Unfinished Business

Abstract

This article explains legislators’ support for electoral reforms reducing electoral irregularities and protecting voters’ autonomy at the ballot box in Britain and Germany in the late 19th century. We argue that the main political cleavage over the adoption of new legislation to limit illicit electoral practices pitted politicians able to take advantage of opportunities for vote buying and intimidation against those who could not do so because of unfavorable political and economic conditions in their district. We examine the political, partisan, and economic factors accounting for candidates’ ability to engage in electoral irregularities and show that, in both countries, resource-constrained candidates were more likely to support the introduction of electoral reforms. Because the primary illicit electoral practice differs across these two cases—vote buying in Britain and economic intimidation in Germany—some of the political and economic factors accounting for legislators’ support for reform differ across these cases.

Keywords

democratization and regime change corruption and patronage quality of democracy European politics

Throughout the 19th century, European electoral practices were remarkably similar to those encountered in developing countries today. Candidates competed by offering goods, favors, or money to voters, and electoral intimidation and coercion were common. To mobilize voters, candidates relied on many intermediaries. These agents included policemen, tax collectors, and both rural and manufacturing employers. Such electoral practices persisted in many countries even after the adoption of the secret ballot.

The elimination of electoral corruption and intimidation in Europe raises several important questions. How did European countries democratize their electoral practices? What accounts for the political impetus behind the adoption of electoral reforms that limited opportunities for vote buying and intimidation? Which politicians demanded the adoption of these reforms? What economic and political factors explain reforming legislators’ support for legislation protecting the autonomy of voters at the ballot box?

Although deceptively simple, these questions have been largely overlooked in the comparative literature on democratization, which focuses on large-scale changes in electoral institutions. One set of studies explains incumbent elites’ decision to extend the franchise (Acemoglu & Robinson, 2000; Boix, 2003). A second strand of research focuses on why incumbents would risk the political uncertainty associated with democratization (Przeworski, 1991). By contrast, the study of reforms guaranteeing de facto political freedom once the right to vote was granted have rarely been the object of rigorous empirical investigation. To address these questions, one must examine the adoption of legislative reforms that protected voters’ autonomy while they cast their ballot and increased punishments for strategies of voter influence.

The adoption of reforms democratizing electoral practices took place in several stages. A first set of reforms replaced open voting with the secret ballot. The contribution by Aidt and Jensen (2017) in this volume focuses on the adoption of the principle of voting secrecy cross-nationally. The nominal adoption of the secret ballot, however, did not fully protect voters’ electoral autonomy nor did it end all electoral irregularities. Politicians continued to sway voters’ choices, taking advantage of both poorly designed legal punishments for election irregularities and of flaws in voting technology that pierced voting secrecy (Mares, 2015). A second generation of electoral reforms adopted several decades after the introduction of voting secrecy attempted to protect voters’ electoral autonomy.

Second-generation electoral reforms improving the integrity of the voting process varied by national political context. In some countries, reforms made it harder for political operatives to monitor ballots cast by altering voting technology. In other cases, reforms changed whether and how operatives would be punished for violating existing electoral laws against vote buying and intimidation. Finally, some electoral reforms gave party activists unimpeded access to voting places, allowing them to monitor abuses at the ballot box and to detect fraudulent vote counts. These differences notwithstanding, the second-generation electoral reforms shared the common overarching goal of ensuring that electoral outcomes reflected voters’ genuine preferences.

We analyze the adoption of electoral reforms improving the integrity of the electoral process in Britain and Germany. Our empirical analysis examines the economic and political factors that correlate with politicians’ support for electoral reforms. In doing so, we seek to leverage both commonalities and differences across these two cases. In both countries, we examine reforms adopted after the introduction of the secret ballot. The principle of voting secrecy was adopted in Germany in 1870 and Britain in 1872. The details of these reforms differed across the two cases, as the modal electoral irregularity was different in German and British national elections. In Germany, the most problematic aspect of electoral politics during the Imperial period was the intimidation of voters by employers and state employees enlisted as agents by candidates. British electoral politics exhibited a broader mix of irregularities, as compared with Germany. Here, electoral irregularities included vote buying, entertaining potential voters, and intimidation. In each country, our article examines the reforms that occupied the center stage of the parliamentary deliberations and that were regarded as most critical for the improvement of electoral integrity and to increasing voter autonomy. In Germany, we analyze the adoption of legislation that introduced changes in voting technology, such as ballot envelopes or isolating spaces. These reforms countered voter intimidation and increased voter autonomy, by protecting voters at the moment they cast their ballot and minimizing post-electoral punishments. In Britain, we study political support for the adoption of the Corrupt and Illegal Practices Act of 1883. This legislation proposed to increase legal punishments for bribery and treating voters.

Using this comparison, we uncover both important similarities and critical differences in the composition of the coalition of politicians supporting a reform of electoral institutions and in the determinants of political support for electoral reform. Consider the commonalities first. We argue that the main political cleavage over the adoption of the new legislation in both countries pitted politicians with the resources to take advantage of old modes of campaigning against those who could not do so due to unfavorable economic and political conditions in their district. The former politicians defended the political status quo. We develop several propositions about the relationship between various economic and political conditions in a district that affected candidates’ ability to engage in various electoral irregularities and their support for electoral reform.

An important determinant of a candidate’s support for electoral reforms was his access to resources that permitted illicit electoral influence. Politicians with access to these political resources opposed electoral reforms. However, the resources that made illicit electoral influence possible differed across these cases. In Britain, the key political resource deployed by corrupt candidates was money. In Britain, we show that politicians with more resources—as measured by their campaign expenditures—were less likely to support reforms that sought to limit their ability to bribe voters. In Germany, the salient electoral resource was the availability of a pool of actors that could be mobilized as brokers and engage in electoral intimidation. Such agents included both employers and state employees, such as policemen or tax collectors. Here, the main variable predicting a candidates’ access to resources that permit illicit electoral influence is his partisanship. Candidates affiliated with parties controlling state bureaucrats or aligned with the interests of employers opposed electoral reforms. By contrast, candidates representing parties that lacked access to these actors were supportive of changes in electoral institutions.

We also show that the degree of electoral competition resource-rich candidates faced deepened their opposition to electoral reform. In Britain, we show that candidates who won close races while outspending their opponents were less likely to support reforms than to high-spending candidates in safe seats. In Germany, we find that candidates with the means to engage in electoral intimidation were even more opposed to reforms if they were elected in a tight race.

Consider now the differences across cases in the factors affecting politicians’ support for electoral reforms. As the modal irregularity varied across these cases, so did the economic and political factors that allowed candidates to engage in electoral irregularities. Due to the greater importance of vote buying in Britain, we expect to find that political and economic conditions in a district that affect the cost of vote buying were important predictors of support for electoral reform in Britain. Such factors included both the size of the electorate and the relative wealth of a district. By contrast, in Germany, a case where economic intimidation was the key electoral irregularity, we focus on economic conditions affecting voters’ alternative employment prospects.

