Public Economics Workshop
| Centre for Development Economics
Delhi School of Economics
University of Delhi
| Centre for research on the Economics of
Climate, Food, Energy, and Environment
Indian Statistical Institute, Delhi
AmEx Room (# 206)
Department of Economics, Delhi School of Economics
Inspired by fertilizer use, we investigate behavior in settings where externalities arise from individual activities crossing an uncertain individual threshold. We show within an analytic model that only threshold uncertainty causes individual and social interests to diverge with individuals accepting a larger probability of crossing the threshold than is socially optimal. Risk-aversion is predicted to have a non-monotonic impact in the willingness-to-invest in Information systems. We then theoretically investigate the impact of information on thresholds as well as taxing the caused externalities or fertilizer use. We test our predictions within an experiment with farmers as well as with students. We find that, even without any regulation in place, farmers invest significantly less than their payoff maximizing level, but on average close to the social optimum. The probability of exceeding the individual threshold and the resulting damages significantly decrease when subjects are provided free informative signals or have to pay for caused damages. In contrast, taxing inputs (fertilizer use) does not decrease the risk of causing external effects.
We construct an integrated assessment model with multiple energy sources- including fossil fuels and “green energy” and multiple world regions. The energy sources are imperfect substitutes and their production involve structures that are endogenous. In particular, firms can decide to lower the marginal cost of producing one form of energy at the expense of the marginal costs of other energy sources: there is directed technical change. In the lowering of these marginal costs, there are also spillovers, which are international. We analyze how (potentially region-specific) taxes affect output and the climate with and without the endogeneity of technology. We emphasize the second-best nature of taxation when optimal world-wide technology subsidies are not implemented.
What are the long-run implications of coercive colonial institutions on female empowerment? In this paper, we study the short- and long-term effects of a colonial experiment - the Dutch Cultivation System on Java (1830-1870) - that forced the native population to switch from subsistence rice production to sugar and coffee cultivation for exports. First, we show that the cultivation of sugar, but not the cultivation of coffee, is associated with stronger historical women empowerment. An IV strategy, exploiting the suitability of districts to sugar production, provides an even larger (albeit imprecisely estimated) effect. Second, on the basis of the existing literature, we hypothesize that this institutional shock might have affected current female empowerment through: i) An increase in female labor demand (driven by an increase in infrastructure and the emergence of a modern sector). ii) An increase in female labor supply (driven by an decrease in their bargaining power within the household). Consistent with the labor demand channel, we show that sugar and coffee cultivation is associated with greater density of roads and railways, which, in turn, are associated with a several measures of female empowerment. However, contrary to the predictions of the labor supply channel, we show that sugar cultivation, while having a negative effect on the historical male-to-female ratio, does not affect current female empowerment indicators.
The delivery of basic health products and services remains abysmal in many parts of the world where child mortality is high. This paper shows the results from a large-scale randomized evaluation of a novel “social entrepreneurship” approach to health care delivery. In randomly selected villages a sales agent was locally recruited and incentivized to conduct home visits, educate households on essential health behaviors, provide medical advice and referrals, and sell preventive and curative health products. Results after three years show substantial health impact: under-5 child mortality was reduced by 27% at an estimated average cost of $6871 per life-year saved.
Behavioral responses to taxation affect the trade-off society faces between implementing equality and efficiency. Several influential theoretical papers have used heterogeneity in beliefs about behavioral responses to taxation to explain variation in people’s support for redistribution of income in society. In this study, we use a purposefully simple task to elicit incentivized beliefs from a representative sample of over 9,000 Americans about how taxes affect people’s effort choices. We also elicit incentivized equality-efficiency preferences. We find that beliefs about behavioral responses to taxation fail to explain variation in people’s support for redistribution of income in society. We also find that Democrats and Republicans have virtually identical beliefs about behavioral responses to taxation. By contrast, equality-efficiency preferences strongly predict both people’s support for redistribution of income in society and their political affiliation. In a follow-up experiment, we explore the role of motivated beliefs and group identity by priming the respondents about the political debate on taxation and party views on behavioral responses to taxation.
This paper examines how voluntary contributions to a public good are affected by the contributors’ heterogeneity in beliefs about the uncertain impact of their contributions. It assumes that contributors have Savagian preferences that are represented by a two-state-dependent expected utility function and different beliefs about the benefit that will result from the sum of their contributions. Under some conditions imposed on preferences, we establish general comparative static results on the effect of specific changes in the distribution of beliefs on the (Nash) equilibrium provision of public good. We specifically shows that the equilibrium public good provision is increasing with respect to first-order and second order stochastic dominance changes in the distribution of beliefs.
