Search

I develop a sequential trading model with ambiguity-averse market makers and provide a theoretical explanation to the historical coincidence of ambiguous events, asset illiquidity, and price variability. My model implies that the bid-ask spread of an asset contains an additive component of ambiguity premium. As a result, higher ambiguity generally leads to lower asset liquidity. More interestingly, asset prices are variable under particular conditions: specifically, only mixed-strategy equilibria exist, such that market makers probabilistically set multiple prices. Further analysis confirms that, compared with risk, ambiguity plays a unique role in explaining price variability.

We examine the microstructure of liquidity provision in the COVID-19 corporate bond liquidity crisis. During the two weeks leading up to Federal Reserve System interventions, volume shifted to liquid securities, transaction costs soared, trade-size pricing inverted, and dealers, particularly non-primary dealers, shifted from buying to selling, causing dealers’ inventories to plummet. Liquidity provisions in electronic customer-to-customer trading increased, though at prohibitively high costs. By improving dealer funding conditions and providing a liquidity backstop, the Primary Dealer Credit Facility and the Secondary Market Corporate Credit Facility (SMCCF) stabilized trading conditions. Most of the impact of SMCCF on bond liquidity seems to have materialized following its announcement. We argue that the Federal Reserve's actions reflect a new role as market maker of last resort.

Hedging short gamma exposure requires trading in the direction of price movements, thereby creating price momentum. Using intraday returns on over 60 futures on equities, bonds, commodities, and currencies between 1974 and 2020, we find strong market intraday momentum everywhere. The return during the last 30 minutes before the market close is positively predicted by the return during the rest of the day (from previous market close to the last 30 minutes). The predictive power is economically and statistically highly significant, and reverts over the next days. We provide novel evidence that links market intraday momentum to the gamma hedging demand from market participants such as market makers of options and leveraged ETFs.

Micro-mobility market-making represents an under-studied but important aspect of urban transport sustainability transitions. Micro-mobility roll-out combines several critical elements: decarbonisation, digitalisation and public space interventions. We theorise the emergence of a micro-mobility market, drawing on innovation studies, micro-politics scholarship and commoning mobility literature, and critically discuss the relationship between innovation and regulation in this emerging market. We examine e-scooter roll-out in Bergen, Norway, using a structured analysis that interrogates knowledge, authority and power. Our empirical analysis employs expert interviews and focus group discussions with key transport sector stakeholders, including policymakers, practitioners and urban residents. We discuss insights in terms of implications for how micro-mobility markets are made and in turn make urban transport sustainability transitions. We argue that while micro-mobility can enable just low-carbon mobility transitions, market forces drive outcomes that may undermine urban sustainability agendas. We identify scope for dynamic regulation to engender low-carbon mobility in wider public interest. publishedVersion

This paper derives a liquidity cost process in a non-cooperative cost competition game among market makers and discusses its implication for the structure of a dealer market. The main result shows that there does not exist an equilibrium supporting both multiple market makers earning strictly positive profits and a well-behaved liquidity cost process (i.e. strictly increasing and convex) as documented in the price impact literature. Bertrand price competition arises as an equilibrium phenomenon, which naturally leads to the dominance of a cost-efficient market maker in the dealer market.

Motivated by the theory of demand-based option pricing in imperfect markets, we examine the relation between short-sale constraints and equity option returns, conditional on the level of mispricing in the underlying stock. We report a monotonic relation between various measures of short-sale constraints and delta-hedged returns of put options on overpriced stocks. This relation is robust to controls for firm attributes and limits to arbitrage proxies. Our findings suggest that while investors drive up the demand for these put options, dealers command a high premium as compensation for the increased market making risk. We do not find a robust relation for either put options on underpriced stocks or call options.

We investigate the impact of an exogenous trading glitch at a high-frequency market-making firm on standard measures of stock liquidity (spreads, price impact, turnover, and depth) and institutional trading costs (implementation shortfall and volume-weighted average price slippage). Stocks in which the firm accumulates large long (short) positions increase (decrease) by about 4% during the glitch and become substantially more illiquid. It takes one day for prices and spread-based liquidity measures to revert. Institutional trading costs, however, remain significantly higher for more than one week. Both liquidity measures are also weakly correlated outside the glitch period, suggesting they capture different aspects of liquidity.

This paper finds that window dressing, bypass opportunities, market sentiment and index adjustments drive seasonal bond market liquidity effects. It uses a unique data set that combines a number of recurring event types. After controlling for other seasonal factors, liquidity is lower prior to quarter ends, which is attributed to market makers’ window dressing. Seasonal liquidity effects are smaller in case investors can bypass secondary market venues, and market sentiment influences both the direction and the size of most seasonal effects. Additions to an index spur the liquidity of the bond included, and index adjustments are also important for the liquidity of other bonds, that have been issued by the same institution. These results give important advice for the trading of market participants, they explain why the primary market is essential for secondary venues and why too much regulation can be detrimental to market capacity.

