Research

Publication

[SSCI & SCI]

1. Jungtaek Lee (forthcoming). Remote Work and Psychological Distress across Gender, Age, and Employment Sector. International Journal of Organization Theory & Behavior (With Jaehee Jong)

2. Jungtaek Lee (2024). Age-Specific Changes in the Effects of Social Connectedness and Loneliness on Depressive Symptoms: Evidence from the Korean Longitudinal Study of Ageing before and during the COVID-19 Pandemic. Psychiatric Investigation (SSCI) (With Juyeon Kim)

This study investigates the longitudinal effects of changes in the associations between two distinct aspects of social connections—i.e., social connectedness and loneliness—on depressive symptoms among community-dwelling Korean adults. This study also examines whether these associations vary across three age groups (45-64, 65-74, and 75 or above). Using data from the Korean Longitudinal Study of Ageing collected between 2014 and 2020 (N = 3,642 individuals), fixed effects models were used to examine the age-specific associations between the two distinct aspects of social connections and depressive symptoms (CES-D) while, accounting for time-invariant individual heterogeneity. Social connectedness is measured by asking the frequency of interactions with friends, relatives, or neighbors. The findings indicate that the impact of loneliness on depressive symptoms outweighs that of social disconnectedness. Notably, this study unveils age-specific patterns concerning the impact of the COVID-19 pandemic on depressive symptoms and changes in the association between loneliness and depressive symptoms. Specifically, middle-aged individuals reported higher levels of depressive symptoms and loneliness along with a heightened impact of loneliness on depressive symptoms, despite maintaining stable social connections. Conversely, the oldest adults experienced reductions in both depressive symptoms and loneliness, despite a significant decrease in socializing. These findings shed light on the differential effects of loneliness on depressive symptoms within distinct age groups before and during the pandemic. The implications of these findings are discussed with a focus on informing the development of targeted policy interventions tailored to the specific needs of different age groups.

3.  Jungtaek Lee (2023). Effects of private health insurance on healthcare services during the MERS Pandemic: Evidence from Korea. Heliyon, Volume 9, Issue 12, e22241 (SCI)

This study investigates how private health insurance impacted healthcare services during the MERS pandemic in Korea. Using the Korea Health Panel Study (KHPS), this study examines the difference in healthcare utilization between insured and uninsured individuals during the pandemic. If insured individuals use fewer healthcare services than the uninsured during the MERS pandemic, it could be evidence of moral hazard. During the MERS outbreak, the probability of outpatient medical services utilization was lower by 19 % than during non-pandemic periods. All individuals decreased the number of outpatient visits by 7 %. Insured individuals reduced outpatient visits more than the uninsured in response to the MERS pandemic. The increased outpatient utilization by private health insurance could be attributed to both moral hazard and adverse selection. However, given that people with poor health cannot enroll in private health insurance due to the insurance company’s screening process, moral hazard leads to increase healthcare utilization rather than adverse selection. 

4. Jungtaek Lee (2023). The effect of vaccination on social isolation during the COVID-19 Pandemic. Applied Economic Letters, Volume 30, Issue 15, pp.1987-1994 (SSCI)

The objective of this study is to investigate the effect of vaccination on social isolation and loneliness. Data from 6,049 individuals were collected from the Understanding America Study (UAS). I use the fixed effect model with the panel data to control time-constant unobserved characteristics which affect vaccination status as well as social isolation. Vaccination was related to about 2 percentage points increase in the proportion of those who met family or friends in person. Days in the past week interacting with family or friends increased. Vaccination was also associated with about 1 percentage point decrease in the proportion of those who felt lonely for more than one day in the past week. The degree of loneliness measured by how often respondents felt lonely in the past week decreased by 2 percentage points for the vaccinated respondents. I found that vaccination decreased subjective social isolation (loneliness) and objective social isolation (interaction with people in person). I also consider the dynamic effect of vaccination status on social isolation. Results suggest that the effect of vaccination became bigger with increasing duration of vaccination. 

5.  Jungtaek Lee, Juyeon Kim (2023). Social connectedness and cognitive function before and during COVID-19: A longitudinal study of Korean older adults with an Instrumental Variable Regression. Psychiatric Investigation, Volume 20, Issue 4, pp. 325-333 (SSCI)

We estimate the causal effect of social connectedness (i.e., the frequencies of meeting with friends, relatives, or neighbors) on cognitive function (the Korean version of Mini-Mental State Exam) among Korean older adults. We used longitudinal panel data collected before and during the coronavirus disease-2019 (COVID-19) to set up the fixed (FE) or random effect (RE) models. To overcome omitted variable bias or reverse causality, we used COVID-19 pandemic period as an instrumental variable to estimate the causal effect of social connectedness on cognitive function. Social distancing measures during the COVID-19 period decreased social interaction. The results showed that an increase in the frequency of social interaction led an increase in cognitive scores. Specifically, an increase of one unit in the frequency of meeting familiar people increased cognitive scores by 0.1470 and 0.5035 in the RE and FE models, respectively. Social distancing policies due to the global pandemic may have increased the risk of social isolation and cognitive decline among older adults. The government and local communities need to increase their effort to develop way to connect adults through the remainder of the pandemic and beyond. 

