EHS Conference Version (Preliminary).
Fiscal Centralization, Limited Government, and Public Finances in Europe, 1650-1914
5 December 2006
Abstract: In 1650, most European states suffered from fiscal fragmentation and absolutist rule, though by 1914 many of these same countries were fiscally centralized, with limited government. This paper examines the evolution of fiscal institutions within European states over this period, focusing on two common political transformations: fiscal centralization and limited government. Using a newly assembled panel data set of per-capita revenues and expenditures for five of the most important European players at the time – Britain, France, the Netherlands, Prussia, and Spain – I carry out a systematic cross-country analysis of the public finance effects of these institutional changes. The results indicate a significant increase in per-capita revenues and expenditures associated with each political transformation, even after controlling for a variety of other political and economic factors that influenced public finance trends over this period.
JEL Classification: N23, P48, O16
In 1650, most European countries faced two fiscal problems. On one hand, nearly all rulers had absolute fiscal discretion, meaning that they were able to spend state funds as they wished. As a result, rulers often chose personal consumption (e.g. foreign military adventures) over public services that would most benefit society (e.g. roads). To solve this problem, the power of rulers had to be limited. On the other hand, European states also suffered from fiscal fragmentation, meaning that central governments were forced to bargain with numerous local bodies over tax amounts. Indeed, each locality attempted to free-ride on the contributions of others, leaving states starved for revenues. To solve this problem, tax systems had to be centralized, enabling single authorities to take control of collection.
Much of the contemporary literature on political institutions and economic outcomes centers on the problem of absolute fiscal discretion, suggesting that limited government itself matters to economic development.1 Yet as Acemoglu (2005) points out, wealthy economies are typically supported by powerful states, where central governments are able to raise large tax amounts and play a valuable role in the economy. For example, as figure 1 shows, there is a strong positive relationship between central government tax revenues and per-capita GDP, suggesting that the state’s ability to tax matters to economic welfare.
Indeed, fiscal fragmentation remains a serious issue in poor parts of the world. Herbst (2000), among others, has argued that divided fiscal authority and economic stagnation are directly linked for many African nations.2 Dictatorial regimes also plague several of the same countries. In such cases, there is a direct link with pre-19th century European states, when rulers had great control over expenditures, but less command over revenues.
One useful way to improve our understanding of the effects of absolutism and fragmentation on public finances is to examine the evolution of state fiscal systems over the long term. In this regard, Europe is ideal to study. First, there was a clear pattern of political and economic changes from 1650 to 1914, as European states replaced Old Regime institutions with modern ones. Second, European forms of fiscal governance have been imitated throughout the world, making such an inquiry valuable.
With that said, much of the literature on the historical political economy of Europe suffers from another sort of one-sidedness, attempting by induction to generalize the results of specific case studies. Hoffman and Norberg (1994), Bonney (1999), and Bordo and Cortès-Conde (2001), among others, describe the fiscal evolution of European states in intimate detail, focusing on case histories and dividing up states by chapter. There is a cost to this painstaking approach, however. In particular, the emphasis on particular cases makes it difficult to convincingly generalize any findings, since no systematic effort has been made to do so.3
For instance, based on their reading of English history, North and Weingast (1989) claim that oversight by parliament after the Glorious Revolution of 1688 enabled the crown to make a credible commitment to responsible fiscal policies.4 Following this political transformation, the state had access to more plentiful, cheaper loans, because investors had greater confidence in the ruler’s ability to repay them. Since the time of the article, many other scholars have examined the link between limited government and public debt.5 This line of research, however, has overlooked a more basic point, which is the direct impact of limited government on state revenues and expenditures.
Moreover, England was centralized from early medieval times, making it anomalous among European states.6 Elsewhere, there was divided fiscal authority, meaning that central governments had to negotiate with local bodies over tax revenues. Indeed, the exceptional nature of English political and economic institutions calls the causal link between limited government and improvements in public finance into question.
In this light, Epstein (2000) uses the financial histories of pre-modern Italian states to argue that North and Weingast conflate the economic benefits of limited government with those of institutional centralization, claiming that political and economic fragmentation within states rather than fiscal abuse by rulers was the main cause of economic distortions in Europe prior to the 19th century.7 Certainly, in this context it makes less sense to discuss the fiscal power of rulers, which falsely presupposes the existence of strong, centralized institutions.
On the Continent, centralization of political and economic institutions often occurred during the French Revolutionary and Napoleonic eras (1789-1815). Afterwards, many European states more closely resembled pre-1688 England, possessing centralized but absolutist institutions. Hence, for the 19th century it is no longer anachronistic to consider how constitutional constraints on rulers influenced public finances.
