This analysis addresses the research questions on bank borrowing
behavior across Federal Reserve emergency facilities (BTFP and Discount
Window) during the March 2023 banking crisis.
Key Additions Based
on Advisor Feedback
- Jiang et al. Insolvency Link: Comprehensive
analysis linking insolvent banks (by adjusted equity and IDCR measures)
to LOLR borrowing by period
- GSIB-Specific Analysis: Separate analysis for
G-SIBs (with custodial bank identification)
- MTM Total Only: Extensive margin models now use
total MTM loss (not split by eligible/non-eligible)
- Acute Period Deep Dive: Descriptive stats by
subsamples for the acute phase
- Run Risk Dummy: Added to temporal
specifications
- Insolvency Specs for DW: Complete insolvency
analysis for both BTFP and DW (without book equity)
- FHLB Borrowing Analysis: Analysis of banks with
FHLB increases, focused on $10-100B banks
# ============================================================================
# LOAD PACKAGES
# ============================================================================
pkgs <- c("tidyverse", "data.table", "lubridate", "fixest", "sandwich",
"lmtest", "broom", "margins", "pROC", "sampleSelection",
"modelsummary", "kableExtra", "patchwork", "scales",
"viridis", "psych")
invisible(lapply(pkgs, library, character.only = TRUE))
cat("All packages loaded successfully.\n")
All packages loaded successfully.
# ============================================================================
# HELPER FUNCTIONS
# ============================================================================
# Winsorization function
winsorize <- function(x, probs = c(0.025, 0.975)) {
if (all(is.na(x))) return(x)
q <- quantile(x, probs = probs, na.rm = TRUE, names = FALSE)
pmax(pmin(x, q[2]), q[1])
}
# Safe division
safe_div <- function(num, denom, default = NA_real_) {
ifelse(is.na(denom) | denom == 0, default, num / denom)
}
# Significance stars
add_stars <- function(pval) {
case_when(pval < 0.01 ~ "***", pval < 0.05 ~ "**", pval < 0.10 ~ "*", TRUE ~ "")
}
# Publication theme
theme_pub <- function(base_size = 12) {
theme_minimal(base_size = base_size) +
theme(
plot.title = element_text(face = "bold", size = base_size + 2),
plot.subtitle = element_text(color = "gray40"),
legend.position = "bottom",
panel.grid.minor = element_blank(),
axis.title = element_text(face = "bold")
)
}
# Color palettes
COLORS <- list(
btfp = "#2E86AB", dw = "#A23B72", both = "#F18F01", neither = "gray70",
acute = "#FC9272", post = "#A1D99B", arb = "#9ECAE1", winddown = "#DADAEB"
)
facility_colors <- c(
"BTFP" = "#2E86AB", "DW" = "#A23B72", "Discount Window" = "#A23B72",
"Both" = "#F18F01", "Neither" = "gray70"
)
# ============================================================================
# OUTPUT DIRECTORY SETUP
# ============================================================================
OUTPUT_PATH <- "C:/Users/mferdo2/OneDrive - Louisiana State University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/final_result_04"
if (!dir.exists(OUTPUT_PATH)) dir.create(OUTPUT_PATH, recursive = TRUE)
TABLE_PATH <- file.path(OUTPUT_PATH, "tables")
FIG_PATH <- file.path(OUTPUT_PATH, "figures")
if (!dir.exists(TABLE_PATH)) dir.create(TABLE_PATH)
if (!dir.exists(FIG_PATH)) dir.create(FIG_PATH)
cat("Output directories set up:\n")
