index.qmd

---
title: ""
execute:
  echo: false

---

```{r}
#| include: false

library(tidyverse)
library(lubridate)
library(hrbrthemes)
library(cowplot)
library(ggtext)
library(realtalk)
library(assertr)

# necessary for cowplot
set_null_device("png")
```


```{r}
# raw datasets
nominal_min_wage_data <- us_minimum_wage_monthly %>% 
  transmute(
    month_date = ym(paste(year, month)), 
    nominal_fed_mw = minimum_wage
  ) 

cpi_data <- c_cpi_u_extended_monthly_sa %>%
  rename(cpi_sa = c_cpi_u_extended) %>% 
  full_join(c_cpi_u_extended_monthly_nsa, by = join_by(year, month)) %>% 
  rename(cpi_nsa = c_cpi_u_extended) %>% 
  transmute(
    month_date = ym(paste(year, month)),
    cpi = if_else(year < 1947, cpi_nsa, cpi_sa)
  )

# parameters
# current_cpi_date
current_cpi_date <- cpi_data %>% 
  filter(month_date == max(month_date)) %>% 
  pull(month_date)

# cpi base
cpi_base <- cpi_data %>% 
  filter(month_date == current_cpi_date) %>% 
  pull(cpi)

# inflation-adjusted data
real_min_wage_data <- nominal_min_wage_data %>% 
  filter(month_date <= current_cpi_date) %>% 
  inner_join(cpi_data, by = "month_date") %>% 
  mutate(real_fed_mw = nominal_fed_mw * cpi_base / cpi) %>% 
  select(month_date, real_fed_mw, nominal_fed_mw, cpi)

real_min_wage_data %>% 
  mutate(
    across(matches("mw"), ~ round(.x, 2)),
    month = month(month_date),
    year = year(month_date),
  ) %>% 
  select(
    year, 
    month, 
    nominal_federal_mw = nominal_fed_mw, 
    real_federal_mw = real_fed_mw,
    extended_chained_cpi = cpi,
  ) %>% 
  write_csv("real_federal_minimum_wage.csv")
```


```{r}
# create data for figures

# for figure we usually label current date as month after cpi date
current_cpi_date_label <- format(current_cpi_date + months(1), "%b. %Y")

# current fed mw
current_fed_mw <- real_min_wage_data %>% 
  filter(month_date == current_cpi_date) %>% 
  pull(nominal_fed_mw)

# date break data
date_breaks <- tribble(
  ~month_date, ~date_label, ~date_hjust, ~date_description,
  ym("1938m10"), "Oct. 1938", 0, NA,
  ym("1968m2"), "Feb. 1968", 0.5, "Peak value", 
  ym("2009m7"), "July 2009", 0.5, "Last increase",
  current_cpi_date, current_cpi_date_label, 1, "Today"
)

line_graph_data <- real_min_wage_data %>% 
  full_join(date_breaks, by = "month_date") %>% 
  mutate(
    date_break = if_else(!is.na(date_label), month_date, NA),
    date_value = if_else(!is.na(date_label), real_fed_mw, NA),
    date_value = round(date_value, 2),
    date_value_string = if_else(
      !is.na(date_value),
      scales::label_dollar(accuracy = 0.01)(date_value),
      ""
    )
  )

bar_graph_data <- line_graph_data %>% 
  # grab only relevant dates for bar graph
  filter(!is.na(date_description)) %>% 
  mutate(date_year = year(month_date)) %>% 
  mutate(date_description = case_when(
    date_year == 1968 ~ "Peak value of minimum wage",
    date_year == 2009 ~ "Last minimum wage increase",
    date_year == year(current_cpi_date) ~ "Today"
  )) %>% 
  mutate(date_value_string = case_when(
    date_year == 1968 ~ paste0(date_value_string, "/hr"),
    TRUE ~ date_value_string
  )) %>% 
  mutate(
    annual_value_string = prettyNum(round(date_value*2080), big.mark=","),
    date_value_string = paste0("**", date_value_string, "**"),
    date_value_string = paste0(date_value_string, " ($", annual_value_string, ")")
  ) %>% 
  arrange(month_date) %>% 
  mutate(display_order = 4-row_number())

value_peak <- bar_graph_data %>% 
  filter(year(month_date) == 1968) %>% 
  pull(real_fed_mw)
decline_since_peak <- abs(current_fed_mw / value_peak - 1) * 100
decline_since_peak <- paste0(round(decline_since_peak), "%")

value_last <- bar_graph_data %>% 
  filter(year(month_date) == 2009) %>% 
  pull(real_fed_mw)
decline_since_last <- abs(current_fed_mw / value_last - 1) * 100
decline_since_last <- paste0(round(decline_since_last), "%")
```


```{r}
my_theme <- function(){
  theme_minimal() +
  theme(