Indeed, new costs arose when European rulers gained undivided fiscal authority. Even in absolutist regimes, organized bodies such as parliament exercised control over taxation, inducing rulers to seek extra-legal means to increase revenues, which they used to wage foreign wars and stifle internal opposition. Following fiscal centralization, rulers may have abused newfound powers to further undermine private property rights, making it easier for them to pursue larger revenues. In fact, this is just the sort of problem that North and Weingast envisioned when describing the virtues of limited government, which constrained the behavior of rulers.
In this paper, I perform a systematic cross-country analysis of the effects of fiscal centralization and limited government on per-capita central government revenues and expenditures in Europe from 1650 to 1914, finding that both political transformations improved public finances. To measure these impacts, I first identify fiscal centralization and limited government within European states, each of which may be dated with relative precision. Next, I use regression techniques to carry out an analysis of the public finance effects of these political transformations on the assembled panel, which is comprised of data for five of the most important players in Europe at the time: Britain, France, the Netherlands, Prussia, and Spain.8 For robustness, controls are implemented for a variety of other political and economic factors that also influenced public finance trends in Europe over this period. To round out the analysis, afterwards I break up the panel and discuss the French and Dutch cases in detail. In total, the results indicate that fiscal centralization and limited government each led to significant increases in per-capita revenues and expenditures, suggesting that both political transformations mattered to public finance improvements.
The rest of the paper proceeds as follows. Section 2 discusses the possible effects of fiscal centralization and limited government on public finances within European states from 1650 to 1914. Section 3 examines the relative levels of per-capita revenues and expenditures that we would expect to see associated with each of the three types of political regimes that existed in Europe at the time. Section 4 discusses the data used and the sample states selected. Section 5 performs the econometric tests and section 6 examines the two case studies. Section 7 concludes.
2. Political Transformations
In 1650, most European states suffered from fiscal fragmentation and absolutist rule, though by 1914 many of these same countries were fiscally centralized, with limited government. In this section, I discuss the possible effects of political transformations on public revenues and expenditures in Europe over this period. The first part of this section lays out the historical framework within which to understand problems of fragmentation and absolutism in Europe over the 18th and 19th centuries. The second part examines fiscal fragmentation in greater detail and dates centralization within European states. Finally, the third part takes up the problem of absolutism and dates the rise of limited government.
2.1 Historical Framework
To start, we must clearly distinguish between divided fiscal authority and absolute fiscal discretion. In particular, divided fiscal authority was a problem that encompassed geographical space, involving negotiations over tax revenues between central and local governments. On the other hand, absolute fiscal discretion was a problem contained within the central government itself, involving interactions between the ruler and organized national bodies like parliament over fiscal policy.9
Though distinct, both problems were closely linked before the 19th century. Fragmented sovereignty in the form of powerful local laws, assemblies, and notables constrained the predatory capacities of European absolutists, suggesting that divided fiscal authority “limited” the ruler’s power in this regard.10 Once tax deals between central and local bodies had been struck and revenues collected, however, European absolutists had great control over how to spend state funds.
Indeed, my definition of limited government refers exclusively to the problem of absolute discretion, meaning that a political regime could have been divided in terms of fiscal authority but unlimited in terms of fiscal policy.11 Following North and Weingast (1989), a state was limited if and only if there was a constitutional arrangement institutionalizing parliament’s ability to constrain the ruler’s spending habits. Otherwise, many sorts of institutional arrangements could be considered limited, since most governments faced a variety of political and economic constraints. In section 2.5, I will make this parliamentary definition of limited government more explicit.
Before the 1800s, the predatory urges of European absolutists were stifled by a plethora of local obstacles, often making it is difficult to draw a meaningful distinction between the terms “limited” and “absolutist.” As a result of French revolution and conquest, however, European fiscal structures underwent radical change by the end of the Napoleonic era (1815). In particular, most systems were rationalized, vesting a great deal more power in the central government. Hence, at this institutional juncture the notion of limited government in terms of parliamentary constitutional constraints on rulers becomes relevant.12
Indeed, one might argue that in terms of asymmetric information limited government represented a notable institutional improvement upon previous credible commitment mechanisms. Beforehand, it was uncommon for European governments to release financial data on revenues and debt levels to parliament or to the public at large. With the emergence of limited government in the 19th century, however, these statistics became widely available, reducing the amount of private information held by rulers.
2.2 Fiscal Fragmentation
Prior to the 19th century, most European states were fiscally fragmented. Indeed, Dincecco (2006a) has performed a quantitative cross-country analysis of fragmentation and centralization in Europe from 1700 to 1871, finding that fiscal zones were small as a result of divided fiscal authority through the start of the French Revolution (1789).
