House Sales

time series

A series of events, such as the Financial Crisis and the 2016 Brexit vote, that damped down residential property sales in London

Author

Carl Goodwin

Published

December 17, 2017

Modified

February 24, 2023

Various events have impacted house sales in London. There has been a series of increases in stamp duty and the impact of the financial crisis. More recently Brexit and the consequences of Covid-19.

How is London postal area SW10 coping with all this?

library(conflicted)
library(tidyverse)
conflict_prefer_all("dplyr", quiet = TRUE)
conflict_prefer("as_date", "lubridate")
library(scales)
library(SPARQL)
library(clock)
conflict_prefer("date_format", "clock")
library(wesanderson)
library(glue)
library(tsibble)
library(patchwork)
library(ggmosaic)
library(usedthese)

conflict_scout()

theme_set(theme_bw())

(cols <- wes_palette(name = "Darjeeling1"))

House prices paid data are provided by HM Land Registry Open Data.

endpoint <- "https://landregistry.data.gov.uk/landregistry/query"

query <- 'PREFIX  text: <http://jena.apache.org/text#>
PREFIX  ppd:  <http://landregistry.data.gov.uk/def/ppi/>
PREFIX  lrcommon: <http://landregistry.data.gov.uk/def/common/>
  
SELECT  ?item ?ppd_propertyAddress ?ppd_hasTransaction ?ppd_pricePaid ?ppd_transactionCategory ?ppd_transactionDate ?ppd_transactionId ?ppd_estateType ?ppd_newBuild ?ppd_propertyAddressCounty ?ppd_propertyAddressDistrict ?ppd_propertyAddressLocality ?ppd_propertyAddressPaon ?ppd_propertyAddressPostcode ?ppd_propertyAddressSaon ?ppd_propertyAddressStreet ?ppd_propertyAddressTown ?ppd_propertyType ?ppd_recordStatus

WHERE
{ ?ppd_propertyAddress text:query _:b0 .
  _:b0 <http://www.w3.org/1999/02/22-rdf-syntax-ns#first> lrcommon:postcode .
  _:b0 <http://www.w3.org/1999/02/22-rdf-syntax-ns#rest> _:b1 .
  _:b1 <http://www.w3.org/1999/02/22-rdf-syntax-ns#first> "( SW10 )" .
  _:b1 <http://www.w3.org/1999/02/22-rdf-syntax-ns#rest> _:b2 .
  _:b2 <http://www.w3.org/1999/02/22-rdf-syntax-ns#first> 3000000 .
  _:b2 <http://www.w3.org/1999/02/22-rdf-syntax-ns#rest> <http://www.w3.org/1999/02/22-rdf-syntax-ns#nil> .
  ?item ppd:propertyAddress ?ppd_propertyAddress .
  ?item ppd:hasTransaction ?ppd_hasTransaction .
  ?item ppd:pricePaid ?ppd_pricePaid .
  ?item ppd:transactionCategory ?ppd_transactionCategory .
  ?item ppd:transactionDate ?ppd_transactionDate .
  ?item ppd:transactionId ?ppd_transactionId
  
  OPTIONAL { ?item ppd:estateType ?ppd_estateType }
  OPTIONAL { ?item ppd:newBuild ?ppd_newBuild }
  OPTIONAL { ?ppd_propertyAddress lrcommon:county ?ppd_propertyAddressCounty }
  OPTIONAL { ?ppd_propertyAddress lrcommon:district ?ppd_propertyAddressDistrict }
  OPTIONAL { ?ppd_propertyAddress lrcommon:locality ?ppd_propertyAddressLocality }
  OPTIONAL { ?ppd_propertyAddress lrcommon:paon ?ppd_propertyAddressPaon }
  OPTIONAL { ?ppd_propertyAddress lrcommon:postcode ?ppd_propertyAddressPostcode }
  OPTIONAL { ?ppd_propertyAddress lrcommon:saon ?ppd_propertyAddressSaon }
  OPTIONAL { ?ppd_propertyAddress lrcommon:street ?ppd_propertyAddressStreet }
  OPTIONAL { ?ppd_propertyAddress lrcommon:town ?ppd_propertyAddressTown }
  OPTIONAL { ?item ppd:propertyType ?ppd_propertyType }
  OPTIONAL { ?item ppd:recordStatus ?ppd_recordStatus }
}'

data_lst <- SPARQL(endpoint, query)

The focus is on the standard price paid.

