War, Democracy, and Government Size Over the Long Run: A Structural Breaks Analysis∗ Mark Dincecco†

Mauricio Prado‡

14 April 2010 Abstract This paper examines the links between warfare, democracy, and government size over the long run, from the early nineteenth century to the present. We assemble new time series for government size for eight European countries. We then examine individual data trends using a combination of narrative and statistical techniques. Our key methodological contribution is to employ structural breaks tests, which assume no a priori knowledge of major turning points in the time series, but let the data “speak” for themselves. Our results indicate that warfare, and not franchise extension, is the key driver of long-run increases in government size. Keywords: wars, voting franchise, government size, structural breaks tests, European history.



We thank Leonardo Baccini, Jing-Yuan Chiou, Peter Lindert, Thomas Piketty, and Jean-Laurent Rosen-

thal. †

Corresponding author: IMT Lucca Institute for Advanced Studies; email: [email protected]



University of Cambridge, and King’s College; email: [email protected]

1

1

Introduction

Over the past two centuries, there has been dramatic growth in the size of governments in developed countries.1 One standard factor that is offered to explain this increase is democratization. As the government extends the franchise to the less affluent, the relative income of the median voter falls. In turn, there is greater political demand for fiscal policies that redistribute tax amounts from the rich to the poor. But is suffrage sufficient by itself to explain larger government? Another argument emphasizes the effects of warfare. In the face of military competition, states may implement fiscal changes that increase their ability to raise revenues. To finance its historical campaigns against France, for instance, Britain introduced a modern income tax in 1799. Though repealed in 1802, this tax was reintroduced the next year with the start of the Napoleonic Wars (1803-15). If fiscal innovations endure after conflict ends, then the expansion of the public sector may be permanent. There are also links between warfare and suffrage. In the face of major conflicts, governments may confront a new set of commitment problems. Since franchise extension enables governments to credibly commit to greater redistribution in the future, citizens may be more willing to pay higher tax amounts or fight hard to win the war. Indeed, many twentieth-century episodes of suffrage reform took place during or near the time of large-scale conflicts. To improve our understanding of the relationships between warfare, democracy, and growth in government, this paper performs an empirical analysis that runs from the end of the Napoleonic era to the present. To our knowledge, it is the first systematic study of individual trends in government size over the long run for a large set of sample countries. Our investigation employs a combination of qualitative and statistical techniques. We use analytic narratives to document changes in government size over time in relation to external wars and franchise reforms. Our key methodological contribution is to perform structural breaks tests, which assume no a priori knowledge of major turning points in the government size series, but let the data “speak” for themselves. Breaks tests do not pinpoint causal relationships, but if wars and/or franchise reforms are associated with turning points, then there is reason to think that such factors are important. By documenting the broad temporal order in which events occur, the 1

See Lindert (2004).

2

narratives also help to clarify casuality. Our approach is thus a novel way to circumvent problems of endogeneity that affect standard regression analyses. Our results indicate that warfare, and not franchise extension, is the key driver of long-run increases in government size. Major (i.e., universal and equal) franchise reforms were sometimes associated with structural breaks, but only if they were also associated with wars, and in particular World Wars I (1914-8) and II (1939-45). Wars, on the other hand, were sometimes independently associated with breaks, though this outcome was more likely when major franchise reforms had already occurred. Elite franchise reforms, in which less than 40 percent of the adult population had the right to vote, were almost never associated with breaks. In total, our findings imply that franchise extension is at best a necessary, but not sufficient, condition for persistent increases in government size. The rest of the paper proceeds as follows. Section 2 reviews the key theoretical arguments. Section 3 describes the data and descriptive facts. Section 4 examines the time series for government size by country. Section 5 presents the structural breaks methodology and performs the breaks tests on government size levels. For robustness, Section 6 performs similar tests on growth in government. Section 7 concludes.

2

Theoretical Arguments

Three theoretical arguments guide our inquiry. The first theory, which relates growth in government to democratization, draws on classic theoretical works by Black (1948) and Downs (1957), as well as later contributions by Romer (1975), Roberts (1977), Meltzer and Richard (1981), and Persson and Tabellini (2000). By extending the franchise, the government becomes more accountable to poorer voters, who demand greater amounts of collective goods (education, health, housing, transportation) and transfers (unemployment benefits, old age pensions). In consequence, we would expect to observe a permanent expansion of the public sector.2 We summarize the hypothesis associated with this argument as follows: Hypothesis 1. The extension of the franchise increases government size. The second theory, which emphasizes the impact of warfare on fiscal capacity, draws 2

This argument finds support in panel data analyses for Europe, the United States, and elsewhere. See Husted and Kenny (1997), Kenny and Lott (1999), Lindert (2004), Aidt et al. (2006), and Aidt and Jensen (2009a,b).

3

on historical and theoretical works such as Tilly (1990), O’Brien (2005), Dincecco (2009), and Besley and Persson (2008, 2009, 2010). To wage war, the government may implement changes to its tax system, such as the income tax. Fiscal capacity, defined as the ability of the state to raise tax revenues, thus rises. Though the public sector typically expands during military conflicts, government size may shrink to prior levels at war’s end.3 If fiscal innovations become permanent, however, and the government continues to play a large economic role, then expansion may endure.4 We formulate the hypothesis associated with this argument as follows: Hypothesis 2. Interstate conflicts increase government size. The final theory, which builds upon the previous two arguments, claims that warfare promotes democratization, and thus redistribution. Several potential mechanisms underpin this claim.5 We describe two possibilities, both of which see franchise extension as way for governments to solve commitment problems that arise in the face of major conflicts. The first draws on works by Hoffman and Rosenthal (1997a,b) that investigate the emergence of nascent forms of representative government in European history. If tax changes under the status quo political regime are unfeasible, or provide insufficient funds for military efforts, then the government may undertake suffrage reforms as part of a credible bargain by which newly enfranchised elites agree to higher tax levels. By extending the franchise, the government commits to provide greater future amounts of public goods that such voters value. As a consequence, the public sector permanently expands. The second possibility draws on work by Ticchi and Vindigni (2009) that examines suffrage reforms within political systems that are already partially representative. By virtue of the large size of mass armies, conscripted soldiers are imperfectly monitored by central military authorities. To induce middleand lower-class citizens to fight hard in battle, elites may thus extend the franchise as a 3

This possibility draws on the tax-smoothing literature, which claims that, to minimize supply-side disincentives caused by sudden changes in taxation, governments should finance large, temporary increases in spending with loans funded by peacetime surpluses. See Barro (1979, 1987, 1989). 4 There are many historical examples in which warfare led to permanent tax changes. The Austrian Empire implemented a permanent income tax during the First War of Italian Independence (1848-9). Similarly, the U.S. government established the Internal Revenue Service during its war against Southern states (1861-5). Finally, Japan introduced a permanent income tax during the large-scale Satsuma Rebellion (1887), in which the Meiji government defeated the ex-samurai. See Clodfelter (2002) and Aidt and Jensen (2009b). 5 The literature that relates external threats to democratization has a distinguished pedigree. See McKinley (1934), Weber (1950), Andreski (1968), Marwick (1974), Mann (1988), Porter (1994), Halperin (2004), and Scheve and Stasavage (2009).

