Search

We explore how information ambiguity and traders’ attitudes toward ambiguity affect expectations and asset prices under three different market institutions. Specifically, we test the prediction of Epstein and Schneider (2008) that information ambiguity will lead market prices to overreact to bad news and to underreact to good news. We find that such an asymmetric reaction exists and is strongest in individual prediction markets. It occurs to a lesser extent in single price call markets. It is weakest of all in double auction markets, where buyers’ asymmetric reaction to good/bad news is cancelled out by the opposite asymmetric reaction of sellers.

We report the results of an experiment designed to study the role of trading institutions in the formation of bubbles and crashes in laboratory asset markets. We employ three trading institutions: Call Market, Double Auction, and T\^atonnement. The results show that bubbles are significantly smaller in uniform-price institutions than in Double Auction. We reproduce this and other critical patterns of the data by calibrating a heterogeneous agent model with fundamental and myopic-noise traders. The model produces larger bubbles under Double Auction because multiple trades occur within a period, amplifying the impact of myopic traders with positive bias on transaction prices.

We study uniform-price double auctions augmented with a class of subsidy schemes. Using these subsidies, any profile of linear demand schedules can be implemented as an equilibrium outcome. By revenue equivalence, all other mechanisms which implement linear auction equilibria are essentially equivalent to some subsidy scheme in our class. We show that, under a linear dependency condition on primitives, fully efficient, budget balanced, and individually rational trade is possible. Thus, the welfare loss from equilibria in uniform-price auctions without subsidies is a non-fundamental distortion, which can be fixed with better mechanism design. However, we show that monopolist trading platforms have incentives to reduce allocative efficiency to increase revenue, even as the number of traders becomes large.

We test whether anchoring affects people's elicited valuations for a bottle of wine in individual decision making and in markets. We anchor subjects by asking them if they are willing to sell a bottle of wine for a transparently uninformative random price. We elicit subjects' Willingness-To-Accept for the bottle before and after the market. Subjects participate in a double auction market either in a small or a large trading group. The variance in subjects' Willingness-To-Accept shrinks within trading groups. Our evidence supports the idea that markets have the potential to diminish anchoring effects. However, the market is not needed: our anchoring manipulation failed in a large sample.

Information dissemination and aggregation are key economic functions of financial markets. How intelligent do traders have to be for the complex task of aggregating diverse information (i.e., approximate the predictions of the rational expectations equilibrium) in a competitive double auction market? An apparent ex-ante answer is: intelligent enough to perform the bootstrap operation necessary for the task—to somehow arrive at prices that are needed to generate those very prices. Constructing a path to such equilibrium through rational behavior has remained beyond what we know of human cognitive abilities. Yet, laboratory experiments report that profit motivated human traders are able to aggregate information in some, but not all, market environments (Plott and Sunder 1988, Forsythe and Lundholm 1990). Algorithmic agents have the potential to yield insights into how simple individual behavior may perform this complex market function as an emergent phenomenon. We report on a computational experiment with markets populated by algorithmic traders who follow cognitively simple heuristics humans are known to use. These markets, too, converge to rational expectations equilibria in environments in which human markets converge, albeit slowly and noisily. The results suggest that high level of individual intelligence or rationality is not necessary for efficient outcomes to emerge at the market level; the structure of the market itself is a source of rationality observed in the outcomes.

This paper provides a micro-foundation for approximate incentive compatibility using ambiguity aversion. In particular, we propose a novel notion of approximate interim incentive compatibility, approximate local incentive compatibility, and establish an equivalence between approximate local incentive compatibility in a Bayesian environment and exact interim incentive compatibility in the presence of a small degree of ambiguity. We then apply our result to the implementation of efficient allocations. In particular, we identify three economic settings---including ones in which approximately efficient allocations are implementable, ones in which agents are informationally small, and large double auctions---in which efficient allocations are approximately locally implementable when agents are Bayesian. Applying our result to those settings, we conclude that efficient allocations are exactly implementable when agents perceive a small degree of ambiguity.

Standardized classroom economics experiments provide a treasure trove of evidence about how well results reproduce when nearly-identical methods are used. We use a sample of around 20,000 observations to test reproducibility in bargaining and trading. Ultimatum bargaining exhibits some geographical variation, and shows that equal split offers are accepted more often and more quickly than slightly unequal offers. Double auction results are highly reproducible and are close to equilibrium predictions. Our large sample shows robust, strong correlations between how individual surplus and trading order, and autocorrelation ( 0.5) of successive price changes, consistent with trading dynamics based on limited rationality.

