No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Matching As An Econometric Evaluation Estimator
Abstract
This paper develops the method of matching as an econometric evaluation estimator. A rigorous distribution theory for kernel-based
matching is presented. The method of matching is extended to more general conditions than the ones assumed in the statistical
literature on the topic. We focus on the method of propensity score matching and show that it is not necessarily better, in
the sense of reducing the variance of the resulting estimator, to use the propensity score method even if propensity score
is known. We extend the statistical literature on the propensity score by considering the case when it is estimated both parametrically
and nonparametrically. We examine the benefits of separability and exclusion restrictions in improving the efficiency of the
estimator. Our methods also apply to the econometric selection bias estimator.
... He claimed that the program failed to generate the anticipated effect when the transfer level fell below the target level by using PSM. According to Heckman (1998), the success of PSM depends on two factors-the identification of non-beneficiary groups that are readily comparable and the isolation of outcome variables from other possible correlated variables (Heckman 1998). ...
... He claimed that the program failed to generate the anticipated effect when the transfer level fell below the target level by using PSM. According to Heckman (1998), the success of PSM depends on two factors-the identification of non-beneficiary groups that are readily comparable and the isolation of outcome variables from other possible correlated variables (Heckman 1998). ...
... The larger the common support, the better the matching. In fact, matching can be improved by dropping both treatment and control observations whose estimated propensity score is greater than the maximum or less than the minimum of the comparison group propensity scores (Heckman 1998). ...
In Bangladesh Social Safety Net (SSN) interventions are intended to achieve specific goals, e.g., reduction in poverty, decrease in early school dropout, and nutrition supplementation. This study investigates the effectiveness of SSN participation on less explored goals of child labor reduction and school retention in Bangladesh. Empirical evidence is provided to evaluate the success of existing programs to combat child labor and early school dropout. This study utilizes the Household Income and Expenditure Survey (HIES), 2016 of Bangladesh. We employ the Propensity Score Matching (PSM) framework to show the causal relationship between SSN participation and its impact on the use of child labor and school retention. Given the nature and availability of data, the matching method provides us an estimate of the impact of SSN based on a quasi-experimental setting. We estimate counterfactuals based on observed characteristics of the household that affect both SSN participation and our outcome variables. Our results suggest that households that receive SSN tend to employ less child labor domestic or paid compared to their counterparts. Our Propensity Score Matching (PSM) results show that SSN participation reduces child labor prevalence in Bangladesh by 2.4% to 2.9%. This is an indication that families essentially substitute SSN income with child labor earnings. Thus, SSN programs in Bangladesh are effective in reducing child labor though the magnitude is not overwhelming. On the contrary, our results show that children from SSN-participating families have a higher incidence of early school dropout. This may be due to inefficiency or poorly designed programs that effectively keep the children in school. We estimate early school dropout is higher by 3.6% to 6.6% for households that receive SSN compared to the control group. Our findings are important for policymakers to redesign the programs that retain children in school. We suggest targeted interventions that will help to keep these children in school, which is conducive to human capital accrual and increasing lifetime earnings. Additionally, our insights will assist policymakers in rethinking SSN coverage, efficient selection of beneficiaries, and initiation of the correct intervention.
... When applying the double difference method to evaluate a fishing ban compensation policy, endogenous issues may cause estimation errors. Therefore, this study used the PSM-DID proposed by Heckman et al. (1998) [56] for estimation, by referring to other studies on biodiversity conservation policies [10,57]. First, PSM was performed on the samples of prohibited and non-prohibited groups, and a first-stage estimation was obtained. ...
... When applying the double difference method to evaluate a fishing ban compensation policy, endogenous issues may cause estimation errors. Therefore, this study used the PSM-DID proposed by Heckman et al. (1998) [56] for estimation, by referring to other studies on biodiversity conservation policies [10,57]. First, PSM was performed on the samples of prohibited and non-prohibited groups, and a first-stage estimation was obtained. ...
The conflict between biodiversity conservation and economic development is intense in developing countries. With the decline in biodiversity, the Chinese government imposed a 10-year fishing ban in the priority waters of the Yangtze River Basin, which resulted in many fishermen losing their livelihood. However, a compensation policy was subsequently introduced. To investigate the impact of the ban, we conducted a two-year study using balanced panel data and the PSM-DID model for 365 households in the Poyang Lake area and introduced fixed effects to discuss the impact of the Yangtze River fishing ban compensation policy on the income of returning fishermen and their choices for alternative income. We also explored the interactive effects of fishing ban compensation and other biodiversity conservation policies on their income. The results showed that the fishing ban negatively impacted the agricultural income of households but not the total household income, and the compensation somewhat subsidized the fishermen’s loss of income. Moreover, this was an opportunity for farmers to adjust their livelihood strategies. These findings improve our understanding of the impact of this fishing ban and compensation policy on the household incomes of returning fishermen and their adaptive strategies for alternative income.
... Since food safety variables that measure similar concepts can introduce a multicollinearity problem in the regression analysis, the study first used principal component analysis to combine multiple food safety statements into a single statement representing an FSA index. Following Child (2006) and Harman (1976), we retained the factor of eigenvalues greater than 1, and loading above 0.30 (in absolute terms) was considered as factoring together (see Appendix Table A2). Thus, the FSA variable for i th family is created by multiplying the j th variable with the respective loading factor, and summing across 14 variables (X): Figure 5 shows the density of the FSA index in the sample. ...
... -on-observables assumption commonly used in binary treatment (for example, Firpo 2007;Imbens 2004;Hirano, Imbens, and Ridder 2003;Heckman, Ichimura, and Todd 1998). According to , if the assignment to the treatment group is weakly unconfounded, given the covariates Z, then it also is weakly unconfounded given the GPS. ...
Despite significant improvements in consumers’ food safety awareness, the literature provides little or no evidence of the impact of food safety awareness on the quantity of fluid milk consumers purchased. This is especially true for India’s consumers, where the economy is changing rapidly regarding food marketing, incomes, urbanization, and increased demand for food safety.The study is based on primary data from a survey of consumers from urban and peri-urban districts in India. The survey collected information on socioeconomic characteristics, dairy consumption, purchasing patterns, and food safety awareness. This study first creates a food safety awareness (FSA) index to measure consumers’ awareness of food safety attitudes and perceptions. It investigates the factors affecting consumers’ food safety awareness and FSA’s impact on fluid milk demand. The study uses a newly developed impact evaluation method, continuous treatment matching estimation, and dose-response functions (DRFs) to assess the impact of FSA on consumers’ fluid milk purchases. Findings show that educational attainment, sources of information on food safety, age, and gender of the head of the household significantly influenced consumers’ food safety awareness. Further, the study finds that food safety awareness positively impacts the demand for fluid milk. After controlling for milk prices and family income, Indian consumers aware of and adopt seven or more food safety habits or practices are likely to demand more fluid milk. Results are robust to specifications, income levels, and the location of households. The link between food safety awareness and the quantity of milk purchased implies a latent consumer demand for food safety. Any consumer awareness programs and policies supporting milk and its products’ hygienic production will likely increase demand for milk and dairy products. At a policy level, credible systems can be put in place for food certification and labeling, which enhance the availability of safe and hygienic food.
... (PSM), which is known for its robustness (Heckman, Ichimura, and Todd 1998). According to Table A1 in the Appendix, there are no significant differences in the balance test results between the treatment and control groups after matching. ...
Due to the combined pressures of reducing carbon emissions and enhancing high-quality international trade, the influence of environmental regulations on foreign direct investment has to be assessed. This study focused on China's low-carbon cities development and analysed city-level panel data from 2003 to 2019 using a multi-period difference-indifferences approach to investigate its impact on foreign direct investment. The findings suggest that the implementation of the low-carbon cities policy resulted in both a reduction in the number of foreign-invested enterprises and a less effective utilization of foreign direct investment. Furthermore, western China was more profoundly affected. Finally, some valuable recommendations are presented.
... The weights were inversely proportional to the distance between the treated and control group propensity scores. Following Heckman and Todd [32], the Kernel matching methods can be written as follows: ...
