The electoral costs of registration and elections may be difficult to calculate and compare as they often vary depending on a number of factors, such as the level of responsibility and tasks of the EMB, number of elections in an election year, the timing and sequence of elections, personnel and operational costs, and more.  

The Cost of Registration and Elections Encyclopaedia topic evaluates the methods by which election budgets are established, tracked and funded; identifies the cost-management practices that can be adopted by EMBs; establishes a methodology for the comparative assessment of electoral costs with a goal of helping election organizers and policy makers to identify ways of managing financial resources, measure their relative costs of operations with other EMBs, and reduce the cost of electoral events. 

According to the current standard theory and practice of elections, the main expenses are incurred for the following activities: voter registration, boundary delimitation, the voting operation, counting and transmission of results, dispute adjudication, voter education and information, campaigning by political parties and candidates, and vigilance or oversight by party representatives and domestic or international observers (Goodwin-Gill 1994; López-Pintor 2000; OSCE 2001; EU 2002; IDEA 2002). Consequently, electoral costs include all the costs incurred in undertaking such activities, regardless of the kind of agency involved, whether national or local, public or private.

Voter registration, boundary delimitation, the voting operation, counting and transmission of results and dispute adjudication are almost invariably conducted by various forms of EMBs (i.e., the executive branch of government, officials under the supervision of an electoral commission or an independent electoral commission). Providing voter education and information, may be shared by electoral authorities, political parties and civil society organizations, while the seventh (campaign activities) is exclusively conducted by political parties and candidates. The main variable for cost analysis may therefore be directly related to the specific organization or entity in charge of a given electoral activity—one or several organizations within the national government, local governments, judiciaries, private firms (e.g., quasi-public postal service and telecommunications), NGOs and political parties. 

The electoral budget during an election year may cover one or several elections, and in the latter case one should consider whether elections are held simultaneously or separately on different dates. These factors heavily influence the overall budget figures as well as their distribution by budget line item. 

Number and time sequence of elections are variables that make it difficult to compare electoral costs among different countries. It is not always easy to split budgets and assign costs to different elections. 

A distinction between personnel costs and operational costs is generally used as the main structuring criteria in an electoral budget. 

Over time, there are fixed costs for the ordinary functioning of an electoral administration; these costs are incurred independently of the occurrence of elections in a given year. Variable costs, meanwhile, are those related to the actual conduct of elections. Almost the entire budget for a specific election consists of variable costs. 

With the term funding source, three main distinctions can be made. One is between a) a separate electoral budget that is part of the consolidated budget of the nation and b) election-tied funds that form part of the budget of public agencies whose main responsibilities have little or nothing to do with elections (i.e. civil registries, police and postal services). Another distinction, between national and international funding of electoral budgets, is of utmost relevance in the case of post-conflict elections and even second-generation elections in emerging democracies. A third distinction, between public and private funding, is applicable primarily in regards to the funding of political parties and electoral campaigns. 

Electoral-budget costs that can be readily identified on a budget document are called direct costs. There are also diffuse costs—those that may prove difficult or impossible to accurately assess even if properly identified. This category can be further divided by degrees of diffusiveness; for example, some costs for activities can be clearly identified, yet still cannot be disentangled from within the general budget of the agency involved (i.e. the contribution of civil registries in providing information to EMBs for the production of voter lists; or the production of voter lists by the national agency in charge of censuses and statistics). Obtaining specific information about such costs is frequently impossible because activity-focused cost audit are not often practiced by organizations responsible for a multiplicity of programs, such as those mentioned above. As noted above in the funding source category, other diffuse costs may include actual costs hiding beneath the ordinary operations of agencies that lend various forms of support to the electoral process (i.e., police force, postal services, school

systems, local governments and public TV). While these are real costs, they are neither included in the electoral budget nor are they easy to assess in many cases. 

An additional and very important distinction is between integrity costs and core costs. This distinction may be essential for an adequate understanding of the funding of elections, most notably in emerging and post-conflict democracies. It has much to do with conditions ensuring a safe, politically neutral environment and a level playing field. 

The integrity of the voting operation is mainly a function of voter security and ballot security. Voter security includes ensuring the safety of individual voters and of voting and counting facilities; removing threats and intimidation factors; and guaranteeing accessibility to polling stations. Security of the ballot implies arranging the voting and counting in such a way that the voter lists, ballot papers, tallies and other result records are tamper-proof. The main objectives are to preserve secrecy of the vote (disposition of the voting booth), to avoid double voting (a safe ballot box, use of indelible ink), and to eliminate undue manipulation of voting materials (printing control, storage and transport). The presence of party representatives and electoral observers—both domestic and international—may be required at the voting and counting locations. 

In summary, integrity costs generally concern security arrangements for registration and polling places. They may include funding for international personnel serving as part of the electoral administration; tamper-resistant electoral materials necessitated by a low level of trust among contenders; long-term electoral observer missions; intensive voter education campaigns and election publicity; and assistance to political parties at national and local levels as part of a broader approach to capacity building. 

Those costs routinely associated with carrying out elections are designated as core costs. They are incurred independently of the degree of uncertainty and security of the political environment and have to do with voter lists, voting materials, competence among polling officials, voter information, and organizational and logistical arrangements. Core costs are assumed to be fixed rather than variable; integrity costs are incurred when special and often unexpected expenses are required to ensure that the process works efficiently. 

The relative proportion of overall election costs that can be attributed to integrity or core costs generally depends on how far removed the politics of a given country are from conflict and lack of security. In a recent assessment of post-conflict elections, the following dummy exercise offered a useful indication of how costs may be structured.

The classification of electoral costs are defined in Types of Electoral Costs serve as the basis for developing cost-assessment methodologies. The following steps are required: 

• Define the political environment—to be taken as a constant factor—a necessary consideration regardless of whether a democracy is stable, transitional or post-conflict.
• Consider whether the electoral budget of an election year is covering one or several elections; and in the latter case, whether elections are to be held simultaneously or separately.
• Separate fixed costs of the ordinary functioning of the EMB from variable costs as specific for a given election—and then distinguish between personnel and operational costs if relevant for the budget of a particular  event.
• Identify the list of election-related activities, which may include voter registration, boundary delimitation, the voting operation, counting and transmission of results, dispute adjudication, voter education and information, and campaigning by political parties and candidates.
• Separate core costs and integrity costs in relation to each of the activities listed in D above.
• Consider separately the direct and diffuse costs incurred by different organizations involved in electoral activities as core and integrity costs. Quantify direct costs precisely, and estimate diffuse costs based on contextual figures or best-educated guesses.
• Identify the funding source for each cost category, whether national or external, public or private.
• Finally, make the necessary adjustments in comparing cost evolution over time (increase/decrease in the overall cost or in particular budget line items) for within a given country or among different countries, by including capital and equipment investments as well as amortizations. In both cases, indexing the currency in constant values for a given year may be necessary. 

Table 1 (below) shows the frequency and intensity with which different costs are incurred, depending on the democracy environment.  

Three trends are apparent when comparing electoral scenarios in stable, transitional and post-conflict democracies. 

First, integrity costs are relatively higher in transitional than in stable democracies, and still higher in post-conflict situations. This is primarily, although not exclusively, due to the requirement of police and military to handle security at every stage of the electoral process, frequently with the presence of an international force. 

Second, diffuse core costs are particularly extensive in stable democracies where a large part of the actual electoral expense is covered by budgets from different state administrations; in such environments, it is not always easy or feasible to quantify which part of their budgets are devoted to electoral activities. On the other hand, diffuse integrity costs are greater in transitional and post-conflict environments because of the need for funding from a peacekeeping budget, which are difficult to quantify. Diffuse costs attributable to political party finance are substantial and difficult to quantify in both stable and other democracies.  

Third, the existence and amount of vigilance costs could be reduced to some core costs for party agents in stable democracies, but would expand to domestic and international observers in the other democracy environments, especially in post-conflict situations.

The Cost of Registration and Elections (CORE) Project evaluates budgets of Electoral Management Bodies (EMB’s) in order to identify budgeting practices and techniques that influence cost control and transparency and identifies the revenue sources for funding the administration of election processes. 

For more information covered in this section, see:

• Election funding: legal framework and practice; election budget processes and cycles
• Resource-sharing arrangements with other public agencies
• How much do elections cost?
• Electoral Budget histories in the 1990s
• Public funding of political parties
• Electoral Assistance from the international community

As a rule, electoral budgets tend to be part of the consolidated budget of the nation on an annual cycle. In a non-election year, the budget for the electoral authority— whatever shape it takes—is usually a line item of the national budget or is included within the budget of the larger agency of which the electoral administration is part (e.g., the Ministry of Interior). In an election year, the corresponding budget is funded from the national budget by following ordinary or extraordinary procedures, depending on whether the elections could be anticipated or were called unexpectedly.  Flexibility is particularly necessary in parliamentary systems in which governments may collapse unexpectedly or a prime minister may call an election at any time within a given term of office¹. 

Almost invariably, electoral budgets are prepared by the electoral authority and processed through the finance ministries for approval in the legislature. Most finance ministries do not have the authority, at least formally, to curtail or amend an electoral budget prepared by electoral authorities. Some exchange and bargaining between a ministry and electoral officials may nonetheless ensue. In Australia, for example, the Australian Electoral Commission (AEC) receives the major part of its funding through government appropriations, which are acts of Parliament that authorize expenditures and appropriate money from the Consolidated Revenue Fund to provide agreed levels of budget funds to government agencies. Budget levels are agreed for the budget year and three succeeding years, according to the stages of the election cycle. Adjustments to funding to cover changes in operational and legislative requirements occur through a formal process of New Policy Proposals involving ministers and the Expenditure Review Committee made up of senior cabinet ministers. Appropriations are made on an outcomes basis, and annual reporting is made in accordance with the outcomes and outputs framework, including performance indicators and targets against which performance can be assessed. Appropriations and all reporting occur on an accrual base, so non-cash expenses (such as depreciation) form part of annual funding levels. 

Improved information technology (IT) platforms have helped limit increases in the costs of registration and elections in Australia over the past five years—operational costs in a non-election year have increased by approximately 0.8 percent, and in an election year by 1.3 percent. The Government and the Parliament place constant pressure on the AEC (along with other publicly funded agencies) to continually review its operational performance and to contain costs. The AEC will upgrade its IT capabilities in the next three years to take advantage of the scanning and optical character recognition technologies that can further enhance its roll management systems and its management of elections. 

