Quantitative easing and the functioning of the gilt repo market

Figures & data

Figure 1. QE transmission channels.

Figure 2. Average daily amounts outstanding of reverse repo in the gilt repo market Source: Bank of England Database: https://www.bankofengland.co.uk/boeapps/database/default.asp. Series codes: YWQZM5L YWQZM5M YWQZM5O YWQZM5P YWQZM5Q YWQZM5R YWQZM5S YWQZM5T YWQZM6E.

Table 1. Descriptive statistics of the dataset.

Variable		Obs.	mean	St. dev	Min	P25	median	P75	Max
Trade-level
Amount	Log of amount borrowed or lent (cash leg of the trade)	4,109,107	16.91378	1.301885	12.27746	16.1159	17.07136	17.78374	22.1519
Spread* (%)	Spread over the reference rate index (to calculate interest rate at issuance)	4,109,107	−0.0169279	0.1354192	−3.1665	−0.0501	−0.008	0.03	182.542
maturity bucket	Maturity of the trade at issuance in years	4,109,107	0.0161702	0.0741871	0.0013699	0.0013699	0.0013699	0.0013699	1
Bank-level
totAssets	Log of total assets	352	25.16366	2.714524	13.91972	23.98957	25.92452	27.22304	28.22435
Rwa	Log of risk weighted assets	352	24.27514	3.021395	9.977574	23.64253	25.24683	25.99111	27.41535
Lem	Log of leverage ratio exposure measure	352	25.65179	2.632744	13.91972	25.10503	26.48531	27.35131	28.47689
cet1Cap	Log of core tier 1 capital	352	22.58127	2.359405	12.4285	22.01448	23.2494	24.106	25.36766
tier1Cap	Log of tier 1 capital	352	22.69106	2.389325	12.4285	22.01448	23.44739	24.21319	25.53545
trdAssets	Log of total trading assets	285	23.59532	3.308831	9.283219	22.72902	24.77246	25.3961	27.82987
securitiesAssets	Log of total securities holdings	349	22.83931	3.071773	9.92564	21.67911	23.60409	25.08148	27.05477
loansBanks	Log of total lending to banks	343	22.10879	3.101762	9.828926	20.0111	22.6904	24.53215	27.1546
totLoans	Log of total loans	346	23.83968	2.959396	11.9591	21.92683	24.96898	26.27623	27.4483
Npl	Non-performing loans/total loans		0.0367954	0.1163406	0	0.0030009	0.0139795	0.0298776	0.0929210
totLiabs	Log of total liabilities	351	25.06824	2.945941	8.994103	23.96959	25.87486	27.18127	28.17011
totDeposits	Log of total deposits	169	24.7924	3.320296	6.907755	24.0883	25.94362	26.96937	27.58602
Amount	Log of total amount lent/borrowed by a bank in a day per maturity bucket, LR nettability and collateral type	195,589	19.42061	1.752594	13.81551	18.30774	19.48539	20.74307	23.98591
Spread (%)	Weighted average spread on lending/borrowing by a bank in a day per maturity bucket, LR nettability and collateral type	195,589	0.0112572	0.1953133	−2.374435	−0.0414	0.003	0.0491144	53.5329

Source: Bank of England Sterling Money Market database, asset purchase facility operations database, and internal regulatory data.

*Spread: the difference between the rate charged and the relevant reference rates. Amounts are in logarithmic scale.

Table 2. Propensity score matching (pre Brexit QE).