The remaining part of the article is organized as follows. We begin by providing historical background on electoral rules and electoral practices in Britain and Imperial Germany in the period following the adoption of the secret ballot. We discuss how existing electoral laws created opportunities for electoral irregularities that persisted even after the secret ballot was introduced. The next section develops our hypotheses about the economic and political determinants of support for electoral reform. Next, we offer an analysis of support for electoral reforms that sought to democratize electoral practices in both countries. We conclude by drawing the implications of our analysis for existing theories of democratization.

Historical Background

The adoption of the secret ballot was a dramatic turning point in democratization in Europe. Nevertheless, these reforms did not eliminate undue electoral influence. Commenting on the persistence of electoral irregularities in Britain, Seymour (1915) noted that “the main purpose of the ballot was not completely realized and corrupt influence was by no means banned from elections” (p. 433). The mix of electoral irregularities differed significantly across countries as did the legal sanctions to which violators of the law exposed themselves. We begin by characterizing the key violations of voter integrity in the two countries studied in this article, Britain and Germany.

Britain

In Britain, the franchise was extended during a long historical process. Consequently, reforms of electoral irregularities were linked to other changes in electoral law, such as the extension of suffrage and electoral redistricting. The Reform Acts adopted in 1832 and 1867, respectively, gradually extended the franchise. Prior to these reforms, only 15% of the male population had the right to vote; the share of enfranchised men increased to 20% and 30% in 1832 and 1867, respectively. In contrast to Germany, which adopted both universal suffrage and the secret ballot in national elections in 1870, Britain adopted electoral secrecy in 1872 prior to the adoption of full universal male suffrage in 1921.

Compared with their German counterparts, British politicians drew from a broader array of illicit electoral strategies. Corruption and bribery were the most common tactics used by candidates to influence voter choice. Throughout this period, British Parliament enacted a number of laws and statutes designed to curtail bribery. Parliament’s most systematic attempt to address bribery in British elections was the Corrupt Practices Prevention Act of 1854 (17& 18 Vict c.102). The Act provided a systematic definition of bribery and identified seven instances of corruption. Consider one example of bribery identified in the law:

Every person who shall, directly or indirectly, give, procure, agree to give, agree to procure, offer, promise, promise to produce or promise to endeavor to procure any money or valuable consideration to or for any voters, to or for any person on behalf of any voter to induce any voter to vote or refrain from voting shall be deemed guilty of bribery.

This comprehensive definition of bribery was adopted in the Corrupt Practices Act of 1883 with little modification.

Lawmakers viewed treating—the provision of “drinks or entertainment at times of elections”—as corruption distinct from vote buying. As Williams (1906), a prominent legal historian characterizes this distinction, “bribery is directed to obtain the adverse or fix the doubtful voters, treating is resorted to confirm the good intentions and keep up the party zeal of those believed to be already in the interest of the candidate” (p. 332). The Treating Resolution of 1677 prohibited “excessive entertainment of voters to be given at any other place than the giver’s own dwelling house” (Williams, 1906). The Treating Act (7 & 8 Will 3, c.4) and the Bribery Election Act took additional steps in defining punishments for these electoral practices.

A final strategy politicians used to influence voters was more coercive—the use of threats and intimidation. The legal term used by British legislators to encompass all these different strategies was “undue influence.” The Corrupt Practices Prevention Act of 1854 was the first legislation to define undue influence. Under this Act, people were guilty of exerting “undue influence” on voters if they “made use of or threatened to make use of any force violence or restraint or threatened to inflict any temporal or spiritual injury in order to induce such person to vote or to not vote.” Undue electoral influence could be perpetuated by many types of actors. Landlords and industrial employers pressured their tenants and employees, respectively, to vote correctly (O’Malley & Henry, 1869, Vol. 6). Members of the clergy exercised undue electoral influence on voters by threatening to excommunicate or to withhold sacraments from those who voted the wrong way.

In Figure 1 below, we illustrate the distribution of electoral irregularities in England and Scotland after the adoption of the secret ballot in 1872 and prior to the adoption of the Corrupt Practices Act. The source of these data is a report compiled by the British Parliament in 1883 that tabulated all electoral irregularities based on election petitions tried in Britain, Scotland, and Ireland, respectively (Great Britain. Parliament, 1883). Each of the 57 petitions alleged more than one type of irregularity. The bars in Figure 1 show the percentage of petitions reporting at type of irregularity. The figure illustrates a fairly even distribution across different irregularities. Together, bribery and treating—the irregularities we consider vote buying—comprise 54% of the 185 illicit practices alleged in these petitions.

Figure 1.

Percentage electoral irregularities mentioned in election petitions.

Germany

Germany introduced universal manhood suffrage for national elections to the second chamber of the Reichstag in 1870. The electoral law adopted in 1870, following the creation of the German Empire, initially established 382 single-member constituencies. According to Germany’s electoral law, politicians were elected in single-member districts. Electoral rules allowed for runoffs if no candidate received a majority in the first round. The number of districts increased to 397 following the annexation of Elsass and Lothringen, but remained unchanged until World War I. In contrast to Britain, German lawmakers did not modify the boundaries of the existing electoral districts during the period between 1870 and 1918.

In Germany, the secret ballot was introduced at the same time as universal manhood suffrage. Article 10 of the Germany’s electoral law mandated that “the right to vote will be exercised in person, through covered and unsigned ballots that have to be placed in an urn.” However, this law represented only a commitment to an abstract principle. The imperfect regulation of many details of the voting process, such as the design of ballots and urns, allowed politicians and their supporters to violate electoral secrecy routinely (Mares, 2015).

German electoral law punished electoral irregularities unevenly. On one hand, German lawmakers mandated harsh punishments for vote buying, which they viewed as a particularly pernicious form of electoral intervention. People using money or gifts to buy votes could be punished under the penal code (Article 109) and faced up to 2 years of imprisonment (Freudenthal, 1895; Hatschek, 1920). By contrast, other irregularities were punished less stringently. The most striking omission of the German electoral law was the absence of any punishment for the electoral participation of employers. This resulted in unprecedented electoral intimidation by private actors in German elections, which included threats of layoffs and other post-electoral punishments. Private actors’ threats were highly credible due to the observability of the vote.

Vote buying was almost non-existent in German elections due to its stringent punishment under German electoral law (Anderson, 2000; Arsenschek, 2003; Kühne, 1994). Many studies of electoral politics in 19th century Germany conducted by historians, legal scholars, or political scientists testify to the absence of vote buying in German elections (Anderson, 2000; Freudenthal, 1895; Hatschek, 1920). Vote buying was so rare that there were very few attempts to limit it in the Reichstag. Consider the characterization of Lavinia Anderson (2000), a prominent historian of the period:

Germany presented a very different picture [from England]. Here too, of course, there were scattered examples of “treating.” Cotters in the east Elbian Herrschaft Pnuwno were given a half bushel or dried peas for voting Conservative. Voters in the Rhenisch village of Rodenkirchen who accepted “loyal” ballots were handed a sausage from a large hamper at the door of the polls. A dram of schnapps at the beginning of our period and much larger quantities of “Freibier” at the end were common, although not always considered respectable features of German elections. Nevertheless, I know of only three contests in the entire Imperial Period in which corruption in the form of treating or bribery was even rumored to have affected the outcome of the election—and in two of them the rumors seem untrustworthy, given that the losing party did not bother to file a complaint. Whatever the incidence of these little reciprocities, it is clear that German voters were not considered venal even by parties hoping to overturn an election, nor did enough money change hands to give Reichstag deputies, unlike their British counterparts, any stake in regulating it. Bribery simply did not play a role in what Germans thought was wrong in their political process. (pp. 26-27)

In both Britain and Germany, the extension of suffrage and the adoption of the secret ballot marked only the beginning of a long process of democratization. The adoption of secret voting did not protect voters from actors seeking to unduly influence on their choices. A second set of democratizing reforms adopted in the aftermath of secret ballot reform attempted to eliminate electoral irregularities such as vote buying and economic intimidation. We now turn to an analysis of the determinants of the support for these electoral reforms.

