Brokers and Order Flow Leakage: Evidence from Fire Sales

• DOI: http://doi.org/10.1111/jofi.12840
• Published date: 01 December 2019
• Author: FRANCESCO FRANZONI,ANDREA BARBON,AUGUSTIN LANDIER,MARCO DI MAGGIO
• Date: 01 December 2019

THE JOURNAL OF FINANCE •VOL. LXXIV, NO. 6 •DECEMBER 2019

Brokers and Order Flow Leakage:

Evidence from Fire Sales

ANDREA BARBON, MARCO DI MAGGIO, FRANCESCO FRANZONI,

and AUGUSTIN LANDIER∗

ABSTRACT

Using trade-level data, we study whether brokers play a role in spreading order flow

information in the stock market. We focus on large portfolio liquidations that result

in temporary price drops, and identify the brokers who intermediate these trades.

These brokers’ clients are more likely to predate on the liquidating funds than to

provide liquidity. Predation leads to profits of about 25 basis points over 10 days and

increases the liquidation costs of the distressed fund by 40%. This evidence suggests

a role of information leakage in exacerbating fire sales.

“LARGE INSTITUTIONAL ORDERS”ARE TYPICALLY split into smaller amounts over

time to avoid moving the market (see Garleanu and Pedersen (2013), Di Mascio,

Lines, and Naik (2016)). One concern when executing an order over time is that

other traders might anticipate the intent to trade the stock in the near future

and trade in the same direction to benefit from the future price impact. This

problem is particularly pronounced in the case of fire sales, during which the

seller is forced to bring to the market a large quantity of assets in a limited

amount of time (Coval and Stafford (2007), Ellul, Jotikasthira, and Lundblad

(2011)). Moreover, if the liquidation occurs during a time of market stress,

predatory trading can make the market more illiquid and amplify adverse

shocks (Greenwood, Landier, and Thesmar (2015)). Given this possibility, some

observers suggest that reducing the frequency of portfolio disclosure can help

prevent predatory behavior (Brunnermeier and Pedersen (2005)).

∗Andrea Barbon is at USI Lugano and Swiss Finance Institute. Marco Di Maggio is with Har-

vard Business School and NBER. Francesco Franzoni is at USI Lugano, Swiss Finance Institute,

and Center for Economic Policy and Research. Augustin Landier is with Toulouse School of Eco-

nomics. We thank Malcolm Baker; John Campbell; Laurent Fr´

esard; Slava Fos (discussant); Joel

Hasbrouck; Terry Hendershott (discussant); Gary Gorton; Owen Lamont; Jongsub Lee (discus-

sant); Andrew Lo; TobyMoskowitz; Abhiroop Mukherjee (discussant); Cameron Peng (discussant);

Erik Stafford; and seminar/conference participants at Aalto University, the Becker Friedman In-

stitute CITE conference on New Quantitative Models of Financial Markets, Bocconi University,

EIEF, FCA, FINRA Market Structure conference, FIRS, LAEF 2nd OTC Markets and Securities

Workshop, INSEAD, Macro-Prudential Conference ECB, New York Federal Reserve Bank, AQR,

UC Dublin, and WFA for helpful comments. Augustin Landier directs financial research at AXA-

Investment Management Chorus, a quantitative asset management affiliate of Axa-IM. Neither

Andrea Barbon, Marco Di Maggio, Francesco Franzoni nor any of their relatives have received

significant financial support from any interested party,or have positions in relevant organizations.

The authors have signed a nondisclosure agreement for the proprietary data used in this study.

DOI: 10.1111/jofi.12840

C2019 the American Finance Association

2707

2708 The Journal of Finance R

However, the market may possess information about forced liquidations

due to broker’s close relationship with the liquidating managers. Brokers are

uniquely able to observe the daily trades of a fund. In the case of hedge funds,

prime brokers also operate as lenders and risk managers, and thus they know

when a fund is about to breach some risk limit and deleverage its portfolio.

They can also infer their client’s trading habits, such as whether a client tends

to cut trades into small orders over several days when executing a large or-

der. Due to this information, brokers are well placed—indeed they may be best

placed among market participants—to predict the future trades of their clients.

In an effort to establish a reputation as a source of valuable information

and attract new business, brokers may leak the news that a client’s large

trade is likely to extend over time, as other investors can use this information

to predate on the distressed fund. Alternatively, brokers may be reluctant to

foster predatory trading against a client, as doing so may work against their

reputation. According to the latter argument, brokers should instead invite

other traders to provide liquidity and take the other side of the slow trade. Thus,

whether brokers foster predatory trading or whether they support liquidity

provision in the case of slow trading by a client is an open empirical question.

