Journal Article Trading Fast and Slow: Colocation Liquidity Get access Jonathan Brogaard, Brogaard Foster School of Business, University Washington Search for other works by this author on: Oxford Academic Google Scholar Björn Hagströmer, Hagströmer Stockholm Business School, Lars Nordén, Nordén Ryan Riordan Queen's The Review Financial Studies, Volume 28, Issue 12, December 2015, Pages 3407–3443, https://doi.org/10.1093/rfs/hhv045 Published: 12 August 2015

We study performance and competition among firms engaging in high-frequency trading (HFT). construct measures of latency find that differences relative account for large HFT firms’ performance. improve their rank due to colocation upgrades see improved The stronger associated with speed comes through both the short-lived information channel risk management channel, is useful various strategies, including market making cross-market arbitrage. empirical support many predictions regarding competition.

The effective bid-ask spread measured relative to the midpoint overstates true in markets with discrete prices and elastic liquidity demand. average bias is 13%–18% for S&P 500 stocks general, depending on estimator used as benchmark, up 97% low-priced stocks. Cross-sectional variation across stocks, trading venues, investor groups can influence research inference. use of also undermines timing performance evaluations, lead non-sophisticated investors overpay liquidity. To overcome these...

The regulatory debate concerning high-frequency trading (HFT) emphasizes the importance of distinguishing different HFT strategies and their influence on market quality. Using data from NASDAQ-OMX Stockholm, we compare market-making HFTs to opportunistic HFTs. We find that makers constitute lion's share volume (65-71%) limit order traffic (81-86%). Furthermore, have higher order-to-trade ratios lower latency than In a natural experiment based tick size changes, activity mitigates intraday...

We study performance and competition among high-frequency traders (HFTs). construct measures of latency find that differences in relative account for large HFTs' trading performance. HFTs improve their rank due to colocation upgrades see improved The stronger associated with speed comes through both the short-lived information channel risk management channel, is useful various strategies including market making cross-market arbitrage. empirical support many predictions regarding competition.

Fragmented trading of the same security across multiple venues is prevalent in modern stock markets. This paper investigates how market quality effects associated with such fragmentation depend on type exchange competition. Using a natural experiment where goes from fragmented to centralized overnight, we show that liquidity falls, but surprisingly, effect unrelated degree fragmentation. Instead, reduction concentrated among stocks before event had most intense competition supply. The...

We study order aggressiveness of market-making high-frequency traders (HFTs), opportunistic HFTs, and non-HFTs. find that HFTs follow their own group's previous submissions more than they other traders’ orders. Opportunistic non-HFTs tend to split market orders into small portions submitted in sequence. submit (less) aggressive when the same-side (opposite-side) depth is large, supply liquidity bid-ask spread wide. Thus, adhere strongly trade-off between waiting cost immediate execution....

Crude oil markets witness growing disparity between the quality of crudes supplied and demanded in market. The market share low‐quality is increasing due to depletion old fields demand. This unnerving practitioners affecting relevance traditional benchmark lack lower benchmarks (Montepeque, 2005 Montepeque, J. 2005. Sour crude pricing: a pressing global issue. Middle East Economic Survey, 48: 1–42. [Google Scholar]). In this article, we apply Granger causality tests study price dependence 32...

We exploit an optional colocation upgrade at NASDAQ OMX Stockholm to assess how speed affects market liquidity. Liquidity improves for the overall and even noncolocated trading entities. find that is pursued mainly by participants who engage in making. Those use their enhanced reduce exposure adverse selection relax inventory constraints. In particular, upgraded entities remain competitive best bid offer when inventories are top decile. Our results suggest increasing of market-making benefit

Abstract We study order aggressiveness of market‐making high‐frequency traders (MM‐HFTs), opportunistic HFTs (Opp‐HFTs), and non‐HFTs. find that MM‐HFTs follow their own group's previous submissions more than they other traders’ orders. Opp‐HFTs non‐HFTs tend to split market orders into small portions submitted in sequence. submit (less) aggressive when the same‐side (opposite‐side) depth is large, supply liquidity bid–ask spread wide. Thus, adhere strongly tradeoff between waiting cost...

