Batch Auctions Research Articles

Frequent batch auction versus continuous time auction under order cancellation and maker‐taker fee

AbstractWe investigate the consequences of different market designs and policies on market quality in a high‐frequency market. Based on an extensible theoretical framework, high‐frequency traders in our model can be either market makers or arbitragers, which leads to a Nash equilibrium between their utilities. We consider the optimal strategies of different market agents in various market conditions based on the equilibrium. We find frequent batch auctions benefit market liquidity but harm market volatility compared to continuous‐time auctions. Order cancellation ban harms market quality while the taker fee is beneficial. We also find that market design and policy changes are only effective in low‐latency markets. Finally, we address that the total effect of high‐frequency trading is positive, so it is vital to reap high‐frequency traders' benefits while minimising their harms in high‐frequency markets.

The market quality effects of sub-second frequent batch auctions: Evidence from dark trading restrictions

Recent European regulatory restrictions on dark trading induced an increase in sub-second frequent batch/periodic auctions (PA). We exploit this development to investigate the effects of PA on market quality. The restrictions are linked to an observable increase in PA and an economically meaningful loss of liquidity. PA is also associated with a significant decline in liquidity and informational efficiency. However, consistent with Budish et al. (2015 – The Quarterly Journal of Economics, 130, 1547), increased execution via PA leads to a decline in adverse selection costs, which underscores its potential as a trading mechanism for addressing latency arbitrage and the technological arms race.

Insights on the Statistics and Market Behavior of Frequent Batch Auctions

This paper extends previous research performed with the SHIFT financial market simulation platform. In our previous work, we show how this order-driven, distributed asynchronous, and multi-asset simulated environment is capable of reproducing known stylized facts of real continuous double auction financial markets. Using the platform, we study a pricing mechanism based on frequent batch auctions (FBA) proposed by a group of researchers from University of Chicago. We demonstrate our simulator’s capability as an environment to experiment with potential rule changes. We present the first side-by-side comparison of frequent batch auctions with a continuous double auction. We show that FBA is superior in terms of market quality measures but we also discover a potential problem in the technical implementation of FBA.

Frequent Batch Auctions vs. Continuous Trading: Evidence from Taiwan

We exploit the shift from frequent batch auctions to continuous trading at the Taiwan Stock Exchange to show that liquidity deteriorated in large-cap and efficiency significantly improved in mid-cap and small-cap after trading became continuous. Our results reveal that the migration to the continuous limit order book had heterogeneous effects across stocks. The lack of negative abnormal returns in large-cap for which both liquidity and marginally, efficiency declined, suggests investors value highly the opportunity to trade continuously. Continuous trading increased profits for fast investors that engage in liquidity provision while losses incurred by individual investors did not change significantly.

Can the European intraday market be designed as a congestion management tool?

The growth in intermittent renewable power in Europe has increased the need to trade closer to real-time. In recent years, intraday markets have been integrated across Europe, contributing to a more efficient market. On the other hand, European day-ahead and intraday markets are still based on zonal pricing, where the physical characteristics of transmission networks are only partly taken into account, and congestion problems may remain until close to real-time. In this paper, we suggest an intraday market design based on the coordinated multilateral trade (CMT) approach, using power transfer distribution factors and other network information provided by the transmission system operators, to generate profitable, but feasible, and possibly multilateral trades based on nodal bids. Profitable trades can be found by independent brokers or by having the power exchange running frequent batch auctions at discrete-time intervals. Each trade is accepted by the TSO if no violation occurs in the network, or is curtailed. If the network flow is feasible at the start of the intraday market, the procedure converges to the optimal economic dispatch. In any case, this suggestion for intraday trading, taking into account simple information about network constraints, in a procedure consistent with the functioning of the European intraday market, can help bridge the gap between the zonal day-ahead markets and the real-time constraints of the power system.

Frequent Batch Auctions vs Continuous Time Auctions Under Order Cancellation and Maker-Taker Fee

Frequent Batch Auctions vs Continuous Time Auctions Under Order Cancellation and Maker-Taker Fee

Frequent Batch Auctions and Informed Trading

Frequent Batch Auctions and Informed Trading

To batch or not to batch? The release of USDA crop reports

AbstractIn 2012 USDA began releasing crop reports during trading hours. Prior reports were released during a trading halt, and in the electronic trading platform traders competed in a batch auction for 2 h. Compared to the current real‐time release policy, the batch auction reduced speed advantage, lowered market volatility, and improved liquidity around report releases, at the cost of delaying price discovery. The results reported here support the argument that the development of a shorter batch auction could improve on the lengthy price discovery process while maintaining the batch auction's other advantages. These findings are relevant to not only developed countries with well‐established exchanges and public information systems, but to developing countries who have recently launched or are planning to launch futures commodity markets.

