Time-invariant Individual Characteristics Research Articles

Retirement, housing mobility, downsizing and neighbourhood quality - A causal investigation

This paper provides the first causal evidence on the impact of retirement on housing choices. Our empirical strategy exploits the discontinuity in the eligibility ages for state pension as an instrument for the endogenous retirement decision and controls for time-invariant individual characteristics. The results show that retirement leads to a statistically significant and sizable increase in the probability of making a residential move or the likelihood of becoming outright homeowners. We also find that individuals downsize both physically and financially and tend to move to better neighbourhoods or closer to the coast upon retirement. We additionally discover that some housing adjustments take place up to 6 years before retirement. Moreover, our results reveal significant heterogeneity in the retirement impact by gender, marital status, education, housing tenue, income and wealth. Within traditional heterosexual couple households, housing mobility choices are primarily influenced by the wife's retirement while housing downsizing decisions are only affected by the husband's retirement.

Social Climate, Uncertainty and Fertility Intentions: from the Great Recession to the Covid-19 Crisis

The literature on fertility in context of crises considers major crises exclusively as economic experiences, however, they are also social phenomena, affecting communities, morality and social interactions. When changes in the social climate are of a sufficient magnitude, they tend to break down the social fabric and generate additional uncertainty, more of a social form, which may affect reproductive decisions beyond economic uncertainty alone. Applying Fixed Effects Models to 18 waves of the Swiss Household Panel (2004–2021), this study evaluates the relationship between changes in social climate and social uncertainty and first and second order childbearing intentions, net of economic uncertainty, sociodemographic determinants and unobserved time-invariant individual and local area characteristics. Canton-level mean and variance of generalized trust and optimism about the future are used as proxies of the quality and the unpredictability of the social climate respondents live in. Besides parity, the study explores period variation by comparing the time around the Great Recession (before, during and after) and the years around the Covid-19 pandemic. Results show that the worsening of the social climate and its growing uncertainty correlate with lower and more uncertain first and second birth intentions. Yet, important parity-period interactions emerge.

Effects of digital skills and other individual factors on retirement decision-making and their gender differences

Increasing the pension age as a dominant solution to population ageing does not bring desirable outcomes, if not accompanied by other essential measures in lifelong learning and fighting age discrimination. Moreover, rapid digitalisation and automation in the labour market bring additional uncertainties for the growing group of older workers. The analysis is based on the SHARE data from Waves 5, 6, and 7 and examines predictors of retirement intentions by two different estimation methods. While digital skills are positively associated with a willingness to stay in the labour market in the random-effect modelling, fixed-effects regression shows no correlation between digital skills and retirement intentions. This difference means that digital skills do not correlate with retirement intentions once we control for time-invariant individual characteristics. Thus, increasing ICT literacy among older workers can have a very limited potential for extending working lives. In contrast to this result, starting to be self-employed, health improvement, having an additional grandchild, and losing a partner increase the willingness to work longer. The study identifies the factors shaping retirement intentions, which should be reflected in any effective social policy.

Depressive and anxiety symptoms in adults during the COVID-19 pandemic in England: A panel data analysis over 2 years.