In this paper, we shed light on such a question using data on forced liquidations

of portfolio holdings.1Specifically, we exploit proprietary trade-level data to

identify asset managers who sell a significant fraction of their portfolio over a

relatively short amount of time. We restrict attention to asset managers whose

order flow is abnormally negative for at least five days in a row. Moreover, we

focus on managers who liquidate multiple stocks (on average about 20 stocks)

at a significantly faster pace than usual. We identify approximately 400 events

satisfying these criteria over the period 1999 to 2014. We verify that the stock

price movements resulting from such sales are only temporary, consistent with

the identification of liquidity events—the price impact would have to display a

permanent component if sales were motivated by fundamentals.

Not all brokers employed by a liquidating fund will be aware that the fund

is in distress. The liquidating fund has little incentive to disclose its intention

to liquidate a large fraction of its portfolio. Indeed, it is likely to use multiple

brokers (on average 29) to minimize the price impact and information leak-

age. We therefore label only those brokers as aware who intermediate a large

enough fraction of volume. We find that the probability of predatory behavior is

significantly higher for orders executed through aware brokers. In particular,

clients of aware brokers are much more likely to execute sell trades in the same

1We focus on large liquidations (which for convenience we label “fire sales”), that is, we do not

include large purchases in our analysis, because we aim to cleanly identify liquidity-motivated

trades. In our data, the majority of institutional investors are long-only (about 90%) and thus a

sale is less likely to be information motivated (as the manager would need to already have the stock

in her portfolio) than a buy transaction. In addition, large cash inflows can be allocated slowly over

time and are, therefore, less likely to impose a concentrated liquidity demand on the market than

large outflows. Fire sales can also pose a systemic threat if they cause a propagation of idiosyncratic

shocks to the balance sheets of other investors. Hence, studying the effect of information leakage

on fire sales is especially relevant, including from a regulatory perspective.

Brokers and Order Flow Leakage 2709

stocks with the same broker over the same period. Although clients of aware

brokers also engage in liquidity provision, this activity does not appear to be

as prevalent as predatory trading.

We next explore the extent of heterogeneity across different clients of aware

brokers. If brokers leak information about order flow, they are more likely to

leak information to their best clients, that is, to those clients from which they

extract the highest rents. As proxies for the strength of the investor-broker

relation, we employ both the trading volume and the commissions generated

by a client. Using both measures we find that best clients of the aware brokers

are significantly more likely than other clients to sell the stocks that the liq-

uidating manager is offloading during the fire sale compared to immediately

before the fire sale.2,3The magnitude of this result is economically significant:

the net probability of predation for the best clients of aware brokers is more

than twice that for the small clients of aware brokers. This evidence suggests

that predation is more likely than liquidity provision among the best clients

of the brokers who intermediate fire sales. As additional evidence of preda-

tory trading, we find that a significant fraction of the positions sold by other

managers than the distressed fund during the fire-sale period (30% to 42%) is

bought back in the 10 days following the fire sale.

We conduct several robustness checks to rule out the possibility that a

fire sale’s originator and followers are trading in response to a common

information signal. We first exclude from our sample all events that occurred

during recessions. We next exclude all events that occurred around earnings

announcements, changes in analyst recommendations, or any other type of

negative news as reported by the press and classified by the data provider

Ravenpack. We further exclude stocks with negative momentum and high

short interest to mitigate the concern that selling managers follow similar

trading strategies founded on a negative signal on the stock. In each case, our

results continue to hold.

To strengthen our identification of fire-sale events, we examine a natural ex-

periment in which some mutual funds were forced to liquidate their holdings.

Specifically, as a consequence of the late-trading scandal of 2003, 27 fund fami-

lies experienced significant outflows. Anton and Polk (2014) use these outflows

to identify an exogenous determinant of mutual funds’ selling activity. Kisin

(2011) estimates that funds of implicated families lost 14.1% of their capital

within one year and 24.3% within two years. Important for our purposes, the

brokers of the liquidating funds were aware of both the specific stocks that

were being sold and the timing of these liquidations. We show that, after the

2We find that these relations are extremely persistent, consistent with the findings in Goldstein

et al. (2009), which suggests that brokers might have an incentive to nurture such relations.

3We control for time, manager, event, stock, and broker fixed effects. Hence, differences across

stocks, such as their liquidity, or across brokers, such as their ability to execute, cannot explain

our results. We also run a specification in which we control for broker-manager relationship fixed

effects, which controls for the matching between asset managers and brokers. The results, which

are in the Internet Appendix, are qualitatively similar. The Internet Appendix is available in the

online version of this article on the Journal of Finance website.