R/viz_oecd_minwage.R

In MattCowgill/macrodashboard: Shiny dashboard for Australian macroeconomic data

viz_oecd_minwage <- function(data = load_data(),
                             arg1 = c("mean", "median"),
                             arg2 = TRUE) {
  df <- data$oecd_min2ave
  .series <- match.arg(arg1)
  .show_range <- arg2

  df <- df %>%
    mutate(series = tolower(.data$series)) %>%
    filter(
      .data$series == .series,
      .data$time >= 1985
    )

  if (.show_range == TRUE) {
    non_aus <- df %>%
      filter(.data$country != "Australia") %>%
      group_by(.data$time) %>%
      summarise(
        min = min(.data$obs_value),
        max = max(.data$obs_value)
      )

    chart_data <- df %>%
      filter(.data$country == "Australia") %>%
      select(.data$time, aus = .data$obs_value) %>%
      left_join(non_aus, by = "time")

    base_chart <- chart_data %>%
      ggplot(aes(x = .data$time)) +
      geom_ribbon(aes(ymin = .data$min, ymax = .data$max),
        fill = grattantheme::grattan_grey3,
        alpha = 0.33,
        size = 0
      ) +
      geom_line(aes(y = .data$aus)) +
      grattan_label_repel(
        data = ~ filter(., time == max(.data$time)) %>%
          mutate(mid = (.data$min + .data$max) / 2.5) %>%
          select(time,
            "Australia" = .data$aus,
            "Other OECD countries" = .data$mid
          ) %>%
          gather(
            key = "key", value = "value",
            -.data$time
          ),
        aes(
          y = .data$value,
          label = stringr::str_wrap(.data$key, 5),
          col = .data$key
        ),
        hjust = 0,
        direction = "y",
        segment.size = 0,
        size = 14,
        nudge_x = 0.2
      ) +
      scale_colour_manual(values = c(
        "Australia" = grattantheme::grattan_lightorange,
        "Other OECD countries" = grattantheme::grattan_grey3
      )) +
      theme_grattan()
  } else {
    main_countries <- c(
      "Australia" = grattantheme::grattan_lightorange,
      "Colombia" = "black",
      "New Zealand" = grattantheme::grattan_darkorange,
      "Canada" = grattantheme::grattan_red,
      "France" = grattantheme::grattan_lightyellow,
      "United Kingdom" = grattantheme::grattan_darkred,
      "United States" = grattantheme::grattan_yellow
    )

    other_countries <- unique(df$country)[!unique(df$country) %in% names(main_countries)]

    other_countries <- rep(grattantheme::grattan_grey2, length(other_countries)) %>%
      purrr::set_names(other_countries)

    cols <- c(main_countries, other_countries)

    base_chart <- df %>%
      ggplot(aes(
        x = .data$time, y = .data$obs_value,
        group = .data$country, col = .data$country
      )) +
      geom_line() +
      geom_line(data = ~ filter(., .data$country %in% names(main_countries))) +
      geom_point(
        data = ~ filter(
          .,
          .data$country %in% names(main_countries),
          .data$time == max(.data$time)
        ),
        fill = "white",
        size = 3, stroke = 1.5, shape = 21
      ) +
      grattan_label_repel(
        data = ~ filter(
          .,
          .data$country %in% c(
            names(main_countries),
            "Ireland"
          ),
          .data$time == max(.data$time)
        ) %>%
          mutate(label = ifelse(.data$country == "Ireland", "Other OECD countries", .data$country)),
        aes(label = stringr::str_wrap(.data$label, 7)),
        size = 14,
        hjust = 0,
        nudge_x = 0.33,
        segment.size = 0,
        direction = "y"
      ) +
      scale_colour_manual(values = cols) +
      theme_grattan()
  }


  base_chart +
    grattan_y_continuous(
      limits = c(0, ceiling(max(df$obs_value) / .2) / 5),
      labels = function(x) paste0(x * 100, "%")
    ) +
    labs(
      title = "Australia's minimum wage has been drifting towards the middle of the OECD pack",
      subtitle = paste0(
        "Minimum wage as a percentage of ",
        ifelse(.series == "mean", "average", "median"),
        " wage in OECD countries"
      ),
      caption = "Source: OECD.Stat"
    ) +
    theme(
      axis.title = element_blank(),
      axis.title.x = element_blank()
    ) +
    scale_x_continuous(expand = expansion(mult = c(0, .12)))
}