6. Jungtaek Lee, Juyeon Kim (2020).  The role of health insurance in mental health care for young adults. Applied Economics, Volume 52, Issue42, pp.4577-4593 (SSCI)

This study investigates the effect of health insurance on young adults’ utilization of mental health care, by exploiting the dependent coverage expansion that was an early provision of the Affordable Care Act. Using the National Health Interview Survey (NHIS) from 2011 to 2013, we utilize the regression discontinuity design to overcome the problem of endogeneity. There are two key findings. First, health insurance has increased young adults’ mental health care. Second, young adults’ physical health care has not been affected by health insurance. The results suggest that the use of mental health care might be more responsive to changes in health insurance coverage than the use of physical health care is. Furthermore, the effect of health insurance on mental health care utilization is heterogeneous across metal health statuses. 

7.  Jungtaek Lee (2018). Effects of health insurance coverage on risky behaviors. Health Economics, Volume 27, Issue 4, pp.762-777 (SSCI)

Prior to implementation of the Patient Protection and Affordable Care Act, dependent health insurance coverage was typically available only for young adults under the age of 19. As of September 2010, the Affordable Care Act extended dependent health insurance coverage to include young adults up to the age of 26. I use the National Health Interview Survey for the sample period from 2011 to 2013 to analyze the causal relationship between the expansion of dependent coverage and risky behaviors including smoking and drinking as well as preventive care. I employ a regression discontinuity design to estimate the causal effect of health insurance coverage and overcome the endogeneity problem between insurance status and risky behaviors. When young adults become 26 years old, they are 7 to 10 percentage points more likely to lose health insurance than young adults under the age of 26. Although young adults over the age of 26 are generally aged out of insurance coverage, presence or absence of health insurance does not affect their smoking and drinking behaviors and their access to preventive care.

8.  Yoruk, Baris and Lee, Jungtaek (2018). Does legalization of Sunday alcohol sales increase crime? Evidence from seven states. Journal of Studies on Alcohol and Drugs. Volume 76, Issue 6, pp.816-825 (SSCI)

Recently, several states repealed their laws restricting the sale of alcohol on Sundays. We investigate the effect of this policy change on crime trends in seven states using data from FBI’s National Incident-Based Reporting System (NIBRS).We identify the impact of the legalization of Sunday alcohol sales on several different types of criminal activity by exploiting the variation in the implementation of this policy change across different states at different times. Using difference-in-differences type models, we show that the repeal of the ban on Sunday alcohol sales is associated with significant increases in total property and violent crimes committed on Sundays. In particular, we find that states that legalized Sunday sales of alcohol experienced 13% to 20% increase in the total number of violent and property crimes committed on Sundays. However, the aggregate impact of this policy change on crimes committed on all days of the week is not significant due to either positive or statistically insignificant spillover effects of the repeal of Sunday alcohol sales bans on crimes committed on Mondays through Saturdays. These results are robust under alternative model specifications and several falsification tests.

[KCI]

1. Jungtaek Lee (2023). Health Insurance is healthy: An effect of obtaining Medicare on self-rated health. Journal of Insurance and Finance, Volume 34, Issue 1, pp. 65-99

I examine the causal effect of health insurance on self-rated health in the short run, based on a concept of the fuzzy regression discontinuity(RD) research design with the age of 65 as an instrumental variable. To focus on the causal effect of obtaining Medicare, I limit the sample to those without private insurance around the age of 65 - the Medicare eligibility starting age. First, the result shows that the proportion of those covered by health insurance increases by about 27 percentage points at that age. Second, the self-rated health of those newly covered by health insurance at the age of 65 improves. Third, although there is no discrete change in health care services, including surgery and inpatient service, elderly individuals increase their use of preventive care services. These results suggest that obtaining health insurance strengthens the assurance about getting medical and preventive care services and results in better self-rated health. 

2.  Jungtaek Lee (2023). The Impact of Social Factors on Adherenece to Mask-Wearing during the COVID-19 outbreak. The Journal of Women and Economics, Volume 20, Issue 1, pp.61-86

This study provides new evidence on the relationship between health insurance status and mask-wearing during the COVID-19 pandemic. This study also considers different response to state mask mandating by health insurance status, gender, and race. The Understanding America Study (UAS) panel data is used to set up fixed or random effect models. The existence of health insurance actually increases the probability of mask-wearing by 3.9 to 4.8 percentage points. Individuals who identified as male, white or uninsured were less likely to wear masks before statewide mask mandating. However, the mask-wearing rate of these groups increased more than for females, other racial groups, and insured individuals after statewide mask mandating. Statewide mask mandating is effective for those who have not worn masks before the mandating. 

