1
2
3
4
5
#

Name:

Description:

It is commonly claimed that Franklin Roosevelt (FDR) did not end the Great Depression: World War II (WW2) did. This is supported by the 10.6 percent growth per year in real Gross Domestic Product (GDP) per capita seen in the standard GDP estimates from 1940 to 1945. It is also supported by the rapid decline in unemployment during the war.

However, no comparable growth spurts in GDP per capita catch the eye in a plot of log(GDP per capita) from 1790 to 2015, whether associated with a war or not, using data from Measuring Worth. The only other features of that plot that seem visually comparable are the economic disaster of Herbert Hoover's presidency (when GDP per capital fell by 10 percent per year, 1929-1932), the impressive growth of the US economy during the first seven years of Franklin Roosevelt's presidency (6.4 percent per year, 1933-1940), and the post-World War II recession (when GDP per capita fell by 7.9 percent per year, 1945-1947). (NOTE: The web site for Measuring Worth, https://measuringworth.com/ still works, but is not currently maintained to current internet security standards. Therefore, the link is provided here in text but not as a not link.)

Closer inspection of this plot suggests that the US economy has generally grown faster after FDR than before. This might plausibly be attributed to "The Keynesian Ascendancy 1939-1979".

Unemployment dropped during the First World War as it did during WW2. Comparable unemployment data are not available for the U.S. during other major wars, most notably the American Civil War and the Mexican-American War.

This data set provides a platform for testing the effects of presidency, war, and Keynes. It does this by combining the numbers for US population and real GDP per capital dollars from Measuring Worth with the presidency and a list of major wars and an estimate of the battle deaths by year per million population. (As noted above, the web address for measuring worth, https://measuringworth.com/, often gives security warnings but still seems to provide the data as before.)

US unemployment is also considered.

Variables:

A data.frame containing 259 observations on the following variables:

Year

integer: the year, c(seq(1610, 1770, 10), 1774:2015)

CPI

Numeric: U. S. Consumer Price Index per Officer and Williamson (2015). Average 1982-84 = 100.

GDPdeflator

numeric: Implicit price deflators for Gross Domestic Product with 2012s = 100 per Johnston and Williamson.

population.K

integer: US population in thousands.

Population figures for 1610 to 1780 came from Springston (2013). The rest came from Johnston and Williamson. (The early population figures reflect only the European settlers in the British colonies that eventually became the US.)

realGDPperCapita

numeric: real Gross Domestic Product per capita in 2012 dollars

executive

ordered: Crown of England through 1774, followed by the "ContinentalCongress" and the "ArticlesOfConfederation" until Washington, who became President under the current base constitution in 1789. Two nineteenth century presidents are not listed here (William Henry Harrison and James A. Garfield), because they died so soon after inauguration that any contribution they made to the economic growth of the nation might seem too slight to measure accurately in annual data like this; their contributions therefore appear combined with their replacements (John Tyler and Chester A. Arthur, respectively). The service of two other presidents is officially combined here: "Taylor-Fillmore" refers to the 16 months served by Zachary Taylor with the 32 months of Millard Fillmore. These modifications make Barack Obama number 41 on this list, even though he's the 44th president of the U.S.

war

ordered: This lists the major wars in US history by years involving active hostilities. A war is "major" for present purposes if it met two criteria:

(1) It averaged at least 10 battle deaths per year per million US population.

(2) It was listed in one of two lists of wars: For wars since 1816, it must have appeared in the Correlates of War. For wars between 1790 and 1815, it must have appeared in the Wikipedia "List of wars involving the United States".

The resulting list includes a few adjustments to the list of wars that might come readily to mind for people moderately familiar with US history.

A traditional list might start with the American Revolution, the War of 1812, the Mexican-American war, the Civil War, the Spanish-American war, World Wars I and II, Korea, and Vietnam. In addition, the Northwest Indian War involved very roughly 30 battle deaths per year per million population 1785-1795. This compares with the roughly 100 battle deaths per year 1812-1815 for the War of 1812.