ABSTRACT How does information get revealed in decentralized markets? We test several hypotheses inspired by recent dealer‐network theory. To do so, we construct an empirical map of revelation where two dealers are connected based on the synchronicity their quote changes. The tests, euro to Swiss franc spot rate (EUR/CHF) data including 2015 crash, largely support theory: strongly (i.e., central) more informed. Connections weaker when there is less be learned. crash serves identify how a...

Abstract We investigate the effects from introduction of a closing call auction (CCA) at index futures market. Limit order book models, where trader patience determines trading strategies, predict that CCA increases and, hence, improves price accuracy and end‐of‐day liquidity. find leads to increased patience, improved accuracy, unaffected tightness resiliency, decreased depth. Decreased depth is likely due less fishing activity. With CCA, opportunistic patient traders' posting limit orders...

Full-Scale Optimization (FSO) is a utility maximization approach to portfolio choice problems that has theoretical appeal but suffers from computational burden in large scale problems. We apply the heuristic technique differential evolution solve FSO-type asset selection of 97 assets under complex functions rendering rough search surfaces. show this problem computationally feasible and solutions retrieved with random starting values are converging one optimum. Furthermore, study constitutes...

We investigate the economic rationale behind limit order cancellations from perspective of liquidity suppliers. predict that an is cancelled whenever its expected revenue no longer exceeds cost and we model how profitability variation can be determined changes in state book queue position. Our empirical evidence supports predictions general for orders submitted by high-frequency trading firms particular. Consistent with our approach, find cancellation patterns are more consistent market...

We develop a structural model for the price formation and liquidity supply of an asset. Our facilitates decompositions both bid–ask spread return variance into components related to adverse selection, inventory, order processing costs. Furthermore, shows how fragmentation trading volume across venues influences inventory pressure discovery. use analyze intraday gold futures traded at Shanghai Futures Exchange. find that costs explain about 50% spread, whereas remaining is equally due...

How does information get revealed in decentralized markets? We test several hypotheses inspired by recent dealer-network theory. To do so we construct an empirical map of revelation where two dealers are connected based on the synchronicity their quote changes. The tests, EUR/CHF data including 2015 crash, largely support theory: strongly (i.e., central) more informed. Connections weaker when there is less to be learned. crash serves identify how a network forms transitioned from no-learning...

Portfolio choice by full‐scale optimization applies the empirical return distribution to a parameterized utility function, and maximum is found through numerical optimization. Using portfolio setting of three UK equity indices we identify several functions featuring loss aversion prospect theory, under which substantially better approach than mean–variance approach. As have distributions with small deviations from normality, findings indicate much broader usefulness has earlier been shown....

Abstract To improve the efficiency of closing price, many equity exchanges apply volatility extensions to their call auctions (CCAs). If an imminent auction execution implies a large price change, order submission period is extended let traders reconsider orders. This paper uses introduction at NASDAQ Nordic provide first analysis effects such mechanisms. We find that reduce transitory and deter manipulation close. Consistent with increased trust in mechanism, CCA attracts higher volumes...

Recently, the Shanghai Futures Exchange (SHFE) introduced gold futures trading in China. This paper is first to study SHFE futures, and evaluate hedging effectiveness since introduction. The results show that with reduces variance of a hedged spot position by about 88% its two years existence. During second half 2008, however, when global financial crisis escalated, reduction dropped 70%. Overall, new Chinese prove be attractive well-needed vehicles for domestic producers, refiners,...

This is the internet appendix for "How Aggressive are High-Frequency Traders".

This paper investigates whether investors are compensated for taking on commonality risk in equity portfolios.A large literature documents the existence and causes of illiquidity, but implications less understood.We find a return premium NYSE stocks that is both economically statistically significant.The independent illiquidity level effects, robust to variations measurement systematic estimation.We also show precision estimation can be increased by use daily measures, instead monthly.

Abstract Almost all limit orders are canceled. We examine two economic channels that can motivate cancellations: reductions in the expected profit at execution, and probability of execution. An order‐level analysis shows changes depth best bid offer prices, as well order queue position, influence cancellation a way consistent with former channel, market makers monitor execution each order. Although buy‐side investors use passive extensively, our findings indicate cancellations on aggregate...