Corporate social responsibility（CSR）has been a highly debated topic in these past years,as businesses are looking to move towards more “sociallyresponsible” business models.Generally,CSR has been viewed as a multidimensional construct.For example,Aguinis and Glavas（2012）define CSR as “context-specific organizational actions and policies that take into account stakeholder’s expectations and the triple bottom line of economic,social and environmental performance”.One main stream of CSR research focuses on investigating the antecedents of CSR.That is,scholars have examined the factors that are able to influence whether and why firms initiate CSR practices.Those studies have largely highlighted external pressures.Influence from the organizational field,which includes groups of organizations to which the firm is aligned,is a key determinant of CSR（Hoffman 1999）.In the realm of philanthropic giving,firms take cues from their institutional peers（Marquis and Tilcsik 2016）.In addition,the context/climate of the firm’s surroundings is identified as a critical factor for determining the firm’s CSR activities,particularly in country-specific corporate governance structures（Williams and Aguilera2009）,country context（Brammer,Pavelin,and Porter 2009）and sociocultural environment（Victor and Cullen 1988）.Many of these predictors are found in institutional environments and operate externally.While these external factors have been demonstrated to influence CSR engagement,the decision-making process of a firm to initiate and engage in CSR practices has not been thoroughly examined.In order to both understand actual decision-making processes and extend the research on CSR,this study places CSR in the context of firm decision-making.Drawing from the behavioral theory of the firm（BTOF）,which is a theory particularly relevant to firm decision-making,this study aims to explore how financial performance feedback can play a role in determining CSR of a firm.In this way,a behavioral-theory-based view of CSR can further elaborate on the “how” of CSR practices,rather than viewing them as a response to external environment.Through this,the study hopes to contribute to the understanding of how firms make CSR-related decisions and provide a foundation for firms and stakeholders to utilize in engaging more effectively with CSR.Several key theoretical concepts formulate the framework of this study.First,this study explores the foundational research of CSR,looking at its multilevel definition and antecedents,along with usage in related models and other research.This study then deviates from existing literature by examining CSR through BTOF.As institutional theory of a firm has commonly been applied in analyzing CSR,the BTOF can help uncover more of the underlying mechanisms within a firm that encourage it to change or modify its strategy and business operations.Within BTOF,the main ideas of bounded rationality,aspiration levels,performance feedback,and search are highlighted.Firms operate within a certain scope and would extend this scope in cases where they require more information in order to modify firm behavior.To begin a process of changing firm strategy or not depends on certain goals the firm creates for itself.These goals come in the form of aspiration levels,which is the base for the firm to evaluate its performance.In this study,historical and social aspiration levels are discussed,with historical meaning comparison of current and past performances and social indicating comparison among peers in the same industry.From these aspirations,firms can determine whether they are performing above or below aspirations,equivalent to success or failure.The feedback is used as the main factor in the decision-making process to embark on a search.Search in this case refers to an action taken by the firm in response to the feedback.Search may take the form of problemistic or slack-driven.Problemistic search occurs when firms perceive that they are performing below aspirations and need to remedy their situation.Slack-driven search is propelled by the fact that the firm has excess resources due to performing above aspirations,and it chooses to use this buffer to engage in more innovative or exploratory means.The outcome of these two types of searches that this study examines is CSR.Each firm’s CSR activities are divided into two types.Technical CSR deals with primary stakeholders while institutional CSR is related to secondary stakeholders.Using BTOF as a theoretical basis,the framework in this study suggests how firms will respond to CSR engagement or involvement depending on the type of performance feedback received.For the purpose of this study,performing below aspirations is assumed to be equivalent to negative performance feedback while performing above aspirations is equal to positive performance feedback.Performance feedback in this analysis is in the form of financial performance feedback.Generally,positive financial feedback can be expected to increase engagement in technical and institutional CSR.However,in the case of negative performance feedback,engagement with technical CSR is still expected to increase while institutional CSR is expected to decrease.The case of ambiguous feedback for firms