Setup and Load
Packages
library(tidyverse)
library(lubridate)
library(kableExtra)
library(pander)
# Set pander options for clean text tables
panderOptions('table.split.table', Inf) # Don't split wide tables
panderOptions('table.style', 'simple') # Simple style like Stata output
Directory Setup
Helper Functions
Period Definitions
# ==============================================================================
# PERIOD DEFINITIONS
# ==============================================================================
periods <- tribble(
~period_num, ~period_name, ~start_date, ~end_date, ~description,
0, "Pre-BTFP", "2023-03-01", "2023-03-10", "DW only available",
1, "Acute", "2023-03-13", "2023-05-01", "Peak crisis phase",
2, "Post-Acute", "2023-05-02", "2023-10-31", "Stabilization",
3, "Arbitrage", "2023-11-01", "2024-01-24", "BTFP rate < IORB",
4, "Wind-down", "2024-01-25", "2024-03-11", "BTFP closing announced"
) %>%
mutate(across(c(start_date, end_date), as.Date))
# Key dates
BTFP_LAUNCH <- as.Date("2023-03-12")
BTFP_CLOSE <- as.Date("2024-03-11")
DW_DATA_END <- as.Date("2023-12-31")
BASELINE_DATE <- "2022Q4"
# Display periods
periods %>%
kable(col.names = c("Period", "Name", "Start", "End", "Description"),
caption = "Analysis Period Definitions") %>%
kable_styling(bootstrap_options = c("striped", "hover"), full_width = FALSE)
Analysis Period Definitions
|
Period
|
Name
|
Start
|
End
|
Description
|
|
0
|
Pre-BTFP
|
2023-03-01
|
2023-03-10
|
DW only available
|
|
1
|
Acute
|
2023-03-13
|
2023-05-01
|
Peak crisis phase
|
|
2
|
Post-Acute
|
2023-05-02
|
2023-10-31
|
Stabilization
|
|
3
|
Arbitrage
|
2023-11-01
|
2024-01-24
|
BTFP rate < IORB
|
|
4
|
Wind-down
|
2024-01-25
|
2024-03-11
|
BTFP closing announced
|
Load Data
Define Exclusions
Prepare Baseline Data
(2022Q4)
Identify FHLB Borrowers
(Q1 2023)
Table Functions
Table 1: March 10,
2023 - DW Borrowers vs Non-Borrowers
# ==============================================================================
# TABLE 1: March 10, 2023 Only - DW Borrowers vs Non-Borrowers
# ==============================================================================
dw_mar10 <- dw_loans_raw %>%
filter(dw_loan_date == as.Date("2023-03-10")) %>%
distinct(rssd_id) %>%
pull(rssd_id)
table1_data <- baseline_q %>%
mutate(
borrower_type = if_else(idrssd %in% dw_mar10, "DW Borrower", "Non-Borrower")
)
table1 <- create_summary_table_v2(table1_data, "borrower_type",
group_levels = c("DW Borrower", "Non-Borrower"))
# Display
table1 %>%
select(-Row_Type) %>%
kable(caption = "Table 1: Pre-Crisis Bank Characteristics - DW Borrowers vs Non-Borrowers (March 10, 2023)") %>%
kable_styling(bootstrap_options = c("striped", "hover"), full_width = FALSE)
Table 1: Pre-Crisis Bank Characteristics - DW Borrowers vs Non-Borrowers
(March 10, 2023)
|
Variable
|
DW Borrower
|
Non-Borrower
|
|
N
|
49
|
4647
|
|
Book Equity Ratio
|
8.17 (8.45)
|
10.90 (9.01)
|
|
|
49
|
4647
|
|
MTM Loss (HTM)
|
0.22 (0.00)
|
0.20 (0.00)
|
|
|
49
|
4629
|
|
MTM Loss (Total)
|
5.70 (5.66)
|
5.47 (5.26)
|
|
|
49
|
4629
|
|
Adjusted Equity
|
2.48 (2.44)
|
5.26 (3.82)
|
|
|
49
|
4629
|
|
Uninsured Leverage
|
418.36 (352.70)
|
330.72 (249.57)
|
|
|
49
|
4647
|
|
MTM Insolvent
|
24.49 (0.00)
|
18.41 (0.00)
|
|
|
12
|
852
|
|
Run Risk (Median Cutoff)
|
38.78 (0.00)
|
23.26 (0.00)
|
|
|
19
|
1079
|
|
Run Risk (Q75 Cutoff)
|
8.16 (0.00)
|
4.18 (0.00)
|
|
|
4
|
194
|
|
IDCR Insolvent (s=0.5)
|
2.04 (0.00)
|
4.41 (0.00)
|
|
|
1
|
202
|
|
IDCR Insolvent (s=1.0)
|
32.65 (0.00)
|
27.99 (0.00)
|
|
|
16
|
1283
|
|
Capital Insolvent (s=0.5)
|
4.08 (0.00)
|
2.31 (0.00)
|
|
|
2
|
107
|
|
Capital Insolvent (s=1.0)
|
16.33 (0.00)
|
7.63 (0.00)
|
|
|
8
|
353
|
# Export (silently)
invisible(export_table(
table1,
"table1_dw_mar10",
caption = "Pre-Crisis Bank Characteristics: DW Borrowers vs Non-Borrowers (March 10, 2023)",
label = "tab:dw_mar10",
note = "Bank characteristics from 2022Q4 Call Reports. Values shown as Mean (Median) with N observations below. Continuous variables in percentage points. Dummy variables show proportion (\\%) with value = 1. DW borrowers identified as banks with Discount Window loans on March 10, 2023 only. Sample excludes G-SIBs and failed banks."