	Lending (reverse repo)								Borrowing (repo)
	All maturities		Overnight		2 weeks to 1 month		3 months or more		All maturities		Overnight		2 weeks to 1 month		3 months or more
	(1)	(2)	(1)	(2)	(1)	(2)	(1)	(2)	(1)	(2)	(1)	(2)	(1)	(2)	(1)	(2)
VARIABLES	treat	treat	treat	treat	treat	treat	treat	treat	treat	treat	treat	treat	treat	treat	treat	treat
Total amount	−0.495	−0.057	−14.443	−2.408	−1.343	−1.202	0.050	−0.396	0.182	0.080	0.255	0.062	−0.257	−0.491*	−0.463	−0.688
	(0.303)	(0.200)	(19.972)	(1.973)	(1.089)	(1.657)	(0.190)	(0.411)	(0.179)	(0.180)	(0.174)	(0.169)	(0.231)	(0.286)	(0.368)	(0.488)
w. avg. maturity	−22.642	2.548							−13.379***	−7.885
	(15.940)	(14.380)							(4.725)	(29.879)
pct. Of total portfolio			157.854	22.484	−6.141*	−2.626	−5.263	−0.261			0.258	0.548	3.468	4.968*	13.907	21.223
			(220.803)	(15.317)	(3.236)	(4.350)	(8.013)	(10.664)			(0.169)	(0.337)	(2.431)	(2.661)	(12.193)	(15.022)
Total assets	−0.806***	−0.724*	−21.692	−3.463	−2.014*	−1.920	−0.662***	−0.500	−0.125	−0.153	0.059	−0.080	0.365	0.497*	0.220	0.386
	(0.302)	(0.439)	(30.499)	(1.942)	(1.059)	(1.818)	(0.218)	(0.385)	(0.121)	(0.202)	(0.123)	(0.161)	(0.285)	(0.271)	(0.217)	(0.296)
RWAs	1.365**	2.440**	37.062	6.695	4.322	4.078	1.107**	2.144*
	(0.642)	(1.133)	(51.809)	(3.335)	(2.726)	(3.930)	(0.505)	(1.199)
CET1 ratio	13.628	65.345	77.036	61.118	49.267	48.434	9.183	62.038
	(11.477)	(40.846)	(76.983)	(41.781)	(47.046)	(62.553)	(6.782)	(45.579)
Constant	−2.037	−30.212*	−141.664	−44.790**	−30.965	−32.095	−13.781	−40.927*	−0.726	1.258	−7.590	−0.649	−1.364	0.107	6.425	6.727
	(15.448)	(18.366)	(194.956)	(18.494)	(27.054)	(28.914)	(11.227)	(23.686)	(5.510)	(8.353)	(5.350)	(6.733)	(8.981)	(6.101)	(9.138)	(9.883)
Matching	-pre	-post	-pre	-post	-pre	-post	-pre	-post	-pre	-post	-pre	-post	-pre	-post	-pre	-post
$\mathrm{Adj}{R}^2$	0.498	0.494	0.669	0.598	0.586	0.442	0.430	0.340	0.355	0.126	0.338	0.244	0.242	0.364	0.291	0.432
p-value	0.000	0.101	0.688	0.151	0.002	0.333	0.007	0.489	0.027	0.619	0.046	0.115	0.368	0.233	0.496	0.473
N	17	14	17	13	16	12	16	13	13	12	13	13	12	12	11	11

Probit regressing the treatment on lending/borrowing (reverse repo/repo) portfolio characteristics and bank characteristics. The dependent variable is QE treatment status treat for QE banks. Model (1) reports the pre-matching results while model (2) reports the post matching results with matching ratio 1:2. Coefficients and standard errors are reported for each variable. Standard errors are robust and reported in brackets, *p < 0.10, **p < 0.05, ***p < 0.01.

Table 3. Propensity score matching (pre Covid QE).