This study investigated the nexus among the profitability, sales, and willingness to pay (WTP) more for weather index-based agricultural insurance premiums of flower farmers. In addition, the effect of sociodemographic and farm characteristics of flower farmers on their WTP more for insurance premiums was also estimated. A total of 160 flower farmers were selected from Bangladesh using the purposive random sampling technique. Propensity score matching technique was employed to identify the sales difference depending on WTP for insurance premiums while the profitability differences of flower farmers were assessed from different points of view. A binary logistic regression model was used to estimate the effect of sociodemographic and farm characteristics of flower farmers on their WTP more for insurance premiums, while a Likert scale was used to identify the major problems faced by flower farmers. Flower growers are willing to pay a higher premium for insurance when their sales decline. Farmers with lower profitability show a greater WTP higher insurance premiums, whereas those with relatively higher profitability are less inclined to do so. Farmers’ WTP more for insurance premiums decreases with age, education, and farm area, while farmers’ WTP more for insurance premiums increases with experience, training, earning members, marigold farming, and consciousness about natural calamities, pests, and diseases. The most significant problems faced in flower production are high input costs, demand fluctuation, pest and disease attacks, price fluctuation, and loss of production. Thus, the introduction of crop insurance in flower farming may increase profitability and reduce the exposure to risks involved in flower farming. The involvement of younger and more trained farmers in flower farming will increase their WTP more for insurance premiums.
... In other words, the key matching assumption is that selection occurs only on observables. Thus, by construction, matching eliminates two of the three sources of selection bias identified by Heckman et al. (1998), namely the bias resulting from having different ranges of the matching covariates for treated and control samples, and the bias resulting from having different distributions of these covariates across their common support. The remaining source of bias, differences in unobservables across groups, is ruled out by the matching assumptions. ...
We use rich Portuguese data to analyse the relationship between the use of different management practices and worker pay in a large representative sample of firms. We find that the overall score on the use of management practices is significantly associated with both higher average wages and higher within-firm wage dispersion. The positive relationship between management practices and average pay is present throughout the wage distribution and for all occupational skill groups, but is stronger for workers higher up in the wage distribution and in higher-skilled occupations. These results are driven by incentives-based management practices, in particular practices related to reward and promotion of high performers in the firm. We also identify a potential moderating role of works councils. Our results suggests that, in firms with works councils, a larger share of the gains from productivity-enhancing management practices is shared with workers, and these gains are more evenly distributed across the workforce.
... However, the pilot of the MSSR may not be a random event, which would not satisfy the parallel trend assumption and thus lead to biased estimates of policy effects. This paper draws on the PSM-DID model proposed by Heckman et al. (1998) to study the effect of MSSR on farmers' maize productivity. ...
Introduction
China is the largest producer, consumer, and trader of grain. Changes in China’s agricultural policies will affect global food trade and thus impact food security. In this paper, we use China’s maize subsidy system reform (MSSR) as a quasi-natural experiment to investigate the impact of market-oriented reforms in price support policy on the productivity of grain.
Methods
We use official Chinese government panel data on farm households and a PSM-DID model to overcome the endogeneity problem of policy change.
Results and discussion
The empirical results show that MSSR can increase maize productivity. The MSSR is divided into two phases: eliminating the maize purchase price and implementing maize producer subsidies. The policy effect of eliminating the purchase price exceeds the implementation of producer subsidies. Further analysis reveals that for farmers with a larger scale of cultivation, higher level of specialization, and higher degree of part-time employment, the MSSR enhances their productivity more significantly. In the high quartile, the MSSR reduces farmers’ productivity. In the low quartile, the MSSR raises farmers’ productivity, suggesting that the MSSR reduces the productivity differences among farmers. The results of our study suggest that market-based reform of price subsidies is an effective institutional arrangement to mitigate resource mismatch and increase food productivity, and point to the need to continue to improve the MSSR, explore diversified maize producer subsidy policies, and take into account the impact of other subsidies on farmers’ maize production behavior.
... The second method is Propensity Score Matching (PSM), a statistical technique used in observational studies to estimate the causal effects of interventions, policies, or treatments (Heckman et al., 1998). This method is particularly relevant in scenarios where randomized controlled trials are not feasible (Curto-Grau, 2017;Cusimano et al., 2021), such as urban planning interventions like the Superblocks in Barcelona. ...
... The propensity score matching method is adopted to eliminate the bias caused by the significant differences in corporate characteristics between those in the non-pilot areas and the pilot areas [15]. The covariates used in predicting the propensity scores are all control variables that reflect the firm characteristics. ...
To achieve the carbon goals, the Chinese government initially implemented the carbon ETS in 2013 in 7 pilot provinces and cities. Using the firm-level financial and management data of the A-share listed companies in the 30 provinces of mainland China from 2008 to 2020, this paper examines the ETS impact on corporate financialization by constructing a DID model. The result supports the “crowd-out” effect that the implementation of ETS decreases corporate financialization and this negative impact is weaker on the state-owned firms, located in the eastern region of China, and are not in the manufacturing industry. These findings imply that other than the original target to reduce carbon emissions, the ETS, by its market-based nature, is effective in reducing the risk of over-financialization.
... Fonte: elaborazione Ismeri Europa su dati di monitoraggio, ASIA e COB Per ovviare e risolvere questo problema, ristimiamo il modello DDD pesando le osservazioni con pesi generati da una stima di propensity-score matching (PSM) 7 di tipo Kernel (PSM-DDD) in modo da bilanciare le imprese dei gruppi di controllo su quelle dei trattati sfruttando le caratteristiche predeterminate delle imprese presenti nel dataset e che sono state inserite nel bando come discrimini per acquisire punteggio nel bando della policy analizzata. In questo modo affrontiamo il problema del selection bias da caratteristiche non osservabili e produciamo una stima dell'impatto della policy sui trattati (ATT) double robust (tra gli altri guardare Heckman et al.1997Heckman et al. , 1998Blundell e Costa Dias, 2009) e, quindi, consistente poiché basta che uno dei due modelli di stima rispetti le proprie assunzioni per far convergere la stima rendendola unbiased 8 . ...
... Econometric analysis and impact of urban agriculture on the food security status of households According to the literature, impact assessment studies based on cross-sectional data were evaluated by the propensity score-matching model. This situation holds if the households have similar characteristics except for the treatment variable (Heckman et al., 1998). Similarly, in this study, the propensity score matching (PSM) model assessed the impact of urban agriculture on food security. ...
Article type: With the rise in capital wealth and human populations, cities need more food input. This study assessed the impact of urban agriculture on improving food security in four town administrations of Gurage Zone. Qualitative and quantitative data were collected from primary and secondary sources. The primary data was collected from 340 sampled adopters using interviews, focus group discussions, and observation. Chi-square and t-tests enabled comparisons among percentages and mean differences between adopters and non-adopters of urban agriculture. The propensity score matching (PSM) model enabled an assessment of the urban agricultural impact on food security. The statistical analysis revealed a statistically significant mean difference between adopters and non-adopters in job status, perception toward urban agriculture, and family size. The logit model showed that family size, ownership of living home, job status of HH head, perception toward urban agriculture, market demand, and training and support significantly determined the adoption of urban agriculture. The results of PSM indicated that adopting urban agriculture improved food security. Thus, the likelihood of being food secure would increase by a factor of 0.60, ceteris paribus. The study suggested that education, urban agriculture, and cooperative offices can teach urban dwellers via an integrated functional adult education program for training purposes and support them more in cooperative organization. The adoption of urban agriculture helps to win the battle against food insecurity.
... To some extent, the selection of urban regeneration projects is not a perfect quasiexperiment. The Propensity Score Matching (PSM) is a dominant way to avoid selection bias on account of observable variables and capture more reliable average treatment effects [86]. Its basic idea is to match individuals in treatment groups to those in nontreatment groups that have the most similar covariables [87]. ...
Urban regeneration is a sound, sustainable urban development strategy globally. In China, promoting urban regeneration has become the national sustainable urban strategy. Under the resource constraint, it is necessary to understand what benefits different types of urban regeneration projects can contribute to urban development. Much research has contributed to evaluating the benefits of urban regeneration on the project scale. The systematic investigation of their effects on surrounding economic vitality, especially the comparison between different types of projects, is relatively downplayed. This paper aims to evaluate and compare the effects of urban regeneration projects on surrounding economic vitality by calculating the change in housing prices. Chongqing is selected as the case city. Based on the housing transaction data from 2015 to 2021, a staggered difference-in-difference method is employed to capture the results. The finding shows that the overall effects are significantly negative, especially within a 400–800 m radius. Higher investment or better project location relates to stronger negative effects. Moreover, the differences in effects among the three types of urban regeneration projects are clearly revealed from diverse perspectives. It provides a valuable reference for policymakers and urban planners to make urban regeneration planning better by considering comprehensive benefits.