Output directories set up:
cat(" Tables:", TABLE_PATH, "\n")
Tables: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/final_result_04/tables
cat(" Figures:", FIG_PATH, "\n")
Figures: C:/Users/mferdo2/OneDrive - Louisiana State
University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025/03_documentation/final_result_04/figures
# ============================================================================
# CANONICAL 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"
periods %>%
kable(col.names = c("Period", "Name", "Start", "End", "Description"),
caption = "Table 0: Analysis Period Definitions") %>%
kable_styling(bootstrap_options = c("striped", "hover"), full_width = FALSE)
Table 0: 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
# ============================================================================
BASE_PATH <- "C:/Users/mferdo2/OneDrive - Louisiana State University/Finance_PhD/DW_Stigma_paper/Liquidity_project_2025"
DATA_PROC <- file.path(BASE_PATH, "01_data/processed")
call_q <- read_csv(file.path(DATA_PROC, "final_call_gsib.csv"), show_col_types = FALSE) %>%
mutate(idrssd = as.character(idrssd))
btfp_loans_raw <- read_csv(file.path(DATA_PROC, "btfp_loan_bank_only.csv"), show_col_types = FALSE) %>%
mutate(rssd_id = as.character(rssd_id), btfp_loan_date = mdy(btfp_loan_date))
dw_loans_raw <- read_csv(file.path(DATA_PROC, "dw_loan_bank_2023.csv"), show_col_types = FALSE) %>%
mutate(rssd_id = as.character(rssd_id), dw_loan_date = ymd(dw_loan_date))
cat("=== DATA LOADED ===\n")
=== DATA LOADED ===
cat("Call Report:", nrow(call_q), "| BTFP Loans:", nrow(btfp_loans_raw), "| DW Loans:", nrow(dw_loans_raw), "\n")
Call Report: 61002 | BTFP Loans: 6734 | DW Loans: 10008
# ============================================================================
# ASSIGN PERIODS TO LOANS
# ============================================================================
assign_period <- function(date, periods_df) {
sapply(date, function(d) {
if (is.na(d)) return(NA_integer_)
match <- periods_df %>% filter(d >= start_date & d <= end_date) %>% pull(period_num)
if (length(match) == 0) NA_integer_ else match[1]
})
}
btfp_loans_raw <- btfp_loans_raw %>% mutate(period = assign_period(btfp_loan_date, periods))
dw_loans_raw <- dw_loans_raw %>% mutate(period = assign_period(dw_loan_date, periods))
# ============================================================================
# DEFINE EXCLUDED BANKS AND GSIB IDENTIFICATION
# ============================================================================
# Identify GSIBs at baseline (KEEP SEPARATE FOR GSIB ANALYSIS)
gsib_banks <- call_q %>%
filter(quarter == BASELINE_DATE, gsib == 1) %>%
pull(idrssd)
# Identify failed banks
failed_banks <- call_q %>%
filter(quarter == BASELINE_DATE, failed_bank == 1) %>%
pull(idrssd)
# Exclude failed + GSIBs for main analysis
excluded_banks <- call_q %>%
filter(quarter == BASELINE_DATE, failed_bank == 1 | gsib == 1) %>%
pull(idrssd)
cat("=== EXCLUSIONS ===\n")
=== EXCLUSIONS ===
cat("G-SIBs:", length(gsib_banks), "\n")
G-SIBs: 33
cat("Failed banks:", length(failed_banks), "\n")
Failed banks: 8
cat("Total excluded:", length(excluded_banks), "\n")
Total excluded: 41