	Lending (reverse repo)								Borrowing (repo)
	All maturities		Overnight		2 weeks to 1 month		3 months or more		All maturities		Overnight		2 weeks to 1 month		3 months or more
	(1)	(2)	(1)	(2)	(1)	(2)	(1)	(2)	(1)	(2)	(1)	(2)	(1)	(2)	(1)	(2)
VARIABLES	treat	treat	treat	treat	treat	treat	treat	treat	treat	treat	treat	treat	treat	treat	treat	treat
Total amount	0.050	−0.189	0.059	−0.212	0.001	−0.169	−0.074	0.103	0.080	−1.352***	0.117	−0.029	−0.005	−0.119	−0.024	−0.101
	(0.171)	(0.172)	(0.164)	(0.176)	(0.215)	(0.272)	(0.218)	(0.236)	(0.127)	(0.422)	(0.107)	(0.132)	(0.155)	(0.177)	(0.128)	(0.146)
w. avg. maturity	−23.157**	−9.880							−5.720	413.900
	(11.559)	(16.021)							(8.014)	(172.524)
pct. Of total portfolio			4.647**	1.191	−6.003	1.183	−10.392*	−9.771*			−0.041	−0.009	−1.487	0.334	−4.041	6.796
			(2.289)	(2.786)	(3.834)	(3.274)	(5.786)	(5.341)			(0.074)	(0.063)	(2.681)	(2.337)	(7.919)	(9.241)
Total assets	−3.026***	−2.783*	−3.259***	−2.594*	−3.187**	−2.210**	−2.699***	−1.941**	−2.199***	−1.183	−2.160***	−2.516***	−2.072***	−2.295***	−2.026***	−2.335***
	(0.842)	(1.457)	(1.047)	(1.400)	(1.322)	(0.912)	(0.793)	(0.835)	(0.814)	(0.893)	(0.733)	(0.956)	(0.676)	(0.695)	(0.686)	(0.860)
RWAs	3.370***	2.683*	3.595***	2.487*	3.491**	2.119**	3.028***	1.813**	2.319***	2.244*	2.392***	2.427**	2.343***	2.360***	2.269***	2.339**
	(0.924)	(1.405)	(1.135)	(1.367)	(1.439)	(0.947)	(0.914)	(0.802)	(0.826)	(1.254)	(0.775)	(0.986)	(0.723)	(0.742)	(0.732)	(0.926)
CET1 ratio	10.964	20.929	13.785	22.331	14.644	20.694	10.865	32.142	4.067***	63.367**	4.671***	5.506	6.681	9.260	5.264	5.799
	(7.097)	(18.458)	(9.761)	(18.338)	(11.567)	(16.546)	(7.196)	(24.323)	(1.360)	(25.104)	(1.310)	(8.314)	(6.455)	(6.201)	(5.743)	(5.821)
Constant	−7.754	6.906	−12.456*	6.033	−5.640	4.818	−4.737	−2.157	−2.998	−0.427	−6.937	4.443	−4.820	1.729	−3.630	3.189
	(6.750)	(13.684)	(7.350)	(12.329)	(6.038)	(13.952)	(4.546)	(5.392)	(8.794)	(11.803)	(4.231)	(9.482)	(5.757)	(5.990)	(5.424)	(6.477)
Matching	-pre	-post	-pre	-post	-pre	-post	-pre	-post	-pre	-post	-pre	-post	-pre	-post	-pre	-post
$\mathrm{Adj}{R}^2$	0.550	0.483	0.563	0.475	0.545	0.487	0.530	0.439	0.465	0.675	0.465	0.489	0.445	0.446	0.434	0.468
p-value	0.008	0.257	0.031	0.383	0.109	0.152	0.019	0.206	0.003	0.022	0.008	0.149	0.033	0.026	0.032	0.032
N	33	28	33	28	31	26	31	26	30	26	30	29	29	29	28	28

Probit regressing the treatment on lending/borrowing (reverse repo/repo) portfolio characteristics and bank characteristics. The dependent variable is QE treatment status treat for QE banks. Model (1) reports the pre-matching results while model (2) reports the post matching results with matching ratio 1:2. Coefficients and standard errors are reported for each variable. Standard errors are robust and reported in brackets, *p < 0.10, **p < 0.05, ***p < 0.01.

Table 4. QE treatment effect – bank-level.