Demand for Electoral Reform

What accounts for the adoption of legislation protecting voters’ autonomous choices against coercion and monetary inducements? Which politicians supported the adoption of legislation democratizing electoral practices? Which politicians resisted the adoption of these reforms? In the section that follows, we present our hypotheses regarding the correlates of support for reform. Our hypotheses highlight both similarities and differences across the two cases. Such differences arise because the modal electoral irregularity differed across countries.

Resource Asymmetries, Competition, and Support for Reform

In evaluating whether to support a change in electoral rules, politicians considered the likely consequences of a change in electoral legislation on their ability to compete in future races and on the electoral success of their respective party. The opportunities available to engage in electoral irregularities and the cost of engaging in such practices varied across individual candidates, across districts and across parties.

We conjecture that support for electoral reform originated with politicians who found it harder to engage in electoral irregularities under existing rules due to unfavorable conditions in their districts. Besides the specific conditions in a district, politicians considered how changes in electoral institutions affected the electoral fate of their party. We hypothesize that members of parties disadvantaged by existing electoral rules were likely to support the adoption of reforms. By contrast, members of parties benefitting from existing electoral rules were likely to oppose electoral reform. This hypothesis points to important commonalities across these cases in the determinants of support for electoral reforms.

Consider now the differences across the two countries. As discussed above, the modal electoral irregularity differed between Germany and Britain. Candidates engaged in the production of electoral irregularities deployed different political resources. Vote buying required access to monetary resources, while electoral intimidation required access to agents, such as state employees or employers, who would campaign on behalf of candidates and mobilize voters to the polls. These differences in the modal type of electoral corruption across these cases, led to variation in how candidates and parties thought about the relationship between political and social characteristics at the district level and the desirability of electoral reform.

Consider first Britain, a case where the dominant irregularity was vote buying. The most important resources for candidates to engage in the treating or buying of voters were monetary resources. The availability of these resources influenced candidates’ preferences about the desirability of electoral reforms. All things equal, we expect that candidates able to outspend their opponents were more likely to oppose reforms. To test this hypothesis, we use a variable that measures the resources spent by candidates in a race. It is important to note that campaign expenditures were financed by individual candidates and not mediated by parties.

Consider parties’ calculations regarding the desirability of electoral reforms in a country where the primary electoral irregularity is vote buying. Here, parties form their preferences on the attractiveness of a change in electoral rules, by considering the likely impact of this change in institutions on their vote share. Parties finding that their ability to win office has been systematically hampered by existing electoral rules are likely to support electoral reforms. By contrast, parties experiencing an advantage under the current electoral system are likely to support the status quo. In Britain, the Liberal Party considered that their candidates were systematically disadvantaged by electoral rules that did not limit campaign expenditures. Liberals considered that such electoral rules strengthened the electoral advantage of the Tories, who could select wealthier candidates with the means to engage in these corrupt practices. As Henry James, the deputy general who guided the new legislation through the parliamentary deliberations remarked, the proposals of the new legislation “could return to the House those who would represent the majority of the people and not so much the wealth of the country” (Great Britain. Parliament, 1882).

Finally, let us consider the relationship between district-level political conditions and support for reform. A central conjecture in the literature examining electoral irregularities is that politicians competing in tighter races are more likely to deploy all available strategies to ensure their victory (Lehoucq, 2003; Lehoucq & Molina, 2002; Scott, 1972). Examining the legislative activity of British members of Parliament (MPs) between 1832 and 1918, Eggers and Spirling (2014) find that MPs in closely contested seats both speak and vote more in Parliament. Yet, whether a legislator who narrowly acquired his seat under the old system favors reform limiting campaign expenditures depends on whether he was better at using illicit campaign strategies than his opponents.¹ Therefore, we expect electoral competition to affect legislators’ support for reform if they won their seats in a close race against an opponent less able to deploy money to win the race.

Consider now the political determinants of support for electoral reforms in Germany, where the dominant electoral irregularity was intimidation. In contrast to vote buying, candidates’ ability to engage in electoral intimidation did not depend on monetary resources or wealth, but their ability to enlist political support from employers or state employees, such as policemen or the Landräte. We hypothesize that candidates who had closest political ties to employers or state employees were more likely to oppose reforms.

In Germany, we also expect to find partisan differences in support of electoral reforms that reduced opportunities for intimidation. The parties on the right of the political spectrum—Conservatives, Free Conservatives, and National Liberals—benefitted from the existing electoral law. As documented by the electoral commission of the Reichstag, the overwhelming share of electoral irregularities were committed by candidates of these parties. By contrast, all other political parties, including Zentrum, Free Liberals, and Social Democrats were likely to anticipate gains in their vote share in the aftermath of these reforms. As a result, we expect candidates from these parties to support the introduction of legislation protecting voter autonomy.

One important difference between Germany and Britain comes from the ways in which partisanship affects candidates’ support for reform. In Britain, parties did not influence the resources available to candidates who engaged in electoral irregularities. Here, partisan differences in support for electoral reforms reflect only calculations about the likely effect of a change in electoral rules on the aggregate vote share of the party. By contrast, in Germany, parties also affected the resources deployed by candidates. In this case, partisan differences in support for electoral reforms reflected both considerations about the likelihood of future electoral gain and differential access to political resources of candidates.

Finally, one additional political variable that might affect support for electoral reform is the population of a district. The size of the electorate grew as a result of population growth and, in Britain, due to the extension of the franchise. Cox (1987, 1997) and Stokes, Dunning, Nazareno, and Brusco (2013) have argued that a larger electorate makes vote buying harder for candidates to organize and easier for the authorities to detect. Economic intimidation may also be more difficult in large more populous cities, if workers have a higher number of outside employment opportunities. Therefore, we expect that politicians elected in more populous districts to be more likely to support electoral reforms.

Economic Conditions in a District and Support for Reform

Besides political and partisan variables, economic conditions in a district were also likely to affect politicians’ calculations about the desirability of electoral reforms. We draw on and extend arguments from the literature on the economic origins of democracy and derive some implications of these studies for the analysis of reforms that democratize electoral practices. First, we discuss a set of explanations emerging out of the literature on modernization, or endogenous democratization (Boix, 2003; Przeworski & Limongi, 1997). Second, inequality—particularly inequality in land ownership—has been identified as an important factor impeding the elites’ willingness to embrace democracy (Acemoglu & Robinson, 2000, 2006; Boix, 2003). Finally, we discuss a different aspect of economic structure—the concentration of employment opportunities in a few firms and sectors. As Mares (2015) argues, voters are more susceptible to economic intimidation if they have few alternative employment opportunities. This perspective suggests that the economic heterogeneity of a district affects the costs of intimidation and, consequently, the demand for electoral reform.