We study markets based on the uniform-price double auction with T periods and I traders who have private information about their demands. The model accommodates the heterogeneity in the traders' risk preferences, the statistics of outcomes they condition their demands on, and general interdependence in traders' valuations. We show that the traders' price impacts are sufficient statistics for traders’ equilibrium demands, including inference coefficients. This allows a non-recursive equilibrium characterization which eliminates the information about the joint distribution of future prices that a recursive characterization would require. A trader optimizes with respect to a directional derivative. Equilibrium strategies depend on the entire history. In markets with heterogeneous traders, changes in transparency affect the traders’ price impact and the distribution of risk among the traders. For any information structure, increasing transparency in early rounds increases efficiency provided that there are sufficiently many trading rounds; introducing sufficiently many rounds improves welfare.

This article compares centralized with disconnected markets in which n > 2 strategic agents trade two perfectly divisible goods. In a multi-goods uniform-price double auction (centralized market) traders can make their demand for one good contingent on the price of the other good. Interlinking demands across goods is - by design - not possible when each good is traded in separate, single-good uniform-price double auctions (disconnected market). Here, agents are constrained in the way they can submit their joint preferences. I show for a class of models that equilibrium allocations and efficiency of centralized and disconnected markets nevertheless coincide when the total supply of the goods is known or perfectly correlated. This suggests that disconnected markets perform as well as centralized markets when the underlying uncertainty that governs the goods’ market prices is perfectly correlated.

We investigate how the network structure of financial linkages and uncertainty about the location of a shock affect the likelihood of contagion and the formation of prices in a double auction market experiment. Core-periphery networks are highly susceptible to contagion and generate fire sales of assets that exacerbate financial contagion beyond the mechanical role of network structure. In contrast, contagion is minimal on circle networks and market prices remain stable. Uncertainty on the location of the shock has little influence. The traders' comprehension level of the network-driven risk is predictive of their behavior and the likelihood of bankruptcy.

Privacy is a common concern in markets. We study dynamic markets based on the uniform-price double auction with I heterogeneous traders who have private information. Traders’ demand schedules can condition on the statistics of past and current outcomes. Equilibrium strategies depend on the entire history. Limited transparency is generally necessary for equilibrium existence. With sufficiently many trading rounds, more transparent designs increase efficiency. For any information structure, one can design a privacy-preserving auction that is ex post efficient. The model allows analysis of how transparency and, more generally, the design of conditioning statistics affects equilibrium trading behavior and efficiency.

We report boundary experiments testing the robustness of price convergence in double auction markets for non-durable goods in which there is extreme earnings inequality at the competitive equilibrium (CE). Following up on a conjecture by Smith (1980), we test whether the well-known equilibrating power of the double auction institution is robust to the presence of complete information about traders' values and costs and the presence of symmetric market power. We find that complete information is insufficient to impede convergence to CE prices; however, introducing market power consistently causes prices to deviate from the CE, whether or not subjects possess complete information. Our design highlights the value of boundary experiments in understanding how market institutions shape behavior, and our findings help delineate the limits of the double auction institution to generate competitive outcomes.

Voluntary carbon offset schemes have sprung up in the last decade offering individuals opportunities to neutralize their own carbon footprint. These schemes strongly appeal to the personal responsibility of individuals in reducing the carbon emissions they cause. In this paper we report on a controlled laboratory experiment to better understand the behavioral motivations driving the purchase of carbon offsets, i.e., payments towards the reduction of damages to the environment. We show that the opportunity to offset damages does not affect the total damages created by the individuals when individuals trade in competitive markets. At the same time, we find a stable demand for carbon offsets when the price is sufficiently low. Therefore, introduction of carbon offsets increases efficiency by eliminating some of the damages ex-post. Behavior, however, is very heterogeneous. Individuals with a high (low) personal-responsibility index increase their offset purchases as their own damage (total damages) increases, but do not condition their offsetting behavior on the total damages (own damages) created.