Panel (a): reverse repo (lending) amounts and pricing
	Amounts				Spreads
Variables	All 1(a)	Overnight 1(b)	2 weeks to 1 month 1(c)	3 months or more 1(d)	All 1(a)	Overnight 1(b)	2 weeks to 1 month 1(c)	3 months or more 1(d)
$\mathit{Treated}\times \mathit{Q}{\mathit{E}}_{\mathit{Brexit},\mathit{t}}$	−0.149	1.852**	−0.149	−0.528***	−0.0879**	−0.133***	−0.0661***	−0.0310***
	(0.519)	(0.915)	(0.0139)	(0.162)	(0.0428)	(0.00755)	(0.0140)	(0.00723)
$\mathrm{Treated}\times Q{E}_{\mathrm{Brexit},t}\times L{R}_t$	0.253	0.547	−0.0648	0.195	−0.0662***	−0.0927***	−0.0242***	0.0265***
	(1.0797)	(1.115)	(0.129)	(0.253)	(0.0162)	(0.00895)	(0.00261)	(0.00590)
Obs.	33,679	16,428	10,956	5,468	33,679	16,428	10,956	5,007
R-squared	0.117	0.111	0.124	0.111	0.021	0.198	0.012	0.094
$\mathrm{Treated}\times Q{E}_{\mathrm{Covid},t}$	0.336***	0.383***	−0.297	0.376*	0.0120	0.0000384	−0.0210	0.00780
	(0.157)	(0.0817)	(0.256)	(0.203)	(0.0145)	(0.0222)	(0.0237)	(0.0204)
.$\mathrm{Treated}\times Q{E}_{\mathrm{Covid},t}\times L{R}_t$	0.684***	0.975**	−0.0493	0.699***	0.000850	−0.00322	−0.00616	0.0121
	(0.254)	(0.385)	(0.222)	(0.159)	(0.0129)	(0.0246)	(0.0182)	(0.0150)
Obs.	8,978	4,382	2,881	1,654	8,978	4,382	2,881	1,654
R-squared	0.269	0.576	0.260	0.348	0.178	0.663	0.232	0.141
QE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
LR	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Ctrls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Ctrls * QE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Bank FEs	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Coefficient estimates of amounts and spreads on reverse repo (lending) on daily basis (Jan 2016 to Apr 2021). $\mathrm{Treate}{d}_i$ equals to 1 for QE banks and 0 for non-QE banks. Controls include bank-level controls (e.g. total assets, RWAs) and portfolio-level controls (e.g. maturity bucket, collateral type). Robust standard errors reported between parentheses, *p < 0.10, **p < 0.05, ***p < 0.01.
$\mathrm{Treated}\times Q{E}_{\mathrm{Brexit},t}$	0.428***	0.900***	−1.538***	−0.0725	0.0000424	0.00899	−0.0709*	0.0532
	(0.143)	(0.162)	(0.209)	(0.550)	(0.0263)	(0.00760)	(0.0395)	(0.0597)
$\mathrm{Treated}\times Q{E}_{\mathrm{Brexit},t}\times L{R}_t$	−0.170*	−0.158	0.659***	0.808***	−0.0753***	−0.0578***	−0.131***	−0.0781***
	(0.0999)	(0.126)	(0.125)	(0.171)	(0.0222)	(0.00564)	(0.0469)	(0.00698)
Obs.	13,847	24,952	16,623	8,748	49,246	24,952	16,623	8,748
R-squared	0.295	0.233	0.141	0.101	0.017	0.065	0.013	0.068
$\mathrm{Treated}\times Q{E}_{\mathrm{Covid},t}$	0.122	−0.0378	0.329	0.134	0.0373***	0.0409***	0.0186	−0.0109
	(0.117)	(0.172)	(0.225)	(0.220)	(0.00852)	(0.0130)	(0.0130)	(0.0130)
.$\mathrm{Treated}\times Q{E}_{\mathrm{Covid},t}\times L{R}_t$	0.128	0.0914	0.0520	0.125	−0.00200	−0.0422***	0.0116	0.0320***
	(0.100)	(0.174)	(0.186)	(0.212)	(0.00521)	(0.00851)	(0.00746)	(0.00983)
Obs.	19,692	10,126	6,520	3,877	19,692	10,126	6,520	3,877
R-squared	0.380	0.403	0.382	0.490	0.600	0.662	0.609	0.706
QE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
LR	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Ctrls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Ctrls * QE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Bank Fes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes

Coefficient estimates of amounts and spreads on repo (borrowing) on daily basis (Jan 2016 to Apr 2021). $\mathrm{Treate}{d}_i$ equals to 1 for QE banks and 0 for non-QE banks. Controls include bank-level controls (e.g. total assets, RWAs) and portfolio-level controls (e.g. maturity bucket, collateral type). Robust standard errors reported between parentheses, * p < 0.10 ** p < 0.05 *** p < 0.01.