4

credible commitment to the greater future provision of collective goods and transfers.6 We operationalize the hypothesis associated with this argument as follows: Hypothesis 3. Interstate conflicts induce suffrage reforms that increase government size.

3

Data and Descriptive Facts

3.1

Government Size

To study changes in government size over time, we constructed a database from Mitchell (2007). Our key measure, central government expenditures (inclusive of social security) as a share of GDP, follows Persson and Tabellini (2003). Our analysis runs from the first year that data were available to 1998, the last year before the euro currency was introduced. Our set of sample countries builds on Aidt et al. (2006) and covers eight European countries: Denmark, Finland, France, Italy, the Netherlands, Norway, Sweden, and the United Kingdom. We focused on spending by central governments rather than general government spending (that incorporated local and regional spending) for reasons of data availability, comparability, and reliability. Persson and Tabellini (2003) find that there is a very strong correlation between the size of central and general governments. Moreover, all of our sample countries had centralized (versus federalist) political structures over the sample period. We are thus confident that the use of this sort of data does not bias our inferences. Table 1, which displays the descriptive statistics, indicates that there was substantial growth in government size from the early nineteenth century to the present. Prior to 1913, mean government spending amounted to 10 percent of GDP. During the period that encompassed the Two World Wars (1914-45), this average more than doubled. After 1945, mean government spending rose to nearly 30 percent of GDP. For the post-World War II period, the Netherlands and Norway had the largest governments, at 34 percent of GDP, and France had the smallest, at 22 percent of GDP. Government size in Denmark, Finland, Sweden, and the United Kingdom ranged from 28 to 30 percent of GDP. 6

Factors beyond wars may also influence suffrage reforms. See Acemoglu and Robinson (2000, 2002, 2006), Conley and Temimi (2001), Boix (2003), Lizzeri and Persico (2004), Llavador and Oxoby (2005), Greif (2006), Jack and Lagunoff (2006), and Gradstein (2007).

5

3.2

Warfare

Our conflict data are from Correlates of War Database of Sarkees (2000) and include all interstate wars fought over the sample period. Table 2, which summarizes these data, indicates that France fought the most conflicts, at 17, followed by Italy, at 12, and the United Kingdom, at 10. All three of these countries participated in both World Wars. Denmark, Finland, the Netherlands, and Norway, on the other hand, each fought two or less conflicts, and Sweden did not fight any wars. All five of these countries were neutral during World War I, and only Finland and Norway participated in World War II. However, the proximity of these countries to the war theater, and hence the threat of invasion, may have prompted military mobilization and other defensive actions.7

3.3

Franchise Extension

Our franchise data are from Flora (1980) and refer to the extension of voting rights to citizens with no previous political representation at the national level. Restrictions on universal franchise occurred by excluding certain population groups, by economic, educational, or occupational census requirements, or by high voting age limits. Restrictions on voting equality occurred by dividing the electorate into estates with disproportionate numbers of representatives, by giving certain groups additional votes through plural voting, or by gerrymandering. Table 3, which summarizes these data, divides franchise reforms into two types. Elite reforms were non-universal and/or unequal. After they took place, less than 40 percent of the adult population had the right to vote. Major reforms, in contrast, were both universal and equal, and granted the suffrage to all adult men and/or women. On average, sample countries experienced roughly two elite reforms, which typically occurred during the nineteenth century. Major reforms, on the other hand, typically occurred within five years of the start or end of World Wars I and II. The exceptions were France, which established universal suffrage for all adult men in 1848, and Finland and the United Kingdom, which established universal suffrage for all adult men and women in 1907 and 1928, respectively. 7

See Ticchi and Vindigni (2009).

6

4

Analytic Narratives

Figures 1 through 8 document changes in government size levels relative to external wars and franchise reforms from the early nineteenth century to the present for each sample country. To further explore the general patterns described in Section 3, we now turn to an examination of the time series data on a country-by country basis.

4.1

Denmark

Figure 1, which plots government size in Denmark from 1854 to 1998, indicates that government spending as a share of GDP was typically low, at less than 10 percent, through the start of World War II in 1939. Though Denmark was neutral during this conflict, there was a steady increase in government size over the next four decades. Government spending as a share of GDP peaked at over 50 percent at the start of the 1980s, but fell since. Both instances of franchise extension in Denmark came during times of war. Universal and equal adult suffrage was established at the end of World War I in 1918. While this set of events was associated with an uptick in government size, government spending as a share of GDP soon fell to pre-1913 levels. Though an earlier, elite reform was made in 1849, as Denmark fought the First Schleswig-Holstein War, we cannot assess the impact of this change on government size, since the data series began in 1854. Finally, there was also an uptick in government spending as a share of GDP during the Second Schleswig-Holstein War in 1864. Franchise extension, however, did not accompany this conflict.8

4.2

Finland

The time series for Finland, which Figure 2 plots from 1882 to 1998, indicates four major jumps in government size.9 The first occurred at the end of the nineteenth century, when government spending as a share of GDP roughly doubled to 10 percent. The second occurred at the end of the first World War, when government size increased 8

On the contrary, King Christian IX made the franchise more restrictive in 1866 by enacting higher wealth requirements, though this change did not significantly affect the share of the adult population that had the right to vote. See Carstairs (1980) and Flora (1980). 9 Though Finland was part of the Russian Empire after 1809, it was largely autonomous. At the end of the nineteenth century, however, the Russian government began to take a heavy-handed approach to Finnish domestic affairs. Finland became independent in 1917 following the Russian Revolution. See Carstairs (1980).

7

to around 20 percent. The third occurred with World War II. During this conflict, in which Finland participated, government spending reached nearly 60 percent of GDP. Though government spending fell in the aftermath, it remained between 25 and 30 percent of GDP from 1945 onwards. Finally, a fourth jump occurred at the end of the 1980s, with government size peaking at over 40 percent of GDP in the early 1990s. There are two recorded franchise reforms for Finland. The first, an early, elite reform, took place in 1869, before the data series began. The second, a major reform that established universal and equal adult suffrage, took place in 1907. As Figure 2 shows, it occurred between the first and second major jumps described above, and was not associated with any notable change in government size.

4.3

France

Though France fought nine wars from 1816 to 1913, and established universal and equal suffrage for adult men, government spending remained low at just over 10 percent of GDP (see Figure 3). The most notable uptick over this period came during the FrancoPrussian War (1870-1), when government size increased to nearly 15 percent. By 1890, however, it had fallen to pre-war levels. The instance of major franchise reform in 1848, which coincided with the First Italian War of Independence, was also associated with a slight rise in government spending as a share of GDP. Since Napoleon III overthrew the Second Republic and established an authoritarian regime in 1852, though, it is difficult to attribute this increase to democratization. The major jump in government size in France occurred with World War I. During the interwar period, average government spending as a share of GDP, at 20 percent, was roughly double that of the pre-1913 era. The second instance of major franchise reform, which extended the suffrage to women, took place at the end of World War II in 1945. Unlike the first World War, however, there was no major jump in government size associated with this set of events. Government spending as a share of GDP hovered at just over 20 percent to the present.