))
LaTeX table saved: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/borrowing_stat/tables/table1_dw_mar10.tex
HTML table saved: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/borrowing_stat/tables/table1_dw_mar10.html
CSV table saved: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/borrowing_stat/tables/table1_dw_mar10.csv
Table 2: March 10-13,
2023 - DW Borrowers vs Non-Borrowers
# ==============================================================================
# TABLE 2: March 10-13, 2023 - DW Borrowers vs Non-Borrowers
# ==============================================================================
dw_mar10_13 <- dw_loans_raw %>%
filter(dw_loan_date >= as.Date("2023-03-10") & dw_loan_date <= as.Date("2023-03-13")) %>%
distinct(rssd_id) %>%
pull(rssd_id)
table2_data <- baseline_q %>%
mutate(
borrower_type = if_else(idrssd %in% dw_mar10_13, "DW Borrower", "Non-Borrower")
)
table2 <- create_summary_table_v2(table2_data, "borrower_type",
group_levels = c("DW Borrower", "Non-Borrower"))
# Display
table2 %>%
select(-Row_Type) %>%
kable(caption = "Table 2: Pre-Crisis Bank Characteristics - DW Borrowers vs Non-Borrowers (March 10-13, 2023)") %>%
kable_styling(bootstrap_options = c("striped", "hover"), full_width = FALSE)
Table 2: Pre-Crisis Bank Characteristics - DW Borrowers vs Non-Borrowers
(March 10-13, 2023)
|
Variable
|
DW Borrower
|
Non-Borrower
|
|
N
|
93
|
4603
|
|
Book Equity Ratio
|
8.65 (8.46)
|
10.91 (9.01)
|
|
|
93
|
4603
|
|
MTM Loss (HTM)
|
0.24 (0.00)
|
0.20 (0.00)
|
|
|
93
|
4585
|
|
MTM Loss (Total)
|
5.50 (5.59)
|
5.47 (5.26)
|
|
|
93
|
4585
|
|
Adjusted Equity
|
3.15 (2.72)
|
5.28 (3.83)
|
|
|
93
|
4585
|
|
Uninsured Leverage
|
435.81 (382.71)
|
329.53 (247.81)
|
|
|
93
|
4603
|
|
MTM Insolvent
|
20.43 (0.00)
|
18.43 (0.00)
|
|
|
19
|
845
|
|
Run Risk (Median Cutoff)
|
43.01 (0.00)
|
23.03 (0.00)
|
|
|
40
|
1058
|
|
Run Risk (Q75 Cutoff)
|
11.83 (0.00)
|
4.07 (0.00)
|
|
|
11
|
187
|
|
IDCR Insolvent (s=0.5)
|
5.38 (0.00)
|
4.36 (0.00)
|
|
|
5
|
198
|
|
IDCR Insolvent (s=1.0)
|
33.33 (0.00)
|
27.94 (0.00)
|
|
|
31
|
1268
|
|
Capital Insolvent (s=0.5)
|
5.38 (0.00)
|
2.27 (0.00)
|
|
|
5
|
104
|
|
Capital Insolvent (s=1.0)
|
16.13 (0.00)
|
7.55 (0.00)
|
|
|
15
|
346
|
# Export (silently)
invisible(export_table(
table2,
"table2_dw_mar10_13",
caption = "Pre-Crisis Bank Characteristics: DW Borrowers vs Non-Borrowers (March 10--13, 2023)",
label = "tab:dw_mar10_13",
note = "Bank characteristics from 2022Q4 Call Reports. Values shown as Mean (Median) with N observations below. DW borrowers identified as banks with Discount Window loans between March 10--13, 2023. Sample excludes G-SIBs and failed banks."
