Abstract

Introduction:

As population aging intensifies, elder care has emerged as a central issue affecting socioeconomic sustainability, social harmony, and stability. The coordinated development of social care and family-based care has become pivotal to addressing the challenges of elder care. The rise of community-based home care services, while bridging gaps in family-provided care, has also exerted far-reaching effects on intergenerational family support. Therefore, examining the impact of community-based home care service utilization on intergenerational family support carries significant implications for addressing the elder care challenge.

Methods:

Drawing on data from the 2020 Chinese Longitudinal Aging Social Survey (CLASS) and given the ordinal nature of the dependent variables, this study employs an ordered Probit model, instrumental variable estimation (IV ordered Probit), and propensity score matching (PSM) to examine the effect of community-based home care service utilization on intergenerational family support.

Results:

Community-based home care service utilization significantly promotes intergenerational economic support, yet has a crowding-out effect on intergenerational instrumental and emotional support. Moderation analysis reveals that co-residence with adult children mitigates the adverse effects of community-based home care service utilization on intergenerational instrumental and emotional support. Heterogeneity analysis indicates that the negative effects on intergenerational instrumental and emotional support are primarily concentrated among urban households, with no statistically significant effect observed among rural households. In contrast, the positive effect on intergenerational economic support is more pronounced in urban households.

Discussion:

This study provides empirical evidence for understanding the relationship between community-based home care services and intergenerational family support. It delineates the differentiated effects and heterogeneous characteristics of community-based home care service utilization across distinct types of intergenerational support, yielding important policy implications for optimizing the elder care service system and fostering the coordinated development of social care and family-based care.

1 Introduction

The rise of community-based home care services, while bridging gaps in family-provided care, has also had far-reaching effects on intergenerational family support. As China’s socioeconomic development and demographic transition continue, population aging has become increasingly acute. By the end of 2025, the population aged 60 and above reached 323.38 million, accounting for 23.0% of the total population, while the population aged 65 and above reached 223.65 million, representing 15.9% of the total (47). China has entered a stage of moderate aging, and as household nuclearization and the prevalence of “empty- nest” households intensify, the scale of elder care demand continues to expand, and its composition becomes increasingly diverse (1). The traditional family-based care model faces mounting challenges stemming from shrinking household sizes and diminishing intergenerational support, rendering it increasingly inadequate to meet the growing complexity and multilayered nature of elder care needs, and thus necessitating broader societal engagement. In recent years, the Chinese government has introduced a series of policies that, while reaffirming the foundational role of family-based care, actively promote the development of community-supported home care. Against this backdrop, community-based home care services—by virtue of their proximity and accessibility—have emerged as a vital component of China’s elder care service system. Nevertheless, as professionalized social forces accelerate their involvement in family care provision, scholarly consensus has yet to be reached on whether intergenerational family support is strengthened by alleviating caregiving burdens or, conversely, weakened as elder care responsibilities shift outward from the family. A deeper investigation of this question not only contributes theoretically to advancing understanding of the “family–community” care nexus in the context of population aging, but also holds practical significance for informing the implementation of silver economy policies and refining a multi-tiered elder care service system.