4.4

Italy

Figure 4, which plots government size in Italy from 1862 to 1998, indicates that government spending as a share of GDP was higher than in France, but still relatively low, at roughly 13 percent, through the start of World War I. Over this period, Italy fought seven wars, and made two elite franchise reforms, none of which had a lasting

8

impact on government size. All three instances of major franchise extension in Italy were associated with World Wars I and II. Nearly universal and equal male suffrage was established in 1913, just before the start of World War I. In turn, the share of the adult population with the right to vote rose from 15 to 40 percent. In 1919, male suffrage became fully universal and equal. The franchise was extended to women in 1946, just after the end of World War II. Both World Wars were associated with large increases in government size. In the former case, government spending peaked at over 50 percent of GDP, and in the latter, at just under 50 percent. The Second Italo-Ethiopian War (1935-6), also led to a substantial rise in government size. From the 1960s onwards, there was a steady increase in government spending, which peaked at 45 percent of GDP at the end of the 1980s.

4.5

The Netherlands

In the aftermath of the Napoleon Wars, King William I was established as the head of a constitutional monarchy with absolutist leanings.10 The king spent large amounts on the military, infrastructure, and the monarchy itself, and government expenditures as a share of GDP roughly doubled to 20 percent by the 1820s (see Figure 5). By 1839, the king’s budget problems had become public, and William was forced to abdicate. Over the 1840s and 1850s, government spending fell, stabilizing around 10 percent of GDP through 1913. Elite franchise reforms in 1849, 1887, and 1897 did not have a notable effect on government size. Major franchise reforms occurred at the end of World War I. Universal and equal male suffrage was established in 1918, and voting rights were extended to women in 1922. During the first World War, government size more than doubled, though it fell to pre-war levels in the aftermath. The largest increase in government spending took place during World War II, when it rose to more than 100 percent of GDP. Though government spending fell during the 1950s, it always amounted to at least 25 percent of GDP. From the 1960s onwards, there was a steady rise in government spending, peaking at 46 percent of GDP in the early 1980s. 10 See Van Zanden and Van Riel (2004). Due to availability, the 1816-1913 data are from the Dutch National Accounts.

9

4.6

Norway

Figure 6 indicates that Norway is the only sample country for which elite franchise reforms were plausibly associated with tangible, though minor, increases in government size. The first of these reforms occurred in 1885. Over the 1890s, government spending rose to nearly 10 percent of GDP. A more substantial, but still elite, reform took place in 1900, when almost universal and equal male suffrage rights were granted. Afterward, the share of the adult population with the right to vote rose from 17 to 35 percent, and there was a small jump in government size. Major franchise reforms occurred before the start of and during World War I. In 1909, non-universal suffrage rights were extended to women, and in 1915, nearly universal and equal suffrage rights were established for all adults. The share of the adult population with the right to vote grew from 35 to 58 percent with the first of these, and from 60 to 77 percent with the second. This set of events was associated with a moderate increase in government size, though it fell to pre-war levels by the mid-1920s. World War II was associated with the largest jump in government size. After, there was a level difference in government spending, which amounted to 15 percent or more of GDP by the 1950s. From the 1960s onwards, there was a steady rise in government spending, which peaked at 45 percent of GDP in 1992.

4.7

Sweden

Figure 7, which plots government size in Sweden from 1881 to 1998, shows that government spending remained low at roughly 7 percent of GDP through 1913. The elite reform that occurred in 1909 did not have a notable effect. Though an earlier, elite reform was made in 1866 (while the Austro-Prussian War was fought in close proximity), we cannot assess the impact of this change on government size, since the data series began in 1881. Major franchise reform, which granted voting rights to all adult men and women, occurred in the aftermath of World War I in 1921. During this conflict, there was temporary jump in government spending to nearly 20 percent of GDP, which fell in the aftermath. There was also a moderate jump at the start of the 1930s. The major increase in government size took place with World War II, when government spending reached 25 percent of GDP. Following a drop just after the war ended, government size began to steadily rise from 1950 onwards. Government spending peaked at 39 percent of GDP in 1983, and then declined through the late 1980s. By the start of the 1990s, government size had begun to rebound, rising to levels similar to those

10

of the early 1980s.

4.8

United Kingdom

There were three elite franchise reforms (1832, 1867, 1884) in the United Kingdom during the nineteenth century. As Figure 8 indicates, none of them had a notable impact on government spending, which fell from roughly 10 to 8 percent of GDP over the 1800s.11 Two short-lived jumps before 1913 were associated with military conflict: the Crimean and Anglo-Persian Wars during the 1850s and the South African War around 1900. World War I was associated with a level increase in government spending, which peaked at over 60 percent of GDP in 1916. Major franchise reforms occurred at the end or after that conflict. In 1918, nearly universal and equal male suffrage was established, increasing the share of the adult population with the right to vote from 29 to 75 percent. In 1928, the suffrage was extended to women. During the interwar period, government spending remained steady at roughly 20 percent of GDP. World War II precipitated a second large rise in government spending, which peaked at nearly 70 percent of GDP in the early 1940s. Government size fell from the 1950s to the mid-1960s. Afterward, there was a steady increase through 1980, when government size fell, rose again, and fell once more. At the end of the 1990s, government spending amounted to 37 percent of GDP.

5

Structural Breaks Tests

Our analytic narratives suggest that warfare, and not franchise extension, is the key driver of the long-run rise in government size. Wars were sometimes independently associated with increases in government spending as a share of GDP, though this outcome was more likely when major franchise reforms had already occurred. Major franchise reforms, on the contrary, were sometimes associated with increases in government size, but only if they were also associated with the First or Second World Wars. Elite franchise reforms were almost never associated with size increases. To subject these findings to a more rigorous analysis, we now turn to structural breaks tests, which assume no a priori knowledge of major turning points in the time 11

Unlike other European countries, a substantial portion of U.K. public investments prior to 1913 were made by local governments. See Lizzeri and Persico (2004).

11

series for government size, but let the data “speak” for themselves.12

5.1

Methodology

Our breaks test methodology is from Bai and Perron (2003), and identifies multiple structural changes in means while allowing for serial correlation. It thus improves upon the “moving windows” technique that relies upon sequential single structural change methods. A program created by Thomas Doan for the Regression Analysis of Time Series (RATS) software performs the Bai-Perron procedure by estimating the following regression for each sample country: EXPt = β0 +

L X

βl EXPt−l + t ,

(1)

l=1

where EXPt is central government expenditures as a share of GDP in year t, β0 through βL are parameters to be estimated, and t is the disturbance term. We allowed up to five significant yearly lags of the dependent variable (L ≤ 5). The RATS routine uses a dynamic programming algorithm to evaluate which final partitioning of the time series data achieves a global minimization of the overall sum of squared residuals. It then returns the optimal set of break points. The RATS procedure calls for the selection of a maximum number of “best” turning points in each time series subject to a minimum number of observations between data segments. Willard et al. (1996) discuss the trade-off that occurs when choosing parameter values. A minimum space of two observations sharply reduces the chance of confounding the effects of different events but ends up analyzing blips (false positives that characterize certain events as “long-lasting” that were really not) rather than turning points. Longer periods of analysis, however, increase the likelihood of missing important shifts (false negatives). To minimize the likelihood of generating false positives or negatives, we selected the best one, two, or three structural breaks with at least five, 10, or 15 observations (i.e., five, 10, or 15 years) per segment. Table 4 compares the structural breaks that the RATS routine identifies for each of the different sets of parameter values for each sample country. They are very stable. For instance, the major turning points for Denmark are always 1864, 1939, and 1982 (see Panel A). Here the only exception is 12 For historical applications, see Willard et al. (1996), Brown and Burdekin (2000), Sussman and Yafeh (2000), Mauro et al. (2002), and Dincecco (2009).

