An R package for downloading data from the Bank of England Statistical Database.
The Bank of England is the United Kingdom's central bank. Founded in 1694, it is responsible for setting monetary policy (including Bank Rate), issuing banknotes, supervising the banking system, and maintaining financial stability. Its Monetary Policy Committee meets eight times a year to set the interest rate that ripples through every mortgage, savings account, and bond in the UK economy.
The Bank publishes thousands of statistical time series through its Interactive Statistical Database - covering interest rates, exchange rates, money and credit, gilt yields, and housing market indicators. This data underpins monetary policy analysis, financial research, and economic journalism in the UK.
The bbk package on CRAN provides a single generic function for Bank of England data (bbk::boe_data()), but it is primarily a Bundesbank client - the Bank of England is one of seven central banks it covers, and its BoE support amounts to a raw API wrapper. You still need to know the series codes, and the output requires further processing.
This package is different. It is built specifically for the Bank of England and provides named, documented functions for the series people actually use - boe_bank_rate(), boe_mortgage_rates(), boe_yield_curve(), and so on. You don't need to know that Bank Rate is IUDBEDR or that a 2-year fixed mortgage rate is IUMBV34. The package handles series codes, date formatting, caching, and error handling internally.
Beyond the IADB wrappers, it also ships:
boe_curve(): the full Anderson-Sleath fitted yield curves at all maturities, with five curve types (nominal, real, implied inflation, OIS, commercial bank liability) and full historical archive coverage back to 1979 (nominal), 1985 (real), 2000 (BLC), or 2009 (OIS).boe_curve_panel()reshapes to a wide panel at chosen pillar maturities for time-series modelling.boe_search()/boe_browse(): a built-in catalogue of wrapped series so you can find codes from R rather than the website.- A
boe_tblS3 class so every returned data frame carries provenance metadata (series codes, date range, frequency, fetch timestamp).
The data is freely available, but using it programmatically requires knowing obscure series codes, constructing query URLs with a non-standard date format (DD/Mon/YYYY), parsing CSV responses with irregular date formats, and handling HTML error pages returned with HTTP 200 status codes. Every analyst who works with this data writes the same boilerplate.
This package replaces all of that with named functions that return clean data frames.
# Without this package
url <- paste0(
"https://www.bankofengland.co.uk/boeapps/database/",
"_iadb-fromshowcolumns.asp?csv.x=yes",
"&SeriesCodes=IUDBEDR&UsingCodes=Y&CSVF=TN",
"&Datefrom=01/Jan/2020&Dateto=01/Jan/2025"
)
raw <- read.csv(url)
# ... parse dates, rename columns, handle errors ...
# With this package
library(boe)
boe_bank_rate(from = "2020-01-01")install.packages("boe")
# Or install the development version from GitHub
# install.packages("devtools")
devtools::install_github("charlescoverdale/boe")Data access:
| Function | Description | From | To |
|---|---|---|---|
boe_get() |
Fetch any series by BoE series code | Any | Present |
boe_bank_rate() |
Official Bank Rate (daily or monthly) | 1975 | Present |
boe_sonia() |
SONIA risk-free reference rate (daily, monthly, or annual) | 1997 | Present |
boe_yield_curve() |
Nominal and real gilt yields at 5yr, 10yr, 20yr maturities | 1985 | Present |
boe_curve() |
Full Anderson-Sleath fitted curves (nominal / real / inflation / OIS / BLC, spot or forward) at all maturities | 1979 | Present |
boe_curve_panel() |
Wide panel of boe_curve() at chosen pillar maturities (default 0.5y, 1y, 2y, 5y, 10y, 20y) |
1979 | Present |
boe_exchange_rate() |
Daily sterling spot rates for 27 currencies | 1975 | Present |
boe_mortgage_rates() |
Quoted mortgage rates (2yr/3yr/5yr fixed, SVR) | 1995 | Present |
boe_mortgage_approvals() |
Monthly mortgage approvals for house purchase | 1993 | Present |
boe_consumer_credit() |
Consumer credit outstanding (total, cards, other) | 1993 | Present |
boe_money_supply() |
M4 broad money amounts outstanding | 1982 | Present |
Monetary policy:
| Function | Description | From | To |
|---|---|---|---|
boe_mpc_decisions() |
MPC rate-change events: date, new rate, change in bps, direction | 1997 | Present |
boe_mpc_votes() |