This paper studies the impact of increasing trading frequency in financial markets on allocative efficiency. We build and solve a dynamic model of sequential double auctions in which traders trade strategically with demand schedules. Trading needs are generated by time-varying private information about the asset value and private values for owning the asset, as well as quadratic inventory costs. We characterize a linear equilibrium with stationary strategies and its efficiency properties in closed form. Frequent trading (more double auctions per unit of time) allows more immediate asset reallocation after new information arrives, at the cost of a lower volume of beneficial trades in each double auction. Under stated conditions, the trading frequency that maximizes allocative efficiency coincides with the information arrival frequency for scheduled information releases, but can far exceed the information arrival frequency if new information arrives stochastically. A simple calibration of the model suggests that a moderate market slowdown to the level of seconds or minutes per double auction can improve allocative efficiency for assets with relatively narrow investor participation and relatively infrequent news, such as small- and micro-cap stocks.

Existing models of divisible double auctions typically require three or more traders -- when there are two traders, the usual linear equilibria imply market breakdowns unless the traders' values are negatively correlated. This paper characterizes a family of nonlinear ex post equilibria in a divisible double auction with only two traders, who have interdependent values and submit demand schedules. The equilibrium trading volume is positive but less than the first best. Closed-form solutions are obtained in special cases. Moreover, no nonlinear ex post equilibria exist if: (i) there are n>=4 symmetric traders or (ii) there are 3 symmetric traders with pure private values. Overall, our nonlinear equilibria fill the "n=2" gap in the divisible-auction literature and could be a building block for analyzing strategic bilateral trading in decentralized markets.

In an auction market, the auctioneer exerts significant influence in choosing and administering a selling strategy. We make the case for viewing the auctioneer as a market maker, whose success depends on how well he manages externalities without jeopardizing the trust of the buyers and sellers. We illustrate that incentives of the market maker may not be aligned with that of individual sellers. Using a unique data set, from tea auctions in Chittagong, Bangladesh, we argue that an auctioneer's actions maintain a careful balance of his own incentives vs. those of his clients, and the auction outcomes are affected by how much discretion the auctioneer has in choosing the selling strategy. Specifically, we find that raising the reserve price for a lot exerts a positive price externality on subsequent lots within the auction. To manage the momentum of market prices, the auctioneer chooses higher reserve prices for tea produced by tea estates in which he has an ownership stake. While these teas receive a higher price when sold, they sell less frequently creating a short run cost to the auctioneer but an overall positive impact on market prices. Thus, consistent with the role of a market maker, a desire to appear non-opportunistic, rather than opportunism, seems to better explain the auctioneer's actions.

A single unit of a good is to be sold by auction to one of two buyers. The good has either a high value or a low value, with known prior probabilities. The designer of the auction knows the prior over values but is uncertain about the correct model of the buyers’ beliefs. The designer evaluates a given auction design by the lowest expected revenue that would be generated across all models of buyers’ information that are consistent with the common prior and across all Bayesian equilibria. An optimal auction for such a seller is constructed, as is a worst-case model of buyers’ information. The theory generates upper bounds on the seller’s optimal payoff for general many-player and common-value models.

This paper studies a dual that applies to a class of quasilinear exchange economies with indivisible (or divisible) goods in search for an equilibrium. Our model aims at an economy with a large scale and an agent's revealed demand or supply is contaminated with stochastic errors (noises). Thus, it is computationally expensive, if not inapplicable, to run a Walrasian auction based on truthfully revealed total demand and supply. Instead, we study a probabilistic α-double auction and are interested in the convergence of a price process it generates. We allow the weight α to be a random variable with time-varying unknown distributions over [0,1] and show a convergence result when the two step sizes are either deterministically diminishing or probabilistically diminishing in the mean. An error bound is estimated when the two step sizes are constant, bounded away from zero, while α is at random. We provide conditions under which the double auction generates a price process that converges in probability to the set of Walrasian equilibria of the primal economy.

This paper establishes existence and uniqueness of Quantal Response Equilibrium (QRE) in a double auction. The concept of QRE has the intuitive property that a deviation from best response is less likely the higher the cost associated with the deviation itself. Thanks to such property, the QRE accommodates stochastic elements in the analysis of the strategic decision-making that arises in the double auction. By providing a theoretical alternative to the Bayesian Nash Equilibrium model, the QRE model offers an appealing tool for analyzing data of double auction experiments.