Economic growth and industrialization affect the relative cost of vote buying intimidation through a few different mechanisms. We consider two explanatory variables that are central to modernization theory: a districts’ residents’ wealth and the share a district’s workforce employed in more modern occupations. Economic development is likely to increase the cost and limit the likely effectiveness of vote buying. This hypothesis was advanced by Gary Cox (1997) and further explored by Stokes and co-authors (2013). These authors argue that wealthier people will demand a higher price for their votes, making programmatic appeals relatively more attractive to politicians than vote buying. If vote buying becomes less effective as voters become wealthier, we should expect politicians from wealthier districts to be more supportive of reforms limiting vote buying. Increasing wealth is not the only relevant dimension of modernization. Economic development also entails an increase in manufacturing employment. The growth in the number of people working in modern sectors also limits the effectiveness of vote buying because politicians may find it harder to monitor votes in densely populated industrialized areas containing many voters (Cox, 1997; Stokes et al., 2013).

While economic development is likely to decrease the effectiveness of vote buying, the relationship between economic development and the incidence of economic intimidation is ambiguous. Economic intimidation is premised on threats made by employers to lay off workers if they support the wrong political candidate. Such threats are powerful if voters have fewer outside options, but are negligible if voters’ outside options are good. Voters’ outside options are a better predictor of how easily they can be intimidated than the income level of workers. An employer may have greater influence over a relatively highly paid worker if this worker has few alternative sources of employment. As a vast historical literature on economic intimidation in Imperial Germany has demonstrated, electoral intimidation was a pervasive electoral phenomenon in the areas of concentrated employment, such as the Ruhr or Saar (Bajohr, 1988; Klein, 2003; Mares, 2015). The level of industrial development per se is unlikely to affect incidence of intimidation and the resulting demand for electoral reform.

Focusing on voters’ outside economic options, we study an additional district-level economic variable that is likely to affect the cost of intimidation, the economic heterogeneity of a district. Economic heterogeneity, we conjecture, affects the incidence of intimidation, but not of vote buying. Economic heterogeneity is the combination of two variables, the number of firms (I) and the number of occupations (J). We define a district with high levels of economic concentration (and low occupational heterogeneity) as one in which a few firms and sectors control employment and output. To illustrate this argument, let us consider two hypothetical districts that differ in their number of firms (I) and number of occupations (J). One extreme case is a district in which both the output and employment levels of firms are controlled by a few economic actors both I → 1 and J → 1. The opposite is the case in a district where both I and J are very large. We refer to the latter as a district with high occupational heterogeneity and low levels of economic concentration. The former district, where I → 1 and J → 1, is a district with low occupational heterogeneity and high levels of economic concentration.

All things equal, we hypothesize that the cost of electoral intimidation declines as economic concentration increases. Three main factors lower the cost of electoral intimidation by employers in districts with high levels of economic concentration. First, due to their scale, large firms can more easily carrying out political activities, such as control of electoral turnout or distributing political material on behalf of a particular candidate. Second, in such a district, workers have fewer employment opportunities outside the firm, as there are few competing firms that can rehire them. Finally, the concentration of employment in the hands of a few actors also reduces the coordination problems faced by employers in punishing workers with “dangerous” political views by denying them employment opportunities. As firms in concentrated districts face relatively lower costs of political intimidation, we expect these firms will be more likely to engage in political activities during elections. In these localities, firms face a lower cost of electoral intimidation. We expect lower levels of workplace pressure in localities with higher levels of economic heterogeneity. Economic heterogeneity is unlikely to affect the cost or incidence of vote buying but is likely to affect the ease with which politicians and their supporters can engage in electoral intimidation.

Landholding inequality is a central variable in existing explanations of democratization. Existing accounts conjecture that landholding inequalities are likely to enhance opposition to democratic reforms through two related mechanisms. Such opposition may result from different preferences, on one hand, and different resources. Higher landholding inequality is expected to translate into electoral opposition to democratic reforms because landholding elites are hypothesized to fear the redistributive consequences of democracy. Tax-and-transfer models suggest that democracy will be more redistributive under conditions of high inequality and, therefore, elites are less likely to support democracy, particularly if the cost of repressing violent revolts is high (Acemoglu & Robinson, 2000; Acemoglu & Robinson, 2006). Boix (2003) extends this redistributive argument noting that inequality in land ownership is problematic for democracy because land is a fixed asset. This assumption is central to existing explanations of democratization. It implies that candidates from districts with higher levels of inequality are likely to oppose reforms that democratize electoral practices because they anticipate that the adoption of these reforms will lead to the election of candidates that are likely to support higher taxation and more public spending.

The second mechanism by which landholding inequality may affect the preference of politicians for electoral reforms is via access to resources. According to this logic, higher levels of economic inequality increase candidates’ ability to coerce voters by encouraging political support from landowners. Baland and Robinson (2008) provide empirical evidence from Chile consistent with the argument that the political influence of landlords over the electoral choices of rural voters was higher in districts where more people were employed on large farms. Ziblatt (2009) shows that the incidence of electoral corruption in Imperial Germany was higher in districts with higher levels of inequality and attributes this relationship to politicians’ ability to control voters and to influence state officials’ electoral decisions. Ardanaz and Mares (2014) illustrate that the ability of Prussian landholding employers to affected electoral decisions is affected by immediate labor market conditions (such as labor scarcity) and not by levels of landholding inequality per se. These studies predict that candidates elected in districts with high levels of landholding inequality will oppose electoral reforms enhancing voter autonomy.

To summarize the above discussion, Table 1 presents our predictions about the economic and political factors that affect candidates’ support for electoral reforms. We expect legislators who enjoy a resource advantage in the production of electoral irregularities to oppose reforms. However, the specific factors affecting candidates’ differential access to resources varies across our cases. In Britain, the main resource advantage is monetary resources that can be used during campaigns. By contrast, the main source of the resource asymmetry in the German came from differential access of candidates to actors who could be deployed as brokers. Finally, we expect legislators from districts with large electorates to support reforms in both cases.

Table 1.

Correlates of Support for Electoral Reform.

	Vote buying	Intimidation
Political
Resource advantage	−	−
Margin of victory	~	~
Resource Advantage × Margin of Victory	+	+
Size of the electorate	+	+
Economic
District wealth	+	~
Percentage of the population in manufacturing	+	~
Economic concentration	~	−
Landholding inequality	−	−

The lower part of the table summarizes the expectations about the economic conditions in a district and support for electoral reforms. We expect to find a positive relationship between economic development and support for electoral reforms in Britain. Economic development reduces the attractiveness of vote buying as the monitoring of voters becomes more difficult in industrialized and densely populated settings. At the same time, the costs of vote buying increase in wealthier areas. By contrast, we expect to find no relationship between economic development and support for electoral reforms limiting intimidation. The incidence of economic intimidation is affected by the availability of other sources of employment. Therefore, we expect economic concentration in a district to be negatively correlated with its legislator’s support for reforms. Finally, we expect a negative relationship between landholding inequality and support for electoral reforms.