12

minor: the combination of the best three structural breaks with at least 15 observations per segment identifies 1869 rather than 1864 as the third-best break. Overall, there is at least one common break (i.e., within five years or less) among different sets of parameter values in more than 95 percent of the cases. There are at least two common breaks more than 85 percent of the time, and at least three common breaks more than 80 percent of the time.

5.2

Results

We now describe the results of the structural breaks tests for each sample country. Table 5 shows the individual results for the combination of the best three structural breaks with at least ten observations per segment. We chose this set of parameter values because it is representative of the general patterns that the structural breaks tests identify. In Denmark, two of the three best breaks coincided with wars. The 1864 break occurred in the same year as the Second Schleswig-Holstein War. Figure 1 indicates that government spending as a share of GDP rose during this conflict. During the decade after this break, however, it fell significantly by 24 percent. The next break came at the start of World War II in 1939, and led to a significant increase in government spending as a share of GDP of 45 percent. The final break (1982) occurred with the dismantlement of the post-war welfare state at the start of the 1980s. Figure 1 indicates that government size fell thereafter. Neither the elite franchise reform of 1849 or the major franchise reform of 1918 were associated with breaks in the Danish time series. In Finland, all three best breaks coincided with World Wars I or II. The first break occurred in the midst of the First World War in 1917. Though Finland was neutral during this conflict, government spending as a share of GDP increased significantly by 44 percent. The second (1939) and third (1949) breaks coincided with the start and end of the Second World War, in which Finland participated. Unsurprisingly, the former break led to a significant increase in government spending as a share of GDP by 138 percent, and the latter break to a significant fall by 38 percent. Neither the elite franchise reform of 1869 or the major franchise reform of 1907 were associated with breaks in the Finnish time series. Data gaps in the French time series during the First and Second World Wars prevent us from performing structural breaks tests for France from the early nineteenth century to the present. Hence, we conduct our analysis from 1816 to 1913, the longest stretch

13

for which continuous yearly data are available. Due to the shorter time span, we limit ourselves to the best two breaks over this period. Once more, both breaks coincided with wars. The 1869 break occurred just before the start of the Franco-Prussian War (1870-1). Similarly, the 1879 break occurred within five years of the start of the SinoFrench War (1884-5). Both conflicts led to significant increases in government spending as a share of GDP of roughly 15 percent. Neither the elite franchise reforms of 1824 and 1831 or the major franchise reform of 1848 were associated with breaks in the French time series. In Italy, two of the three best breaks coincided with major franchise extensions that took place during or near the time of World Wars I and II. The 1919 break occurred with the establishment of universal and equal male suffrage, just one year after the end of the First World War. Similarly, the break in 1945 that extended the franchise to women coincided with the last year of World War II. Neither suffrage reform, however, led to an increase in government size. In fact, the 1945 reform was associated with a significant fall in government spending as a share of GDP of 39 percent. The third break coincided with the start of the Second Italo-Ethiopian War in 1935, and was associated with a significant increase in government spending as a share of GDP of 85 percent over the next decade (which also included World War II). Finally, the elite franchise reform of 1882 was not associated with a break in the Italian time series. In the Netherlands, two of the three best breaks coincided with the start and end of World War II. The first break, which occurred in 1936, led to a significant increase in government spending as a share of GDP of 247 percent. The 1946 break, instead, was associated with a significant fall in government spending as a share of GDP of 40 percent. The final break (1969) occurred with the development of the post-war welfare state. In response, government spending as a share of GDP rose significantly by 21 percent over the next decade. Neither the elite franchise reforms of 1849, 1887, and 1897 or the major franchise reforms of 1918 and 1922 were associated with breaks in the Dutch time series. As for France, a data gap in the Norwegian time series during the Second World War prevents us from performing structural breaks tests for Norway from the early nineteenth century to the present. Hence, we conduct our analysis from the start of the time series in 1865 to 1939, the longest stretch for which continuous yearly data are available. Due to the shorter time span, we again limit ourselves to the best two breaks over this period. The first break took place in the midst of World War I in 1917, two

14

years after the establishment of nearly universal and equal voting rights for all adults. This set of events was associated with a significant increase in government spending as a share of GDP of 26 percent. The second break, which occurred in 1927, did not coincide with any known external war or franchise reform, and was not associated with any significant change in government size. Neither the elite franchise reforms of 1885 and 1900 or the major franchise reform of 1915 were associated with breaks in the Norwegian time series. In Sweden, the first break occurred at the end of World War I in 1918, three years before the establishment of universal and equal voting rights for all men and women. Over the 1920s, there was a (moderately) significant increase in government spending as a share of GDP of 33 percent. The second break coincided with the start of the Second World War in 1939, and led to a significant rise in government spending as a share of GDP of 96 percent. The third break (1964) occurred with the development of the post-war welfare state, and was associated with a significant rise in government spending as a share of GDP of 16 percent. Finally, the elite franchise reforms of 1866 or 1909 were not associated with breaks in the Swedish time series. In the United Kingdom, two of the three best breaks coincided with major franchise extensions that took place during or just after the time of World War I. The 1913 break occurred one year before the start of the First World War, and led to a significant increase in government spending as a share of GDP of 348 percent. In 1918, nearly universal and equal male suffrage was established. The second break took place in 1923, just after the end of World War I, and five years before the suffrage was extended to women. Government spending as a share of GDP, however, fell significantly by 45 percent over the 1920s. The final break (1940) came near the start of World War II, and led to a significant increase in government spending as a share of GDP of 158 percent. The elite franchise reforms of 1832, 1867, and 1882 were not associated with breaks in the British time series.

6

Robustness: Government Growth

Our main goal is to characterize structural breaks in government size levels relative to external wars and franchise reforms over the long run. For robustness, we now perform a similar set of structural breaks tests for growth in government size, which measures first differences from year to year. We estimate the following regression for each sample

15

country: ∆EXPt = β0 +

L X

βl ∆EXPt−l + t ,

(2)

l=1

where ∆EXPi ≡ EXPi − EXPi−1 is the annual change in central government expenditures as a share of GDP, β0 through βL are parameters to be estimated, and t is the disturbance term. Once more, we allowed up to five significant yearly lags of the dependent variable (L ≤ 5). Table 6 shows the results of our analysis for the combination of the best three structural breaks with at least ten observations per segment. The years marked in bold identify structural breaks that are within five years or less of the structural breaks for government size levels as described in Table 5. There is substantial overlap between the best breaks for the two time series. Nearly 70 percent of the breaks are common between them. Over 80 percent of the breaks for growth in government size, moreover, are within ten years or less of the breaks for government size levels. The breaks that the tests for growth identify that differ from those for levels by a distance of more than five years are as follows: Denmark, 1966; Finland, 1931; France, 1826; Italy, 1927; Norway, 1890 and 1901; and Sweden, 1929. Of these seven different breaks, three are within ten years or less of the structural breaks for levels (Finland, 1931; Italy, 1927; and Sweden, 1929). Each of these breaks occurred between the First and Second World Wars, and did not coincide with any known franchise reforms. Of the four different breaks that remain, the Danish break (1966) occurred with the development of the post-war welfare state in the 1960s. Perhaps surprisingly, each of the remaining three breaks coincide with elite (but not major) franchise reforms. The 1826 break in France occurred two years before and five years after the elite franchise reforms of 1824 and 1831, respectively. It also occurred two years after the Franco-Spanish War of 1823. This set of events, however, was associated with a significant fall in government spending as a share of GDP by 12 percent. The 1890 break in Norway occurred five years after the elite franchise reform of 1885, and the 1901 break one year after the elite franchise reform of 1900. Though nearly universal and equal male suffrage rights were established, the latter reform was elite in that the share of the adult population with the right to vote remained less than 40 percent. Both reforms led to significant increases in government spending as a share of GDP of over 20 percent.