is also examined where firms receiving conflicting feedback may exercise self-enhancement or negative bias.Applying these underlying concepts,five main hypotheses are formulated.The first two cover historical financial performance feedback.Positive historical feedback is expected to increase both technical and institutional CSR.Negative historical feedback is expected to also increase technical CSR score but decrease institutional CSR score instead.The next two hypotheses propose a similar relationship.Positive social feedback is expected to increase both types of CSR.Negative social feedback will have a positive effect on technical CSR but decrease institutional CSR.The last hypothesis concerns firm biases.On one hand,firms may choose to focus more on the positive feedback they receive,thereby adhering to the self-enhancement bias.On the other hand,firms can focus on their negative feedback instead,falling into the negative bias.These hypotheses will serve as the main arguments for this study.To empirically examine the arguments in this study,Chinese listed firms are used as the sample.CSR in this study is represented by each listed firm’s CSR scores,indicating the level of engagement firms have with the respective CSR activity types.CSR scores are obtained from Hexun CSR ratings system which collects comprehensive data from annual reports submitted by Chinese listed firms on both the Shanghai and Shenzhen Stock Exchange.The rating system gives firms a score for their CSR engagement based on five specific sub-indicators of CSR that spans across various types of stakeholders.Technical,institutional and overall CSR scores are collected.The rest of the firm data for this analysis,which includes financial performance and other control variables,is obtained from the China Stock Market & Accounting Research（CSMAR）database.The models chosen for this analysis are fixed effect and random effect models which are suitable for interpreting panel data,the format that is used for the data set.The analysis is conducted on STATA,where results demonstrate that model selection had no impact,and results from both models are viable.The results from STATA demonstrate three important findings regarding the influence of financial performance feedback on CSR.The first interesting and important finding is negative performance feedback tends to have greater influence on CSR（i.e.,institutional CSR,technical CSR,and overall CSR）.The second finding is that firms with negative social feedback tend to improve technical CSR because technical CSR has the potential to improve financial performance.The third finding is that firms with negative historical feedback tend to decrease institutional CSR because institutional CSR is unlikely to improve financial performance but instead reduction of institutional CSR could help reduce costs.The findings highlight that performing below historical aspirations and below social aspirations possess different meanings for firms and as a result firms take different actions.It is supported that when firms perform below historical aspirations,they tend to decrease involvement in CSR.Furthermore,below historical aspiration performance also leads to lower involvement with institutional CSR.On the contrary firms performing below social aspirations increase involvement with CSR.Breaking it down further,below social aspiration performance leads to higher involvement with technical CSR.What the results are not able to demonstrate is what happens in the case of firms receiving positive feedback both historically and socially.While there are some assumptions and implications that can be deduced,the results are not able to support the proposed outcomes of firms receiving positive feedback.This would require further research.Certain practical implications can be made from the findings.It can educate firms that are hesitant or unfamiliar with CSR to understand how other firms are engaging with CSR and suggest similar actions they can also implement.The findings also elevate the knowledge among firms that are already engaging with CSR,providing a rough model to interpret CSR performance.Investors can also benefit by understanding more on how CSR fits into firm strategy and decision-making processes.This study does have limitations.Although some control variables have been specifically chosen to make the analysis more robust,there can still be better approaches to examining the net relationship between the predictor and outcome variables.The data chosen is in the form of a score that indicates CSR performance.Yet,as this score depends on annual reports,certain firms may categorize and present their CSR activities in different ways.This is a tough research obstacle due to the complexity of CSR but can be pursued further.Overall,this study lends support to extending the research in using behavioral theory of a firm to understand CSR.While previous research places CSR in the predictor category and also under influence from external stakeholders,this study looks at CSR as an outcome from firm internal decision-making processes.Through this,the multi-dimensionality of CSR can be further understood and offer an interesting view on how CSR has been integrated into new business models.