Regarding the impact of community-based home care service utilization on intergenerational family support, existing research has generated extensive debate about crowding-out and crowding-in effects. However, a growing body of empirical evidence suggests that these two effects are not mutually exclusive; rather, they exhibit structural differentiation in opposing directions across distinct types of intergenerational support (2, 3). In the context of community-based home care services in particular, service provision may substitute for children’s instrumental care inputs while simultaneously reducing caregiving burdens in ways that facilitate emotional interaction and higher-quality companionship, thereby giving rise to a compound effect structure characterized by instrumental crowding out and emotional crowding in Sun et al. (4). This suggests that treating intergenerational support as a single, undifferentiated variable risks obscuring critical mechanistic heterogeneity in the context of formal service intervention. Intergenerational support theory, rooted in the sociological altruism model (5), is typically analyzed along three dimensions: economic, instrumental, and emotional support. With respect to economic support, existing studies have demonstrated that children’s financial transfers exert a significantly positive effect on older adults’ health and subjective well-being (6, 7); however, the extant literature has seldom directly examined, from a crowding-in perspective, how community-based service utilization affects intergenerational economic support through its influence on children’s labor time allocation. With respect to instrumental support, older adults’ demand for daily care increases steadily with age, and most exhibit a strong preference for family-based care arrangements (8); yet, given the substantial functional overlap between professional community services and children’s day-to-day caregiving, the crowding-out logic driven by substitution effects carries considerable empirical grounding along this dimension. With respect to emotional support, children’s psychosocial interaction with older adults is essential to the latter’s psychological well-being; in practice, however, emotional care is frequently overshadowed by material support (9, 10). More importantly, scholarly attention to the emotional support dimension remains relatively limited, with few studies subjecting it to systematic examination as an independent outcome variable (11).

Meanwhile, prevailing scholarship holds that the effective provision of community-based home care services should be grounded in the shared responsibilities of multiple stakeholders (12, 13). Service content has largely come to encompass such foundational domains as daily living assistance, healthcare, and psychological support (14, 15); nevertheless, a significant misalignment persists between supply and the actual needs of older adults (16, 17). Nevertheless, existing studies still have the following limitations. First, most scholars regard social care for older adults and family-based elder care as two independent systems, lacking an integrated analysis of their interactive mechanisms. Second, prior research predominantly focuses on macro-level policy effects, while insufficient attention has been paid to how the micro-level behavior of service utilization differentially affects the internal mechanisms of intergenerational support across various dimensions. Finally, there may be potential endogeneity between service utilization and intergenerational support, which has not been adequately addressed in most studies, further undermining the credibility of causal identification. Against this backdrop, this study draws on data from the 2020 Chinese Longitudinal Aging Social Survey (CLASS). It employs an Ordered Probit model in conjunction with instrumental variable estimation to disaggregate intergenerational support into three dimensions—economic, instrumental, and emotional—and systematically examines the effects of community-based home care service utilization on intergenerational family support and its underlying mechanisms, to provide empirical evidence and policy guidance for the development of an elder care service system suited to China’s national conditions.

Compared with existing studies, the marginal contributions of this research are evident in three respects. First, this study systematically uncovers the dimensional differentiation effects of community-based home care service utilization on intergenerational support by conducting a multidimensional decomposition of intergenerational support and separately evaluating the impact of service utilization across the economic, instrumental, and emotional dimensions. Second, by situating family-based and community-based care within a unified analytical framework and examining their interaction mechanisms from a micro-level perspective on service utilization behavior, this study addresses the limitations of prior research, characterized by the artificial separation of social care from family-based care and the disconnect between macro-level policy effects and household-level response mechanisms. Third, the application of instrumental variable estimation effectively addresses the endogeneity of service utilization; further examination of the moderating role of co-residence patterns and the heterogeneous effects of the urban–rural dual structure enhances the credibility of causal identification and deepens the understanding of differentiated pathways through which the “family-community integration” model of elder care may be realized.

The remainder of this paper is structured as follows. Section 2 presents the theoretical analysis and research hypotheses, drawing on relevant theories to examine how utilization of community-based home care services affects intergenerational family support and to derive the study’s hypotheses. Section 3 outlines the research design, detailing the data sources, model specification, and variable definitions. Section 4 reports the empirical results and analysis, including baseline regression, endogeneity issues, robustness tests, and further analysis. Section 5 discusses the effects of community-based home care service utilization on intergenerational family support, elaborating on the study’s arguments in light of the empirical findings and laying the groundwork for the conclusions. Section 6 presents the conclusions and policy recommendations, summarizing the key findings and proposing corresponding policy implications.