16

7

Conclusion

This paper investigates the links between conflict, franchise, and government size over the long run. We examine individual time trends for eight European countries using a combination of narrative and statistical techniques. Our key methodological contribution is to perform structural breaks tests, which assume no a priori knowledge of major turning points in the government size series, but let the data “speak” for themselves. The results indicate that warfare is the key driver of long-run increases in government size. Our analysis highlights the role that conflict plays in the emergence of democratic societies. Major franchise reforms were sometimes associated with structural breaks in the time series for government size, but only if they were also associated with wars, and in particular the First and Second World Wars. This result indicates that there are important links between warfare, suffrage, and growth in government. More generally, it suggests that major shocks go hand in hand with institutional change. Though the structural breaks tests identify strong empirical relationships between conflict, franchise, and government size, they do not pinpoint casuality. The next step would be to establish the precise mechanisms by which interstate wars induce suffrage reforms. In Section 2, we described two possible channels, both of which see franchise extension as a way for governments to solve commitment problems particular to major conflicts. To determine which mechanism is valid, one useful area for future work would be to reconstruct detailed wartime histories of franchise reforms for a variety of cases. In addition to the sheer scale of government spending, conflict and democratization also have important implications for tax design.13 One may expect that, to raise greater revenues to wage war, the share of direct taxes in total taxes would rise, as elites take on greater tax burdens. We may also observe similar effects for franchise extension and redistribution. To test such hypotheses, another promising area for future work would be to use the structural breaks methodology to examine the long-run relationships between warfare, suffrage, and tax composition.

13

See Lindert (2004), Aidt and Jensen (2009a,b), and Scheve and Stasavage (2009).

17

References Acemoglu, D. and J. Robinson (2000). “Why Did the West Extend the Franchise? Democracy, Inequality, and Growth in Historical Perspective.” Quarterly Journal of Economics, 115: 1167-99. Acemoglu, D. and J. Robinson (2002). “A Theory of Political Transitions.” American Economic Review, 91: 938-63. Acemoglu, D. and J. Robinson (2006). Economic Origins of Dictatorship and Democracy. Cambridge, UK: Cambridge University Press. Aidt, T., J. Dutta, and E. Loukoianova (2006). “Democracy Comes to Europe: Franchise Extension and Fiscal Outcomes, 1830-1938.” European Economic Review, 50: 249-83. Aidt, T. and Jensen, P. (2009a). “Tax Structure, Size of Government, and the Extension of the Voting Franchise in Western Europe, 1860-1938.” International Tax and Public Finance, 16: 362-94. Aidt, T. and Jensen, P. (2009b). “The Taxman Tools Up: An Event History Study of the Introduction of the Personal Income Tax, 1815-1938.” Journal of Public Economics, 93: 160-75. Andreski, S. (1968). Military Organization and Society. London: Routledge. Bai, J. and P. Perron (2003). “Computation and Analysis of Multiple StructuralChange Models.” Journal of Applied Economics, 18: 1-22. Barro, R. (1979). “On the Determination of Public Debt.” Journal of Political Economy, 87: 940-71. Barro, R. (1987). “Government Spending, Interest Rates, Prices, and Budget Deficits in the United Kingdom, 1701-1918.” Journal of Monetary Economics, 20: 221-48. Barro, R. (1989). “The Neoclassical Approach to Fiscal Policy.” In R. Barro, ed., Modern Business Cycle Theory, Cambridge, UK: Cambridge University Press. Besley, T. and T. Persson (2008). “Wars and State Capacity.” Journal of the European Economic Association, 2: 522-30.

18

Besley, T. and T. Persson (2009). “The Origins of State Capacity: Property Rights, Taxation, and Politics.” American Economic Review, 99: 1218-44. Besley, T. and T. Persson (2010). “State Capacity, Conflict, and Development.” Econometrica, 78: 1-34. Black, D. (1948). “On the Rationale of Group Decision-making.” Journal of Political Economy, 56: 23-34. Boix, C. (2003). Democracy and Redistribution. Cambridge, UK: Cambridge University Press. Brown, W. and R. Burdekin (2000). “Turning Points in the U.S. Civil War: A British Perspective.” Journal of Economic History, 60: 216-31. Carstairs, A. (1980). A Short History of Electoral Systems in Western Europe. London: George Allen. Clodfelter, M. (2002). Warfare and Armed Conflicts: A Statistical Reference, 15002000. Jefferson: McFarland. Conley, J. and A. Temimi (2001). “Endogenous Enfranchisement When Group Preferences Conflict.” Journal of Political Economy, 109: 79-102. Dincecco, M. (2009). “Fiscal Centralization, Limited Government, and Public Revenues in Europe, 1650-1913.” Journal of Economic History, 69: 48-103. Doan,

T.

(2009).

“Bai-Perron

Procedure

for

RATS

Software.”

Estima,

www.estima.com. Downs, A. (1957). An Economic Theory of Democracy. New York: Harper and Row. Flora, P. (1983). State, Economy, and Society in Western Europe, 1815-1975, Volume I. Frankfurt: Campus Verlag. Gradstein, M. (2006). “Inequality, Democracy, and the Protection of Property Rights.” Economic Journal, 117: 252-69. Greif, A. (2006). Institutions and the Path to the Modern Economy. Cambridge, UK: Cambridge University Press.