We study experimentally the effect of bargaining power in two sequential mechanisms that offer the possibility to trade at a fixed price before an auction. In the “Buy-It-Now” format, the seller has the bargaining power and offers a price prior to the auction; whereas in the “Sell-It-Now” format, it is the buyer. Both formats are extensively used in online and offline markets. Despite very different strategic implications for buyers and sellers, results from our experiment suggest no effects of bargaining power on aggregate outcomes. There is, however, substantial heterogeneity within sellers. Sellers who ask for high prices not only benefit from having the bargaining power but also earn revenue above those expected in the auction.

We investigate how price ceilings and floors affect outcomes in continuous time, double auction markets with discrete goods and multiple qualities. When price controls exist, the existence of competitive equilibria (the solution concept of classical market theory) is no longer guaranteed; hence, we investigate the nature of non-price competition and how markets might evolve in its presence. We develop a quality competition model based on matching theory. Equilibria of the quality competition model always exist in such price-constrained markets; moreover, they naturally correspond to competitive equilibria when competitive equilibria exist. Additionally, we characterize the set of equilibria of the quality competition model in the presence of price restrictions. In a series of experiments, we find that market outcomes closely conform to the predictions of the model. In particular, price controls induce non-price competition between agents both in theory and in the experimental environment; market behaviors result in allocations close to the predictions of the model.

This paper studies the coexistence of two competing mechanisms in the same market, where one follows the posted-offer rule and the other one incorporates a double-auction mechanism. We explore this coexistence within a sports betting example in which bettors are free to choose between a bookie (posted-offer market) and a betting exchange. Our findings imply that i) bettors' risk aversion parameter is instrumental in whether these two mechanisms coexist or not, ii) most bettors are strictly better off, and none is worse off, when they have access to both of these competing mechanisms rather than just one, and iii) these results hold even when we allow the bookie to make a positive profit instead of following a zero expected profit pricing rule.

The analysis of equilibrium in double auctions is complicated, with almost no closed-form examples and with formal results limited to bounds on bidding behavior. We consider the buyer's bid double auction in a correlated, private values model. Sellers have the incentive to submit their costs as their asks while buyers strategically underbid. We determine (i) the asymptotic distribution of the equilibrium market price and (ii) the asymptotic limits of terms in the first order condition for a buyer's selection of his bid. Part (i) reveals the properties of the strategically determined market price as an estimator of the rational expectations equilibrium price. Part (ii) provides a simple formula for a buyer's bid that is shown numerically to closely approximate his equilibrium bid even in relatively small markets. This formula reveals how equilibrium varies with the numbers of buyers and sellers and the distribution of their values/costs.

This paper studies the double auction (DA) mechanism in Ma and Li (2011) for a class of exchange economies. We extend their results to more general cases where sellers and buyers each form a complex time non-homogeneous Markovian chain, as specified in Ram et al. (2009), in the communication of their private information. A computational experiment is provided. The results of the experiment show that the DA mechanism can quickly integrate new information to an equilibrium. But formation of bubbles and crashes is also observed in the experiment, consistent with our theorems, which together with the experimental results provide new evidence that a DA mechanism, widely utilized in real exchange markets, may contribute to the excess volatility identified in Shiller (1981) and LeRoy and Porter (1981).

We investigate equilibrium in the buyer's bid double auction (BBDA) in a model with correlated signals and either private or interdependent values. Using a combination of theorems and numerical experiments, we demonstrate that simple equilibria exist even in small markets. Moreover, we bound traders' strategic behavior as a function of market size and derive rates of convergence to zero of (i) inefficiency in the allocation caused by strategic behavior and (ii) the error in the market price as an estimate of the rational expectations price. These rates together with numerical experiments suggest that strategic behavior is inconsequential even in small markets in its effect on allocational efficiency and information aggregation. The BBDA thus simultaneously accomplishes both the informational and allocational goals that markets ideally fulfill; it does this perfectly in large markets and approximately in small markets, with the error attributable mainly to the smallness itself and not the strategic behavior of traders.

This paper studies information aggregation of double auction mechanisms in general interdependent values economic environments. The main findings of the paper is as follows. First, when buyers and sellers have unit demand and supply and one dimensional conditional iid signals, since inferences are symmetric, even with a small private value effect, every perfect equilibrium of double auction games converges to an ex post monotone equilibrium in the limit economy. Second, in more general economic environments with heterogeneous distributions of signals, multidimensional signals, and multiple units of demand and supply, information aggregation requires stronger private value effects that outweighs the heterogeneous inference effects. Without these conditions, equilibrium prices in the limit economy can only be partially revealing. Third, when these conditions hold, equilibrium prices of double auction markets provide consistent and asymptotic estimators of the fundamental value of the assets.