Full MPC voting record, one row per (meeting, member), with dissent flag | 1997 | Present |
boe_mpr_forecasts() |
Monetary Policy Report forecast paths (CPI inflation, GDP growth, GDP level, unemployment, Bank Rate) | 2019 | Present |
Discovery:
| Function | Description |
|---|---|
boe_series |
Exported catalogue of every wrapped series (code, title, category, frequency, unit, start date) |
boe_search() |
Keyword search over boe_series |
boe_browse() |
Filter boe_series by category or frequency |
list_exchange_rates() |
Currency codes available to boe_exchange_rate() |
Cache:
| Function | Description |
|---|---|
boe_cache_info() |
Report cache directory, file count, total size |
clear_cache() |
Delete locally cached data files |
library(boe)
# Bank Rate since 2000
br <- boe_bank_rate(from = "2000-01-01")
tail(br, 6)
#> date rate_pct
#> 2026-02-26 3.75
#> 2026-02-27 3.75
#> 2026-03-02 3.75
#> 2026-03-03 3.75
#> 2026-03-04 3.75
#> 2026-03-05 3.75# GBP/USD and GBP/EUR
fx <- boe_exchange_rate(c("USD", "EUR"), from = "2024-01-01", to = "2024-01-31")
head(fx, 6)
#> date currency rate
#> 2024-01-02 EUR 1.1536
#> 2024-01-03 EUR 1.1580
#> 2024-01-04 EUR 1.1591
#> 2024-01-05 EUR 1.1615
#> 2024-01-08 EUR 1.1623
#> 2024-01-09 EUR 1.1636
# See all 27 available currencies
list_exchange_rates()# 10-year nominal gilt yield
yc <- boe_yield_curve(from = "2024-01-01", to = "2024-01-31", maturity = "10yr")
head(yc, 5)
#> date maturity yield_pct
#> 2024-01-02 10yr 3.7190
#> 2024-01-03 10yr 3.7638
#> 2024-01-04 10yr 3.8006
#> 2024-01-05 10yr 3.8398
#> 2024-01-08 10yr 3.8619
# Full curve: 5yr, 10yr, and 20yr
boe_yield_curve(from = "2024-01-01")
# Real yields
boe_yield_curve(from = "2024-01-01", type = "real", measure = "zero_coupon")For the complete yield curve at every published maturity (typically 0.5 years to 25 or 40 years, in 0.5-year steps), use boe_curve(). This parses the BoE's published Excel archive and covers four curves: nominal gilt, real (index-linked) gilt, implied inflation (breakeven), and overnight index swap (OIS).
# Latest nominal spot curve at all maturities
nc <- boe_curve(curve = "nominal", measure = "spot")
head(nc, 6)
#> date maturity_years rate_pct
#> 2026-04-01 0.5 3.95
#> 2026-04-01 1.0 4.10
#> 2026-04-01 1.5 4.13
#> 2026-04-01 2.0 4.15
#> 2026-04-01 2.5 4.16
#> 2026-04-01 3.0 4.17
# Implied inflation curve (breakeven inflation)
boe_curve(curve = "inflation", measure = "spot")
# OIS forward curve
boe_curve(curve = "ois", measure = "spot")Requires the readxl package (loaded lazily). Reference: Anderson and Sleath (2001), New estimates of the UK real and nominal yield curves, Bank of England Working Paper No. 126.
# All mortgage rate types
mr <- boe_mortgage_rates(from = "2023-01-01")
# Latest rates (as of December 2024)
#> 2yr_fixed: 4.60%
#> 3yr_fixed: 4.48%
#> 5yr_fixed: 4.37%
#> svr: 7.47%# Monthly mortgage approvals - a leading indicator of housing activity
ma <- boe_mortgage_approvals(from = "2019-01-01")
tail(ma, 6)
#> date approvals
#> 2025-08-31 64588
#> 2025-09-30 65436
#> 2025-10-31 64634
#> 2025-11-30 64018
#> 2025-12-31 61007
#> 2026-01-31 59999# Total consumer credit outstanding
cc <- boe_consumer_credit(type = "total", from = "2024-01-01")
tail(cc, 6)
#> date type amount_gbp_m
#> 2024-01-31 total 476154
#> 2024-02-29 total 479974
#> 2024-03-31 total 484269
#> 2024-04-30 total 490106
#> 2024-05-31 total 494904
#> 2024-06-30 total 498639
# Credit card debt only
boe_consumer_credit(type = "credit_card", from = "2024-01-01")# M4 amounts outstanding
m4 <- boe_money_supply(from = "2024-01-01")
head(m4, 6)
#> date amount_gbp_m
#> 2024-01-31 2986264
#> 2024-02-29 2999033
#> 2024-03-31 3025146
#> 2024-04-30 3030412
#> 2024-05-31 3028825
#> 2024-06-30 3044464 # ← £3 trillion# SONIA replaced LIBOR as the UK's benchmark interest rate
sonia <- boe_sonia(from = "2024-01-01", to = "2024-01-31")
head(sonia, 6)
#> date rate_pct
#> 2024-01-02 5.1863
#> 2024-01-03 5.1863
#> 2024-01-04 5.1870
#> 2024-01-05 5.1869
#> 2024-01-08 5.1869
#> 2024-01-09 5.1867
# Monthly or annual average
boe_sonia(from = "2020-01-01", frequency = "monthly")# If you know the BoE series code, use boe_get() directly
# Series codes: https://www.bankofengland.co.uk/boeapps/database/
# Multiple series in one call - Bank Rate vs SONIA
boe_get(c("IUDBEDR", "IUDSOIA"), from = "2024-01-01", to = "2024-01-10")
#> date code value
#> 2024-01-02 IUDBEDR 5.2500
#> 2024-01-03 IUDBEDR 5.2500
#> 2024-01-04 IUDBEDR 5.2500
#> ...