2 Theoretical analysis and research hypotheses

The intervention of community-based home care services fundamentally alters the allocation of time, income, and emotional resources within the family. Its impact on intergenerational family support is not unidirectional but rather constitutes a reallocation process shaped by the joint operation of multiple mechanisms. Existing studies have largely treated intergenerational support as a single, undifferentiated variable; however, a growing body of evidence suggests that distinct types of intergenerational support may exhibit divergent, even opposing, trajectories in response to service intervention. Against this backdrop, this study disaggregates intergenerational support into three dimensions—economic support, instrumental support, and emotional support—and, drawing on Becker's (18) household decision-making model and Grossman's (19) health demand model, together with social psychological theory, constructs an integrated analytical framework encompassing three mechanisms: resource release, functional substitution, and emotional motivation. Specifically, community-based home care services operate primarily through two economic rationality mechanisms and one social psychological mechanism: first, the resource release mechanism, whereby service provision substitutes for a portion of household caregiving inputs, thereby freeing up children’s time and influencing their labor supply and income levels; second, the functional substitution mechanism, whereby the functional overlap between community services and children’s caregiving leads households to engage in cost-efficiency trade-offs across different modes of care provision; and third, the emotional motivation mechanism, whereby service intervention alters individuals’ perceptions of family obligations and emotional needs, thereby influencing intergenerational interaction behaviors. These three mechanisms operate differently across distinct support dimensions and collectively constitute the theoretical foundation of this study. Therefore, this study constructs a theoretical analytical framework to elaborate the relationship between community-based home care service utilization and family intergenerational support, as shown in Figure 1.

Theoretical analytical framework for the impact of community-based home care service utilization on intergenerational family support.

2.1 Community-based home care service utilization and intergenerational economic support

Becker’s household decision-making model (18) conceptualizes the family as a unified unit of production and consumption, in which household members maximize family utility by allocating time and material resources. Central to this model is the insight that the distribution of resources within the household is subject to dual constraints of time and income, compelling family members to make trade-offs between labor market participation and family caregiving. Within this framework, the time children devote to parental care directly affects their labor supply and income-earning capacity, thereby further constraining the level of intergenerational economic support they can provide. In the absence of community-based home care services, children must dedicate substantial time to their parents’ daily care, which crowds out their market work hours, reduces their wage income, and, to some extent, diminishes their capacity to provide economic support to their parents. When community-based home care services are introduced and used, they assume some of the caregiving functions previously performed by children. This substitution process, operating through the resource release mechanism, effectively alleviates children’s time constraints and enables them to allocate more time to labor market participation. Holding wage rates constant, the increase in market working hours directly raises children’s wage income (20). Under the household budget constraint, higher income implies an expansion of disposable household resources; driven by altruistic motives or intergenerational exchange mechanisms, a substantial share of this incremental income is likely to be channeled into transfers from children to parents, thereby generating an income effect. Concurrently, the utilization of community-based home care services may itself entail certain financial costs, and as older adults attain a degree of material security through service provision, the perceived necessity of children’s economic transfers may correspondingly diminish, giving rise to a partial substitution effect. Nevertheless, compared with the aforementioned substitution mechanism, the income growth pathway generated by resource release is more direct and sustained; particularly within an institutional context in which the family continues to bear primary responsibility for elder care, increases in children’s income do not typically translate into a complete displacement of their filial economic support obligations (21). In sum, community-based home care service utilization alleviates household caregiving burdens. It releases family members’ labor time, thereby facilitating the rational reallocation of resources within the household, enhancing overall household welfare, and ultimately strengthening intergenerational economic support. Accordingly, the following research hypothesis is proposed.

H1: Community-based home care service utilization positively promotes intergenerational economic support by releasing children’s labor time and increasing household income.

