are estimated at 6.4-9.1%. However, this is highly sensitive to the ability of the economy to absorb new workers, returns were especially high in Indonesia due to the high growth rate.

Duflo et al (2007): Peer Effects, Pupil-Teacher ratios and teacher-incentives, evidence from Kenya

Contract teachers are present more often than civil service teachers. Students of contract teachers performed 0.2 standard deviations better than those taught by civil service teachers. Smaller classes taught by civil service teachers showed no significant higher score, suggesting limited impact of student-teacher ratios. Students at both the low and high ends of the achievement spectrum benefited by being sorted into classes by initial achievement, performing 0.12 deviations better than the unsorted sample. Civil servants are more likely to be present in such classes where there is less differential in abilities, and also where school boards were empowered to monitor and punish them.

Morduch (1999): The Microfinance Promise

Criticially surveys the results of social banks and microfinance institutions. Social banks such as BancoSol in Bolivia, BRI and Kredit Desa in Indonesia generally lend to the “richer of the poor”, but are generally sustainable businesses, charging about 45% per annum. Microfinance institutions such as Grameen Bank charge between 12-16.6% per annum, but would need to charge 18-22% to cover costs without direct grants. To cover the subsidies on capital costs, Grameen needs to charge 32-45% per annum. However, every dollar lent seems to raise incomes by 17 cents to the dollar, lower than the 22 cent subsidy. However, the 5 cents on the dollar can be seen as a “cost of targeting, especially as they lend to the poorest of the poor. Members need to save 0.5% of the amount lent into an “emergency fund”, and 5% is a “group tax”, which is refunded upon successful repayment of the loan. Loans are disbursed primarily to groups of women (who have higher repayment rates), and if one member of the group defaults, lending to the rest of the group is ceased, thus encourages peer selection. Default rate is very low, at only 1.8%, and the bank is mostly owned by the poor it lends to, through a cooperative structure. Microfinance also cannot generate jobs on its own or solve the inherent seasonal nature of agriculture.

Burgess and Pande (2004) Can Rural Banks reduce poverty?

Attempts to estimate the impact of rural bank branches on poverty. Banks prefer to set up outlets in richer areas, with more clients, so determining the effect of financial institutions on rural poverty is complicated by the double causation. They isolated the causal effect of banks on poverty by examining trend breaks caused by legislation in India mandating the opening of 4 rural branches in unbanked rural areas for every bank opened in an already-banked location, and establish that they significantly reduced poverty. Opening a bank branch in a rural unbanked location reduces the poverty headcount by 4.74%

Udry (1991) Credit Markets in Northern Nigeria

Anthropological-style study of informal credit markets in Northern Nigeria. Villages in rural Northern Nigeria deal with the problem of incomplete information and enforcement in credit markets by taking advantage of the free-flow of information within villages and social enforcement mechanisms. In addition, the villager’s need for insurance is interlinked to credit, where the terms and frequency of repayment are adjusted to cater for adverse shocks, in accordance with syariah law. However, this is unable to insure against aggregate village risk, such as drought. This requires linkages across villages, to which only a select few Hausa traders have links to. Thus most credit is disbursed through them.

Burgess and Zhuang (2003): Modernisation and Son Preference

Examines intrahousehold allocation decisions to see if income is allocated evenly to all members of the households. Investigates gaps in welfare attainment between boys and girls by drawing on household expenditure from both a poor province (Sichuan), and a richer one (Jiangsu), and comparing with census outcomes. There is evidence of gender bias against girls in health spending in the poorer province, while evidence of bias in education spending against girls exists in both provinces. Split-sample analysis suggests that poorer, less diversified households exhibit stronger biases against girls, and suggests that son preference is not driven solely by cultural factors, pointing to a potential role for public policy. Rothbarth method is used to determine if adults give up adult good expenditure, such as cigarettes and alcohol, for their kids, however, this shows insignificant results, contradiciting census data. There could be forms of undetectable discrimination, such as girls having greater needs than boys when young, or critical interventions being made only for boys, and parents are more willing to go into debt to preserve an asset (a boy) but not for a liability (a girl). Also, evidence from India shows that younger female siblings tend to be the ones most discriminated against. Need to check as well that the lack of gender bias in Jiangsu is driven by better access to ultrasound technology and thus higher pre-birth abortion.

