Liquidity Commonality

Question 1

Chordia, Roll, Subrahmanyam 2000

Answer 1

• Mot: trading vol determinant of dealer invty → vol variation induces co-move in opt invty lvls → co-move in bid-ask, depth, etc.
• Invty carry costs depend on mkt int rates → shd also co-move. Ex: Institutional funds with sim invest styles → corr trade patterns → induce chgs in invty across mkt sectors → invty flux corr across assets ⇒ liq also co-moves
• Find: (1) indiv liq measures co-move with each other (2) stk spreads pos related to num of indiv txs but neg related to aggr lvl of trading
• Interp: manifestation of (1) decr in invty risk from greater mkt-wide trade activity by uninformed (2) incr assymm info risk from informed hiding trades by breaking into small units → incr numb of txs

Question 2

Hasbrouck and Seppi 2001

Answer 2

• How important are cross-stock common factors in the price discovery/liquidity provision process in equity mkts?
• investigate two aspects of this Q for the 30 Dow stocks.
• (1) using principal components & canonical corr analyses → commonality order flows explains roughly 2/3 commonality in returns.
• (2) examine variation & common covariation in various liq proxies & mkt depth (trade impact) coefs. → liq proxies such bid-ask spread & bid-ask quote sizes help explain time variation in TI, BUT common factors in these liq proxies are relatively small.

Question 3

Domowitz, Hansch, and Wang 2005

Answer 3

• Mot: rather than just describe liq commonality, examine its cause & evaluate its asset pricing imps
• Arg: two types of co-move result of diff economic forces (1) order type: mkt orders always consume liq, limit orders always supply it, regardless of order flow (2) order flow: buy orders tend to raise prices, sell to lower it, regardless of order type
• Method: simulation & empirics
• Find: (1) order-type co-move leads to liq comm, BUT not return comm (2) order-flow co-move leads to return comm, but not liq comm.
• Imp: liq co-move poses problem for traditional diversification strats based solely on return interactions.

Question 4

Kamara, Lou and Sadka 2008

Answer 4

• Method: (1) Chordia, Roll, Subrahmanyam (2000) mkt model of liq to est each firm’s sensitivity to changes mkt liq. (2) Amihud (2002) measure of firm’s illiq (3) institutional holdings data
• Find: cross-sec variation of liq comm incrd 1963-2005
• divergence of sys liq can be explained by patterns in institutional ownership
• findings associated with similar patterns in sys risk → ability to diversify sys risk & aggr liq shocks by holding large-cap stocks has declined.
• suggests fragility of US equity mkt to unanticipated events has incrd over time

Question 5

Brunnermeier and Pedersen 2009

Answer 5

• Intuition: trade req cap → funding of traders profoundly affects/affected by mkt liq.
• When funding liq tight → traders become reluctant to take on positions → lowers mkt liq → incr vol.
• Low future mkt liq → incr risk of financing trade → incr margins
• Result: margins destabilizing and mkt & funding liquidity mutually reinforcing
• ⇒ liq spirals: Initial losses → funding problems → reduced positions → prices move away from fundamentals → higher margins, & losses on existing positions → more funding problems → …
• Matches Emp Obs: mkt liq (i) can suddenly dry up (ii) has comm across stocks (iii) relates to vol (iv) subj to “flight to quality” (v) co-moves with mkt

Question 6

Brockman, Chung and Perignon 2009

Answer 6

• Bloomberg data: 1.47 bill txs, 47 exchanges, 38 countries
• Firm-lvl changes in liq influenced by exch-lvl changes across most world’s exchs
• Emerging Asian exchs have exceptionally strong comm, while those of Latin America exhibit little if any
• Find evidence of a distinct, global component in bid-ask spreads & depths.
• Local (exch-level) sources of comm repr roughly 39% of firm’s total comm in liq, while global sources contribute an additional 19%.
• Comm driven by both domestic & U.S. macro news

Question 7

Karolyi, Lee and van Dijk (2012)

Answer 7

• 40 dev & emerg countries 1995-2009.
• Comm is greater in those with weaker investor protections & a more opaque info environment
• Comm greater during (a) times of high mkt vol, (b) large mkt declines, (c) heightened presence of intnl & institutional investors, (d) when investor sentiment pos
• Results more reliably consistent with demand-side explanations for comm, especially for stks in emerg countries.

Question 8

Hameed, Kang, Viswanathan 2010

Answer 8

• Motivation: collateral-based theo models – mkt makers make mkts by absorbing temp liq shocks. Face funding contraints → obtain financing by posting margins & pledging stks they hold as collateral → when big price drop mkt makers hit margin constraints → forced to liquidate when mkt most needs their liq svcs
• Find neg mkt returns decr stk liq, esp during tight funding mkts
• Interindustry spillover effects in liq, which likely to arise from cap constraints in mkt making sector.
• Find economically significant returns to supplying liq following periods of large drops in mkt valuations.

Question 9

Naes, Skjeltorp and Odegaard 2011

Answer 9

• Use long US time-series & granular Norway data
• Intuition as to why liq better predictor of real economy than stocks: extra info contained in order flow data explained by agg order flows bringing together dispersed info from heterogenously informed investors.
• Find a strong relation between stock market liquidity & biz cycle.
• Investors’ pf compositions change with biz cycle & investor participation is related to mkt liquidity → sys liq variation is related to a “flight to quality” during economic downturns. ⇒ provide alt explanation for liq commonality