Replicating a specific plot - ggplot axis not as they should be

Question

I have some data and I am trying to re-create a plot. However I cannot seem to get the axis aligned.

I am trying to plot something very similar to the following:

A bar and a line plot with two different axis. However my attempt does not seem to work:

ggplot(df, aes(x = years)) +
  geom_col(aes( y = IPOs_sum, fill="redfill")) +
  #geom_text(aes(y = IPOs_sum, label = IPOs_sum), fontface = "bold", vjust = 1.4, color = "black", size = 4) +
  geom_line(aes(y = returns_mean, group = 1, color = 'blackline')) +
  #geom_text(aes(y = returns_mean, label = round(returns_mean, 2)), vjust = 1.4, color = "black", size = 3) +
  scale_y_continuous(sec.axis = sec_axis(trans = ~ . / 20)) +
  scale_fill_manual('', labels = 'IPOs_sum', values = "#C00000") +
  scale_color_manual('', labels = 'returns', values = 'black') +
  theme_minimal()

The problem I am having is that the returns line plot uses the same scale as the bar plot which makes the line plot seem very small. I have tried scale_y_continuous

https://site.warrington.ufl.edu/ritter/files/2018/03/UnitedStates1980-2017.pdf

Data:

df <- structure(list(years = c(1980, 1981, 1982, 1983, 1984, 1985, 
1986, 1987, 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996, 
1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 
2008, 2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017), 
    returns_mean = c(49.525, 16.7583333333333, 15.2416666666667, 
    23.5916666666667, 11.6833333333333, 13.2, 6.39166666666667, 
    5.77272727272727, 4.65833333333333, 8.61666666666667, 9.56363636363636, 
    14.25, 10.6416666666667, 13.0916666666667, 9.86666666666667, 
    20.5166666666667, 17.3083333333333, 13.7916666666667, 39.6916666666667, 
    75.8416666666667, 49.0916666666667, 13.3363636363636, 8.48, 
    14.9666666666667, 13.3166666666667, 10.35, 11.3333333333333, 
    17.4083333333333, 4.47777777777778, 13.0888888888889, 7.99166666666667, 
    14.6272727272727, 16.6583333333333, 21.1666666666667, 15.8666666666667, 
    18.4583333333333, 11.0181818181818, 12.3583333333333), IPOs_mean = c(19.8333333333333, 
    37.5, 18.5, 73.5833333333333, 46, 42.25, 79.4166666666667, 
    52.5, 18.9166666666667, 17, 14.3333333333333, 30.5833333333333, 
    42.4166666666667, 52.25, 47.3333333333333, 47.1666666666667, 
    70.4166666666667, 51, 32.6666666666667, 45.3333333333333, 
    35.3333333333333, 11, 13.3333333333333, 11, 25.25, 23.3333333333333, 
    21.25, 20.75, 4.5, 6.33333333333333, 16.4166666666667, 15, 
    14.9166666666667, 21, 24.3333333333333, 14.3333333333333, 
    8.5, 16.0833333333333), returns_sd = c(30.9637067607164, 
    15.6027653920319, 18.6855538917749, 15.2870984424082, 2.74684391896437, 
    7.93702486051062, 4.59277264907165, 3.22275996906096, 4.27263136578371, 
    4.64872090586282, 4.9828250475554, 6.13299570875737, 6.54404077745316, 
    3.15204358914638, 3.6317622402488, 5.84587396581918, 6.69225581390818, 
    5.55361607559899, 47.3886725692838, 27.7436137887732, 31.6808935346135, 