data_df <- data_lst$results |>
  as_tibble() |>
  mutate(
    trans_date = as_datetime(ppd_transactionDate) |> as_date(),
    amount = ppd_pricePaid,
    prop_type = str_extract(ppd_propertyType, "(?<=common/)[\\w]+"),
    est_type = str_extract(ppd_estateType, "(?<=common/)[\\w]+"),
    cat = str_remove(ppd_transactionCategory, "<http://landregistry.data.gov.uk/def/ppi/"),
    prop_type = recode(prop_type, otherPropertyType = "Other")
  ) |>
  filter(str_detect(cat, "standard"))

A Telegraph article entitled Timeline: 20 years of stamp duty increases for home buyers pinpoints many of the key event dates.

events <- tribble(
  ~event_date, ~change,
  "96-07-31", "Stamp Duty £250k (1.5%) £500k (2%)",
  "98-03-31", "£250k (2%) £500k (3%)",
  "99-03-31", "£250k (2.5%) £500k (3.5%)",
  "00-03-31", "£250k (3%) £500k (4%)",
  "11-04-30", "£250k (3%) £500k (4%) £1m (5%)",
  "12-03-31", "£250k (3%) £500k (4%) £1m (5%) £2m (7%)",
  "14-12-31", "£250k (5%) £925k (10%) 1.5m (12%)",
  "07-08-09", "Financial Crisis",
  "16-06-23", "Brexit Vote",
  "20-03-23", "Covid-19 Lockdown"
) |>
  mutate(event_date = date_parse(event_date, format = "%y-%m-%d"))

events |> 
  rename("Date" = event_date, "Event" = change)

Date	Event
1996-07-31	Stamp Duty £250k (1.5%) £500k (2%)
1998-03-31	£250k (2%) £500k (3%)
1999-03-31	£250k (2.5%) £500k (3.5%)
2000-03-31	£250k (3%) £500k (4%)
2011-04-30	£250k (3%) £500k (4%) £1m (5%)
2012-03-31	£250k (3%) £500k (4%) £1m (5%) £2m (7%)
2014-12-31	£250k (5%) £925k (10%) 1.5m (12%)
2007-08-09	Financial Crisis
2016-06-23	Brexit Vote
2020-03-23	Covid-19 Lockdown

Visually, it appears that the financial crisis had a big impact on sales volume, with the Brexit vote sucking much of the remaining oxygen out of the market. Stamp duty increases in between probably slowed any intermediate recovery.

to_date <- data_df |> 
  summarise(max(trans_date)) |> 
  pull() |> 
  date_format(format = "%b %d, %Y")

data_df |>
  ggplot(aes(trans_date, amount, colour = est_type)) +
  geom_point(alpha = 0.2, size = 0.7, show.legend = FALSE) +
  geom_smooth(se = FALSE, aes(linetype = est_type), size = 1.2) +
  labs(
    title = "SW10 Standard House Prices",
    subtitle = glue("Prices Paid to {to_date} (Prices > £5m Not Shown)"
    ),
    x = NULL,
    y = NULL,
    colour = "Type", linetype = "Type",
    caption = "Source: HM Land Registry"
  ) +
  geom_vline(xintercept = events$event_date, 
             size = 0.5, lty = 2, alpha = 0.4) +
  annotate("text", events$event_date, 5000000,
    angle = 90,
    label = events$change, vjust = 1.4, hjust = 1, size = 3, fontface = 2
  ) +
  coord_cartesian(ylim = c(0, 5000000)) +
  scale_colour_manual(values = cols[c(2, 3)]) +
  scale_x_date(date_breaks = "2 years", date_labels = "%Y") +
  scale_y_continuous(labels = label_dollar(accuracy = 0.1, prefix = "£", 
                                           scale_cut = cut_short_scale()))

An alternative way of looking at this is by median quarterly prices (with upper and lower quartiles), supplemented by sales volumes.