There has been much research into the mental health impact of the Coronavirus Disease 2019 (COVID-19) pandemic and how it is related to time-invariant individual characteristics. However, there is still a lack of research showing long-term trajectories of mental health across different stages of the pandemic. And little is known regarding the longitudinal association of time-varying factors with mental health outcomes. This study aimed to provide a longitudinal profile of how mental health in adults changed across different stages of the COVID-19 pandemic and to examine their longitudinal associations with time-varying contextual (e.g., COVID-19 policy response and pandemic intensity) and individual level factors. This study used data from a large panel study of over 57,000 adults living in England, who were followed up regularly for 2 years between March 2020 and April 2022. Mental health outcomes were depressive and anxiety symptoms. Depressive symptoms were assessed by the Patient Health Questionnaire (PHQ-9) and anxiety symptoms by the Generalized Anxiety Disorder assessment (GAD-7). Entropy balancing weights were applied to restore sample representativeness. After weighting, approximately 50% of participants were female, 14% from ethnic minority backgrounds, with a mean age of 48 years. Descriptive analyses showed that mental health changes were largely in line with changes in COVID-19 policy response and pandemic intensity. Further, data were analysed using fixed-effects (FE) models, which controlled for all time-invariant confounders (observed or not). FE models were fitted separately across 3 stages of the COVID-19 pandemic, including the first national lockdown (21/03/2020-23/08/2020), second and third national lockdowns (21/09/2020-11/04/2021), and "freedom" period (12/04/2021-14/11/2021). We found that more stringent policy response (measured by stringency index) was associated with increased depressive symptoms, in particular, during lockdown periods (β = 0.23, 95% confidence interval (CI) = [0.18 to 0.28], p < 0.001; β = 0.30, 95% CI = [0.21 to 0.39], p < 0.001; β = 0.04, 95% CI = [-0.03 to 0.12], p = 0.262). Higher COVID-19 deaths were also associated with increased depressive symptoms, but this association weakened over time (β = 0.29, 95% CI = [0.25 to 0.32], p < 0.001; β = 0.09, 95% CI = [0.05 to 0.13], p < 0.001; β = -0.06, 95% CI = [-0.30 to 0.19], p = 0.655). Similar results were also found for anxiety symptoms, for example, stringency index (β = 0.17, 95% CI = [0.12 to 0.21], p < 0.001; β = 0.13, 95% CI = [0.06 to 0.21], p = 0.001; β = 0.10, 95% CI = [0.03 to 0.17], p = 0.005), COVID-19 deaths (β = 0.07, 95% CI = [0.04 to 0.10], p < 0.001; β = 0.04, 95% CI = [0.00 to 0.07], p = 0.03; β = 0.16, 95% CI = [-0.08 to 0.39], p = 0.192). Finally, there was also evidence for the longitudinal association of mental health with individual level factors, including confidence in government/healthcare/essentials, COVID-19 knowledge, COVID-19 stress, COVID-19 infection, and social support. However, it is worth noting that the magnitudes of these longitudinal associations were generally small. The main limitation of the study was its non-probability sample design. Our results provided empirical evidence on how changes in contextual and individual level factors were related to changes in depressive and anxiety symptoms. While some factors (e.g., confidence in healthcare, social support) clearly acted as consistent predictors of depressive and/or anxiety symptoms, other factors (e.g., stringency index, COVID-19 knowledge) were dependent on the specific situations occurring within society. This could provide important implications for policy making and for a better understanding of mental health of the general public during a national or global health crisis.

Understanding the effects of informal caregiving on health and well-being: Heterogeneity and mechanisms

This study examines the heterogenous effects of informal caregiving on caregivers' health and well-being and the mechanisms of the effects, which remain largely undiscussed in previous literature. We used a combined estimation of fixed effects and the instrumental variables to address unobserved time-invariant individual characteristics and the endogeneity problem between caregivers' health and caregiving status. Using data from the four waves of the Japanese Study of Aging and Retirement collected in 2007, 2009, 2011, and 2013, and covering individuals aged 50 to 75 at the baselines, we found robust evidence of the negative effects of informal caregiving on caregivers' mental health and life satisfaction, but not on their physical health. Regarding heterogenous effects, we found that informal caregiving adversely affected female but not male caregivers' mental health and life satisfaction. Our results also showed that informal caregiving had greater effects on individuals providing care for their mothers-in-law, with a higher socioeconomic status, living with their in-laws, and belonging to younger groups. Our results indicated that the loss of social networks and leisure and social activities were channels through which informal caregiving might negatively affect caregivers’ health and well-being. This study provides suggestions that policy makers may use to mitigate the negative effects of caregiving with targeted interventions, while formulating policies to support informal caregivers.