We propose a dynamic auction mechanism for efficiently allocating multiple heterogeneous indivisible goods. These goods can be split into two distinct sets so that items in each set are substitutes but complementary to items in the other set. The seller has a reserve value for each bundle of goods and is assumed to report her values truthfully. In each round of the auction, the auctioneer announces the current prices for all items, bidders respond by reporting their demands at these prices, and then the auctioneer adjusts simultaneously the prices of items in one set upwards but those of items in the other downwards. We prove that although bidders are not assumed to be price-takers and thus can strategically exercise their market power, this dynamic auction always induces the bidders to bid truthfully as price-takers, yields an efficient outcome and also has the merit of being a detail-free, transparent and privacy preserving mechanism.

This paper studies the impact of heterogeneity in interdependence of trader values on price inference and welfare. A model of double auction with quasilinear-quadratic utilities is introduced that allows for arbitrary Gaussian information structures. With heterogeneous interdependence, some traders learn more from prices whereas others from private signals; thus, heterogeneity separates informed and uninformed trading. Changes in market structure can improve both informativeness of prices and private signals of a trader and make some traders learn more from prices than others. We characterize conditions on the information structure for price and signal inference to involve no tradeoff.

An impatient, risk-neutral monopolist must sell one unit of an indivisible good within a fixed number of periods and privately informed myopic buyers with independent values enter the market over time. In each period, the seller can either run a reserve price auction incurring a cost or post a price without the cost. We characterize the optimal sequence of mechanisms that maximizes the seller's expected profits. When there is an infinite number of periods, repeatedly running auctions with the same reserve price or posting a constant price is optimal. When there is a finite number of periods, the optimal sequence is a sequence of declining prices, a sequence of auctions with declining reserve prices converging to the static optimal monopoly reserve price, or the combination of the two. Most interestingly, a sequence of auctions before a sequence of posted prices is never optimal. The mechanism sequence of posted prices followed by auctions remains optimal under various extensions of the basic setting and resembles a Buy-It-Now option.

In this paper we study a pure exchange economy with commodity money. Each agent submits on each market the quantity that he wants to buy in that market, the quantity that he wants to sell on the market, a bid price and an asking price for the commodity traded on the market. A rationing rule decides the allocation of goods in the economy. In this framework we show that under four very plausible conditions on rationing rules every competitive equilibrium is a Nash equilibrium of our proposed class of market games. Conversely, if each market has at least two active buyers and two active sellers at a Nash equilibrium, then subject to the satisfaction of three more plausible conditions, the resulting allocation is a competitive equilibrium that is supported by the prices that active players choose on each market.

This study quantifies the eciency of a real-world bargaining game with two-sided incomplete information. Myerson and Satterthwaite (1983) and Williams (1987) derived the theoretical ecient frontier for bilateral trade under two-sided uncertainty, but little is known about how well real-world bargaining performs relative to the frontier. The setting is wholesale used-auto auctions, an 80 billion industry where buyers and sellers participate in alternating-oer bargaining when the auction price fails to reach a secret reserve price. Using 300,000 auction/bargaining sequences, this study nonparametrically estimates bounds on the distributions of buyer and seller valuations and then estimates where bargaining outcomes lie relative to the ecient frontier. Findings indicate that the observed auction-followed-by-bargaining mechanism is quite ecient, achieving 88‐96% of the surplus and 92‐99% of the trade volume which can be achieved on the ecient frontier.

We develop a competitive equilibrium theory of a market for votes. Before voting on a binary issue, individuals may buy and sell their votes with each other. We define the concept of Ex Ante Vote-Trading Equilibrium, identify weak sufficient conditions for existence, and construct one such equilibrium. We show that this equilibrium must always result in dictatorship and the market generates welfare losses, relative to simple majority voting, if the committee is large enough. We test the theoretical implications by implementing a competitive vote market in the laboratory using a continuous open-book multi-unit double auction.