Data and Empirical Analysis

In the empirical analysis that follows, we examine the political and economic determinants of the legislators’ support for electoral reform. In both Britain and Germany, we use legislators’ positions on key aspects of electoral reform. The outcome variable in Germany is whether a member of the Reichstag was a cosigner of a proposal recommending changes to electoral law.² In Britain, we measure support for reform using votes taken by MPs on specific aspects of the legislation that became the Corrupt and Illegal Practices Act (1883).

Our explanatory variables (X_idrp) vary over MPs (i), electoral districts (d), and geographic regions (r). We record each legislator’s position on multiple proposals or motions (p).³ We estimate the following model and include random effects for each vote or proposal (or motion; $α_{p}$ ) and dummies for each region. In our analysis of Germany, regions are 27 large regions, and in the British, the relevant geographic regions are England, Scotland, and Wales.⁴ We show below, our main estimates are substantively similar using motion (or proposal)-fixed effects.⁵

Ρ (y_{i} = 1) = \log i t^{1} [β_{0} + β Χ_{i d r} + α_{p}] .

α_{p} = N (0, σ_{p}) .

In the analysis that follows, all continuous variables were rescaled to have a mean of 0 and a standard deviation of 0.5.

Votes Against Electoral Corruption in England

The adoption of the Secret Ballot Act (1872) did not eliminate electoral irregularities. By precluding opportunities for closely monitored intimidation, the Secret Ballot Act created the incentives for substitution into less coercive forms of electoral influence, such as “bribing” and “treating.” As Seymour (1915) describes these practices,

A host of canvassers, bookmen, messengers and watchers were hired by each party far in excess of the number actually required and receiving munificent wages for little or no work. At Chester, where 88 clerks were admitted to be all that the situation demanded, 488 people were hired and paid. At Boston, 1,200 people were employed. As in many other constituencies, the amount spent upon this political employment was large. The tradesmen and laborers also profited from making flag poles, flags and rosettes for which the contending parties cheerfully paid exorbitant prices.

Politicians’ use of bribery and treating increased in the 1880 election. As a commission established to investigate these electoral practices concluded, “The corruption at the late election was far more widespread and far more open than had been the case at any previous parliamentary elections” (Great Britain. Parliament, 1881, p. 36). Another report investigating irregularities in Sandwich noted bribery remained unchecked after introducing the secret ballot and that sometimes, voters took bribes from both sides (Seymour, 1915). Other election reports noted the increase of other types of corrupt practices, such as employing voters.

The evidence of electoral irregularities uncovered by various Royal Commissions created an outcry for electoral reform. Gladstone’s Liberal government responded to this outcry in a swift and decisive manner. When Parliament reassembled in January 1881, Sir Henry James, the attorney general, submitted a first draft of the legislation that became the Corrupt and Illegal Practices Act. The most significant principle of the new legislation was the effort to increase the punishment for many electoral irregularities. The bill distinguished between “corrupt practices” that were punished severely and “illegal practices” for which lower punishments were enacted. In the new legislation, “corrupt practices” included bribery, treating, undue influence, and personation. The legislation also defined specific schedules that attempted to regulate the electoral expenses of politicians in different constituencies.

We consider key variables predicting MPs support for significant aspects of the legislation taken between 1882 and 1883. Consider first the political variables, which we construct using Craig (1989) as our primary source. We have argued that political variables included the size of the electorate, whether a legislator faced an opponent with fewer resources to buy votes, and the interaction between a legislator’s resource advantage over his opponent and electoral competition. We measure the size of the electorate as the logged number of electors on the voting register (Craig, 1989). We measure electoral competitiveness using the Margin of Victory, the difference between the MPs vote share and the vote share of the first runner-up. Where candidates run unopposed, Margin takes on a value of 1.⁶ Twenty-three percent of candidates in our data set run unopposed; therefore, we include a dummy variable that takes on a value of 1 if the candidate faced no challenger in his constituency (Unopposed). Because the primary form of electoral corruption in Britain was vote buying, we measure an MP’s resource advantage over his opponents (Share of All Spending) using his share of the officially reported electioneering expenses of all candidates contesting his seat in the 1880 election (Great Britain. Parliament, 2006).⁷

Besides variables testing our political hypotheses, we include control for a potential confounding variable, whether corrupt practices were identified and punished in an earlier period. We measure Previous Irregularities using an indicator variable that captures whether any electoral area in that constituency was disenfranchised because of the prevalence of electoral corruption using information in Craig (1989).

As regards economic determinants of support for reform, we expect economic development to reduce the attractiveness of vote buying as a political strategy and, therefore, to increase support for reform. Because vote buying may be not be cost-effective in richer districts, we include Income per Capita in the county the electoral district was part of in the 1881 census using data from Hechter (1984). To measure economic development, we use the share of the population of a county employed in manufacturing (Prop. Manufacturing) also from the 1881 census (Hechter, 1984). To examine whether rural inequality affects political support for reform, we compute the Gini of inequality in landholding (Land Gini) for each county using Britain’s 1896 agricultural census (Great Britain. Board of Agriculture, 1896). Although we would ideally use a land inequality data from an earlier period, Peter Lindert (1987) has convincingly shown that rural inequality experienced low levels of change throughout the 19th century. Therefore, we do not believe that measuring land inequality a decade after the adoptions of these reforms will pose a significant problem.

Table 2 presents the results of our analysis. Model 1 includes only a measure of the size of the electorate. As vote buying is less feasible in constituencies with larger electorates, we expected a positive relationship between the size of the electorate and support for electoral reform (Cox, 1997; Stokes et al., 2013). Model 2 includes other political variables in an electoral district we believe lower support for electoral reform. Politicians elected from constituencies that include areas disenfranchised due to corruption were more likely to oppose the adoption of the new legislation. The legacy of corruption persists, depressing demand for electoral reform.

Table 2.

Voting for Electoral Reform in Britain.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)
Log No. of Electors	0.26*** (0.079)	0.31*** (0.083)	0.26*** (0.085)	0.27*** (0.086)	0.26*** (0.088)	0.26*** (0.089)	0.29*** (0.096)
Margin		0.16 (0.20)	0.60*** (0.22)	0.61*** (0.22)	0.70*** (0.23)	−0.48* (0.29)	−0.11 (0.87)
Unopposed		−0.73*** (0.26)	−1.19*** (0.29)	−1.20*** (0.29)	−0.86*** (0.29)	−1.45*** (0.31)
Previous Irregularities		−1.47*** (0.25)	−1.52*** (0.25)	−1.49*** (0.25)	−1.47*** (0.26)	−1.27*** (0.26)	−1.81*** (0.37)
Land Gini			−0.36*** (0.097)	−0.30*** (0.10)	−0.28*** (0.11)	−0.31*** (0.11)	−0.31*** (0.12)
Income per Capita				−0.012 (0.082)	0.041 (0.083)	0.077 (0.084)	0.095 (0.091)
Prop. Manufacturing				−0.15* (0.089)	−0.17* (0.091)	−0.15* (0.092)	−0.086 (0.11)
Share of All Spending					−0.63*** (0.095)	−0.83*** (0.10)	−0.90*** (0.20)
Share of Spending × Margin						2.54*** (0.39)	2.08 (1.60)
n	3,384	3,269	3,257	3,257	3,216	3,216	2,740
No. of proposals	20	20	20	20	20	20	20
σ_u	0.97	1.01	1.02	1.02	1.05	1.07	1.13
BIC	4,033.5	3,834.4	3,799.7	3,813.0	3,721.7	3,687.0	3,093.3

SEs are in parentheses. Continuous variables are standardized to have a mean of 0 and a standard deviation of 0.5. BIC = Bayesian information criterion.

p < .10. **p < .05. ***p < .01.