# ============================================================================
# IDENTIFY CUSTODIAL BANKS AMONG GSIBs
# Note: Custodial banks (State Street, BNY Mellon) may not have typical
# uninsured deposit profiles - flag for separate analysis
# ============================================================================
# These are known custodial banks - verify RSSDIDs in your data
custodial_banks <- call_q %>%
filter(quarter == BASELINE_DATE, gsib == 1) %>%
filter(grepl("STATE STREET|MELLON|NORTHERN TRUST", toupper(bank_name), ignore.case = TRUE)) %>%
pull(idrssd)
cat("Custodial banks identified:", length(custodial_banks), "\n")
Custodial banks identified: 6
# Non-custodial GSIBs
gsib_non_custodial <- setdiff(gsib_banks, custodial_banks)
cat("Non-custodial GSIBs:", length(gsib_non_custodial), "\n")
Non-custodial GSIBs: 27
# Create filtered loan datasets
btfp_loans <- btfp_loans_raw %>% filter(!rssd_id %in% excluded_banks)
dw_loans <- dw_loans_raw %>% filter(!rssd_id %in% excluded_banks)
# GSIB-only loan datasets
btfp_loans_gsib <- btfp_loans_raw %>% filter(rssd_id %in% gsib_banks)
dw_loans_gsib <- dw_loans_raw %>% filter(rssd_id %in% gsib_banks)
# ============================================================================
# AGGREGATE BORROWER-LEVEL DATA
# ============================================================================
# Helper function for aggregation
aggregate_btfp <- function(loans_df) {
loans_df %>%
filter(!is.na(period), period >= 1) %>%
group_by(rssd_id) %>%
summarise(
btfp = 1L,
btfp_amount = sum(btfp_loan_amount, na.rm = TRUE),
btfp_first_date = min(btfp_loan_date),
btfp_first_period = first(period[btfp_loan_date == min(btfp_loan_date)]),
btfp_n_loans = n(),
btfp_collateral = sum(btfp_total_collateral, na.rm = TRUE),
.groups = "drop"
) %>% rename(idrssd = rssd_id)
}
aggregate_dw_post <- function(loans_df) {
loans_df %>%
filter(!is.na(period), period >= 1) %>%
group_by(rssd_id) %>%
summarise(
dw = 1L,
dw_amount = sum(dw_loan_amount, na.rm = TRUE),
dw_first_date = min(dw_loan_date),
dw_first_period = first(period[dw_loan_date == min(dw_loan_date)]),
dw_omo_coll = sum(dw_omo_eligible, na.rm = TRUE),
dw_non_omo_coll = sum(dw_non_omo_eligible, na.rm = TRUE),
.groups = "drop"
) %>% rename(idrssd = rssd_id)
}
aggregate_dw_pre <- function(loans_df) {
loans_df %>%
filter(!is.na(period), period == 0) %>%
group_by(rssd_id) %>%
summarise(
dw_pre = 1L,
dw_pre_amount = sum(dw_loan_amount, na.rm = TRUE),
dw_pre_first_date = min(dw_loan_date),
.groups = "drop"
) %>% rename(idrssd = rssd_id)
}
# Period-specific aggregation for temporal analysis
aggregate_btfp_by_period <- function(loans_df, target_period) {
loans_df %>%
filter(!is.na(period), period == target_period) %>%
group_by(rssd_id) %>%
summarise(
btfp_period = 1L,
btfp_period_amount = sum(btfp_loan_amount, na.rm = TRUE),
.groups = "drop"
) %>% rename(idrssd = rssd_id)
}
aggregate_dw_by_period <- function(loans_df, target_period) {
loans_df %>%
filter(!is.na(period), period == target_period) %>%
group_by(rssd_id) %>%
summarise(
dw_period = 1L,
dw_period_amount = sum(dw_loan_amount, na.rm = TRUE),
.groups = "drop"
) %>% rename(idrssd = rssd_id)
}
# Create aggregations
btfp_agg_all <- aggregate_btfp(btfp_loans_raw)
btfp_agg <- aggregate_btfp(btfp_loans)
btfp_agg_gsib <- aggregate_btfp(btfp_loans_gsib)
dw_post_agg_all <- aggregate_dw_post(dw_loans_raw)
