唧唧堂：MS 营销科学2022年5-6月论文摘要11篇

消费者通过多种设备、浏览器和机器与公司互动；这些交互记录通常成为同一个消费者的不同身份标识符。如未能正确匹配一个消费者的不同身份将导致消费者识别的碎片化。本文研究了身份碎片偏差，即使用碎片化数据产生的估计偏差。利用一个形式化的框架，我们分解了数据碎片导致的估计偏差的影响因素，并讨论了偏差的方向。与传统观点相反，在标准假设下，这种偏差不能被定义或限制。相反，即使在实验环境中，也可能发生向上偏差和符号反转。文章比较了几种纠正措施，并讨论了它们的优点和注意事项。

Consumers interact with firms across multiple devices, browsers, and machines; these interactions are often recorded with different identifiers for the same consumer. The failure to correctly match different identities leads to a fragmented view of exposures and behaviors. This paper studies the identity fragmentation bias, referring to the estimation bias resulted from using fragmented data. Using a formal framework, we decompose the contributing factors of the estimation bias caused by data fragmentation and discuss the direction of bias. Contrary to conventional wisdom, this bias cannot be signed or bounded under standard assumptions. Instead, upward biases and sign reversals can occur even in experimental settings. We compare several corrective measures and discuss their advantages and caveats.

论文原文：Tesary Lin, Sanjog Misra (2022) Frontiers: The Identity Fragmentation Bias. Marketing Science 41(3):433-440.

https://doi.org/10.1287/mksc.2022.1360

2.前沿：使用自然语言生成支持内容营销

自然语言生成（natural language generation, NLG）的进步促进了数字语音助理和聊天机器人等技术的发展。在这项研究中，我们展示了NLG如何通过在搜索引擎优化（search engine optimization, SEO）中为网站的登录页起草内容来支持内容营销。传统的SEO项目依赖于手动生产的内容，这既耗时又昂贵。为了解决与生成SEO内容相关的成本问题，我们提出了一种使用最先进的NLG的半自动化方法，并证明了内容编写机可以创建独特的、类似人类的SEO内容。作为研究的一部分，我们证明，尽管机器生成的内容在搜索引擎中表现良好，但人工编辑器的作用仍然至关重要。通过将机器生成内容与人类细化后的文本与传统人类编写的SEO文本进行比较，我们发现，经过修订的机器生成文本在许多人类感知维度上与SEO专家创建的文本几乎无法区分。我们在两个行业进行了实地实验，以证明我们的方法，并表明最终的SEO内容在搜索引擎排名中优于人类作者（包括SEO专家）创建的内容。此外，我们还说明了我们的方法如何大幅降低与内容营销相关的生产成本，提高其投资回报。

Advances in natural language generation (NLG) have facilitated technologies such as digital voice assistants and chatbots. In this research, we demonstrate how NLG can support content marketing by using it to draft content for the landing page of a website in search engine optimization (SEO). Traditional SEO projects rely on hand-crafted content that is both time consuming and costly to produce. To address the costs associated with producing SEO content, we propose a semiautomated methodology using state-of-the-art NLG and demonstrate that the content-writing machine can create unique, human-like SEO content. As part of our research, we demonstrate that although the machine-generated content is designed to perform well in search engines, the role of the human editor remains essential. Comparing the resulting content with human refinement to traditional human-written SEO texts, we find that the revised, machine-generated texts are virtually indistinguishable from those created by SEO experts along a number of human perceptual dimensions. We conduct field experiments in two industries to demonstrate our approach and show that the resulting SEO content outperforms that created by human writers (including SEO experts) in search engine rankings. Additionally, we illustrate how our approach can substantially reduce the production costs associated with content marketing, increasing their return on investment.

论文原文：Martin Reisenbichler, Thomas Reutterer, David A. Schweidel, Daniel Dan (2022) Frontiers: Supporting Content Marketing with Natural Language Generation. Marketing Science 41(3):441-452.

https://doi.org/10.1287/mksc.2022.1354

3.耐用品采用的可视化和同侪影响

对同侪影响在耐用品采用中发挥作用的潜在机制的理解可能会增强营销人员利用这些渠道的能力。在本文中，我们评估了同伴采纳决策的可见性是否会导致更大的同伴影响。我们的研究对象是住宅屋顶太阳能电池板。我们确定屋顶太阳能电池板相对于邻近道路和视觉障碍物（如植被）的外生位置和方向，以探究地理上邻近的同行安装是否会增加家庭采用太阳能的可能性（如果从道路上可以看到）。我们发现，在至少500米的距离内，同一街道上对等设施的总可见度会对太阳能采用决策产生积极影响。相比之下，我们还发现100米范围内的不可见太阳能电池板也产生了积极影响，这可能是通过其他渠道（如口碑传播）产生的，或局部的不可观测因素造成的。同伴可见性的主效应受安装太阳能的经济价值的调节，这通过视觉信息展示了社会学习的证据。