... Within the domain of impact evaluation, the propensity score matching (PSM) technique has found extensive application across diverse fields. For instance, Heckman et al. (1998), Lechner (1999), Dehejia and Wahba (2002) and Smith and Todd (2005) utilized PSM to estimate the impact of the labour market and training programmes on income dynamics. Similarly, Jalan and Ravallion (2003) explored the efficacy of anti-poverty workfare programmes; Trujillo et al. (2005) investigated the impact of health insurance on medical-care participation; Lavy (2002) delved into the effect of performance incentives for teachers on student outcomes and Persson et al. (2003) scrutinized the consequences of electoral reform on the prevalence of corruption. ...
The COVID-19 pandemic significantly impacted Bangladesh’s economy, leading to disruptions in supply chains and a decline in business activities. To counter these effects, the government introduced stimulus packages aiming to aid business recovery. This article evaluates the impact of these measures using a unique dataset from the Business Confidence Index survey conducted by the South Asian Network on Economic Modeling in April 2021. This survey covered 503 firms across eight divisions in Bangladesh. Using the Propensity Score Matching technique for causal estimation, findings suggest that firms that availed the stimulus packages demonstrated around 9 percentage points higher recovery than those that did not. Such insights are instrumental for future policy design and strengthening evidence-based policymaking.
... Outcome regression methods were first developed to estimate conditional effects accounting for covariate imbalance between treatment arms [23,24]. However, standard adjustment by parametric regression models is sensitive to model mis-specification [8]. ...
Controlling for confounding bias is crucial in causal inference. Causal inference using data from observational studies (e.g., electronic health records) or imperfectly randomized trials (e.g., imperfect randomization or compliance) requires accounting for confounding variables. Many different methods are currently employed to mitigate bias due to confounding. This paper provides a comprehensive review and tutorial of common estimands and confounding adjustment approaches, including outcome regression, g-computation, propensity score, and doubly robust methods. We discuss bias and precision, advantages and disadvantages, and software implementation for each method. Moreover, approaches are illustrated empirically with a reproducible case study. We conclude that different scientific questions are better addressed by certain estimands. No estimand is uniformly more appropriate. Upon selecting an estimand, decisions on which estimator can be driven by performance and available background knowledge.
... In order to mitigate the systematic divergence in the patterns of road infrastructure investment between villages with and without such investments, as well as to reduce the inherent selection bias of the DID method, the PSM-DID approach was employed for a robustness evaluation in this work. First, the Heckman et al. (1998) kernel technique for specific estimating was used to generate the propensity score, yielding the following estimates: ...
To investigate the effects of road infrastructure investments on rural household welfare, this study combines DID analysis with Propensity Score Matching (PSM) and Difference-in-Differences (DID). Initial DID results point to a non-significant impact of road infrastructure investment on household well-being; however, PSM-DID analysis yields different conclusions. The non-significant DID results show a possible mismatch between theoretical predictions and actual results, challenging preconceived notions and contradicting empirical data. Nonetheless, the next PSM-DID analysis shows that investments in road infrastructure have a notable and beneficial effect on household welfare, especially when it comes to real food spending per capita. These findings highlight the need to use reliable approaches to precisely evaluate the effects of infrastructure investments. Furthermore, well-being is favorably influenced by control variables including household business, education, and urbanization. Welfare, however, is adversely affected by the size of the home and the number of individuals residing there. These results underline the complexity of rural development and the need for more study to fully comprehend the intricate relationships between these factors.
... (16) Lastly, we selected for each individual in the treatment group, at least a non-treated individual with similar characteristics (the same propensity scores). (17) ...
Introduction: since 2008, Morocco’s Tayssir program has been a key public initiative aimed at combating school dropout rates, by offering conditional cash transfers to households with school-aged children, particularly targeting rural communities with high poverty rates. This initiative seeks to ensure equitable access to education, regardless of socioeconomic status, and boosted school attendance rates. Objective: to assess the impact of the Tayssir program on reducing school dropout rates in rural Morocco and to examine the effectiveness of targeting strategies and incentives provided to families. Methods: the study utilized cross-sectional data from the Household Survey Panel Data. Propensity score matching (PSM) techniques were employed to estimate the program’s impact on school dropout rates, comparing beneficiaries with a control group not participating in the program. Various statistical analyses were conducted to explore the characteristics of participants and to validate the logistic model used. Results: the propensity score matching analysis revealed a statistically significant reduction in school dropout rates among beneficiaries of the Tayssir program. The average treatment effect on the treated (ATET) demonstrated a decrease in dropout rates by approximately 43 % using one-to-one matching, 42,7 % with k-nearest neighbor, and 38,6 % via kernel matching methods. Furthermore, no significant gender differences were observed in the program’s impact. Conclusions: the Tayssir program has significantly contributed to reducing school dropout rates in rural Morocco, ensuring better access to education for children from disadvantaged backgrounds. The program’s effectiveness underscores the importance of targeted interventions and conditional cash transfers in promoting educational attainment. Future recommendations include expanding the beneficiary base, refining targeting mechanisms, and establishing a unified social registry to improve program governance.
... In this paper, the relationship between the implementation of EPT and urban CI is demonstrated using the DID model, which aims to estimate the effect of treatment on the causal effect by comparing the difference between the treatment and control groups before and after the treatment. However, there may be systematic differences in choices between the treatment and control groups, resulting in the estimated causal effects being subject to self-selection bias (Lyu et al. 2022), so the use of propensity score matching (PSM) is needed to reduce self-selection bias (Heckman et al. 1998). Specifically, the core idea of propensity score matching is to estimate the probability of each individual receiving the treatment by modeling the propensity score between the treatment and control groups. ...
In the context of increasingly severe global climate change, finding effective carbon emission reduction strategies has become key to mitigating climate change. Environmental Protection Tax (EPT), as a widely recognized method, effectively promotes climate change mitigation by encouraging emission reduction behaviors and promoting the application of clean technologies. Based on data from 282 cities in China, this paper takes the official implementation of the EPT in 2018 as the policy impact and the cities with increased tax rates for air taxable pollutants as the treatment group and uses DID model to systematically demonstrate the relationship between the implementation of the EPT and carbon intensity (CI) and further explores the possible pollutant emissions and green innovation mediating effects. The findings show that (1) the implementation of EPT can effectively reduce CI by about 4.75%, and this conclusion still holds after considering the robustness of variable selection bias, elimination of other normal effects, policy setting time bias, and self-selection bias. (2) The implementation of EPT can reduce CI by reducing pollutant emissions and improving the level of green innovation. (3) There is obvious regional heterogeneity in the carbon reduction effect of EPT, and the implementation of EPT has a more significant effect on CI in medium-tax areas, low environmental concern areas, general cities, and eastern regions. This paper not only provides a new analytical perspective for systematically understanding the carbon emission reduction effect of EPT but also provides policy insights for promoting regional green transformation and advancing carbon peak carbon neutralization.
... as an alternative would lead to selection bias. Thus, in this study, we estimated ATT using the following formula [38,39]: ...
Background
Social pensions, social assistance systems for older people in rural areas, have been put into place in many nations and have positively impacted health. The long-term health consequences of social pension programs in China are uncertain. The aim of this study is to evaluate the long-term health consequences of the new rural social pension (NRSP) for the rural older people in China.
Methods
Based on the 2011 and 2018 China Health and Retirement Longitudinal Study, we compared the scores on eight Health-Related Quality of Life (HRQoL) subscales of the rural older people before and after participation in the NRSP. The propensity score matching and difference-in-difference methods were used in data analysis. We also conducted a heterogeneity analysis for subgroups with different characteristics and pension enrolment times.
Results
The NRSP significantly enhanced scores on physical functioning, role-physical, and self-rated mental health of old rural participants by 1.90 (p < 0.01), 2.05 (p < 0.01), and 2.93 (p < 0.05), respectively. After excluding newly enrolled individuals, the beneficial health effects of the NRSP remained significant. There were no significant changes due to NRSP in the other five scores on the HRQoL subscale of the rural older people. The NRSP had more health benefits for older people in underdeveloped areas without formal schooling.
Conclusions
The NRSP reduced health disparities and had long-term benefits on the physical and mental health of the rural older people. We suggest continuously expanding the NRSP throughout rural China and further improving the social support system to enhance the overall quality of life of the rural older people. Comparable social pension programs aimed at underprivileged groups could also be conducted in other low- or middle-income nations.