Table 5. QE treatment effect on spreads and amounts of reverse repos – transaction-level.

Panel (a): Spreads
	Overnight			2 weeks to 1 month			3 months or more
Variables	1(a)	1(b)	1(c)	2(a)	2(b)	2(c)	3(a)	3(b)	3(c)
.$\mathrm{Treated}\times Q{E}_{\mathrm{Brexit},t}$	−0.0989***	−0.0868***	−0.0868***	−0.828	−1.375	−1.375	−0.500	−0.0554	−0.0554
	(0.0215)	(0.0213)	(0.0213)	(0.605)	(0.982)	(0.982)	(0.546)	(0.251)	(0.251)
$\mathrm{Treated}\times Q{E}_{\mathrm{Brexit},t}\times L{R}_t$	−0.0399***	−0.0281***	−0.0281***	0.880	1.190	1.190	0.0456	0.0526	0.0526
	(0.00170)	(0.00165)	(0.00165)	(0.772)	(1.896)	(1.896)	(0.0735)	(0.205)	(0.205)
.$\mathrm{Transactionamount}$	0.00132***	0.00169***	0.00169***	−0.00547***	−0.00179	−0.00179	−0.00894***	−0.00274***	−0.00274***
	(0.0000905)	(0.0000795)	(0.0000795)	(0.000216)	(0.00131)	(0.00131)	(0.00191)	(0.000497)	(0.000497)
Obs.	567,711	567,711	567,711	221,523	221,523	221,523	43,591	752	752
R-squared	0.079	0.147	0.147	0.113	0.169	0.169	0.241	0.296	0.296
.$\mathrm{Treated}\times Q{E}_{\mathrm{Covid},t}$	0.00334	0.00426	0.00426	−0.00216	0.00689	0.00689	−0.0714	−0.204***	−0.204***
	(0.00511)	(0.00506)	(0.00506)	(0.00790)	(0.00665)	(0.00665)	(0.0596)	(0.0596)	(0.0596)
.$\mathrm{Treated}\times Q{E}_{\mathrm{Covid},t}\times L{R}_t$	−0.0316***	−0.0328***	−0.0328***	−0.0157***	−0.00216	−0.00216	−0.147***	−0.149***	−0.149***
	(0.00566)	(0.00558)	(0.00558)	(0.00436)	(0.00410)	(0.00410)	(0.0443)	(0.0391)	(0.0391)
.$\mathrm{Transaction}\mathrm{amount}$	0.00223***	0.00225***	0.00225***	−0.00527***	0.00132*	0.00132*	−0.0173***	−0.00555	−0.00555
	(0.000736)	(0.000738)	(0.000738)	(0.000237)	(0.000759)	(0.000759)	(0.00584)	(0.00611)	(0.00611)
Obs.	43,591	43,591	43,591	221,523	221,523	221,523	43,591	752	752
R-squared	0.192	0.194	0.194	0.006	0.013	0.013	0.192	0.254	0.254
QE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
LR	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Ctrls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Ctrls * QE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Bank Fes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
CP sec. Fes	No	Yes	Yes	No	Yes	Yes	No	Yes	Yes
LR nettability	No	Yes	No	No	Yes	No	No	Yes	No
Coefficient estimates of spreads on reverse repo transactions on daily basis (Jan 2016 to Apr 2021). Model specifications: (a) bank fixed-effect only; (b) bank, counterparty sector and LR nettability fixed-effects; and (c) bank and counterparty sector fixed-effects and LR nettability as a control. Treatment status$\mathrm{Treate}{d}_i$ equals to 1 for QE-banks and 0 for non-QE-banks. Controls include collateral type and central clearing status. Robust standard errors reported between parentheses, *p < 0.10, **p < 0.05, ***p < 0.01.
.$\mathrm{Treated}\times Q{E}_{\mathrm{Brexit},t}$	−0.587	−1.849***	−1.849***	−7.777**	−3.308	−3.308	0.386	0.392	0.3858684
	(0.466)	(0.461)	(0.461)	(3.500)	(4.791)	(4.791)	(0.751)	(0.700)	(0.708254)
.$\mathrm{Treated}\times Q{E}_{\mathrm{Brexit},t}\times L{R}_t$	−0.639***	−0.927***	−0.927***	−0.00913	0.294	0.294	1.434	−1.012	−0.9687634
	(0.0656)	(0.0645)	(0.0645)	(0.362)	(0.350)	(0.350)	(1.351)	(1.293)	(1.321237)
Obs.	567,711	567,711	567,711	57,536	57,536	57,536	34,168	34,168	34,168
R-squared	0.069	0.118	0.118	0.076	0.205	0.205	0.043	0.141	0.141
.$\mathrm{Treated}\times Q{E}_{\mathrm{Covid},t}$	0.303***	0.305***	0.305***	1.223***	1.074***	1.074***	0.109	0.127	0.127
	(0.0565)	(0.0564)	(0.0564)	(0.0941)	(0.0918)	(0.0918)	(0.807)	(0.602)	(0.602)
.$\mathrm{Treated}\times Q{E}_{\mathrm{Covid},t}\times L{R}_t$	0.0674	0.0566	0.0566	−0.243***	−0.295***	−0.295***	0.109	0.203	0.203
	(0.0614)	(0.0614)	(0.0614)	(0.0378)	(0.0346)	(0.0346)	(0.455)	(0.360)	(0.360)
Obs.	43,591	43,591	43,591	221,523	221,523	221,523	752	752	752
R-squared	0.143	0.145	0.145	0.093	0.247	0.247	0.276	0.432	0.432
QE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
LR	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Ctrls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Ctrls * QE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Bank FEs	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
CP sec. FEs	No	Yes	Yes	No	Yes	Yes	No	Yes	Yes
LR nettability	No	Yes	No	No	Yes	No	No	Yes	No