))
LaTeX table saved: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/borrowing_stat/tables/table2_dw_mar10_13.tex
HTML table saved: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/borrowing_stat/tables/table2_dw_mar10_13.html
CSV table saved: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/borrowing_stat/tables/table2_dw_mar10_13.csv
Table 3: March 9-14,
2023 - DW Borrowers vs Non-Borrowers
# ==============================================================================
# TABLE 3: March 9-14, 2023 - DW Borrowers vs Non-Borrowers
# ==============================================================================
dw_mar9_14 <- dw_loans_raw %>%
filter(dw_loan_date >= as.Date("2023-03-09") & dw_loan_date <= as.Date("2023-03-14")) %>%
distinct(rssd_id) %>%
pull(rssd_id)
table3_data <- baseline_q %>%
mutate(
borrower_type = if_else(idrssd %in% dw_mar9_14, "DW Borrower", "Non-Borrower")
)
table3 <- create_summary_table_v2(table3_data, "borrower_type",
group_levels = c("DW Borrower", "Non-Borrower"))
# Display
table3 %>%
select(-Row_Type) %>%
kable(caption = "Table 3: Pre-Crisis Bank Characteristics - DW Borrowers vs Non-Borrowers (March 9-14, 2023)") %>%
kable_styling(bootstrap_options = c("striped", "hover"), full_width = FALSE)
Table 3: Pre-Crisis Bank Characteristics - DW Borrowers vs Non-Borrowers
(March 9-14, 2023)
|
Variable
|
DW Borrower
|
Non-Borrower
|
|
N
|
129
|
4567
|
|
Book Equity Ratio
|
9.03 (8.68)
|
10.92 (9.01)
|
|
|
129
|
4567
|
|
MTM Loss (HTM)
|
0.19 (0.00)
|
0.20 (0.00)
|
|
|
129
|
4549
|
|
MTM Loss (Total)
|
5.55 (5.59)
|
5.47 (5.26)
|
|
|
129
|
4549
|
|
Adjusted Equity
|
3.48 (2.86)
|
5.29 (3.83)
|
|
|
129
|
4549
|
|
Uninsured Leverage
|
411.08 (352.70)
|
329.39 (247.44)
|
|
|
129
|
4567
|
|
MTM Insolvent
|
20.93 (0.00)
|
18.40 (0.00)
|
|
|
27
|
837
|
|
Run Risk (Median Cutoff)
|
40.31 (0.00)
|
22.94 (0.00)
|
|
|
52
|
1046
|
|
Run Risk (Q75 Cutoff)
|
10.85 (0.00)
|
4.04 (0.00)
|
|
|
14
|
184
|
|
IDCR Insolvent (s=0.5)
|
5.43 (0.00)
|
4.35 (0.00)
|
|
|
7
|
196
|
|
IDCR Insolvent (s=1.0)
|
34.88 (0.00)
|
27.85 (0.00)
|
|
|
45
|
1254
|
|
Capital Insolvent (s=0.5)
|
4.65 (0.00)
|
2.26 (0.00)
|
|
|
6
|
103
|
|
Capital Insolvent (s=1.0)
|
15.50 (0.00)
|
7.50 (0.00)
|
|
|
20
|
341
|
# Export (silently)
invisible(export_table(
table3,
"table3_dw_mar9_14",
caption = "Pre-Crisis Bank Characteristics: DW Borrowers vs Non-Borrowers (March 9--14, 2023)",
label = "tab:dw_mar9_14",
note = "Bank characteristics from 2022Q4 Call Reports. Values shown as Mean (Median) with N observations below. DW borrowers identified as banks with Discount Window loans between March 9--14, 2023. Sample excludes G-SIBs and failed banks."
))
LaTeX table saved: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/borrowing_stat/tables/table3_dw_mar9_14.tex
HTML table saved: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/borrowing_stat/tables/table3_dw_mar9_14.html
CSV table saved: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/borrowing_stat/tables/table3_dw_mar9_14.csv
Table 4: Acute Period
- BTFP, DW, Both, Non-Borrowers
# ==============================================================================
# TABLE 4: Acute Period (March 13 - May 1, 2023)
# ==============================================================================
acute_start <- as.Date("2023-03-13")
acute_end <- as.Date("2023-05-01")
btfp_acute <- btfp_loans_raw %>%
filter(btfp_loan_date >= acute_start & btfp_loan_date <= acute_end) %>%
distinct(rssd_id) %>%
pull(rssd_id)
dw_acute <- dw_loans_raw %>%
filter(dw_loan_date >= acute_start & dw_loan_date <= acute_end) %>%
distinct(rssd_id) %>%
pull(rssd_id)
