As the role of foreign subsidiaries in capability development of multinational corporations (MNCs) has gained importance, there has been active discussion regarding reverse knowledge transfer—the transfer of knowledge acquired or created by subsidiaries in host countries back to headquarters. Previous studies have simply assumed reverse knowledge transfer as knowledge transfer between organizations within an MNC and have examined factors influencing reverse knowledge transfer only as individual factors depending on the researcher's context. However, knowledge transfer between organizations within an MNC involves substantial hierarchical elements and temporal and spatial differences. Therefore, rather than examining reverse knowledge transfer as simple knowledge transfer between organizations within a firm, it is also important to treat subsidiaries as independent entities and examine factors influencing reverse knowledge transfer from an inter-firm knowledge transfer perspective. Moreover, since knowledge transfer involves transferring the knowledge provider's knowledge to the recipient, thereby enhancing the recipient's competitive advantage, it is important to identify the factors that the knowledge provider must consider for effective knowledge transfer. Accordingly, this study examined factors influencing reverse knowledge transfer from the perspective of foreign subsidiaries as knowledge providers, limiting the transferred knowledge to marketing knowledge characterized by strong tacit properties. The study constructed a research model that comprehensively considers subsidiary characteristics, headquarters characteristics, and the relationship characteristics between headquarters and subsidiaries, and empirically tested it from an inter-firm knowledge transfer perspective. For this purpose, 253 foreign subsidiaries of multinational corporations operating in Korea were surveyed, and the collected data were analyzed using SPSS (version 15.0). The analysis results confirmed that characteristics of both the knowledge provider and the knowledge recipient—including subsidiaries' marketing knowledge creation, subsidiaries' willingness to transfer knowledge, headquarters' absorptive capacity, and headquarters' learning intent—all positively influenced reverse knowledge transfer. Additionally, among the relationship characteristics between headquarters and subsidiaries, socialization mechanisms, social capital, and knowledge relevance positively influenced reverse knowledge transfer, whereas formal integration mechanisms did not significantly affect reverse knowledge transfer. Taken together, as posited in the research model, it can be concluded that the characteristics of foreign subsidiaries as knowledge providers, the characteristics of headquarters as knowledge recipients, and the relationship characteristics between headquarters and subsidiaries all influence reverse knowledge transfer.