The underlying channels through which peer influence operates in durable good adoption can affect the ability of marketers to leverage them. In this paper, we assess whether the visibility of peers’ adoption decisions leads to greater peer influence. The context we study is residential rooftop solar panels. We exploit the plausibly exogenous location and orientation of peers’ rooftop solar panels relative to proximate roadways and visual obstructions, such as vegetation, in order to determine whether geographically proximate peer installations increase a household’s probability of solar adoption more if they are visible from the road. We find that the total angle of visibility of peer installations on the same street positively affects solar adoption decisions at distances of at least 500 meters (m). In contrast, we only find a positive effect of nonvisible solar arrays within 100 m, which may be due to causal peer influence via other channels, such as word of mouth, or very localized unobservable effects. The effect of peer visibility is moderated by the economic value that the peers receive from installing solar, providing suggestive evidence of social learning through visual information.

论文原文：Bryan Bollinger, Kenneth Gillingham, A. Justin Kirkpatrick, Steven Sexton (2022) Visibility and Peer Influence in Durable Good Adoption. Marketing Science 41(3):453-476.

https://doi.org/10.1287/mksc.2021.1306

4.网络效应个性化定价的风险

本研究考察了网络效应是如何影响个性化定价对企业和消费者的影响的。网络效应是一种现象，即对产品或服务的需求越大，该产品或服务对每个消费者的价值就越高。我们从价格、需求、利润和消费者剩余等方面研究了网络效应对个性化定价的影响。公司普遍认为，个性化定价可以捕捉消费者支付意愿的异质性，满足不同消费群体的需求，从而增加利润。利用博弈论模型，我们发现了个性化定价不仅会减少需求，还会降低企业利润，减少所有消费者剩余的条件。具体而言，我们的模型显示，在具有网络效应的市场中，除非预期消费者异质性的影响高于网络效应的影响，否则个性化定价会减少需求，降低利润，减少所有消费者的剩余，从而导致市场不足。

This research examines how network effects, a phenomenon whereby the larger the demand for a product or service, the higher the value of that product or service for each consumer, influence the impact of price personalization on firms and consumers. We study the impact of network effects on price personalization in terms of price, demand, profit, and consumer surplus. One belief commonly held by companies is that price personalization increases their profit by providing means for capturing heterogeneity in consumers’ willingness to pay and enticing demand from heterogeneous consumer segments. Using a game-theoretical model, we find conditions under which price personalization not only shrinks demand, but also decreases the profit of the firm and reduces surplus of all consumers. Specifically, our model shows, in markets with network effects, unless when the impact of expected consumer heterogeneity is higher than the impact of network effects, personalized pricing reduces demand, lowers profit, reduces surplus of all consumers, and thus causes a deficiency in the market.

论文原文：Bita Hajihashemi, Amin Sayedi, Jeffrey D. Shulman (2021) The Perils of Personalized Pricing with Network Effects. Marketing Science 41(3):477-500.

https://doi.org/10.1287/mksc.2021.1323

5. 垂直市场中的网络形成和议价：以健康保险中的窄带网络为例

本文建立了一个具有多个上下游企业的垂直市场模型。模型显示，纵向关系网络、为这些关系协商的合同条款以及向最终客户收取的下游价格都是内生的。此外，文章还提供了一个估计过程，以计算可伸缩的方式为观测到的网络和合同提供了充分的解释。本文将该框架实际应用于医院和健康保险市场，并使用该模型量化了“网络充分性监管”的影响，该监管旨在扩大患者的医院选择范围，但可能会因价格上涨而适得其反。文章证明了“严格”的监管会产生不利影响。文章还表明，市场反应在很大程度上受到涉及多家保险公司之间以及上游谈判和下游定价之间相互作用的复杂机制的影响。

This paper develops a model of vertical markets with multiple upstream and downstream firms. Networks of vertical relationships, negotiated contract terms for those relationships, and downstream prices charged to end customers all arise endogenously. In addition, I provide an estimation procedure to fully rationalize the observed network and contracts in a computationally scaleable manner. I empirically apply the framework to the market between hospitals and health insurers. I use the model to quantify the effects of “network adequacy regulations,” which are designed to broaden patients’ hospital choice sets but can backfire by increasing prices. I show that “tight” regulations have adverse effects. I also show market response is substantially impacted by complex mechanisms involving the interplay among multiple insurers and between upstream negotiation and downstream pricing.

