Table 7.

Difference-in-differences regression for U.K. and U.S. ETFs: Falsification test .

	Quoted spread	Effective spread	Realized spread	Price impact
U.K.	21.65	1.07	0.31	0.77
	(14.36)	(2.79)	(1.76)	(2.56)
Margin increase	−6.48	−1.68	−0.30	−1.64
	(11.16)	(2.17)	(1.37)	(1.99)
U.K. × Margin increase	2.73	5.75${}^{\mathrm{\ast }}$	1.11	4.10
	(17.01)	(3.31)	(2.09)	(3.03)
Volatility	1,353.01${}^{\mathrm{\ast \ast }}$	361.58${}^{\mathrm{\ast \ast \ast }}$	36.67	380.69${}^{\mathrm{\ast \ast \ast }}$
	(581.08)	(112.98)	(71.24)	(103.58)
Return	−722.45	−174.68	−176.56${}^{\mathrm{\ast \ast }}$	−10.41
	(609.68)	(118.54)	(74.75)	(108.68)
Constant	−4.98	−1.20	−0.14	−1.29
	(9.37)	(1.82)	(1.15)	(1.67)
Observations	69	69	69	69
${\overline{\text{R}}}^2$	0.14	0.26	0.04	0.20

The table reports the results to a falsification test in August 2015, when the S&P 500 fell over 10% with no corresponding change to exchange margin requirements. The CSPX.L (U.K.) and SPY (U.S.) are ETFs that track the performance of the S&P 500, traded on the LSE and NYSE, respectively. Specifically, we report the results for the following difference-in-differences regression:

${Liquidity}_{i,t}={\alpha }_0+{\beta }_1{U.K.}_{i,t}+{\beta }_2{Marginincrease}_t+{\beta }_3{U.K.}_{i,t}\times {Marginincrease}_t+{Volatility}_{i,t}+{Return}_{i,t}+{\varepsilon }_{i,t}$,

where ${Liquidity}_{i,t}$ is one of the following liquidity variables: Quoted spread, Effective spread, Realized spread or Price impact for the U.S. or U.K. ETF, $i$. The dependent variables are calculated over one hour intervals, and are expressed in basis points. ${U.K.}_{i,t}$ is an indicator variable equal to 1 if the ETF trades in the U.K., and 0 otherwise. Margin increase is an indicator variable equal to 1 for the period August 19 to August 25, 2015 and 0 for the period August 12 to 18, 2015. $Volatilit{y}_{i,t}$ and $Retur{n}_{i,t}$ are based on the FTSE 100 and S&P 500 index levels for the U.K. and U.S. markets, respectively. $Volatilit{y}_{i,t}$ and $Retur{n}_{i,t}$ are calculated over one-hour intervals as $log({HighPrice}_{i,t}/{LowPrice}_{i,t})$ and $log({ClosePrice}_{i,t}/{OpenPrice}_{i,t})$, respectively. ***, **, and * denote statistical significance at the 1%, 5%, and 10% levels, respectively.