... In order to establish a valid counterfactual, two matching assumptions must be satisfied: the Conditional Independence Assumption (CIA) and the assumption of Common Support (Heckman et al., 1998, Imbens & Wooldridge, 2009). The CIA requires that the observable characteristics of households that influence their participation in CCT (C) are conditioning factors. ...
This paper analyzes the effectiveness of the Conditional Cash Transfer (CCT) program in the Philippines in terms of its effect on conditionality goods as reflected by food, health, and education expenditures of households from rural and urban areas that benefitted the program. The CCT program in the Philippines is known as Pantawid Pamilyang Pilipino Program (4Ps) that seeks to address the problem on poverty by improving the socioeconomic status of poor households through targeted investments in health and education. This study used the Propensity Score Matching methodology in estimating the average treatment effect on the treated to capture the effect of CCT on conditionality goods. The study finds that CCT has a significant effect on education for household beneficiaries in rural areas and has improved the quality of food consumed by household beneficiaries in urban areas. Also, a decreased in the per capita total expenditure and per capita food expenditure of the household beneficiaries is revealed in urban areas driven by their improved saving behavior. Thus, the CCT program, at some point, is effective in meeting its short-term goal, but it must be more targeted in order to improve its impact on other conditionality goods.
... The value of the counterfactual effect of no treatment on the treated E Y D 1 X ( | , ) 0 = is approximated by the average result of the self-selected group of untreated individuals E Y D 0 X ( | , ) 0 = (Heckman et al., 1998). We employ the propensity score P X Pr D 1 X ( ) ( | ) i = = , namely the probability of belonging to the groups under analysis given some determined observed characteristics mentioned previously. ...
While policy evaluation is essential for improving labor programs, a significant knowledge gap exists regarding the impact of local labor market policies in developing countries. This article analyzes the impact of three employment programs in Chile aimed at enhancing employability and wages, namely Youth Employment Subsidy (SEJ), Women's Employment Subsidy (BTM), and Job or Skill Training (CT). We evaluate data from the CASEN Survey in 2015 and 2017 using matching techniques, presenting results for each program separately as they target different segments of the labor market. Fixed effects by territorial units, time effects, and geographic factors are all employed as controls per each matching procedure. For wages, variables such as the worker's productive sector, type of contract, and other socioeconomic controls are considered as well. Robustness analyses through different matching strategies are included. The results indicate that the SEJ and BTM have positive and significant effects on employability, while their impact on wages is null. As for CT, there are positive and significant effects on wages and employability, though with nuanced variations across territories. These findings are relevant in terms of effective labor market support programs, crucial for improving working conditions, narrowing gaps, and increasing employment opportunities.
... To enhance confidence in the evaluation of the DID model, the PSM-DID method is also employed to match the samples and test the robustness of the empirical results. PSM-DID was initially proposed by Heckman et al. (1998). The underlying assumption of the model is that the mean value can be disregarded, i.e.: ...
Rural areas in developing countries often grapple with energy shortages, low energy efficiency, and significant environmental pollution, all of which are interconnected with the issues of rural poverty. Policies geared toward rural green development are closely linked to energy transformation and economic growth. China's “Beautiful Countryside Construction” initiative prioritizes the enhancement of rural living conditions, the promotion of rural economic progress, energy transformation, and environmental amelioration through rural–urban integration. This study utilizes panel data encompassing 1,787 Chinese counties spanning from 2008 to 2015, employing the difference-in-difference (DID) method to analyze the economic, social, and environmental impacts of the “Beautiful Countryside Construction” program. The findings indicate that this initiative has contributed to the enhancement of the rural environment, the promotion of environmentally friendly economic development, and the advancement of rural energy transformation. These outcomes hold significant implications for how developing nations can design policies that achieve a harmonious balance between economic advancement and environmental preservation, fostering sustainable economic growth worldwide.
... The upper green region represents the treated support observations, the upper yellow represents treated off support, the lower blue shows untreated off support and the lower red indicates untreated on support. It is evident that the vast majority of observations are concentrated within the region's common support, verifying that there is enough overlap between treated and untreated components to effectively find matches (Heckman et al., 1998). This means that the assumption of common support has been satisfied ( Figure 2). ...
Women empowerment in agricultural endeavors is deemed to be a powerful solution for improving global sustenance. Thus, this survey seeks to analyze the impact that women’s participation in agriculture has on the food security of 336 households from Anna Sorraa District, Guji zone of Oromia Region, Ethiopia. Quantitative and qualitative data were collected using household surveys, focus group discussions, and data collection methods using key informant interviews. Statistical measures such as the Women Empowerment in Agriculture Index and household calorie availability were used for descriptive analysis, while binary logistic regression and Propensity Scores Matching provided insights through econometric modeling. Results indicated that 58.6% of women’s were disempowered. Household calorie intake statistics revealed that 53.9% of those surveyed had food insecure households, while 46.1% had food intake available to them. A regression analysis found most factors, including age, female head of household, female education level, Husband’s education and annual household income positively impacted women empowerments—with one exception being whether subject lived near a market or not. Propensity Scores Matching further demonstrated that households with empowered women consumed 560.89 kcal/AE/day compared to those without such an asset. Research concluded that women’s involvement in agricultural activities is a crucial factor to improve food security in this area. Consequently, it was strongly recommended that government and other stakeholders focus their attention on empowering women. Additionally, further research should be conducted to gain an even more comprehensive understanding of how important female empowerment really is.
Existing research has shown that political crisis events can directly impact the tourism industry. However, the current methods suffer from potential changes of unobserved variables, which poses challenges for a reliable evaluation of the political crisis impacts. This paper proposes a panel counterfactual approach with Internet search index, which can quantitatively capture the change of crisis impacts across time and disentangle the effect of the event of interest from the rest. It also provides a tool to examine potential channels through which the crisis may affect tourist outflows. This research empirically applies the framework to analyze the THAAD event on tourist flows from the Chinese Mainland to South Korea. Findings highlight the strong and negative short-term impact of the political crisis on the tourists’ intentions to visit a place. This paper provides essential evidence to help decision-makers improve the management of the tourism crisis.
In theory, market-based regulatory instruments correct market failures at least cost. However, evidence on their efficacy remains scarce. Using administrative data, we estimate that, on average, the EU ETS – the world's first and largest market-based climate policy – induced regulated manufacturing firms to reduce carbon dioxide emissions by 14-16% with no detectable contractions in economic activity. We find no evidence of outsourcing to unregulated firms or markets; instead, firms made targeted investments, reducing the emissions intensity of production. These results indicate that the EU ETS induced global emissions reductions, a necessary and sufficient condition for mitigating climate change. We show that the absence of any negative economic effects can be rationalized in a model where pricing the externality induces firms to make fixed-cost investments in energy-saving capital that reduce marginal variable costs.
Purpose
This paper aims to understand if buy-now-pay-later (BNPL) services, a digital type of credit that targets young consumers, acts as a protective or a risk factor for food insecurity among young consumers in Australia.
Design/methodology/approach
The study uses survey data from a representative sample of young consumers aged 18–24 from all internal states and territories in Australia. Propensity score matching is used to test two hypotheses: BNPL drives young consumers to food insecurity, and food insecurity leads young consumers to use BNPL.
Findings
There is evidence that BNPL use is driving young Australian consumers to experience food insecurity, but there is no evidence of food insecurity driving the use of BNPL services.
Practical implications
The evidence of BNPL driving young consumers to experience food insecurity calls for the adoption of practices and stronger regulation to ensure that young users from being overindebted.
Originality/value
Although the link with more traditional forms of credit (such as personal loans) and consumer wellbeing has been explored more broadly, this project is the first attempt to have causal evidence of the link between BNPL and food insecurity in a high-income country, to the best of the authors’ knowledge. This evidence helps to fill the gap about the protective or risky nature of this type of digital financial product, as experienced by young Australians.
Social programs are designed to reach beneficiaries and achieve expected objectives. There is a need to understand whether development programs work and their level of impact on the beneficiaries involved. Along these lines, the objective of this study is to evaluate the impact of Improved Technology adoption in Traditional Poultry Farming (ITTPF) on farmers’ potential outcomes in Togo. Baseline and follow-up data were collected from 400 farmers and analyzed using difference-in-differences models. The study reveals that five years after the implementation of the program, the annual gross profit increased on average by US$ 1294 for each program participant. The results of the heterogeneous impacts assessment indicate that participating in the program is a necessary condition for ITTPF adoption, but not sufficient for profit optimization. Overall, the program has a positive and significant impact on the potential outcomes of farmers in Togo. The government in its agricultural policy should mobilize more resources to enable considerably more farmers to adopt improved agricultural technologies. In addition, agricultural policymakers should implement the instruments of the chain planning, programming, budgeting, execution, monitoring and evaluation of all agricultural development programs and projects to make progressive adjustments for optimal results achievement and sustainable agricultural development.