Coefficient estimates of amounts of reverse repo transactions on daily basis (Jan 2016 to Apr 2021). Model specifications: (a) bank fixed-effect only; (b) bank, counterparty sector and LR nettability fixed-effects; and (c) bank and counterparty sector fixed-effects and LR nettability as a control. Treatment status$\mathrm{Treate}{d}_i$ equals to 1 for QE-banks and 0 for non-QE-banks. Controls include collateral type and central clearing status. Robust standard errors reported between parentheses, *p < 0.10, **p < 0.05, ***p < 0.01.

Table 6. QE treatment effect on spreads and amounts of repos (borrowing) – transaction-level.

Panel (a): Spreads
	Overnight			2 weeks to 1 month			3 months or more
Variables	1(a)	1(b)	1(c)	2(a)	2(b)	2(c)	3(a)	3(b)	3(c)
.$\mathrm{Treated}\times Q{E}_{\mathrm{Brexit},t}$	0.0143***	0.0156***	0.0156***	−0.0751***	−0.0732***	−0.0732***	−0.150***	−0.146***	−0.146***
	(0.00242)	(0.00245)	(0.00245)	(0.0186)	(0.0184)	(0.0184)	(0.0571)	(0.0435)	(0.0435)
$\mathrm{Treated}\times Q{E}_{\mathrm{Brexit},t}\times L{R}_t$	−0.0338***	−0.0385***	−0.0385***	−0.0920***	−0.0829***	−0.0829***	−0.120***	−0.0715***	−0.0715***
	(0.00144)	(0.00138)	(0.00138)	(0.00331)	(0.00328)	(0.00328)	(0.0189)	(0.0161)	(0.0161)
.$\mathrm{Transactionamount}$	0.00181***	0.00224***	0.00224***	−0.000891***	−0.00581***	−0.00581***	0.0108***	−0.0000914	−0.0000914
	(0.0000971)	(0.0000999)	(0.0000999)	(0.000241)	(0.000255)	(0.000255)	(0.000826)	(0.000835)	(0.000835)
Obs.	1,174,064	1,174,064	1,174,064	168,563	168,563	168,563	11,897	11,897	11,897
R-squared	0.059	0.068	0.068	0.053	0.110	0.110	0.157	0.298	0.298
..$\mathrm{Treated}\times Q{E}_{\mathrm{Covid},t}$	−0.258***	0.00443	0.00443	−0.117***	−0.0220	−0.0220	−0.474***	0.0518***	0.0518***
	(0.0190)	(0.00273)	(0.00273)	(0.0421)	(0.0157)	(0.0157)	(0.0490)	(0.0112)	(0.0112)
.$\mathrm{Treated}\times Q{E}_{\mathrm{Covid},t}\times L{R}_t$	0.00672**	−0.00746***	−0.00746***	−0.0198	−0.0134*	−0.0134*	0.0496***	−0.0369***	−0.0369***
	(0.00275)	(0.00197)	(0.00197)	(0.0155)	(0.00736)	(0.00736)	(0.0120)	(0.00783)	(0.00783)
.$\mathrm{Transactionamount}$	0.00506***	0.00448***	0.00448***	0.00161***	−0.000531	−0.000531	0.0113***	0.000468	0.000468
	(0.000198)	(0.000207)	(0.000207)	(0.000518)	(0.000618)	(0.000618)	(0.000666)	(0.000716)	(0.000716)
Obs.	130,763	130,763	130,763	14,332	14,332	14,332	19,337	19,337	19,337
R-squared	0.153	0.167	0.167	0.198	0.219	0.219	0.164	0.273	0.273
