Devika Hazra

with Jose Aranzazu

Journal of Policy Modeling, Vol 44 (2) 2022

Abstract: As federal, state and local governments continue to allocate a significant share of their resources to law enforcement and correctional spending, concerns have risen that spending in education and welfare is declining. With fiscal pressure in the United States mounting, it is important to determine the effectiveness of public spending in deterring crime. This paper compares the effectiveness of the impact of government spending on welfare and education with that of law enforcement and correction on crime. Using panel data from 50 U.S. states over a time period of 1994-2014, results of linear regression with panel corrected standard errors as well as GMM estimation reveals that public welfare and education spending can potentially lower violent and property crime rates but law enforcement spending can only deter property crime. However, correctional spending can exacerbate both types of crimes. There is little to no evidence of the presence of crowding out of one category of spending by another. This results in the policy implication that more resources be allocated towards welfare and education programs.

with Patricia Gallagher

Economics Modelling, Volume 108, March 2022, 105768

Abstract: With wildfire-risk rising globally, the role of home insurance continues to remain under- studied in the search for mechanisms to mitigate loss from wildfire. This study investigates whether insurance policies effectively discourage homeownership in fire-prone zones. Using zip-code level data from Los Angeles between 2011-2018, we employ linear regression, GMM and machine-learning to examine the extent to which wildfire-risk impacts FAIR plan vis-a-vis market insurance. Results indicate that insurance policies do not effectively disincentivize homeowners away from high-risk zones which is novel to the existing literature. This is because the marginal impact of wildfire-risk on both insurance premiums, is relatively lower compared to other factors, especially for private insurance. Moderate-to-high risk areas continue to remain underinsured. Evidence of underinsurance was found in racially diverse neighborhoods. Recommendations include designing policies such that difference in risk-premium reflects difference in wildfire-risk to discourage homeownership in high-risk zones. [Media Mention: The Atlantic]​

International Journal of Social Economics, 2021, Vol. 48 Issue 12

Abstract: Using data from 2009 – 2016 across 31 states and union territories, this paper investigates determinants of juvenile delinquency in India as well as explores the nature of the complex relationship between economic variables and crime rate. The paper employs a panel corrected standard error model due to the presence of heteroskedasticity and contemporaneous correlation. Additionally, due to possible feedback effect from independent variables resulting in endogeneity, a two-step Generalized Method of Moments (GMM) is utilized to estimate a system of equations. Estimation results indicate that macroeconomic factors – GSDP per capita and adult unemployment rate – are significant in explaining the juvenile crime rate in India. Higher poverty rate and percentage of slums were found to increase juvenile crime. This paper also demonstrates the harmful effects that domestic violence has on juvenile delinquency. Finally, education has a deterring impact on crimes relating to juveniles but deterrence factors do not. While some implications are consistent with those found in previous studies of crime in developed and developing countries, the analysis in this paper also reveals unique results. For example, the adult unemployment rate was negatively correlated with juvenile crime, and an increase in police density exhibits a positive association with the juvenile crime rate. Further analysis of crimes by type (property and violent), reveals additional insights. In addition to that, contrary to hypothesis, by employing GMM estimation the paper finds no evidence of a negative impact of juvenile delinquency on economic growth.

International Journal of Social Economics, 2020, Volume 47 Issue 4

Abstract: This paper uses data from 2010–2016 across 32 states and union territories to investigate the determinants of crime in India. Results indicate the significance of demographic factor – population density and socioeconomic factors – poverty, income inequality and literacy rate – in accounting for crime in India. Among the macroeconomic factors, only GSDP per capita was found relevant. The study also reveals the importance of deterrence variables – chargesheeting rate, conviction rate, pendency in police as well as court cases – in explaining crime rates. Unlike the existing studies which use fixed effects or random effects, this paper utilizes a panel corrected standard errors model due to the presence of heteroskedasticity and contemporaneous correlation. While some implications are in line with those found in previous studies of crime in developed and developing countries, the effects of literacy rates and deterrence variables on crime are unique to the current analysis.

