Prospective associations between the state minimum wage and cognitive function in older Americans Authors: Boram Lee, Lucia C. Petito, Jiaxing Bao, Daniel Kim Published: 2026 ABSTRACT ObjectivesLimited research has explored how social policies influence cognitive function at older ages. This study examined the association between the U.S. state minimum wage and subsequent cognitive function among older American adults.MethodsUsing data from the U.S. Health and Retirement Study, we studied non-self-employed workers who were aged 50-75 without dementia at baseline and followed between 1998 and 2020 (N = 7263). We employed fixed-effects linear models to estimate associations between the state minimum wage and cognitive function, measured by the modified version of Telephone Interview for Cognitive Status (TICS; higher values indicating better cognitive function), using 2, 6, and 10-year lags. We further stratified our models by levels of earning wages, working hours, race/ethnicity, and sex.ResultsAmong all workers (n = 7263), the state minimum wage was not associated with cognitive function (β = 0.05; 95% CI = −0.01 to 0.12). However, for low-wage workers (n = 880) and part-time workers (n = 1432), a $1 higher state minimum wage was associated with a 0.29 point (95% CI = 0.14 to 0.45, p = .001) and a 0.17 point (95% CI = 0.04 to 0.30, p = .014) higher TICS score 6 years later, respectively, controlling for time-invariant individual factors and time-varying individual- and state-level covariates. Among Hispanic low-wage workers, a higher state minimum wage was associated with lower cognitive function in a 2-year lagged model (β = −0.41; 95% CI = −0.77 to −0.06) but with higher cognitive function in a 6-year lagged model (β = 0.41, 95% CI = 0.03, 0.79, p = .036).DiscussionOur findings suggest that increases in the state minimum wage may benefit the cognitive health of low-earning older adults, though heterogeneities by race/ethnicity may exist. Introduction As American society ages rapidly, the burden of cognitive impairment has burgeoned (Rajan et al., 2021). Alzheimer's Disease and related dementias are the sixth leading cause of death in the United States (US), costing $305 billion in 2020 and projected to increase to $2.2 trillion in 2060 (Nandi et al., 2024; Rajan et al., 2021). The burden of cognitive impairment is disproportionately higher among those with low incomes and among racial/ethnic minorities (Rajan et al., 2021; Wang et al., 2023). Most prior studies on cognitive impairment have primarily focused on individual-level factors such as smoking, hearing loss, and chronic diseases (Livingston et al., 2020), and little is known about how social policies may influence the cognitive function of older adults. Increases in the minimum wage may serve as a policy lever to influence cognitive health of workers. The stress process model suggests that socioeconomic status, including income, shapes individuals’ exposure to stressors, vulnerability to stress, and access to coping resources, thereby influencing long-term health outcomes (Pearlin et al., 1981). Meanwhile, life course theory posits that health in older adulthood is influenced by cumulative experiences across the life course (Richards & Deary, 2005). This perspective suggests that even modest socioeconomic changes during mid-adulthood could compound over time and ultimately produce meaningful differences in cognitive function in later life (Cheng et al., 2025; Lyu & Burr, 2016). Despite this theoretical support, no prior studies have directly examined associations between minimum wage increases and cognitive function. Prior research on minimum wages nevertheless suggests potential pathways through which minimum wage increases might influence cognitive health. For example, higher earnings may reduce financial stress, a known risk factor for cognitive impairment, and facilitate health-promoting behaviors that support cognitive health (e.g., purchasing high-quality food) (Blavin & Gangopadhyaya, 2022; Lenhart, 2017; McCarrier et al., 2011). Conversely, there may be unintended consequences of minimum wage increases, such as unemployment or loss of eligibility for social welfare benefits, which could adversely affect cognitive health (Gangopadhyaya et al., 2019; Vélez-Coto et al., 2021). The mixed expectations about the effects of minimum wage increase on cognitive function and the scarcity of empirical evidence on this topic underscore the need for further investigation. To address these gaps, we aimed to examine the relationships between the state-level minimum wage and subsequent cognitive