论文原文：Soheil Ghili (2022) Network Formation and Bargaining in Vertical Markets: The Case of Narrow Networks in Health Insurance. Marketing Science 41(3):501-527.

https://doi.org/10.1287/mksc.2021.1331

6. 可跳过广告：数字媒体平台上的互动广告

数字媒体平台通常使用可跳过的广告，这使观众可以在看到一些信息后跳过部分广告，直接跳转到所需的内容。使用可跳过广告的平台有时仅可在观众观看整个广告后向广告商收费。本文开发了一个观众从广告商接收增量信息的动态模型，并将其嵌入到双边广告市场环境中。研究结果表明，可跳过广告在客户转化率上效果较差。然而，可跳过广告为平台带来了更多的观众，进而吸引了更多的广告商。这表明，对于一个新兴和不断增长的平台来说，使用可跳过广告是一种有利可图的策略，但对于一个处于饱和市场的平台来说，这并不一定是有效策略。

Digital media platforms commonly use the skippable ad format, which gives a viewer the option to skip part of an advertisement after seeing some limited information and jump directly to the desired content. It also enables these platforms to charge advertisers only when viewers attend to the entire ad. We develop a dynamic model of a viewer receiving incremental information from the advertiser and embed it in a two-sided market setting with an advertising market. Our results show that skippable ads can be less effective overall in converting existing viewers to advertisers. However, skippable ads bring more viewers to the platform and, in turn, induce more advertisers. This suggests that a switch to the skippable format is a profitable strategy for an emerging and growing platform but not necessarily for one in a saturated market.

论文原文：Anthony Dukes, Qihong Liu, Jie Shuai (2021) Skippable Ads: Interactive Advertising on Digital Media Platforms. Marketing Science 41(3):528-547.

https://doi.org/10.1287/mksc.2021.1324

7.成为金牌用户：常旅客身份激励与道德风险

本文研究了飞行常旅客在身份升级的过程中购买行为的改变。利用美国一家领先航空公司的数据，本文对理论预测进行了实证检验，即旅行者的转换成本随着他们获得身份升级的进程而动态变化。文章发现当消费者接近达到某身份所需的积分累积目标时转换成本会增加。这些转换成本反映了预定航空公司时行为的变化：即使该航空公司的吸引力不如竞争对手，旅客（在到即将到达下一积分身份时）也更有可能选择该航空公司，并支付比其他航空公司更高的价格。当旅行者以远远超过目标配速的速度累积积分时，这一效应会减弱。这一转换成本增加的效应对位于航空枢纽城市的消费者和商务旅行者来说更为明显。此外，当消费者出差且不由自己承担机票费用时，他们展示了更强烈的支付意愿。这一效应表明，商务旅行引发的非对称激励在很大程度上解释了商务旅行者和休闲旅行者之间的异质性，而道德风险可能是常旅客身份激励能够盈利的重要原因。

We study how frequent-flyer program members change their purchase behaviors as they progress toward achieving elite status. Using data from a leading U.S. airline, we empirically test the theoretical prediction that travelers’ switching costs vary dynamically with their progress toward attaining status. We show evidence for increased switching costs as the consumer approaches the target pace of point accumulation required to attain status. These switching costs reflect changes in booking behavior with the airline: Travelers become more likely to choose the airline even when it is less appealing than its competitors, and to pay higher prices than they otherwise would. These responses are reduced when travelers accumulate points at a rate substantially ahead of the target pace. The increase in switching costs is more pronounced for consumers at a hub of the airline and for business travelers. Moreover, we document a stronger willingness-to-pay response when consumers are less likely to shoulder the ticket costs themselves because they are traveling for business. This response suggests that asymmetric incentives induced by business travel explain much of the heterogeneity between business and leisure travelers, and moral hazard may be responsible for a large part of the profitability of frequent-flyer status incentives.