We examine market making behavior of dealers for 55,988 corporate bonds, many of which trade infrequently. Dealers have a substantially higher propensity to offset trades within the same day rather than committing capital for longer periods for riskier and less actively traded bonds. Dealers’ holding periods do not decline with a bond's prior trading activity and in fact are lowest for some of the least active bonds. As a result, cross-sectional estimates of roundtrip trading costs do not increase as prior trading activity declines. Our results suggest that dealers endogenously adjust their behavior to mitigate inventory risk from trading in illiquid and higher risk securities, balancing search and inventory costs in equilibrium such that observed spreads can appear invariant to expected liquidity.

As the role of designated market makers in liquidity supply has diminished over time, algorithmic traders (ATs) have stepped up to fill that gap in modern equity markets. Prompted by concerns of fragility introduced by such voluntary market making, we examine ATs’ liquidity-provision role during the COVID-19 crisis. Amidst the turmoil as overall market liquidity declined, we find that ATs did not (disproportionately) withdraw liquidity supply. Stocks with the highest algorithmic trading (AT) activity experienced lower liquidity reduction compared to stocks with low AT activity, although profits from market making in high AT stocks fell more. High AT stocks did not experience greater reduction in either competition for liquidity provision or price improvements than low AT stocks. Multiple tests indicate that high AT activity did not associate with any greater deterioration in price efficiency vis-a-vis low AT stocks. Stocks in industries hardest hit by COVID-19 did not see any less AT competition for liquidity supply or price efficiency than stocks in the least affected ones. Overall, our results allay some recent concerns that the current levels of AT have made markets more susceptible to liquidity withdrawal in times of crises.

This paper analyzes how informed traders balance leverage and liquidity in making cross-market trading decisions. Our findings are three-fold. Using daily data on the actively traded North American CDX Investment Grade and High Yield indexes from 2010 to 2017, first, we find robust evidence that, when the credit default swap (CDS) market is illiquid, informed traders do not hesitate to exploit their information advantage in the option market, and this manifests mainly for negative credit news. Second, we find that informed trading occurs predominantly in the option market for low-rated firms due to high adverse selection risk in the CDS market. Third, there is strong evidence of informed trading taking place in the CDS market for financial firms and we argue that this is due to the interdealer network through which CDS dealers obtain private information about other financial firms and their privileged position as market makers to be able to disguise their informed trades as ordinary market-making activities. Finally, we propose re-thinking market efficiency and how to advance it in a direction which does not privilege a small circle of financiers or create monopoly while keeping the markets liquid and tranquil at all times.

I examine the impact of the V-shaped disposition effect, which results in uninformed sales when investors realize large gains and losses. Adverse selection risk is reduced in the presence of more uninformed sales. I show in a model that market makers should post tighter bid-ask spreads and quote prices that are less sensitive to sales. Using stocks with merger and acquisition announcements, which increase the magnitude of investors' unrealized gains or losses and trigger disposition sales in the post-announcement period, I find evidence supporting these predictions.

This paper presents an application of a linear quadratic stochastic differential game to a financial model that describes trading behaviors of different types of players in a high frequency stock market. Stability of the stock market in a high frequency environment is a central issue in financial markets. Building a model that expresses the trading behaviors of the different types of players and the price actions in a high frequency market helps introduce better regulations for a healthy market. Firstly, we represent trading behaviors of the three types of players, algorithmic traders, general traders, and market makers as well as the mid-price process of a risky asset by a linear quadratic stochastic differential game. Secondly, we obtain a Nash equilibrium for open loop admissible strategies by solving a forward-backward stochastic differential equation (FBSDE) derived from the stochastic maximum principle. Finally, we present numerical examples of the Nash equilibrium for open loop admissible strategies and the corresponding price action of the risky asset, which agree with the empirical findings on the mechanism of high frequency markets. This model can be used to investigate the impact of regulation changes on the market stability as well as trading strategies of the players.