High-quality development is the paramount task for comprehensively building a socialist modernized country. The low-carbon city pilot policy, with cities as the unit of action, provides new opportunities for high-quality economic transformation. Based on panel data from 261 cities between 2005 and 2018, this study calculates the level of high-quality economic development in Chinese cities, constructs a multi-period difference-in-differences model, analyzes the impact of the low-carbon city pilot policy on high-quality economic development, and explores the policy’s heterogeneous effects on high-quality development in different types of cities and its transmission mechanism. The research findings show that the low-carbon city pilot policy can significantly promote high-quality economic development in cities and has heterogeneous effects in terms of regional differences, city types, and city scale. The effects are relatively greater in the eastern region, non-resource-based cities, and mega-cities. The low-carbon city pilot policy promotes high-quality economic development through mechanisms such as technological progress effects, resource agglomeration effects, and government action improvement effects. Combining theoretical analysis with empirical results, this study proposes policy recommendations to enhance the effectiveness of the low-carbon city pilot policy in promoting high-quality development.
The aging population has been rising rapidly in every country, slowing the labour force and causing lower per capita growth. Many policies incentivize working in old age as it can alleviate the challenges of the aging population. This study examines the impacts of the extended maternity leave reform in Vietnam on elderly labour market outcomes. As grandparents' childcare responsibility may be reduced by maternal childcare following the extended maternal leave, especially for elderly living with children smaller than 1 year old, the reform gives older adults or grandparents more flexibility to join the labour market. Using the difference‐in‐differences model, we find that older people in the treatment group are more likely to return to the labour market following the reform implementation. The effect is positive and most robust for the self‐employment group rather than waged employment and for males rather than females. Our study suggests that there is a need for a policy design to shift public finance into the healthcare and pension system.
Research Question/Issue
The purpose of this paper is to analyze the long‐run determinants of the corporate structure of Italian firms to explain the persistent role of a long‐run tradition of civic capital that has favored interpersonal trust, fostered cooperation outside of the narrow ties of family members and limited the diffusion of family businesses managed predominantly by family members.
Research Findings/Insights
We examined a large sample of Italian listed and not listed firms and identified those that operate in current municipalities that in the past used to be independent communes. Such firms featuring experiences of civic engagement are today less likely to be owned by a family and run by family management.
Theoretical/Academic Implications
Our findings highlight the role of institutions as drivers of corporate governance and signal that long forgotten institutions, by modifying local social capital, may interact with family social capital and have important persistent effects on current corporate governance arrangements. Therefore, significant elements of path dependency may explain current patterns of unbundling of ownership and management.
Practitioner/Policy Implications
Persistent corporate governance structures are difficult even for policymakers to modify. Our findings suggest that political measures should favor the accumulation of social capital at the local level when aiming to change ownership and management arrangements and limit the misallocation of resources due to family management.
Many advocates highlighted place-based housing policies as a remedy for urban decay (Katz & Nowak, 2017) despite mixed evidence on their long-term efficacy (Neumark & Simpson, 2015). This paper is the first national-level study investigating changes in the trajectory of the housing market caused by Housing Opportunities for People Everywhere Programs (HOPE VI), the largest federal investment in public housing revitalization. I combine a stratified version of the synthetic difference-in-difference method (Arkhangelsky et al., 2021) with a stacked event study model to identify the effects of HOPE VI redevelopment on housing values and rents. The analysis reveals a 9 and 13 percent increase in housing values and rents in the redeveloped neighborhoods, respectively. These increases varied by program budgets and neighborhood characteristics. Moreover, I find a significant shift in housing supply towards lower density and a significant 6.5 percent increase in median household income. Coincident with these changes, poverty rates declined, and racial diversity increased in the redeveloped neighborhoods.
RÉSUMÉ
L’aménagement hydro-agricole de la Tapoa représente une opportunité d’amélioration des conditions de vie des populations riveraines mais également une cause de dégradation des écosystèmes naturels. Cette thèse a pour objectif d’analyser les incidences de l’aménagement hydro-agricole sur les écosystèmes et sa contribution à l’amélioration de la situation socioéconomique des populations du bassin versant de la Tapoa. La méthodologie adoptée a consisté en l’analyse diachronique d’images multi-dates de Landsat Thematic Mapper, Sentinel et Google Earth de 1975, 1998 ETM+ et OLI-TIRS de 2018. Plusieurs scènes ont été utilisées et une période d’écart de 43 ans répartie en deux périodes séquentielles a permis la comparaison de l’évolution des unités d’occupation des terres dans la zone humide aménagée de rayon de 15 Km et sur le bassin versant de la Tapoa. Les images ont été traitées à l’aide des logiciels ENVI 4.7 et ArcGis 10.3. La classification du Millenium Ecosystems Assessment et les enquêtes semi-directives auprès de 80 chefs de ménages ont permis d’analyser les services écosystémiques fournis et la perception des populations riveraines vis à vis de ces services écosystémiques. Les résultats ont révélé qu’entre 1975 et 2018, les savanes arborées et arbustives se sont réduites au profit de l’extension des cultures et du plan d’eau. Les services d’approvisionnement et les services de régulation y sont dominants et plus utilisés respectivement par les agriculteurs (42,86 %), les arboriculteurs (26,88 %) et les pêcheurs (25,81 %). Les services de soutien à la production et les services culturels sont moins représentés dans la zone humide de la Tapoa. L’usage de ces services écosystémiques par les populations riveraines participe à leurs stratégies d’adaptation aux changements climatiques. Les revenus annuels de 20,93 % des utilisateurs des services écosystémiques sont compris entre 100 000 FCFA et 200 000 FCFA, destinés à l’alimentation et aux manifestations sociales. Les populations ont une bonne connaissance des services écosystémiques fournis par les écosystèmes de l’aménagement hydro-agricole du bassin versant de la Tapoa. Ceux-ci se déclinent dans les bienfaits procurés par les écosystèmes et les avantages créés dans le développement de leurs activités socioéconomiques. Un grand nombre de la population reconnait leurs qualités « Très Bonne et « Bonne » mais il s’est avéré qu’ils subissent une dégradation progressive en raison du manque d’entretien (95%), la pression du cheptel (85%) et la baisse des pluies (63,33%). De ce fait, 58,33% des populations sont disposées à contribuer à la sensibilisation, 33,33% aux travaux physiques et seulement 10% pour une contribution financière pour la préservation des écosystèmes en présence. Au regard des effets environnementaux consécutifs à la réalisation de cet aménagement du bassin versant de la Tapoa et de la valeur des écosystèmes et des services écosystémiques rendus, sa gestion participative contribuera durablement à l’amélioration des conditions socioéconomiques des populations locales.
Mots-clés : Aménagement hydro-agricole, unités d’occupations des terres, Services écosystémiques, perception, bassin versant, Tapoa
ABSTRACT
The mpact of hydro-agricultural development on ecosystems and its contribution to improving the socio-economic situation of the inhabitants of the Tapoa watershed
Hydro-agricultural layout of the Tapoa represents an opportunity to improve the living conditions of the local populations but also a source of degradation of the natural ecosystems. The aim of this thesis was to analyze the impact of hydro-agricultural development on ecosystems and its contribution to improving the socio-economic situation of the inhabitants of the Tapoa watershed. The methodology deployed consisted in the diachronic analysis of multi-date images of Landsat Thematic Mapper, Sentinel and Google Earth from 1975, 1998 ETM+ and OLI-TIRS from 2018. Several scenes were used and a 43-year gap period divided into two sequential periods allowed comparison of the evolution of land use units in both the 15 km radius developed wetland and the Tapoa watershed. The images were processed using ENVI 4.7 and ArcGis 10.3 software. The classification of the Millennium Ecosystems Assessment and the semi-directive surveys of 80 householders made it possible to analyze the ecosystem services provided and the perception of local populations with regard to these ecosystem services. The results revealed that between 1975 and 2018, the tree and shrub savannahs were reduced in favor of the extension of crops and the body of water. Supply and regulation services are predominant and more used respectively by farmers (42.86 %), arborists (26.88 %) and fishermen (25.81 %). Production support and cultural services are less represented in Tapoa wetland. The use of these ecosystem services by local populations contributes to their adaptation strategies against climate change. The annual income of 20.93% of users of ecosystem services is between 100,000 FCFA and 200,000 FCFA, used for food and social events. The populations have a clear understanding of the ecosystem services provided by the ecosystems of the hydro-agricultural development of the Tapoa watershed. These are expressed in the benefits provided by the ecosystems and the advantages created in the development of their socio-economic activities. A large number of the population recognize their qualities as “Very Good” and “Good”, but it turned out that they are undergoing progressive degradation due to lack of maintenance (95%), livestock pressure (85%) and the decrease in rainfall (63.33%). As a result, 58.33% of the population is willing to contribute to sensitization, 33.33% to physical work and only 10% for a financial contribution for the preservation of the existing ecosystems. In view of the environmental effects resulting from the realization of this hydro-agricultural layout of the Tapoa watershed and the value of the ecosystems and ecosystem services generated, its concerted management will contribute sustainably to the improvement of the socio-economic conditions of the local populations.