In Model 3, we add a measure of landholding inequality (Land Gini). Consistent with our expectations, we find that politicians from districts with higher levels of landholding inequality are less likely to support reforms limiting vote buying. In Model 4, we include controls for economic development. Contradictory to our expectations, we find no relationship between the district wealth and a legislator’s support for reform. The cost of monitoring of voters may be higher in manufacturing, as compared with rural districts, due to a higher presence of recent migrants in these districts that may be harder to integrate in existing political machines. Contrary to what we expected, the aspect of modernization relating to the number of people in non-traditional occupations (Prop. Manufacturing) is negatively correlated with MPs support for measures to limit electoral corruption.⁸

In Model 4, we explore whether politicians better placed to engage in vote buying oppose reform. We control for an MP’s resource advantage over his challengers using his share of all officially reported election expenses in his constituency (Share of All Spending). Candidates who outspent their rivals did not favor punishing vote buying and treating. High-spending candidates in close races were especially unlikely to support reform (Model 6). This could be the case because unopposed candidates report few election expenses. When we limit our analysis to seats contested, we find that politicians with an advantage at using electoral corruption are less likely to support reform if they were in a close race in 1880 (Model 7).

Germany: Support for Reforms Protecting Voters From Economic Intimidation

As discussed in above section, the most significant shortcoming in the design of Germany’s electoral law were imperfections in the technology of voting, such as the design of the urn or regulation of the size and shape of the ballot. This allowed candidates and electoral agents campaigning on their behalf to monitor the choices people made while voting. The investigation of electoral practices by the electoral commission of the Reichstag revealed the existence of electoral irregularities perpetuated both by private actors and by public election officials. Here, we find ample reports of voters being forced to enter the voting area while holding ballots over their heads and of parallel voting lists created by employers’ representatives in the voting area that recorded the types of ballots held by individual voters. The reports also described evidence that people who voted the wrong way were punished after the election, often using “electoral layoffs.” These reports demonstrated that the protections for electoral secrecy provided by the electoral law were only nominal.

Proposals securing voters’ autonomy were on the agenda of the Reichstag in eight out of the 13 legislative periods between 1870 and 1914. As Heinrich Rickert, one of the politicians who relentlessly pushed for these reforms argued in one of the parliamentary debates over electoral reform,

The freedom of the vote is a guarantee of one of the most important constitutional provisions in Germany. It is the foundation of our constitutional life. If we do not provide sufficient guarantees for it, then our public life will be based on hypocrisy, on pressure and on force and this has never led to good outcomes (Stenographische Berichte des Deutschen Reichstages, January 15, 1896). (Deutscher Reichstag, 1871-1918)

The central element of all proposals to increase the political autonomy of voters was the reform of the technology of voting. We use these proposals to measure support for reform, the dependent variable of our analysis.⁹

Similar to the British case, we construct a novel data set measuring economic and political variables that may predict legislators’ preferences for reform. Consider the political variables first. We measure the size of the electorate as the logged population of an electoral district. Our measurement of the competitiveness of a district is the winner’s margin of victory (Interuniversity Consortium for Political and Social Research [ICPSR], 1984). Given Germany’s runoff elections, we compute Margin of Victory as the difference in vote share between the winner and the runner-up in the decisive electoral round. We also include a variable that takes the value 1 if a politician ran unopposed. Finally, to measure the legacies of local practices of corruption, we construct a variable that sums the cumulative cases of electoral fraud in the district, as reported by the commission verifying electoral irregularities (Wahlprüfungskomission).

Our economic variables capture many dimensions of a district’s economic structure. We measure the economic development of a district, by using the share of the population not employed in agriculture (nonagricultural workforce) based on data reported in Reibel (2007). Although measured using slightly different criteria, the Nonagricultural Workforce variable captures the proportion of the population in more modern sectors as Prop. Manufacturing does in Britain. We also control for inequality in landholding in 1895 using data from Ziblatt (2009).

We have hypothesized that politicians in districts with lower levels of economic concentration have a higher demand for electoral reform. To test this hypothesis, we take advantage of the information collected by the German statistical agency as part of its occupational census (Betriebszählung) in 1895 and 1905 (Kaiserliches Statistisches Amt, 1898, 1907). The occupational census presents precise measures of the employment shares in more than 180 occupations across more than 1,000 communes. We digitized this source (which totals above 400,000 lines of information for each census) to develop very precise measures of the Germany’s occupational landscape. We aggregate these variables to the 397 electoral districts, using the mapping of localities into districts reported in Reibel (2007). Germany experienced no redistricting for the elections to the national parliament, which makes the mapping of localities to districts relatively straightforward.

To examine the relationship between economic concentration and demand for electoral reform, we construct a Herfindahl–Hirschman index (HHI) measure of employment concentration. As conventionally used in the industrial organization literature, the HHI is defined as the sum of the squares of employment shares of all occupations in a particular district. We use

Η Η Ι = \sum_{i = 1}^{n} S_{i}^{2},

where $S_{i}$ stands for the employment share of each occupation found in the district, while i denotes 182 occupations. Higher values of the HHI indicate more concentrated economic districts.¹⁰

To ensure comparability to the results reported for Britain, we estimate similar models to those reported in Table 2. The outcome variable is whether a member of the Reichstag was a cosigner of a proposal recommending changes to the electoral law and estimate whether our variables of interest increase the probability a politician cosigns an electoral reform bill or not. The results are presented in Table 3.

Table 3.