        plot.margin = margin(
          t = 5, 
          r = 5,
          b = 5, 
          l = 5, 
          unit = "pt"),
        axis.title.x = element_blank(),
        axis.title.y = element_blank(),
        legend.position = "none"
  )
}

p1 <- line_graph_data %>% 
  ggplot() +
  geom_segment(
    aes(
      x = date_break, 
      xend = date_break, 
      y = 3.75, 
      yend = 14
    ),
    color = "grey80",
    linetype = "dashed",
    na.rm = TRUE
  ) +
  geom_line(
    aes(x = month_date, y = real_fed_mw), 
    data = filter(line_graph_data, month_date <= ym("2009m7")),
    linewidth = 1.5,
    color = "#1273af",
    lineend = "round"
  ) +
  geom_line(
    aes(x = month_date, y = real_fed_mw), 
    data = filter(line_graph_data, month_date >= ym("2009m7")),
    linewidth = 1.5,
    color = "#c01f41",
    lineend = "round"
  ) +
  my_theme() +
  scale_x_date(
    breaks = NULL,
    minor_breaks = NULL
  ) +
  scale_y_continuous(
    breaks = c(6, 8, 10, 12),
    minor_breaks = NULL,
    labels = scales::label_dollar()
  ) +
  coord_cartesian(
    xlim = c(ym("1938m1"), my(current_cpi_date_label)), 
    ylim = c(3, 13.75),
    expand = FALSE
  ) +
  geom_text(
    aes(
      x = month_date, 
      y = 3.5, 
      label = date_label, 
      hjust = date_hjust
    ),
    na.rm = TRUE,
    color = "grey40",
  ) +
  geom_text(
    aes(
      x = month_date, 
      y = 13.5, 
      label = date_value_string, 
    ),
    hjust = 1,
    nudge_x = -300,
    na.rm = TRUE,
    color = "#2b2b2b",
    fontface = "bold"
  ) +
  geom_text(
    aes(
      x = month_date, 
      y = 13, 
      label = date_description
    ),
    hjust = 1,
    nudge_x = -300,
    na.rm = TRUE,
    color = "grey40",
  ) +
  annotate(
    "text", 
    x = ym("1938m10") + 300, 
    y = 13.5, 
    label = "Minimum wage",
    color = "grey40",
    hjust = 0
  ) + 
  annotate(
    "text", 
    x = ym("1938m10") + 300, 
    y = 13, 
    label = "established",
    color = "grey40",
    hjust = 0
  )
```

```{r}
p2 <- bar_graph_data %>% 
  ggplot(aes(x = display_order, y = date_value, fill = factor(display_order))) +
  geom_col(width = 0.5) + 
  my_theme() +
  theme(plot.margin = margin(t = 20, r = 5, l = 25, b = 5, unit = "pt")) +
  coord_flip() +
  scale_fill_manual(values=c("#c01f41", "#1273af", "#1273af")) +
  scale_x_continuous (
    breaks = NULL,
    minor_breaks = NULL
  ) +
  scale_y_continuous(
    expand = c(0,0),
    breaks = NULL,
    minor_breaks = NULL
  ) +
  geom_richtext(
    aes(
      x = display_order, 
      y = date_value, 
      label = date_value_string, 
    ),
    size = 5,
    hjust = 1,
    vjust = 0.57,
    nudge_y = -0.1,
    color = "#ffffff",
    label.color = NA
  ) +
  geom_text(
    aes(
      x = display_order, 
      y = 0, 
      label = date_year, 
    ),
    size = 5,
    hjust = 0,
    vjust = 0.57,
    nudge_x = 0.45,
    fontface = "bold"
  ) +
  geom_text(
    aes(
      x = display_order, 
      y = 0, 
      label = date_description, 
    ),
    size = 5,
    hjust = 0,
    vjust = 0.57,
    nudge_x = 0.45,
    nudge_y = 0.90,
    color = "grey40"
  )
```


### After the longest period in history without an increase, the federal minimum wage today is worth `r decline_since_last` less than 15 years ago--and `r decline_since_peak` less than in 1968 
#### <b>Real value of the minimum wage (adjusted for inflation)</b>

```{r fig.height=7, warning=FALSE}
plots <- plot_grid(p1, p2, nrow = 2, rel_heights = c(3,2))
plots
```
<small>All values in `r format(current_cpi_date, format = "%B %Y")` dollars, adjusted using the chained CPI-U since 2000, extended retroactively to the CPI-U-RS (1978-1999), CPI-U-X1 (1967–1977), and CPI-U (1966 and before). Calculations by Ben Zipperer, Economic Policy Institute. Download the [data](real_federal_minimum_wage.csv) or view the [code](https://github.com/Economic/real_minimum_wage) that produces this figure.</small>