Key words: Hydro-agricultural development, land occupation units, ecosystem services, perception, watershed, Tapoa
Purpose
The purpose of this study is to examine the relationship between the corporate social responsibility (CSR) performance of job-hopping executives at their former and subsequent firms.
Design/methodology/approach
We conduct regression analyses using a sample of firms listed on the Shanghai and Shenzhen Stock Exchanges from 2010 to 2020 to examine whether CSR performance is similar from one firm to the next as executives switch jobs.
Findings
We find a positive relationship between the CSR performance of former and subsequent firms under job-hopping executives. This relationship is the strongest in the year of the job switch; it weakens in the second year and eventually disappears in the third year. In addition, we show that this relationship benefits different CSR stakeholder groups and is contingent on executive and subsequent firm attributes and job-hopping characteristics. Furthermore, we demonstrate that firms that hire a new chief executive officer from a firm with a strong track record in CSR, the new firm experiences a significant surge in CSR performance compared with firms that do not experience such a shock.
Practical implications
This study has implications for executive hiring decisions.
Originality/value
This study extends the understanding of CSR determinants through the lens of inter-organisational ties associated with job-hopping executives.
Globalization is motivating many emerging market firms (EMFs) to expand operations through outward foreign direct investment (OFDI). General FDI theories, such as internalization theory and the OLI paradigm, are based on industrialized countries, leading to inevitable gaps in explaining the OFDI activities of EMFs. We extend the literature by establishing a connection between international finance and the internationalization of EMFs. We argue that foreign banks operating in emerging markets offer a diverse set of resources that can help EMFs expand globally. Under gradually easing foreign bank entry restrictions in China, utilizing a panel dataset of Chinese listed companies spanning 2001–2018, we find a positive association between foreign bank entry and the OFDI activities of Chinese companies. The identified channels include direct loan support from foreign banks, information support from foreign banks’ global branch networks, a high level of internationalization, and close proximity to local companies, as well as loan support from domestic banks under foreign bank competition. The entry of foreign banks particularly stimulates OFDI in non-state-owned, higher profitability companies, in medium-sized, low financial marketization, and node cities designated in the Belt and Road Initiative. These findings provide new insights into emerging market financial openness and the internationalization of EMFs.
Existing methods that use propensity scores for heterogeneous treatment effect estimation on non-experimental data do not readily extend to the case of more than two treatment options. In this work, we develop a new propensity score-based method for heterogeneous treatment effect estimation when there are three or more treatment options, and prove that it generates unbiased estimates. We demonstrate our method on a real patient registry of patients in Singapore with diabetic dyslipidemia. On this dataset, our method generates heterogeneous treatment recommendations for patients among three options: Statins, fibrates, and non-pharmacological treatment to control patients' lipid ratios (total cholesterol divided by high-density lipoprotein level). In our numerical study, our proposed method generated more stable estimates compared to a benchmark method based on a multi-dimensional propensity score.
Government agencies rely on general deterrence to protect the public. Firms utilize lobbying for influence and information purposes. This paper explores the intersection of general deterrence and lobbying by firms while investigating whether general deterrence efforts of regulators are met with a lobbying response. Specifically, we propose that following a competitor firm being sanctioned, the non-sanctioned peer firms will increase their amount of lobbying targeted at the sanctioning agency, and the key drivers of these increases in targeted lobbying will be penalty severity, concerns over likely reputational damage, and value alignment. We test our hypotheses using U.S. Environmental Protection Agency violations as a context, and they largely receive support.
The objective of this study is to evaluate the impact of the Value School bonus program of the Government of Paraíba State, 2016 edition, on some school indicators. The methods of Propensity Score Matching and Differences in Differences were applied on Saeb microdata and School Census for the years 2015 and 2017. The results of the estimates revealed a positive impact of the bonus on the score of students from award-winning schools, as well as on the pass rate and dropout rate of students. Furthermore, the statistical significance of the interaction dummies between the number of awards and time suggests that the program tends to have a positive and cumulative effect on school performance.
Objectives
Existing evidence from high-income countries suggests that policies aimed at enhancing access to formal care can reduce the burden on informal carers and facilitate their reentry into the labor market. However, there is limited evidence regarding the specific carers who have been most impacted by such insurance. This study focuses on China’s long-term care insurance (LTCI) and examines its effects on informal care burden and the labor market participation of different types of informal carers.
Methods
Drawing data from the China Health and Retirement Longitudinal Study of 2011, 2013, 2015, and 2018, we employ a staggered difference-in-differences model with propensity score matching to analyze the impact of LTCI. To explore time-varying DID estimates, we adopted the DID event study design.
Results
Our study demonstrates that LTCI substantially alleviates the burden on informal carers while markedly boosting labor market participation. Notably, we found a more pronounced decrease in care burden among spouses, amounting to a reduction of 8.5 hours per month. Concurrently, LTCI’s impact on enhancing labor market participation was more significant among younger household members, reflected in an average income increase of 4,534 yuan per year. Furthermore, subgroup analysis highlights that LTCI primarily benefits informal carers providing care for older people with low income or those who were farmers or previously engaged in informal sectors.
Discussion
Our study demonstrates that LTCI has led to a reduction in care burdens and an enhancement in labor market participation. The impact is especially pronounced for informal carers of older people with low income or those with backgrounds in farming or informal work sectors.
We study the possibility of identifying general linear and nonlinear time series models using nonparametric methods. The kernel estimators of the conditional mean and variance are used as a basis, and the properties of these quantities as model indicators are briefly discussed. Some drawbacks are pointed out, and motivated by these we introduce projections as tools of identification. The projections are especially useful for additive modeling. Expressions for the asymptotic bias and variance are obtained. The projection of the conditional variance is suggested as a tool for identifying heteroscedastic time series. The results are illustrated by simulations for both the estimators of the projections and the estimators of the conditional mean and variance.
This paper demonstrates that even under ideal conditions, social experiments in general only uniquely determine the mean impacts of programs but not the median or the distribution of program impacts. The conventional common parameter evaluation model widely used in econometrics is one case where experiments uniquely determine joint the distribution of program impacts. That model assumes that everyone responds to a social program in the same way. Allowing for heterogeneous responses to programs, the data from social experiments are consistent with a wide variety of alternative impact distribution. We discuss why it is interesting to know the distribution of program impacts. We propose and implement a variety of different ways of incorporating prior information to reduce the wide variability intrinsic in experimental data. Robust Bayesian methods and deconvolution methods are developed and applied. We analyze earnings and employment data on adult women from a recent social experiment. In order to produce plausible impact distributions, it is necessary to impose strong positive dependence between outcomes in the treatment and in the control distributions. Such dependence is an outcome of certain optimizing models of the program participation decision.
This paper introduces a new stochastic process, a collection of $U$-statistics indexed by a family of symmetric kernels. Conditions are found for the uniform almost-sure convergence of a sequence of such processes. Rates of convergence are obtained. An application to cross-validation in density estimation is given. The proofs adapt methods from the theory of empirical processes.
Nonparametric regression using locally weighted least squares was first discussed by Stone and by Cleveland. Recently, it was shown by Fan and by Fan and Gijbels that the local linear kernel-weighted least squares regression estimator has asymptotic properties making it superior, in certain senses, to the Nadaraya-Watson and Gasser-Muller kernel estimators. In this paper we extend their results on asymptotic bias and variance to the case of multivariate predictor variables. We are able to derive the leading bias and variance terms for general multivariate kernel weights using weighted least squares matrix theory. This approach is especially convenient when analysing the asymptotic conditional bias and variance of the estimator at points near the boundary of the support of the predictors. We also investigate the asymptotic properties of the multivariate local quadratic least squares regression estimator discussed by Cleveland and Devlin and, in the univariate case, higher-order polynomial fits and derivative estimation.