论文原文：A. Yeşim Orhun, Tong Guo, Andreas Hagemann (2022) Reaching for Gold: Frequent-Flyer Status Incentives and Moral Hazard. Marketing Science 41(3):548-574.

https://doi.org/10.1287/mksc.2021.1341

8. 赌场头奖事件的社会和时空影响

头奖（Jackpos）是赌场博彩业中一种重要的广告和促销工具。本文使用一个独特的数据集来衡量头奖事件对后续赌博行为的影响。使用双重差分法，研究针对三类玩家研究了头奖的后续影响：（1）头奖中奖者，（2）他们的同伴或伙伴，以及（3）头奖事件的旁观者。研究发现，头奖事件增加了赌博支出（平均老虎机下注金额）和头奖中奖者的游戏频率。在中奖活动后的两个小时内，中奖者的每次下注金额增加39美元，下注次数增加33%，头奖中奖者的同伴同一时间内下注次数增加21%。对于旁观者来说，效果较弱，影响在头奖后大约一小时就会消失。对头奖投资回报的研究表明，49%的头奖的发生对赌场是有利可图的。本文对赢家反应的潜在机制的研究支持了热手效应（人们根据近期大幅出现的情况，而不采取分析就直接采取措施），而非赌资效应或赌徒谬误。

Jackpots are an important advertising and promotional tool in the casino gambling industry. In this paper, we use a unique data set to measure the impact of a slot machine jackpot event on subsequent gambling behavior. We use a difference in differences method to partition jackpot value into its impact on the post-jackpot behavior of three categories of players: (1) jackpot winners, (2) their peers or partners, and (3) bystanders who are in the proximity of the jackpot event. We find that jackpot events increase gambling expenditures (average slot machine bet amount) and frequency of plays by jackpot winners. The average impact on the jackpot winner is a $39 increase in bet amount per play and a 33% increase in the number of plays for the two-hour period after the jackpot event, whereas the impact on peers is a 21% increase in number of plays in the same period. For bystanders, effects are weaker and dissipate about an hour after the jackpot. Our study of jackpot return on investment shows that 49% of jackpots are profitable for the casino. Our study of the underlying mechanism of winners’ response favors the hot hand effect rather than the house money or gambler’s fallacy effects.

论文原文：Hee Mok Park, Joseph Pancras (2022) Social and Spatiotemporal Impacts of Casino Jackpot Events. Marketing Science 41(3):575-592.

https://doi.org/10.1287/mksc.2021.1342

9.通过从属关系影响社交媒体影响者

社交媒体影响者是经常发布产品推荐的品类爱好者。公司有时会付钱给有影响力的人，让他们给出偏向于该公司的产品评论。本文提出以下研究问题。首先，从公司的角度来看，与社交媒体影响力者的最佳关联关系（如果有的话）是什么？关联关系可能对社交媒体影响者的评论有积极影响，但是也会负面影响评论的说服力。其次，由于关联评论往往对公司有利，那对消费者福利有何影响？研究发现，关联决策取决于获取信息的成本、消费者的先验知识和知名度以及披露制度。当消费者先验信念较低时，公司需要减少关联关系或根本不关联，以保持影响者的说服力，以及保持消费者对影响者的跟随。相反，当消费者先验信念很高时，公司与影响者充分合作，可以最大限度地提高知名度并减少负面评论。本文还表明，如果信息获取成本相对较高，企业的参与度可以得到帕累托改善，即部分披露规则可能会增加消费者福利。

Social media influencers are category enthusiasts who often post product recommendations. Firms sometimes pay influencers to skew their product reviews in favor of the firm. We ask the following research questions. First, what is the optimal level of affiliation (if any) from the firm’s perspective? Affiliation introduces positive bias to the influencer’s review but also decreases the persuasiveness of the review. Second, because affiliated reviews are often biased in favor of the firm, what is the impact of affiliation on consumer welfare? We find that the affiliation decision depends on the cost of information acquisition, the consumer’s prior and awareness, and the disclosure regime. When the consumer’s prior belief is low, the firm needs to affiliate less closely or not at all in order to preserve the influencer’s persuasiveness, the change in the consumer’s belief following the influencer’s review. In contrast, when the consumer’s prior belief is high, the firm fully affiliates with the influencer to both maximize awareness and prevent a negative review. We also show that the firm’s involvement can be Pareto improving if the information acquisition cost is relatively high, and a partial disclosure rule may increase consumer welfare.