This paper studies a first price package auction in which multiple sellers participate in addition to multiple buyers. We generalize the notion of the profit-target strategy which is first introduced as a truthful strategy in a first price package auction with a single seller by Bernheim and Whinston (1986b). We then show that the set of equilibrium payoffs in profit-target strategies is equal to the bidder-optimal core, and is also equal to the set of coalition-proof Nash equilibria. Using this result, we find that any equilibrium payoff vector is weakly Pareto-dominated by the VCG payoff vector for buyers, and that the Walrasian competitive equilibrium payoff vector is weakly Pareto-dominated by some equilibrium payoff vector for buyers, even if goods are substitutes. This contrasts with the first price package auction with a single seller, in which it is shown that if goods are substitutes, then those three outcomes are payoff-equivalent.

The computational tractability of markets with computerized agents requires they be subject to artificial boundary conditions that are not found in analogous markets with humans in both natural and laboratory settings. In particular, double-auction markets with zero-intelligence agents, which are the baseline for most agent-based modeling, constrain all bids, offers, and trade prices to lie between a fixed upper and lower bound. This paper shows how changes in these boundary conditions influence pricing and allocative efficiency under the most favorable conditions for such agents. Increasing the upper bound on prices (while maintaining the natural lower bound of zero) will initially increase mean trade prices and decrease allocative efficiency. Further increases in the upper bound will not increase mean prices beyond the competitive equilibrium; however, they will continue to reduce allocative efficiency by hindering the price discovery process. Moreover, contrary to standard microeconomic theory and well-established results in human experimental markets, boundary conditions consistent with competitive pricing are found to be inconsistent with the most efficient operation of those markets.

We study the design of settlement auctions for credit default swaps (CDS). We nd that the one-sided design of CDS auctions currently used in practice gives CDS buyers and sellers strong incentives to distort the nal auction price, in order to prot from existing CDS positions. In the absence of frictions, the current auction mechanism tends to overprice defaulted bonds given an excess supply and underprice them given an excess demand. We propose a double auction to mitigate price biases and provide robust price discovery. The predictions of our theory on bidding behavior are consistent with CDS auction data.

We study a sealed-bid auction between two bidders with asymmetric independent private values. The two bidders own asymmetric shares in a partnership. The higher bidder buys the lower bidder's shares at a per-unit price that is a convex combination of the two bids. The weight of the lower bid is denoted by k in [0,1]. We partially characterize equilibrium strategies and show that they are closely related to equilibrium strategies of two well-studied mechanisms: the double auction between a buyer and a seller and the first-price auction between two buyers (or two sellers). Combining results from those two branches of the literature enables us to prove equilibrium existence. Moreover, we find that there is a continuum of equilibria if k in (0,1) whereas the equilibrium is unique if k in {0,1}. Our approach also suggests a procedure for numerical simulations.

A limit order market enables an early seller to trade with a late buyer by leaving a price quote. Information arrival in the interim period creates adverse selection risk for the seller and therefore hampers trade. Entry of high-frequency traders (HFTs) might restore trade as their machines can refresh quotes quickly on (hard) information. Empirically, HFT entry reduced adverse selection by 23% and increased trade by 17%. Model calibration shows that one percentage point more of the gains from trade were realized. Finally, we show that a well-designed double auction raises this to ten percentage points.

This paper presents a market with asymmetric information where a privately revealing equilibrium obtains in a competitive framework and where incentives to acquire information are preserved. The equilibrium is efficient, and the paradoxes associated with fully revealing rational expectations equilibria are precluded without resorting to noise traders. The rate at which equilibria in finite replica markets with n traders approach the equilibrium in the continuum economy is 1/[square root]n, slower than the rate of convergence to price-taking behavior (1/n); and the per capita welfare loss is dissipated at the rate 1/n, slower than the rate at which inefficiency due to market power vanishes (1/n[squared]). The model admits a reinterpretation in which behavioral traders coexist with rational traders, and it allows us to characterize the amount of induced mispricing.

We study the interaction between competitive markets that produce large but unequally distributed welfare gains and elections through which the poor majority can redistribute income away from the rich minority. In our simple laboratory democracy, subjects first earn their income by trading in a double auction market and thereafter vote on redistributive policies in two-candidate elections. In addition, in one of the treatments subjects can attempt to influence the candidates’ policy choices by transferring money to them. We observe very high levels of redistribution – even when transfers to candidates are possible – with little effect on market efficiency. Overall, the experimental results are explained by our equilibrium predictions.