3. Jungtaek Lee, Juyeon Kim (2022). The Effect of Health Insurance on Self-Rated Health: Evidence from Older Adults with Functional Limitation.  The Korean Journal of Health Economics and Policy, Volume 28, Issue 4, pp.123-156

This study examines the health effect of health insurance, based on the disparity in self-rated health between those with and those without any daily-life limitation. We use the National Health Interview Survey to construct the sample of older adults aged between 60 to 69 years with the period between 2004 and 2013. To address omitted variable bias caused by unobserved characteristics, we use regression discontinuity design to estimate the causal effect of health insurance on self-rated health, which exploits the Medicare eligibility age as an instrumental variable. We find that when older adults with functional limitation reached the age of 65, their self-reported health improved more than those without any functional limitation. Improvements in self-rated health are immediately implemented at the age of 65 for those who have any functional limitation. Moreover, health care or preventive care services cannot be attributed to better health for older adults with functional limitation in the short run. Results imply that being able to use health care or preventive care services increases perception of well-being. Health insurance can play an important role in improving assurance in health-managing ability, which in turn influences self-rated health.

4. Jungtaek Lee (2022). The Effect of an Increase in Child Care Needs on Women’s Labor Market Outcomes during the COVID-19 Pandemic. The Korean Journal of Health Economics and Policy, Volume , Issue , pp.  

I use data collected in the Understanding Coronavirus in America (UAS) to consider the gender effect of an increase in child care needs on labor market outcomes during the COVID-19 pandemic. First, I find that this pandemic leads to an increase in child care needs, which in turn provides more burdens on mothers than fathers. Mothers are more likely to lose their job or exit out of the labor market during the pandemic due to the increase in child care needs. Second, coupled women with children are less likely to be employed than single mothers. Because there is a male earner in the family, a decision to quit job is likely to less financially severe than single mothers. As a result, a dramatic increase in child care caused by the COVID-19 pandemic may lead to be out of labor market for coupled mothers.

5. Jungtaek Lee (2022). The Causal Effect of Depression on Cognitive Funciton. The Korean Journal of Health Economics and Policy, Volume 28, Issue 1, pp. 64-86

The objective of this study is to investigate the causal effect of depression on cognitive function. However, estimating the causal effect is empirically challenging. There exists confounding factors which affect both depression and cognitive function (Omitted Variable Bias). Moreover there is a possibility that cognitive function also influences depressive symptoms (Reverse Causality). To overcome the endogeneity caused by reverse causality or omitted variable bias, I used KLOSA( Korean Longitudinal Study of Ageing) to set up fixed effect model with instrumental variable (spousal death). I found that depression has an negative effect on poor cognitive function. An unit increase in CES-D10 score which measures respondent’s depressive symptoms decreases MMSE(mini-mental state examination) scoe by 0.6081.

6. Jungtaek Lee, Taejun Lim (2021). Macroeconomic Impacts of the Reform of National Basic Livelihood Security System. Journal of Insurance and Finance, Volume 32, Issue 3, pp. 95-122

Since introduced in 2000, National Basic Livelihood Security System (NBLSS) has expanded the scope of beneficiaries and increased the level of allowance payment. However, the recent spread of the poor and the acceleration of income polarization has lead to discussions on another system reform. In this study, we propose a Dynamic Stochastic General Equilibrium model to quantify the macroeconomic effects of NBLSS reform in the forms of (i) the relaxed beneficiary selection criteria and (ii) the increased allowance payment per household. We find that the NBLSS reform reduces households’ incentive to work and save, which results in a decrease in aggregate labor and capital. Despite a decrease in aggregate output, the level of social welfare is increased. Importantly, relaxing the beneficiary selection criteria turns out to be more effective in enhancing social welfare than increasing the allowance payment although it brings down the production level more severely.

7.  Kim, Juyeon and Lee, Jungtaek (2018). Living Arrangements and Cognitive Function among Korean Older Adults. Survey Research, Volume 19, Issue 1, pp.97-127

Social relationships are hypothesized to protect against cognitive decline through their association with intellectual stimulation and with reduced stress-related physiological reactivity. Living arrangement is one of social relationships that can be a source of cognitive stimuli and/or a basis for emotional support. However, research on the relationship between living arrangements and cognitive functioning has received little attention. This study distinguishes four living arrangements according to living with a spouse and with children and others, and examines the association between living arrangements and cognitive function using data from 2008 2014 Korean Longitudinal Study of Ageing (KLoSA). The study finds that older adults living with children and others show significantly lower cognitive function than those living with a spouse only but living alone do not show lower cognitive function. However, the rate of decline in cognitive function is significantly larger for those living alone than those living with a spouse only. These findings show that living with a spouse and/or children are associated with cognition, and suggest the need for a prevention plan for those living alone facing a rapid cognitive decline.