qtr_start <- date_today("Europe/London") |> 
  as_year_quarter_day() |> 
  calendar_start("quarter") |> 
  as_date()

qtile_df <- 
  data_df |> 
  filter(trans_date < qtr_start) |> 
  mutate(yr_qtr = yearquarter(trans_date)) |> 
  reframe(price = quantile(amount, c(0.25, 0.5, 0.75)), 
            quantile = c("lower", "median", "upper") |> factor(),
            n = n(),
            .by = yr_qtr) |> 
  pivot_wider(names_from = quantile, values_from = price)

last <- qtile_df |> summarise(max(yr_qtr)) |> pull()
first <- qtile_df |> summarise(min(yr_qtr)) |> pull()

p1 <- qtile_df |> 
  ggplot(aes(yr_qtr, median)) +
  geom_ribbon(aes(ymin = lower, ymax = upper), fill = cols[5]) +
  geom_line(colour = "white") +
  geom_hline(yintercept = 1000000, linetype = "dashed") +
  annotate("text", x = yearquarter("2020 Q2"), y = 300000, 
           label = "Covid-19\nLockdown", size = 3) +
  scale_x_yearquarter(date_breaks = "2 years") +
  scale_y_log10(labels = label_dollar(prefix = "£", 
                                      scale_cut = cut_short_scale())) +
  labs(
    title = glue("Median Quarterly SW10 Property Prices ({first} to {last})"), 
    subtitle = "With Upper / Lower Price Quartiles & Sales Volume",
    x = NULL, y = "Price (Log10)"
    ) +
  theme(axis.text.x = element_blank())

p2 <- qtile_df |> 
  ggplot(aes(yr_qtr, n)) +
  geom_line() +
  annotate("text", x = yearquarter("2007 Q3"), y = 180, 
           label = "Financial\nCrisis", size = 3) +
  annotate("text", x = yearquarter("2016 Q3"), y = 130, 
           label = "Brexit\nVote", size = 3) +
  scale_x_yearquarter(date_breaks = "2 years") +
  labs(x = NULL, y = "Transactions",
       caption = "Source: HM Land Registry") +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

p1 / p2 + plot_layout(heights = c(2, 1))

A ggmosaic (Jeppson, Hofmann, and Cook 2021) visualisation of the composition of SW10 reveals the postal area to be overwhelmingly dominated by leasehold flats.

num_trans <- data_df |> nrow()

data_df |> 
  ggplot() +
  geom_mosaic(aes(product(prop_type, est_type), fill = prop_type), 
              offset = 0.02, divider = mosaic("h")) +
  scale_fill_manual(values = cols[c(2:5)]) + 
  labs(
    title = "SW10 Transactions by Estate & Property Types",
    subtitle = glue("{comma(num_trans)} Transactions to {to_date}"),
    x = "", y = "", fill = "Property Type", 
    caption = "Source: HM Land Registry"
  ) +
  theme_minimal()

Other blog posts on quantum jitter look at SW10 property from different perspectives: Digging Deep considers the correlation between house sales and planning applications; and Bootstraps & Bandings uses a sample of recent house sales to infer whether property bands are as representative of property values today as they were three decades ago.

R Toolbox

Summarising below the packages and functions used in this post enables me to separately create a toolbox visualisation summarising the usage of packages and functions across all posts.

used_here()

Package	Function
SPARQL	SPARQL[1]
base	c[6], factor[1], library[11], max[2], min[1], nrow[1]
clock	as_year_quarter_day[1], calendar_start[1], date_format[1], date_parse[1], date_today[1]
conflicted	conflict_prefer[2], conflict_prefer_all[1], conflict_scout[1]
dplyr	filter[2], mutate[3], n[1], pull[3], recode[1], reframe[1], rename[1], summarise[3]
ggmosaic	geom_mosaic[1], mosaic[1], product[1]
ggplot2	aes[6], annotate[4], coord_cartesian[1], element_blank[1], element_text[1], geom_hline[1], geom_line[2], geom_point[1], geom_ribbon[1], geom_smooth[1], geom_vline[1], ggplot[4], labs[4], scale_colour_manual[1], scale_fill_manual[1], scale_x_date[1], scale_y_continuous[1], scale_y_log10[1], theme[2], theme_bw[1], theme_minimal[1], theme_set[1]
glue	glue[3]
lubridate	as_date[2], as_datetime[1]
patchwork	plot_layout[1]
scales	comma[1], cut_short_scale[2], label_dollar[2]
stats	quantile[1]
stringr	str_detect[1], str_extract[2], str_remove[1]
tibble	as_tibble[1], tribble[1]
tidyr	pivot_wider[1]
tsibble	scale_x_yearquarter[2], yearquarter[4]
usedthese	used_here[1]
wesanderson	wes_palette[1]

References

Jeppson, Haley, Heike Hofmann, and Di Cook. 2021. “Ggmosaic: Mosaic Plots in the ’Ggplot2’ Framework.” https://CRAN.R-project.org/package=ggmosaic.