In the classical Cournot model, each firm tries to maximize its own payoff by deciding an optimal strategy for determining the quantity of goods produced during each time period–i.e. turn, of the game. In the typical application, all firms compete in the same market. In more recent economic models, firms compete across a number of markets simultaneously. In this situation, a Networked Cournot Competition (NCC) graph models the relationship between firms and markets. This paper describes a model of competition among demand response aggregators ( DRA s) as firms to sell energy (as a homogeneous good) stored in aggregated residential batteries in a networked environment where market constraints are effected and trades are generally facilitated through the actions of a market maker who’s turn is sequentially distinct from the other players. We call this game Networked Stackelberg Competition (NSC). The impact of strategic anticipative behavior in networked markets is of paramount importance in distinguishing NSC from other competition models. For each firm, the optimal bidding strategy and Nash equilibrium are obtained through analyses of in an incomplete-information game. DRAs submit quantity bids and the market maker (system operator) controls the transaction power and transaction price over the network subject to transmission constraints and other market policies. Criteria required for existence of uniqueness of a Nash equilibrium are presented, and effectiveness of the game is also studied in the paper demonstrating demand response scheduling improves market situations. The details are presented in the application of a NSC model to real world case study data taken from the island of Maui, Hawaii.

This paper differentiates order imbalances based on trader categories. The daily order imbalances are highly persistent, especially for the number-measured imbalances. That the price pressure caused by imbalances cannot last beyond a trading day indicates that China's stock market is efficient enough to absorb the imbalances. We find that large individuals, small individuals and small institutions act frequently as market makers by submitting non-marketable limit orders, and the market making activities are profitable for small individuals and institutions. The evidence indicates that individuals are noise or liquidity traders, while institutions are more likely to be informed traders.

We present a model where an insider and multiple informed outsiders share some common information about a risky asset to analyze the insider’s incentive for voluntarily leaking information to the market maker. We find that the ex-ante uncertainty of the traded asset and the number of informed outsiders play a decisive role in the insider’s choice of information leakage. When trading an asset with low risk against very few informed outsiders, the insider never chooses to leak any information. However, if sufficient outsiders coexist in the market, strategically releasing a signal to the market maker helps the insider anticipate the price and consequently earn more profits. In this case, if the insider’s monopoly information advantage is large, the insider would choose to fully leak the shared information to the market maker. Our work shows that information leakage to the market maker is more efficient than the public disclosure of the shared private information as a way for the insider to profit from insider trading. The results also provide the theoretical explanation for the insider’s leaking information to the underwriter in seasoned equity offerings (SEOs).

This article develops an insider-trading model with the presence of rational and representativeness heuristic insiders in the market with and without market regulation. We find that, regardless of whether the market regulation exist or not, the heuristic insider trades more aggressively on his information, than the rational insider does, with the result that there is a higher probability for him to earn greater profits. Further, a higher heuristic bias of the heuristic insider leads to lower expected profit for rational insider, but higher expected profits for the heuristic insider. Moreover, the heuristic insider and rational insider prefer to trade less aggressively on their information when there is market regulation, but, they are likely to get more ex-ante expected profit. Furthermore, we find that the price efficiency and market stability may all be sacrificed if insider trading is regulated.

We apply Reinforcement Learning algorithms to solve the classic quantitative finance Market Making problem, in which an agent provides liquidity to the market by placing buy and sell orders while maximizing a utility function. The optimal agent has to find a delicate balance between the price risk of her inventory and the profits obtained by capturing the bid-ask spread. We design an environment with a reward function that determines an order relation between policies equivalent to the original utility function. When comparing our agents with the optimal solution and a benchmark symmetric agent, we find that the Deep Q-Learning algorithm manages to recover the optimal agent.

This paper analyzes differences between the regular and extended trading sessions in the high-frequency reaction of equity markets to potential news. Using presidential tweets as market-stirring events, I find that generally volatility increases and liquidity deteriorates within less than a second after a tweet. Compared to the regular trading session, the reduction in market quality is stronger and faster during the extended trading hours, when liquidity is lower and designated market maker participation is optional.

Blockchain-based decentralized exchanges (DEXes) are a pre-requisite for the nascent decentralized finance ecosystem. The most successful form of DEX in terms of trading volume are swap exchanges, smart contracts that pool liquidity and that price transactions with a deterministic function. Almost all swap exchanges use a pricing rule that is conceptually problematic as it gives rise to intrinsically profitable front-running opportunities as well as cross-DEX arbitrage. In practice, these shortcomings cause significant network congestion, a negative externality that raises costs for all users and that threatens the long-term viability of the blockchain ecosystem. Calibrated to a low liquidity but frequently used trading pair on UniSwap, the largest DEX for much of 2020/21, about 14% of transactions see an implicit excess cost of at least 50bps, which is orders of magnitude larger than the common trading costs for this pair on centralized exchanges.I further present an alternative pricing rule, based on insights from the market microstructure literature, that does not suffer from these flaws.