Accurately measuring the consumers’ willingness-to-pay (WTP) for improved attributes in a newly- developed product is important for the developers and marketers of the product. This paper reports and provides an explanation for a discrepancy between two theoretically equivalent, frequently used, and incentive compatible methods of measuring such premia: the full bidding and endow-and-upgrade meth- ods. We found the following reverse endowment effect in a WTP elicitation Becker-Degroot-Marschak (BDM) experiment for the newly-developed biofortified High Iron Pearl Millet (HIPM) conducted in ru- ral India. The WTP for exchanging local pearl millet (LPM) for HIPM (the endow-and-upgrade measure of premium for HIPM over LPM), was significantly greater than the difference between the WTPs for HIPM and LPM (the theoretically equivalent full bidding measure). Our explanation is an expectations- based reference dependence model of loss aversion for the new product. We identify and structurally estimate the loss aversion parameter, thus obtaining latent values of the premium for HIPM over LPM, the WTP for HIPM, and the effect of nutritional information on the WTP for HIPM. We find that the accuracy of both methods depends on the location of true values with respect to the BDM parameters. In our study, the full bidding measures were downward biased.
Proponents of group-based preferential treatment argue that such policies help equalize life chances for members of historically disadvantaged groups. Opponents argue that they violate meritocratic ideals and worsen economic performance. A better understanding of how institutions allocate positions across individuals and groups helps clarify what is at stake. We study this process of allocation using a model in which individuals differ along two dimensions, ability and training, both of which matter for performance. Training also depends on ability and additionally on resources, such as school and neighborhood quality. One group is disadvantaged in that a smaller fraction of its members have access to superior resources. Training is observable but ability is not. We call an allocation psuedomeritocratic if selection is based on a single training threshold common to the two groups. It is group egalitarian if the same share of each group is selected. We show that performance-maximizing allocations can take a variety of different forms. They are generally not psuedomeritocratic and often favor the disadvantaged group. Moreover, in the absence of legal restrictions, selection can be non-monotonic even within groups, meaning that individuals with low and high training levels may be hired while those in between are excluded. We see the main contribution of our paper as clarifying the concept of meritocracy and questioning policies based on its common interpretations.
We study optimal labor income taxation when the objective of the planner combines the ethical values of Pareto efficiency, responsibility for one's preferences and poverty reduction. We capture the objective of poverty reduction by imposing that a transfer from an agent above the poverty line to another agent below the poverty line is a social improvement. We show that there are ways of defining such a requirement and make it compatible with Pareto efficiency and preferences responsibility. We characterize the way any social preferences satisfying the three requirements and separability should rank allocations in which some agents consume bundles in the neighborhood of the poverty line. Then, assuming that agents have quasi-linear preferences, and that there is only one dimension of behavioral heterogeneity, we derive the formula of the optimal labor income tax scheme and we calibrate the formula to the U.S. economy.
We study the impact on income poverty of the unequal growth experienced in the US over 1989-2013. The conventional wisdom is that unequal growth reduces absolute poverty but increases relative poverty. Our normative evaluation is based on a family of poverty measures that accounts for both the absolute and relative aspects of income poverty. These measures always implicitly consider that being absolutely poor is worse than being relatively poor. Crucially, this family of measures is defined by a unique parameter quantifying the weight given to each of these two aspects. We first develop simple theoretical conditions under which the poverty evaluation is entirely robust to the value taken by the weight parameter. For non-robust cases, we derive a formula returning the threshold parameter value at which the poverty evaluation is reversed. We then apply our measure to study the evolution of poverty in the US over the recent period of unequal growth. Our empirical results highlight two main points. First, including the relative aspects of poverty changes our view on the evolution of poverty in the US for a very large interval of values of the weight parameter. Therefore, even a mildly relativist view is often sufficient to disagree with the conclusion obtained by the official absolute poverty measure. Second, this does not prevent that our poverty evaluation robustly agrees with the official measure in a non-negligible 20% of cases featuring a dispute between absolute measures and relative measures. Overall, the results allow us to conclude that the unequal growth in the US has been more unequal than growing for the most part. Interestingly, poverty judgments over this period are largely robust to the choice of the weight parameter.
This paper documents that growth in the extensive margin is on average lower in the agricultural sector than in other activities. We introduce this new fact into a simple model of trade to show its relevance for regions specialized in the lagging sector. Diversity-loving consumers endogenously reduce the share of their expenditure devoted to that sector. The region specialized in it receives a decreasing share of world income, which results in diverging income and welfare trajectories with respect to the rest of the world. Appropriating a decreasing share of world value pushes downward the relative wage of the agricultural region and lowers the price of its exports relative to that of its imports, resulting in terms of trade deterioration. This result, supported by empirical evidence, separates our theoretical results from those obtained in a similar model of uneven output growth between sectors. We present empirical evidence for the main testable results of the model. Our model is the first replicating these facts without the need of heterogeneous consumers or products, nor resorting to political or institutional explanations.
This paper examines the effects of extreme temperatures on mortality rate using random year-to-year variation in temperature based on a county-level panel in China. We find a robust U-shaped relationship between temperature and mortality rates, indicating that extreme cold or hot temperature will lead to excess deaths, but the effect is two and a half times larger than previous findings that use U.S. data. The effects are especially large for the old, mainly due to excess deaths caused by cardiovascular diseases. We also find evidence that residents living in cold counties are less vulnerable to cold days, but are more vulnerable to hot days. We then explore households’ adaptation behaviors to extreme temperatures. Interestingly, we find that although urban households increase energy consumption when exposed to cold temperatures and purchase more air conditioners both on hot or cold days, rural households are irresponsive to temperature fluctuations, implying that rural people may suffer more from extreme temperatures.