dw_post_agg <- aggregate_dw_post(dw_loans)
dw_post_agg_gsib <- aggregate_dw_post(dw_loans_gsib)
dw_pre_agg_all <- aggregate_dw_pre(dw_loans_raw)
dw_pre_agg <- aggregate_dw_pre(dw_loans)
dw_pre_agg_gsib <- aggregate_dw_pre(dw_loans_gsib)
# Acute period borrowers
btfp_acute_agg <- aggregate_btfp_by_period(btfp_loans, 1)
dw_acute_agg <- aggregate_dw_by_period(dw_loans, 1)
cat("\n=== AGGREGATED BORROWERS ===\n")
=== AGGREGATED BORROWERS ===
cat("Main sample - BTFP:", nrow(btfp_agg), "| DW post:", nrow(dw_post_agg), "| DW pre:", nrow(dw_pre_agg), "\n")
Main sample - BTFP: 1316 | DW post: 1389 | DW pre: 106
cat("GSIB sample - BTFP:", nrow(btfp_agg_gsib), "| DW post:", nrow(dw_post_agg_gsib), "| DW pre:", nrow(dw_pre_agg_gsib), "\n")
GSIB sample - BTFP: 10 | DW post: 17 | DW pre: 0
# ============================================================================
# CONSTRUCT ANALYSIS DATASETS
# ============================================================================
create_analysis_df <- function(baseline_data, btfp_agg, dw_post_agg, dw_pre_agg) {
df_raw <- baseline_data %>%
left_join(btfp_agg, by = "idrssd") %>%
left_join(dw_post_agg, by = "idrssd") %>%
left_join(dw_pre_agg, by = "idrssd") %>%
mutate(
btfp = replace_na(btfp, 0L),
dw = replace_na(dw, 0L),
dw_pre = replace_na(dw_pre, 0L),
any_fed = as.integer(btfp == 1 | dw == 1),
both = as.integer(btfp == 1 & dw == 1),
btfp_only = as.integer(btfp == 1 & dw == 0),
dw_only = as.integer(btfp == 0 & dw == 1),
facility_choice = factor(
case_when(both == 1 ~ "Both", btfp_only == 1 ~ "BTFP_Only",
dw_only == 1 ~ "DW_Only", TRUE ~ "Neither"),
levels = c("Neither", "BTFP_Only", "DW_Only", "Both")
),
# Period-specific entry indicators
btfp_acute = as.integer(!is.na(btfp_first_period) & btfp_first_period == 1),
btfp_post = as.integer(!is.na(btfp_first_period) & btfp_first_period == 2),
btfp_arb = as.integer(!is.na(btfp_first_period) & btfp_first_period == 3),
btfp_winddown = as.integer(!is.na(btfp_first_period) & btfp_first_period == 4),
dw_acute = as.integer(!is.na(dw_first_period) & dw_first_period == 1),
dw_post = as.integer(!is.na(dw_first_period) & dw_first_period == 2),
dw_arb = as.integer(!is.na(dw_first_period) & dw_first_period == 3),
# RAW KEY VARIABLES
mtm_btfp_raw = mtm_loss_omo_eligible_to_total_asset,
mtm_other_raw = mtm_loss_non_omo_eligible_to_total_asset,
mtm_total_raw = mtm_loss_to_total_asset,
borrowing_subsidy_raw = mtm_loss_omo_eligible_to_omo_eligible,
uninsured_lev_raw = uninsured_deposit_to_total_asset,
uninsured_share_raw = uninsured_to_deposit,
omo_ratio_raw = omo_eligible_to_total_asset,
non_omo_ratio_raw = non_omo_eligible_to_total_asset,
mv_asset = mm_asset,
change_fhlb_fwd_q_raw = change_fhlb_adv_fwd_q,
# Jiang et al. insolvency measures
adjusted_equity_raw = book_equity_to_total_asset - mtm_loss_to_total_asset,
mv_adjustment_raw = if_else(mv_asset == 0 | is.na(mv_asset), NA_real_, (total_asset / mv_asset) - 1),
idcr_1_raw = safe_div(mv_asset - 0.5 * uninsured_deposit - insured_deposit, insured_deposit),
idcr_2_raw = safe_div(mv_asset - 1.0 * uninsured_deposit - insured_deposit, insured_deposit),
insolvency_1_raw = safe_div((total_asset - total_liability) - 0.5 * uninsured_deposit * mv_adjustment_raw, total_asset),
insolvency_2_raw = safe_div((total_asset - total_liability) - 1.0 * uninsured_deposit * mv_adjustment_raw, total_asset),