The qualitative evidence supports this conclusion. In France, for example, the tax structure was antiquated and inefficient. Hemmed in by powerful local authorities, the French crown had to negotiate independently over tax amounts with regional bodies. Based upon the particular deal struck, the state applied uneven fiscal pressure across its domain. Whole towns and provinces often avoided certain duties, as did elites. From the middle of the 15th century onwards, for instance, nobles in central and northern France were exempt from the taille, an important land tax. Southern nobles only paid the taille on certain holdings. In time, royal officers won comparable exemptions. Other privileged groups, such as urban elites, avoided sales taxes.13
The tax scheme in Spain was equally complicated, where it was impossible for the various kingdoms united under the sovereign to agree upon a standard tax system. Hence, the Spanish crown had to implement new taxes on top of old ones, rather than make much-needed structural changes. Unable to overcome regional resistance, for instance, Bourbon reformers in the early 1700s chose to superimpose additional taxes in Spain’s eastern provinces, called the catastro in Catalonia, the contribucìon ùnica in Aragon, and the equivalente in Valencia. As in France, tax rates varied across regions, dependent upon the bargain made. Each tax was overseen by separate administrations, which sub-contracted out tax collection to local individuals, reducing tax yields.14 Surprisingly, 18th century per-capita taxes were actually lower in France and Spain, which were ruled by absolute monarchs, than in Britain, where parliamentary institutions held sway, indicating the magnitude of the problem of divided fiscal authority.15
In these fragmented states, the power of local authorities to tax was closely intertwined with political autonomy. In turn, local elites resisted fiscal reforms that threatened their incumbent tax rights, regardless of the negative effect on central government revenues. The result was a classic public goods problem, since each locality attempted to free-ride on the tax contributions of others. In turn, per-capita revenues collected by central governments remained low.
2.3 Fiscal Centralization
Fiscal centralization within European states was generally a centuries-long process which included all of the measures taken to assess, collect, and seize control of tax revenues. In many European countries, however, fundamental structural changes to political and economic institutions were imposed during French Revolutionary and Napoleonic times (1789-1815), meaning that there was a remarkable difference in the degree of centralization before and after this era. Indeed, Dincecco (2006a) also finds a large systematic increase in the size of fiscal zones during this period, which he attributes to fiscal centralization within states.
To make systematic comparisons across states possible, I have chosen a simple definition of fiscal centralization, which occurred when the state’s central government secured its revenues through a tax system with uniform rates throughout the country.16 Table 1 indicates that this political transformation took place swiftly and permanently on the Continent from 1789 to 1815.17 In France itself, the introduction of a national tax system occurred during the Revolutionary decade of the 1790s. The French conquered the Dutch Republic in 1795. Afterwards, its federalist-style fiscal and legal institutions were replaced with centralized ones.18 Indeed, the simple threat of takeover by France was at times enough to instigate institutional change. For instance, Prussia made quick fiscal and legal reforms after French defeat in battle in 1806.19
Two exceptions bear mention, however. As discussed, England was fiscally centralized from very early onwards. On the other hand, Napoleon did little to implement political and economic reforms on the Iberian Peninsula. Hence, fiscal centralization in Spain did not occur until the 1840s. For additional details, refer to appendix 1.
With undivided fiscal authority, the ruler overcame the free-riding problem associated with fragmentation, because all localities had to adhere to national tax rates. Hence, per-capita revenues rose so long as tax standards were equalized at relatively high levels. So long as additional revenues were spent, per-capita expenditures would increase as well.
Even in absolutist regimes where spending was decided entirely by royal administrations, organized bodies like parliament exercised some control over taxation, prompting rulers to find alternative revenue sources, including through predatory means. Such strategies were not simply a response to divided fiscal authority. For example, in centralized England, King Charles I (1625-1649) pursued a variety of extra-legal revenues to bypass parliament. One source was loans under threat (i.e. “forced loans”), which were repaid in a highly unpredictable manner and in terms altered from the original agreement. From 1626 to 1627, Charles raised 260,000 pounds by this method. Similarly, from 1634 onwards, he gathered an average of 107,000 pounds per year. Charles also seized private goods to cover expenditures. In 1640, he confiscated 130,000 pounds in bullion stored on government property from private merchants. Other measures to skirt parliament included new customs impositions, and sale of monopolies, government lands, and offices. Charles also kept parliament in the dark about the state of English finances.20
As discussed, pre-19th century fiscal fragmentation elsewhere in Europe meant that predatory tactics may be attributed to problems of divided fiscal authority. By the beginning of the 19th century, however, fiscal centralization had greatly strengthened the power of Continental monarchs, who with few exceptions no longer negotiated over taxation with various provincial bodies. With undivided fiscal authority, sovereigns secured new tax revenues, ostensibly making it easier to fulfill debt obligations. At the same time, there were no constitutional constraints to limit the way in which rulers were allowed to spend state funds.