with Carol Cui

Business Forum, Volume 27, Issue 1, 2018, Pages 33-38 (2018)

Abstract: This study uses the 2007 and 2013 Annual Social and Economic Supplement of the Current Population Survey to examine the effect of the Affordable Care Act (ACA) on health insurance coverage among adults in the U.S. It finds that the ACA has improved coverage for men, youth, minorities, and low-income and less-educated individuals. However, those who are self-employed or do not work full-time have been negatively impacted.

with Carol Cui

Journal of Quantitative Economics, 16, pages 187-198 (2018)

Abstract: This paper uses data from 1991–2015 to examine the relationship between crime, GDP per capita, inflation, and unemployment rate in India. The Johansen cointegration test confirms the presence of cointegration relationship between the variables. The Toda-Yamamoto Granger causality test suggests that all the macroeconomic variables can significantly affect the crime level in India, and vice versa.​

The Quarterly Review of Economics and Finance, Volume 86, November 2022, Pages 65-86

Abstract: This paper examines whether the impact of monetary policy on welfare is altered when labor skill heterogeneity is incorporated in a limited participation framework. By endogenizing labor-leisure decisions and including labor heterogeneity, we find that not only is there a risky component to the income of financial market participants, but the wage income of non-participants is also subject to risk. This generates a diminished impact of monetary policy on welfare along with a stronger redistributive impact. The redistributive effect is further strengthened for higher levels of non-interest bearing savings among non-participants of financial market. In particular, expansionary monetary policy redistributes consumption and leisure from low to high skilled workers via an increased wage premium along with direct gains accrued from monetary transfers by high skilled agents who are connected to financial markets. Thus, welfare gains attributed to monetary expansion are far less when considering skill heterogeneity. Further, such gains are subject to the share of high skilled labor in the production of goods – increasing that share above a threshold renders a welfare-improving role for monetary tightening. To explore the implications of these results, we also analyze the impact of monetary policy in a crisis scenario.

Research in Economics, Volume 72, 2018, Issue 4, Pages 494-510

Abstract: The paper explores the redistributive effects of a monetary policy shock in a limited participation framework with limited credit access. Expansionary monetary policy redistributes consumption from traders to non-traders. This redistribution is the largest when only financial market participants have a choice between multiple means of payments while non-participants do not. Welfare analysis reveals that the effectiveness of monetary policy on the economy is the greatest when all agents (financial market participants and nonparticipants) can choose from alternative means of payment in a financially segmented model. The model is calibrated to the US economy for quantitative analysis.​ 

Journal of Economic Research, 23 (2017) 253-267

Abstract: Several advanced economies are heading towards a period of fiscal stress; aging population raises government transfers which in turn increases nominal government debt. Existing literature studies how alternative combinations of monetary and fiscal policies can stabilize real debt in the face of exponentially rising transfers. This paper develops an overlapping generations (OG) model to study the implications of endogenous labor supply on the path of real government debt under alternative policy regimes. This paper shows, theoretically, that switching to an alternative policy regime where monetary authority is passive in the face of rising transfers, even before the economy is at the fiscal limit, might stabilize inflation better than an active monetary regime. The model in the paper has been calibrated to the US economy to demonstrate dynamic effects.

The Quarterly Review of Economics and Finance, Volume 65, August 2017, Pages 378-387

Abstract: This paper captures the trade-offs between alternative payment instruments where each is associated with costs and benefits. Most models of cash-credit choice assume cash is a safe non interest-bearing asset and credit is interest-bearing but costly. Here, I consider the risk of loss from using cash resulting from theft and foregone interest earnings. I use a cash-in-advance model to analyze the channel through which monetary policy could have a positive impact on the economy by altering the incentives for cash-credit choice. The model indicates that although expansionary monetary policy increases total consumption, the resulting substitution toward credit might increase transactions cost, which may not result in improving welfare. The net effect depends on the change in transactions cost of using credit relative to the responsiveness of theft to inflation. The assumption of fixed cost of credit is crucial to these results. Calibration of the model to the US and Polish economy confirms the results.​.