Using rule 605 reports from major market makers that receive retail order flow, we examine the impact of zero-commission trades on order execution quality. We find that the effective spread for marketable orders between 100 and 499 shares marginally decreased following the zero-commission cut. We also find that realized spreads increased and adverse selection decreased for market makers relative to exchanges. Consistent with better alignment of incentives between brokers and market makers, our results suggest that the elimination of commissions for retail investors improved execution quality for orders directed to third-party market makers.

We study two aspects of the transition to electronic markets for large uninformed traders: (1) the increase in trading frequency, and (2) the move to anonymous trading, which challenged the implementation of reputation-based strategies. We present a model of optimal execution where a risk-neutral trader must sell over multiple trading periods before a fixed deadline to several risk-averse market makers. We model trading frequency as the number of trading periods over a fixed time intervial, and reputationbased strategies as the ability to commit ex-ante to a sequence of market orders, what we refer to as scheduling. While scheduling and more trading periods are both valuable, a trader typically prefers scheduling over more trading periods. Moreover, the benefits of scheduling significantly increase when coupled with more frequent trading, and outweigh the potential costs associated with predatory trading against predictable orders. Our model suggests that transparent execution algorithms, which could be used to implement scheduling in anonymous electronic markets, may have important benefits for large, uninformed traders.

The delivery of information is as important as the content itself. We present evidence that an announcement's writing literacy is as critical as its information content. We show that the text's readability in firm-specific news affects trading decision, particularly those of algorithmic traders. We find that algorithmic participants trade more aggressively when the news items are more readable. This change in their trade decisions leads to several market impacts. First, we suggest that readability affects algorithmic traders' liquidity provision. Algorithmic traders can capture more rents when the news is more readable while other participants enjoy lower adverse selection costs. The market makers' profit is on average less than the reduced selection costs. Therefore, there is an improvement in overall market welfare in terms of the transaction cost. Finally, price efficiency is improved when the market can trade on highly readable news.

In this study, we analyze the trading behavior of banks with lending relationships. We combine detailed German data on banks' proprietary trading and market making with lending information from the credit register and then examine how banks trade stocks of their borrowers around important corporate events. We find that banks trade more frequently and also profitably ahead of events when they are the main lender (or relationship bank) for the borrower. Specifically, we show that relationship banks are more likely to build up positive (negative) trading positions in the two weeks before positive (negative) news events, and also that they unwind these positions shortly after the event. This trading pattern is more pronounced for unscheduled earnings events, M&A transactions, and after borrower obtain new bank loans. Our results suggest that lending relationships endow banks with important information, highlighting the potential for conflicts of interest in banking, which has been a prominent concern in the regulatory debate.

We propose a new approach to automated market making (AMM) by synchronizing two constant function market makers (CFMM). Our methodology combines the advantages of each individual CFMM and reduces the downside risk for liquidity providers (LP). The approach can be extended to any combination of AMMs. The application of our technique described in this paper is suitable for assets with high volatility that can deviate from the entry price for LPs. The deviation risk, compared to a buy-and-hold strategy and often referred to as impermanent loss can counterbalance the benefit of liquidity mining, that is the earning of associated trading fees. We show in this paper how our approach provides a way to largely limit the impact of such risk and compare the behavior of the new AMM to the well known exchanges, such as Uniswap and Balancer.

Automated market makers (AMM) have grown to obtain significant market share within the cryptocurrency ecosystem, resulting in a proliferation of new products pursuing exotic strategies for horizontal differentiation. Yet, their theoretical properties are curiously homogeneous when a set of basic assumptions are met. In this paper, we start by presenting a universal approach to deriving a formula for liquidity provisioning for AMMs. Next, we show that the constant function market maker and token swap market maker models are theoretically equivalent when liquidity reserves are uniform. Proceeding with an examination of AMM market microstructure, we show how the non-linear price effect translates into slippage for traders and impermanent losses for liquidity providers. We proceed by showing how impermanent losses are a function of both volatility and market depth and discuss the implications of these findings within the context of the literature.

This paper studies competition between strategic high-frequency traders (HFTs) and multiple for-profit exchanges. In the model, HFTs play a dual role as liquidity snipers and market makers and strategically decide on their trading venue, the intensity of market monitoring, a bid-ask spread, and speed technologies. With the strategic liquidity provision and HFTs' dual role in the market, I show that the expected bid-ask spread can shrink when adverse selection becomes more severe. I also derive the HFTs' demand for speed services and demonstrate that it can be an increasing function of the length of intentional delays imposed on trade execution by exchange platforms (e.g., speed bumps). In the second part, for-profit exchanges try to maximize their revenues from supplying speed services to HFTs by controlling the speed of order execution. Since the demand for speed services can positively react to the intentional delays in order execution, exchanges are willing to introduce them to boost their profits. Thus the imposition of delays, which mitigates adverse selection and improves liquidity, is supported as an equilibrium outcome even without government intervention.