Maximal inequalities for degenerate $U$-processes of order $k, k \geq 1$, are established. The results rest on a moment inequality (due to Bonami) for $k$th-order forms and on extensions of chaining and symmetrization inequalities from the theory of empirical processes. Rates of uniform convergence are obtained. The maximal inequalities can be used to determine the limiting distribution of estimators that optimize criterion functions having $U$-process structure. As an application, a semiparametric regression estimator that maximizes a $U$-process of order 3 is shown to be $\sqrt n$-consistent and asymptotically normally distributed.
SUMMARY We define a simple kernel procedure based on marginal integration that estimates the relevant univariate quantity in both
additive and multiplicative nonparametric regression.
A variety of criteria are relevant for evaluating alternative policies in democratic societies composed of persons with diverse values and perspectives. In this paper, we consider alternative criteria for evaluating the welfare state, and the data required to operationalize them. We examine sets of identifying assumptions that bound, or exactly produce, these alternative criteria given the availability of various types of data. We consider the economic questions addressed by two widely-used econometric evaluation estimators and relate them to the requirements of a comprehensive cost-benefit analysis. We present evidence on how the inference from the most commonly used econometric evaluation estimator is modified when the direct costs of a program are fully assessed, including the welfare costs of the taxes required to support the program. Finally, we present evidence of the empirical inconsistency of alternative criteria derived from evaluations based on on self-selection and attrition decisions, and on self-reported evaluations from questionnaires when applied to a prototypical job training program.
Semiparametric methods are developed to estimate the bias that arises from using nonexperimental comparison groups to evaluate social programs and to test the identifying assumptions that justify matching, selection models and the method of difference in differences. Using data from an experiment on a prototypical social program and data from nonexperimental comparison groups, the authors reject the assumptions justifying matching and their extensions of it. The evidence supports the selection bias model and the assumptions that justify a semiparametric version of the method of difference-in-differences. The authors extend their analysis to consider applications of the methods to ordinary observational data. Journal: Econometrica
This paper considers whether it is possible to devise a nonexperimental procedure for evaluating a prototypical job training programme. Using rich nonexperimental data, we examine the performance of a two-stage evaluation methodology that (a) estimates the probability that a person participates in a programme and (b) uses the estimated probability in extensions of the classical method of matching. We decompose the conventional measure of programme evaluation bias into several components and find that bias due to selection on unobservables, commonly called selection bias in econometrics, is empirically less important than other components, although it is still a sizeable fraction of the estimated programme impact. Matching methods applied to comparison groups located in the same labour markets as participants and administered the same questionnaire eliminate much of the bias as conventionally measured, but the remaining bias is a considerable fraction of experimentally-determined programme impact estimates. We test and reject the identifying assumptions that justify the classical method of matching. We present a nonparametric conditional difference-in-differences extension of the method of matching that is consistent with the classical index-sufficient sample selection model and is not rejected by our tests of identifying assumptions. This estimator is effective in eliminating bias, especially when it is due to temporally-invariant omitted variables.
The paper develops a statistical procedure to provide consistent estimators for the average impact of an intervention or treatment (e.g., earnings) among subjects of a target population (e.g., young high school dropouts). The procedures we study belong to a class of estimators which can be expressed as a DIFFERENCE between the average outcome in the treated sample and an adequately chosen weighted average of the outcomes in the control group. We refer to estimators in this class as D-estimators. We show that D-estimators should only be used in circumstances in which subjects are randomly assigned to treatment given the vector of observed covariates. This condition has been referred to as the \strongly ignorable assumption" or \selection on observable model". The consistency of D-estimators is proved under very weak restrictions on the distribution of the covariates.
Preliminary Notation and Definitions Modes of Convergence of a Sequence of Random Variables Relationships Among the Modes of Convergence Convergence of Moments; Uniform Integrability Further Discussion of Convergence in Distribution Operations on Sequences to Produce Specified Convergence Properties Convergence Properties of Transformed Sequences Basic Probability Limit Theorems: The WLLN and SLLN Basic Probability Limit Theorems: The CLT Basic Probability Limit Theorems: The LIL Stochastic Process Formulation of the CLT Taylor's Theorem; Differentials Conditions for Determination of a Distribution by Its Moments Conditions for Existence of Moments of a Distribution Asymptotic Aspects of Statistical Inference Procedures Problems
Social scientists never have access to true experimental data of the type sometimes available to laboratory scientists.1 Our inability to use laboratory methods to independently vary treatments to eliminate or isolate spurious channels of causation places a fundamental limitation on the possibility of objective knowledge in the social sciences. In place of laboratory experimental variation, social scientists use subjective thought experiments. Assumptions replace data. In the jargon of modern econometrics, minimal identifying assumptions are invoked.
This paper considers the use of instrumental variables to estimate the mean effect of treatment on the treated, the mean effect of treatment on randomly selected persons and the local average treatment effect. It examines what economic questions these parameters address. When responses to treatment vary, the standard argument justifying the use of instrumental variables fails unless person-specific responses to treatment do not influence decisions to participate in the program being evaluated. This requires that individual gains from the program that cannot be predicted from variables in outcome equations do not influence the decision of the persons being studied to participate in the program. In the likely case in which individuals possess and act on private information about gains from the program that cannot be fully predicted by variables in the outcome equation, instrumental variables methods do not estimate economically interesting evaluation parameters. Instrumental variable methods are extremely sensitive to assumptions about how people process information. These arguments are developed for both continuous and discrete treatment variables, and several explicit economic models are presented.
Local high-order polynomial fitting is employed for the estimation of the multivariate regression function m(x1,…xd) =E{φ(Yd)φX1=x1,…,Xd=xd}, and of its partial derivatives, for stationary random processes {Yi, Xi}. The function φ may be selected to yield estimates of the conditional mean, conditional moments and conditional distributions. Uniform strong consistency over compact subsets of Rd, along with rates, are established for the regression function and its partial derivatives for strongly mixing processes.
We consider the estimation of the multivariate regression function m(x1, ..., xd) = E[[psi](Yd)X1 = x1, ..., Xd = xd], and its partial derivatives, for stationary random processes Yi, Xi using local higher-order polynomial fitting. Particular cases of [psi] yield estimation of the conditional mean, conditional moments and conditional distributions. Joint asymptotic normality is established for estimates of the regression function and its partial derivatives for strongly mixing and [varrho]-mixing processes. Expressions for the bias and variance/covariance matrix (of the asymptotically normal distribution) for these estimators are given.
The propensity score is the conditional probability of assignment to a particular treatment given a vector of observed covariates. Both large and small sample theory show that adjustment for the scalar propensity score is sufficient to remove bias due to all observed covariates. Applications include: (i) matched sampling on the univariate propensity score, which is a generalization of discriminant matching, (ii) multivariate adjustment by subclassification on the propensity score where the same subclasses are used to estimate treatment effects for all outcome variables and in all subpopulations, and (iii) visual representation of multivariate covariance adjustment by a two-dimensional plot.
For the class of single-index models, I construct a semiparametric estimator of coefficients up to a multiplicative constant that exhibits -consistency and asymptotic normality. This class of models includes censored and truncated Tobit models, binary choice models, and duration models with unobserved individual heterogeneity and random censoring. I also investigate a weighting scheme that achieves the semiparametric efficiency bound.
This paper presents methods for estimating the impact of training on earnings when non-random selection characterizes the enrollment of persons into training. We explore the benefits of cross-section, repeated cross-section and longitudinal data for addressing this problem by considering the assumptions required to use a variety of new and conventional estimators given access to various commonly encountered types of data. We investigate the plausibility of assumptions needed to justify econometric procedures when viewed in the light of prototypical decision rules determining enrollment into training. We examine the robustness of the estimators to choice-based sampling and contamination bias.
Consider a $p$-times differentiable unknown regression function $\theta$ of a $d$-dimensional measurement variable. Let $T(\theta)$ denote a derivative of $\theta$ of order $m$ and set $r = (p - m)/(2p + d)$. Let $\hat{T}_n$ denote an estimator of $T(\theta)$ based on a training sample of size $n$, and let $\| \hat{T}_n - T(\theta)\|_q$ be the usual $L^q$ norm of the restriction of $\hat{T}_n - T(\theta)$ to a fixed compact set. Under appropriate regularity conditions, it is shown that the optimal rate of convergence for $\| \hat{T}_n - T(\theta)\|_q$ is $n^{-r}$ if $0 < q < \infty$; while $(n^{-1} \log n)^r$ is the optimal rate if $q = \infty$.