table4_data <- baseline_q %>%
mutate(
btfp_user = idrssd %in% btfp_acute,
dw_user = idrssd %in% dw_acute,
borrower_type = case_when(
btfp_user & dw_user ~ "Both",
btfp_user & !dw_user ~ "BTFP Only",
!btfp_user & dw_user ~ "DW Only",
TRUE ~ "Non-Borrower"
)
)
table4 <- create_summary_table_v2(table4_data, "borrower_type",
group_levels = c("BTFP Only", "DW Only", "Both", "Non-Borrower"))
# Display
table4 %>%
select(-Row_Type) %>%
kable(caption = "Table 4: Pre-Crisis Bank Characteristics by Borrower Type (Acute Period: March 13 - May 1, 2023)") %>%
kable_styling(bootstrap_options = c("striped", "hover"), full_width = FALSE)
Table 4: Pre-Crisis Bank Characteristics by Borrower Type (Acute Period:
March 13 - May 1, 2023)
|
Variable
|
BTFP Only
|
DW Only
|
Both
|
Non-Borrower
|
|
N
|
382
|
316
|
96
|
3902
|
|
Book Equity Ratio
|
8.19 (8.06)
|
9.15 (8.90)
|
8.09 (8.17)
|
11.34 (9.14)
|
|
|
382
|
316
|
96
|
3902
|
|
MTM Loss (HTM)
|
0.23 (0.00)
|
0.22 (0.00)
|
0.32 (0.00)
|
0.20 (0.00)
|
|
|
382
|
316
|
96
|
3884
|
|
MTM Loss (Total)
|
6.18 (6.13)
|
5.69 (5.64)
|
5.98 (5.79)
|
5.37 (5.14)
|
|
|
382
|
316
|
96
|
3884
|
|
Adjusted Equity
|
2.01 (2.03)
|
3.47 (3.06)
|
2.11 (2.61)
|
5.77 (4.10)
|
|
|
382
|
316
|
96
|
3884
|
|
Uninsured Leverage
|
396.32 (323.88)
|
383.68 (293.78)
|
575.26 (372.85)
|
315.09 (235.29)
|
|
|
382
|
316
|
96
|
3902
|
|
MTM Insolvent
|
30.10 (0.00)
|
18.67 (0.00)
|
22.92 (0.00)
|
17.20 (0.00)
|
|
|
115
|
59
|
22
|
668
|
|
Run Risk (Median Cutoff)
|
39.79 (0.00)
|
33.86 (0.00)
|
48.96 (0.00)
|
20.34 (0.00)
|
|
|
152
|
107
|
47
|
792
|
|
Run Risk (Q75 Cutoff)
|
7.85 (0.00)
|
6.33 (0.00)
|
12.50 (0.00)
|
3.49 (0.00)
|
|
|
30
|
20
|
12
|
136
|
|
IDCR Insolvent (s=0.5)
|
3.40 (0.00)
|
7.28 (0.00)
|
3.12 (0.00)
|
4.27 (0.00)
|
|
|
13
|
23
|
3
|
164
|
|
IDCR Insolvent (s=1.0)
|
28.27 (0.00)
|
32.91 (0.00)
|
25.00 (0.00)
|
27.70 (0.00)
|
|
|
108
|
104
|
24
|
1063
|
|
Capital Insolvent (s=0.5)
|
1.57 (0.00)
|
4.11 (0.00)
|
7.29 (0.00)
|
2.14 (0.00)
|
|
|
6
|
13
|
7
|
83
|
|
Capital Insolvent (s=1.0)
|
9.69 (0.00)
|
12.03 (0.00)
|
12.50 (0.00)
|
7.05 (0.00)
|
|
|
37
|
38
|
12
|
274
|
# Export (silently)
invisible(export_table(
table4,
"table4_acute_period",
caption = "Pre-Crisis Bank Characteristics by Borrower Type: Acute Crisis Period",
label = "tab:acute_period",
note = "Bank characteristics from 2022Q4 Call Reports. Acute crisis period: March 13 -- May 1, 2023. Banks classified by facility usage during this period. Sample excludes G-SIBs and failed banks."
))
LaTeX table saved: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/borrowing_stat/tables/table4_acute_period.tex
HTML table saved: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/borrowing_stat/tables/table4_acute_period.html
CSV table saved: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/borrowing_stat/tables/table4_acute_period.csv
Table 5: Q1 2023 - All
Facilities
# ==============================================================================
# TABLE 5: Q1 2023 (Jan 1 - Mar 31) - BTFP, DW, FHLB, Combinations
# ==============================================================================
q1_start <- as.Date("2023-01-01")
q1_end <- as.Date("2023-03-31")
btfp_q1 <- btfp_loans_raw %>%
filter(btfp_loan_date >= q1_start & btfp_loan_date <= q1_end) %>%
distinct(rssd_id) %>%
pull(rssd_id)
dw_q1 <- dw_loans_raw %>%
filter(dw_loan_date >= q1_start & dw_loan_date <= q1_end) %>%
distinct(rssd_id) %>%
pull(rssd_id)
table5_data <- baseline_q %>%
mutate(
btfp_user = idrssd %in% btfp_q1,
dw_user = idrssd %in% dw_q1,
fhlb_user = idrssd %in% fhlb_borrowers,
borrower_type = case_when(
btfp_user & dw_user & fhlb_user ~ "All Three",