19

Halperin, S. (2004). War and Social Change in Modern Europe. Cambridge, UK: Cambridge University Press. Hoffman, P. and J.L. Rosenthal (1997a). “The Political Economy of Warfare and Taxation in Early Modern Europe: Historical Lessons for Economic Development.” In J. Drobak and J. Nye, eds., The Frontiers of the New Institutional Economics, St. Louis: Academic Press. Hoffman, P. and J.L. Rosenthal (1997b). “Divided We Fall: The Political Economy of Warfare and Taxation.” Mimeo, California Institute of Technology. Husted, T. and L. Kenny (1997). “The Effect of the Expansion of the Voting Franchise on the Size of Government.” Journal of Political Economy, 105: 54-82. Jack, W. and Lagunoff, R. (2006). “Dynamic Enfranchisement.” Journal of Public Economics, 90: 551-72. Kenny, L. and J. Lott (1997). “Did Women’s Suffrage Change the Size and Scope of Government?” Journal of Political Economy, 107: 1163-98. Lindert, P. (2004). Growing Public: Social Spending and Economic Growth Since the Eighteenth Century, Volume I. Cambridge, UK: Cambridge University Press. Lizzeri, A. and N. Persico (2004). “Why Did the Elite Extend the Suffrage? Democracy and the Scope of Government with an Application to Britain’s Age of Reform.” Quarterly Journal of Economics, 119: 707-65. Llavador, H. and R. Oxoby (2005). “Partisan Competition, Growth, and the Franchise.” Quarterly Journal of Economics, 120: 1155-92. Mann, M. (1988). States, War, and Capitalism. Oxford, UK: Blackwell. Marwick, A. (1974). War and Social Change in the Twentieth Century. London: Collier-MacMillan. Mauro, P., N. Sussman and Y. Yafeh (2002). “Emerging Market Spreads: Then Versus Now.” Quarterly Journal of Economics, 117: 695-733. McKinley, S. (1934). Democracy and Military Power. New York: Vanguard Press.

20

Meltzer, A. and S. Richard (1981). “A Rational Theory of the Size of Government.” Journal of Political Economy, 89: 914-27. Mitchell, B. (2007). International Historical Statistics: Europe, 1750-2005. New York: Palgrave Macmillan. National Accounts of the Netherlands. www.nationalaccounts.niwi.knaw.nl. O’Brien, P. (2005). “Fiscal and Financial Pre-conditions for the Rise of British Naval Hegemony: 1485-1815.” Mimeo, London School of Economics. Persson, T. and G. Tabellini (2000). Political Economics. Cambridge, MA: MIT Press. Persson, T. and G. Tabellini (2003). The Economic Effects of Constitutions. Cambridge, MA: MIT Press. Porter, B. (1994). War and the Rise of the Modern State. New York: Free Press. Roberts, K. (1977). “Voting Over Income Tax Schedules.” Journal of Public Economics, 8: 329-40. Romer, T. (1975). “Individual Welfare, Majority Voting, and the Properties of a Linear Income Tax.” Journal of Public Economics, 7: 163-8. Sarkees, M. (2000). “The Correlates of War Data on War: An Update to 1997.” Conflict Management and Peace Science, 18: 123-44. Scheve, K. and D. Stasavage (2009). “The Conscription of Wealth: Mass Warfare and the Demand for Progressive Taxation.” Mimeo, New York University. Sussman, N. and Y. Yafeh (2000). “Institutions, Reforms, and Country Risk: Lessons from Japanese Government Debt in the Meiji Era.” Journal of Economic History, 60: 442-67. Ticchi, D. and A. Vindigni (2009). “War and Endogenous Democracy.” Mimeo, Princeton University. Tilly, C. (1990). Coercion, Capital, and European States, 990-1990. Cambridge, UK: Blackwell.

21

Van Zanden, J.L. and A. Van Riel (2004). The Strictures of Inheritance: The Dutch Economy of the Nineteenth Century. Princeton: Princeton University Press. Weber, M. (1950). General Economic History. Glencoe: Free Press. Willard, K., T. Guinnane and H. Rosen (1996). “Turning Points in the Civil War: Views from the Greenback Market.” American Economic Review , 86: 1001-18.

22

Figure 1: Government Spending as a Share of GDP, Denmark, 1854-1998

0.60

War, 1848-9 Elite Franchise Reform, 1848

0.50

World War I, 1914-8 Major Franchise Reform, 1918

War, 1864

World War II, 1939-45

0.40

0.30

0.20

0.10

Sources: See text.

23

1994

1984

1974

1964

1954

1944

1934

1924

1914

1904

1894

1884

1874

1864

1854

1844

0.00

Figure 2: Government Spending as a Share of GDP, Finland, 1882-1996

0.60

Elite Franchise Reform, 1869

Major Franchise Reform, 1907

0.50

World War I, 1914-8

World War II, 1939-45

0.40

0.30

0.20

0.10

Sources: See text.

24

1992

1982

1972

1962

1952

1942

1932

1922

1912

1902

1892

1882

1872

1862

0.00

Figure 3: Government Spending as a Share of GDP, France, 1816-1993

0.30

War, 1823 Elite Franchise Reforms, 1824, 1831

0.25

Major Franchise Reform, 1848 Wars, 1849, 1854-6, 1859, 1862-7

0.20

Wars, 1870-1, 1884-5, 1893

War, 1900

World War II, 1939-45, Major Franchise Reform, 1945 Wars, 1951-3, 1956

World War I, 1914-8 War, 1919-21

War, 1991

0.15

0.10

0.05

Sources: See text.

25

1996

1986

1976

1966

1956

1946

1936

1926

1916

1906

1896

1886

1876

1866

1856

1846

1836

1826

1816

0.00

Figure 4: Government Spending as a Share of GDP, Italy, 1862-1998

0.60

Wars, 1848-9, 1855-6, 1859, 1860-1

0.50

War, 1866

Elite Franchise Reform, 1882

0.40

Major Franchise Reforms, 1913, 1919 World War I, 1914-8 War, 1911-2

War, 1935-6

World War II, 1939-45 Major Franchise Reform, 1946

War, 1991

0.30

0.20

0.10

Sources: See text.

26

1992

1982

1972

1962

1952

1942

1932

1922

1912

1902

1892

1882

1872

1862

1852

1842

0.00

Figure 5: Government Spending as a Share of GDP, The Netherlands, 1816-1998

1.10

Elite Franchise Reform, 1849

1.00

Elite Franchise Reforms, 1887, 1897

0.90 0.80

World War I, 1914-8 Major Franchise Reforms, 1918, 1922

World War II, 1939-45

War, 1951-3

0.70 0.60 0.50 0.40 0.30 0.20 0.10

Sources: See text.

27

1996

1986

1976

1966

1956

1946

1936

1926

1916

1906

1896

1886

1876

1866

1856

1846

1836

1826

1816

0.00

Figure 6: Government Spending as a Share of GDP, Norway, 1865-1997

0.50

Elite Franchise Reform, 1885

Elite Franchise Reform, 1900

World War I, 1914-8 Major Franchise Reforms, 1909, 1915

0.40

World War II, 1939-45

0.30

0.20

0.10

Sources: See text.

28

1995

1985

1975

1965

1955

1945

1935

1925

1915

1905

1895

1885

1875

1865

0.00

Figure 7: Government Spending as a Share of GDP, Sweden, 1881-1998

0.40

Elite Franchise Reform, 1866

0.35

Elite Franchise Reform, 1909

World War II, 1939-45

World War I, 1914-8 Major Franchise Reforms, 1921

0.30

0.25

0.20

0.15

0.10

0.05

Sources: See text.

29

1991

1981

1971

1961

1951

1941

1931

1921

1911

1901

1891

1881

1871

1861

0.00

Figure 8: Government Spending as a Share of GDP, United Kingdom, 1830-1998

0.70

Elite Franchise Reform, 1832

0.60

0.50

Elite Franchise Reform, 1867

War, 1882 Elite Franchise Reform, 1884

War, 1900

WWI, 1914-8, Major Franchise Reforms, 1918, 1928

Wars, 1950-3, 1956

World War II, 1939-45

Wars, 1982, 1991

Wars, 1854-6, 1856-7

0.40

0.30

0.20

0.10

Sources: See text.