Support for Electoral Reform in Germany.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)
Log Population	−0.27*** (0.095)	−0.50*** (0.099)	−0.45*** (0.10)	−0.55*** (0.10)	−0.55*** (0.10)	−0.55*** (0.10)	−0.56*** (0.10)
Margin		−0.10 (0.089)	−0.14 (0.091)	−0.11 (0.091)	−0.096 (0.092)	−0.16 (0.098)	−0.16 (0.098)
Unopposed		−0.54 (1.07)	−0.53 (1.07)	−0.45 (1.07)	−0.57 (1.06)	−0.50 (1.07)
Previous Irregularities		−0.085 (0.11)	−0.074 (0.11)	−0.041 (0.11)	0.057 (0.12)	0.073 (0.12)	0.075 (0.12)
Employer Aligned Parties		−1.07*** (0.098)	−1.04*** (0.099)	−0.97*** (0.10)	−0.32*** (0.12)	−0.37*** (0.12)	−0.37*** (0.12)
Land Gini			−0.29** (0.13)	−0.31** (0.13)	−0.19 (0.13)	−0.21* (0.13)	−0.21 (0.13)
Nonagricultural Employment				0.60*** (0.11)	0.56*** (0.11)	0.57*** (0.11)	0.57*** (0.11)
HHI				−0.22** (0.11)	−0.27** (0.12)	−0.27** (0.12)	−0.27** (0.12)
Employer Parties vs. Non-Employer Parties					−1.27*** (0.14)	−1.19*** (0.15)	−1.19*** (0.15)
Margin × Employer vs. Non-Employer						0.54** (0.27)	0.54** (0.27)
n	7,123	7,123	7,123	7,123	7,123	7,123	7,100
No. proposals	18	18	18	18	18	18	18
σ_u	0.81	0.80	0.80	0.80	0.81	0.81	0.81
BIC	4,662.7	4,560.8	4,564.2	4,545.0	4,465.6	4,470.8	4,456.7

SEs are in parentheses. Continuous variables are standardized to have a mean of 0 and a standard deviation of 0.5. HHI = Herfindahl–Hirschman index; BIC = Bayesian information criterion.

p < .10. **p < .05. ***p < .01.

Model 1 examines the relationship between the size of the electorate and support for electoral reform using population size as a proxy for the size of the electorate. We expected that politicians competing in more populous districts would support electoral reforms. This hypothesis is not confirmed: population size stands in a negative relationship with support for reforms, and the relationship is significant at conventional levels. One potential explanation for the absence of this relationship is that Social Democratic deputies who often elected in larger urban areas were less likely to support reforms, because many German Social Democrats rejected legislative routes to reform.

In Model 2, we control for the legislator’s margin of victory, unopposed candidates, and a legacy of electoral irregularities in the district. Unlike in Britain, German legislators in competitive races are more likely to support electoral reform, though this finding is not robust across specifications. Unlike in Britain, neither running unopposed nor a legacy of corruption in a district predicts support for reform.

In Model 3, we add a control for the landholding inequality in a district. We find a negative and statistically significant relationship between landholding inequality and support for reform. In Model 4, we introduce variables measuring additional economic covariates in a district. First, the findings in this model follow predictions from the literature on endogenous democratization, according to which higher levels of economic development (measured by a higher share of a nonagricultural workforce) increase support for electoral reform. We also find support for our hypothesis that demand for electoral secrecy was lower in districts with high levels of economic concentration the incidence of private economic intimidation was high.

Finally, in Model 5, we evaluate the extent to which resource advantages shape politicians’ support for reforms. We include a dummy variable for whether a member of the Reichstag represents a party traditionally aligned with employers (Employer-Aligned Parties) and find that, unsurprisingly, members of these parties tend not to support electoral secrecy. In addition, we included a dummy variable indicating when a candidate from an employer-aligned party was pitted against a candidate from a party not associated with employers (Employer Parties vs. Non-Employer Parties). This dummy variable is negative and statistically significant at conventional levels. We also include an interaction between the electoral competitiveness of his race and a legislator’s resource advantage (Model 6). As in Britain, facing electoral competition makes members of resource-advantaged parties even less likely to support reform.

Conclusion

Why did politicians adopt electoral reforms reducing their ability buy votes or intimidate voters during the first wave of democratization? In this article, we have studied these questions by analyzing legislative support for some of the most significant legislation that attempted to limit corrupt electoral practices. Using recently digitized data from Britain and Imperial Germany, we have examined the economic and political determinants of legislative support for electoral reform.

In both Britain and Germany, the adoption of the secret ballot did not fully eliminate electoral irregularities. Nevertheless, both the menu of corrupt electoral strategies and the modal strategy by which candidates appealed to voters differed significantly across these countries. In Britain, the dominant non-programmatic strategy was the bribery and treating of voters. In Germany, the dominant non-programmatic strategy was electoral intimidation. In examining the determinants of demand for electoral reforms that sought to limit these electoral irregularities, our article seeks to leverage both commonalities and differences across these cases. Our overarching hypothesis is that in both countries, demand for electoral reforms originated with politicians who lacked political and economic resources to carry out electoral irregularities under existing electoral rules. Examining the political, partisan, and economic factors that account for candidates’ ability to engage in electoral irregularities, we show that in both countries, resource-constrained candidates were more likely to support the introduction of electoral reforms. In Britain, where most irregular electoral practices required money, we show that candidates who could outspend their opponents opposed electoral reforms limiting electoral bribery. In Germany, resource-constrained politicians, who supported reform, lacked the ability to enlist state employees, or private actors as their brokers.

Economic as well as political factors explain support for electoral reform. Our micro-historical analysis allows us to test the predictions from existing economic explanations of democratization in a more precise way than analyses premised on cross-national data. Consistent with our overarching hypothesis, we find that economic conditions in a district that make it harder for candidates to succeed using electoral irregularities, predict legislators support for reform. Such factors include the level of wealth and development of a district, its level of economic heterogeneity, or the level of landholding inequality. While confirming some existing hypotheses advanced by economic explanations of democratization, our article suggests, however, that the explanation for the adoption of democratizing reforms is not reducible to economic factors alone and that it should consider a combination of economic and political variables.

Footnotes

Acknowledgements

We are very grateful to Lucas Leemann and Long Tran for providing research assistance on an earlier version of the article. We also acknowledge extremely helpful comments from Jan Teorell and Dan Ziblatt and three anonymous reviewers. Data and replication materials can be found at .

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Kimuli Kasara is an associate professor of political science at Columbia University. She studies ethnic demography, political violence, and distributive politics with a focus on sub-Saharan Africa.

Isabela Mares is professor of political science in the Department of Political Science at Columbia University. She is currently working on a number of projects examining various dimensions of electoral corruption and the adoption of reforms limiting electoral irregularities. Her most recent book is From Open Secrets to Secret Voting: Democratic Electoral Reforms and Voter Autonomy, which was published by Cambridge University Press in 2015.

References

Acemoglu

Robinson

J. A.

(2000). Why did the West extend the franchise? Democracy, inequality, and growth in historical perspective. The Quarterly Journal of Economics, 115, 1167-1199.

Acemoglu

Robinson

J. A.

(2006). Economic origins of dictatorship and democracy. Cambridge, UK: Cambridge University Press.

Aidt

T. S.

Jensen

P. S.

(2017). From open to secret ballot: Vote buying and modernization. Comparative Political Studies, 50, 555-593.

Anderson

M. L.

(2000). Practicing democracy: Elections and political culture in Imperial Germany. Princeton, NJ: Princeton University Press.

Ardanaz

Mares

(2014). Labor shortages, rural inequality, and democratization. Comparative Political Studies, 47, 1639-1669. doi:10.1177/0010414013512609

Arsenschek

(2003). Der Kampf um die Wahlfreiheit im Kaiserreich: zur parlamentarischen Wahlprüfung und politischen Realität der Reichstagswahlen 1871-1914 [The struggle for electoral freedom in Imperial Germany: on the electoral verification in the German parliament and political realities during the period between 1871-1914]. Düsseldorf, Germany: Droste.