For present purposes, the Spanish-American War is combined with the lesser-known American-Philippine War: The latter involved 50 percent more battle deaths but over a longer period of time and arguably with less impact on the stature of the US as a growing world power. However, its magnitude suggest it might have impacted the US economy in a way roughly comparable to the Spanish-American war. The two are therefore listed here together as "Spanish-American-Philippine" war.

The Correlates of War (COW) data include multiple US uses of military force during the Vietnam War era. It starts with "Vietnam Phase 1", 1961-65, with 506 battle deaths in the COW data base. It includes the "Second Laotian" war phases 1 and 2, plus engagement with a "Communist Coalition" and Khmer Rouge as well as actions in the Dominican Republic and Guatemala. The current data.frame includes only "Vietnam", referring primarily to COW's "Vietnam War, Phase 2", 1965-1973. The associated battle deaths include battle deaths from these other, lesser concurrent conflicts.

The COW data currently ends in 2007. However, the post-2000 conflicts in Afghanistan and Iraq averaged less than 1,000 battle deaths per year or roughly 3 battle deaths per year per million population. This is below the threshold of 10 battle deaths per year per million population. This in turn suggests that any impact of those conflicts on the US economy might be small and difficult to estimate.

battleDeaths

numeric: Numbers of battle deaths by year estimated by allocating to the different years the totals reported for each major war in proportion to the number of days officially in conflict each year. The totals were obtained (in August-September 2015) from The Correlates of War data for conflicts since 1816 and from Wikipedia for previous wars, as noted above.

battleDeathsPMP

numeric: battle deaths per million population = 1000*battleDeaths/population.K.

Keynes

integer taking the value 1 between 1939 and 1979 and 0 otherwise, as suggested by the section entitled "The Keynesian Ascendancy 1939-1979" in the Wikipedia article on John Maynard Keynes.

unemployment

Estimated US unemployment rate

unempSource

ordered giving the source for US unemployment:

1800-1889

Lebergott

1890-1929

Romer

1930-1939

Coen

1940-present

BLS

Clearly, the more recent numbers should be more accurate.

Details

rownames(USGDPpresidents) = Year

Link To Google Sheets:

Rows:

Columns:

License Type:

References/Notes/Attributions:

Author(s)

Spencer Graves

Source

Louis Johnston and Samuel H. Williamson, "What Was the U.S. GDP Then?", Measuring Worth, accessed 2015-09-08. (NOTE: This came from https://www.measuringworth.org/usgdp/. this web link generally works as of 2020-02-09 but routinely returns a warning, e.g., "SSL certificate problem". The web site seems to be good but not maintained to current security stqndards.)

Sarkees, Meredith Reid; Wayman, Frank (2010). "The Correlates of War Project: COW War Data, 1816 - 2007 (v4.0)", accessed 2015-09-02.

Wikipedia, "List of wars involving the United States", accessed 2015-09-13.

Wikipedia, "Unemployment in the United States". See also https://en.wikipedia.org/wiki/User_talk:Peace01234#Unemployment_Data. Accessed 2016-07-08.

Stanley Lebergott (1964). Manpower in Economic Growth: The American Record since 1800. Pages 164-190. New York: McGraw-Hill. Cited from Wikipedia, "Unemployment in the United States", accessed 2016-07-08.

Christina Romer (1986). "Spurious Volatility in Historical Unemployment Data", The Journal of Political Economy, 94(1): 1-37.

Robert M. Coen (1973) Labor Force and Unemployment in the 1920's and 1930's: A Re-Examination Based on Postwar Experience", The Review of Economics and Statistics, 55(1): 46-55.

The unemployment data since 1940 are from series LNS14000000 from the Current Population Survey. These data are available as a monthly series from the Current Population Survey of the Bureau of Labor Statistics.

Chuck Springston, "Population of the 13 Colonies 1610-1790", October 28, 2013

R Dataset Upload:

Use the following R code to directly access this dataset in R.

d <- read.csv("https://www.key2stats.com/US_GDP_per_capita_with_presidents_and_wars_672_41.csv")

R Coding Interface:


Datasets Tag Questions & Instructional Blocks

NumberContentType
No results found.