Contribution of compositional changes in the workforce to sickness absence trends in Finland

Abstract Background Information on factors driving work disability trends helps to evaluate the potential of interventions to improve the health and work ability of the workforce. We assessed whether the long-term decrease in sickness absences in Finland is explained by observed and unobserved compositional changes in the workforce. Methods Utilising register-based panel data on Finnish private and public sector employees aged 30-62, we examined the annual onset of compensated full sickness absence (granted after 10 working days) in the period 2005-2016. We applied random effects models adjusting for changes in observed sociodemographic factors of the study population. We also applied fixed effects models, with corrections of the estimates for cohort ageing, to additionally account for unobserved time-invariant characteristics of the study population over the years. Results Of the observed factors, increases in educational level partly explained the decreasing trend in sickness absences, and more so among women than men and among private than public sector employees. Changes in occupational class and industrial sector played little role in the public sector and only slightly further explained the sickness absence trend in the private sector. The decreasing trend in sickness absences appeared to be largely explained by unobserved time-invariant individual characteristics. Conclusions The decrease in sickness absences appeared to be more strongly influenced by compositional changes in factors that are established before fully entering the labour market - such as educational level as well as unmeasured individual characteristics that remain unchanged after childhood and early adulthood - than in the work environment or other factors contributing at working age. Attempts to improve the health and work ability of the workforce should not only rely on interventions directed at the working age population, but also on those carried out early during the life course. Key messages • Compositional changes in the workforce should be taken into account when assessing sickness absence trends. • Interventions aiming to improve the health and work ability of the workforce should be implemented already early in the life course and not only in working age.

Linking Eight Decades of Canadian Census Collections.

IntroductionLinking the many decades of census data collected during Canada’s settlement allows researchers to investigate the movement patterns of early settlers, changes in regional demographics, and intergenerational mobility. This research leverages new methodologies and previously untranscribed individual attributes to link for the first time Canadian censuses from 1852 to 1921.&#x0D; Objectives and ApproachThis work aims to build upon prior efforts (1871-1901 linking) to link the decennial Canadian censuses spanning from 1852 to 1921. We use a more complete transcription of the censuses, that has recently become available to researchers through The Canadian People’s project.&#x0D; Our approach to this task begins with reproducing the results of previous work using data from this new transcription. From there, we add additional time-invariant individual characteristics as features to our classification model. We also explore newer methodologies designed to leverage household information during the linkage process, with the goal of increasing the linkage rate.&#x0D; ResultsWe describe the differences between the different methodologies we use, and the steps we took to clean and standardize the data. We compare the links produced by the different methodologies in terms of the number of links yielded, their quality (false positive rate), and certain aspects of the bias present in the resulting collections of links. We discuss the challenges and potential approaches to dealing with sections of the population who remain difficult to link.&#x0D; We expect the advancements in record linkage methodologies for historical populations coupled with this more complete transcription of the censuses to offer advantages over prior approaches when linking these censuses. We expect the resulting linked data to offer new insight into Canada during this time period.&#x0D; Conclusions/ImplicationsThe resulting collection of linked data over this time period should characterize approximately three generations of early Canadians. This linked data will be passed on to other researchers and will allow us to better understand the changing experiences of the Canadian population during these early stages of Canada’s development.

Retirement, Housing Mobility, Downsizing and Neighbourhood Quality – A Causal Investigation

This paper provides the first causal evidence on the impact of retirement on housing choices. Our empirical strategy exploits the discontinuity in the eligibility ages for state pension as an instrument for the endogenous retirement decision and controls for time-invariant individual characteristics. The results show that retirement leads to a statistically significant and sizable increase in the probability of making a residential move or the likelihood of becoming outright homeowners. We also find that individuals downsize both physically and financially and tend to move to better neighbourhoods or closer to the coast upon retirement. We additionally discover that some housing adjustments take place up to 6 years before retirement. Moreover, our results reveal significant heterogeneity in the retirement impact by gender, marital status, education, housing tenue, income and wealth. Within couple households, housing mobility choices are primarily influenced by the wife’s retirement while housing downsizing decisions are only affected by the husband’s retirement. The results suggest that failing to address the endogeneity of retirement often under-states the retirement impact on such housing arrangements.