btfp_user & dw_user & !fhlb_user ~ "BTFP+DW",
btfp_user & !dw_user & fhlb_user ~ "BTFP+FHLB",
!btfp_user & dw_user & fhlb_user ~ "DW+FHLB",
btfp_user & !dw_user & !fhlb_user ~ "BTFP Only",
!btfp_user & dw_user & !fhlb_user ~ "DW Only",
!btfp_user & !dw_user & fhlb_user ~ "FHLB Only",
TRUE ~ "Non-Borrower"
)
)
table5 <- create_summary_table_v2(table5_data, "borrower_type",
group_levels = c("BTFP Only", "DW Only", "FHLB Only",
"BTFP+DW", "BTFP+FHLB", "DW+FHLB",
"All Three", "Non-Borrower"))
# Display
table5 %>%
select(-Row_Type) %>%
kable(caption = "Table 5: Pre-Crisis Bank Characteristics by Borrower Type (Q1 2023: Jan 1 - Mar 31)") %>%
kable_styling(bootstrap_options = c("striped", "hover"), full_width = FALSE) %>%
scroll_box(width = "100%")
Table 5: Pre-Crisis Bank Characteristics by Borrower Type (Q1 2023: Jan
1 - Mar 31)
|
Variable
|
BTFP Only
|
DW Only
|
FHLB Only
|
BTFP+DW
|
BTFP+FHLB
|
DW+FHLB
|
All Three
|
Non-Borrower
|
|
N
|
167
|
286
|
772
|
45
|
51
|
106
|
27
|
3242
|
|
Book Equity Ratio
|
8.13 (7.92)
|
9.27 (8.77)
|
9.80 (9.32)
|
7.37 (7.39)
|
8.90 (8.67)
|
8.99 (8.83)
|
9.13 (8.88)
|
11.56 (9.02)
|
|
|
167
|
286
|
772
|
45
|
51
|
106
|
27
|
3242
|
|
MTM Loss (HTM)
|
0.22 (0.00)
|
0.27 (0.00)
|
0.17 (0.00)
|
0.32 (0.00)
|
0.14 (0.00)
|
0.21 (0.00)
|
0.34 (0.00)
|
0.20 (0.00)
|
|
|
167
|
286
|
772
|
45
|
51
|
106
|
27
|
3224
|
|
MTM Loss (Total)
|
6.07 (6.03)
|
5.47 (5.40)
|
5.96 (5.70)
|
6.18 (5.83)
|
6.19 (6.17)
|
5.85 (5.96)
|
5.98 (5.68)
|
5.28 (5.08)
|
|
|
167
|
286
|
772
|
45
|
51
|
106
|
27
|
3224
|
|
Adjusted Equity
|
2.06 (2.05)
|
3.80 (3.06)
|
3.84 (3.66)
|
1.19 (1.86)
|
2.71 (2.90)
|
3.14 (3.05)
|
3.15 (3.04)
|
6.05 (4.08)
|
|
|
167
|
286
|
772
|
45
|
51
|
106
|
27
|
3224
|
|
Uninsured Leverage
|
379.88 (330.91)
|
424.90 (290.72)
|
332.00 (236.69)
|
614.35 (388.70)
|
364.65 (330.89)
|
341.05 (316.04)
|
454.96 (316.34)
|
315.06 (238.34)
|
|
|
167
|
286
|
772
|
45
|
51
|
106
|
27
|
3242
|
|
MTM Insolvent
|
31.14 (0.00)
|
18.18 (0.00)
|
18.01 (0.00)
|
31.11 (0.00)
|
21.57 (0.00)
|
18.87 (0.00)
|
29.63 (0.00)
|
17.62 (0.00)
|
|
|
52
|
52
|
139
|
14
|
11
|
20
|
8
|
568
|
|
Run Risk (Median Cutoff)
|
38.92 (0.00)
|
29.37 (0.00)
|
24.87 (0.00)
|
55.56 (100.00)
|
49.02 (0.00)
|
33.96 (0.00)
|
40.74 (0.00)
|
20.41 (0.00)
|
|
|
65
|
84
|
192
|
25
|
25
|
36
|
11
|
660
|
|
Run Risk (Q75 Cutoff)
|
9.58 (0.00)
|
8.04 (0.00)
|
4.27 (0.00)
|
6.67 (0.00)
|
5.88 (0.00)
|
5.66 (0.00)
|
11.11 (0.00)
|
3.43 (0.00)
|
|
|
16
|
23
|
33
|
3
|
3
|
6
|
3
|
111
|
|
IDCR Insolvent (s=0.5)
|
3.59 (0.00)
|
5.24 (0.00)
|
5.70 (0.00)
|
2.22 (0.00)
|
3.92 (0.00)
|
6.60 (0.00)
|
7.41 (0.00)
|
3.96 (0.00)
|
|
|
6
|
15
|
44
|
1
|
2
|
7
|
2
|
126
|
|
IDCR Insolvent (s=1.0)
|
27.54 (0.00)
|
28.67 (0.00)
|
28.76 (0.00)
|
20.00 (0.00)
|
21.57 (0.00)
|
29.25 (0.00)
|
29.63 (0.00)
|
28.01 (0.00)
|
|
|
46
|
82
|
222
|
9
|
11
|
31
|
8
|
890
|
|
Capital Insolvent (s=0.5)
|
1.80 (0.00)
|
3.15 (0.00)
|
3.11 (0.00)
|
2.22 (0.00)
|
1.96 (0.00)
|
2.83 (0.00)
|
14.81 (0.00)
|
1.99 (0.00)
|
|
|
3
|
9
|
24
|
1
|
1
|
3
|
4
|
64
|
|
Capital Insolvent (s=1.0)
|
11.38 (0.00)
|
10.14 (0.00)
|
9.46 (0.00)
|
4.44 (0.00)
|
11.76 (0.00)
|
14.15 (0.00)
|
18.52 (0.00)
|
6.58 (0.00)
|
|
|
19
|
29
|
73
|
2
|
6
|
15
|
5
|
212
|
# Export (silently)
invisible(export_table(
table5,
"table5_q1_2023",
caption = "Pre-Crisis Bank Characteristics by Borrower Type: Q1 2023",
label = "tab:q1_2023",
note = "Bank characteristics from 2022Q4 Call Reports. Q1 2023: January 1 -- March 31, 2023. FHLB borrowers identified as banks with increased FHLB advances in 2023Q1 vs 2022Q4. Sample excludes G-SIBs and failed banks."