A report prepared for the Centre for California Studies at Sacramento State University, November 2022

with E. Shiau

Abstract: Income and wealth inequality have been a persistent and growing problem in the state of California over the last 40 years. This report focuses on the role of economic recessions—primarily the 2008 Great Recession and the 2020 COVID-19 Recession—on income and wealth inequality with particular attention to their impacts on housing access and affordability. By reviewing the literature, this report outlines the relationship between economic recessions and inequality, how the unique characteristics of economic recessions lead to disparate impacts, and the particular impact of economic recessions on housing access and affordability. Through a literature review, this report also provides a discussion and assessment of federal and state policies designed to address the impact of economic recessions. It concludes by presenting empirical findings from original research that demonstrate the unique impact of economic recessions on housing access and affordability in California and Los Angeles.

Available at SSRN

Abstract: This paper documents the effects of employment protections for victims of domestic violence as established by California’s Senate Bill 400 implemented in January 2014 on the reporting of intimate partner violence. Results indicate that rates of intimate partner violence reports increased by 14 percent following the enactment of the law. This appears to be primarily driven by Black and Hispanic victims. Further, the increase is concentrated in tracts with higher unemployment rate and poverty rate as well as non-White and renter population, and those with below median income and education. Arrests for intimate partner violence also rose, driven by felony rather than misdemeanor cases, while 911 calls for domestic abuse remained unchanged – consistent with increased reporting rather than higher incidence. These findings suggest that strengthening job-retention rights for victims can lower the economic costs of disclosure, improve engagement with formal institutions, and reduce disparities in reporting across communities.

Available at SSRN, Under Review

Abstract: Violence against children (VAC) is a pervasive public health concern with profound and lasting consequences for physical, mental, and social well-being. While prior research has linked economic stressors such as unemployment to child maltreatment, less is known about the association between broader measures of economic development and VAC, particularly in low- and middle-income countries. This study examines the association between state-level GDP per capita and crimes against children across Indian states over time, incorporating police per capita as a secondary indicator of institutional capacity. Using longitudinal administrative crime data from, results of a linear regression with panel correction standard errors, indicate a nonlinear relationship between GDP per capita and crimes against children. Additionally, there is a significant association between policing capacity and reported crimes. By distinguishing between economic development and state capacity, this study contributes new longitudinal evidence from a large LMIC context and underscores the need for caution in interpreting official crime data when designing child protection policy.

Available at SSRN, Invited Book Chapter – Under Review

Abstract: This article critically examines youth delinquency in South Asia, arguing that criminological models rooted in the Global North offer an inadequate framework to capture its dynamics. While collective youth groups in South Asian lack the territorial, stable structures of archetypal street gangs, they exhibit pervasive and fluid youth networks whose activities, ranging from petty crime to organized violence, are structurally embedded within broader systems of political patronage, clientelism, and informal governance. Drawing on empirical analyses and regional case studies from India, Bangladesh, Nepal, Pakistan, and Sri Lanka, the study advances two central findings. First, youth delinquency in South Asia emerges as a rational response to socio-economic exclusion, rapid urbanization, and the absence of reliable welfare provision. Second, institutional responses involve the simultaneous cultivation of these processes for electoral mobilization and territorial control while publicly stigmatizing and prosecuting youth in conflict with the law. Juvenile delinquency is, therefore, a manifestation of institutional failure and uneven governance rather than a deviant subculture. These analyses urge a policy shift away from punitive enforcement and toward an integrated approach centered on prevention, education, and employment generation.