The persistence of child and adolescence mental healthcare: results from registry data

BackgroundPrevious studies on the persistence of child and adolescent mental healthcare do not consider the role of time-invariant individual characteristics. Estimating persistence of healthcare using standard linear models yields biased estimates due to unobserved heterogeneity and the autoregressive structure of the model. This study provides estimates of the persistence of child and adolescent mental healthcare taking these statistical issues into account.MethodsWe use registry data of more than 80,000 Dutch children and adolescents between 2000 and 2012 from the Psychiatric Case Registry Northern Netherlands (PCR-NN). In order to account for autocorrelation due to the presence of a lagged dependent variable and to distinguish between persistence caused by time-invariant individual characteristics and a direct care effect we use difference GMM-IV estimation. In further analyses we assess the robustness of our results to policy reforms, different definitions of care and diagnosis decomposition.ResultsAll estimation results for the direct care effect (true state-dependence) show a positive coefficient smaller than unity with a main effect of 0.215 (p<0.01), which indicates that the process is stable. Persistence of care is found to be 0.065 (p<0.05) higher for females. Additionally, the majority of persistence of care appears to be associated with time-invariant characteristics. Further analyses indicate that (1) results are robust to different definitions of care and (2) persistence of care does not differ significantly across subgroups.ConclusionsThe results indicate that the majority of mental healthcare persistence for children and adolescents is due to time-invariant individuals characteristics. Additionally, we find that in the absence of further shocks a sudden increase of 10 care contacts in the present year is associated with an average of less than 3 additional care contacts at some point in the future. This result provides essential information about the necessity of budget increases for future years in the case of exogenous increases in healthcare use.

Association between methylphenidate treatment and risk of seizure: a population-based, self-controlled case-series study

Individuals with attention-deficit hyperactivity disorder (ADHD) are at increased risk of seizures. Stimulant medications such as methylphenidate are the most commonly prescribed treatment for ADHD, but the association between their therapeutic use and the risk of seizures is unclear. We aimed to investigate the association between methylphenidate treatment and the risk of seizure. For this population-based observational study, we used the electronic medical record database of the Hong Kong Clinical Data Analysis And Reporting System to identify individuals aged 6-25 years who received at least one methylphenidate prescription during the study period. Individuals with records of seizure or epilepsy before the study period were excluded. Individuals treated with methylphenidate who had seizures during the study period were included in the subsequent analyses, and a self-controlled case-series design was used to control for time-invariant individual characteristics. We did additional analyses using skin infection as a negative control outcome. We compared relative incidence of seizure during periods when individuals were exposed to methylphenidate with that during non-exposed periods. Of 29 604 individuals prescribed methylphenidate between Jan 1, 2001, and Dec 31, 2017, 269 (199 males and 70 females) had incident seizures. The mean age at baseline was 6·66 years (SD 2·01) and the median age at the incident seizure was 9·69 years (IQR 7·62-12·99). The overall incidence of seizure during methylphenidate treatment was 4·4 per 10 000 patient-years. We detected an increased risk of seizure during the first 30 days of methylphenidate treatment compared with that during non-exposed periods, with an incidence rate ratio of 4·01 (95% CI 2·09-7·68). No increase in risk was identified during the following 31-180 days of treatment (1·13, 0·56-2·25) or during subsequent treatment (1·38, 0·92-2·07). We did not identify an increased risk in any risk window for the negative control outcome analysis. No individuals died because of a seizure during the study period. The incidence of seizures was higher in the period immediately after the start of methylphenidate treatment than in the non-exposed period. No increased risk was observed during continuation of methylphenidate treatment. The association between methylphenidate treatment and seizures immediately after initiation of medication can be seen as a potential safety signal. Monitoring of neurological outcomes in individuals with ADHD is recommended when they first start methylphenidate treatment. Hong Kong Research Grants Council.