Bajohr

(1988). Zwischen Krupp und Kommune: Sozialdemokratie, Arbeiterschaft und Stadtverwaltung in Essen vor dem Ersten Weltkreig [Between Krupp and the Commne: Social democracy, working class and local administration in Essen prior to World War I]. Essen, Germany: Klartext.

Baland

J.-M.

Robinson

J. A.

(2008). Land and power: Theory and evidence from Chile. American Economic Review, 98, 1737-1765. doi:10.1257/aer.98.5.1737

Boix

(2003). Democracy and redistribution. New York, NY: Cambridge University Press.

10.

Cox

(1997). Making votes count: Strategic coordination in the world’s electoral systems. Cambridge, UK: Cambridge University Press.

11.

Craig

F. W. S.

(1989). British parliamentary election results 1832-1885 (2nd ed.). Aldershot, UK: Dartmouth.

12.

Deutscher Reichstag. (1870-1912). Stenographische Berichte des Deutschen Reichstags [Stenographical reports of the German Parliament]. Berlin, Germany: Norddeutschen Allgemeinen Zeitung.

13.

Deutscher Reichstag. (1871-1918). Stenographische Berichte über die Verhandlungen des Reichstages [Stenographical reports of the deliberations of the German Reichstag]. Berlin, Germany: J. Gittenfeld.

14.

Eggers

A. C.

Spirling

(2014). Electoral security as a determinant of legislator activity, 1832-1918: New data and methods for analyzing British political development. Legislative Studies Quarterly, 39, 593-620. doi:10.1111/lsq.12059

15.

Freudenthal

(1895). Die Wahlbestechung. Breslau, Germany: Schletter.

16.

Great Britain. Board of Agriculture. (1896). Returns as to the number and size of agricultural holdings in Great Britain in the year 1895. London, England: Her Majesty’s Stationery Office.

17.

Great Britain. Parliament. (1881). Report of the commissioners appointed under Her Majesty’s royal sign manual to inquire into the existence of corrupt practices in the borough of Macclesfield I.—Report reports of commissioners. Great Britain. Parliament. Session 1881. Retrieved from http://gateway.proquest.com/openurl?url_ver=Z39.88-2004&res_dat=xri:hcpp&rft_dat=xri:hcpp:rec:1881-057168

18.

Great Britain. Parliament. (1882). Hansard’s parliamentary debates. London, England: Hansard.

19.

Great Britain. Parliament. (1883). Election petitions. Return of all election petitions tried in England, Scotland, and Ireland, respectively, by election judges, or decided by the court on special cases, under “the Parliamentary Elections Act, 1868,” since 30 January 1874; &c [1883]. Retrieved from http://parlipapers.proquest.com/parlipapers/docview/t70.d75.1883-059630

20.

Great Britain. Parliament. (2006). Election charges. Return of charges made to candidates at the late elections by returning officers, specifying in each case the number of members returned, and, in cases of contest, the number of candidates; also the total expenses of each candidate, both exclusive and inclusive of returning officer’s charges, delivered to the returning officer pursuant to the corrupt practices acts, and the number of votes polled for each candidate (in continuation of parliamentary paper, no. 358 of session 1874) [electronic resource]. Cambridge, UK: Proquest LLC.

21.

Hatschek

(1920). Kommentar zum Wahlgesetz und zur Wahlordnung im deutschen Kaiserreich [Comments on the electoral law in the German Empire]. Berlin, Germany: de Gruyter.

22.

Hechter

(1984). Internal colonialism study: National integration in the British Isles, 1851-1966. Retrieved from http://doi.org/10.3886/ICPSR07533.v1

23.

Interuniversity Consortium for Political and Social Research. (1984). German Reichstag election data, 1871-1912. Retrieved from http://doi.org/10.3886/ICPSR00043.v1

24.

Kaiserliches Statistisches Amt. (Imperial Statistical Office). (1898). Gewerbestatistik der kleineren Verwaltungsbezirke nach der Zaehlung vom 14. [Imperial Statistical Office] Juni 1895 (Volumes 117 and 118). Berlin, Germany: Puttkammer und Muhlbrecht.

25.

Kaiserliches Statistisches Amt. (Imperial Statistical Office). (1907). Gewerbliche Betriebsstatistik nach der Zahlung vom 12. [Imperial Statistical Office] Juni 1907 (Volume 219). Berlin, Germany: Putkammer und Muhlbrecht.

26.

Klein

T. G. G.

(2003). “Gültig - ungültig”: die Wahlprüfungsverfahren des Deutschen Reichstages 1867-1918 [Valid-invalid: Procedures of electoral verification in German Empire 1867-1918]. Marburg, Germany: N.G. Elwert.

27.

Kühne

(1994). Dreiklassenwahlrecht und Wahlkultur in Preussen 1867-1914: Landtagswahlen zwischen korporativer Tradition und politischem Massenmarkt [Censitary electoral law and electoral cultural in Prussia 1867-1914: Elections to the Prussian Parliament between corporatist tradition and political mass market]. Düsseldorf, Germany: Droste.

28.

Lehoucq

F. E.

(2003). Electoral fraud: Causes, types and consequences. Annual Review of Political Science, 6, 233-256.

29.

Lehoucq

F. E.

Molina

(2002). Stuffing the ballot box: Fraud, electoral reform, and democratization in Costa Rica. Cambridge, UK: Cambridge University Press.

30.

Lindert

P. H.

(1987). Who owned Victorian England? The debate over landed wealth and inequality. Agricultural History, 61(4), 25-51.

31.

Mares

(2015). From open secrets to secret voting: Democratic electoral reforms and voter autonomy. New York, NY: Cambridge University Press.

32.

O’Malley

E. L.

Henry

(1869). Reports of the decisions of the judges for the trial of election petitions, pursuant to the parliamentary elections act, 1868. London, England: Stevens & Haynes.

33.

Przeworski

(1991). Democracy and the market: Political and economic reforms in Eastern Europe and Latin America. New York, NY: Cambridge University Press.

34.

Przeworski

Limongi

(1997). Modernization: Theories and facts. World Politics, 49, 155-183.

35.

Reibel

C.-W.

(2007). Handbuch der Reichstagswahlen 1890-1918: Bündnisse, Ergebnisse, Kandidaten [Handbook for the elections to the Reichstag 1890-1918: Alliances, results, candidates]. Düsseldorf, Germany: Droste.

36.

Scott

J. C.

(1972). Patron-client politics and political change in Southeast Asia. American Political Science Review, 66, 91-113.

37.

Seymour

(1915). Electoral reform in England and Wales; the development and operation of the parliamentary franchise, 1832-1885. New Haven, CT: Yale University Press.

38.

Stokes

Dunning

Nazareno

Brusco

(2013). Brokers, voters, and clientelism: The puzzle of distributive politics. New York, NY: Cambridge University Press.

39.

Williams

C. W.

(1906). Parliamentary elections & petitions (Vol. 2). London, England: Stevens & Sons.

40.

Ziblatt

(2009). Shaping democratic practice and the causes of electoral fraud: The case of nineteenth century Germany. American Political Science Review, 103, 1-21.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.17 MB

0.00 MB