    5.95605116285492, 5.27863618750146, 10.4857045542968, 8.05298739975458, 
    5.26402887530074, 5.75141616289309, 13.283992987689, 15.0286208430596, 
    9.68948456374802, 5.44951346174105, 8.37568993087625, 7.3368879126251, 
    5.83022427969243, 7.73672861724965, 14.1409436700239, 15.1387461952315, 
    12.0595837658711), IPOs_sd = c(9.44682085372768, 12.6383255507711, 
    8.74382899275514, 26.6473888652028, 11.7008158223729, 9.66836453218809, 
    24.9743808125386, 23.1025382312696, 5.31649804995276, 6.66060330327789, 
    6.9325757161827, 15.4358398383487, 11.7508864913815, 16.7610207977264, 
    12.6371266392709, 20.0264975984471, 19.965690268027, 14.709304414677, 
    18.7778076623993, 14.2148023574233, 18.6953243222778, 4.26401432711221, 
    5.39921430872263, 7.92005509622709, 8.48662048596067, 8.15010689649175, 
    7.9444091261488, 8.1700673191841, 4.07876986803174, 4.05268336096498, 
    5.07145905728292, 7.54381143117263, 7.06410044885512, 7.90856842579329, 
    7.77330318617783, 7.15202874375622, 6.18649555667239, 6.94731253904353
    ), returns_min = c(12.7, 2.2, -0.9, 2.5, 7.2, 3.6, 1, 0.5, 
    -0.6, 0.6, 0.6, 6.4, 3.2, 8.9, 6.5, 9.2, 8.9, 6, 9.3, 37.1, 
    15.8, 5.7, 1.9, -3.3, 0.5, 4.5, 0.4, 5.2, -19.9, 0.3, -3.5, 
    1.8, 2.4, 13.6, 5.2, -6, -4.3, -4.9), IPOs_min = c(8, 20, 
    11, 24, 28, 26, 37, 7, 11, 8, 4, 4, 22, 22, 26, 18, 29, 33, 
    6, 22, 9, 4, 6, 1, 11, 13, 10, 5, 0, 1, 8, 3, 6, 10, 13, 
    2, 0, 7), returns_sum = c(594.3, 201.1, 182.9, 283.1, 140.2, 
    158.4, 76.7, 63.5, 55.9, 103.4, 105.2, 171, 127.7, 157.1, 
    118.4, 246.2, 207.7, 165.5, 476.3, 910.1, 589.1, 146.7, 84.8, 
    134.7, 159.8, 124.2, 136, 208.9, 40.3, 117.8, 95.9, 160.9, 
    199.9, 254, 190.4, 221.5, 121.2, 148.3), IPOs_sum = c(238L, 
    450L, 222L, 883L, 552L, 507L, 953L, 630L, 227L, 204L, 172L, 
    367L, 509L, 627L, 568L, 566L, 845L, 612L, 392L, 544L, 424L, 
    132L, 160L, 132L, 303L, 280L, 255L, 249L, 54L, 76L, 197L, 
    180L, 179L, 252L, 292L, 172L, 102L, 193L)), class = c("tbl_df", 
"tbl", "data.frame"), row.names = c(NA, -38L))

Answer 1

Consider showing the data in a different way, possibly with a connected scatterplot:

dplyr::arrange(df, years) %>% 
  dplyr::mutate(col = ifelse(years >= 2000, "#08519c", "#74c476")) %>% 
  ggplot() +
  geom_path(aes(IPOs_sum, returns_mean)) +
  geom_label(aes(IPOs_sum, returns_mean, label=years, fill=I(col)), color = "white") +
  ggalt::geom_encircle(data = dplyr::filter(df, years > 2000), aes(IPOs_sum, returns_mean)) +
  labs(
    x = "Number of Offerings (IPOs)", y = "Average First-day Returns",
    title = "IPO Volume (Both Annual Count and Day-1 Returns)\nHas Been Very Low in the U.S. Since 2000"
  ) +
  hrbrthemes::theme_ipsum_rc(grid="XY")

Answer 2

You are right, the geom_* will all use the same y axis value. The secondary axis is just for display as far as I know.

What you can do is transform the value of returns to make it fits the left axis. If you don't want to modify the data, you can directly scale the value of returns in the geom_line's aes.

geom_line(aes(y = returns_mean * 20, group = 1, color = 'blackline'))