This paper decomposes the conventional measure of selection bias in observational studies into three components. The first two components are due to differences in the distributions of characteristics between participant and nonparticipant (comparison) group members: the first arises from differences in the supports, and the second from differences in densities over the region of common support. The third component arises from selection bias precisely defined. Using data from a recent social experiment, we find that the component due to selection bias, precisely defined, is smaller than the first two components. However, selection bias still represents a substantial fraction of the experimental impact estimate. The empirical performance of matching methods of program evaluation is also examined. We find that matching based on the propensity score eliminates some but not all of the measured selection bias, with the remaining bias still a substantial fraction of the estimated impact. We find that the support of the distribution of propensity scores for the comparison group is typically only a small portion of the support for the participant group. For values outside the common support, it is impossible to reliably estimate the effect of program participation using matching methods. If the impact of participation depends on the propensity score, as we find in our data, the failure of the common support condition severely limits matching compared with random assignment as an evaluation estimator.
The role of propensity score in the efficient estimation of the average treatment effects is examined. If the treatment is ignorable given some observed characteristics, it is shown that the propensity score is ancillary for estimation of the average treatment effects but not for estimation of average treatment effects on the treated. Efficient semiparametric estimators take the form of relevant sample averages of the data completed by the nonparametric imputation method. Projection on the propensity score is not necessary for efficient semiparametric estimation of the average treatment effects on the treated even if the propensity score is known.
Large Sample Estimation And Hypothesis Test-ing , in Handbook of Econometrics Vol IV
- W K Newey
- D L Mcfadden
NEWEY, W. K. and McFadden, D. L. (1994) Large Sample Estimation And Hypothesis Test-ing , in Handbook of Econometrics Vol IV, edited by R. F. Engle and D. L. McFadden, Elsevier Science B. V., Amsterdam.
Evaluating The Welfare State , Frisch Symposium Pa-per
- J Heckman
- J Smith
HECKMAN, J. and SMITH, J. (1997), Evaluating The Welfare State , Frisch Symposium Pa-per, Oslo Norway (1995), forthcoming in Frisch Centenary, Econometric Monograph Series, Cam-bridge: Cambridge University Press.
- B Barnow
- G Cain
- A Goldberger
BARNOW, B., CAIN, G. and GOLDBERGER, A. (1980), \Issues in the Analysis of Selectivity
Bias," in Evaluation Studies Review Annual, Volume 5, ed. by E. Stromsdorfer and G. Farkas
(San Francisco: Sage).
Nonparametric Characterization of Selection Bias Using Experimental Data, Part 11: Econometric Theory and Monte Carlo Evidence
- J Heckman
- H Ichimura
- J Smith
- P Todd
HECKMAN, J., ICHIMURA, H., SMITH, J. and TODD, P. (1996a), " Nonparametric Characterization of Selection Bias Using Experimental Data, Part 11: Econometric Theory and Monte Carlo Evidence " (Unpublished manuscript, University of Chicago).
Controlling Bias In Observational Studies Conditional Independence in Statistical Theory Local Linear Regression Smoothers and Their Minimax Efficiencies Direct Estimation of Low Dimensional Components in HAHN Shadow Prices
- W G Cochrane
- D B Sankyha
- A P Dawid
- J Fan
- W Hardle
- E Mammen
COCHRANE, W. G. and RUBIN, D. B. (1973), " Controlling Bias In Observational Studies ", Sankyha, 35, DAWID, A. P. (1979), " Conditional Independence in Statistical Theory ", Journal of The Royal Statistical FAN, J. (1993), " Local Linear Regression Smoothers and Their Minimax Efficiencies ", The Annals of Statistics, FAN, J., HARDLE, W. and MAMMEN, E. (1996), " Direct Estimation of Low Dimensional Components in HAHN, J. (1996), " On the Role of the Propensity Score in the Efficient Semiparametric Estimation of Average HECKMAN, J. (1974), " Shadow Prices, Market Wages, and Labor Supply ", Econometrica, 42, 679-694.
Evaluating The Welfare State Frisch Centenary, Econometric Mono-graph Series (Cambridge Making the Most Out of Social Experiments: Reducing the Intrinsic Uncertainty in Evidence From Randomized Trials With An Application to the National JTPA Experiment
- J Heckman
- J Smith
- Forthcoming
- J Heckman
- J Smith
- N Clements
HECKMAN, J. and SMITH, J. (1998), " Evaluating The Welfare State ", Frisch Centenary, Econometric Mono-graph Series (Cambridge : Cambridge University Press), forthcoming. HECKMAN, J., SMITH, J. and CLEMENTS, N. (1997), " Making the Most Out of Social Experiments: Reducing the Intrinsic Uncertainty in Evidence From Randomized Trials With An Application to the National JTPA Experiment ", Review of Economic Studies, 64, 487-535.
Varieties of Selection Bias Instrumental Variables: A Study of the Implicit Assumptions Underlying One Widely Used Estimator for Program Evaluations
- J Heckman
- J Heckman
HECKMAN, J. (1990), " Varieties of Selection Bias ", American Economic Review, 8 0, 31 3-3 18. HECKMAN, J. (1997), " Instrumental Variables: A Study of the Implicit Assumptions Underlying One Widely Used Estimator for Program Evaluations ", Journal of Human Resources, 32, 441-462.
\Nonparametric Characterization of Selection Bias Using Experimental Data, Part I: De¯nitions, Applications and Empirical Results
- J Ichimura
- H Smith
HECKMAN, J., ICHIMURA, H., SMITH, J. and TODD, P (1994, revised 1996a), \Nonparametric Characterization of Selection Bias Using Experimental Data, Part I: De¯nitions, Applications
and Empirical Results", unpublished manuscript, University of Chicago.
Impact on 1977 Earnings of New FY 1976 CETA Enrollees in Selected Program Activities
WESTAT, INC. (1980), \Net Impact Report No. 1, Impact on 1977 Earnings of New FY 1976
CETA Enrollees in Selected Program Activities", Rockville, Maryland, Westat Inc.
- W G Cochrane
- D B Rubin
COCHRANE, W. G. and RUBIN, D. B. (1973), \Controlling Bias In Observational Studies,"
Sankyha, 35, 417-446.
- J Heckman
HECKMAN, J., (1974), \Shadow Prices, Market Wages, and Labor Supply", Econometrica, 42,
679-694.
revised 1996a), \Nonparamet-ric Characterization of Selection Bias Using Experimental Data, Part I: De¯nitions, Applications and Empirical Results
- J Heckman
- H Ichimura
- J Smith
- P Todd
HECKMAN, J., ICHIMURA, H., SMITH, J. and TODD, P (1994, revised 1996a), \Nonparamet-ric Characterization of Selection Bias Using Experimental Data, Part I: De¯nitions, Applications
and Empirical Results", unpublished manuscript, University of Chicago.
revised 1996b), \Nonpara-metric Characterization of Selection Bias Using Experimental Data, Part II: Econometric Theory and Monte Carlo Evidence
- J Heckman
- H Ichimura
- J Smith
- P Todd
HECKMAN, J., ICHIMURA, H., SMITH, J. and TODD, P (1994, revised 1996b), \Nonpara-metric Characterization of Selection Bias Using Experimental Data, Part II: Econometric Theory
and Monte Carlo Evidence", unpublished manuscript, University of Chicago.
\Asymptotic Distribution of Nonparametric and Semiparametric Estimators with Data-Dependent Smoothing Parameters
- H Ichimura
ICHIMURA, H. (1995), \Asymptotic Distribution of Nonparametric and Semiparametric Estimators with Data-Dependent Smoothing Parameters", unpublished manuscript, University of
Chicago.
- D Pollard
POLLARD, D. (1990), Empirical Processes: Theory and Applications, IMS, Hayward, CA.
Months After Entry), Postprogram Experiences and Pre/Post Comparisons For Terminees Who Entered CETA During
- Inc Westat
WESTAT, INC. (1982), \CLMS Follow up Report No. 7 (18 Months After Entry), Postprogram
Experiences and Pre/Post Comparisons For Terminees Who Entered CETA During July 1976