30

1990

1980

1970

1960

1950

1940

1930

1920

1910

1900

1890

1880

1870

1860

1850

1840

1830

1820

0.00

Table 1: Descriptive Statistics for Government Size

Totals

Denmark Finland France Italy Netherlands Norway Sweden United Kingdom

1816-1998

Pre-WWI

1914-1945

Post-WWII

No of Obs

1215

506

248

461

Mean

0.19

0.10

0.21

0.29

Std Dev

0.13

0.03

0.16

0.09

Min

0.03

0.03

0.04

0.10

Max

1.09

0.21

1.09

0.57

No of Obs

145

60

32

53

Mean

0.15

0.08

0.07

0.28

No of Obs

115

32

32

51

Mean

0.21

0.09

0.22

0.28

No of Obs

162

98

19

45

Mean

0.15

0.11

0.20

0.22

No of Obs

137

52

32

53

Mean

0.23

0.13

0.30

0.29

No of Obs

183

98

32

53

Mean

0.21

0.13

0.27

0.34

No of Obs

127

49

32

53

Mean

0.21

0.13

0.27

0.34

No of Obs

118

33

32

53

Mean

0.18

0.07

0.13

0.28

No of Obs

169

84

32

53

Mean

0.21

0.08

0.33

0.28

Source: Mitchell (2007). Notes: Government size is central government spending as a share of GDP.

31

Table 2: Wars in Europe, 1816-1998 Denmark

1848-9, 1864

Finland

1939-40, 1941-4

France

1823, 1849, 1854-6, 1859, 1862-7, 1870-1, 1884-5, 1893, 1900, 1914-8, 1919-21, 1939-40, 1940-1, 1944-5, 1951-3, 1956, 1991

Italy

1848-9, 1855-6, 1859, 1860, 1860-1, 1866, 1911-2, 1915-8, 1935-6, 1940-3, 1943-5, 1991

Netherlands

1940, 1951-3

Norway

1940

Sweden

No interstate wars

United Kingdom

1854-6, 1856-7, 1882, 1900, 1914-8, 1939-45, 1950-3, 1956, 1982, 1991

Source: Correlates of War Database of Sarkees (2000). Notes: All interstate wars are included.

32

Table 3: Democratization in Europe Panel A: Elite Franchise Reforms Denmark

1849

Finland

1869

France

1824, 1831

Italy

1882

Netherlands

1849, 1887, 1897

Norway

1885, 1900

Sweden

1866, 1909

United Kingdom

1832, 1867, 1884 Panel B: Major Franchise Reforms

Denmark

1918

Finland

1907

France

1848, 1945

Italy

1913, 1919, 1946

Netherlands

1918, 1922

Norway

1909, 1915

Sweden

1921

United Kingdom

1918, 1928

Source: Flora (1980). Notes: After elite franchise reforms, which were non-universal and/or unequal, less than 40 percent of the adult population had the right to vote. Major reforms, in contrast, were both universal and equal, and granted the suffrage to all adult men and/or women.

33

Table 4: Comparison of Best 1, 2, or 3 Breaks in Government Size with 5, 10, or 15 Minimum Observations per Segment Best 1 5

10

Best 2 15

5

10

Best 3 15

5

10

15

Panel A: Denmark, 1854-1998 1982

1982

1982

1982

1982

1982

1982

1982

1982

1939

1939

1939

1939

1939

1939

1864

1864

1869

Panel B: Finland, 1882-1996 1941

1941

1941

1939

1939

1938

1939

1939

1938

1947

1949

1953

1949

1949

1953

1944

1917

1919

Panel C: France, 1816-1913 1874

1874

1874

1869

1869

1869

1878

1879

1887

Panel D: Italy, 1862-1998 1917

1917

1917

1935

1935

1919

1919

1919

1919

1941

1945

1939

1941

1945

1950

1935

1935

1935

Panel E: Netherlands, 1816-1998 1838

1944

1944

1946

1946

1946

1946

1946

1944

1939

1936

1930

1939

1936

1928

1822

1969

1964

Panel F: Norway, 1865-1939 1917

1917

1917

1917

1917

1917

1923

1927

1890

Panel G: Sweden, 1881-1998 1930

1930

1930

1930

1930

1931

1930

1939

1939

1918

1918

1916

1918

1918

1918

1992

1964

1964

34

Table 4, Continued: Best Breaks in Government Size with Different Minimum Observations per Segment Best 1 5

10

Best 2 15

5

10

Best 3 15

5

10

15

Panel H: United Kingdom, 1830-1998 1915

1915

1915

1939

1930

1923

1915

1913

1910

1945

1940

1940

1945

1940

1940

1939

1923

1925

Notes: Government size is central government spending as a share of GDP.

35

Table 5: Best 3 Breaks in Time Series for Government Size with 10 Minimum Observations per Segment Year

% Change

Event Panel A: Denmark, 1854-1998

1864

−24%∗∗

1939

+45%∗∗∗ (5.09)

Start of World War II (1939-45)

1982

+9% (1.29)

Dismantlement of post-war welfare state (1980s)

(2.16)

2nd Schleswig-Holstein War (1864)

Panel B: Finland, 1882-1996 1917

+44%∗∗∗ (6.41)

World War I (1914-8)

1939

+138%∗∗∗

Start of World War II (1939-45)

1949

−38%∗∗∗ (3.89)

(6.02)

End of World War II (1939-45) Panel C: France, 1816-1913

1869

+15%∗∗ (2.29)

Start of Franco-Prussian War (1870-1)

1879

+16%∗∗∗

Start of Sino-French War (1884-5)

(3.31)

Panel D: Italy, 1862-1998 1919

+6% (0.25)

Major franchise reform (1919) / End of WWI (1914-8)

1935

+85%∗∗∗

Start of 2nd Italo-Ethiopian War (1935-6)

1945

−39%∗∗∗ (4.02)

(4.44)

End of WWII (1939-45) / Major franchise reform (1946) Panel E: Netherlands, 1816-1998

1936

+247%∗∗∗ (3.41)

Start of World War II (1939-45)

1946

−40%∗

End of World War II (1939-45)

1969

+21%∗∗∗

1917

+26%∗∗

1927

−11% (1.39)

(1.88) (3.19)

Development of post-war welfare state (1960s) Panel F: Norway, 1865-1939

(2.27)

World War I (1914-8) / Major franchise reform (1915) Unidentified event Panel G: Sweden, 1881-1998

1918

+33%∗ (2.05)

World War I (1914-8) / Major franchise reform (1921)

1939

+96%∗∗∗

(6.86)

Start of World War II (1939-45)

1964

+16%∗∗∗

(5.54)

Development of post-war welfare state (1960s)

* significant at 10%, ** significant at 5%, *** significant at 1%

36

Table 5, Continued: Best 3 Breaks for Government Size with 10 Minimum Observations per Segment Year

% Change

Event Panel H: United Kingdom, 1830-1998

1913

+348%∗∗∗ (4.35)

Start of WWI (1914-8) / Major franchise reform (1918)

1923

−45%∗∗

Major franchise reforms (1918, 1928) / End of WWI (1914-8)

1940

+158%∗∗∗

(2.50) (6.83)

World War II (1939-45)

* significant at 10%, ** significant at 5%, *** significant at 1% Notes: The first column shows the years for the 3 best structural breaks with 10 minimum observations according to the algorithm described in the text. The second column shows the percentage change in average government size for the decades before and after each break. T-statistics in absolute values are in parentheses. The third column shows brief “explanations” for the turning points, which the text elaborates upon. Government size is central government spending as a share of GDP.

