Markets with Imperfect Information Transmission

当潜在购房者发现一套在他看来完美无缺、但已经上市一段时间的房子时，他自然会对早先没有出售的原因产生怀疑。(这里提出的分析可以应用于各种物品的销售，如耐用设备、企业或艺术品。这场讨论是在房地产市场的背景下进行的，只是为了具体。)有三种可能的解释(两种好的和一种不好的)。首先，他可能是第一个发现这座房子的潜在买家。其次，考虑到早期潜在买家的品味，这套房子的价格可能过高了。第三(也是最令人担忧的)，之前的潜在买家可能已经发现了房子里一个他看不出来的瑕疵。他认为每一种可能性的权重取决于市场的特征，如总体需求水平、买家品味的分布，以及他是否能观察到之前公布的价格的历史。此外，房屋的卖家可能会通过她的定价策略和信息披露来影响潜在买家的质量评估。本课题在一个双边不对称信息和消费者学习的模型中研究这些不完全信息传递问题，结构如下。卖家被认为知道她的房子的质量，但在不知道消费者品味的情况下发布了一段时间的询价。消费者在报价时知道自己的特殊偏好，但不知道房子的质量。出价最高的消费者承诺在质量上得到有利的检查后才购买房屋。此外，在销售季节晚些时候发现这套房子上市的消费者，会利用上市时间来更新他们对质量的评估。初步发现有几个耐人寻味的现象。首先，当检查结果没有公开记录，而消费者观察到历史上的要价时，卖家有动机在销售季节早期发布异常高的价格，以“抑制”传递给潜在买家的信号，这些买家可能会在销售季节后期发现待售房屋。如果一位在季节晚些时候到达的潜在买家知道卖家最初的要价相对较高，那么他倾向于认为房子没有早点售出是因为缺乏合适的买家(即，一个足够喜欢它的人支付了很高的要价)。相反，如果一位在季节晚些时候到达的潜在买家知道卖家最初的要价相对较低，那么他很容易相信这套房子没有早点售出，因为检测到了低质量。另一种说法是，离开高价房屋的潜在买家向随后的买家传达的关于质量的信息很少，但离开低价房屋的潜在买家传达了关于他对质量的评估的非常强烈的信号。在检验结果没有公开记录，消费者没有观察到询价历史的情况下，卖家有动机在销售季节早期发布异常低的价格，以便及早销售，从而避免不利信息的传播。这一发现的精髓体现在人们经常听到的民间智慧中，即卖家应该注意不要将初始价格定得太高。前期定价过高的危险在于，晚到、不观察价格历史的购房者自然会对房子没有早点卖出的原因产生怀疑。与一开始公布较低的初始价格相比，在这个节骨眼上纠正最初的过高定价的成本很高。事实上，如果卖家发布了过高的初始价格，可能会受到诱惑，在后期继续过高定价，导致报价被拒绝和价格下跌的恶性循环。最终，卖家可能被迫将她的房子从市场上撤下，或者与她早期获得的价格相比，以非常低的价格出售。初步调查结果还表明，要求公开提交检查结果的法律和使用信誉良好的经纪人的法律，在理论上得到了支持。这些措施可以缓解潜在买家对质量和随之而来的低效率的担忧。本文还提出了对基本模型的几个扩展以供研究。首先，调查检查成本的影响是很自然的。据推测，较高的检查成本将产生更多的低效率，但涉及信息传输的问题将在一定程度上得到改善。其次，建议研究的环境是，高品质房屋的卖家可以通过张贴高要价来表明其房屋的价值。这里一个有趣的可能性是，随着高品质房屋的卖家从均衡的“合用”阶段转向“分离”阶段，房价实际上可能会在两个时期之间上涨。这种异常的定价行为有时会在房地产和其他市场上观察到。提出的研究基本模型的最后一个扩展是考虑卖方承诺较少和讨价还价能力较小的环境。

When a prospective home buyer discovers a house that appears to him to be flawless but that has been on the market for some time, he is naturally suspicious about the reason it did not sell earlier. (The analysis proposed here can be applied to the sale of a variety of items such as a durable piece of equipment, a business, or a work-of-art. The discussion is set in the context of a housing market merely for concreteness.) There are three possible explanations (two good and one bad). First, he may be the first potential buyer to have discovered the house. Second, the house simply may have been over-priced given the tastes of earlier potential buyers. Third (and most worrisome), earlier potential buyers may have detected a flaw in the house not apparent to him. The weight he ascribes to each of these possibilities depends upon the characteristics of the market such as the overall level of demand, the distribution of buyer tastes, and whether he can observe the history of previously posted prices. Also, the seller of the house may be able to influence a potential buyer's quality assessment through her pricing strategy and disclosure of information. This project studies these issues of imperfect information transmission in a model of two-sided asymmetric information and consumer learning with the following structure. The seller is presumed to know the quality of her house, but posts asking-prices over time without knowing the realization of consumer tastes. Consumers know their own idiosyncratic preferences when making offers, but do not know the quality of the house. The consumer who makes the highest offer commits to buy the house subject to a favorable inspection on quality. In addition, consumers who discover the house on the market late in the selling season use time-on-the-market to update their assessments of quality. Preliminary findings suggest several intriguing phenomena. First, when inspection outcomes are not publicly recorded and consumers observe the history of asking prices, the seller has an incentive to post inordinately high prices early in the selling season in order to ``dampen'' the signal transmitted to prospective buyers who may discover the house for sale late in the season. If a potential buyer who arrives late in the season knows that the seller's initial asking price was relatively high, then he is apt to believe that the house did not sell earlier due to lack of a suitable buyer (i.e., one who liked it enough to pay the high asking price). Conversely, if a potential buyer who arrives late in the season knows that the seller's initial asking price was relatively low, then he is apt to believe that the house did not sell earlier due to detection of low quality. Another way of saying this is that a prospective buyer who walks away from a high-priced house conveys little information concerning quality to subsequent buyers, but a prospective buyer who walks away from a low-priced house transmits a very strong signal regarding his assessment of quality. In the case when inspection outcomes are not publicly recorded and consumers do not observe the history of asking prices, the seller has an incentive to post inordinately low prices early in the selling season in order to make an early sale and thereby avoid the transmission of unfavorable information. The essence of this finding is captured in the often-heard folk wisdom that a seller should beware of setting her initial price too high. The danger from over-pricing in early periods is that a buyer who arrives late and who does not observe the price history will naturally be suspicious about the reason the house did not sell earlier. The cost of correcting the initial over-pricing at this juncture is high as compared with the cost of posting a lower initial price to begin with. Indeed, a seller who posts too high an initial price may be tempted to continue over-pricing in later periods resulting in a vicious circle of rejected offers and falling prices. Ultimately, the seller may be forced to remove her house from the market or to make a sale at a very low price as compared with what she could have secured in early periods. Preliminary findings also suggest theoretical support for laws that require inspection outcomes to be publicly filed and for the use of reputable brokers. Such measures can ease potential buyers' concerns about quality and the concomitant inefficiency. Several extensions to the basic model are also proposed for study. First, it is natural to investigate the impact of inspection costs. It is conjectured that higher inspection costs will generate more inefficiency but that problems involving the transmission of information will be ameliorated to some degree. Second, environments in which the seller of a high-quality home can signal the value of her house by posting a high asking-price are proposed for study. An intriguing possibility here is that the price of a house might actually rise between periods as the seller of a high-quality house switches from a `pooling` to a `separating` phase of the equilibrium. Such anomalous pricing behavior is sometimes observed in real estate and other markets. A final extension of the basic model proposed for study is to consider environments in which the seller has less commitment and bargaining power.