# Controls
ln_assets_raw = log(total_asset),
cash_ratio_raw = cash_to_total_asset,
securities_ratio_raw = security_to_total_asset,
loan_ratio_raw = total_loan_to_total_asset,
book_equity_ratio_raw = book_equity_to_total_asset,
tier1_ratio_raw = tier1cap_to_total_asset,
roa_raw = roa,
loan_to_deposit_raw = loan_to_deposit,
fhlb_ratio_raw = fhlb_to_total_asset,
wholesale_raw = safe_div(
fed_fund_purchase + repo + replace_na(other_borrowed_less_than_1yr, 0),
total_liability, 0
) * 100,
# Size categorization
size_cat = factor(size_bin, levels = c("small", "large")),
prior_dw = dw_pre,
# State for clustering
state = if("state" %in% names(.)) state else NA_character_
)
# WINSORIZE
df <- df_raw %>%
mutate(
mtm_btfp = winsorize(mtm_btfp_raw),
mtm_other = winsorize(mtm_other_raw),
mtm_total = winsorize(mtm_total_raw),
borrowing_subsidy = winsorize(borrowing_subsidy_raw),
uninsured_lev = winsorize(uninsured_lev_raw),
uninsured_share = winsorize(uninsured_share_raw),
omo_ratio = winsorize(omo_ratio_raw),
non_omo_ratio = winsorize(non_omo_ratio_raw),
fhlb_change = winsorize(change_fhlb_fwd_q_raw),
ln_assets = winsorize(ln_assets_raw),
cash_ratio = winsorize(cash_ratio_raw),
securities_ratio = winsorize(securities_ratio_raw),
loan_ratio = winsorize(loan_ratio_raw),
book_equity_ratio = winsorize(book_equity_ratio_raw),
tier1_ratio = winsorize(tier1_ratio_raw),
roa = winsorize(roa_raw),
loan_to_deposit = winsorize(loan_to_deposit_raw),
fhlb_ratio = winsorize(fhlb_ratio_raw),
wholesale = winsorize(wholesale_raw),
adjusted_equity = winsorize(adjusted_equity_raw),
mv_adjustment = winsorize(mv_adjustment_raw),
idcr_1 = winsorize(idcr_1_raw),
idcr_2 = winsorize(idcr_2_raw),
insolvency_1 = winsorize(insolvency_1_raw),
insolvency_2 = winsorize(insolvency_2_raw),
# Insolvency indicators
mtm_insolvent = as.integer(adjusted_equity < 0),
insolvent_idcr_s50 = as.integer(idcr_1 < 0),
insolvent_idcr_s100 = as.integer(idcr_2 < 0),
insolvent_cap_s50 = as.integer(insolvency_1 < 0),
insolvent_cap_s100 = as.integer(insolvency_2 < 0),
# Run risk measures (continuous and dummy)
run_risk = uninsured_share * mtm_total,
high_uninsured = as.integer(uninsured_lev > median(uninsured_lev, na.rm = TRUE)),
high_mtm = as.integer(mtm_total > median(mtm_total, na.rm = TRUE))
) %>%
# Calculate run risk dummy based on sample medians
mutate(
median_uninsured = median(uninsured_share, na.rm = TRUE),
median_mtm = median(mtm_total, na.rm = TRUE),
run_risk_dummy = as.integer(uninsured_share > median_uninsured & mtm_total > median_mtm)
)
# Intensive margin (BTFP only)
df <- df %>%
mutate(
btfp_pct = ifelse(btfp == 1, 100 * btfp_amount / (total_asset * 1000), NA_real_),
dw_pct = ifelse(dw == 1, 100 * dw_amount / (total_asset * 1000), NA_real_),
btfp_util = ifelse(btfp == 1 & omo_eligible > 0, btfp_amount / (omo_eligible * 1000), NA_real_),
maxed_out = as.integer(!is.na(btfp_util) & btfp_util > 0.90),
total_fed_pct = ifelse(any_fed == 1,
100 * (replace_na(btfp_amount, 0) + replace_na(dw_amount, 0)) / (total_asset * 1000),
NA_real_)
)
# Remove raw variables
df <- df %>% select(-ends_with("_raw"), -median_uninsured, -median_mtm)
return(df)
}
# Create datasets
baseline_all <- call_q %>% filter(quarter == BASELINE_DATE)