function among US adults aged 50 and older who were followed between 1998 and 2020. During this time period, 45 states and the District of Columbia increased their minimum wages and 5 states did not establish a minimum wage (University of KentuckyCenterfor Poverty Research, 2022, Feb). We hypothesized that a higher state minimum wage would be associated with better cognitive function in later life among US older adults. We further hypothesized that the association between the state minimum wage and cognitive function would differ by earnings, working hours, race/ethnicity, and sex. Methods Data and sample We used data from the Health and Retirement Study (HRS), a U.S. population-based nationally representative longitudinal multi-cohort study of Americans aged 50 years and older. The HRS, conducted by the University of Michigan, has collected data on health, economic, and family status via biannual interviews since 1992 (Servais, 2010). We constructed the analytic dataset by combining the RAND HRS longitudinal files (user-friendly dataset that harmonizes all waves of HRS, produced by RAND) with residential data from the HRS restricted geographic files (Health and Retirement Study, May 2024). We used the HRS cohorts enrolled in 1992, 1998, 2004, and 2010 (the minority supplemental sample). Our analytical sample consisted of dementia-free older American workers born between 1924 and 1953, with complete data on cognitive function collected at least twice during the study period. We excluded those whose baseline scores on the Telephone Interview for Cognitive Status (TICS) were consistent with the presence of dementia (0-6 score) (Langa et al., 2017). We restricted the sample to workers aged 50 to 75 years at baseline, excluding those who were retired, unemployed, or self-employed. The baseline year was either 1998 when the enrolled cohorts expanded to cover the full range of ages 50 and older, or a subsequent year when individuals entered the study. The study participants were followed until death, dropout, reaching the age of 85, or the end of follow-up in 2020, whichever occurred first. Individuals were censored upon reaching the age of 85 to reduce the potential for survival bias (Banack et al., 2019). The final analytical sample size was 7263 (see Fig. 1 for details).Fig. 1Selection of the final analytical sample (n = 7263) from the Health and Retirement Study, 1998-2020. Selection of the final analytical sample (n = 7263) from the Health and Retirement Study, 1998-2020. Measures Outcome Cognitive function was assessed using a score derived from a previously validated version of the Telephone Interview for Cognitive Status (TICS), by being shown to successfully predict the diagnosis of dementia (Crimmins et al., 2011; Langa et al., 2017). TICS scores ranged from 0 to 27, with higher scores indicating better cognitive function. The TICS score is calculated as the sum of the composite scores with subcategories: immediate and delayed word recall (0-20), serial 7's test (0-5), and backward count (0-2). The respondents were asked to recall as many words as possible immediately after reading a list of 20 words (immediate word recall) and again 5 min later (delayed word recall). The serial 7's test asked respondents to subtract 7 from 100 five times. Lastly, the respondents were asked to count backward from 20 as quickly as possible for 10 continuous numbers. Exposure The effective state minimum wage refers to the higher of the state minimum wage or the federal minimum wage each year. If a state did not have a minimum wage or had a lower minimum wage than the federal minimum wage, the federal minimum wage for that year was assigned. All effective state minimum wages were adjusted to 2021 U.S. dollars based on the Bureau of Labor Statistics’ Consumer Price Index (The United States Census Bureau, 2024). Data on minimum wages were obtained from the University of Kentucky Center for Poverty Research (University of Kentucky Center for Poverty Research, 2022) (Supplementary Table 1). To investigate delayed effects, we used lag periods of two, six, and ten years for the state minimum wage. The lag period represents the number of years the state minimum wage preceded the TICS score measurement, ensuring temporality of the association (Buszkiewicz et al., 2023). Covariates All models were adjusted for individual- and state-level covariates that could potentially confound the association of interest. Individual-level covariates included age, marital status, the number of household members (1, 2, and 3 or more), the quintile of household income, urban/rural residence, and