Exchange-traded funds (ETF or ETFs) are presented as securities with the same trading, clearing and settlement features as listed equity securities. The CFA Institute Research Foundation A Comprehensive Guide to Exchange-Traded Funds (ETFs) notes: “…Confusion abounds about how ETFs work. Claims have even appeared in the popular press that somehow ETFs are a special class of securities subject to different rules when it comes to the back-office processes. Although ETFs are truly unique, the reality is that from the perspective of an investor buying them on the open market, they go through the same settlement and clearing process as any other stock listed on the US stock markets….” As such, the expectation is that operational overhead associated with ETFs would not produce an operational outcome pattern different from the operational outcome pattern of listed equity securities. However, using available public fails-to-deliver data, the data demonstrates ETFs have an operational outcome that is not comparable to the operational outcome pattern for listed equity securities groups. Organizations such as The CFA Institute Research Foundation and Blackrock believe “…a market maker can benefit from delaying settlement for as long as possible…” or “…higher failure-to-deliver rates in ETF shares may merely represent greater market making activity….” A "market maker" is a firm that stands ready to buy or sell a stock at publicly quoted prices . On June 7, 1995 when T+3 settlement came into existence, bona fide market making was allowed an additional 3 days after contractual settlement date to close-out a position. The additional 3 days were to allow for the physical processing and movement of securities by couriers carrying satchels of physical securities. On September 5, 2017, T+2 settlement took effect to reduce settlement risk by a shortened period to convert market liquidity into settlement liquidity. Despite with the change to T+2 settlement, bona fide market making was allowed to continue an additional 3 days to close-out a position. Currently, DTCC is considering a change from T+2 settlement to T+1 settlement to improve liquidity, margin usage, and risk control. White papers from organizations such as DTCC and SIFMA make no mention of improvements to market making close-out efficiency to accompany overall settlement and clearing process improvements. Even with the technological advances in trading, clearing, and settlement since the mid-1990s, market maker close-out efficiency has not improved. This suggests market making close-out efficiency, in particular with respect to ETFs, is a matter for consideration to accompany overall settlement and clearing process improvement efforts.

We consider a stochastic game between three types of players: an inside trader, noise traders and a market maker. In a similar fashion to Kyle's model, we assume that the insider first chooses the size of her market-order and then the market maker determines the price by observing the total order-flow resulting from the insider and the noise traders transactions. In addition to the classical framework, a revenue term is added to the market maker's performance function, which is proportional to the order flow and to the size of the bid-ask spread. We derive the maximizer for the insider's revenue function and prove sufficient conditions for an equilibrium in the game. Then, we use neural networks methods to verify that this equilibrium holds. We show that the equilibrium state in this model experience interesting phase transitions, as the weight of the revenue term in the market maker's performance function changes. Specifically, the asset price in equilibrium experience three different phases: a linear pricing rule without a spread, a pricing rule that includes a linear mid-price and a bid-ask spread, and a metastable state with a zero mid-price and a large spread.

This paper develops a strategic trading model in which the market maker has a monopoly on short-lived information. Given that modern market makers are high frequency traders, it is assumed that market makers are able to process any short-lived information event faster than other traders. Since the market maker’s information is short-lived, informed traders sequentially learn about it and adjust their strategies accordingly. The correlation structure of the market maker’s short-lived information has significant effects on the dynamic trading equilibrium and can generate new stylized facts like negative trading intensities and non-monotonic news sensitivities.

This paper constructs a model of the FX swap valuation, the cost of synthetic dollar borrowing, based on the dealers’ behavior and the hierarchy of international finance. In this model, three fundamentals- market-making costs (measured by dealers’ bid-ask spreads), dollar funding liquidity risk (measured by CIP deviation), and the FX swap market liquidity (measured by the dealers’ competition)- derive the valuation of FX swap. The goal is to understand the financial instability spillovers between FX swaps and offshore US-Dollar funding markets through the “dealers” channel. FX dealers are vital institutions that connect different national monetary jurisdictions in the international monetary system. For currencies where FX turnover is low, market makers are central banks. On the other hand, those dealers are primarily private banks for currencies with the highest FX turnover, such as the US dollar. Nevertheless, studying the “dollar funding gap” through the lens of FX swap dealers is a feature that often gets overlooked in International Political Economy scholarship.

Designated market makers (DMMs) are utilized in many stock markets to provide liquidity and to reduce volatility, but convincing evidence regarding their effects on volatility is scarce. We estimate the effects of DMMs on securities by exploiting exogenous variation in DMM assignment to securities generated by a cutoff rule in the eligibility criteria of securities for DMM designation in the Korean Stock Exchange market. Using a regression discontinuity design, we find that DMMs not only improve the liquidity of securities but that they also increase return volatility. We investigate potential channels for the DMM-induced volatility increase and find evidence suggesting that it is a consequence of increased presence of funds whose managers, as previous studies suggest, engage in noise trading.