2.2 Community-based home care service utilization and intergenerational instrumental support

Grossman’s health demand model (19) conceptualizes health as a form of human capital that households produce and maintain through investment in time and market goods. Extending this model to the elder care context, older adults’ capacity for independent living may be understood as a form of “welfare capital” that requires continuous production, with its generation depending on multiple inputs—encompassing both intergenerational instrumental support from within the family and professional care services provided by the community. From the perspective of the technical characteristics of the production function, community-based home care services and children’s intergenerational support exhibit a high degree of functional overlap along the instrumental dimension, as both primarily aim to meet older adults’ basic needs in areas such as daily living and household assistance. Within a household utility maximization framework, different modes of care provision are clearly substitutable, and rational households will select the optimal input combination subject to cost and efficiency constraints. Given that community professional services typically benefit from economies of scale and a specialized division of labor, their output efficiency per unit cost tends to exceed that of children’s non-professional, fragmented caregiving. Consequently, households are inclined to substitute community-based home care services for intergenerational instrumental support—that is, to reduce children’s caregiving inputs and rely instead on community services (22). In other words, as the frequency and coverage of community-based home care service utilization increase, the probability and frequency with which children provide instrumental support, such as household help and daily care, decline significantly. Older adults themselves may also proactively reduce their reliance on children’s care as they become accustomed to the convenience and professionalism of formal services, giving rise to “demand-side substitution” and resulting in a net reduction in intergenerational instrumental support at the household level (23). In sum, as the accessibility and professionalism of community-based home care services continue to improve, older adults gradually become more reliant on these services and reduce their demand for instrumental care from children. Accordingly, the following research hypothesis is proposed.

H2: Community-based home care service utilization reduces intergenerational instrumental support through a substitution effect, attributable to its functional overlap with children’s caregiving.

2.3 Community-based home care service utilization and intergenerational emotional support

The impact of community-based home care service utilization on intergenerational emotional support is examined through two theoretical lenses: the responsibility dilution hypothesis and socioemotional selectivity theory. On the one hand, the responsibility dilution hypothesis holds that when a task or obligation can be shared among multiple actors, each actor’s perceived personal responsibility diminishes. In the elder care context, once community-based care services intervene, caregiving responsibility expands from the family as a sole provider to a pluralistic “family–community” arrangement. Children may subconsciously perceive that their parents are already receiving care and companionship from service providers and, consequently, feel justified in moderating their own emotional investment (24). This dilution of responsibility, however, does not necessarily reflect indifference on the part of children; rather, it represents an unconscious psychological adjustment. On the other hand, drawing on socioemotional selectivity theory (25, 26), when community-based home care services can provide stable, reliable emotional comfort, older adults’ psychological need for emotional dependence on their children may diminish. They may partially redirect their emotional expectations toward service providers, thereby reducing how often they actively seek emotional support from their children. In this process, the emotional demand signals that children receive weaken, further attenuating their motivation to invest emotionally. In sum, the proliferation of community-based home care services may, through the responsibility dilution mechanism, erode children’s sense of filial obligation and lead them to redirect greater energy toward personal pursuits; the socioemotional selectivity mechanism further reinforces older adults’ reliance on external care services, thereby weakening their emotional bonds with children. The joint operation of these two mechanisms may result in a net reduction in intergenerational emotional support following service utilization. It should be noted that such a reduction does not necessarily signify a deterioration in intergenerational relations. Still, it may instead reflect a reconfiguration of caregiving responsibilities and a diversification of channels for meeting emotional needs. Accordingly, the following research hypothesis is proposed.

H3: Community-based home care service utilization reduces intergenerational emotional support through the mechanisms of responsibility dilution and socioemotional selectivity.