We study the coordination of production decisions for multiple products among many manufacturers and many suppliers, each with private information about its own objective and its own production capabilities. Our methodology does not require a probabilistic model of the beliefs of each decision maker (manufacturer or supplier) regarding asymmetric information. We discover a game form (decentralized mechanism) which implements in Nash Equilibria the solution of the corresponding centralized supply chain problem. In our mechanism the buyers and sellers in the supply chain submit bids/o ers for the quantities they would like to buy/sell and the prices they would like other buyers/sellers to pay/receive per unit of a product. We provide rules for allocating orders and determining payments based on the bids/o ers. We show that, in equilibrium, the unit price for all rms supplying a particular product is equal to the unit price for all rms buying the product. One could interpret this price as the clearing price for the product in the internal market among rms or among divisions of the same rm. Unlike prior work on coordinating rm's production and sales decisions through internal markets, our model does not require a centralized planner to set the prices in the internal markets. Instead, the clearing prices are part of the equilibrium outcome.

Standard setters and regulators generally assume that making accounting easier to process leads to more efficient markets, thereby benefiting traders. I test that assumption in markets that are prone to price bubbles. Market efficiency will obtain only if reducing complexity in accounting leads traders to acquire and process that information more readily, and then to trade in a manner that moves price closer to fundamental value. This reasoning implies two causal links - (1) the link from less complex information to greater processing of that information, and (2) the link from greater processing of information to more efficient markets. I test these causal links in an experiment by examining the effect of reducing accounting complexity on market efficiency, as mediated by processing accounting information and as moderated by the market’s propensity to form price bubbles. Traders buy and sell shares of an asset in a multiperiod, double auction market in which speculative capital gains are possible. I find that reducing complexity in information increases processing. However, this processing increases market efficiency only in markets that are less prone to a price bubble. Conversely, in markets more prone to bubble some traders that process information trade according to value, while other informed traders utilize profitable non-value trading strategies that exacerbate price inefficiencies.

We define and examine three minimal market games (sell-all, buy-sell, and double auction) in the laboratory relative to the predictions of theory. These closed exchange economies have some cash to facilitate transactions, and include feedback. The experiment reveals that (1) the competitive general equilibrium (CGE) and non-cooperative (NCE) models are reasonable anchors to locate most but not all the observed outcomes of the three market mechanisms; (2) outcomes tend to get closer to CGE predictions as the number of players increases; (3) prices and allocations in double auctions deviate persistently from CGE predictions; (4) the outcome paths across the three market mechanisms differ significantly and persistently; (5) importance of market structures for outcomes is reinforced by algorithmic trader simulations; and (6) none of the three markets dominates the others across six measures of performance. Inclusion of some mechanism differences into theory may enhance our understanding of important aspects of markets.

A finite number of sellers (n) compete in schedules to supply an elastic demand. The cost of each seller is random, with common and private value components, and the seller receives a private signal about it. A Bayesian supply function equilibrium is characterized: The equilibrium is privately revealing and the incentives to rely on private signals are preserved. Supply functions are steeper with higher correlation among the cost parameters. For high (positive) correlation, supply functions are downward sloping, price is above the Cournot level, and as we approach the common value case, price tends to the collusive level. As correlation becomes maximally negative, we approach the competitive outcome. With positive correlation, private information coupled with strategic behavior induces additional distortionary market power above full information levels. Efficiency can be restored with appropriate subsidy schemes or with a precise enough public signal about the common value component. As the market grows large with the number of sellers, the equilibrium becomes price-taking, bid shading is on the order of 1/n, and the order of magnitude of welfare losses is 1/n2. The results extend to inelastic demand, demand uncertainty, and demand schedule competition. A range of applications in product and financial markets is presented.

This paper presents and analyzes the results of a recent field experiment in which residential electricity customers in Washington State with price-responsive in-home devices could use those devices to change their electricity consumption autonomously. Doing so also required an important institutional change: the regulatory institutions had to change to allow dynamic pricing. Customers could choose a retail pricing contract from a portfolio of contracts, instead of the fixed, regulated retail rate. Here we focus on the results of the real-time contract, under which homeowners participate in a double auction with a market clearing occurring every five minutes. These customers saved money, and their peak demand (and pressure on infrastructure at peak capacity) fell by 15 percent. Moreover, this combination of technology and institutional design enabled decentralized coordination, and we use complexity science to interpret results that show that the real-time market outcomes were those of a self-organizing and scalable complex adaptive system. We also draw policy implications from these results.

