Management Insights

Is Noise Trading Cancelled Out by Aggregation? (p. 1047)

Hongjun Yan

Can individual investor bias cause speculative stock price bubbles in aggregate? Conventional wisdom suggests that investors' independent biases should cancel each other out and have little impact on aggregate stock price. Dr. Yan finds the opposite; biases often have a significant impact on the equilibrium even if they are independent across investors. Across a number of different assumptions on the form of investor bias, Dr. Yan shows that a modest amount of bias can have a large impact on the equilibrium. Investor bias can cause an initial run-up of the stock price and make optimistic investors richer in the short term, which then further pushes the stock price up and leads to lower future returns in the longer term. This effect can lead to such a level of price overshooting that a stock might have a negative expected future return. Conversely, an initial drop of the stock price leads to higher future returns. The insight for management: Knowing the market psychology as well as the market fundamentals can be important for earning superior returns.

eBay's Crowded Evenings: Competition Neglect in Market Entry Decisions (p. 1060)

Uri Simonsohn

A Smashburger and a Five Guys Burgers and Fries open on the same block next to the University of Dayton in the same month. Do firms such as these neglect competition when making entry decisions? Dr. Simonsohn examines this question by analyzing the time of day at which eBay sellers set their auctions to end for nearly 15,000 auctions of DVDs. Conventional wisdom (and the advice of many ``how to sell on eBay'' books) is to set auction ending times consistent with peak bidding times, but on eBay the strategy is more conventional than wise. Dr. Simonsohn shows that the peak for sellers is actually more pronounced than the peak for buyers; a disproportionate share of auctions end during peak bidding hours. Worse still for sellers: Peak hours exhibit lower selling rates and prices. This overcrowding of sellers at peak times appears to be by choice and not by chance; peak listing is more prevalent among sellers likely to have chosen ending time strategically as demonstrated by their willingness to pay a premium for their auction to close during peak hours. This result suggests that assessing rather than assuming the rationality of firm behavior may be wise. The insight for management: A seller must know the competition as well as the customer.

Vertically Differentiated Simultaneous Vickrey Auctions: Theory and Experimental Evidence (p. 1074)

Ravi Bapna, Chrysanthos Dellarocas, Sarah Rice

Vickrey auctions are commonplace: They are used on eBay, for procurement of services by major organizations, and even for oil and mineral rights. Vertical differentiation can arise from differences in item quality, seller reliability, etc. Drs. Bapna, Dellarocas, and Rice study settings where a number of sellers simultaneously offer vertically differentiated Vickrey auctions for imperfect substitute goods. They find that low bidders will tend to place positive bids in all available auctions whereas higher bidders tend to follow more targeted strategies, focusing their "serious" bids on fewer and, generally, higher-quality auctions. The lack of coordination in auction selection among bidders might lead to high bidding in one auction and no bidding in another. They test the theory in a controlled laboratory experiment and find that low-risk-type bidders tend to crowd on the highest auction and will pay a premium for the certainty that it offers, whereas high-risk-type bidders fail to appropriately adjust for risk associated with the lowest auction, leading to overbidding. These behaviors lead to an interesting result whereby bids are concentrated on the highest and lowest auctions, bypassing intermediate auctions. The insight for management: Although auctions are touted for extracting surplus and driving higher revenues, the ``price of anarchy'' in a multiple-auction setting is the loss of efficiency arising from uncoordinated buyer behavior.

Staffing Call Centers with Uncertain Demand Forecasts: A Chance-Constrained Optimization Approach (p. 1093)

Itai Gurvich, James Luedtke, Tolga Tezcan

Call centers often must staff for multiple customer and agent types while trying to achieve high-quality service in the face of high call volume uncertainty. Drs. Gurvich, Luedtke, and Tezcan introduce a formulation of the staffing problem that requires that the minimum service levels are met with high probability rather than meeting some average level of performance over time. Despite uncertain demand rates, their two-step solution allows them to approach the problem as if arrival rates are known. The result is a solution that is feasible with respect to the uncertain demand constraint and nearly optimal for large call centers. The insight for management: Clever approaches to a problem with high demand uncertainty and dedicated resource types can be used to reduce errors due to demand forecast and thereby improve staffing and customer service.

Workplace Peers and Entrepreneurship (p. 1116)

Ramana Nanda, Jesper B. Sørensen

How much do entrepreneurial tendencies transfer between coworkers? Is an individual more likely to become an entrepreneur if his or her coworkers have been entrepreneurs before? Drs. Nanda and Sørensen examine whether the likelihood of entrepreneurial activity is related to the prior career experiences of an individual's coworkers. They test their theory with a unique panel data set from Denmark covering more than a quarter of a million coworkers over an 18-year span. They find that coworkers can increase the likelihood that an individual will perceive entrepreneurial opportunities and can increase his or her motivation to pursue those opportunities. Furthermore, they find that peer influences are strongest for those who have less exposure to entrepreneurship in other aspects of their lives. The insight for management: A more entrepreneurial environment might be fostered through the juxtaposition of experienced entrepreneurs and employees with a less entrepreneurial background.