Such a budgetary practice is better established in countries where the state administration has achieved a certain degree of organizational maturity and the legal system is stable. More frequently than not, standard budgetary practices are formally followed at all levels of government (after influencing international financial and monetary agencies as a requirement for aid and trade agreements). Otherwise, the electoral budget for a fresh election after civil conflict, or elections in the midst of unstable times, is usually made on an ad hoc basis and defined with participation of international agencies (e.g., Haiti, Afghanistan). A kind of transitional situation might be that of Cambodia, where elections are funded through a special account at the National Treasury called the Trust Fund Account for Elections. Practically all contributions for elections—private and public, national and international—are deposited in this fund. Guatemala offers an interesting case where a fixed percentage of the national budget (0.5 percent) is allocated by law for the ordinary functioning of the electoral commission. In an election year, the commission follows standard budgetary procedures in defining and requesting the necessary funds for the election. 

Afghanistan presents a recent example of how elections are organized in a post-conflict scenario. The October 2004 presidential poll bore many similarities to other electoral processes where the international community has been heavily involved in both the political process and the practicalities of organizing elections. Those processes are inclined to choose high-tech solutions, rather than low-tech alternatives, and a large central election administrative bureaucracy with a distinct international component— which together contribute to a staggering price tag for the electoral process. A unique feature of the Afghanistan election was how early and suddenly the responsibility for planning and implementing the election was handed to the government itself, at least on paper. This marked the first time that a national election authority with no previous experience in organizing elections was charged with implementing its first election—even though the United Nations had a staff of more than 100 people on the ground tasked to work on the electoral process and had the explicit mandate to conduct voter registration prior to polling. The early transfer of responsibility for organizing the elections from the United Nations Assistance Mission in Afghanistan (UNAMA) to the national government became known as the ‘Afghanization’ of the electoral process.

Notes: 

¹A notable exception is Switzerland, where the electoral budget stays the same in election and non-election years. This is due to the fact that there are almost always three to four referenda at the national level in a non-election year and two to three referenda in an election year.

Next: Resource-sharing arrangements with other public agencies

A particularly common situation in stable democracies with highly developed state apparatuses at the national and sub-national levels is that a number of electoral costs are shared among the electoral administration and other public agencies. This is the case with the making and updating of voter lists and district boundary delimitation, voter information, the conduct of external and proxy voting, early transmission of results, dispute adjudication, and security of the polling operation. Sometimes such costs are easily identifiable within an electoral budget of the national EMB or of another public agency. Frequently, however, there are diffuse costs within the state and local administrations that cannot be sorted out as electoral costs since they are part of standard costs of functioning state apparatuses (i.e., local governments, police)  or of private organizations (i.e., private schools functioning as polling centers at no cost). Such costs cannot be properly assessed unless program-focused cost accounting is made by the implementing agency, which is not usually the case in most governmental agencies dealing with a variety of activities (e.g., local governments, census and statistical authorities or civil registries). A typical agency budget is normally structured by separating personnel and non-personnel services (materials and procurements), but it generally difficult to determine how much of each is devoted to a given election-related activity (i.e. security, arranging for polling stations, supplying information for voter lists) as part of the larger cost of operations of the entire organization.

For example, in Spain and Sweden, voter lists are compiled in a national office of statistics. Although the electoral authorities pay a fee for the lists, this does not cover the full cost of the operation simply because there are hidden costs that could only be ascertained by program-focused cost accounting. There are other hidden costs in both countries, as in many others, related to the conduct of the polling operation by provincial and local municipal administrations, the postal services, police and embassies abroad.

Such a wide range of hidden costs is less likely in countries with small state infrastructures, where almost any electoral cost incurred at any level of government can be easily spotted and referred to a general electoral budget for provision or refund by whatever authority implemented a given election-related activity. For example, in Cambodia and Guatemala, a local employee at the municipal level is exclusively in charge of helping with civil registry and voter lists. Although the municipality provides certain office support, both the main office and salary costs are covered in the national electoral budget. A number of templates of electoral budgets are included as illustrative examples of budget structures in different democracy environments.

Next: How much do elections cost?

One major finding regarding the overall cost of elections is the importance of the type of democracy environment (i.e. stable, transitional and post-conflict) in determining both the kind and amount of electoral expenses. The conclusions from previous research, have been validated in the current study and are described according to the two categories below: 

Previous experience with multi-party elections

Electoral costs do not tend to decrease overall once civil conflict and severe political turmoil have been overcome, although some line items of expense may prove to be more sensitive than others to cost-cutting measures.  In other words, stability is much less expensive than civil conflict, but it cannot be taken for granted that electoral costs will decrease once democracy is stabilized and a state administration continues to develop. There are no grounds to expect that such massive processes as voter registration and voting would not be subject to the same challenges and expectations as other undertakings by a modern state administration (i.e., tax collection, educational services, postal services and communications).

There are more countries where electoral costs have increased during the last five years than those where costs have decreased. In about half of the countries, the budget was merely inflation-dependent and did not change in substance. The case of Switzerland, with a constant electoral budget over the last five years, illustrated this situation. In contrast, Canada, Spain and Sweden experienced high budget increases due largely to automation of voter lists and early transmission of results, expansion of postal voting, and intensive voter-information activities.

Although second- and third-generation elections following civil conflict tend to cost less than the first post-conflict elections, there are cases showing the opposite trend. In three Latin American countries (Nicaragua, 2001; Ecuador, 2002; Guatemala, 2003) later elections were more expensive than earlier ones.

Electoral expenses showing the largest growth tend to be for personnel and professional services. Sometimes this occurs following the establishment of an increasingly consolidated electoral administration, as in Cambodia and Guatemala; at other times, it stems from outsourced professional expertise in computer and telecommunication fields, as in Australia, Spain and Sweden. Voter information and postal voting are other expenses experiencing growth, especially in well-established democracies where sophisticated voter-information efforts offer individualized notification of date and place of the polling. For example, in Switzerland, voter information accounts for as much as 50 percent of the entire electoral budget. According to the survey response from the Australian Electoral Commission (AEC), the areas of operation experiencing the greatest increase in costs related to upgrading the AEC Web site (developing the ’virtual tally room’), administering a 40 percent increase in the volume of postal voting, and also administering the increased volume of overseas voting together with attendant security issues.

Postal voting within country and abroad is becoming increasingly popular, yet is quite expensive. In Spain, for example, the cost per registered voter grew from $2.1 in 1996 to $4.1 in 2004. While reporting and accounting factors may explain a large part of the growth, there is still considerable room for actual cost increases, which can be explained by significant changes in the management of the electoral process. In any case, cost increases have been much larger than the 10 percent increase in the number of eligible voters from 31.4 million in 1996 to 34.5 million in 2004. Some of the discrepancy is related to the change in Spain’s currency from the peseta to the euro as well as the fall in the US dollar’s value against the euro over the past couple of years.  These external factors are responsible for the estimate that at least 25 percent of the current cost per registered voter should be considered ‘inflated’ by mere accounting factors; taking this into consideration, the actual 2004 cost is $3 per registered voter.

Other factors are also responsible for the higher costs in 2004 compared with 1996. The major significant changes in the management of elections in Spain over those eight years included:

• A dramatic increase in the number of registered voters abroad for whom electoral material was mailed and processed (1.1 million people), plus other mail services to normal voters. Moreover, information facilities to electors in and outside the country were improved (i.e., by providing online information). This explains why the largest single line item in the most recent election budget was postal and telecommunications expenses, which accounted for over 21 percent of the entire budget;
• The increasing volume of printing and mailing by the Office of the Voter Registry accounted for approximately 13 percent of the electoral budget, the largest growth item in the budget. This increase resulted from higher levels of external voting; and
• Rapid transmission of preliminary results on election night. This activity has been improving technologically over time and now accounts for almost 10 percent of the total budget. 

In Sweden, the cost of technology and Web communications is fairly high due to the large investments that have been made. In addition, most of this work is done by consultants instead of permanent staff—yet another reason for the high cost. Voter-information efforts are mainly focused on the production of a magazine in 14 different languages (in addition to Swedish); placing information on the official Web site; producing brochures; and advertising on television and radio. Brochures and films in sign language and Braille are also produced. The cost for voter information is around 18 million kroner ($2.3 million) for each general election; this number increased by 60 percent between 1998 and 2002. Before the election to the European Parliament in June 2004, the central EMB participated in the government’s Democracy Campaign by organizing information seminars for representatives from municipalities, organizations, government agencies and schools. The authority focusing on schools included information brochures in its general dissemination plan. The central EMB is prepared to develop its information activities further before the general election in 2006; among the suggestions additions are the preparation of special information packages for schools and immigrants.

Special investments in new technology and Web-based communication have been made in recent years in Sweden. To fund them, a relatively high annual budget has been granted to the central EMB. The cost of IT support and technology in 2002, more than 20 million kroner, was almost double its usual cost. The new technology is mainly used to produce a voter register, assist in accurate boundary delimitation and in managing election results. In addition, election results are published electronically immediately after the close of polls on Election Day.

Some countries show dramatic reductions in electoral budgets. The case of Cambodia looks spectacular because the first elections were held as part of peacekeeping operation in 1993 and had an exceptionally high cost—nearly $46 per registered voter. Subsequently, the cost dropped to $5 in 1998 and to $2 in 2003. Expense-line items undergoing the greatest reductions in recent years include staff savings (Canada), voter education (Australia) and voter registration after establishing a permanent registry (Cambodia).

In other countries such as Spain, however, no significant cost reduction was noted from one election to the next—although in the longer term a number of cost-reducing measures have been singled out. Spain’s electoral authorities have gained experience over the past 20 years with a number of cost-effective measures:

• Considerable savings have come from the decentralized printing of ballot papers since the late 1980s (party lists are different in each of the 52 provinces), and by widening the bidding to a larger number of firms. Savings of 50 percent or more have been reported, moving the cost from 1.6 pesetas to 0.60 per ballot, a savings of around $2 million per election;
• The cost of voter-information campaigns was reduced enormously in the early 1990s by limiting campaigning to the state-owned media where airtime is free. Expenditures were reduced from approximately $13 million to around $1 million;
• As indicated above, decentralizing the procedures for the preparation and application of the budget, and making each agency formally accountable for its own budget and accounting, appears to have been cost-effective; and
• The electoral authorities conducted a study on the feasibility of substituting disposable ballot boxes for the currently used hard plastic boxes. Savings on storage costs, as well as on some production costs, are expected as a result; currently, about 30 percent of all boxes are unrecoverable after an election regardless. This type of cost-effective measure has already been introduced in other countries like Australia.

