本文主張經理人不同的文化背景以及生涯經歷會影響公司在社會企業社會責任(Corporate Social Responsibility, CSR)相關政策並檢驗經理人生涯三階段的論點，分別是出生階段、大學階段以及工作階段(以前一份工作為代表)，此作法可解答經理人與CSR決策相關之文化認同於生涯何階段形成，並在實證上改進現有以國籍為主的分類方式。 利用Hofstede以及Schwartz等人提出之文化尺度辨別經理人與公司之文化分數，本文確認了在公司所在環境之國家文化之外，經理人文化認同對公司CSR相關政策之影響。另外，文化並非永久不變的，其中於大學階段形成的文化認同對於公司在CSR相關政策的執行上最具影響力。經初步實證發現，權力距離於三階段皆與CSR政策呈現正相關、不確定性規避則於大學階段以及工作階段與CSR政策呈現正相關，和諧主義與公司CSR政策之執行於出生階段以及大學階段皆呈現正相關；而縱容主義與CSR相關政策於大學以及工作階段呈現負相關，個人主義以及陽剛主義主要於大學階段與公司CSR政策呈現負相關、保守主義則是於出生階段與CSR政策呈現負相關。上述發現有些文化維度在經理人同時為董事長的公司中更強，於第四章說明。最後我針對其中六個文化維度挑選合適之工具變數進行兩階段迴歸，分析結果顯示，個人主義於生涯三階段的結果皆保留；權力距離在工作階段轉為顯著負向關係；保守主義保留大學階段的結果；和諧主義則是出生階段結果保留。 ;This paper finds that CEO’s cultural identity formed in different life stages affects his/her firm’s Corporate Social Responsibility (CSR) decisions. I examine CEO’s cultural identity formed in three different stages throughout his/her life: We childhood stage, college stage, and working stage (use the last job before current position). With this empirical design, I can answer which stage is the most relevant in forming cultural identity which affects CSR policies and I provide an improvement over the commonly used method according to nationality. Utilizing Hofstede and Schwartz’s cultural dimensions to identify the CEO and the firm’s cultural identity, I confirm that the CEO’s culture identity has impact on the CSR policies in addition to the firm’s cultural identity. Furthermore, the CEO’s cultural identity is not constant over time and the cultural identity formed in college stage has more impact. I found that Power distance is positively correlated with the CSR policies in all three stages, and the Uncertainty avoidance is positively related to the CSR policies at the college stage and working stage. Harmony is positively related to the CSR policies in childhood stage and college stage; Indulgence is negatively related to the CSR policies at college stage and working stage, Individualism and Masculinity is negatively related to CSR policies in college stage and Embeddedness is negatively related to CR policies in childhood stage. The effect is stronger for some culture dimensions when the CEO is also the chairman of the board. Lastly, I selected appropriate instrument variables and ran two-stage regressions to confirm the above findings. The results regarding Individualism are robust in all three stages; the results of Power Distance reverse sign in working stage; the results of Embeddedness (Harmony) remain only in college stage.