QE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
LR	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Ctrls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Ctrls * QE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Bank FEs	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
CP sec. FEs	No	Yes	Yes	No	Yes	Yes	No	Yes	Yes
LR nettability	No	Yes	No	No	Yes	No	No	Yes	No
Coefficient estimates of spreads on repo transactions on daily basis (Jan 2016 to Apr 2021). Model specifications: (a) bank fixed-effect only; (b) bank, counterparty sector and LR nettability fixed-effects; and (c) bank and counterparty sector fixed-effects and LR nettability as a control. Treatment status$\mathrm{Treate}{d}_i$ equals to 1 for QE-banks and 0 for non-QE-banks. Controls include collateral type and central clearing status. Robust standard errors reported between parentheses, *p < 0.10, **p < 0.05, ***p < 0.01.
.$\mathrm{Treated}\times Q{E}_{\mathrm{Brexit},t}$	0.0161	0.112***	0.112***	−0.506**	−0.344*	−0.344*	1.974***	1.225***	1.225***
	(0.0318)	(0.0316)	(0.0316)	(0.221)	(0.199)	(0.199)	(0.350)	(0.360)	(0.360)
.$\mathrm{Treated}\times Q{E}_{\mathrm{Brexit},t}\times L{R}_t$	0.0588***	0.121***	0.121***	0.0316	−0.0245	−0.0245	−0.743***	−0.538**	−0.538**
	(0.0189)	(0.0180)	(0.0180)	(0.0540)	(0.0489)	(0.0489)	(0.276)	(0.223)	(0.223)
Obs.	1,174,064	1,174,064	1,174,064	168,563	168,563	168,563	11,897	11,897	11,897
R-squared	0.110	0.149	0.149	0.073	0.205	0.205	0.162	0.345	0.345
.$\mathrm{Treated}\times Q{E}_{\mathrm{Covid},t}$	0.00474	−0.136***	−0.136***	0.719***	0.293***	0.293***	0.683***	0.341**	0.341**
	(0.0346)	(0.0352)	(0.0352)	(0.105)	(0.0955)	(0.0955)	(0.171)	(0.138)	(0.138)
.$\mathrm{Treated}\times Q{E}_{\mathrm{Covid},t}\times L{R}_t$	0.416***	0.274***	0.274***	−0.309***	−0.0924	−0.0924	0.340***	−0.189**	−0.189**
	(0.0346)	(0.0332)	(0.0332)	(0.105)	(0.0977)	(0.0977)	(0.113)	(0.0963)	(0.0963)
Obs.	130,763	130,763	130,763	14,332	14,332	14,332	19,337	19,337	19,337
R-squared	0.169	0.262	0.262	0.233	0.377	0.377	0.175	0.420	0.420
QE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
LR	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Ctrls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Ctrls * QE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Bank FEs	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
CP sec. FEs	No	Yes	Yes	No	Yes	Yes	No	Yes	Yes
LR nettability	No	Yes	No	No	Yes	No	No	Yes	No