The expenses that remain constant vary greatly among countries. Examples include fees to registration and polling officials in Canada or Guatemala; training in Australia; general voter-information campaign by the EMB (Spain); voter education (Cambodia); and every single budget line item (Switzerland).

Next: Public funding of political parties

The source of funds for political parties varies around the world based on culture, precedent and legal standards. In most countries in continental Europe, political parties are at least partially funded from the national budget in the form of various types of allowances, including those for regular party operation, functioning of parliamentary groups of party representatives, and for campaign expenses in an election year¹. In the United Kingdom and the United States, private financing of parties represents the largest portion of campaign expenses. In emerging democracies, there are examples of the different legal models as well as of compliance in actual practice. A mixed model with public and private financing is frequently established with an obligation for disclosure by political parties and controlling authority held by the EMB. More often than not, a legal vacuum exists regarding sanctions for non-compliance or other enforcement mechanisms at the disposal of the implementing authorities.

In Australia, political parties and independent candidates receive public funding for campaigning, as per the Commonwealth Electoral Act 1918 (the Act). Under the Act, candidates or Senate groups are eligible for election funding if they receive a minimum of 4 percent of the formal first preference vote in the division (House of Representatives) or the state or territory (Senate) in which they stand for election. The amount payable is calculated by multiplying the number of eligible votes by the current election-funding rate, which is indexed to increases in the consumer price index every six months and was set at $1.97² per vote for the 2004 election. The total amount payable that election was $42 million. Following an election, key participants in the electoral process are required to lodge with the Australian Electoral Commission (AEC) various returns disclosing certain campaign transactions. These transactions include donations received and electoral expenditures by candidates and Senate groups; details of electoral expenditures, donations received; donations made to candidates and others by third parties; electoral advertising by broadcasters; and published electoral advertisements.

Regarding political party financing, the disclosure provisions of the Act are intended to improve the integrity of the electoral process by allowing the electorate to be well informed about the major donors to political parties, groups and candidates as well as the levels of some kinds of expenditures by those involved in the electoral process. The AEC receives and processes the various returns required under the Act, makes them publicly available for inspection (on the AEC Web site), and conducts a program of compliance reviews of disclosures made by the political parties and associated entities. Even as the AEC’s workload in administering these requirements and activities continues to grow, it is frequently criticized by parties and members of Parliament who are often interested in exposing the financial arrangements of their political opponents while seeking at the same time to limit the transparency of their own arrangements.

During the 2003–2004 period, some 184 annual returns were processed; 1,190 donor returns were processed; and 102 compliance reviews were conducted. According to the AEC Portfolio Budget Statements, the cost associated with funding and disclosure activities during 2004-2005 was estimated to be $2.8 million.

In Spain, a mixed model applies, with public funding provided to political parties for ongoing and electoral operations. Public funding for campaigning is not directly a part of the electoral budget, and parties can also receive funds from private sources. Campaign subsidies are calculated in proportion to the number of votes and seats obtained. Political parties also receive funds from their membership, private donors, their own investment revenues, and bank loans. Access to state radio and television is free of charge and is managed by a special commission supervised by the electoral committees.

Sweden has a rather peculiar mixed model that is based largely on informal agreements among political parties. For one thing, any political party, candidate or organization can conduct an election campaign. Organizations other than political parties were involved in campaigning prior to the referendum in 2003 on replacing the Swedish currency, the kroner, with the euro. Any business was able to fund campaign organizations prior to this vote, and many did so³. Regarding the financing of political parties, there is no system of regulation, although there is one law containing rules and moral guidelines for political parties. There are no provisions for disclosure of contributions to political parties; however, the political parties represented in parliament reached a mutual agreement in the 1960s to share information about sources of contributions with each other, but not with the general public. Political parties are also obligated to produce an annual report, which is also not made public.

There are four main sources of funding for Swedish political parties: direct public funding, indirect public funding, income from party membership, and lotteries conducted by parties or their affiliated organizations. The main source is direct public funding given to political parties during election periods and between elections, with the amount based on performance in the previous election and current representation in the legislature. Indirect party funding is typically provided through the party-affiliated press, which is publicly subsidized. Although membership in political parties has declined, income from membership still remains important, representing between 5-10 percent of the total budget of the two main parties in the early 1990s. Some political parties or their affiliated organizations receive income from lotteries; the Social Democrats are particularly successful in this respect. Although no law prohibits political parties from receiving funds from private businesses, party leaders agreed in the 1970s to refuse such contributions.

In Guatemala, public financing of political parties is established by the law. Parties are entitled to free postal and telecomunnications services from the calling of an election until one month after its conclusion. Moreover, political parties receive public funding in proportion to the number of votes obtained in the first round of the presidential election, at a rate of 2 quetzals per vote, if and only if a party receives at least 4 percent of total valid votes. Disbursement is made under the control of the national electoral commission, Tribunal Supremo Electoral (TSE), in four annual installments between elections. Political parties may use public funds for either ordinary operations or campaign expenses. The TSE has the responsibility for audit and control of party finances and expenses. Before each annual disbursement, parties must submit a detailed report of expenses. In the event that a political party does not abide by its accounting and reporting obligations, the TSE may bring the party before the courts in an effort to force compliance. Private financing is neither contemplated nor prohibited, and no limit on campaign expenses is established in the law. It follows that it would be extremely difficult to estimate campaign expenses except for the public subsidies component, even if all parties complied with the obligation for disclosure after elections.

In Cambodia, the Law on Political Parties states that parties should be funded through their members’ contributions, from income generated by lawful business activities, from donations of private enterprises or individuals, and from the political party’s own assets. Contributions from foreign firms, public or government institutions, and NGOs are forbidden. Regardless of their participation in the electoral process, all political parties are required to report annually to the Ministry of Interior their income and expenditures, balance sheets, statements of bank accounts, and assets. These documents are not available to the public. Regarding campaigns, the same law establishes that the State shall provide funding for campaigning to political parties on an equal basis; to date, however, this provision has never been applied. Existing laws also require parties to report campaign expenditures and maintain a special account registry showing their sources of income and expenditures. The registry must remain available to the election commission for examination, if requested. The only actual government contribution to campaigns of political parties comes under the provision of free airtime on state radio and TV. The electoral authorities arrange free equal access to airtime for political party campaign messages during the 30-day campaign period. This, however, does not apply to local commune elections.

In Haiti, few provisions or policies had been determined regarding political party campaign activities by the time the electoral budget was drafted and international assistance was pledged for the 2005 elections. This oversight illustrates the uncertainties and challenges that are typical of post-conflict environments.

In Afghanistan, the picture is also rather bleak even though laws regarding political party contributions have been passed. The political entities registered to take part in an election must have a dedicated bank account where all campaign contributions are deposited, as stipulated in the Joint Electoral Management Body (JEMB) Regulation on Political Campaign Finances. Each Afghan citizen or organization may contribute up to 2 million Afghani ($47,000) to a candidate per campaign period. Any contribution greater than 1,000 Afghani must be recorded and reported to the JEMB. The only kind of campaign contribution from a public source that can be accepted is in the form of security services. Thus, current circumstances hold that no public money is available to political parties, its candidates, or independent candidates running for office. Even though political entities are not allowed to accept funding from abroad, international NGOs have been offering courses and advice on how to organize a political party, develop a political program, and how best to run an election campaign in a democratic environment.

Notes:

¹Private funding in most of these countries is allowed, but it tends to be secondary in importance.

²All references to Australian electoral costs are in Australian dollars.

³The vast majority of businesses supported replacing the kroner with the euro, thus greatly tilting the balance of funding in favor of the “yes” vote. Some analysts attribute voters’ rejection of the euro at least in part to a backlash against the huge amount of money spent by euro supporters.

Next:  Electoral Assistance from the international community

In emerging democracies, the funding of an electoral budget by the international community, totally or in part, is an important component of democracy assistance. The amount varies from practically funding an entire election in a post-conflict scenario to more limited support of a specific phase of the electoral process, such as voter registration, civic education, and the provision of some materials like ballot papers and indelible ink. The following are illustrative country examples.

In Afghanistan, as in Kosovo since 2000, elections in 2004 were financed completely by foreign donors, with an overall cost per registered voter of more than $20. An additional peculiarity was the financial management of the election. Even though the JEMB was responsible for the election, it did not have complete authority over finances; that authority rested with UNDP and its Voter Registration and Elections Project. According to a conservative estimate by the unit responsible for the election budget, more than $190 million was spent to obtain 8,128,940 ballots, resulting in a cost of $23 per vote. This figure excluded all costs related to the security operations conducted by 18,000 US troops and 8,000 ISAF (International Security Assistance Force) soldiers. Other costs not included were related to expansive civic education activities and logistic assistance provided by international NGOs but paid by individual donor countries; the total amount of such assistance may have amounted to more than $30 million. More than 2,000 domestic electoral monitors were separately funded by USAID. Special election support teams from the European Union ($2 million) and the Organization for Security and Cooperation in Europe (OSCE) were also separately funded by their respective organizations.

The estimated total cost for the direct polling operation inside the country was $55 million. In addition, UNAMA spent almost $21 million, and support to the Ministry of Interior for security cost a further $10 million. The latter funds were primarily spent on salaries, new vehicles fitted with communication equipment, and other equipment (except weapons and ammunition) necessary for the national police to perform its duties. Not included in these totals was the $23 million contract to Global Risk Strategies (GRS) to field 96 international staff to conduct security coordination and perform logistical planning for the EMB in the field. In fact, to a very large extent, GRS became the JEMB’s field structure.

There was also the issue of funding external voting processes for Afghan citizens living elsewhere, particularly refugees in neighboring Iran and Pakistan. Significant resources were needed to set up and run, in fewer than 80 days, a registration and voting operation in those two countries for more than one million potential voters,. In Pakistan an in-person registration exercise was conducted prior to Election Day (this was not feasible in Iran). On Election Day, both Pakistan and Iran offered in-person voting at a limited number of polling centers, primarily near locations with large Afghan refugee populations. The cost, excluding resources spent by host countries (in particular, for security arrangements), amounted to nearly $30 million. This exercise resulted in 818,189 votes being cast.