8.   Lee, Jungtaek and Juyeon Kim (2018). The effect of Retirement on Health Behaviors, Mental Health and Subjective Health Status. Public Health and Social Science, Volume 47, Issue 1, pp.5-29

Social relationships are hypothesized to protect against cognitive decline through their association with intellectual stimulation and with reduced stress-related physiological reactivity. Living arrangement is one of social relationships that can be a source of cognitive stimuli and/or a basis for emotional support. However, research on the relationship between living arrangements and cognitive functioning has received little attention. This study distinguishes four living arrangements according to living with a spouse and with children and others, and examines the association between living arrangements and cognitive function using data from 2008 2014 Korean Longitudinal Study of Ageing (KLoSA). The study finds that older adults living with children and others show significantly lower cognitive function than those living with a spouse only but living alone do not show lower cognitive function. However, the rate of decline in cognitive function is significantly larger for those living alone than those living with a spouse only. These findings show that living with a spouse and/or children are associated with cognition, and suggest the need for a prevention plan for those living alone facing a rapid cognitive decline.

Working Papers

1. Lee. Jungtaek, Medical Marijuana Laws and Crime. 2019. 

Using Uniform Crime Report (UCR) data, I first show that arrests for illegal marijuana possession increase in states, where a medical marijuana law (MML) has been enacted, compared to states without a MML. Using difference-in-differences method, I find that the MML leads to an increase in the number of arrests for illegal marijuana possession by 6-10 percent. Second, possession arrests for hard drugs, such as "cocaine, opium, and other derivatives," do not increase due to the MML, as opposed to what a gateway hypothesis expects. Moreover, using detailed information on arrests for hard drugs possession from the National Incident-Based Reporting System (NIBRS), I find that arrests for use of crack cocaine has decreased sharply after the introduction of the MML. However, I find no significant relationship between the MMLand arrests for use of other hard drugs. Third, while driving under influence (DUI) arrests are negatively affected by the MML, other violent and property crimes are not significantly affected by the policy change. I find that the implementation of the MML leads to increased illegal marijuana possession arrests, but externalities related to the implementation of the MML, such as consumption of hard drugs or the incidence of non-drug related crimes, are relatively limited.

On going Projects

1. Do the Minimum Legal Drinking and Smoking Age Laws Save Babies? 

Using the National Vital Statistics Natality Birth Data, I estimate the impacts of smoking and drinking during pregnancy on infant birth outcomes. However, there might be some risky factors that affect both smoking (or drinking) behavior and infant birth outcomes. To deal with this concern, I exploit the minimum legal tobacco purchasing age (MLTPA) and the minimum legal drinking age (MLDA) as instruments for smoking and drinking status of young pregnant women around the legal cutoff ages for smoking and drinking. My research design is based on a simple idea: to the extent that MLDA and MLTPA laws have been enforced, age cutoffs for drinking and purchasing cigarette yields different probabilities to access alcohol and cigarette for young pregnant women on either side of legal age cutoffs. Hence, I can identify the effect of MLDA and MLTPA laws on infant birth outcomes using the discontinuous jump in drinking and smoking during pregnancy at cutoff ages, caused by MLDA and MLTPA laws.

2. Violence and Infant Health: The Effect of Crime on Infant Birth Outcomes.

This study examines how maternal exposure to local violent or property crime affects infant birth outcomes. The research design is based on a simple idea: pregnant women who live in communities with high crime rates are more likely to have babies with poor health. However, those who reside in dangerous neighborhood are also more likely to be poor and minority. That might lead to poor birth outcomes and there might exist unobserved factors that affect both infant birth outcomes and local crime rate. To address this concern, I exploit the number of police officers at the county-level as an IV for county-level crime rates. I collect the information on individual birth outcomes, county-level crime rates, and the number of police officers at police agency from the National Natality Birth Data, Uniform Crime Report (UCR), and Law Enforcement Management and Administrative Statistics (LEMAS). Using these data, I can identify the causal effect of crime rates on infant birth outcomes by using the variation on crime rates across counties and time.

3. The Effect OF Rankings on Hospital Choices

Using hospital rankings released by US News and World Report, this study estimates the effect of the U.S. News hospital rankings on both patient volume and hospital revenues. Contrary to other markets, consumers of health care or hospital services are unresponsive to change in hospital quality, because of some restrictions to use health services, such as distance from home, health plan network, and doctor referrals. Therefore, whether rankings have a significant effect on consumer decision of hospitals is an empirical question. I use a regression discontinuity design to understand the effect of the rankings relative to other important factors of hospital choice such as distance to hospital.