3 Research design3.1 Data sources

The data for this study are drawn from the 2020 China Longitudinal Aging Social Survey (CLASS). CLASS is a nationally representative longitudinal survey of the aging population, employing a stratified, multi-stage probability sampling design that covers 28 provinces, autonomous regions, and municipalities. The survey encompasses data on the socioeconomic conditions, health status, family circumstances, and social support of adults aged 60 and above, providing a reliable data foundation for the present study. To ensure sample validity and analytical reliability, the raw data were subjected to a sequential screening procedure as follows (see Figure 2). In the first step, an initial sample screening was conducted. The raw sample of the 2020 CLASS dataset comprised 11,398 observations; after excluding invalid samples with refused or unavailable responses, 10,177 observations remained. In the second step, observations with missing values on core variables were removed. The core dependent variables are intergenerational economic, instrumental, and emotional support, while the core independent variable is community-level utilization of elder care services. After excluding observations with missing or anomalous values on any of these four variables, 6,782 observations remained. In the third step, observations with missing values on key control variables were addressed. Key control variables include gender, educational attainment, pension insurance coverage, self-rated health, activities of daily living (ADL), number of chronic conditions, marital status, number of children, and number of co-residing household members. Observations with a substantial number of missing values on control variables were removed. In contrast, those with relatively few missing values were imputed using multiple imputation. Finally, a final analytic sample of 5,220 observations was obtained for the subsequent empirical analysis.

Schematic diagram of the sequential screening procedure for the analytic sample.

3.2 Variable description3.2.1 Independent variables

Community-Based Home Care Service Utilization. The core measure for this variable is derived from the survey question: “Over the past year, have you used any elder care services provided by the community?” Older adults who reported having utilized such services were assigned a value of 1, while those who had not were assigned a value of 0.

3.2.2 Dependent variables

Intergenerational Support. Drawing on the classification of intergenerational support proposed by Wang et al. (27) and integrating Maslow’s hierarchy of needs, this study disaggregates intergenerational support into three dimensions: economic, instrumental, and emotional. Specifically, intergenerational economic support is measured by the survey question: “Over the past year, how much cash or in-kind support did your children provide to you in total?” Intergenerational instrumental support is captured by the survey question: “Over the past year, how frequently did your children help with household chores or assist with your daily care in total?” Intergenerational emotional support is assessed by the survey question: “Over the past year, how frequently did you communicate with your children in total, through means such as in-person visits, telephone calls, or the internet?” It should be noted that the level of intergenerational support recorded here reflects the aggregate support provided by all children. For older adults with multiple children, the level of support each child provides is first calculated, then either summed or averaged across all children.

3.2.3 Control variables

Beyond the utilization of community-based home care services, a wide range of factors may influence intergenerational family support. To minimize estimation bias arising from omitted variable bias, this study draws on prior research (28, 29) and selects a set of control variables at both the individual and household levels. At the individual level, the following variables are included: gender, level of education, presence of pension insurance, self-rated health, activities of daily living (ADL), and number of chronic conditions, to capture both the subjective perceptions and objective circumstances that shape older adults’ demand for elder care services. At the household level, marital status, number of children, and number of household members are included to account for the potential influence of intra-household human resource allocation and structural characteristics on elder care decisions. Variable definitions and descriptive statistics are presented in Table 1.

VariableVariable nameAbbreviationsDefinitionMeanS.D.NIndependent variableIntergenerational economic supportIECSTotal value of material or cash support provided by children over the past year: Less than 1,000 yuan = 1; 1,000 ~ 2,000 yuan = 2; 2,000 yuan and above = 31.9900.8655,220Intergenerational instrumental supportIINSFrequency of children providing domestic assistance and personal care over the past year: Rarely = 1; Occasionally = 2; Frequently = 32.2150.8295,220Intergenerational emotional supportIEMSFrequency of contact with children over the past year: Rarely = 1; Occasionally = 2; Frequently = 32.3960.5295,220Dependent variableCommunity-based home care service utilizationCBHCSOver the past year, have you utilized any community-based elder care services: Yes = 1; No = 00.1080.3105,220Instrumental variableCommunity-based home care provisionCBHCPDoes the community provide social care for older adults: Yes = 1; No = 00.4860.5005,220Control variablesGenderGenderMale = 1; Female = 00.5040.5005,220Level of educationEduNo education = 1; Primary or literacy education = 2; Education beyond primary level = 32.1650.7585,220Presence of pension insuranceInsuranceYes = 1; No = 00.8810.3245,220Self-rated healthHealthHealthy = 1; Fairly well = 2; Unwell = 32.2680.7255,220Activities of Daily LivingADLFully independent = 0; Requires some assistance = 1; Unable to self-care = 20.7910.4075,220Number of chronic conditionsChronic0 item = 1; 1 item = 2; 2 items = 3; 3 items and above = 42.7311.2325,220Marital statusMaritalNo spouse including widowed = 0; With spouse = 10.7550.4305,220Number of childrenChildren0 item = 1; 1 item = 2; 2 items = 3; 3 items = 4; 4 items and above = 53.3161.0755,220Number of household membersHousehold0 item = 1; 1 item = 2; 2 items = 3; 3 items = 4; 4 items and above = 53.5360.9625,220