In addition to disclosure regulation, the Securities and Exchange Commission (SEC) periodically intervenes in the market making process to facilitate fair, orderly, and efficient capital markets. For example, responding to calls for increased market maker competition on the Nasdaq in the early 1990s, the SEC imposed several reforms in 1997 to decrease bid-ask spreads and dealer rents. While prior research documents that this reform was successful in improving market maker competition and stock liquidity, we examine whether this increase in liquidity sufficiently altered firm’s incentives to disclose, such that firms reduced voluntary disclosure. In a differences-in-differences design, we examine whether firms affected by the reforms subsequently reduced their voluntary disclosure and whether this reduction in disclosure increased information asymmetry among investors. Our results suggest that firms impacted by the regulatory intervention subsequently reduced management forecast frequency relative to a set of firms unaffected by the regulation. Further, the firms that reduced disclosure experienced an increase in information asymmetry among investors through an increased probability of informed trading.

The literature offers two foundations for competitive search equilibrium, a Nash approach and a market-maker approach. When each buyer visits only one seller (or each worker makes only one job application), the two approaches are equivalent. However, when each buyer visits multiple sellers, this equivalence can break down. Our paper analyzes competitive search equilibrium with simultaneous search using the two approaches. We consider four cases defined by (i) the surplus structure (are the goods substitutes or complements?) and (ii) the mechanism space (do sellers post fees or prices?). With fees, the two approaches yield the same constrained efficient equilibrium. With prices, the equilibrium allocation is the same using both approaches if the goods are complements, but is not constrained efficient. In the case in which only prices are posted and the goods are substitutes, the equilibrium allocations from the two approaches are different.

We explore the ties between bonds and individual dealers formed through home advantage and the persistence of previous underwriting relationships. Building on these connections, we show that the introduction of the leverage ratio for the European banks had a large impact on exposed bonds’ liquidity. Moreover, based on these ties, we show that bond mutual fund panic following the 2020 pandemic outbreak affected substantially more mutual funds with the larger exposures to dealer banks’ balance sheet constraints.

Kyle (1985) builds a pioneering and influential model, in which an insider observing private information submits an optimal order given the market-maker’s pricing rule, which is assumed a linear function of the aggregated order flow. We propose an extension to Kyle's model where different types of uncertainty regimes exist and where the market maker estimates market uncertainty and uses it to set her price. The model implies that the elasticity of prices to liquidity demand will increase in high uncertainty regimes. We test the outcome of the model empirically by studying the price formation process during the COVID-19 pandemic crash. A period of agitation with important announcements having a major impact on financial markets, such as the state lockdown and the Fed's fiscal response. We find that indeed the elasticity of prices to liquidity demand during the COVID-19 period has increased four times.

We investigate the market microstructure of Automated Market Makers (AMMs), the most prominent type of blockchain-based decentralized exchanges. We show that the order execution mechanism yields token value loss for liquidity providers if token exchange rates are volatile. AMMs are adopted only if their token pairs are highly correlated, or of high personal use for investors. A pricing curve with higher curvature makes the arbitrage problem less severe but also decreases investors’ surplus. Pooling multiple tokens exacerbates the arbitrage problem. We provide statistical support for our main model implications using transaction-level data of AMMs.

I investigate the effect of trading restrictions on price efficiency in a model with short-term shareholders. I show that insider trading constraints set a floor on prices. More importantly, as trade is the mechanism that transfers information into the price, the market maker’s pricing function has nonconstant sensitivity to order flows, with selling orders impacting the price less than positive orders. These two effects decrease price volatility. Prices will be more efficient in good states of the world but lose their informativeness in bad states of the world. Trading constraints will emerge in equilibrium when shareholders have preferences for short-term market prices. In such a case, an increase in cash flow volatility can translate into more severe limitations, while the impact of noise variance depends on the short-term shareholder’s preferences.

This paper explores whether corporate social performance impacts CDS market liquidity. We first document that better corporate social performance attracts more CDS market makers and induces more frequent CDS price updates. Second, we explore the influence mechanism of CSR on CDS market liquidity. Cross-sectional analyses show that the effect of CSR on liquidity provision is stronger for firms with smaller sizes, fewer analysts following, higher stock return volatility, higher systematic risk or higher credit risk. These findings are consistent with the notion that strong social performance attracts CDS dealers by reducing information asymmetry or offsetting firm risk. Our findings are robust to controlling for more variables, remain qualitatively similar for three subperiods around the 2008 financial crisis period and are robust to the acquisition of CSR data vendors. To our knowledge, this is the first study to focus on the relation between corporate social responsibility and financial market microstructure.