Efficient Structures for Innovative Social Networks (p. 1127)

William S. Lovejoy, Amitabh Sinha

What lines of communication among members of an organization are most productive in the early, idea-generation phase of innovation? Drs. Lovejoy and Sinha investigate this question with a model of knowledge transfer operating through a social network. They find that idea generation is accelerated when people in the organization dynamically churn through a large set of conversational partners over time, which reduces information bottlenecks. They suggest that group meetings, in which the content of conversations is available to all for consideration, are another way to learn in parallel and accelerate the ideation process, although for complex problems they may not offer significant advantages over the best decentralized networks. The insight for management: Water cooler conversations and staff meetings might seem unproductive, but they may be an important ingredient of innovation in an organization.

Detecting Management Fraud in Public Companies (p. 1146)

Mark Cecchini, Haldun Aytug, Gary J. Koehler, Praveen Pathak

How can auditors and government regulators better allocate their time investigating companies to avoid corporate disasters like Enron? Drs. Cecchini, Aytug, Koehler, and Pathak provide a data mining methodology for detecting management fraud using basic financial data. They collected a large data set of quantitative financial attributes for 205 fraudulent and 6,427 nonfraudulent public companies. Their approach correctly labeled 80% of the fraudulent cases and 90% of the nonfraudulent cases. The results show that the methodology has predictive value because, using only historical data, it was able to distinguish fraudulent from nonfraudulent companies in subsequent years. The insight for management: Auditing firms and government regulators can better target their efforts through identifying the most likely fraudulent companies before planning their detailed investigations.

Why Are Bad Products So Hard to Kill? (p. 1161)

Duncan Simester, Juanjuan Zhang

RCA's SelectaVision was a disk-based format for home movies at the advent of the VCR, and it was a monumental market failure. Yet RCA continued to invest in marketing and development of that product until it was acquired by General Electric in 1986. Products will fail (as evidenced by the DeLorean and New Coke in the 1980s alone), but it is puzzling that firms often continue to invest in product development projects when they should know that demand will be low. Drs. Simester and Zhang argue that bad products are hard to kill because firms face an inherent conflict when designing managers' incentives. Rewarding success encourages managers to forge ahead even when demand is low. To avoid investing in low-demand products, the firm must also reward decisions to kill products. However, rewarding managers for killing products effectively undermines the rewards for success. The inability to resolve this tension forces the firm to choose between paying an even larger bonus for success and accepting continued investment in low-demand products. The insight for management: To avoid throwing good money after bad, a check-and-balance system should be in place to temper the unbridled optimism typically embodied in product managers.

Structural Estimation of the Effect of Out-of-Stocks (p. 1180)

Andrés Musalem, Marcelo Olivares, Eric T. Bradlow, Christian Terwiesch, Daniel Corsten

What is the cost of a stockout? At its simplest, it is a lost or deferred sale, coupled with some less-quantified customer dissatisfaction. However, in many cases, the purchaser may just purchase a near substitute. Drs. Musalem, Olivares, Bradlow, Terwiesch, and Corsten look at the impact of substitute products on the cost of a stockout. Their estimation method uses store-level data on sales and partial information on product availability. The model allows for flexible substitution patterns and can be applied to data from multiple markets in categories with a relatively large number of alternatives. It also applies to slow-moving products as well as frequent out-of-stocks. They illustrate how the model can be used to assist the decisions of a store manager in two ways. First, they show how to quantify the lost sales induced by out-of-stock products. Second, they provide insights on the financial consequences of out-of-stocks and suggest price-promotion policies that can be used to help mitigate their negative economic impact, which run counter to simple commonly used methods. The insight for management: An accurate estimate of the cost of a stockout is essential for making inventory-carrying decisions and depends on a number of product-family-related factors.

Is Regime Switching in Stock Returns Important in Portfolio Decisions? (p. 1198)

Jun Tu

How should your portfolio change during bull and bear markets? Statistical analysis of historical data might lead to a portfolio decision, but, given that bull markets have both higher mean return and lower variance in returns than bear markets, parameters based on those historical data may be misleading if the market regime turns to a bear market. Dr. Tu shows that accounting for whether you are in a bull or bear market is critical: Ignoring regime switching might cost your portfolio more than 2% per year and as much as 10%. These results suggest that the more realistic regime-switching model is fundamentally different from the commonly used single-state model and hence should be employed instead in portfolio decisions irrespective of concerns about model or parameter uncertainty. The insight for management: To increase returns and reduce the risk of misestimating portfolio parameters, portfolios based on historical performance data should be evaluated whether the current market dynamic matches that over which the portfolios were evaluated.