State owned forest, especially in developing countries, suffer from lack of funds to efficiently manage the forest. In particular, this results in unsustainable rate of deforestation. As a result, decentralization of ownership to local institutions (devolution), for example forest user groups, has been suggested. We propose and experimentally test for the relevance of voluntary participation in the form of opt-out option as a policy tool that could potentially trigger the establishment of cooperative institutions after devolution of forest rights to community members. The basic idea is that community members choose between joining a (devolved) forest user group or an opt-out option. The opt-out option provides some benefits to those who choose it, but potential benefit is higher if one joins a forest user group, provided that there is cooperation among the users. By framing the social dilemma game after devolution as a multiplayer prisoner’s dilemma game, we hypothesize that such an opt-out option could serve as a coordination tool in the presence of social preferences and forward induction. We conduct a lab-in-the field experiment in Ethiopia to test this. Our results show that, not only do most people reject opt-out options that provide a significant amount of money, the level of cooperation is higher among those who reject opt-out options than those who were not offered an opt-out option. Opt-out options can therefore serve as a cheap policy tool to trigger cooperation among forest user groups after devolution reforms.
Missing female patients: An analysis of gender discrimination among tertiary care patients in India, 2016 Gender discrimination in access to healthcare has not been systematically studied in India and other developing countries. This is primarily due to non-availability of regular and reliable data. In this paper we use big data on clinical appointments from a large, public funded, tertiary care hospital with a robust hospital information system, to study the level and extent of gender discrimination in access to healthcare. In particular, we use data on clinical appointments of 2,377,028 outpatients to analyse the likelihood that a female patient would be brought to the hospital vis-à-vis a male patient and relate this to distance from the hospital and the age of the patient.
Bishnu, Monisankar: Revisiting the Annuity Role of Estate Taxes
Previous conclusions that a uniform lump-sum estate tax could implicitly provide annuity income were reached by not including the bequest income that household receives. However, where agents leave behind bequest, they should also receive bequest from their parents. The differential timing and sizes of bequest income will generate unequal wealth effects even with actuarially fair annuity markets. In order to restore the first best, the government has to adopt an estate tax regime that is no longer uniform. Thus once the bequest income is determined with an uniform estate tax, it no longer bears the annuity role.
Coal provided 76% of Indian electricity generation of 1240 Terawatt hours in 2016-17. We compare the social cost of power from coal with power from renewables such as solar photovoltaics and wind. Our lower bound estimate for the domestic environmental and health cost of coal is 1.58 Rs/KWh (0.024 USD/KWh) in 2018. The cost of fuel alone for the cheapest coal generator in India will be more expensive than new solar and wind generators in 2018 if these costs are taken into account. But most coal-fired capacity has a lower marginal cost than solar PV and wind if external costs are not priced, so unless there is policy action to replace coal, it will not happen. Since there is currently more capacity than is needed to meet peak demand, no new coal plants over and above those already planned or under construction will be required until 2026. We argue that by that time, the requirement for new peak capacity will be more economically met by a combination of other sources including concentrated solar power with storage. This implies that the most economical way forward for India’s electricity system is to monotonically reduce its dependence on coal.
This paper deals with tax policy responses to inequality aversion by examining the first-best Pareto-efficient marginal tax structure when people are inequality averse. In doing so, we distinguish between four different and widely used models of inequality aversion. The results show that empirically and experimentally quantified degrees of inequality aversion have potentially very strong implications for Pareto-efficient marginal income taxation. It also turns out that the exact type of inequality aversion (self-centered vs. non-self-centered), and the measures of inequality used, matter a great deal. For example, based on simulation results mimicking the disposable income distribution in the US in 2013, the preferences suggested by Fehr and Schmidt (1999) imply monotonically increasing marginal income taxes, with large negative marginal tax rates for low-income individuals and large positive marginal tax rates for high-income individuals. In contrast, the often considered similar model by Bolton and Ockenfels (2000) implies close to zero marginal income tax rates for all.
Using data from a unique retrospective primary survey in the city of Delhi in 2014-15, this paper examines the role of class and caste in patterns of social mobility. The sample consists of respondents who have finished higher secondary schooling, and thus are eligible to enter higher education. Examining their outcomes 11-12 years after they completed high school, the paper finds that despite the fairly similar starting positions, the educational and occupational trajectories of these individuals have diverged considerably. The paper attempts to disentangle the role of caste and class – usually referred to as “family background” -- in explaining these trajectories. While both are important, caste continues to impact upward mobility even after class has been controlled for, with significant differences between poor, but socially higher ranked Brahmins, and poor and socially lower-ranked Dalits. Probing deeper into pathways that explain later life educational and labour market outcomes, we find that caste inequality manifests itself via final school-leaving examination performance (Class XII scores), which shape future educational attainment. The evidence shows that performance in high-school is critical for later life chances, but that labour market returns to educational characteristics vary even after controlling for a host of factors, including Class XII scores, indicating fresh additional labour market discrimination which could account for caste differences in social mobility.