Available at SSRN

Abstract: This paper examines the impact of India’s Accredited Social Health Activists (ASHAs) on domestic violence (DV). Using a novel dataset that combines administrative data from the Ministry of Health and Family Welfare’s Health Management Information Systems (HMIS), National Crime Records Bureau, and incidence data from waves 3 and 4 of the National Family Health Survey (NFHS) we exploit spatial and temporal variation in ASHA deployment through a difference-in-differences design. Findings suggest that while spousal violence has declined over the past decade, it remains prevalent, with help-seeking behavior still low – particularly toward formal institutions. Our results indicate that the presence of ASHAs is associated with a decrease in actual incidence, demonstrating the importance of community-based interventions. These findings highlight the critical role of localized support systems in empowering women and addressing DV, especially in traditionally patriarchal contexts where formal help-seeking remains culturally constrained.

Available at SSRN

Abstract: This study investigates the impact of a liquidity shock created by demonetization in India in 2016 on domestic violence (DV). Results of interrupted time series analysis, using monthly data from 2011-2019, indicate an immediate short-term decline in calls related to DV reflecting disruptions in reporting and help-seeking. However, the post-demonetization trend increases significantly relative to the pre-policy trajectory, suggesting that financial stress associated with the liquidity shock may have intensified domestic violence over time. 

Available at SSRN, Under Review

Abstract: With crimes in India rising exponentially, particularly violence against women, several states have launched a centralized emergency response system, between 2013 and 2018, to reduce the time taken by law enforcement to respond to calls for service. Exploiting this staggered timing, I use a generalized difference-in-difference approach to estimate its impact on crimes against women in states and districts of India. Results indicate that overall, there is a significant negative impact – about 25% – on overall violence against women in the treated states. Specifically, domestic violence and sexual assault have decreased by 49% and 38%, respectively. This has also improved a secondary outcome – indictment for cases of gender-based violence. Consequently, suicides related to dowry or sexual abuse have also registered a decline. In contrast, there appears to be no such impact on non-gender violence, crimes against men or property crimes.

Available at SSRN, Under Review

Abstract: We examine the causal impact of political majoritarianism on gender-based violence in India by exploiting the Bharatiya Janata Party (BJP)’s rise to national political prominence in 2013 as an exogenous institutional shock. Using a differencein-differences framework and leveraging administrative data from 36 Indian states for 2006-2019, we find that BJP-majority states experienced a 47% increase in genderbased violence relative to non-BJP states after 2013, with no change in gender-neutral crimes. Robustness checks support a causal interpretation. Evidence on mechanisms from survey and mortality data shows rising domestic violence-related female suicides and declining indictment rates as well as rising acceptance of domestic violence in BJP-ruled states. These patterns point to weakening institutional deterrence and ideological shifts that normalize gendered subjugation. Our findings underscore the role of political ideology in shaping institutional behavior and gendered social norms, with broader implications for safeguarding women’s rights under rising populist and religious nationalist regimes.

Available at SSRN – Under Review

Abstract: Can a majoritarian political party being in power lead to an increase in crimes against minorities? I examine this question in the context of India where the right-wing nationalist Bharatiya Janata Party (BJP) won an unprecedented victory in the 2014 national elections and where caste-based crimes are a major concern. With states where BJP won majority votes in 2014 being the treatment group, I use a difference-in-difference research design and administrative crime data from 2006-2019 to estimate the impact of BJP’s electoral victory on violence against caste-based minorities. Results indicate that overall there is a significant increase in caste-based crimes – about 42% – in treated states compared to control states following the 2014 electoral win of BJP. Falsification tests, alternative specifications and additional robustness checks demonstrate the validity of the main results.

Available at SSRN

Abstract: In this paper, I study the impact of establishing a centralized emergency response system (CERS) in a developing country on atrocities towards historically disadvantaged minorities. To answer this question, I exploit the staggered implementation of CERS between 2013 and 2018 in several states in India to estimate its impact on caste-based crimes. Results of a generalized difference-in-difference approach indicate the presence of a significant reporting effect, that is, an increase in reporting of crimes by minority groups – about 38.56% – in the treated states. Thus, the study contributes to the literature that seeks to shed light on factors that could lead to improvement in reporting of crimes among minority groups.