whether the respondent had health insurance. State-level covariates included the percentage of the population self-identifying as Black, the percentage of older adults aged 65 years or older, and median household income, and were derived from the US Census and American Community Surveys (ACS) available through the HRS Contextual Data Resource. We used 1990, 2000, and 2010 US Census data for the 1988-2006 covariates and 2005-2009, 2006-2010, 2008-2012, 2010-2014, 2014-2018, 2016-2020 ACS 5-year estimates for the 2008-2018 covariates. We also controlled for two state policies—the state earned income tax credits (EITC) rate as a percentage of federal credit and the maximum monthly allotment for food stamps adjusted to 2021 U.S. dollars (The United States Census Bureau, 2024). Data on these state policies were drawn from the University of Kentucky Center for Poverty Research National Welfare data (University of Kentucky Center for Poverty Research, 2022). We lagged individual-level covariates by 2 years, except for the baseline year where we used concurrent data. We used the same time lags for state-level covariates as for the minimum wage. For example, the 6-year lagged models included 6-year lagged state-level covariates and state minimum wages. Statistical analysis We calculated proportions and means for the categorical and continuous characteristics of the sample at baseline, respectively. Using individual fixed-effects linear models, we estimated associations between the effective state minimum wage and cognitive function. Fixed-effects models enable us to estimate differences in cognitive function within an individual whose state minimum wage changed over time by controlling for unmeasured individual-level time-invariant factors. We first fit models for all workers (equation (1)) and then assessed heterogeneity (equation (2)) in the associations between the state minimum wage and subsequent (2, 6, and 10 year-lagged) cognitive function by subgroups by adding interaction terms between the state minimum wage and each subgroup indicator (i.e., level of earning wages (low-wage vs. higher-wage), working hours (full-time vs. part-time), and race/ethnicity (non-Hispanic White vs. non-Hispanic Back vs. Hispanic)) in the models. Our models are shown below.(equation 1)Yist=β0+αi+δt+β1Xst+β2Zit+β3Wst+εist(equation 2)Yist=β0+αi+δt+β1Xst+β2Zit+β3Wst+β4(Xst∗Gi)+εistwhere Yist is cognitive function for an individual i residing in state s at time t, β0 is the global intercept, αi represents individual fixed effects, δt denotes fixed effects for time t, Xst represents the state minimum wage in state s at time t, Zit are a set of time-varying individual-level covariates, Wst are a set of time-varying state-level covariates, Gi represents subgroup indicator, and Xst∗Gi captures the interaction between the state minimum wage and subgroup membership, and εist is the error term. Low-wage workers were defined as individuals whose hourly earnings were up to 30% above the state minimum wage in a given year, while higher-wage workers were defined as those earning more. Full-time workers referred to those working at least 35 h per week or 36 weeks per year; otherwise they were considered part-time workers. These subgroups were classified based on the level of earning wages and working hours at the baseline year of observation. Furthermore, we examined the association between the state minimum wage and cognitive function among low-wage workers, with stratification by level of earning wages, working hours, and race/ethnicity. We used two-way clustered standard errors to account for clustering within individual over time and correlation across individuals within the same state in the main analysis; however, we used only robust standard errors clustered at the individual level in subgroup analyses because of convergence issues encountered due to the lack of sufficient number of clusters within smaller subgroups. We applied individual-level sampling weights at baseline to account for the complex survey design. All analyses were conducted using Stata 18.0 (StataCorp, TX). In addition, we performed sensitivity analyses. We repeated the analyses among individuals who stayed in the same state during the lag periods prior to baseline. We further examined the robustness of findings after: 1) restricting the sample to individuals who lived in the same state in the past 6 years; 2) restricting the sample to those aged between 50 and 65 at baseline; 3) excluding the minority oversample who was supplementally sampled in 2010, thereby limiting the sample to the 1992, 1998, and 2004 