37

Table 6: Best 3 Breaks in Time Series for Growth in Government with 10 Minimum Observations per Segment Year

% Change

Event Panel A: Denmark, 1854-1998

1864

−24%∗∗

1966

+79%∗∗∗

1982

+9% (1.29)

(2.16) (5.72)

2nd Schleswig-Holstein War (1864) Development of post-war welfare state (1960s) Dismantlement of post-war welfare state (1980s) Panel B: Finland, 1882-1996

1931

+15% (0.75)

Unidentified event

1941

+89%∗∗∗

World War II (1939-45)

1951

−38%∗∗∗ (4.04)

1826

−12%∗∗

1869

+15%∗∗ (2.29)

(3.66)

End of World War II (1939-45) Panel C: France, 1816-1913

(2.43)

Elite franchise reforms (1824, 1831) / Franco-Spanish War (1823) Start of Franco-Prussian War (1870-1) Panel D: Italy, 1862-1998

1917

+89%∗∗∗ (2.82)

World War I (1914-8) / Major franchise reform (1919)

1927

−25% (1.36)

Unidentified event

1939

+21% (1.18)

Start of WWII (1939-45) / End of 2nd Italo-Ethiopian War (1935-6)

1936

+247%∗∗∗

1946

−40%∗ (1.88)

End of World War II (1939-45)

1968

+29%∗∗∗

Development of post-war welfare state (1960s)

1890

+23%∗∗∗

(4.68)

Elite franchise reform (1885)

1901

+21%∗∗∗ (4.73)

Elite franchise reform (1900)

Panel E: Netherlands, 1816-1998 (3.41)

(5.80)

Start of World War II (1939-45)

Panel F: Norway, 1865-1939

Panel G: Sweden, 1881-1998 1918

+33%∗ (2.05)

World War I (1914-8) / Major franchise reform (1921)

1929

+21%∗∗

Unidentified event

1941

+74%∗∗∗

(2.91) (6.21)

World War II (1939-45)

* significant at 10%, ** significant at 5%, *** significant at 1%

38

Table 6, Continued: Best 3 Breaks for Growth in Government with 10 Minimum Observations per Segment Year

% Change

Event Panel H: United Kingdom, 1830-1998

1915

+282%∗∗∗

1928

−16% (1.23)

Major franchise reform (1923)

1940

+158%∗∗∗

World War II (1939-45)

(4.28) (6.83)

Start of WWI (1914-8) / Major franchise reform (1918)

* significant at 10%, ** significant at 5%, *** significant at 1% Notes: The years marked in bold identify structural breaks that are within five years or less of the structural breaks for government size levels as described in Table 5. Growth in government is the annual change in central government spending as a share of GDP. For further details, see Table 5.

39

War, Democracy, and Government Size Over the Long ...

Apr 14, 2010 - methods. A program created by Thomas Doan for the Regression Analysis of Time. Series (RATS) software performs the Bai-Perron procedure ...

276KB Sizes 3 Downloads 195 Views

Recommend Documents

Does Direct Democracy Reduce the Size of Government?
Jun 9, 2009 - two questions: does direct democracy reduce government spending? .... expenditures exceed a certain monetary threshold (which is defined.

the size and functions of government and economic ...
Grossman (1988), Kormendi and Meguire (1985), Landau (1983, 1986), Peden (1991), Peden and Bradley (1989), and Scully. (1992, 1994). These prior studies ...

Optimal Size of the Government and Imperfect ...
His model does not take into account any other (non#technical) explanatory ...... Therefore we can restrict our study to the open interval /z/.0, &'. ...... [10] Grossman, P. [1988b], lGrowth in Government and Economic Growth: the Australian Experien

The War Over Israel's Influence.pdf
The War Over Israel's Influence.pdf. The War Over Israel's Influence.pdf. Open. Extract. Open with. Sign In. Main menu. Displaying The War Over Israel's ...

Understanding the Size of the Government Spending Multiplier: It's in ...
meterizing the impulse response functions with a small number of Gaussian functions, which offers effi ...... (contractionary) multiplier was close to linear: For small business cycle ..... Evidence from 20th Century Historical Data" American Eco-.

Understanding the Size of the Government Spending ...
and Zubairy, 2016) contains larger and more numerous positive shocks than negative ... (according to our results) do not depend on the state of the business cycle. ..... trend for each variable.20 The error bands cover 90 percent of the posterior ...

Understanding the Size of the Government Spending ... -
do not necessarily reflect those of the Federal Reserve Banks of Richmond and San ...... (contractionary) multiplier was close to linear: For small business cycle ...

Understanding the Size of the Government Spending Multiplier: It's in ...
May 2016. Abstract. Despite intense scrutiny, estimates of the government spending ..... impulse responses may be different depending on the state of the business ..... trend for each variable.18 The error bands cover 90 percent of the posterior ...

Jazz as Democracy? Dave Brubeck and Cold War ... -
ing Jazz Goes to College (1954), Brubeck Time (1955), Jazz: Red, Hot, and. Cool (1955), and ...... anists (Bloomington: Indiana University Press, 2000), 139.

Government Size, Openness, and Output Growth
JEL Classification Numbers: F32, H11, H50, O47. ..... countries that liberalized their stock markets between 1986 and 1993, including. Argentina, Brazil, Chile ...

Government size and macroeconomic stability in life ...
Business Cycle Volatility of Employment by Age Group, OECD ... Canada, Czech Republic, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Japan, South Korea, Netherlands, New Zealand, .... Small open economy: given.

Long-Term Government Debt and Household ... - HEC Lausanne - Unil
risk and high-return assets like stocks than they do according to survey data. ..... 42 The data on real US stock returns is taken from Robert Shiller's online database, which is freely ..... School diploma .... In the data, having earned a high-scho

Long-Term Government Debt and Household Portfolio ... - HEC - Unil
∗Department of Economics, Faculty of Business and Economics (HEC), University .... (2016), Gomes and Michaelides (2005) and Haliassos and Michaelides (2003).3 ..... cyclical fluctuations to time effects and trends to age and cohort effects.

Long-Term Government Debt and Household Portfolio ...
Early contributions by Merton (1969, 1971) and Samuelson (1969) analyse optimal portfolio choice neglecting the asset market participation decision. More recent ... category that mainly includes cash/liquidity, short-term sovereign debt and corporate

United-States-Government-Democracy-In-Action-Student-Edition.pdf ...
total on the web electronic digital local library that offers access to many PDF e-book collection. You might find many. different types of e-reserve along with other literatures from my files database. In case you happen to be planning to understand

Extractors for Polynomials Sources over Constant-Size ...
Sep 22, 2011 - In this work, we construct polynomial source extractors over much smaller fields, assuming the characteristic of the field is significantly smaller than the field size. Theorem 1 (Main — Extractor). Fix a field Fq of characteristic p