It is not surprising that the 2004 elections in Afghanistan were so complicated to arrange and proved quite costly. The election authority charged with organizing the 2004 presidential poll was faced with an electorate that had not experienced an attempt to hold democratic elections in four decades. Instead, more than 25 years of violence had dominated the political landscape and colored the political climate of the country. A significant amount of money was therefore needed to recreate the election authorities’ infrastructure, as well as on civic and voter education to facilitate the electorate’s understanding of a democratic electoral process. The electoral administration had a huge pool of internationals teaching Afghans how to run elections via on-thejob- training, which affected the cost of the election. Security concerns also played a significant role. Preparing and organizing elections is more costly in a post-conflict society, as much of a country’s physical infrastructure has been destroyed and large segments of the population have moved internally or become part of a diaspora. The late decision to allow eligible Afghan refugees in Pakistan and Iran to take part in the poll further increased the costs.

In Haiti, it is likely that a three-election event in November 2005 will be almost entirely financed internationally. As in Afghanistan and Iraq, the electoral budget in Haiti is illustrative of the exceptional electoral setting in post-conflict societies. First, there is the overall cost of approximately $10.5 per registered voter, not including political party funding and international and domestic observer missions. The voter registration operation alone will cost around $2.10 per registered voter. Typical of post-conflict situations, infrastructure and technical equipment costs (i.e. transport, office space rental, communications and computer facilities) will require approximately $2.30 per registered voter. It should be noted that the electoral cost per registered voter in 2005 will be higher than it was ten years ago in 1995, when it reached $4 (López-Pintor 2000, p. 74).

The total resource requirement for the period October 2004 to December 2005 amounts to $48,893,180, which is expected to cover personnel and operational costs for municipal, congressional and presidential elections. The Haitian government will finance the cost of 82 headquarters electoral staff and also contribute $2.9 million to the extraordinary electoral budget, to which three main international donors (Canada, the European Union and the United States) have committed $41 million. International financial resources for operational items have been linked to objectives of quantitative output through a result-based framework created by the electoral commission. Of the $41 million committed by donors, which will be administered by UNDP, $9 million has been allocated for voter registration to be conducted by the Organization of American States (OAS). The international community has pledged additional funding for security services and civic education and election observation programs, including political party activities. The following countries and regional organizations have pledged or are considering technical assistance opportunities prior to the 2005 elections in Haiti: Brazil, Chile, Colombia, France, Japan, Mexico, Spain, Venezuela, the Caribbean Community (CARICOM), and the Organization of French Speaking Countries (OIF). The United Nations Stabilization Mission in Haiti (MINUSTAH) is also preparing to establish a Trust Fund aimed at mobilizing additional resources for the elections as may be required. The substantial contribution of the international community is the only hope to sustain the entire electoral operation, as the national contribution amounts to less than 10 percent of the total budget.

International financial assistance for Cambodia has consistently decreased every election since 1993. In 1998, international assistance funded almost 80 percent of the total cost of the election, but this percentage declined to around 60 percent for the 2002 commune elections and to less than 50 percent for the 2003 national election. The estimated cost per registered voter was $45.50 in 1993, but dropped dramatically to $4.40 in 1998, and continued to decrease to $2.30 in 2002 and $1.70 in 2003.

For the 2003 national election, the budget was funded at 24,344,723,000 riels ($5.68 million) from a Trust Fund for Elections in the National Treasury and 24,931,681,000 riels from the international community. Seven billion riels were provided to the National Election Committee (NEC) from UNDP; six billion riels was provided by one donor country to the NEC through the Ministry of Finance; and only 536 million riels went from donors directly to the NEC. The donors’ money covered, for the most part, the non-salary, non-contingency budget costs. UNDP signed funding agreements with 12 donor countries and coordinated the earmarked funding to cover all the various phases of the process in agreement with each country. The Cambodian government’s money was used almost exclusively for salaries. The NEC accredited over 1,000 international observers, with two most important observation missions being conducted by the Asia Foundation (305 observers) and the European Union (some 120 observers). The approximate cost was $1.5 million for the EU Election Observation Mission and $240,000 for the Asia Foundation observation effort. These costs offer proof that international funding of civil society activities in election monitoring sometimes comes in relatively massive amounts. For example, in the 2003 Cambodian general election, external funding amounted to almost as much as the entire electoral budget of the NEC.

In Guatemala, the electoral budget was generously funded by international assistance up to $9 million (not including international observation missions), with assistance provided mainly by Canada, Japan, Norway, Sweden, Switzerland and the United States. Of the $9 million, $3.5 million flowed through the Organization of American States (OAS) for technical assistance on voter lists, and the remaining $5.5 million was provided directly to the national electoral agency. The distribution of this aid by item was approximately as follows: voter registration, $2.5 million; voter education, $2.5 million; purchasing of equipment, $2 million; and training for poll workers, $2 million. To a lesser extent, external aid was provided to civil society organizations monitoring the elections. All in all, electoral expenditures amounted to around $23 million, which is equivalent to $4.6 per register voter. There were also several large international observer missions, including particularly extensive ones from OAS and EU, with an average cost of around $1 million each.

The main issues regarding voter registration costs have to do with the type of registration system (permanent versus ad hoc, automatic versus show-up update); institutional locus responsible for voter registration (the EMB or a separate agency); and degree of resilience in cost assessment (easily identifiable versus diffuse costs). The following conceptual considerations, which are likely to be relevant to a correct assessment of voter registration costs, are excerpted from a UNDP assessment mission in Cambodia in August 2004.

For more information covered in this section, see:

• General considerations on the purpose of voter list in comparison with other registries of people
• Voter registration systems in a global perspective
• Quality standards of voter lists
  
• Managing the cost of voter/civil registries: Country Experiences

Voter lists are meant for the exclusive purpose of recording the enfranchisement of voters and facilitating the voting operation by which the citizens eligible to vote (with given age and legal conditions) exercise their right to suffrage on a periodic basis. Voter lists are the tangible proof of enfranchisement, and the quality of such lists is to be judged according to their capacity to serve that specific purpose.

Other registries of people exist and are meant to achieve different purposes: civil registries, lists of residents, and different personal identification documents (i.e., identity cards, passports and driver licenses). The number of registries and the institutional locations of each registry vary from country to country, depending on historical and legal traditions.

Civil registries, which are usually based at local government levels, serve the purpose of recording and certifying births, deaths and marital status. Certification by civil registry is the main proof of citizenship, which in turn is generally based on nationality. Civil registries also offer proof of age—and both citizenship and age are the fundamental requirements for voter enfranchisement. Other lists are compiled Voter lists should not be considered per se as registries of citizens or of residents. Voter lists are intended neither to help with the identification of citizens, nor (even less so) to serve as a resource for the quantification of the population of a given country. Voter lists essentially exist for the purpose of allowing individual voters to vote only within a given constituency and at a given polling station.

In the production and maintenance of voter lists, the kind of links to be expected between voter lists and other types of registries basically depends upon two factors. One is the legal administrative tradition of a given country. The other refers to the special circumstances under which voter registration is organized, most frequently in the case of emerging democracies and post-conflict situations. In the latter case, problems stemming from a deficit of population statistics, civil registries and citizen identification, as well as significant displaced populations, may strongly influence the procedures whereby voters are registered. In general, the main link between voter lists and civil registries is the requirement of a birth certificate as proof of citizenship (which is normally defined by nationality) and, at least in the first instance, a proof of residency (which would normally be required for the allocation of individual voters to constituencies, polling centers, and polling stations). Voter lists and personal identification documents are linked by the necessity for the voter to be identified at the ballot box. Finally, the link of voter lists with population censuses can be found in the facilitating role of census data for the organization of early voter registration, and the disaggregating of voter lists into constituencies. Also, population statistics may be crosschecked with voter lists to assess the size or age of segments of the population that will be eligible for enfranchisement at the present time or in the future. Or provide citizens or residents in a country with certain personal documents, such as national identity cards, passports or driver licenses. These activities are frequently undertaken by some branch of the security administration, normally the national police, and these documents exist for the purpose of personal individual identification. A third type of registry is a list of residents, which may be compiled, maintained and updated by local authorities in the districts of large cities, townships, communes or villages. These lists may serve different purposes normally related to the delivery of public services at the local level, and to provide information to higher levels of government for their respective action (i.e., correcting or updating other lists of citizens on which residence information is required, like the voter lists). Finally, another kind of documentary record is the population census, which is usually compiled every 10 years for the production of aggregate statistics of the population at different territorial levels of a country. Aggregate census information (but not data on individual persons) has many different uses, including the definition of electoral constituencies, which may be based on population, and the allocation of parliamentary or local council seats to constituencies according to population.

Next: Voter registration systems in a global perspective

Voter registration can be conducted on an ad hoc basis only for a specific election or organized on a permanent basis either with periodic updating or with updating as a more continuous process. Ad hoc registration occurs least frequently, with the exception of post-conflict elections, which are often conducted with an ad hoc registry of voters. Nevertheless, there are exceptional cases even in well-established democracies where a permanent voter registry was instituted only recently (i.e., Canada in 1997).

In addition, voter registration may be compulsory or voluntary for citizens. A global survey of 124 countries by International Institute for Democracy and Electoral Assistance (IDEA) found that compulsory voter registration is quite common in Western Europe as well as in Central and Eastern Europe. In those nations, it is organized on a continuous basis whereby the voter is automatically registered on a voter list after turning 18 years of age, with no requirement to appear before any public authority. This system is based on individual citizenship and age information, provided from either civil registry records or residents’ rolls compiled at local government levels.

Compulsory registration is not widely practiced in Africa or North America. In South America, the two practices are evenly distributed. Again, much depends on the way civil registries and residence records are organized.

Permanent voter registers are updated either on a continuous basis or at specific time intervals, normally during the period immediately preceding an election. Continuous registration requires an appropriate infrastructure to maintain the register either at the electoral administration or at the civil register. This involves adding the names and other relevant information for those who satisfy eligibility requirements (attaining citizenship, satisfying residency requirements, and attaining voting age) and deleting the names of those who no longer meet the eligibility requirements (usually because of death and change of residency). Continuous registers are used more often than periodic registers in all regions of the world, despite the complex machinery and high cost incurred. It is important to note, however that voter registration through a periodic voter register is ultimately a more expensive operation than maintaining a continuous register (IDEA, 2002, p. 30).