Well-Being During the Transition from Work to Retirement

We investigate the consequences of retirement from work for the overall well-being of individuals aged 50 and above. Well-being is captured by two different concepts: life satisfaction and agency-freedom, i.e. the evaluation of a person’s ability to do the things s/he wants to do and be who s/he wants to be. We use three observation periods of the Survey of Health, Ageing and Retirement in Europe and include nine European countries. The sample counts 62,082 observations (38,344 individuals) of at least 50 years old. A fixed-effects estimation approach is used so that differences in (time-invariant) individual characteristics are taken into account. We control for changes in financial and health situations of the individual and the situations of the partner. When retiring, people do not immediately report (on average) a different level of life satisfaction, but after 2 years, life satisfaction decreases compared to the beginning of the retirement [identified as Atchley’s honeymoon effect (The sociology or retirement, Wiley, New York, 1976)]. If well-being is expressed as agency-freedom, well-being is immediately positively affected, and this effect does not change after 2 years of retirement. This paper also investigates several forms of heterogeneities in the transition from work to retirement. We consider partial, early and joint retirement, part-time employment and self-employment, and job quality. We find that there is no difference in overall well-being between being partially and fully retired, between being retired before or after the normal retirement age or between those who retire simultaneously with their partner and those who do not. However, for some older workers, such as those employed with a low-quality job, retirement can be a relief from their current employment status. In summary, a policy for longer working careers is, on average, not detrimental to well-being, but some specific groups need special treatment.

The relationship between living arrangements and higher use of hospital care at middle and older ages: to what extent do observed and unobserved individual characteristics explain this association?

BackgroundPrevious research has shown that certain living arrangements, such as living alone, are associated with worse health at older ages. We assessed the association between living arrangements and hospital care use among middle-aged and older adults, and investigated to what extent observed and unobserved individual characteristics explain this association.MethodsLongitudinal Finnish registry data for men and women aged 50–89 years were used for the period 1987–2007. The relationship between living arrangements (based on whether an individual lived with a partner, other adults or alone, and whether they lived with minor/adult children) and heavy hospital care use (i.e., having been in hospital for 8 or more days in a year) was studied. First, we applied logistic regression models and linear probability models controlling for observed time-invariant factors (socioeconomic status measured by education, labour force status, and household income; and marital status), and then individual linear probability models with fixed-effects to further account for unobserved time-invariant individual characteristics in the measurement period. Analyses were done separately for 10 year age-groups.ResultsIn the logistic regression models, men and women who lived alone had higher crude odds of heavy hospital care use than those living only with their partner. These odds ratios were highest for men and women in the youngest age category (50–59 years, 1.72 and 1.36 respectively) and decreased with age. Adjusting for observed time-invariant socioeconomic status attenuated these odds by 14–40%, but adjusting for marital status did not affect the results. Lower odds were observed among adults aged 50–59 years who lived with their partner and (minor or adult) children. But odds were higher for individuals aged 60–79 years who co-resided with their adult children, regardless of whether they lived with a partner. Adjusting for observed time-invariant factors generally did not change these results. After further adjusting for unobserved time-invariant individual characteristics in the individual fixed-effects models, most of these associations largely attenuated or disappeared, particularly for ages 80–89 years.ConclusionsThe association between living arrangements and higher use of hospital care at middle and older ages is largely explained by socioeconomic disadvantage and unobserved time-invariant individual characteristics.

The Association Between Income and Psychotropic Drug Purchases: Individual Fixed Effects Analysis of Annual Longitudinal Data in 2003-2013.