Variable definitions and descriptive statistics.

3.3 Model specification3.3.1 Ordered probit model

Given that the dependent variable “intergenerational family support” is a discrete ordinal variable, direct application of OLS regression would yield biased parameter estimates. Accordingly, this study employs the Ordered Probit model for baseline estimation, thereby circumventing the parameter-estimation bias and heteroskedasticity problems that arise from treating ordinal variables as continuous. To examine the effects of community-based home care service utilization on intergenerational economic support, intergenerational instrumental support, and intergenerational emotional support, the following model is specified:where denotes the latent variable for individual on the dimension of intergenerational support (economic support, instrumental support, and emotional support); is the corresponding observed dependent variable; indicates whether individual utilizes community-based home care services; is a vector of control variables; , , and are estimated coefficients; is a random error term following a standard normal distribution; and denotes the threshold parameter.

3.3.2 IV ordered probit model

The core dependent variables in this study (intergenerational economic support, intergenerational instrumental support, and intergenerational emotional support) are all ordinal polytomous variables. In contrast, the core independent variable (community-based home care service utilization) is a binary choice variable. A potential bidirectional causal relationship between these variables may give rise to endogeneity in the model specification described above. Specifically, whether older adults utilize community-based home care services may be influenced by unobservable factors such as health status, economic conditions, and children’s willingness to provide support—factors that may simultaneously affect both the level and composition of intergenerational support, thereby introducing bias into the estimates yielded by the conventional Ordered Probit model. To accurately identify the causal effect of community-based home care service utilization on intergenerational support, this study employs a combined instrumental-variables and conditional mixed process (CMP) approach to address endogeneity in the baseline regression (30). This study selects “whether the community provides elder care services” as the instrumental variable for “whether the individual utilizes such services.” This variable is measured by the CLASS survey question: “Does the community in which you reside provide elder care services?” with communities that provide such services coded as 1 and those that do not coded as 0. This variable reflects the coverage of elder care services at the community level. On the one hand, the availability of community services directly influences the likelihood that older adults will use them; on the other hand, community-level service provision decisions are relatively independent of individual household intergenerational support behavior. The proposed instrumental variable, therefore, satisfies both the relevance and exogeneity conditions.

4 Empirical results and analysis4.1 Baseline regression analysis

Based on Equations 1, 2, we obtain the regression estimation results of the ordered probit model (see Table 2). With respect to intergenerational economic support, the results in column (1) of Table 2 show that the coefficient on community-based home care service utilization is significantly positive at the 1% level. This indicates that, holding other factors constant, older adults who use community-based home care services are significantly more likely to receive higher levels of intergenerational economic support than their non-users. This finding confirms a significant positive effect of community-based home care service utilization on intergenerational economic support, consistent with Han et al. (31) and thereby validating Hypothesis 1. A plausible explanation is twofold: on the one hand, service utilization releases children’s labor time and increases household income, thereby enhancing children’s capacity to provide economic support to their parents; on the other hand, service utilization alleviates household caregiving burdens, enabling families to channel more resources into productive activities and generating a virtuous cycle of strengthened intergenerational economic support.

