Yan XIONG
Prof. Yan XIONG
会计学及法学
金融学
Associate Professor
MAA Programme Director

39171003

KK 1218

Publications
港股向前,恒指向后?

今年外围股市屡创新高,恒生指数却停滞不前,让不少投资者大叹“恨铁不成钢”。大众直觉认为,恒指低迷,代表港股表现差。面对这种失落感,是投资者对恒指执念太深,还是恒指已跟不上市场脉动,急需改革?

数据资产的价值,只能由市场发掘

2026 年,Reddit 与 Google 那份签约价约 6000 万美元/年的数据授权合同即将到期续约。加上同期与 OpenAI 签订的姊妹合同,两份协议目前合计为 Reddit 贡献约 1.3 亿美元的年化数据授权收入。分析师预测,续约后两份合同总价可能跳升至 5.5 亿美元。那么,这些价格是怎么定出来的?更广义地,数据资产应该如何定价?

AI冲击会计前景?港大熊琰教授:两者是赋能关系

港大经管学院熊琰副教授接受《香港经济日报》采访时指出,尽管不少学生担忧会计行业的前景,但收生人数不跌反升,原因在于会计跟AI是赋能而非取代简单关系。她强调,“学生必须找出雇主的真实需求,之后去适应并装备自身。”

The Strength of Weak Commitments: A Theory of Price Preannouncements

Manufacturers often preannounce reference prices for products that have not yet been produced or even developed. These prices are rarely binding, meaning that the manufacturers can make price adjustments in the future, possibly at a cost. In this paper, we argue that price preannouncements can serve as a weak price commitment that, we find, helps the manufacturers secure better deals from their suppliers, thereby lowering their procurement costs and improving their profit. Surprisingly, even an extremely weak price commitment can substantially improve a manufacturer’s profit. On the other hand, when the price commitment is credible enough, the manufacturer forgoes the price preannouncement. Collectively, these results underscore the strategic effects that price preannouncements can have on firms’ marketing decisions.

AI投资泡沫?

确实,虽然大量资本正涌入AI领域,但其中相当一部分投资尚未带来实质性或可见的商业回报。例如,Meta发布财报披露其投入巨资布局AI运算能力、数据中心和人才招募后股价应声下跌,部分反映了投资者对于大量AI投资的疑虑。那么,当下的AI投资是否已经显得过于狂热? 还是如许多支持者所言,“这一次是真的”(This time is real)?

Personalized Pricing, Network Effects, and Commitment

Big data and data technology have facilitated the widespread adoption of personalized pricing practices. While price personalization enables firms to extract greater rent from consumers, it often reduces price transparency, which can negatively impact firm profits in situations involving consumer coordination. In such contexts, a firm's commitment to pricing strategies can become essential for restoring profitability. We explore several commitment devices available to firms and discuss their implications. These devices include delegating pricing decisions to a manager who prioritizes consumer surplus, leveraging existing networks as signals for later consumers or to build reputation, and implementing uniform pricing or price caps in response to regulatory restrictions.

另类数据隐患:AI时代下的信贷市场

人工智能技术高速发展,有机会被部分借款者用来美化其信贷资料以获批贷款。贷款方与其透过多方面大量收集数据审批借款者,倒不如自我设限,集中提高收集数据的质量,同时增加贷款利润。

Disclosing Endogenous Cost Information

We study voluntary cost disclosure by duopoly firms when they can invest in a cost-reduction technology, i.e., when their private cost is endogenously determined. We find that, contrary to most of the literature, firms disclose their endogenous cost information regardless of the type of competition. The underlying mechanisms and welfare implications, however, are different. Under Bertrand competition, cost disclosure helps a firm avoid aggressive investment in cost reduction to coordinate actions to the mutual advantage of the duopoly firms. Under Cournot competition, disclosing cost information enables a firm to show a hardened stance toward the competing firm. Although firms gain from their disclosure decisions under Bertrand competition, their disclosure decisions under Cournot competition place them in a prisoner’s dilemma, as both firms would be better off if they chose not to disclose their information. Consequently, consumers may lose under Bertrand competition but gain under Cournot competition.