baseline_excl <- call_q %>% filter(quarter == BASELINE_DATE, !idrssd %in% excluded_banks)
baseline_gsib <- call_q %>% filter(quarter == BASELINE_DATE, idrssd %in% gsib_banks)
baseline_gsib_nc <- call_q %>% filter(quarter == BASELINE_DATE, idrssd %in% gsib_non_custodial)
# Main analysis samples
df_all <- create_analysis_df(baseline_all, btfp_agg_all, dw_post_agg_all, dw_pre_agg_all)
df <- create_analysis_df(baseline_excl, btfp_agg, dw_post_agg, dw_pre_agg)
df_gsib <- create_analysis_df(baseline_gsib, btfp_agg_gsib, dw_post_agg_gsib, dw_pre_agg_gsib)
df_gsib_nc <- create_analysis_df(baseline_gsib_nc, btfp_agg_gsib, dw_post_agg_gsib, dw_pre_agg_gsib)
cat("\n=== ANALYSIS DATASETS CREATED ===\n")
=== ANALYSIS DATASETS CREATED ===
cat("df (main sample, excl. failed & GSIB):", nrow(df), "\n")
df (main sample, excl. failed & GSIB): 4696
cat("df_gsib (GSIBs only):", nrow(df_gsib), "\n")
df_gsib (GSIBs only): 33
cat("df_gsib_nc (non-custodial GSIBs):", nrow(df_gsib_nc), "\n")
df_gsib_nc (non-custodial GSIBs): 27
cat(" - BTFP users:", sum(df$btfp), "\n")
cat(" - DW users:", sum(df$dw), "\n")
cat(" - Both:", sum(df$both), "\n")
# ============================================================================
# GLOBAL MODEL SETTINGS
# ============================================================================
# Standard controls WITH book equity (for non-insolvency specs)
CONTROLS <- "ln_assets + cash_ratio + securities_ratio + loan_to_deposit + book_equity_ratio + wholesale + fhlb_ratio + roa"
# Controls WITHOUT book equity (for insolvency specs - per advisor feedback)
CONTROLS_INSOLVENCY <- "ln_assets + cash_ratio + securities_ratio + loan_to_deposit + wholesale + fhlb_ratio + roa"
# Cluster at bank level
CLUSTER_VAR <- "idrssd"
# Coefficient labels
COEF_MAP <- c(
# Insolvency measures
"adjusted_equity" = "Adjusted Equity",
"mtm_insolvent" = "MTM Insolvent",
"idcr_1" = "IDCR (s=0.5)",
"idcr_2" = "IDCR (s=1.0)",
"insolvent_idcr_s50" = "Insolvent IDCR (s=0.5)",
"insolvent_idcr_s100" = "Insolvent IDCR (s=1.0)",
"insolvency_1" = "Insolvency Cap (s=0.5)",
"insolvency_2" = "Insolvency Cap (s=1.0)",
"insolvent_cap_s50" = "Insolvent Cap (s=0.5)",
"insolvent_cap_s100" = "Insolvent Cap (s=1.0)",
# MTM measures - USE TOTAL ONLY FOR EXTENSIVE MARGIN
"mtm_total" = "MTM Loss (Total)",
"mtm_btfp" = "MTM Loss (BTFP-Eligible)",
"mtm_other" = "MTM Loss (Non-BTFP)",
"uninsured_lev" = "Uninsured Leverage",
"borrowing_subsidy" = "Borrowing Subsidy",
"run_risk" = "Run Risk (MTM × % Uninsured)",
"run_risk_dummy" = "Run Risk Dummy",
"I(mtm_total * uninsured_lev)" = "MTM × Uninsured",
"fhlb_change" = "FHLB Change (fwd Q)",
"fhlb_borrower" = "FHLB Borrower",
# Controls
"prior_dw" = "Prior DW User",
"omo_ratio" = "BTFP-Eligible Ratio",
"non_omo_ratio" = "DW-Only Eligible Ratio",
"maxed_out" = "Maxed Out (>90%)",
"high_uninsured" = "High Uninsured",
"high_mtm" = "High MTM",
"ln_assets" = "Log(Assets)",
"cash_ratio" = "Cash Ratio",
"securities_ratio" = "Securities Ratio",
"loan_to_deposit" = "Loan/Deposit",
"book_equity_ratio" = "Book Equity",
"wholesale" = "Wholesale Funding",
"fhlb_ratio" = "FHLB Ratio",
"roa" = "ROA"
)
cat("=== MODEL SETTINGS ===\n")
=== MODEL SETTINGS ===
cat("Main sample size:", nrow(df), "\n")
Main sample size: 4696
cat("GSIB sample size:", nrow(df_gsib), "\n")
GSIB sample size: 33