))
LaTeX table saved: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/borrowing_stat/tables/table5_q1_2023.tex
HTML table saved: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/borrowing_stat/tables/table5_q1_2023.html
CSV table saved: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/borrowing_stat/tables/table5_q1_2023.csv
Table 6: Arbitrage
Period - BTFP, DW, Both, Non-Borrowers
# ==============================================================================
# TABLE 6: Arbitrage Period (Nov 1, 2023 - Jan 24, 2024)
# ==============================================================================
arb_start <- as.Date("2023-11-01")
arb_end <- as.Date("2024-01-24")
btfp_arb <- btfp_loans_raw %>%
filter(btfp_loan_date >= arb_start & btfp_loan_date <= arb_end) %>%
distinct(rssd_id) %>%
pull(rssd_id)
# Note: DW data ends 2023-12-31, so only partial coverage
dw_arb <- dw_loans_raw %>%
filter(dw_loan_date >= arb_start & dw_loan_date <= min(arb_end, DW_DATA_END)) %>%
distinct(rssd_id) %>%
pull(rssd_id)
table6_data <- baseline_q %>%
mutate(
btfp_user = idrssd %in% btfp_arb,
dw_user = idrssd %in% dw_arb,
borrower_type = case_when(
btfp_user & dw_user ~ "Both",
btfp_user & !dw_user ~ "BTFP Only",
!btfp_user & dw_user ~ "DW Only",
TRUE ~ "Non-Borrower"
)
)
table6 <- create_summary_table_v2(table6_data, "borrower_type",
group_levels = c("BTFP Only", "DW Only", "Both", "Non-Borrower"))
# Display
table6 %>%
select(-Row_Type) %>%
kable(caption = "Table 6: Pre-Crisis Bank Characteristics by Borrower Type (Arbitrage Period: Nov 1, 2023 - Jan 24, 2024)") %>%
kable_styling(bootstrap_options = c("striped", "hover"), full_width = FALSE)
Table 6: Pre-Crisis Bank Characteristics by Borrower Type (Arbitrage
Period: Nov 1, 2023 - Jan 24, 2024)
|
Variable
|
BTFP Only
|
DW Only
|
Both
|
Non-Borrower
|
|
N
|
644
|
323
|
145
|
3584
|
|
Book Equity Ratio
|
8.57 (8.31)
|
9.55 (9.27)
|
8.15 (8.08)
|
11.51 (9.18)
|
|
|
644
|
323
|
145
|
3584
|
|
MTM Loss (HTM)
|
0.25 (0.00)
|
0.16 (0.00)
|
0.21 (0.00)
|
0.20 (0.00)
|
|
|
644
|
323
|
145
|
3566
|
|
MTM Loss (Total)
|
6.20 (5.96)
|
5.24 (5.09)
|
6.12 (6.12)
|
5.33 (5.12)
|
|
|
644
|
323
|
145
|
3566
|
|
Adjusted Equity
|
2.37 (2.28)
|
4.31 (4.14)
|
2.03 (1.93)
|
5.97 (4.15)
|
|
|
644
|
323
|
145
|
3566
|
|
Uninsured Leverage
|
414.69 (305.51)
|
361.76 (280.41)
|
535.11 (321.47)
|
305.76 (232.24)
|
|
|
644
|
323
|
145
|
3584
|
|
MTM Insolvent
|
27.33 (0.00)
|
14.55 (0.00)
|
31.72 (0.00)
|
16.69 (0.00)
|
|
|
176
|
47
|
46
|
595
|
|
Run Risk (Median Cutoff)
|
36.34 (0.00)
|
24.77 (0.00)
|
35.17 (0.00)
|
20.50 (0.00)
|
|
|
234
|
80
|
51
|
733
|
|
Run Risk (Q75 Cutoff)
|
7.14 (0.00)
|
4.95 (0.00)
|
11.03 (0.00)
|
3.35 (0.00)
|
|
|
46
|
16
|
16
|
120
|
|
IDCR Insolvent (s=0.5)
|
4.66 (0.00)
|
4.02 (0.00)
|
3.45 (0.00)
|
4.40 (0.00)
|
|
|
30
|
13
|
5
|
155
|
|
IDCR Insolvent (s=1.0)
|
31.99 (0.00)
|
29.72 (0.00)
|
32.41 (0.00)
|
26.99 (0.00)
|
|
|
206
|
96
|
47
|
950
|
|
Capital Insolvent (s=0.5)
|
1.86 (0.00)
|
4.64 (0.00)
|
6.21 (0.00)
|
2.05 (0.00)
|
|
|
12
|
15
|
9
|
73
|
|
Capital Insolvent (s=1.0)
|
8.85 (0.00)
|
11.15 (0.00)
|
13.79 (0.00)
|
6.95 (0.00)
|
|
|
57
|
36
|
20
|
248