VariableIntergenerational economic supportIntergenerational instrumental supportIntergenerational emotional support(1)(2)(3)CBHCS0.528*** (0.061)−0.094*** (0.036)−0.087** (0.042)Gender−0.148*** (0.053)−0.210*** (0.060)−0.402*** (0.083)Edu0.378*** (0.037)0.068*** (0.024)0.226*** (0.057)Insurance−0.112 (0.082)−0.034 (0.092)0.548*** (0.108)Health0.272*** (0.044)0.182*** (0.046)−0.215*** (0.058)ADL0.156*** (0.039)0.103** (0.049)−0.176** (0.069)chronic0.074*** (0.024)0.062** (0.028)−0.081* (0.042)Marital0.135** (0.063)−0.642*** (0.078)−0.235** (0.099)Children0.034 (0.026)−0.258*** (0.030)−0.350*** (0.039)Household0.106*** (0.027)0.905*** (0.038)0.627*** (0.051)Constant−1.623*** (0.192)−1.422*** (0.223)0.758*** (0.288)N5,2205,2205,220

Baseline regression results for the effect of community-based home care service utilization on intergenerational family support.

*, **, and *** denote statistical significance at the 10, 5, and 1% levels, respectively; robust standard errors are reported in parentheses.

Regarding intergenerational instrumental support, the results in column (2) of Table 2 show that the coefficient on community-based home care service utilization is −0.094, indicating that utilization of community-based elder care services is associated with a significant 9.40 percentage point reduction in the probability of receiving higher levels of intergenerational instrumental support. This finding demonstrates that community-based home care service utilization exerts a significant crowding-out effect on intergenerational instrumental support, consistent with the conclusions of Wang et al. (32), thereby validating Hypothesis 2. A plausible explanation lies in the substantial functional overlap between community-based home care services and children’s instrumental caregiving: when professional community services can meet older adults’ daily living needs, children correspondingly reduce their direct caregiving. This crowding-out effect lends support to the “functional substitution” hypothesis (33).

Regarding intergenerational emotional support, the results in column (3) of Table 2 show that the coefficient on community-based home care service utilization is −0.087, indicating that utilization of community-based elder care services is associated with a significant 8.70 percentage point reduction in the probability of receiving higher levels of intergenerational emotional support. This finding confirms that community-based home care service utilization likewise exerts a crowding-out effect on intergenerational emotional support, thereby validating Hypothesis 3. The underlying mechanism may be attributed to “responsibility dilution.” Once community services intervene, service utilization attenuates children’s sense of urgency about their caregiving obligations to their parents, thereby reducing the frequency of emotional contact. Concurrently, older adults may experience a diminished psychological need for emotional support from children due to the emotional comfort provided by service personnel (34). As the control variables are not the primary focus of this study, a detailed discussion of their effects is omitted here for brevity.

4.2 Endogeneity issues

Although the baseline regression results indicate that community-based home care service utilization has a significant effect on intergenerational family support, potential reverse causality and omitted variable bias may introduce endogeneity in the core independent variable, rendering the baseline regression coefficient estimates biased and inconsistent. Accordingly, this study employs instrumental-variables estimation using the IV Ordered Probit model to address potential endogeneity. Column (1) of Table 3 reports the first-stage regression results and endogeneity test statistics. The first-stage F-statistic is 46.28, exceeding the conventional threshold of 10, indicating the absence of a weak instrument problem. The estimated coefficient of community-level elder care service provision on service utilization is 0.387 and is statistically significant at the 1% level, satisfying the relevance condition for a valid instrument. Columns (2) through (4) of Table 3 report the IV Ordered Probit estimation results. The coefficient atanhrho is statistically significant at the 1% level, thereby rejecting the null hypothesis that community-based home care service utilization is exogenous and confirming the appropriateness of the selected instrumental variable. The estimation results reveal that the coefficients on community-based home care service utilization with respect to intergenerational economic support, instrumental support, and emotional support are each statistically significant at least at the 5% level and are somewhat larger than the baseline regression estimates. These findings demonstrate that, after accounting for endogeneity, community-based home care service utilization continues to exert significant and differentiated effects on intergenerational family support—effects that are in fact more pronounced—further corroborating the research hypotheses advanced earlier in this study.