Technical Notes For ‘Why The Next Global Crisis…’

If you like knowing that your faithful blogger scours obscure data sources in dark corners of libraries to bring you analysis you’re unlikely to find in the mainstream media, you’ve come to the right place. As we worked on the budget balance chart for “Why The Next Global Crisis Will Be Unlike Any In The Last 200 Years,” one planned trip to the library became two and then three and four.

Here’s a list of the data we used, followed by some details and comments about the start dates for each country in the analysis:

Military spending

• For the U.S. from 1816 until 1928, we used: Historical Statistics of the United States, Earliest Times to the Present: Millennial Edition, edited by Susan B. Carter, Scott Sigmund Gartner, Michael R. Haines, Alan L. Olmstead, Richard Sutch, and Gavin Wright. New York: Cambridge University Press, 2006.
• For the U.S. from 1929 until 2012, we used OMB data.
• For the U.K. from 1816 until 1979, we used: B. R. Mitchell, British Historical Statistics (Cambridge: Cambridge University Press, 1988).
• For all other countries up until 1985 and the U.K. from 1980 until 1985, we used ICPSR data: Singer, J. David, and Melvin Small. NATIONAL MATERIAL CAPABILITIES DATA, 1816-1985 [Computer file]. Ann Arbor, MI: J. David Singer, University of Michigan, and Detroit, MI: Melvin Small, Wayne State University [producers], 1990. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 1993.
• For all countries except the U.S. from 1988 until 2012, we used the World Bank’s indicator for military spending as a % of GDP.
• For all countries except the U.S., we interpolated figures for 1986 and 1987 from the other sources noted above.

Budget balances

• For the U.S. from 1816 until 1900, we used Historical Statistics of the United States, Earliest Times to the Present: Millennial Edition.
• For the U.S. from 1901 until 2012, we used OMB data.
• For the U.K. from 1816 until 1979, we used Mitchell, British Historical Statistics.
• For France from 1816 to 1958, we used: B.R. Mitchell, International Historical Statistics: Europe, 1700-1993 (London: Macmillian, 1998).
• For France from 1959 to 2012, we used Insee data.
• For the Netherlands from 1816 until 2012, Frits Bos kindly supplied the data that he used in: Frits Bos, “The Dutch Fiscal Framework: History, Current Practice and the Role of the Central Planning Bureau,” OECD Journal on Budgeting Vol. 8 No. 1, 2008.
• For Italy from 1862 until 1987, we used Mitchell, International Historical Statistics: Europe, 1700-1993.
• For Canada from 1949 until 1979, we used B.R. Mitchell, International Historical Statistics: The Americas 1750-1988 (New York: Stockton Press, 1993).
• For Australia from 1949 until 1987, we used B.R. Mitchell, International Historical Statistics: Africa, Asia & Oceania 1750-1993 (London: Macmillan, 1988).
• For Germany from 1955 to 1990, we used Destatis Statistisches Bundesamp data.
• For Spain from 1958 to 1979, we used Mitchell, International Historical Statistics: Europe, 1700-1993.
• For Japan, South Korea, the U.K., Canada and Spain from 1980, Italy and Canada from 1988, and Germany from 1991, we used data from the IMF’s October 2013 World Economic Outlook.

Central versus general government

• Budget data from the IMF, Insee and Destatis Statistisches Bundesamp are for general government finances.
• Mitchell’s data is primarily limited to central governments.
• Netherlands data is mixed but “very close to” general government balances after WW2.
• For the U.S., we used central government data for all time periods.

GDP

• U.S. GDP estimates from 1926 to 2012 are from the BEA.
• U.S. GDP estimates from 1816 to 1928 are derived from Treasury Direct debt figures and Reinhart, Rogoff debt/GDP figures and checked against the Smits, Woltjer and Ma database: J.P. Smits, P.J. Woltjer and D. Ma (2009), ‘A Dataset on Comparative Historical National Accounts, ca. 1870-1950: A Time-Series Perspective’, Groningen Growth and Development Centre Research Memorandum GD-107, Groningen: University of Groningen.”
• U.K. budget data through 1979 and military spending data through 1985 are matched with GDP estimates from the Bank of England.
• Dutch military spending data through 1985 is matched with GDP estimates from Smits, Woltjer and Ma.
• Other budget data from Mitchell and military spending data from the ICPSR are matched with GDP estimates from Mitchell, although substituting GNP or NNP in the absence of GDP in certain instances.
• Budget data from Destatis Statistisches Bundesamp are matched with GDP estimates from Mitchell.
• Budget data from the IMF, Insee and Fritz Bos include % GDP figures.

Country weights

• Country weights from 1816 to 1979 are calculated from Angus Maddison’s estimates of world GDP.
• Country weights from 1980 to 2012 are calculated from the IMF’s October 2013 World Economic Outlook figures for U.S. $-denominated GDP, translating other currencies to U.S. $ at market rates.

Exchange rates

• For the U.S. $ to British pound rate, we used Bank of England data.
• For other U.S. $ exchange rates from 1914 to the present, we used data from www.measuringworth.com.
• For pre-1914 British pound to French Franc rates, we used Mitchell, British Historical Statistics.
• We used gold parities for other pre-1914 intra-European exchange rates, using the post-1875 parity for the Dutch Guilder (which isn’t significantly different from pre-1875 gold prices).

Country start dates

• The U.S., U.K., France and Netherlands are included in the global aggregate from the 1816 inception.
• Italy is added in 1862, although it’s excluded from the aggregate in the years 1935 and 1936 (due to inconsistencies between budget and military spending data).
• Canada and Australia are added in 1949.
• Germany is added in 1955.
• Spain is added in 1958.
• Japan is added in 1980.
• South Korea is added in 1995.
• These start dates are based on a combination of data limitations (for instance, some countries’ older budget data blends in financing activities, making it inappropriate for this analysis) and confidence that we’ve taken the database far enough to reach meaningful conclusions.
• For the two largest countries that aren’t included until after World War 2 – Germany and Japan – a longer history would suggest a different type of message, but one that ends in the same place. In other words, we can view current trends as ominous because they’re unprecedented in the data shown in our chart; or, we can view them as ominous because they bring to mind fiscal struggles in, say, the Weimar Republic and Nazi Germany. Either way, today’s dangers should be clear.
• Ideally, we would construct separate budget histories for the Allies and Axis powers of World War 2, but it’s doubtful that we could piece together reliable data through Weimar, Nazi Germany and pre-war Japan.