#> 2024-01-02 IUDSOIA 5.1863
#> 2024-01-03 IUDSOIA 5.1863
#> 2024-01-04 IUDSOIA 5.1870
#> ...# Every Bank Rate change since 1997
decisions <- boe_mpc_decisions()
tail(decisions, 5)
#> date new_rate_pct prev_rate_pct change_bps direction
#> 2024-08-01 5.00 5.25 -25 cut
#> 2024-11-07 4.75 5.00 -25 cut
#> 2025-02-06 4.50 4.75 -25 cut
#> 2025-08-07 4.25 4.50 -25 cut
#> 2026-02-05 4.00 4.25 -25 cut
# Full voting record: who dissented, and how
votes <- boe_mpc_votes()
recent_dissents <- subset(votes, dissent & date >= as.Date("2024-01-01"))
head(recent_dissents)
# How does Catherine L Mann vote?
mann <- subset(votes, member == "Catherine L Mann")
table(mann$dissent)# Latest CPI inflation projections (one row per publication x horizon)
cpi <- boe_mpr_forecasts(series = "cpi_inflation")
head(cpi)
#> date horizon horizon_date series value
#> 2026-02-01 2026 Q1 2026-01-01 cpi_inflation 2.7
#> 2026-02-01 2026 Q2 2026-04-01 cpi_inflation 2.6
#> 2026-02-01 2026 Q3 2026-07-01 cpi_inflation 2.5
# All five headline series for the most recent MPR
all <- boe_mpr_forecasts()
unique(all$series)
#> [1] "bank_rate" "cpi_inflation" "gdp_growth" "gdp_level" "unemployment"Requires the readxl package. Note: this targets the post-2025 MPR file format; older releases use a different archive layout.
# Keyword search across the catalogue
boe_search("mortgage")
# Filter by category and frequency
boe_search(category = "interest_rates", frequency = "daily")
# Browse without a keyword
boe_browse(category = "exchange_rates")
# The full catalogue is exported as a data frame
head(boe_series)
table(boe_series$category)
#> consumer_credit exchange_rates interest_rates
#> 3 27 14
#> monetary_aggregates mortgage_market
#> 2 6Every result from a boe_*() function is a boe_tbl (a data frame with attached metadata). Printing shows a one-line provenance header, but it behaves like a normal data frame for everything else.
br <- boe_bank_rate(from = "2024-01-01", frequency = "monthly")
br
#> # BoE [boe_bank_rate]: 1 series [IUMABEDR] · 16 obs · 2024-01-01 to 2025-04-30 · freq=monthly
#> date rate_pct
#> 2024-01-31 5.25
#> 2024-02-29 5.25
#> ...All downloads are cached locally in your user cache directory. Subsequent calls return the cached copy instantly - no network request is made.
# Inspect the cache (path, file count, size, range)
boe_cache_info()
#> BoE cache
#> * Path: /Users/.../R/boe/cache
#> * Files: 12
#> * Size: 6.4 MB
#> * Range: 2026-04-12 09:14:02 to 2026-04-25 11:30:18
# Force a fresh download
boe_bank_rate(from = "2020-01-01", cache = FALSE)
# Remove files older than 7 days
clear_cache(max_age_days = 7)
# Remove all cached files
clear_cache()| Package | Description |
|---|---|
ons |
UK Office for National Statistics data |
hmrc |
HM Revenue & Customs tax data |
obr |
Office for Budget Responsibility fiscal forecasts |
fred |
US Federal Reserve (FRED) data |
readecb |
European Central Bank data |
yieldcurves |
Yield curve fitting (Nelson-Siegel, Svensson) |
mpshock |
Monetary policy shock series (US/UK/AU) |
inflationkit |
Inflation analysis (decomposition, persistence, Phillips curve) |
nowcast |
Economic nowcasting (bridge, MIDAS, DFM) |
Please report bugs or requests at https://github.com/charlescoverdale/boe/issues.
Bank of England, BoE, interest rates, bank rate, SONIA, yield curve, exchange rates, mortgage rates, consumer credit, money supply, monetary policy, UK economic data, R package