With the rapid development of artificial intelligence technologies, an increasing number of companies have adopted algorithmic systems to replace human managers in organizational decision-making. Although algorithmic organizational decision-making can significantly improve companies’ managerial efficiency, the use of algorithms is often morally condemned by the public, which might in turn affect company performance in the consumer market. Therefore, it is of great practical significance to study the impact of the adoption of algorithmic organizational decision-making on consumers' perceived ethicality of the company.

Unfortunately, little attention has been paid to the relationship between the adoption of algorithmic organizational decision-making and consumers' perceived ethicality of the company. Three current literature streams are related to this topic, but each of them has its specific limitations. First, although prior literature has found organizational activities as an important source of consumer perceived ethicality, no study thus far has shed light on the impact of algorithmic organizational decision-making. Considering the complexity and controversy of algorithmic organizational decision-making, theories regarding how organizational activities affect consumer perceived ethicality await examination in the context of algorithmic organizational decision-making. Second, the extant literature mostly focuses on the context where the outcome of algorithmic or human-based decisions is explicit and examines how consumers’ responsibility attribution processes differ in terms of decision-making agents. However, little attention has been paid to the context where the decision outcome is ambiguous. Third, most of the extant literature examines the impact of algorithmic organizational decision-making on firm performance through the perspective of decision-makers (i.e. managers) and decision receivers (i.e. employees); the perspective of decision observers (such as consumers in this thesis) awaits exploration.

Considering the practical and theoretical backgrounds above, this thesis, inclusive of three related studies, aims to investigate the impact of the adoption of algorithmic organizational decision-making on consumers' perceived ethicality of the company. Based on theory of machine, correspondent interference theory, and moral judgment theory, this thesis proposes that 1) consumers would make less favorable ethicality inferences of companies that rely on algorithms (vs. human) to make organizational decisions, even when the decision outcome of algorithms is not explicitly worse off (Study One). 2) Because consumers tend to attribute the dehumanizing nature of algorithmic decision-making to the company’s lower level of respect toward its employees, which in turn draws consumers’ unfavorable ethicality inferences of the company (Study Two). Meanwhile, the negative impact of the adoption of algorithmic organizational decision-making on consumers' perceived ethicality of the company would draw negative downstream consequences (Study Three).

Study One (three experiments; N = 667) investigated the effect of the type of organizational decision-making agent (algorithm vs. human) on consumer perceived ethicality of the company. We found that consumers tend to perceive the company as less ethical when it relies on algorithms (vs. human) to lay off employees, even when the decision outcome is not explicitly negative (experiment 1-1). In addition, the proposed effect can be generalized between different organizational contexts. Specifically, we found that consumers tend to perceive the company as less ethical when it relied on algorithms (vs. human) to make either layoff or promotion decisions (experiments 1-2). However, with explicit positive or negative decision outcomes, consumers drew the same ethical inferences toward companies that relied on either algorithms or human to make organizational decisions (experiments 1-3).

Study Two (four experiments; N = 921) further explored the mechanism that underlies the effect of the type of organizational decision-making agent (algorithm vs. human) on consumer perceived ethicality. We found that the degree to which the company respects its employees could serve as the mediator (experiment 2-1). However, the perceived respect mechanism has boundaries in terms of algorithms’ delegation type (experiment 2-2), employees’ humanness (experiment 2-3), and consumers’ power distance belief (experiment 2-4). Specifically, we found that the mediating role of perceived respect is attenuated when algorithms were adopted as augmented (vs. automated) agents, the humanness of targeted employees was weak (vs. strong), or the consumers were of high (vs. low) power distance beliefs.

Study Three (three experiments; N = 404) investigated the downstream consequences of consumers’ unfavorable ethicality inferences of the company induced by its adoption of algorithmic organizational decision-making. The results showed that the company’s adoption of algorithmic (vs. human-based) organizational decision-making drew unfavorable consumers’ ethicality inferences of the company, in turn leading to consumers’ more negative attitude toward the company (experiment 3-1), lower willingness to purchase the company’s products (experiment 3-2), and stronger intention to conduct ethical behavior to punish the company (experiment 3-3).

This thesis contributes to the extant literature in three primary ways. First, although prior literature has suggested organizational activities as a major source of consumer perceived ethicality, this thesis focuses on an overlooked but distinctively complex organizational activity, algorithmic organizational decision-making. We found that the adoption of algorithmic organizational decision-making would draw unfavorable consumers’ perceived ethicality of the company. In turn, this thesis enriches the current understanding of the influences of organizational activities on consumers' perceived ethicality of the company.

Second, prior literature on consumers' ethical evaluations of algorithm users mostly focused on the context where the decision is explicitly negative or positive. On the contrary, this thesis focuses on the context where the decision outcome is ambiguous. Based on theory of machine, correspondent inference theory, and moral judgment theory, this thesis proposed and examined the mediating role of perceived respect. By doing so, this thesis provides a novel theoretical explanation to understand how and why consumers draw ethical inferences toward algorithm users.

Third, most of the previous literature studied the consequences of the adoption of algorithmic organizational decision-making from the perspective of decision makers (i.e. managers) or decision receivers (i.e. employees). On the contrary, this thesis focuses on the decision observer perspective (i.e. consumers). We found that the company’s adoption of algorithmic organizational decision-making would draw unfavorable consumers’ ethicality inferences of the company as well as its unfavorable downstream consequences. These findings provide initial evidence that examining the consequences of algorithmic organizational decision-making from the observer's perspective could be promising.

This thesis is practically relevant from two perspectives. First, this thesis proposed and tested the negative impact of algorithmic organizational decision-making on consumers' perceived ethicality of the company, which would in turn draw unfavorable marketing consequences. These findings alert business managers to be aware of the potential negative externalities of algorithm adoption (i.e., the unfavorable ethicality inferences and their downstream consequences). This would help business managers to assess the consequences of the adoption of algorithmic organizational decision-making in a more comprehensive way. Second, this thesis examined the boundaries of the mediating effect of perceived respect. These findings provide business managers with actionable solutions to avoid or mitigate the negative impact of algorithmic organizational decision-making on consumer perceived ethicality. This would guide business managers to conduct algorithmic organizational practices more effectively.