Next: Quality standards of voter lists

Democracy means rule by the people, and the purpose of a voter list is to register the people who are eligible to exercise the right to vote. The following standards of democratic quality are particularly relevant:

• Inclusion of all eligible voters. Any discrimination against individuals who are perceived to be political opponents should be avoided, and special sensitivity should be applied regarding certain sectors of the population that tend to be more vulnerable to disenfranchisement (e.g., women, rural residents, illiterate individuals and ethnic minorities). Comprehensive and accurate voter registers should be considered a prerequisite for free and fair elections (IDEA 2002, p. 33).
• Facilitation of the voting operation. In structuring voter registries, the fundamental guiding principles are simplicity of procedures and convenience for the citizen. Because the exercise of a fundamental right is at stake, governmental and administrative authorities must not hinder the exercise of that right or obstruct election-related authorities. Simple administrative procedures must be put in place, and the process for registering must be made as convenient as possible, for example by reducing the need to travel long distances and eliminating payment of heavy fees on stamps and photographs or having to appear several times at registration centers (IDEA 2002, p. 25). This democracy standard requires that, independently of how the list of voters is produced (for only one election or as a permanent or continuous register), the voter lists must be publicly exposed for a given period of time for additions, corrections or deletions on the initiative of citizens.

There are also standards of technical efficacy such as the following:

• Accessibility to eligible voters (this should be considered a technical as much as a democracy standard);
• Ease of use on polling day. This refers to the way in which the voter rolls are disaggregated by polling center and the ability of voters to find their names on the appropriate voter list;
• Updated lists either periodically or on a continuous basis; and
• A clean registry, which should discourage double voting and help produce accurate electoral statistics. The main issues here are procedures for the deletion of the deceased, the inclusion of changes of residence so that voters are registered in the constituency where they live, and cross-checking to eliminate multiple registrations of the same individual.

Since the achievement of absolute perfection in almost any ongoing population record is unlikely, the main guiding principle for voter rolls, according to IDEA, is that enrollment procedures:

 “need to strike the right balance between the need to be rigorous to ensure integrity of the rolls, and the need for flexibility to ensure that people’s rights to enroll and vote are protected.” (IDEA, 2002, p. 34).

Next: Managing the cost of voter/civil registries: Country Experiences

In Australia, voter lists are automatically updated with civil registry information. The Australian Electoral Commission (AEC) divisional staff continually process enrollment information and enter it into the computerized Roll Management System (RMANS) in order to ensure an accurate and up-to-date electoral roll. This includes information about newly eligible persons, electors changing addresses, and deletions of electors who have left their enrolled addresses or died. The staff also process information received from Continuous Roll Update (CRU) activities and elections. Data matching programs undertaken in conjunction with other Commonwealth, state and territory authorities have also assisted the AEC in confirming enrollment details for approximately 43 percent of occupied addresses that can be enrolled without the need for direct contact with residents. The AEC also provides roll products for joint roll partners (state and territory electoral bodies), senators and members of Parliament, registered political parties, medical researchers, government agencies and authorities, and the public. The costs associated with the roll administration activities in 2003–2004 amounted to $58 million (a period leading up to the national election), and the estimate for the 2004–2005 financial year was $46 million.

In Sweden, the voter register is compiled on the basis of the civil register, which is continuously updated by the National Tax Agency, the organization that since 1971 has been responsible for keeping a correct civil register. Before every election, the central EMB acquires from the agency a complete voter register, which is extracted from the database containing the civil register. The central EMB pays the agency a fee for this information; the amount for the euro referendum in 2003 totaled 500,000 kroner ($64,000). The amount is calculated on the basis of the number of persons in the register, with the per-name being 2-5 öre (1 kroner = 100 öre). The central EMB receives the voter register in an electronic format and produces printed registers for each polling station. This is done with the use of statistics and maps provided by the Land Survey. The cost of dividing and printing the register by polling station is equivalent to 300,000 kroner. Additional costs (almost 13.5 million kroner) involve the printing and dissemination of voting cards to all eligible voters.

In Spain, lists of voters have a permanent automatic character. They are substantially produced and updated with information based on civil registries and lists of residents provided at the municipal level. Voter lists are handed to the provincial office of the Office of the Voter Registry and then compiled at a national level. Lists were updated on a yearly basis until 1998, when monthly updates were established. Lists of voters are distributed to the municipalities for a five-day revision period before every election. Following revision, the Office of the Voter Registry distributes voter lists organized by booth to each polling station. Also, voter lists are distributed to political parties, but not to the Ministry of Interior, which receives only aggregate figures of voters at different levels of government.

Costs related to the routine production and monthly update of voter lists are part of the ordinary budget of the National Institute of Statistics, which totaled 200 million euros in 2004. The costs are not officially quantified separately in the organization of the Institute’s budget, but could be more than 15 percent. In its turn, the Ministry of Interior includes all costs related to the use of voter lists at election times (13 million euros in 2004). These costs cover reproducing the lists to be handed out to political parties; postal communication to voters informing them of the reviewing period before every election and of their specific polling locations; and the management of external voting. This extensive outreach effort explains why voter registration-related expenses have shown the largest actual growth over the last five years. The Ministry must reimburse the Office of the Voter Registry for these costs.

Spain allows external voting by mail, both within the national territory and abroad (the 2004 voter list consisted of 34,557,370 individuals, of whom 1.1 million were voters living abroad). Voters from abroad are included in a separate list of absentee voters that is produced by the Office of the Voter Registry with information provided by Spanish consular offices. At election time, consular offices share management of the ballot abroad with the Office of the Voter Registry. Voters first apply for external voting at the consular office; then most of the ballots are mailed directly to the corresponding local electoral commission, and a smaller amount is handled directly by the consular office for remittance in Spain. All external ballots—including those from within the country—are supposed to be deposited at local electoral commissions within three days following Election Day since the official vote counting must take place on the third day. External voting operations have recently involved more than 600,000 people living in the country and 295,000 from abroad, amounting to 1.8 percent and 0.8 percent of the entire electorate, respectively. Voting abroad has continued to increase over time, while in-country mail voting tends to remain unchanged (Ramos Vadillo, 2003; Brugarolas Masllorens, 2003).

In Guatemala, voter registration costs amounted to 20 percent of the total electoral budget in 2003. The rising cost was due to a particularly complicated method of updating voter lists by which the voter lists and polling stations were split into two sections: updated and non-updated. The process of updating the voter register by allowing people to vote closer to their residence resulted in a split register—nonupdated voters and updated voters. This made the entire operation not only costly, but also created much confusion among the electorate. A technical problem while processing updated voter information produced an unknown number of legally registered voters who could not be found on the updated register but were still on the non-updated list; the problem stemmed from the fact that the electoral law states that citizens must vote in the municipality where they are registered. The assessment by the European Union Observer Mission concluded that a main liability of the electoral process was the failure of organizing voter lists when allocating voters to polling stations in accordance with the identification records previously given to the almost two million voters who had updated their registration or registered for the first time. This shortcoming was recognized as a major issue a few weeks prior to Election Day, but the problem could not be properly addressed by the national election commission. As a consequence, many voters were unable to exercise their right to vote even after the commission decided to allow for a tendered ballot in the first round for those duly registered voters whose names did not appear on the voter list of the corresponding polling station (EU EOM Report, February 2004). Despite recent efforts to improve the rate of voter registration, accounts continue to be received of actual, if not legal, disenfranchisement of certain populations (primarily rural residents and indigenous people). (López-Pintor and Gratschew, 2002, 37).

In Cambodia, voter registration costs in 2002 were part of the commune election budget and were close to $1.00 per registered voter. In 2003, the update of the list was also part of the electoral budget and amounted to $0.50 per registered voter; the update of a permanent registry in 2004 was estimated at $0.14. Therefore, this is a case where registration costs were lowered in comparison to the overall cost of elections, largely due to the establishment of permanent, automatically updated registries. The cost of voter registration operations for the commune elections included the production of photographic identity cards for each voter. The total voter registration operation amounted to 20 billion riels ($4.9 million), which was included in the electoral budget of the commune elections.

In 2003, Cambodia created a permanent voter registry based on the voter list used in 2002 for the commune elections. Registration operations were delegated to the commune administrations but were closely supervised by the National Election Committee (NEC). The operation was an update of the previous voter list: voters already on the list did not need to re-register. Only those who had reached voting age or who had moved from one commune to another were required to register. Under this new system, no voter card was issued. As a result, the total cost of voter registration amounted to 12 billion riels ($2.9 million) and was incorporated in the 2003 electoral budget.

In 2004, as required by the Law on the Election of Members of the National Assembly (LEMNA), updating of the permanent registry entailed an annual voter registration and revision of the list even if there were no upcoming elections. Voter registration and list revision are conducted between October 1 and December 31 of each year. Therefore, a new valid voter registry is generated each year. For these operations in 2004, the NEC prepared a budget of 3.7 billion riels that was approved and provided entirely by the government. This amount was not part of any electoral budget. In future, this recurrent expenditure may either be granted on an ad hoc basis as happened in 2004 or eventually be integrated in the annual functioning budget of the NEC. The cost of voter registration will likely continue to decline as the procedure becomes part of usual commune administration operations. In 2004, savings had already been realized on training and direct assistance to commune clerks.

Since registration operations are no longer conducted directly by the NEC, part of the cost is now indirectly absorbed by the commune councils, which provide the personnel. Although the financial burden of voter registration seemed to be ‘passed on’ to the commune councils, the NEC is bound to provide “appropriate training, capacity building, facility, supplies and materials and budget to the commune and clerk to enable the implementation of these responsibilities.” Regardless, these costs are already lower than those of a periodic registration. The NEC also provides salary supplement incentives to commune clerks for the additional administrative burden during the annual update period.

In Haiti, as in Guatemala recently, the voter registration operation for 2005 will be conducted with technical assistance from the Organization of American States (OAS). Of a total electoral budget of nearly $49 million, some $9 million will be allocated to the task of voter registration, which ideally will include the possibility of creating the basis for a civil register, still non-existent in the country. The overall voter registration process will be automated, and no manual registration will be made. All voters will receive a digital card that will gradually replace the national identification card. The OAS is planning to establish 615 offices for registration activities (400 registration centers, 185 automated registration offices, and 30 automated registration mobile units). The registration centers will be located in rural areas; the automated offices in urban areas; and the mobile units in the semi-urban areas. The number of automated centers will be adjusted as required to reach a larger number of potential voters. It is estimated that the registration process will take four months (March–June 2005). The concept of operations also provides for the establishment of approximately 3,200 polling centers. All automated registration offices will also be considered as polling centers, and voters will be advised where to vote during the registration process.

Haiti’s provisional electoral council and the international donors have approved this project, and the OAS is already implementing it.

The government will directly fund operating costs for approximately 82 staff at the electoral council’s central headquarters. Provision of $2,017,167 under this heading reflects the need to hire voter registration officers. According to the electoral law of 1999, each voter registration office should comprise a staff of five: a president, vice president, secretary, and two voter registration officials. The monthly salary of voter registration officers is approximately $111, while voter registration officials receive $97 for a two-month period.