Prior research documents that analyst revisions are important information events and that some investors trade ahead of them. However, prior studies have been unable to show which investors trade ahead of analyst revisions, the aggregate benefits to trading ahead, and the costs borne by various investors from trading against investors that anticipate revisions. To address these questions, we exploit high-frequency trading information from the Korea Exchange (KRX) that characterizes all purchase and sale transactions of twenty different investor groups. We first validate that analyst revisions contain valuable private information: upgrades increase prices by 2.5% and downgrades decrease prices by 2.9%. We next document that asset managers who are geographically and socially close to analysts anticipate recommendations up to two days before they are released, while other institutions do not. Lastly, we show that this ability to anticipate analyst revisions enables these investors to trade profitably against domestic retail investors and foreign professional investors. Local hedge funds earn $2.5 million more than normal when trading against foreign professionals ahead of analyst revisions, which is more than all other investor groups combined. Notably, local asset managers typically lose money when trading against foreign professionals, who seem to earn profits by engaging in high-frequency market making. In sum, asset managers in Korea appear to receive tips from brokerages about upcoming revisions, which shifts the competitive balance between domestic institutions and skilled foreign market makers, and aggregate profits around revisions depend on investor sophistication, geographic proximity, and cultural affinity.

A growing number of blockchain-based decentralized exchanges have adopted automated market makers to attract liquidity, and they replace the traditional order book system for trade execution. This paper studies the equilibrium liquidity provision via constant product market makers, one of the simplest and widely adopted algorithms of automated market making. The model derives the optimal behavior of liquidity providers and the aggregate size of the liquidity pool on a constant product market. As suggested by the traditional market microstructure literature, I demonstrate that information asymmetry between traders leads to an adverse selection problem and plays a key role in determining equilibrium liquidity on an automated market. Moreover, when liquidity providers are strategic, I find that competition between them exhibits both strategic substitution and complementarity, making the best response function of each liquidity provider non-monotonic in her rivals' liquidity supply. The model also shows that the equilibrium market liquidity tends to be stable and does not diverge or evaporate even if a (small) temporary shock hits the market.

This paper develops a strategic trading model in which uninformed orders are allowed to exhibit a general correlation structure that generates autocorrelation in the order flow. Since the order flow is predictable, informed traders and the market maker not only need to infer information about the asset value, but also forecast future order flows. The correlation structure of uninformed orders has significant effects on trading strategies, market liquidity, and the informativeness of prices. Since the empirical autocorrelation in order flows is likely to come from uninformed traders, strategic trading models should not assume them to be simply noise.

We show that High Frequency Traders (HFTs) are not beneficial to the stock market during flash crashes. They actually consume liquidity when it is most needed, even when they are rewarded by the exchange to provide immediacy. The behavior of HFTs exacerbate the transient price impact, unrelated to fundamentals, typically observed during a flash crash. Slow traders provide liquidity instead of HFTs, taking advantage of the discounted price. We thus uncover a trade-off between the greater liquidity and efficiency provided by HFTs in normal times, and the disruptive consequences of their trading activity during distressed times.

We study algorithmic trading strategies in order driven markets. We make three contributions to the literature. One, we show how a market maker employs information about the momentum in the price of the asset to design liquidity provision strategies. The momentum in the midprice of the asset depends on the arrival of liquidity taking orders and the arrival of news. Buy market orders (MOs) exert a short-lived upward pressure on the midprice and sell MOs exert a downward pressure of the price. We employ high-frequency data to estimate model parameters and show the performance of the market making strategy. Two, we model the trading strategy of an investor who spoofs the limit order book (LOB) to increase the revenue she obtains from selling a position in an asset. The strategy employs, in addition to sell limit orders (LOs) and sell market orders (MOs), a large number of spoof buy LOs to manipulate the volume imbalance of the LOB. Our results show that spoofing considerably increases the revenues from liquidating a position. The spoof strategy employs, on average, fewer sell MOs (than a strategy without spoof LOs) and from executing roundtrip trades that are initiated by buy spoof LOs that are inadvertently filled and subsequently unwound with sell LOs. Spoofing is illegal and difficult to detect. We show that as the financial penalty for spoofing increases, the spoof strategy relies less on spoof LOs. There is a critical point where the gains from spoofing are outweighed by the financial penalty, so it is optimal no not to spoof the LOB. Three, we show how the supply of liquidity in order driven markets is affected if LOs are forced to rest in the LOB for a minimum resting time (MRT) before they can be cancelled. The bid-ask spread increases as the MRT increases because market makers (MMs) increase the depth of their LOs to protect them from being picked off by other traders. The expected profits of the MMs increase when the MRT increases. The intuition is as follows. As the MRT increases, there are two opposing forces at work. (i) The longer the MRT, the more likely the LOs are to be filled and, on average, shares are sold at a loss. (ii) because the depth of the posted LOs increases, the probability that the LO is picked off by other traders before the end of the MRT decreases. The net effect is that a longer MRT leads to a higher expected profit. We also show that the depth of LOs increases when the volatility of the price of the asset increases. Also, the depth of LOs increases when the arrival rate of market orders increases because it is less likely that LOs will be picked off by the end of the MRT. Finally, our model also makes predictions about the overall liquidity of the market. We show that MMs choose to supply the minimum amount of shares per LO allowed by the exchange because expected profits are maximised when liquidity provided is lowest.

