Facing another U.S. government shutdown, I came across an article on Vox.com describing the 18 previous shutdowns. The article stated that “… they’re a relatively recent development … the result of the Congressional Budget Act of 1974.”
I was curious to visualize the timing and the duration of the shutdowns. Happily, the beginning and ending dates were reported in the article. I hand entered the data (it seemed faster than scraping). In terms of writing an R script, the only trick was creating date variables (plot.From and plot.To) for plotting that used a single fiscal year (FY2001).
library(tidyverse)
library(scales)
df <- data.frame(
Shut = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18),
mm1 = c(9, 9, 10, 11, 9, 9, 11, 9, 12, 11, 9, 10, 10, 12, 10, 11, 12, 10),
dd1 = c(30, 30, 31, 30, 30, 30, 20, 30, 17, 10, 30, 3, 16, 18, 5, 13, 5, 1),
yyyy1 = c(1976, 1977, 1977, 1977, 1978, 1979, 1981, 1982, 1982, 1983, 1984,
1984, 1986, 1987, 1990, 1995, 1995, 2013),
mm2 = c(10, 10, 11, 12, 10, 10, 11, 10, 12, 11, 10, 10, 10, 12, 10, 11, 1,
10),
dd2 = c(11, 13, 9, 9, 18, 12, 23, 2, 21, 14, 3, 5, 18, 20, 9, 19, 6, 17),
yyyy2 = c(1976, 1977, 1977, 1977, 1978, 1979, 1981, 1982, 1982, 1983, 1984,
1984, 1986, 1987, 1990, 1995, 1996, 2013)
) %>%
mutate(
From = as.Date(paste(yyyy1, mm1, dd1, sep="-")),
To = as.Date(paste(yyyy2, mm2, dd2, sep="-")),
Duration = as.numeric(To - From),
FY = ifelse(mm2 > 8, yyyy2+1, yyyy2),
# put From and To on a single calendar year for plotting
Plot.year1 = ifelse(mm1 > 8, 2000, 2001),
Plot.year2 = ifelse(mm2 > 8, 2000, 2001),
plot.From = as.Date(paste(Plot.year1, mm1, dd1, sep="-")),
plot.To = as.Date(paste(Plot.year2, mm2, dd2, sep="-"))
)
ggplot(df, aes(x=plot.From, xend=plot.To, y=FY, yend=FY)) +
geom_segment(color="#85bb65", size=2) +
geom_point() +
geom_point(aes(x=plot.To)) +
scale_x_date(labels=date_format("%b %d")) +
labs(x="Date", y="Fiscal year",
title="Timing of Past US Gov't Shutdowns") +
theme_bw(base_size=14)
There have been 18 shutdowns so far (not including this latest one), but sometimes there was more than one shutdown in a year. There have been shutdowns in 13 fiscal years since 1974; that’s an average of 1 shutdown every 3.4 years. The average total duration of the shutdowns in a given fiscal year was 8.4 days.
df %>%
group_by(FY) %>%
summarise(ndays = sum(Duration)) %>%
ggplot(aes(FY, ndays)) +
geom_bar(stat="identity", fill="#85bb65", color="black") +
labs(x="Fiscal year", y="Days",
title="Duration of Past US Gov't Shutdowns") +
theme_bw(base_size=14)