cohorts; 4) restricting the sample to those who reported their cognitive function on at least five occasions over the study period; and 5) using a censored age of 90 years old. This study was deemed exempt by the Institutional Review Board at Northeastern University. Results Among the 7263 individuals in the sample, 51.8% were female, 78.4% were non-Hispanic White, 10.5% were non-Hispanic Black, 73.8% were married, 5.6% were low-wage workers, and 17.3% were part-time workers (Supplementary Table 2). The mean baseline age was 55.4 years, and the mean baseline TICS score was 17.4. The state minimum wage was not associated with subsequent cognitive function among all workers in the 2-, 6, and 10-year lagged models (Table 1). Interaction models indicated significant heterogeneity by working hours (p = .001) and race/ethnicity (p < .001). The interaction between the state minimum wage and the level of earning wages (low wage vs. higher wage) was only marginally significant, providing evidence that the association may be stronger among low-wage workers.Table 1Associations between the state minimum wage and subsequent cognitive function score among older adults, Health and Retirement Study, 1998-2020.Table 1nEffective state minimum wage2-year lag6-year lag10-year lagMean difference in TICS95% CIMean difference in TICS95% CIMean difference in TICS95% CIAll workers7263−0.04(-0.11, 0.02)0.05(-0.01, 0.12)−0.001(-0.06, 0.06)Working hours Full time5831−0.03(-0.10, 0.05)0.02(-0.05, 0.09)0.002(-0.06, 0.07) Part time1432−0.11(-0.26, 0.04)0.17∗(0.04, 0.30)−0.003(-0.11, 0.10)Wage level Low wage880−0.06(-0.27, 0.45)0.29∗∗(0.14, 0.45)0.15(-0.11, 0.41) Higher wage6383−0.04(-0.11, 0.03)0.02(-0.05, 0.09)−0.01(-0.08, 0.05)Race/ethnicity NH White5034−0.05(-0.12, 0.02)0.04(-0.02, 0.11)−0.01(-0.08, 0.05) NH Black12380.11(-0.10, 0.32)−0.02(-0.26, 0.23)−0.24(-0.28, 0.23) Hispanic790−0.09(-0.22, 0.03)0.10(-0.02, 0.22)0.14(-0.02, 0.30)Notes. All models were adjusted for individual-level covariates (i.e., age, marital status, the number of household members, household income, residential area, health insurance) and state-level covariates (i.e., state earned income tax credits rate as a percentage of federal credit, the maximum monthly allotment for food stamps, the percentage of Black in the residing state, the percentage of older adults aged 65 years or older in the residing state, and median household income in the residing state). CI=Confidence Interval; NH=Non-Hispanic; ∗p < .05, ∗∗p < .01. Table 1Associations between the state minimum wage and subsequent cognitive function score among older adults, Health and Retirement Study, 1998-2020.Table 1nEffective state minimum wage2-year lag6-year lag10-year lagMean difference in TICS95% CIMean difference in TICS95% CIMean difference in TICS95% CIAll workers7263−0.04(-0.11, 0.02)0.05(-0.01, 0.12)−0.001(-0.06, 0.06)Working hours Full time5831−0.03(-0.10, 0.05)0.02(-0.05, 0.09)0.002(-0.06, 0.07) Part time1432−0.11(-0.26, 0.04)0.17∗(0.04, 0.30)−0.003(-0.11, 0.10)Wage level Low wage880−0.06(-0.27, 0.45)0.29∗∗(0.14, 0.45)0.15(-0.11, 0.41) Higher wage6383−0.04(-0.11, 0.03)0.02(-0.05, 0.09)−0.01(-0.08, 0.05)Race/ethnicity NH White5034−0.05(-0.12, 0.02)0.04(-0.02, 0.11)−0.01(-0.08, 0.05) NH Black12380.11(-0.10, 0.32)−0.02(-0.26, 0.23)−0.24(-0.28, 0.23) Hispanic790−0.09(-0.22, 0.03)0.10(-0.02, 0.22)0.14(-0.02, 0.30)Notes. All models were adjusted for individual-level covariates (i.e., age, marital status, the number of household members, household income, residential area, health insurance) and state-level covariates (i.e., state earned income tax credits rate as a percentage of federal credit, the maximum monthly allotment for food stamps, the percentage of Black in the residing state, the percentage of older adults aged 65 years or older in the residing state, and median household income in the residing state). CI=Confidence Interval; NH=Non-Hispanic; ∗p < .05, ∗∗p < .01. Associations between the state minimum wage and subsequent cognitive function score among older adults, Health and Retirement Study, 1998-2020. Notes. All models were adjusted for individual-level covariates (i.e., age, marital status, the number of household members, household income, residential area, health insurance) and state-level covariates (i.e., state earned income tax credits rate as a percentage of federal credit, the maximum monthly allotment for food stamps, the percentage of Black in the residing state, the percentage of older adults aged 65 years or older in the residing state, and median household income in the residing state). CI=Confidence Interval; NH=Non-Hispanic; ∗p < .05, ∗∗p < .01. In the subgroup analyses, we found that a $1 higher state minimum wage was associated with a 0.29 (95% confidence interval [CI] = 0.14, 0.45, p = .001) and a 0.17 (95% CI = 0.04, 0.30, p = .014) point increase in the TICS score six years later among low-wage workers (n = 880) and part-time workers (n = 1432), respectively, controlling for time-invariant individual factors and individual- and state-level covariates (Table 1). These associations were not observed in the 2-year and 10-year lagged models. There was no evidence of an association between the state minimum wage and cognitive function among higher-wage and full-time workers in any lagged model. When we stratified low-wage workers by working hours and race/ethnicity (Table 2), the association between a higher state minimum wage and higher cognitive function 6 years later was significant among full-time low-wage workers (β = 0.33, 95% CI = 0.03, 0.62, p = .028) but marginally non-significant among part-time low-wage workers (β = 0.25, 95% CI = −0.02, 0.53, p = .069) and White low-wage workers (β = 0.25, 95% CI = −0.12, 0.50, p = .062). For Hispanic low-wage workers, a higher state minimum wage was associated with a lower TICS score two years later (β = −0.41, 95% CI = −0.77, −0.06, p = .021) but with a higher TICS score in 6-year (β = 0.41, 95% CI = 0.03, 0.79, p = .036) and 10-year lagged models (β = 0.38, 95% CI = −0.01, 0.78, p = .056), although the significance was only marginal in the 10-year lagged model. We found no association among Black low-wage workers.Table 2Among low-wage workers (n = 880), associations between the state minimum wage and subsequent cognitive function score, stratified by working hours, and race/ethnicity, Health and Retirement Study, 1998-2020.Table 2nEffective state minimum wage2-year lag6-year lag10-year lagMean difference in TICS95% CIMean difference in TICS95% CIMean difference in TICS95% CIWorking Hours Full time462−0.12(-0.42, 0.18)0.36∗(0.03, 0.62)0.16(-0.11, 0.44) Part time418−0.02(-0.29, 0.24)0.25a(-0.02, 0.53)0.11(-0.20, 0.43)Race/ethnicity NH White4750.09(-0.16, 0.35)0.25b(-0.12, 0.50)−0.01(-0.28, 0.26) NH Black177−0.20(-0.75, 0.35)0.07(-0.39, 0.53)0.09(-0.53, 0.71) Hispanic202−0.41∗(-0.77, −0.06)0.41∗(0.03, 0.79)0.38c(-0.01, 0.78)Notes. All models were adjusted for individual-level covariates (i.e., age, marital status, the number of household members, household income, residential area, health insurance) and state-level covariates (i.e., state earned income tax credits rate as a percentage of federal credit, the maximum monthly allotment for food stamps, the percentage of Black in the residing state, the percentage of older adults aged 65 years or older in the residing state, and median household income in the residing state). CI=Confidence Interval; NH=Non-Hispanic.∗p < .05, ∗∗p < .01.ap = .069.bp = .062.cp = .056. Table 2Among low-wage workers (n = 880), associations between the state minimum wage and subsequent cognitive function score, stratified by working hours, and race/ethnicity, Health and Retirement Study, 1998-2020.Table 2nEffective state minimum wage2-year lag6-year lag10-year lagMean difference in TICS95% CIMean difference in TICS95% CIMean difference in TICS95% CIWorking Hours Full time462−0.12(-0.42, 0.18)0.36∗(0.03, 0.62)0.16(-0.11, 0.44) Part time418−0.02(-0.29, 0.24)0.25a(-0.02, 0.53)0.11(-0.20, 0.43)Race/ethnicity NH White4750.09(-0.16, 0.35)0.25b(-0.12, 0.50)−0.01(-0.28, 0.26) NH Black177−0.20(-0.75, 0.35)0.07(-0.39, 0.53)0.09(-0.53, 0.71) Hispanic202−0.41∗(-0.77, −0.06)0.41∗(0.03, 0.79)0.38c(-0.01, 0.78)Notes. All models were adjusted for individual-level covariates (i.e., age, marital status, the number of household members, household income, residential area, health insurance) and state-level covariates (i.e., state earned income tax credits rate as a percentage of federal credit, the maximum monthly allotment for food stamps, the percentage of Black in the residing state, the percentage of older adults aged 65 years or older in the residing state, and median household income in the residing state). CI=Confidence Interval; NH=Non-Hispanic.