Given the wide variety of equity funds in the public offering market, fund investors find it difficult to select funds that match their investment preferences. Therefore, equity fund investors pay greater attention to funds that are recommended by sales professionals from banks or securities companies. They rely on brokers' recommendations when selecting equity funds instead of conducting rigorous analysis. Therefore, brokers' fund recommendations carry great importance from the perspective of investor protection. However, due to the limitations of data collection, research on this topic has not yet been conducted in Korea. Analysis of fund recommendations would have valuable implications for investor protection. Studies of equity funds have paid much attention to overall fund performance, fee structures, or the role of fund distributors (e.g., Won, 2009; Kim, 2011; Kho and Baek, 2013). However, although performance analysis of fund recommendations is important, research has paid little attention to the topic. To the best of our knowledge, no study has considered the real performance of recommended funds in Korea, as fund brokers do not voluntarily disclose data related to fund sales or the return performance of recommended funds. As each broker sells tens of funds, it is difficult to estimate the return performance of the funds of each broker. However, fund evaluation companies have recently begun to build data on funds recommended by each vendor, so analysis of the performance of recommended funds is possible. We empirically analyze the determinants of brokers' fund recommendations and the performance of recommended funds in two ways. First, we analyze the factors that determine brokers' fund recommendation decisions. Second, we analyze the future performance of the recommended funds. To confirm our findings, we use propensity score matching to compare the performances by matching recommended and non-recommended funds with similar characteristics. To investigate possible conflicts of interest in the fund recommendations, we focus on the levels of fees and commissions and the relationships between brokers and fund managers. We pay attention to the possibility that the seller, in recommending a fund to a customer, acts on behalf of the seller or affiliated fund manager rather than prioritizing the customer's benefit. The results of this study are as follows. Funds with front-end loads and/or funds of asset management firms with no affiliated brokerage companies are more likely to receive recommendations from brokers. Furthermore, funds of large asset management companies receive more recommendations than those of medium or small management companies. In addition, new funds with short operating periods are selected as recommended funds, and new funds from affiliate management firms are more likely to be recommended. The performance of those recommended funds turns out to be inferior. Based on the results of analysis of the profitability of broker-recommended funds, we find that recommended funds do not perform better than funds that are not recommended by brokers. This result does not support the general perception that brokers have fund-selection ability. However, no evidence supports the view that the inferior performance of broker-recommended funds is caused by the self-interest of the broker who pursues the profit of a subsidiary asset management company. Recommended funds of affiliated asset management companies record better performance than those of non-affiliated companies. In particular, when the seller recommends a newly launched fund of the affiliated management company, its return performance is better. Thus, we conjecture that the inferior performance of a recommended fund is not due to the broker's intention to subsidize an affiliated asset management company. This result suggests that a broker, through an information transfer from affiliated fund managers, plays a role in mitigating the information asymmetry of newly launched funds. It also supports the view that the affiliate broker analyzes the newly launched fund more accurately. Our study shows that broker-recommended funds generally exhibit inferior return performance. This result suggests that fund sellers are required to improve their fund selection ability to enhance the protection of fund investors. Sellers must improve their fund analysis capabilities through internal research and workforce reinforcement. Our study has some limitations in the sense that it covers a relatively short period due to the insufficient accumulation of fund recommendation data. Further research should be conducted to review the robustness of the analytical results over a wider analysis period after equity fund data are accumulated more broadly.