Trading On-Chain: How Feasible is Regulators’ Worst-Case Scenario?

When consumers trade financial products, they typically use well-identified service providers that operate under government regulation. In theory, decentralized platforms like Ethereum can offer trading services ‘on-chain’ without an obvious entry point for regulators. Fortunately for regulators, most trading volume in blockchain-based assets is still on centralized service providers for performance reasons. However this leaves the following research questions we address in this paper: (i) is secure trading (i.e., resistant to front-running and price manipulation) even feasible as a fully ‘on-chain’ service on a public blockchain, (ii) what is its performance benchmark, and (iii) what is the performance impact of novel techniques (e.g., ‘rollups’) in closing the performance gap? To answer these questions, we ‘learn by doing’ and custom design an Ethereum-based call market (or batch auction) exchange, Lissy, with favorable security properties. We conduct a variety of optimizations and experiments to demonstrate that this technology cannot expect to exceed a few hundred trade executions per block (i.e., 13s window of time). However this can be scaled dramatically with off-chain execution that is not consumer-facing. We also illustrate, with numerous examples throughout the paper, how blockchain deployment is full of nuances that make it quite different from developing in better understood domains (e.g., cloud-based web applications).

Latency Arbitrage and Frequent Batch Auctions

Latency Arbitrage and Frequent Batch Auctions

Non-Markov rate kernels: Application to batch auction

Non-Markov rate kernels: Application to batch auction

Batch Auction Design for Cloud Container Services

Cloud containers represent a new, light-weight alternative to virtual machines in cloud computing. A user job may be described by a container graph that specifies the resource profile of each container and container dependence relations. This work is the first in the cloud computing literature that designs efficient market mechanisms for container based cloud jobs. Our design targets simultaneously incentive compatibility, computational efficiency, and economic efficiency. It further adapts the idea of batch online optimization into the paradigm of mechanism design, leveraging agile creation of cloud containers and exploiting delay tolerance of elastic cloud jobs. The new and classic techniques we employ include: (i) compact exponential optimization for expressing and handling non-traditional constraints that arise from container dependence and job deadlines; (ii) the primal-dual schema for designing efficient approximation algorithms for social welfare maximization; and (iii) posted price mechanisms for batch decision making and truthful payment design. Theoretical analysis and trace-driven empirical evaluation verify the efficacy of our container auction algorithms.

High‐frequency trading order cancellations and market quality: Is stricter regulation the answer?

AbstractHigh‐frequency traders (HFTs) frequently submit, cancel and resubmit trading orders in an attempt to stay in front of the queue. This study shows that HFTs cancel a large number of limit orders within 50 ms in order to create arbitrage opportunities in the Australian Stock Exchange (ASX). We find that HFTs generate substantial profits after transaction costs and are capable of extending their strong profitability generating record in the future. Our empirical results indicate that institutional investors lacking speed are more likely to experience higher execution costs. We propose the introduction of batch auctions, one per 50 ms of trading, to impose a queuing risk for HFTs leading to positive market quality effects.

Speed and learning in high-frequency auctions

Faster trading improves liquidity in periodic call auction markets, in contrast to continuous-time markets. We build a model where high-frequency traders (HFTs) engage in duels to trade on stale quotes. More frequent periodic auctions increase the likelihood that a single HFT arrives in any given auction and subsequently acts as a monopolist on information. Higher trading speed increases the expected number of arbitrageurs participating in auctions, promoting competition between snipers and improving liquidity. We find that faster trading and longer auction intervals are substitute instruments to reduce bid-ask spreads. Relative to continuous-time trading, periodic batch auctions reduce HFT informational rents.

High‐Frequency Trading and Market Performance

ABSTRACTWe study the consequences of, and potential policy responses to, high‐frequency trading (HFT) via the tradeoff between liquidity and information production. Faster speeds facilitate HFT, with consequences for this tradeoff: Information production decreases because informed traders have less time to trade before HFTs react, but liquidity (measured by the bid‐ask spread) improves because informational asymmetries decline. HFT also pushes outcomes inside the frontier of this tradeoff. However, outcomes can be restored to the frontier by replacing the limit order book with one of two alternative mechanisms: delaying all orders except cancellations or implementing frequent batch auctions.