Previous cross-sectional studies show that low income is associated with poor mental health. However, longitudinal research has produced varying results. We assess whether low income is associated with increased psychotropic drug use after accounting for confounding by observed time-varying, and unobserved stable individual differences. The longitudinal register-based data comprises an 11% nationally representative random sample of Finnish residents aged 30-62 years between the years 2003 and 2013. The analytic sample includes 337,456 individuals (2,825,589 person-years). We estimate the association between annual income and psychotropic purchasing using ordinary-least-squares and fixed effects models, the latter controlling for all unobserved time-invariant individual characteristics. The annual prevalence of psychotropic purchasing was 15%; 13% among men and 18% among women. Adjusted for age squared, sex and calendar year, the doubling of income decreased the probability of purchases by 4 percentage points (95% confidence interval: 4,4) in the ordinary-least-squares model. We observed no association after further adjusting for observed sociodemographic characteristics and unobserved individual differences in the fixed effects specification. Following adjustment for an extensive set of confounders, no contemporaneous association between variations in annual individual income and psychotropic drug purchasing was observed. Similar results were obtained irrespective of baseline income level and sex. The results imply that indirect selection based on preexisting individual characteristics plays a major role in explaining the association between variations in income measured over the short term, and psychotropic drug purchases. The association appears largely attributable to unobserved, stable individual characteristics. See video abstract at, http://links.lww.com/EDE/B463.

Bias, accuracy and sample size in the systematic linking of historical records

IntroductionLinking distinct historical sources on an automated basis directs attention to the quality and representativeness of the linked data created by these systems. Linking with time-invariant personal characteristics arguably minimizes bias or departures from representativeness even though a wider set of features might generate more links.&#x0D; Objectives and ApproachThe objective of this research is to compare, evaluate and understand bias when two linking methodologies are employed on the same data sources. Our approach to studying this problem is by comparing linked records from Canadian censuses (linking 1871 to 1881) generated by two different linking strategies. The first method is a support vector machine based classification model on time-invariant individual characteristics. Using this method a large number of multiple matches is generated, as records look similar on a small number of time-invariant individual characteristics. The second method adds a second stage of disambiguating multiple matches using family information.&#x0D; ResultsWe compare the links produced by the two methods used in the study and we discuss the results. The comparison is in terms of number of links produced, their quality (false positive rate) and the bias of the linked data produced. A complication is that there are many dimensions of bias. Even time-invariant criteria typically generate some bias.&#x0D; As expected, the two-step process produces a larger linked sample. Interestingly, it also produces a lower error rate and different patterns of bias. Both methods understate the Quebec-born, French-ethnicity, the unmarried and adolescents. Unexpectedly, the bias in favour of married people is larger using individual (first method) than family information (second method). However, family-based linking does over-represent young children.&#x0D; Conclusion/ImplicationsResults suggest that neither method will be universally preferable. Rather, the choice of research question may affect the preferred balance of biases and link rate. Fortunately, the advance of computational capacity allows a researcher to select a method that generate links most appropriate for the problem at hand.

Relationship of Smoking to Hypertension in a Developing Country.

Smoking and hypertension are 2 leading causes of death worldwide, and it has been claimed that smoking is a cause of hypertension despite inconsistent results in the literature. In addition, the literature focuses mostly on developed countries, although more people in developing countries are adversely affected by smoking and hypertension. This study attempted to address critical limitations of the literature by analyzing a nationally representative survey of Indonesians with a large sample size and a long follow-up period. This study drew on the Indonesian Family Life Survey and followed men aged ≥15 years from 1993 to 2014. The sample size was 14,722 in the unbalanced panel data, 7,876 in the balanced data, and 12,715 in the cross-sectional data. Fixed-effects models were used to remove time-invariant individual characteristics for the panel data. These results were compared to those derived from the cross-sectional data in linear probability models, and the importance of controlling for time-invariant individual characteristics was assessed. The results derived from fixed-effects models indicated no relation of smoking to hypertension. The relation remained null whether the unbalanced or balanced panel data, hypertension status or blood pressure, or smoking status or smoking intensity in linear or nonlinear form were analyzed. In contrast, the results derived from linear probability models exaggerated the relation. Given the null relation, smoking was not associated with hypertension in this population.

Interpersonal Relationships as Protective and Risk Factors for Psychopathy: A Follow-up Study in Adolescent Offenders.