Coefficient estimates of amounts of repo transactions on daily basis (Jan 2016 to Apr 2021). Model specifications: (a) bank fixed-effect only; (b) bank, counterparty sector and LR nettability fixed-effects; and (c) bank and counterparty sector fixed-effects and LR nettability as a control. Treatment status$\mathrm{Treate}{d}_i$ equals to 1 for QE-banks and 0 for non-QE-banks. Controls include collateral type and central clearing status. Robust standard errors reported between parentheses, *p < 0.10, **p < 0.05, ***p < 0.01.

Table 7. Synthetic QE treatment effect – reverse repo (lending) amounts and pricing.

	Amounts				Spreads
Variables	All 1(a)	Overnight 1(b)	2 weeks to 1 month 1(c)	3 months or more 1(d)	All 1(a)	Overnight 1(b)	2 weeks to 1 month 1(c)	3 months or more 1(d)
$\mathrm{Treated}\times Q{E}_{\mathrm{Synth},t}$	−0.785	−0.901**	0.285	−0.876	0.0150	−0.00515	0.129**	−0.0178
	(0.510)	(0.425)	(0.694)	(0.713)	(0.0325)	(0.0186)	(0.0560)	(0.0472)
Obs.	5,562	2,942	1,546	1,070	5,562	2,942	1,546	1,070
R-squared	0.335	0.605	0.316	0.394	0.257	0.589	0.337	0.193
QE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
LR	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Ctrls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Ctrls * QE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Bank FEs	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes

Coefficient estimates of amounts and spreads on reverse repo on daily basis (Feb 2017 to Jan 2020, treatment on 1 Aug 2018). $\mathrm{Treate}{d}_i$ equals to 1 for QE banks and 0 for non-QE banks. Controls include bank-level controls and portfolio-level controls. Robust standard errors reported between parentheses, *p < 0.10, **p < 0.05, ***p < 0.01.

Table 8. Alternative QE treatment effect – reverse repo (lending) amounts and pricing.

	Amounts				Spreads
Variables	All 1(a)	Overnight 1(b)	2 weeks to 1 month 1(c)	3 months or more 1(d)	All 1(a)	Overnight 1(b)	2 weeks to 1 month 1(c)	3 months or more 1(d)
$\mathrm{Treated}\times Q{E}_{\mathrm{Synth},t}$	1.802	137.4	2.577	−5.881***	−0.174	30.52	0.0251	0.250**
	(3.365)	(392.916)	(2.778)	(1.672)	(0.271)	(28.04)	(0.188)	(0.107)
Obs.	779	240	1,366	3,259	779	240	1,366	3,259
R-squared	0.106	0.377	0.293	0.186	0.153	0.283	0.187	0.132
QE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
LR	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Ctrls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Ctrls * QE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Bank Fes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes

Coefficient estimates of amounts and spreads on reverse repo on daily basis (Jan 2016 to Apr 2021). $\mathrm{Treate}{d}_i$ equals to 1 for alternative treatment banks and 0 otherwise. Controls include bank-level controls and portfolio-level controls. Robust standard errors reported between parentheses, *p < 0.10, **p < 0.05, ***p < 0.01.