On Frequent Batch Auctions for Stocks

Abstract I show that frequent batch auctions for stocks have the potential to reduce the severity of stock price crashes when they occur. For a given sequence of orders from a continuous electronic limit order book market, matching orders using one-second apart batch auctions results in nearly the same trades and prices. Increasing the time interval between auctions to one minute significantly reduces the severity stock price crashes. In spite of this and other advantages pointed out in the literature, frequent batch auctions have not caught on. There is a need for carefully designed market experiments to understand why and what aspect of reality academic research may be missing.

On the Effects of Continuous Trading

The continuous limit order book, in which messages are processed one by one in the order of receipt, is a prominent design feature of modern securities markets. Theoretical models show that this design imposes a cost on liquidity providers and suggest that this cost may be reduced by switching to batch auctions. We examine a recent opposite move, whereby a stock exchange switches from batch auctions to continuous trading. Consistent with theoretical predictions, we find that the move leads to greater adverse selection. Trading costs increase as a result, while displayed liquidity deteriorates.

Experiments in high-frequency trading: comparing two market institutions

We implement a laboratory financial market where traders can access costly technology that reduces communication latency with a remote exchange. In this environment, we conduct a market design study on high-frequency trading: we contrast the performance of the newly proposed frequent batch auction (FBA) against the continuous double auction (CDA), which organizes trades in most exchanges worldwide. Our evidence suggests that, relative to the CDA, the FBA exhibits (1) less predatory trading behavior, (2) lower investments in low-latency communication technology, (3) lower transaction costs, and (4) lower volatility in market spreads and liquidity. We also find that transitory shocks in the environment have substantially greater impact on market dynamics in the CDA than in the FBA.

Regulations and Technology Behind HFT Latency, Batch Auctions and Payments for Order Flow in the US and EU

Abstract Since the appearance of high-frequency trading in the 1990s, speed has become one of the key issues in trading and with it, the controversy around High-Frequency Trading. In recent years, there have been many discussions and analyses of how high-frequency trading may affect the financial market – but still without any clear conclusions. Leaving these opinions behind, many adjustments have already been made in the US and Europe - both to regulations and market rules, impacting not only High-Frequency Trading but general electronic trading as well. These rules and regulations are the result of technological developments in electronic trading and more specifically, High-Frequency Trading and the practice of Payments for Order Flow. The question remains as to how deep regulations should go, especially in the case of HFT which can be severely affected by harsh regulatory requirements or procedures. Because two of the most important issues in HFT are time and information, some of the rules and regulations affect aspects such as not only what type of information and how it should be gathered, but also clock synchronisation and time-stamp granularity. Another issue that may be considered controversial in the field of HFT (although it is not a practice limited to HFT) is Payment For Order Flows. Under this mechanism, wholesale market makers pay brokers for their client’s order flow – a practice that performed in great amounts and at high speeds may give a considerable level of “inside” information. Regulations, especially from ESMA (MiFID II). try in great part to thus mitigate the practice of Payments For Order Flows. The aim of this paper is to present technological advancements in the field of trading communications used, not only by HFT firms, but also by exchanges. Additionally, the objective is to underline some challenges regarding regulatory changes that try to adapt to the current level of technology – for example, those related to clock synchronisation and data processing. One last issue brought forward is the possibility of converting markets from continuous-time auctions to discrete-time auctions - a concept that is aimed at liquidating the speed advantage and competition only to price level and hence, eliminating HFT advantages.

Machine Learning and High-Frequency Algorithms during Batch Auctions

We present the first direct evidence of algorithmic imprints during batch auctions. Order anticipation is an integral part of high-frequency traders' strategies. Hence, some participants may have economic incentive to encrypt noise in the data. We use machine learning to identify five types of algorithmic imprints that hinder the processing of auction information and have the encrypted noise characteristics. Our approach rests on the shifted wavelet tree (Yunyue and Shasha (2003)), a burst detection indicator, and the dynamic time warping similarity measure (Skutkova, Vitek, et al. (2013)). We show that market participants can adapt their trading to the presence of encrypted noise by filtering data in real time, thus clarifying the price discovery process. This could reveal the presence of informed traders. The methodology deployed is adaptable to different environments, including continuous trading.