The example of King Willem I of the Netherlands proves useful in this respect. Fiscal unification occurred in the Netherlands under French occupation in 1806. At the end of the Napoleonic era (1815), the Kingdom of the United Netherlands emerged, investing King Willem I with hereditary absolutist powers. In the constitution, parliament was granted the right to monitor national budgets just once every ten years, meaning that Willem I was able to use a large portion of the funds that parliament had originally voted for as he saw fit. Moreover, parliament was unable to audit how the government had spent its funds and received little information from the king about the state of public finances.
Unsurprisingly, Willem I was able to include many favored items in his ten-year budgets of 1819 and 1829, spending heavily on the military and infrastructure. Though fiscal centralization nearly doubled the size of the Dutch tax base, and Europe remained politically stable, Willem I found it increasingly difficult to balance the national accounts. Under his reign the public debt increased from roughly 575 million florins in 1814 to 900 million in 1830 and 1200 million in 1840, which amounted to more than 200 percent of GDP, comparable to the debt ratio during the turbulent Napoleonic era. Limited government emerged in the 1840s, partly in response to the excesses of fiscal absolutism. In particular, after the establishment of a new constitution in 1848, the Dutch crown had to submit annual budgets to parliament for approval.21
In the absence of constitutional constraints, organized bodies like parliament feared that rulers would spend the additional funds that it granted in reckless and wasteful ways. Thus, to provide new revenues, they demanded limits to fiscal power. Unwilling to bow to such requests, rulers often resorted to property rights violations. For these reasons, individuals resisted tax requests more fervently, suggesting that per-capita revenues collected by central governments remained low.
2.5 Limited Government
As for fiscal centralization, I rely on a simple definition of limited government to make this political transformation comparable across states. In particular, limited government emerged when parliament gained the constitutional right to control the state’s annual budget. To meet my criteria, parliament’s power had to be stable enough to hold for at least ten consecutive years. Since even absolutist rulers faced a variety of political and economic constraints, I have made the definition specific enough to refer only to a particular set of institutional arrangements that enabled parliament to limit the executive. Otherwise, many sorts of governments could be called “limited.”
Table 1 indicates that in general this political transformation began to take place during the 1830s and 1840s, several decades after fiscal centralization. As mentioned, the major exception is England, which became limited nearly 150 years prior to any of the Continental states, and Spain, where it did not emerge until 1876, following decades of failed constitutional initiatives. For additional details, refer to appendix 1.
Following limited government, parliament’s exclusive right to levy taxes was significantly strengthened and the ruler’s ability to violate private property rights (i.e. financial or otherwise) was dramatically reduced. Thus, individuals had much greater confidence that such rights would be respected. At the same time, parliament controlled the state’s purse strings, substantially reducing the likelihood of misspending by the ruler. Due to these twin developments, individuals were much more willing to submit to tax requests by the state. Hence, per-capita revenues rose. In turn, per-capita expenditures most likely increased as well.
Unlike fiscal centralization, which in many European countries was an involuntary result of French conquest, dating of limited government is sometimes open to debate. With this in mind, I have always chosen the earliest reasonable year to define political regimes as limited. Since central government per-capita revenues and expenditures grew dramatically from the 17th to the 20th centuries, the tests that I will perform have been designed to ensure that any misclassification errors bias against the hypotheses that both political transformations led to significant improvements in public finances, because mean per-capita revenues and expenditures associated with centralized and/or limited regimes will be lower than otherwise. In turn, the results become stronger if they still reveal that centralized and/or limited regimes were associated with significantly higher per-capital revenue and expenditure levels than fragmented and/or absolutist ones.
Similarly, over the 19th century limited government on the Continent was also less resilient to political upheaval than in England. As noted, I ensure a minimum standard for “stable” limited government by requiring that parliament’s constitutional veto power held for at least ten years in a row. Any stronger criterion is impractical, however. Indeed, much of the 19th century data would have to be discarded if limited government was required to have been a “permanent” reform. As mentioned, however, since 20th century per-capita revenues and expenditures levels for central governments in Europe were generally higher than in previous centuries, pushing back the dates for limited government later in time would simply strengthen any results that centralized and/or limited regimes were associated with improvements in public finance.
For additional issues surrounding the dating of limited government, please refer to section 2.5 of Dincecco (2006b) and appendix 1.