Friendships and romantic relationships may function as protective and risk factors for psychopathic traits. To better understand potential causal associations, we investigated whether within-individual changes in relationship characteristics were related to changes in psychopathic traits over time. Data were derived from ten repeated measurements of the Pathways to Desistance longitudinal study of 1354 offending adolescents (14.3% female; 40.1% Black). Analyses were adjusted for age, gender, ethnicity, self-reported offending and living facilities. Relationships of high quality were associated with lower psychopathic traits, whereas antisocial behavior and antisocial influence in relationships were related to higher psychopathic traits. Within-individual analysis indicated that time-invariant individual characteristics did not confound these associations. The findings suggest that the quality and antisocial activities of interpersonal relationships can affect positively or negatively on the levels of psychopathy.

Impact of Early Work Experiences on Subsequent Paid Employment for Young Adults With Disabilities

To better understand how early work experience shapes subsequent employment outcomes for young people (ages 18 to 20) with disabilities, we analyzed longitudinal data from the Youth Transition Demonstration (YTD) evaluation to test whether the employment experiences of 1,053 youth during the initial year after entry affected their employment during the third year after entry. To derive causal estimates, we used a dynamic-panel estimation model to account for time-invariant unobserved individual characteristics that may be correlated with youth’s self-selection into both early and later employment. We also controlled for other socioeconomic and health factors that may affect later employment. We found that early work experience increases the probability of being employed 2 years later by 17 percentage points. This estimate is an important advancement over the correlational approaches that characterize the current literature and provides stronger evidence that early work experience is a key determinant of subsequent labor market success.

Negative wealth shock and short-term changes in depressive symptoms and medication adherence among late middle-aged adults

BackgroundExperiencing a negative wealth shock in late middle age may cause high levels of stress and induce reductions in health-related consumption.MethodsWe used data on late middle age individuals (51–64 years)...

Dietary Habits and Cognitive Impairment Risk Among Oldest-Old Chinese.

This study examined the longitudinal relationship between dietary habits and cognitive impairment among the oldest-old Chinese. Cognitively intact adults aged 80 or older (N = 4,749) came from the Chinese Longitudinal Healthy Longevity Survey. Cognitive impairment was assessed by Mini-mental State Examination. Cox regressions were performed to examine the relationship between dietary habits and cognitive impairment onset during follow-up period of 1998-2012, adjusting for various time-variant and time-invariant individual characteristics. Compared with those who rarely/never consumed fruit, vegetables, meat, and soybean-derived products, participants consuming such products almost every day were 21%, 25%, 17%, and 20% less likely to develop cognitive impairment during follow-up, respectively. Compared with those who rarely/never consumed sugar, participants consuming sugar almost every day were 17% more likely to develop cognitive impairment during follow-up. Consumption of fish, egg, salt-preserved vegetable, tea, and garlic was not found to be associated with cognitive impairment. Dietary habits might profoundly impact cognitive functioning among the oldest-old Chinese. This work has limitations pertaining to study design and measure. Future work adopting experimental design and refined dietary measures is warranted to confirm these findings and inform public nutrition practices aiming at preventing cognitive decline among the oldest-old Chinese population.

The Role of China's Household Registration System in the Urban-Rural Income Differential

Together with the rapid growth of the Chinese economy, there has been a growing divide in the earnings of urban and rural residents. In this paper we focus on China's household registration system, or hukou, as a potential source of the earnings gap. Using multiple waves of data from the Chinese Health and Nutrition Survey from 1993 through 2011, we take advantage of variation in hukou status generated by individual-level changes over time. Unlike previous studies, we are able to control for fixed individual-specific characteristics that determine earnings and focus specifically on estimating an urban hukou premium. For estimates that do not account for time-invariant individual characteristics, urban hukou holders earn almost 30% more than rural hukou holders. After we account for individual-level fixed characteristics, the urban hukou premium drops to 6–8%. Our empirical evidence indicates that the hukou system is a notable component of the urban-rural earnings differential, but its importance should not be overstated. Given long-standing differences in access to government funding and social services between rural and urban populations, relaxing residency restrictions may not be a panacea for curbing rising income inequality.