论文原文：Amy Pei, Dina Mayzlin (2021) Influencing Social Media Influencers Through Affiliation. Marketing Science 41(3):593-615.

https://doi.org/10.1287/mksc.2021.1322

10.数字内容消费中的时间非一致性偏好与策略性自我控制

本研究探讨消费者在数字内容消费中的时间非一致性偏好（time-inconsistent preferences）以及他们的策略性自我控制行为（strategic self-control behaviors）。本文数据来自中国一家数字图书平台，该平台的用户可以按章节购买图书或成为会员。数据显示，三分之一的消费者自始至终选择按章节购买图书，尽管按月付费的会员会大大降低花费。本文提出了一个动态结构模型将时间非一致性行为与策略性自我控制整合来解释这一行为。文章首先通过分析揭示了唯一平衡点的存在，并发现对消费者而言，在稳定状态下过度支付阅读费用可能是最佳选择。数据估计模型的结果表明，有很大一部分对价格敏感的消费者愿意支付过高的价格来抑制未来的消费。这表明，取消每章付费选项将损害消费者福利和平台利润。然而，取消每月订阅计划将增加平台的利润，但会降低消费者福利。本文发现了一种新的非线性定价方案，并解释它如何使消费者和平台共同受益。

This study examines consumers’ time-inconsistent preferences in digital content consumption and their strategic self-control behaviors. We used a unique data set obtained from a major digital book platform in China, where consumers can pay either by chapter or a monthly subscription. A third of consumers consistently chose to pay by chapter, even though a monthly subscription would significantly reduce the monetary cost. We propose a dynamic structural model that incorporates time-inconsistent preferences and strategic self-control behaviors to rationalize such behavior. We first analytically demonstrate the existence of a unique equilibrium and show how, under steady states, overpaying for reading can be optimal for consumers. We then estimate the model from the data. The results show that there is a large segment of price-sensitive consumers who are willing to overpay to curb future consumption. Our counterfactuals show that eliminating the pay-per-chapter option would hurt consumer welfare and the platform’s profit. Eliminating the monthly subscription plan, however, would increase the platform’s profit but reduce consumer welfare. We introduce a novel nonlinear pricing plan with a volume surcharge and illustrate how it can benefit both consumers and the platform.

论文原文：Shuo Zhang, Tat Y. Chan, Xueming Luo, Xiaoyi Wang (2021) Time-Inconsistent Preferences and Strategic Self-Control in Digital Content Consumption. Marketing Science 41(3):616-636.

https://doi.org/10.1287/mksc.2021.1318

11. 用匹配的行政数据衡量收入和财富对自有品牌需求的影响

行业情绪将收入和财富与自有品牌需求联系起来。直觉上我们认为收入和财富的减少增加了对（更便宜的）自有品牌的需求。虽然在加总层面的时间序列分析中很难建立合理的因果关系，但这种分析具有足够的效应量。个人层面的数据虽然能够极大地促进因果估计的合理性，但在测量上具有难度。本文通过将来自荷兰2011-2018年间的家庭扫描仪数据与行政数据进行联系，克服了这些挑战。文章分析了家庭扫描仪数据中测量的个人层面的自有品牌份额，将其作为收入和财富的函数，并对所有家庭成员（而不仅仅是主要收入者）进行汇总。文章发现，与使用行政数据相比，使用调查得到的收入数据中的家庭内部差异显著减弱了收入的影响，但效应仍然存在。此外，财富的变化对自有品牌份额存在具有经济意义的微弱显著效应。

Industry sentiment links income and wealth to private-label demand. The intuition is that decreasing income and wealth increases the demand for (cheaper) private labels. Whereas plausible causality is harder to establish in aggregate time series analyses, such analyses suggest large effect sizes. An individual-level perspective greatly facilitates plausibly causal estimates but poses measurement challenges. We overcome these challenges by linking household scanner data to administrative data. We analyze individual-level private-label shares measured in household scanner data as a function of income and wealth, both from a linked administrative database in the Netherlands in the period from 2011 to 2018 and aggregated over all household members (rather than only from the main earner). We find that relying on within-household variation in surveyed income data significantly attenuates income effects relative to using that from administrative data. Yet, we still find an economically small effect. In addition, changes in wealth have at most an economically small effect on private-label shares.

论文原文：Calogero Brancatelli, Adrian Fritzsche, Roman Inderst, Thomas Otter (2022) Measuring Income and Wealth Effects on Private-Label Demand with Matched Administrative Data. Marketing Science 41(3):637-656.

https://doi.org/10.1287/mksc.2021.1334