∗p < .05, ∗∗p < .01.ap = .069.bp = .062.cp = .056. Among low-wage workers (n = 880), associations between the state minimum wage and subsequent cognitive function score, stratified by working hours, and race/ethnicity, Health and Retirement Study, 1998-2020. Notes. All models were adjusted for individual-level covariates (i.e., age, marital status, the number of household members, household income, residential area, health insurance) and state-level covariates (i.e., state earned income tax credits rate as a percentage of federal credit, the maximum monthly allotment for food stamps, the percentage of Black in the residing state, the percentage of older adults aged 65 years or older in the residing state, and median household income in the residing state). CI=Confidence Interval; NH=Non-Hispanic. ∗p < .05, ∗∗p < .01. p = .069. p = .062. p = .056. In sensitivity analyses, when we restricted our sample to those who had stayed in the same state in the past six years prior to baseline (based on approximately 60% of the sample reporting their residence history), we found a significant association between a higher state minimum wage and better cognitive function among Black low-wage workers (β = 0.81, 95% CI = −0.03, 1.59, p = .042, Supplementary Table 3). The findings of the main analyses were robust in four other sensitivity analyses with varying sample eligibility and censored age specifications (Supplementary Tables 4–7). Discussion To our knowledge, this study is the first to examine the prospective associations between the state minimum wage and cognitive function among older adults in the United States using nationally-representative longitudinal data. We found that higher state minimum wage was associated with better cognitive function six years later, particularly among low-wage and part-time workers. The stratified analyses of low-wage workers revealed that these associations were evident among Hispanic American low-wage workers. Our findings of the beneficial relationship between a higher state minimum wage and better cognitive function among low-wage workers align with those of Kezios et al. who found that individuals who consistently earned low wages over 12 years of midlife experienced faster memory decline in later life, whereas those whose wages fluctuated between low and higher levels did not exhibit the same pattern (Kezios et al., 2022). From a life course perspective, these findings suggest that wage trajectories during mid-adulthood may have enduring implications for later-life cognitive trajectories. Persistent exposure to low wages during mid-adulthood may not simply reflect immediate economic hardship but may also serve as an indicator of cumulative socioeconomic disadvantages. This suggests that increasing the minimum wage may help reduce low-wage workers’ vulnerability to cognitive impairment in older adulthood by interrupting long-term accumulation of income-related disadvantages. While the observed association of a 0.29 TICS increase among low-wage workers is modest, its potential population-level and life course implications should not be dismissed. Although this magnitude would not independently shift individuals across widely used TICS-based cognitive categories, including dementia (0-6), cognitive impairment without dementia (7-11), and normal cognitive functioning (12-27) (Crimmins et al., 2011), small protective benefits sustained over time may accumulate into meaningful differences over the life course, particularly among individuals at elevated risk of dementia. The beneficial association observed among part-time workers might be explained by a large proportion of low-wage workers in part-time workers (18.1% of part-time workers vs. 4.6% of full-time workers) because low-wage workers are most likely to be sensitive to minimum wage changes. This interpretation is consistent with the stress process model, which suggests that individuals in economically disadvantaged positions are particularly vulnerable to chronic stress and have limited resources to buffer its consequences (Meyer et al., 2008; Pearlin et al., 1981). Moreover, our findings are consistent with prior studies reporting associations between higher minimum wages and better health outcomes, including reduced mortality from suicide, diabetes, and cardiovascular disease (Lenhart, 2017; Paul Leigh et al., 2019; Van Dyke et al., 2018), and better birth outcomes (Wehby et al., 2020). This concordance of findings indicates that the mechanisms linking minimum wage increases to better health outcomes in low-wage workers may also apply to cognitive health. Several mechanisms could plausibly account for our findings. First, additional earnings may reduce psychological stress from financial strain, a well-documented contributor to cognitive decline (Samuel et al., 2020). Second, higher earnings may enable low-wage workers to make healthier choices more easily in daily life, such as purchasing high-quality food, engaging in physical activities, and utilizing health care services (Blavin & Gangopadhyaya, 2022; Lenhart, 2017; McCarrier