Susan Athey, Harvard University and NBER, and
Denis Nekipelov, UC, Berkeley
A Structural Model of Sponsored Search Advertising Auctions
Sponsored links that appear beside Internet search results on the major search engines are sold using real-time auctions, where advertisers place standing bids that are entered in an auction each time a user types in a search query. The ranking of advertisements and the prices paid depend on advertiser bids as well as "quality scores" that are assigned for each advertisement and user query. Existing models assume that bids are customized for a single user query; however, in practice, queries arrive more quickly than advertisers can change their bids, and advertisers cannot perfectly predict quality scores. Athey and Nekipelov develop a new model where bids apply to many user queries, while the quality scores and the set of competing advertisements may vary from query to query. In contrast to existing models that ignore uncertainty and produce multiple equilibria, these researchers can provide sufficient conditions for the existence and uniqueness of equilibria. In addition, they propose a homotopy-based method for computing equilibria, given advertiser valuations and the distribution of uncertainty, and a structural econometric model. With sufficient uncertainty in the environment, the valuations are point-identified; otherwise, a bounds approach can be used. The researchers develop an estimator for bidder valuations, which they show is consistent and asymptotically normal, and they provide Monte Carlo analysis to assess the small sample properties of the estimator. Finally, they apply their model to historical data for several keywords. This model yields lower implied valuations and bidder profits than approaches that ignore uncertainty. Bidders have substantial strategic incentives to reduce their expressed demand in order to reduce the unit prices they pay in the auctions. These incentives are asymmetric across bidders, leading to inefficient allocation. For the keywords that Athey and Nekipelov study, the auction mechanism used in practice is not only less efficient than a Vickrey auction but also, for some keywords, it raises less revenue.
Severin Borenstein, UC, Berkeley and NBER
The Redistributional Impact of Non-Linear Electricity Pricing
Economists have long studied the efficiency impact of non-marginal-cost pricing by a regulated utility, but regulators frequently focus as much or more on the distributional impact of such rate structures. The goal of protecting low-income consumers from rising electricity rates has led to the widespread use of increasing-block pricing (IBP), under which the marginal price to the household increases as its daily or monthly usage rises. With recent increases in wholesale power costs and anticipation of significant costs of greenhouse gas emissions in the near future, forms of IBP are being implemented or considered by many utilities. There is no cost basis for differentiating marginal price of electricity by consumption level, so perhaps nowhere is the conflict between efficiency and distributional goals greater than in the use of IBP. Since the 2000-2001 California electricitycrisis, the state has adopted some of the most steeply increasing-block tariffs in electric utility history. Though a primary stated goal of this approach was to protect low-income customers, the distributional effects have not been analyzed in detail. Combining household-level utility billing data with census data on income distribution by area, Borenstein derives estimates of the income redistribution effected by these increasing-block electricity tariffs. He finds that the rate structure does redistribute income to lower-income groups, cutting the bills of households in the lowest income bracket by about 12 percent (about $5 per month). The effect would be about twice as large if not for the presence of another program that offers a different and lower rate structure to qualified low-income households. He finds that the deadweight loss associated with IBP is likely to be large relative to the transfers, at least in comparison to the deadweight losses that have been estimated for other sources of public funds. In contrast, he finds that the means-tested program transfers income with much less economic inefficiency. Most of the revenue redistributed by the IBP tariff, however, comes from the wealthiest quintile of households, while the means-tested program increases bills of middle-class and wealthy households by similar amounts, so IBP may be a more progressive structure of redistribution. In carrying out the analysis, he also shows that a common approach to studying (or controlling for) income distribution effects by using median household income within a census block group may substantially understate the potential effects.
Igal Hendel and Aviv Nevo, Northwestern University and NBER
A Simple Model of Demand Anticipation
In the presence of intertemporal substitution, static demand estimation yields biased estimates and fails to recover long-run price responses. The goal in Hendel and Nevo's paper is to present a computationally simple way to estimate dynamic demand using aggregate data. Previous work on demand dynamics is computationally intensive and relies on (hard to obtain) household level data. These authors estimate the model using store level data on soft drinks and find: 1) a disparity between static and long-run estimates of price responses; and 2) heterogeneity consistent with sales being driven by discrimination motives. Their model's simplicity allows them to compute mark-ups implied by dynamic pricing.