Voter registration in Afghanistan prior to the 2004 election was particularly complicated and even hazardous. The actual size of the electorate was unknown, and identity card ownership was very limited among the population. Therefore, a nationwide registration exercise was implemented, resulting in what was regarded as a largely successful registration drive. Due to the security situation, all registration centers were required to have communications (satellite phones in most cases) and to be well protected and mobile. Much money was spent on acquiring thousands of vehicles, in addition to setting up the entire registration organization to run a computerized register. With a $74 million budget, the voter registration exercise was the single most expensive component of the entire electoral process. An additional $7.4 million was spent on registering refugees in Pakistan the week before the poll. Unfortunately, the voter register was never used in the polling stations on Election Day because voters were allowed to cast their ballots in any polling station in the country. The EMB has yet to decide how to turn the current voter register into a continuous voter registration process.

Next: Electoral cost management practices

A few general considerations are pertinent in a discussion of cost management. First, expense processing and cost management are handled in very different ways in elections that are fully managed by international organizations (i.e., Cambodia in 1993, Kosovo from 2000 to 2002, and Afghanistan in 2004); elections administered by national authorities with substantial financial support from the international community (i.e., Angola in 1992, Guatemala in 2003, and Mozambique from 1994 to 2004); and elections fully handled by national electoral authorities (i.e., well-established democracies). In the first scenario, expense processing is handled directly by the international agency in charge of election finances—the OSCE in the Balkans and UNDP in other regions of the world. Budgeting, procurement and auditing procedures are followed in accordance with the rules of the relevant international organization. In financially sustainable, well-established democracies, expense management is conducted by the relevant department within the electoral administration in accordance with domestic public management regulations, some of which might pertain specifically to election expense management. Finally, in situations where national electoral authorities are only partly subsidized by the international community, management of expenses charged to external funds tends to be totally or partly in the hands of international implementing agencies like UNDP and the International Organization for Migration (IOM), or a regional organization like the OAS.

Another issue concerns the management of political party and campaign finance by public agencies—an important and sensitive issue in both stable and emerging democracies. Significant variations exist among countries in terms of public versus private funding, legal limits on campaign expenses, obligation for disclosure by political parties, and enforcement responsibilities of electoral authorities. Total or partial public financing of political parties is already common in most democracies, and in some countries such financing represents the fastest-growing electoral public expense over the last five years (i.e., Canada). A correct assessment of campaign costs and financing is of utmost relevance for reaching some sound conclusions about whether the playing field is level for all parties and participants. The fact that political finance reporting and control is methodologically complex and politically sensitive should not keep it from the forefront in both election management analysis and democratic quality assessment.

Regarding specific management practices, the following issues deserve attention: procurement procedures, inventory control, cash transactions, cost overruns, budget audit, high-technology investment and cost-effective measures. Each is described in detail below.

Procurement procedures for equipment, services and supplies are generally prescribed by government guidelines and practices, often complemented by EMB regulations and practices. The guidelines have become increasingly standardized around the world since the 1960s in response to regulations and pressure from international assistance and finance agencies like the World Bank, the International Monetary Fund, and UNDP.

An interesting case of adapting general governmental prescriptions to the specificity of election expenses is that of Spain. The state administration’s general procedures for expense management are applied with some modifications: a given expenditure requires financial clearance by the general expense-controlling agency within the executive branch of government. This is to ensure that expenses are handled according to budgetary and public contractual provisions. Until the mid-1980s, electoral expenses followed standard clearance procedures, which made management burdensome and dysfunctional due to the short timeline typical of the electoral process. (Making funds available for an after-expense justification was the procedure actually followed to make elections possible). Further legal reform introduced the modality of permanent financial control; in this system, prior clearance for every expense is not necessary, yet the controlling authority retains the right to examine electoral expenses at any given time in the process. Moreover, the Ministry of Interior must submit a detailed report on expenses to Parliament after the election. As for expense management by the different agencies involved in the process, each agency may spend the anticipated allowance in the budget from its own funds, with the understanding that it will be reimbursed by the Ministry of Interior. Such expenditures might cover voter registration expenses other than the current maintenance of the voter list to external voting—and certainly polling operations at lower levels, which are managed by local authorities.

The situation in Cambodia, an entirely different political-administrative environment, illustrates the difficulties in obtaining disbursements of the budget. In practice, the difficulties concerning electoral funding stem from the Ministry of Finance’s procedural and administrative practices, which are not geared for quick response in the context of election preparations. The standard finance procedures applicable to an ordinary government department are not capable of responding to the National Election Committee’s exceptional time constraints in terms of procuring election-related materials and ensuring unavoidable cash-flow disbursements during the organization of elections. In addition, direct funding from international donors or through UNDP tends to be disbursed quite late in the process, sometimes even after the election, thus creating additional cash-flow problems. In some cases, excessive earmarking may hamper the timely utilization of funds and may require last-minute changes to funding agreements to fit the actual operational expenditures more flexibly.

As in the case of procurement, inventory of unused supplies is conducted according to general government regulations and varies substantially among countries. For example, in Spain an inventory of both used and unused supplies is made following an election. Local municipal authorities are in charge of making an inventory and storing electoral materials. They pass on this information to the national government’s provincial and regional delegates who, in turn, consolidate the inventories at the provincial level and report to the Ministry of Interior. No special or unanticipated costs are incurred in this activity other than current operational costs of the different administrations involved. Supplies that can be reused are stored on municipal premises. Unused supplies, mostly paper products, are discarded or recycled at the local level.

Inventory control in Australia requires that arrangements for the storage of unused supplies be included in the contract when ballot paper is purchased. This is treated as Australian Electoral Commission inventory and is carried on the balance sheet at cost. It is all processed electronically. 

In Canada, electoral supplies and materials are tracked using the Supply Management System, which is an application developed in-house that uses a weighted monthly average cost to estimate the value of unused supplies. In Sweden, an inventory of unused supplies takes place after each election; unused material without text, such as voting envelopes, can be used in future elections. The central EMB always keeps in stock a certain amount of voting envelopes and paper used for printing ballots. This stock is regarded as a reserve. The local EMBs store materials used in polling stations, such as ballot boxes and polling booths. These are normally made of strong, long-lasting material and do not need to be replaced often, which is why they constitute a very low and rare cost in the budget of the EMB.

In Guatemala, sub-national electoral commissions conduct an inventory of used and unused supplies, focusing particularly on long-lasting materials such as screens and ballot boxes, which are stored for the next election. The national electoral authorities keep a record of all inventories. In Cambodia, all unused supplies are inventoried and stored in a central warehouse. Each year, a committee for inventory is established and performs its task for all departments of the National Election Committee. In addition, the department of operations receives monthly reports on the movement of supplies and materials in and out of the warehouse.

Developments in Afghanistan in 2004 illustrated some of the uncertainties and complications that can arise from the electoral system itself. The electoral formula for the presidency requires a run-off between the two top candidates if no single candidate receives 50 percent plus one vote in the first round. In planning for this possibility (which ultimately was not necessary because Hamid Karzai won more than 50 percent in the first round), the electoral Secretariat had to purchase significant quantities of polling day commodities such as tamper-evident bags, seals and indelible ink. These unused materials are now stored in a borrowed UNICEF warehouse in Kabul. Currently, the Joint Election Management Body (JEMB) and its Secretariat are trying to decide where and how to store ballot boxes and other polling material. The JEMB is fully reliant on the UN’s inventory system and has not yet developed its own policies and procedures.

Regarding payroll procedures and cash transactions, the rule is that salaries are seldom paid in cash¹. In general, cash transactions are limited to minor purchases and petty cash. In Australia, for example, the election commission handles very few cash transactions, and these are restricted to the use of petty cash (limited to $100 per transaction) for which there are established controls and accounting procedures within the organization. Likewise, deposits for the nomination of candidates and fees for party registration can be paid in cash.

In Spain, small allowances such as per diem for poll workers are paid with checks and bank transfers, although there have been cases when cash payment has been facilitated by local authorities. Special allowances for security services provided by the National Police are included with salaries in the corresponding monthly payroll. (Per diem assigned to polling and security officers was most recently established at rates of about S$60 and $98, respectively). In Sweden, the EMB handles no cash transaction except for small purchases of certain office supplies.

Note:

¹There are some exceptions, including Cambodia. In that country, salaries are paid in cash but all other transactions are made through bank transfer or by check.

In general, the Cost of Registration and Elections (CORE) Project survey indicated that election managers tend to keep expenses within limits anticipated under budgetary premises. Covering cost overruns does not seem to be a major problem regardless of the kind of democracy setting: stable, transitional or post-conflict. In the rare case of unanticipated expenses, electoral authorities have requested and obtained from the government some extra funds whose potential necessity was anticipated in electoral legislation (e.g., Canada, Guatemala, Spain), or the expenses were netted against surpluses within the agency’s approved budget (most often in Canada).

Accounting procedures and audit practices in the surveyed countries generally follow widely shared international standards, at least formally. Electoral expenses are normally subject to some internal audit at the EMB, plus an external audit by the national accounting control agency. An annual audit cycle is the norm. Accounting generally applies to income and expenses organized by broad chapters or budget lines (i.e., personnel, procurement services) encompassing the entire electoral operation. Program focused accounting is not usually applied (i.e., the overall cost of voter registration as part of activities by the different agencies involved). Some specific country examples and their variations are detailed below.

In Australia, the electoral budget requires approval by the Department of Finance on an annual audit cycle. In Canada, the Office of the Auditor General may decide to perform an audit or evaluation at any time. In Spain, all electoral expenses are audited by the national audit agency on an annual basis. This agency also audits the finances and expenses of political parties. An internal audit by the Finance Ministry’s auditor general can be made at any time during the process. In Switzerland, the budget is audited by the Finance Committee and the Finance Delegation of the national parliament. In Guatemala, there is an internal audit at the election commission and an external audit by the National Controller General on an annual cycle. In Cambodia, the budget is internally audited by the Ministry of Economics and Finance on a yearly basis; donor funds, however, are audited by external agencies. A national accounting audit agency remains to be established in Cambodia.

New technology is not only considered a pressing necessity for electoral modernization, but it has also become a status symbol for many organizations and countries. The question is not whether to employ new technology, but how much to use and at what cost. Examples below from countries at different levels of socioeconomic development examine this feature of electoral systems.

