Piketty and Saez 2003, enhanced fig. 1

incomes.r

## Top incomes data by Piketty and Saez, Sep 2013 update.
## 2014-10-16

## source

# T. Piketty and E. Saez, "Income Inequality in the United States, 1913-1998"
# Quarterly Journal of Economics, 118(1), 2003, 1-39 [ fig. 1, enhanced ]

## packages

library(ggplot2)
library(reshape2)
library(scales)
library(xlsx)

## dataset

file = "incomes.csv"
if(!file.exists(file)) {

  download.file("http://emlab.berkeley.edu/~saez/TabFig2012prel.xls", file, mode = "wb")

  # read XLSX
  ps = read.xlsx(file, sheetName = "Table A1",
                 startRow = 4, endRow = 105, colIndex = 1:7)

  # drop first row
  ps = ps[ -1, ]
  
  # write CSV
  write.csv(ps, file, row.names = FALSE)

}

# read CSV
ps = read.csv(file, stringsAsFactors = FALSE)

# result
str(ps)

## variables

# names
names(ps) = c("Year", paste0("top ", c(10, 5, 1, 0.5, 0.1, 0.01), "%"))

# reshape
ps = melt(ps, id = "Year", variable = "Fractile")

# subset
ps = na.omit(ps)

# result
head(ps)
tail(ps)

## plot

qplot(data = ps, x = Year, y = value / 100, linetype = Fractile, geom = "line") + 
  labs(y = NULL, x = NULL, title = "U.S. Top Income Shares of National Revenue, 1917-2012\n") +
  geom_text(data = subset(ps, Year == 2012), aes(x = 2014, label = Fractile, hjust = 0)) +
  scale_x_continuous(lim = c(1911, 2031), breaks = seq(1910, 2010, by = 20)) +
  scale_y_continuous(labels = percent) +
  theme_minimal(12) +
  theme(legend.position = "none")

# uncomment to save
# ggsave("piketty_saez.png", width = 11, height = 6.375)

## have a nice day