et al., 2011). These choices can directly improve cognitive health and indirectly do so by preventing chronic diseases (e.g., diabetes) associated with cognitive decline (Lenhart, 2017). Third, additional earnings may allow low-wage workers to participate in educational and social activities that promote cognitive function (Marmot, 2002). Lastly, from a population-level perspective, if minimum wage increases reduce income inequality as intended, it could lead to structural changes (e.g., increased social mobility), which may reduce perceived discrimination and deprivation, eventually enhancing health of economically-vulnerable older adults (Kim et al., 2016; Walsh et al., 2023). An additional notable finding is that among Hispanic low-wage workers, a higher state minimum wage was associated with lower cognitive function in the short term but higher cognitive function over longer lag periods. Although this pattern should be interpreted cautiously given the small sample size in subgroup-specific analyses, it may suggest that the potential cognitive benefits of minimum wage increases require sustained exposure to emerge. Short-term negative associations may reflect possible labor market adjustment following wage changes (e.g., reduced hours or job displacement), whereas longer-term positive associations may reflect the gradual accumulation of benefits from higher wages. Indeed, this delayed pattern is broadly consistent with previous studies showing lagged associations between socioeconomic and educational exposures and cognitive function (Chang et al., 2021; Kim et al., 2016). Additional research with larger samples of low-wage workers is warranted to examine racial/ethnic disparities and the temporal dynamics of associations between minimum wage increases and cognitive function. Our study has several limitations. First, although the statistical models controlled for various covariates and unmeasured time-invariant factors, residual confounding may still have been present due to unadjusted time-varying factors (e.g., social integration). Second, we assigned the baseline state's minimum wage during the lag period before baseline, assuming participants remained in the same state. While some participants may have relocated during the lag period before baseline, the sensitivity analysis with non-movers generally supported the robustness of our findings. Third, despite the possibility of interactions between demographics, we could not explore their combined interactions due to sparse data. Last, the HRS core cognitive battery administered to all HRS participants does not include dedicated measures of executive function; hence, our ability to assess this domain of cognitive function was limited. While the modified TICS score is a validated measure, it relies on self-report and is subject to misclassification. Despite these limitations, our study, using nationally-representative data of older Americans over an extended follow-up period, contributes to the minimum wage literature by demonstrating prospective associations between higher state minimum wages and better cognitive function in older life, particularly among low-wage workers. Amidst heated debates around minimum wage increases, these findings provide population-based evidence of potential cognitive health benefits of a higher minimum wage. CRediT authorship contribution statement Boram Lee: Writing – review & editing, Writing – original draft, Project administration, Methodology, Investigation, Formal analysis, Conceptualization. Lucia C. Petito: Writing – review & editing, Supervision, Methodology. Jiaxing Bao: Formal analysis, Methodology. Daniel Kim: Writing – review & editing, Supervision, Project administration, Methodology, Investigation, Funding acquisition. Data statement The data that support the findings of this study are available from the Health and Retirement Study Survey Research Center (http://hrs.isr.umich.edu/). Restrictions apply to the availability of these data, which were used under license for this study. Data are available with the permission of the Health and Retirement Study Survey Research Center after approval of a proposal. Ethical statement This study was deemed exempt by the Institutional Review Board at Northeastern University. Funding This work was supported by the National Institute on Aging (RF1 AG075916 to DK). Conflict of interest This work was supported by the National Institute on Aging (RF1 AG075916 to DK). LCP reports unrelated research funding from Omron Healthcare Co., Ltd. BL reports unrelated research funding from Ewha Womans University (RP-Grant 2025).