Ginger Jin, University of Maryland and NBER, and
Seth Freedman, University of Maryland
Learning by Doing with Asymmetric Information: Evidence from Prosper.com
Freedman and Jin examine the nature of information asymmetry in online peer-to-peer (P2P) lending markets. These markets use the Internet to match individual borrowers and lenders of consumer loans without financial institutions as intermediaries. Like other anonymous interactions, P2P lending may face additional information asymmetry as compared to offline because P2P lenders have less access to "hard" information such as borrower credit history, income, or employment. However, the shortage of "hard" information could be mitigated by "soft" information via online social networks. The researchers examine this tradeoff using data from all requested and funded loans between June 1, 2006 and July 31, 2008 on Prosper.com. They find first that Prosper lenders understand the ordinal difference across credit grades, but the incomplete disclosure of a borrower's credit history leads to additional adverse selection relative to traditional markets. Second, some social networks help to mitigate information asymmetry and others do not, depending on the institutional incentives. Third, lenders, especially those who joined Prosper early, did not fully understand the market risk. The authors estimate, on average, lenders would have expected an annualized internal rate of return of -0.62 percent to -1.38 percent on a dollar invested if they had correctly understood the risk distribution of Prosper loans. However, lender learning is effective in reducing the risk of funded loans over time. As a result, the market has excluded more and more sub-prime borrowers and evolved towards the population served by traditional credit markets.
Ali Hortascu, University of Chicago and NBER;
Gregor Matvos, University of Chicago;
Chad Syverson, University of Chicago and NBER, and
Sriram Venkataraman, Emory University
Are Consumers Affected by Durable Goods Makers' Financial Distress? The Case of Auto Manufact
Theory suggests that the financial decisions of durable goods makers can impose externalities on their consumers. Namely, the consumption stream that durable goods provide frequently depends on services provided by the manufacturer itself (for example, warranties, spare parts availability, maintenance, and upgrades). Bankruptcy of a manufacturer, or even the possibility thereof, threatens this service provision and as a result can substantially reduce the value of its products to their current owners. Hortacsu, Matvos, Syverson, and Venkataraman test whether this hypothesis holds in one of the largest durable goods markets, automobiles. They use data on prices of millions of used cars sold at wholesale auctions around the United States during 2006-8. They find that an increase in an auto manufacturer's financial distress (as measured by an increase in its CDS spread) does result in a contemporaneous drop in the prices of its cars at auction, controlling for a host of other influences on price. The estimated effects are statistically and economically significant. Furthermore, cars with longer expected service lives (lower mileage or better condition cars) see larger price declines than those with shorter remaining lives. These patterns do not seem to be driven solely by reduced demand from auto dealers affiliated with the troubled manufacturers.
Francesco Decarolis, University of Chicago
When the Highest Bidder Loses the Auction: Theory and Evidence from Public Procurement
When bids do not represent binding commitments, the use of a first-price sealed-bid auction favors bidders who are penalized less from reneging on their bids. They are the most likely to win, but also the most likely to default on their bid. Decarolis studies two methods often used in public procurement to deal with this problem: 1) augmenting the first price auction with an ex-post verification of the responsiveness of the bids; and 2) using an average bid auction in which the winner's bid is closest to the simple average of all the bids. The average bid auction is new to economics but has been proposed in civil engineering literature and adopted by several countries. Decarolis shows that when penalties for defaulting are asymmetric across bidders, and when their valuations are characterized by a predominant common component, the average bid auction is preferred over the standard first price by an auctioneer IF the costs due to the winner's bankruptcy are high enough. Depending on the cost of the ex-post verification, the average bid auction can be dominated by the first price with screening. Decarolis uses a new dataset of Italian public procurement auctions, run alternately using a form of the average bid auction or the augmented first price, to structurally estimate the bids' verification cost, the firms' mark up, and the inefficiency generated by the average bid auctions.
Katja Seim, University of Pennsylvania,and
Joel Waldfogel, University of Pennsylvania and NBER
Public Monopoly and Economic Efficiency: Evidence from the Pennsylvania Liquor Control Board's Entry Decisions
When fixed costs are present, markets can deliver an excessive number of products in inefficient product-space locations under free entry. Seim and Waldfogel study product provision under an alternative form of market organization -- state-run monopoly -- using data from the Pennsylvania Liquor Control Board (PLCB), the state's monopolist wine and spirits retailer. Using information on store location choices, prices, wholesale cost, and sales to uncover the goals implicit in entry decisions, they estimate a model of demand for liquor as a function of the price, distance to stores, and other demographic characteristics. In their counterfactual analyses, they calculate the configurations of stores tha: 1) maximize welfare, 2) maximize profit, and 3) mimic free entry. They then compare the actual PLCB store network to these benchmarks. They find that the PLCB's implicit goal is better characterized by welfare than profit maximization. The state has roughly three times the number of stores that would maximize profit and about one fifth more than would maximize welfare. The researchers also find that atomistic location decisions would dissipate between 6.9 and 11.5 percent of welfare from excessive clustering and entry of stores arising from the failure to internalize business stealing.