|
# Export (silently)
invisible(export_table(
table6,
"table6_arbitrage_period",
caption = "Pre-Crisis Bank Characteristics by Borrower Type: Arbitrage Period",
label = "tab:arbitrage_period",
note = "Bank characteristics from 2022Q4 Call Reports. Arbitrage period: November 1, 2023 -- January 24, 2024 (BTFP rate below IORB). Note: DW data available only through December 31, 2023. Sample excludes G-SIBs and failed banks."
))
LaTeX table saved: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/borrowing_stat/tables/table6_arbitrage_period.tex
HTML table saved: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/borrowing_stat/tables/table6_arbitrage_period.html
CSV table saved: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/borrowing_stat/tables/table6_arbitrage_period.csv
Table 7: FHLB Borrower
Characteristics
# ==============================================================================
# TABLE 7: FHLB Borrower Characteristics (Q1 2023)
# ==============================================================================
# Prepare data with FHLB variables
table7_data <- baseline_q %>%
mutate(
# Rename FHLB variables for clarity
fhlb_change = change_fhlb_adv_fwd_q,
fhlb_abnormal_5pct = abnormal_fhlb_borrowing_5pct,
fhlb_abnormal_10pct = abnormal_fhlb_borrowing_10pct,
# Scale total assets to billions for readability
total_asset_bn = total_asset / 1e9,
# Categorize by FHLB usage
fhlb_user = idrssd %in% fhlb_borrowers,
btfp_user = idrssd %in% btfp_q1,
dw_user = idrssd %in% dw_q1,
borrower_type = case_when(
fhlb_user & btfp_user & dw_user ~ "FHLB+BTFP+DW",
fhlb_user & btfp_user & !dw_user ~ "FHLB+BTFP",
fhlb_user & !btfp_user & dw_user ~ "FHLB+DW",
fhlb_user & !btfp_user & !dw_user ~ "FHLB Only",
!fhlb_user & (btfp_user | dw_user) ~ "Fed Only",
TRUE ~ "Non-Borrower"
)
)
# Define variables and labels for summary stats
fhlb_vars <- c(
"fhlb_change", "fhlb_abnormal_5pct", "fhlb_abnormal_10pct",
"total_asset_bn", "book_equity_to_total_asset", "mtm_loss_to_total_asset",
"adjusted_equity", "uninsured_deposit_to_total_asset", "uninsured_leverage",
"mtm_insolvent", "run_risk_median", "run_risk_q75"
)
fhlb_labels <- c(
"fhlb_change", "fhlb_abnormal_5pct", "fhlb_abnormal_10pct",
"total_asset_bn", "book_equity_ratio", "mtm_loss_ratio",
"adjusted_equity", "uninsured_dep_ratio", "uninsured_leverage",
"mtm_insolvent", "run_risk_median", "run_risk_q75"
)
# FHLB borrower breakdown
table7_data %>%
count(borrower_type) %>%
arrange(desc(n)) %>%
pander(caption = "FHLB Borrower Breakdown")
FHLB Borrower Breakdown
| Non-Borrower |
3242 |
| FHLB Only |
772 |
| Fed Only |
498 |
| FHLB+DW |
106 |
| FHLB+BTFP |
51 |
| FHLB+BTFP+DW |
27 |
# Full sample summary stats
full_stats <- summary_stats(table7_data, fhlb_vars, fhlb_labels)
pander(full_stats, caption = "Full Sample Summary Statistics")
Full Sample Summary Statistics
| fhlb_change |
2469 |
159.1 |
3147 |
-90.08 |
-30.51 |
0 |
30.37 |
120 |
| fhlb_abnormal_5pct |
4696 |
0.05 |
0.22 |
0 |
0 |
0 |
0 |
0 |
| fhlb_abnormal_10pct |
4696 |
0.07 |
0.25 |
0 |
0 |
0 |
0 |
0 |
| total_asset_bn |
4696 |
0 |
0.02 |
0 |
0 |
0 |
0 |
0 |
| book_equity_ratio |
4696 |
10.87 |
10.44 |
5.55 |
7.24 |