In Australia, the Australian Electoral Commission (AEC) has proposed to upgrade its Roll Management System (RMANS) and the supporting IT capacity. This will be undertaken over a four-year period and will enable the MANS to integrate with other Web-based systems, exploit image and optical character-recognition technologies, meet e-government standards, and more easily adapt to new technologies as they emerge. Pilot programs have been undertaken to test the use of optical scanning and character recognition to process applications for enrollment and other forms. An automated postal voting system is also envisaged. The trials have successfully collected Continuous Roll Update (CRU) enrollment application data and transmitted the information to a RMANS test database. The results of these tests will be reported to government by June 2005. Over the next two to three years, the AEC will further enhance the security of its IT network with the rollout of additional network attached storage (NAS) servers to each division.

In Sweden, investments have been made in new technology and Web-based communication during the past few years. The cost of IT support and technology almost doubled in 2002, when it was a little more than 20 million kroner ($2.6 million). In 1994, the Swedish central EMB was first in the world to publish preliminary and later final election results in real time on the Internet. Results for the parliamentary elections of that year were reported through a reporting receiver to the central EMB (then located at the National Tax Agency) directly from the polling stations. In the parliamentary elections of 1998, the final election results were reported in this way also. The central EMB was established as an independent authority in 2001, and one reason for the structural change was to facilitate the installation of new technology.

In Canada, all aspects of the administration of federal elections, except voting, have been computerized since the 1990s, and the systems are upgraded constantly. The National Register of Elections was established in 1997 as a permanent voter list to replace a door-to-door enumeration system, and this has resulted in considerable savings. Elections Canada also developed REVISE, a system used for the revision of the preliminary lists of electors during an election and to oversee the integration of changes into the Register. In 2003, Elections Canada developed REVISE-2, which allows changes of address across all electoral districts. Also, a Web site has been developed on which a range of training materials, handbooks, forms, election returns, election results and maps are made available.

In Spain, permanent voter lists are computerized and updated on a monthly basis. Early transmission of results is facilitated by high-tech communication. The electoral unit at the Ministry of Interior has developed a Web site with electoral information and results. The use of electronic voting is being tried on a pilot scale. At the national level, a branch of the National Police—a civil guard composed of 70,000 individuals— voted by Internet for its Staff Advisory Committee in 2002 and 2004. On a much smaller scale, a rehearsal with Internet and cellular telephone voting took place in four small municipalities during general elections in May 2004. Prime Minister Zapatero announced in August 2004 that electronic voting might be used in the referendum on the European Constitution early in 2005. Apparently, electronic voting is seen more as a way to encourage voter turnout than to reduce costs.

In other countries like Guatemala, office management and the voter registries are computerized. Furthermore, quick vote counting and transmission of preliminary results are subcontracted with a private firm. Nevertheless, there are no plans to introduce other new technologies like electronic voting.

In Cambodia, the electoral list is compiled in a centralized database. For the past three elections, lists for each polling station have been generated by computer. (A new computer system costing $220,000 was offered to Cambodia in 2001.) Efforts are ongoing to improve technical aspects related to the sorting of voter names and the cleansing of double registrations; the budget for this is around $20,000. There are plans to change the software platform to one that is less expensive to maintain and to provide computer equipment to all provincial offices, including networking capacity; the prospective budget for these operations is $103,000. Such initiatives are bound to improve the technical quality of the list while reducing the costs of technical maintenance. The production of the list on CD-ROM will enhance its accessibility and considerably reduce the cost for political parties to buy the list for their own use.

Some final considerations regarding cost effectiveness in elections include the following:

1. In the specific scenario of emerging democracies, dramatic cost decreases may be expected once peace is achieved and reconciliation fares well. All steps toward and investments in peace, security and national reconstruction—disarmament, demobilization, integration to civilian life, interethnic reconciliation and infrastructure development—lead to a reduction in the integrity costs of elections. Other electoral costs may remain constant or even increase (i.e., personnel and technology), but significant integrity costs will certainly be reduced with democratic progress. In addition, just as national security, transport and communication infrastructures are of paramount importance in determining integrity costs, state building is a key determination of core electoral costs. Few would doubt that democracy is much less expensive than civil conflict, but decreasing electoral costs should not be assumed once democracy is stabilized and a state administration continues to develop well. There are no grounds to expect that such massive processes as voter registration and voting are not subject to the same challenges and expectations experienced by other undertakings by modern state administrations (i.e., tax collection, educational services, postal services and communications).
2. Previous and current research indicates that duration of electoral practice is in itself a cost-reducing mechanism, perhaps the most important one during the stage of democratic consolidation. Since a longer-term perspective is by definition difficult when assessing election costs in new democracies, the research findings offer strong support for the claim that efforts at capacity building in electoral administrations are probably cost-effective in the longer term. These findings also support the idea that establishing and consolidating a permanent electoral administration as the repository for managerial capacity—within both the political and the administrative systems—is a cost-effective practice.  (Cambodia was among the countries showing a dramatic reduction in the electoral budget. The first election within a peacekeeping operation in 1993 cost approximately $46 per registered voter; subsequently, per-voter costs went down to $5 in 1998 and $2 in 2003.
3. In well-established democracies, electoral costs tend to increase due to personnel expenses, high-tech investment and maintenance, and special efforts to enfranchise certain populations (i.e., aging people and citizens abroad). For example, postal voting within country and abroad is growing and is rather expensive.
4. Concerning the cost of voter lists, a main lesson learned from previous research (and still relevant) is that permanent registries promote both transparency and cost-effectiveness,  particularly when they are periodically updated with corrections, additions and deletions without obliging voters to re-register. Recent reforms in this direction have been implemented in a number of new as well as older democracies including Botswana, Cambodia, Canada and Chile. The single most important cost-cutting measure is probably continuous voter registration. Given the huge costs involved in undertaking voter registration operations for the first time, permanent registers that can be updated periodically will prove cost-effective in the long run (UNDP, 2000, pp. 126, 128).
5. Regarding the use of voter cards, a lesson learned is that allowing citizens to vote with a variety of identification documents, such as a driver’s license or passport, rather than requiring a voter card, should be considered good practice. The use of a voter card does not in itself add anything to democratization; furthermore, the high cost of producing them, in addition to the tasks involved in distribution, can deter voter turnout and otherwise delay or disrupt the electoral process. Moreover, the use of voter cards does not prevent multiple voting, which can be restricted by other means ranging from indelible ink to computerized systems (UNDP, 2000, p. 126).
6. Expense line items undergoing the greatest reductions sometimes refer to staff savings (Canada), voter education (Australia), or voter registration after establishing a permanent registry (Cambodia, Canada). In contrast, in other countries like Spain, no significant cost reduction can be reported from one election to the next, while in the longer term a number of cost-reducing measures have been singled out. Spain’s electoral authorities have developed experience during the last 20 years with a number of cost-effective measures.
7. High-technology investment in itself should not be considered as a cost-effective measure unless it is related to products and processes leading to long-term cost reductions in the electoral operation. Such would be the case with permanent voter lists vis-à-vis ad hoc voter registration before every election. Indeed, electoral authorities may reasonably expect some cost savings by introducing new technologies.
8. Political party finance and campaign expenses fall largely in a grey area concerning electoral costs. Some cost effectiveness can be expected in these areas by filling the legal vacuums prevailing in so many countries (on issues like expense limits, disclosure obligation, sanctioning and enforcement powers of public authorities). Political parties may receive public subsidies and international public aid, which are relatively easy to track and take into account; however, they may also receive funds from their membership, private donors, investment revenues and bank loans, all of which may be hard to quantify or even estimate. Consequently, the issuance of legal provisions is a first step towards transparency and likely cost-reduction of campaign expenses. Recent research efforts in this domain have been undertaken by UNDP, International IDEA, the Federal Electoral Institute of Mexico and IFES.
9. In most well-established democracies and in an increasing number of the new ones, the electoral budget is prepared by the electoral authority and processed through the finance ministry for approval in parliament. The ministry does not have the authority, at least formally, to curtail or amend an electoral budget prepared by electoral authorities. Nevertheless, the government and the legislature may place constant pressure on electoral authorities—along with other publicly funded agencies—to continually review their operational performance and to contain costs. In fact, constant concern over the budget by governments and legislatures should be considered in itself as cost effective and as a tool for good management because it encourages strategic and operational planning by the electoral authority.

Contributors to Focus on Cost of Registration and Elections

The material in this ACE Focus On comes from the Cost of Registration and Elections (CORE) Project initiated by the United Nations Development Programme (UNDP) and the Center for Transitional and Post-Conflict Governance at IFES; written and produced by Jeff Fischer and Dr. Rafael López-Pintor.  The primary goal of CORE is to assist governments and independent organizations in their efforts to identify and examine all forms of election-related costs and funding sources. 

The Center is grateful for the support provided by Linda Maguire of the UNDP and would also like to acknowledge research and editing assistance provided by members of the Center’s staff, including Will Covey, Jamie Crowley, Joanne Faulkner, Maya Serban and Rakesh Sharma. 

To download a PDF copy of this publication in English, please click here 

Other "Focus On..." series

This section of the Cost of Registration and Elections (CORE) Project contains the main findings and conclusions drawn from case studies on electoral costs in 10 countries.  The countries were selected as illustrative of election management within different democracy environments—from stable to transitional to post-conflict, because they were also examined in a previous UNDP study published in 2000 and other countries, meanwhile, were added based on geographic location and size of the electorate. 

Afghanistan - Conflict environment with large- and medium-sized electorates

Australia - Stable democracy with large- and medium-sized electorates

Cambodia - Conflict environment with small-sized electorate

Guatemala - Transitional democracy with small-sized electorate

Haiti - Conflict environment with small-sized electorate

India - Stable democracy with large- and medium-sized electorates

Iraq - Conflict environment with large- and medium-sized electorates

Mexico - Transitional democracy with large- and medium-sized electorates

Spain - Stable democracy with large- and medium-sized electorates

Sweden - Stable democracy with small-sized electorate

Introduction

This case study concerns the first post-war elections in Iraq, held on 30 January 2005. The figures in this paper are based on budget estimates and on partial expenditure reports collected during the implementation of the 30 January 2005 elections. As of the time of the final revisions to the study (March 2006), final expenditure figures are not yet available, both because the IECI has remained busy with a referendum and general election in 2005 and so has not completed its 2005 audits and because so many agencies contributed to the 2005 costs. The budget estimates are, however, reasonably accurate estimates of the incurred costs.