Banerjee, Gertler and Ghatak (2002): Tenancy Reform in West Bengal

Investigates the effect of tenancy reform through Operation Barga, launched in 1977 in West Bengal when the Left Front swept to power. This aimed to legalize a tenant’s rights through formal registration of their contracts, thus increasing the security of tenure over their land, by reducing the risk of random expropriation. Theory suggests that the effect should be ambiguous, as the increased investment the rights give could be counteracted by the lack of incentive that eviction threats used to elicit. However, the evidence shows that the Operation was responsible for an average of 28% of the increase in productivity in the period, or a 20% contribution to agricultural productivity. This suggests that the investment incentives outweigh the eviction incentives.

Lin (1992): Rural Reforms and Agricultural Growth in China

In a different institutional setting, this examines the effect the introduction of the Household Responsibility System (HRS) had on agricultural productivity in China. From 1978-1984, there was a phased abolition of the “danwei” system with large work teams of agriculture and a shift to giving households tenure over land for 15-year periods. This would improve incentives for households. In addition they reduced the quota of grain needed to be ceded to the central authority, increased market prices of output and allowed farmers to trade other products through different channels. HRS contributed up to 48% of the 42% jump in productivity between 1978-1984, when all households had converted to HRS. This was a one-time jump caused by the introduction of incentives.

Besley and Burgess (2000): Land Reform, Poverty and Growth

Another land reform study, again located in India. This examines the effect of different types of land reform. Tenancy reform: to improve a tenant’s rights and legal standing, Abolition of intermediaries: removing the Zamandari (feudal lords) from collecting a share of the surplus by intermediating between the landlord and the tenant, Implementing land ceilings to prevent the amassing of too much land and to ensure some land is for the landless, Land consolidation to stop the over-fragmentation of small plots. It finds that the first 2 reforms are responsible for increasing wages and output, while the other 2 are less significant, due to poor implementation, as well as the circumventing of legislation by holding of plots by family members.

Besley and Burgess (2004): Can Labor Regulation Hinder Economic Performance?

Yes it can! Pro-labour regulation, by increasing worker’s bargaining power, increases the likelihood of strikes and hold ups. This reduces efficiency and increases cost of production, decreasing productive activity in the region, and lowering manufacturing wages. Andhra Pradesh, with the most pro-employer legislation, grew at 6% per annum, 1.9% above trend, while West Bengal, the most pro-labour state, grew at -1.5% per annum after the Left Front gained power, compared to 2.2% trend growth.

Djankov et al (2002): The Regulation of Entry

Finds that greater regulation is associated with worse outcomes, even in terms of the measures it tries to regulate. Collects data on how long it takes to start a business in a given country, and the cost of doing so. Finds that more autocratic governments have higher regulatory costs, and worse social outcomes, in terms of pollution, health and safety standards. Consistent with the grabbing hand theory of regulation, where regulators extract rents from entrepreneurs by hindering start-ups, instead of the Pigouvian one where regulation solves market externalities. This effect is ameliorated by democratic constraints on power.

Djankov et al (2003): Who owns the media?

State ownership of media is associated with poorer quality governments and worse social outcomes. Collects data on ownership structures of international media companies (even including SPH, wow, now I know who owns it)

Besley and Burgess (2002): Political Economy of Government Responsiveness

Shows that increased newspaper circulation, especially in regional languages, increases the amount of public aid and food aid disbursed in the aftermath of a disaster. This effect is stronger the higher the turnout for elections is, and the more marginal the seat is. This is due to the media passing along information on politicians, encouraging opportunistic politicians to behave, while the vulnerable population can learn more about candidates other than simple ideology.