Conjectures of British Investment, Tax Revenues, and Deficit Amounts from the Thirteenth to the Nineteenth Century using the Concept of Economic Surplus

Abstract

This article attempts to estimate trends in the levels of public and private investment, and national government surpluses and deficits from accumulated capital income, taxation, and rents estimated by different economic historians for England and the UK by utilizing the concept of Paul Baran and Paul Sweezy’s economic surplus. The data support historical accounts that income per capita growth begins to increase around the 1600s in Britain, perhaps due to the level of capital, tax, and land income achieving an adequate threshold amount. According to some historians, this would also be about the time of capitalism’s ascent as the dominant economic system in Britain. Even then, dramatic increases in investment and economic growth do not appear until the late eighteenth century when investment and deficits reach even higher levels. The new estimates developed in this article are offered as additional macroeconomic data supplements to works created by other authors and researchers and submitted as a demonstration of the concept of economic surplus and the power of threshold levels of private and public investment. Most of all they also give some support to Baran and Sweezy’s notion of a society’s economic surplus coming from labor exploitation and being used to further investment and government expenditures.

Keywords:

• feudalism
• capitalism
• economic surplus
• deficits
• investments
• mercantilism
• private investment
• public investment
• tax revenues

JEL Classification Codes:

• B50
• B51
• C80
• N13
• O11
• O52

Disclosure Statement

The author confirms that there are no relevant financial or non-financial competing interests to report.

Notes

1 This article uses Britain and British interchangeably for England and the UK and of things pertaining to England and the UK so as to minimize verbiage, although Britain (prior to 1707) refers to only England and Wales alone. England and Wales are separate from Scotland for centuries until James VI of Scotland also becomes King of England in 1603 (his title becomes James I), and then England, Wales, and Scotland mostly become one nation. James I Is the King of Scotland and England at the same time. In the 1700s, unification is made official by the English and Scottish Parliaments under Quenn Anne. The datasets used in the paper mostly make a distinction between England (or England and Wales) and the UK (England, Wales and Scotland). The latter begins around 1700 according to the datasets by Gregory Clark and Broadberry et al.

2 John Hatcher and Mark Bailey (2001) examine what they call the three “super models” or main theories of economic history by Thomas Malthus, Adam Smith, and Karl Marx regarding the middle ages and conclude that each is lacking and somewhat over simplistic. Christopher Dyer (2005) claims that late medieval times might not have been as bad or dire economically as some may think due to his research on probate records. It is not the purpose of this article to critique or support explicitly any one of these three perspectives, yet they are still useful frameworks for understanding medieval economies and are still dominant theories in the historical literature.

3 Which some have claimed were also caused indirectly by a mini-ice age (e.g., Fagan 2000; Blom 2019; among others).

4 There are other data sources which give estimates for similar variables but only for limited periods of time or only going back to the eighteenth century such as those on the Bank of England website in their “A Millennium of Macroeconomic Data,” much of which is comprised of the Stephen Broadberry et al. data (2015) used in this article (www.bankofengland.co.uk).

5 Both Clark (2009) and Broadberry et al. (2015) provide seven centuries worth of estimates or conjectures of British economic activity, although Clark takes a more Malthusian approach in his estimates whereas Broadberry et al. believe that the time period examined is not as bad or as subject to net zero economic growth over prolonged periods of time. The author for this article could not find longer and more comprehensive listings of data to provide a better and more coherent time series.

6 By having direct taxes (mostly income and property taxes) included in gross capital and gross rental income amounts, Clark has local and national government revenues fully accounted for when he includes a separate category for local and national indirect taxes (sales and excise taxes, etc.). The income of unproductive segments (military pay, income of jesters, clergy, etc.) is not disclosed and is difficult to estimate over seven centuries without accurate estimates of their numbers employed or pay. Broadberry et al. (2015) give estimates of the size of clergy and nobility over time in England and the UK but only for a handful of years out of the seven centuries. They also do not give conjectures on rent, capital, and taxation income as does Clark (2009).

7 Gordon Rimmer, J. P. P. Higgins, and Sidney Pollard (1971) assess the year-to-year rates of investment in the eighteenth and nineteenth centuries in the UK and estimate it to be slower than other estimates and believe a lot of capital investment undertaken was due to the rapid deterioration of many forms of plants and equipment. They cite the frequency with which horseshoes and many farm tools had to be replaced. Nonetheless, such replacement was necessary to propel agricultural output to higher levels, and therefore the investment expenditures could still be considered productive.