When capacity-constrained bidders have information about a good sold in a future auction, they need to take the information into account in forming today's bids. The capacity constraint makes even otherwise unrelated objects substitutes and creates an equilibrium link between future competition and current bidding strategy. This paper proves the existence and uniqueness of a symmetric pure-strategy equilibrium under mild conditions on the population distribution of valuations, characterizes general properties of the equilibrium bidding strategy, and provides a simple technique for numerically approximating the bidding strategy for arbitrary valuation distributions. The key property of the equilibrium is that almost all bidders submit positive bids in the first stage, thereby ensuring trade with probability one. Even bidders who strongly prefer the second object submit a positive bid in the first auction, because losing the first auction is informative about the remaining competitors who also lost, and losing with a low bid indicates that these competitors are quite strong. Because of the guaranteed trade, the sequential auction with information about future goods is a very efficient trading mechanism, achieving more than 98 percent of the potential gains from trade across a wide variety of settings.

We describe an auction mechanism in the class of Groves mechanisms that has received attention in the computer science literature because of its theoretical property of being more "learnable" than the standard second price auction mechanism. We bring this mechanism, which we refer to as the "clamped second price auction mechanism," into the laboratory to determine whether it helps human subjects learn to play their optimal strategy faster than the standard second price auction mechanism. Contrary to earlier results within computer science using simulated reinforcement learning agents, we find that both in settings where subjects are given complete information about auction payoff rules and in settings where they are given no information about auction payoff rules, subjects converge on playing their optimal strategy significantly faster in sequential auctions conducted with a standard second price auction mechanism than with a clamped second price auction mechanism. We conclude that while it is important for mechanism designers to think more about creating learnable mechanisms, the clamped second price auction mechanism in fact produces slower learning in human subjects than the standard second price auction mechanism. Our results also serve to highlight differences in behavior between simulated agents and human bidders that mechanism designers should take into account before placing too much faith in simulations to test the performance of mechanisms intended for human use.

Each of n ≥ 1 identical buyers (and m ≥ 1 identical sellers) wants to buy (sell) a single unit of an indivisible good. The core predicts a unique and extreme outcome: the entire surplus is split evenly among the buyers when m > n and among the sellers when m < n; the long side gets nothing. We test this core conjecture in the lab with n + m = 3 or 5 randomly rematched traders and minimal imbalances (m = n ± 1) in three market institutions. In the standard continuous double auction, the surplus indeed goes overwhelmingly towards the short side. The DA-Chat institution allows traders to have cheap talk prior to the double auction, while the DA-Barg institution allows the long siders to negotiate enforceable profit sharing agreements while trading. Despite frequent attempts to collude and occasional large deviations from the core prediction, we find that successful collusion is infrequent in both new institutions. A disproportionate fraction of the successful collusions are accompanied by appeals to fairness.

A number of heterogeneous items are to be sold to several bidders. Each bidder demands at most one item. The price of each item is not completely flexible and is restricted to some admissible interval. In such a market economy with price rigidities, a Walrasian equilibrium usually fails to exist. To facilitate the allocation of items to the bidders, we propose an ascending auction with rationing that yields a constrained Walrasian equilibrium outcome. The auctioneer starts with the lower bound price vector that specifies the lowest admissible price for each item, and each bidder responds with a set of items demanded at those prices. The auctioneer adjusts prices upwards for a minimal set of over-demanded items and chooses randomly a winning bidder for any item if the item is demanded by several bidders and its price has reached its highest admissible price. We prove that the auction finds a constrained Walrasian equilibrium outcome in a finite number of steps.

We prove the existence of monotonic pure strategy equilibrium for many kinds of asymmetric auctions with n bidders and unitary demands, interdependent values and independent types. The assumptions require monotonicity only in the own bidder’s type. The payments can be a function of all bids. Thus, we provide a new equilibrium existence result for asymmetrical double auctions and a small number of bidders. The generality of our setting requires the use of special tie-breaking rules. We present an example of a double auction with interdependent values where all equilibria are trivial, that is, they have zero probability of trade. This is related to Akerlof’s “market for lemmons” example and to the “winner’s curse,” establishing a connection between them. However, we are able to provide sufficient conditions for non-trivial equilibrium existence.