US Postal Service

Description

Numbers of post offices in the US from 1789 to 2020 with their income and expenses in current dollars and proportion of the federal government and of Gross Domestic Product (GDP). Also includes the number of pieces of mail, numbers of periodicals, pieces and periodicals per person, and cost coverage of periodicals for selected years.

It would be interesting to find the total value of the subsidies for newspapers and other periodicals as a proportion of the budgets of the USPS and the federal government as well as of GDP. That is currently absent from the data consulted to produce this.

Usage

data(USPS)

Format

A data.frame containing 232 observations on the following variables:

Year

integer: the year: 1789:2020

Income, Expenses

Income and expenses in millions of current dollars, per Historian (2022).

Income_pFed, Expenses_pFed

Income and Expenses as a proportion of USGDPpresidents[, 'fedReceipts'] and USGDPpresidents[, 'fedOutlays'], respectively.

Income_pGDP, Expenses_pGDP

Income and Expenses as a proportion of GDP, per MeasuringWorth.

Income_cap, Expenses_cap

Income and Expenses per capita in current dollars = Income and Expenses divided by 1000 * USGDPpresidents[, 'population.K'].

realIncome_cap, realExpenses_cap

Income and Expenses per capita in constant 2012 dollars = Income_cap and Expenses_cap divided by USGDPpresidents[, 'GDPdeflator'].

postOffices

Number of post offices per Historian (2022).

KpopPerPostOffice

US population in thousands per post office: USGDPpresidents[, 'population.K'] divided by postOffices.

piecesOfMail, periodicals

numeric: Millions of pieces of mail handled and periodicals mailed. "Pieces of mail"" are from Historian (2022). "Periodicals" are from Historian (2010).

piecesOfMailPerCap, periodicalsPerCap

piecesOfMail and periodicals handled per capita (per human in the US) per year.

costCoveragePeriodicals

Cost coverage of periodicals, per Historian (2010). This is available here only since 1960, though Historian (2010) gave a general outline of these numbers. This included saying, "In 1966, the percentage of its own costs covered by second-class mail (or 'cost coverage'), including the subsidy, was 35 percent [reported as 36 percent here]. Its real coverage was 24 percent." The narrative noted that during parts of the nineteenth century the actual rate was zero. Sometimes it was zero only within county. Sometimes advertising was charged a higher rate than news.

Other than numbers for the period since 1960, we note the coverage in 1951 as 20 percent, based on the following comment:

"In February 1951, in a special message to Congress, President Harry S. Truman argued at length for a rate increase: 'In fiscal year 1952 . . . newspaper and magazine publishers will have 200 million dollars – or 80 percent – of their postal costs paid for them by the general public.'"

Details

rownames(USPS) = year

Data used by McChesney and Nichols (2021-12-13) To Protect and Extend Democracy, Recreate Local News Media (Freepress.net, p. 6, note 10) to estimate that newspaper subsidies averaged roughly 0.216 percent of GDP between 1840 and 1844.

Author(s)

Spencer Graves

Source

Historian (2010-06) Postage Rates for Periodicals: A Narrative History, accessed 2022-04-29.

Historian (2022-02) Pieces of Mail Handled, Number of Post Offices, Income, and Expenses Since 1789.

References

Robert W. McChesney and John Nichols (2010) The Death and Life of American Journalism (Nation Books, pp. 310-311) describe how they computed 0.216 as an estimate of the percent of national income (Gross Domestic Product, GDP) devoted to newspaper subsidies, 1840-1844. The numbers in the current dataset seem essentially equivalent but new and therefore perhaps more accurate. With these numbers, we got 0.209 percent of GDP rather than their 0.216 percent.

Examples

##
## plot Expenses as a percent of the 
## federal budget and of GDP 
##
data(USPS)
plot(Expenses_pFed~Year, USPS, type='l')
plot(Expenses_pGDP~Year, USPS, type='l')
plot(100*periodicals/piecesOfMail~Year, 
      USPS, type='l', ylab='', 
      main='periodicals as percent of mail')
      
# Select a year 
# as a charcter string not a number:
USPS['1850',]

##
## Plot Expenses_pGDP with 
## USGDPpresidents[, 'fedOutlays_pGDP']
##
str(yrs2 <- intersect(USPS$Year, 
              USGDPpresidents$Year))
yrs2a <- as.character(yrs2)

str(USPS_fed <- cbind(USPS[yrs2a, "Expenses_pGDP"], 
      USGDPpresidents[yrs2a, "fedOutlays_pGDP"]))

matplot(yrs2, USPS_fed, log='y', 
  ylab='', las=1, type='l', xlab='')
abline(v=c(1840, 1844), lty='dotted', col='grey')
text(1842, 6e-3, cex=.7,
  'McChesney & Nichols analysis', srt=90, col='grey')

abline(v=c(1861, 1865), lty='dotted', col='grey')
text(1863, 6e-3, 'Civil War', srt=90, col='grey')
sel1 <- (USGDPpresidents$war=='World War I')
(yr1 <- USGDPpresidents$Year[sel1])
abline(v=yr1, col='grey', lty='dotted')
text(mean(yr1), 2e-3, 'WWI', col='grey', srt=90)

sel2 <- (USGDPpresidents$war=='World War II')
(yr2 <- range(USGDPpresidents$Year[sel2]))
abline(v=yr2, col='grey', lty='dotted')
text(mean(yr2), 2e-3, 'WWII', col='grey', srt=90)

abline(h=c(.001, .01, .1), lty='dotted', col='grey')
legend("bottomright", 
    c('USPS Expenses_pGDP', 'fedOutlays_pGDP'), 
    col=1:2, lty=1:2, bty='n')