8 See article by Thomas Lambert (2020) for more background. Admittedly, this is assuming somewhat that the real rate of interest is a reward to capital based on capital’s productivity, which is a neoclassical economics tenet applied to pre-capitalist and partially capitalist time periods. This neoclassical tenet is a debatable concept since it does not address interest rate setting or targeting by lenders or other factors that influence interest rates. However, despite its flaws, this is one way absent any others to estimate a long run time series and to develop a trend of capital stock and investment spending over seven centuries absent other data that does not cover as long of a period. It should be emphasized that the goal of the conjectures developed in this article is to develop general trends that can support other historical observations and accounts.

9 A listing of all the data estimated by the author is provided in the appendix of the article.

10 Elise Brezis (1995, 57) gives estimates from Charles Feinstein (1978 and 1981) and Feinstein and Pollard (1988) of nominal investment in the UK from the eighteenth to twentieth centuries. Although different in magnitude from the estimates for this article, for 1740 to 1860 the Pearson correlation coefficient between Brezis’ estimates and the estimates for this article is +0.938, and after adjusting for serial correlation using Newey-West standard errors, the time series regression equation is Predicted Brezis Estimates = –45.97 + 12.48 Author Estimates with the independent variable being statistically significant at alpha = 0.05 and with the model having an adjusted r-square of around 0.87. Augmented Dickey-Fuller tests or unit root tests for each variable indicate failure to reject the null hypothesis of non-stationarity. Co-integration tests, however, indicate that the correlation between the two variables is probably not spurious.

11 The results of these tests can be provided upon request.

12 Direct taxes such as national income taxes, wealth and gift taxes are not used to a large extent in England and/or the UK until toward the end of eighteenth and sometimes during the nineteenth centuries to finance various wars and are not used on a consistent basis until the twentieth century (Seely 1995; HM Revenues and Customs 2010).

13 The author could not find definitive sources for English and UK budget deficits and surpluses before 1800, although estimates of net public debt can be found from B. R. Mitchell (1988) and Christopher Chantrill (n.d.) that go back to 1692. The surplus and deficit estimates that are calculated in this article have a +0.735 correlation coefficient with the inflation adjusted debt level (base year of 1860 using Clark’s price index) estimates from Mitchell when correlated on a decadal basis from 1700 to 1820 (no more estimates are provided after 1829 although debt as a percentage of GDP is presented). As deficits increased so did corresponding debt levels. The furthest Mitchell’s data goes back in time appears to be 1688, usually 1692. Predicting Mitchell’s estimates using the author’s deficit/surplus numbers yields a regression equation with Newey-West standard errors (to correct for serial correlation) of Predicted Mitchell Debt = 157.86 – 1.06 Author Estimates with an adjusted r-square of 0.498. The independent variable is statistically significant at alpha = 0.05. Augmented Dickey-Fuller tests or unit root tests for each variable indicate failure to reject the null hypothesis of non-stationarity. However, Johansen tests of cointegration indicate that the relationship is probably not spurious. These results can be furnished upon request.

14 Interestingly one thing that Barro finds is that as long as currency could be converted to gold, money supply growth and inflation are not problems resulting from the budget deficits or the temporary rises in government spending mostly due to military spending. He claims that such deficits are associated with increases in long term interest rates, however, except for the deficits associated with the slavery buy out and the income tax dispute. In those two cases, long term rates do not rise. Clark (2001) in estimating deficits from the 1720s to the 1830s finds no “crowding out” effects of British deficits. Figure A1 in the appendix also plots Clark’s estimates of real interest rates from 1200 to 1860.

15 Javier Esteban (2001) writes that the French wars would have been very difficult for Britain to finance had it not been for trade credits from India.

16 There is of course a simultaneous relationship among many of these variables with investment not only leading to higher real output/income, but the latter also leads to higher investment, consumption, etc., in turn. That is, a feedback loop exists between investment and output.

17 The high negative correlation between investment and government surpluses/deficits is not surprising, of course, since surpluses/deficits were derived by taking the savings plus taxes minus investment estimates.

18 The results of these tests can be provided upon request.

19 Using income (A. H. Studenmund (2017, 421–434).

20 Although the scatterplots show this somewhat, Gregory Chow (1960) tests indicate the year 1780 as the strongest demarcation or “break” in the time series versus other years.

21 Graphs of real GDP per capita using the Clark and Broadberry et al. data showed pretty much a flat line trend during both feudalistic and mercantilistic (or transition period) epochs.

22 David Richardson (1987) and David Etlis and David Richardson (2008), among other scholars, estimate that the British slave trade of the seventeenth to nineteenth centuries had a big impact on British economic growth.

Additional information

Notes on contributors

Thomas E. Lambert

Thomas E. Lambert is at the University of Louisville in Kentucky. The author expresses gratitude to the editors and reviewers for their help. Any remaining errors in the paper are strictly those of the author.