Following the Dodd-Frank Act, central clearing and centralized trading became mandatory for a class of the most liquid interest rate swaps (IRS). Nevertheless, IRS market making in the dealer-to-client sector remained concentrated at a few regulated banks that transfer their funding costs to end users. I develop an equilibrium model of IRS markets with imperfect competition among capital-constrained dealers. Using proprietary data on cleared IRS transactions and the dealers' daily margin requirements, I empirically investigate the impact of recent regulatory changes on market liquidity. By exploiting the variation in margin requirements across swaps with different risk profiles intermediated by bank-affiliated dealers, my estimates show that a 1 percentage point increase in the supplementary leverage ratio (SLR) leads to an increase of $1.5 billion in the annual cost of hedging for the end-users.

We document a significant, up to 10-fold increase in the intra-day correlation of firm-specific and market returns over the last decade. This surge in the intra-day correlation of returns coincided with the advent of electronic, automated trading in U.S. markets. Using changes to the S&P500 index, we establish evidence of a causal relationship. When firms are included in this major index, they enter the radar of high frequency arbitrageurs and market making algorithmic traders. These trading robots, who monitor prices in major securities closely and continuously, increase their quoting activities significantly and cause individual stocks’ returns to align more closely with the market.

We establish connections between optimal transport theory and the dynamic version of the Kyle model, including new characterizations of informed trading profits via conjugate duality and Monge-Kantorovich duality. We use these connections to extend the model to multiple assets, general distributions, and risk-averse market makers. With risk-averse market makers, liquidity is lower, assets exhibit short-term reversals, and risk premia depend on market maker inventories, which are mean reverting. We illustrate the model by showing that implied volatilities predict stock returns when there is informed trading in stocks and options and market makers are risk averse.

This study provides a liquidity-supply side model of options markets for illustrating how options pricing uncertainty affects moment risk premia. The model is based on micro-structure theory such that a representative market maker dynamically replicates options prices, hedges risky positions, and seeks the minimum variance in duplicating errors. Therefore, unlike previous studies extending the pricing kernel of underlying assets to explain moment risk premia, our model focuses on options' hedging uncertainty: i.e., the pricing kernel that distinguishes options from ordinary equities. The risk premia in our model are attributed to options pricing uncertainty. Our model also provides a new perspective on the effects of market friction and the non-normality assumption on moment risk premia and the implied volatility curves.

Economists usually assume that price and quantity are continuous variables, while most market designs, in reality, impose discrete tick and lot sizes. We study a firm’s trade-off between these two discretenesses in U.S. stock exchanges, which mandate a one-cent minimum tick size and a 100-share minimum lot size. A uniform tick size favors high prices because the bid–ask spread cannot be lower than one cent. A uniform lot size favors low prices because low prices reduce adverse selection costs for market makers when they have to display at least 100 shares. We predict that a firm achieves its optimal price when its bid–ask spread is two ticks wide, when the marginal contribution from discrete prices equals that from discrete lots. Empirically, we find that stock splits improve liquidity when they move the bid–ask spread towards two ticks; otherwise, they reduce liquidity. Liquidity improvements contribute 95 bps to the average total return on a split announcement of 272 bps. Optimal pricing can increase the median U.S. stock value by 69 bps and total U.S. market capitalization by $54.9 billion.

We analyze a market with private information about cash flows as well as their betas. The unique feature of beta information is that it multiplicatively affects cash flow volatility conditional on order flows. This aspect, together with market makers' risk aversion and betas' non-negativity, implies a buy-sell asymmetry in the price impact function, and a positive covariance between stochastic price impacts and order flows. We predict a negative relation between this covariance and expected returns, and an attenuation of the beta anomaly in asset returns net of this relation. Empirical tests confirm these predictions.