9 |
11.18 |
14.95 |
| mtm_loss_ratio |
4678 |
5.47 |
4.34 |
2.65 |
3.8 |
5.27 |
6.96 |
8.48 |
| adjusted_equity |
4678 |
5.24 |
11.39 |
-1.72 |
0.97 |
3.81 |
6.73 |
10.46 |
| uninsured_dep_ratio |
4696 |
23.27 |
12.38 |
8.87 |
14.81 |
21.93 |
30.17 |
39.09 |
| uninsured_leverage |
4696 |
331.6 |
823.5 |
75.6 |
152.6 |
250.9 |
374.1 |
541.4 |
| mtm_insolvent |
4678 |
0.18 |
0.39 |
0 |
0 |
0 |
0 |
1 |
| run_risk_median |
4688 |
0.23 |
0.42 |
0 |
0 |
0 |
0 |
1 |
| run_risk_q75 |
4689 |
0.04 |
0.2 |
0 |
0 |
0 |
0 |
0 |
# Summary stats by borrower type
for (btype in c("FHLB Only", "FHLB+BTFP", "FHLB+DW", "FHLB+BTFP+DW", "Fed Only", "Non-Borrower")) {
group_data <- table7_data %>% filter(borrower_type == btype)
if (nrow(group_data) > 0) {
group_stats <- summary_stats(group_data, fhlb_vars, fhlb_labels)
pander(group_stats, caption = paste0(btype, " (N = ", nrow(group_data), ")"))
}
}
Notes: Bank characteristics from 2022Q4 Call
Reports. FHLB borrowers identified as banks with increased FHLB advances
in 2023Q1 vs 2022Q4. Abnormal FHLB borrowing is measured using Z-scores
calculated as (current quarter change - mean) / std, where mean and
standard deviation are computed from the prior 8 quarters
(2021Q1-2022Q4). Abnormal FHLB (5%) indicates Z > 1.65 (top 5
percentile); Abnormal FHLB (10%) indicates Z > 1.28 (top 10
percentile). Fed Only = BTFP and/or DW but no FHLB increase. Sample
excludes G-SIBs and failed banks.
Summary
Statistics
# ==============================================================================
# SUMMARY OF BORROWER COUNTS
# ==============================================================================
summary_counts <- tribble(
~Table, ~Period, ~Description, ~N_Borrowers, ~N_Total,
"Table 1", "March 10, 2023", "DW only", length(dw_mar10), nrow(baseline_q),
"Table 2", "March 10-13, 2023", "DW only", length(dw_mar10_13), nrow(baseline_q),
"Table 3", "March 9-14, 2023", "DW only", length(dw_mar9_14), nrow(baseline_q),
"Table 4", "Acute (Mar 13 - May 1)", "BTFP + DW", length(union(btfp_acute, dw_acute)), nrow(baseline_q),
"Table 5", "Q1 2023 (Jan-Mar)", "BTFP + DW + FHLB", length(union(union(btfp_q1, dw_q1), fhlb_borrowers)), nrow(baseline_q),
"Table 6", "Arbitrage (Nov 1 - Jan 24)", "BTFP + DW", length(union(btfp_arb, dw_arb)), nrow(baseline_q),
"Table 7", "Q1 2023", "FHLB focus", length(fhlb_borrowers), nrow(baseline_q)
)
summary_counts %>%
kable(caption = "Summary of Borrower Counts Across Tables") %>%
kable_styling(bootstrap_options = c("striped", "hover"), full_width = FALSE)
Summary of Borrower Counts Across Tables
|
Table
|
Period
|
Description
|
N_Borrowers
|
N_Total
|
|
Table 1
|
March 10, 2023
|
DW only
|
52
|
4696
|
|
Table 2
|
March 10-13, 2023
|
DW only
|
98
|
4696
|
|
Table 3
|
March 9-14, 2023
|
DW only
|
137
|
4696
|
|
Table 4
|
Acute (Mar 13 - May 1)
|
BTFP + DW
|
813
|
4696
|
|
Table 5
|
Q1 2023 (Jan-Mar)
|
BTFP + DW + FHLB
|
1471
|
4696
|
|
Table 6
|
Arbitrage (Nov 1 - Jan 24)
|
BTFP + DW
|
1132
|
4696
|
|
Table 7
|
Q1 2023
|
FHLB focus
|
956
|
4696
|
# Save summary (silently)
invisible(write_csv(summary_counts, file.path(TABLE_PATH, "summary_borrower_counts.csv")))
Session Info