I. Background on the electoral process 

The Law of Administration for the State of Iraq for the Transitional Period (Transitional Administrative Law, TAL) called for five elections during the transitional period (defined as beginning with the formation of a fully sovereign Iraqi Interim Government on 28 June 2004 and ending with the formation of an elected government under the terms of a permanent constitution, as of the final revisions to this paper, the assembly elected on 15 December 2005 has not yet formed a government). All of the required elections were held according to the schedule set in the TAL, which was in itself a notable achievement: 

• Elections to a Transitional National Assembly, Kurdistan National Assembly were held on 30 January 2005;
• A referendum on a constitution drafted by the Transitional National Assembly was held on 15 October 2005; and
• A general election under the new constitution was held on 15 December 2005. 

Through a consultative process facilitated by the United Nations, Iraqi political leaders decided to empower an independent electoral management body to conduct the elections during the transitional period. The Independent Electoral Commission of Iraq (IECI) was created by Coalition Provisional Authority (CPA) Order no. 92, which describes the IECI as the “exclusive electoral authority throughout Iraq during the Transitional Period.

II. Structure of the EMB

The IECI is an “independent and autonomous, non-partisan, neutral and professional government office” empowered “to organize, oversee, conduct, and implement all elections set forth in the TAL.” (CPA 92) The IECI has two major components: the Board of Commissioners regulates the electoral process and has ultimate responsibility for the planning and implementation of the elections; the Electoral Administration, headed by the Chief Electoral Officer (also a non-voting member of the Board), is responsible for carrying out the electoral process. The members of the Board of Commissioners were selected in June 2004 after an extensive process of public nominations; a panel of respected international election officials appointed by the United Nations (UN) proposed a slate of commissioners to the then-Governing Council, which accepted the panel’s recommendations.

The Electoral Administration is divided into a National Electoral Office, 18 Governorate Electoral Offices, and 542 District Electoral Offices/Voter Registration Centres. In addition, there is a Kurdistan Regional Office and two sub-governorate offices in Baghdad, the largest governorate. Each District Electoral Office is responsible for, on average, 10 polling centres with up to 3,000 voters each.

The national office has divisions for Finance, Administration, Public Outreach (including services for electoral contestants and observers), Capacity Building, and Operations. In addition, a small planning and reporting cell assists the Chief Electoral Officer. After the first elections on 30 January 2005, legal and audit units were established in the national office reporting directly to the Board of Commissioners. 

The national office has a staff of approximately 220 persons, the governorates a staff of 22 each, and the districts a staff of 10 each; most of these positions have been filled. In the governorates of al-Anbar and Ninewah, the security situation has precluded the regular operation of the District Electoral Offices/Voter Registration Centres.

III. Legal framework

CPA Order no. 92 grants the IECI broad authority over its finances: 

The Iraqi government shall ensure that the Commission receives all of the resources necessary for administering elections throughout the Transitional Period.  Notwithstanding the foregoing, the Commission may seek to enlist the international community for appropriate assistance in this regard, including the direct provision of additional funds or resources.  All Commission resources shall be solely managed and obligated by the Commission.  The CEO shall be responsible for the accounting of these resources.  

The IECI presented a budget for the 30 January 2005 elections to the Interim Government’s Ministry of Finance in September 2004. While the budget did include some approximate costs for FY2005 to aid the government’s financial planning, these figures were of course very general given the uncertainty surrounding the election calendar for 2005. The 30 January 2005 budget also contained several important unknowns which, in the end, proved to be substantial cost. One of these was secure logistics – the delivery and retrieval of 3.3 million kilograms of election material to more than 5,000 polling centres in the midst of severe ongoing conflict. These costs were eventually borne by UNOPS. In addition, the original budget did not provide for out-of-country voting. The IECI decided to conduct out-of-country voting in October, and contracted the International Organization for Migration (IOM) as its implementing partner. The costs for this operation (US$92 million, of which some portion was unspent and returned) were provided by the Government of Iraq. 

The IECI operates under the procurement, human resources, and accounting rules of the Government of Iraq. 

IV. Electoral costs

a) Establishment of the IECI 

As a new institution, the IECI required facilities, office furniture and equipment, communications and information technology, and other capital investments. These costs amounted to approximately $28 million, of which approximately $12.5 million was provided by the United Nations for office furniture and equipment and communications and information technology and $3.5 million was provided from Iraqi budget funds by the CPA for the renovation of a headquarters facility. 

b) Voter registration 

The voter register for the 30 January 2005 elections was based on the existing Public Distribution System (PDS) database, used by the Ministry of Trade to manage a monthly rations program. To be used as a voter register, the database required a substantial period for corrections and additions. The costs of this period amount to approximately $83 million. These costs were mainly for staff, printing, data entry, and secure logistics. The bulk of these costs was borne by the IECI through Iraqi budget funds, but the UN funded the data entry project, and the United States Agency for International Development (USAID) funded printing of registration forms, procurement of other registration material, and public information. 

c) Polling 

Polling costs were approximately $180 million. The largest line items were for polling staff, material procurement (including polling kits, ballots, and voters lists), and secure logistics. Again, the majority of costs were funded by the Iraqi budget, but the UN funded a large section of the secure logistics costs (approximately $50 million) by contracting with a global logistics firm, which in turn, sub-contracted with security firms working in Iraq. USAID made several donations in kind, as did the Governments of Denmark and the Peoples’ Republic of China. Other support has been provided by the Italian and Japanese governments for training of the election staff. 

d) External voting 

The IECI concluded a Memorandum of Understanding with the International Organization for Migration (IOM) for the implementation of an external registration and voting program in 14 countries. This program was implemented on very rigorous timelines, and the Memorandum allowed for costs as high as $92 million. This total was reached by estimating up to $50 per potential voter, based on recent external voting projects. In addition, $3 million was budgeted for each country in security costs based on IECI experience with an external data entry centre. IOM returned $20 to the IECI, so the final cost of the external registration and voting operation was $72. 

e) International assistance

The IECI received international technical assistance under the leadership of the UN. In addition to the UN, IFES, the UK Department for International Development (DFID), and the European Commission provided staff to the international assistance team.

f) Security costs

The security environment in Iraq imposed a number of security costs. These included personal protection for the Board of Commissioners and international technical advisors, but also markedly higher than normal costs for the storage and transportation of election materials, which required a reliance on air movements and private security contractors. In addition, a number of projects, including data entry, took place abroad and at higher cost due to domestic security concerns. 

g) Support from the Iraqi government

In addition to direct budget support, the Iraqi government provided office and warehouse space, especially through the Ministry of Trade, and support from the security services. The Ministry of Education provided most of the polling locations.

About the author

Jarrett Blanc has been the Chief-of-Party for IFES’ Center for Transitional and Post-Conflict Governance in Iraq and in the Palestinian Authority.  He advised the CPA and the Iraqi Governing Council on the establishment of the Independent Electoral Commission of Iraq and on the establishment of the electoral calendar and legal framework. He was a senior advisor to the IECI in the planning and implementation of the 30 January 2005 elections. In the Palestinian Authority, he advised the Central Elections Commission, Palestinian Legislative Council, and diplomatic community and coordinated IFES’ technical assistance program. Blanc also has field experience in Afghanistan, Kosovo, Lebanon, and Nepal.

Acknowledgement

This case study is based on information from various publications and reports of international bodies, including the Organization of American States (OAS), UNDP and the National Democratic Institute for International Affairs (NDI), on the May 2000 elections in Haiti. Additional information was obtained through personal interviews with members of the Provisional Electoral Council, (Conseil Electoral Provisoire). The author would like to extend special recognition to the Chief of the Electoral Mission of the United Nations Stabilization Mission in Haiti (MINUSTAH) and his staff, who provided the most accurate financial information for the elections to be held in 2005.

This study was revised in March 2006 by Pierre-André Guillaume, electoral consultant of UNDP for MINUSTAH /Electoral Assistance Section.

STRUCTURE OF THE EMB

The Conseil Electoral Permanent, Haiti’s permanent electoral authority, was established in the 1987 Constitution (articles 191 to 199). It was conceived as an independent council tasked with developing a voter register, organizing elections and serving as the tribunal for electoral disputes. However, although it was created by law, this permanent council has not been formed yet, because the Constitution calls for a full election cycle as prerequisite for establishing the council—a situation that has not occurred since the creation legal creation of the council. Therefore, since 1988 all elections have been organized by a provisional electoral authority (Conseil Electoral Provisoire, to be referred to henceforth in this section by the acronym ‘CEP’).

Since 30 April 2004, the CEP has been composed of a board of nine members: one appointed by the private sector; three by the Catholic, Episcopal and Lutheran churches; one by human rights organizations; one by the Supreme Court; and three by political parties.

The composition of the board was agreed following a political settlement in June 2002 and incorporated into the Resolution 822 of the Permanent Council of the Organization of American States (OAS) in September 2002. This board began to function after taking oath at the Supreme Court in May 2004,  and  it will continue functioning until the permanent electoral authority is set up under the constitutional procedures scheduled to begin after the newly elected president of Haiti takes office (article 134:1 Constitution).

After the resignation in December 2004 of the president of the CEP, the council continued working under the leadership of an interim president who will be confirmed as president of the council in August 2005. A Comité d’Appui to help the executive functions of the CEP was appointed as indicated in internal by-laws.

STRUCTURE OF THE CEP OF 2005

The Provisional Electoral Council (CEP) is a hierarchical structure, with the nine board members acting as the highest level. The CEP has an electoral structure composed of departmental electoral offices at the provincial level (the Bureaux Electoraux Departementaux - BED); communal electoral offices at the district level (the Bureaux Electoraux Communaux -BEC); and voting centres (CV), formed of voting bureaus (BV) at the polling station level on Election Day.

The CEP has 11 department electoral offices (BEDs), which follow the geographical division of Haiti in departments: Ouest, Sud-Est, Nord, Nord-Est, Artibonite, Centre, Sud, Grande-Anse, Nord-Ouest and Nippes. Each department has one BED, with the only exception of the department of Ouest—the largest one— which has two: BED Ouest I and BED Ouest II.

In the departments, the BEDs are divided into Bureaux Electoraux Communaux (BECs). The number of BECs within each department varies by the number of the communes. Although Haiti has 140 communes, the CEP has 142 BECs, because Port-au-Prince, the capital, has three BECs.

Each BEC has several voting centres (CV). For the elections of 2006, the CEP will establish 803 voting centres around the country, which in turn will have several voting bureaus (BVs). For the 2006 elections, the CEP has planned to establish 9,214 BVs throughout the country.

THE BEDs and BECs

The role of the BEDs and the BECs, is to organize, manage and monitor the electoral process and to solve procedural challenges presented by political parties in a timely manner in the administrative territories under their jurisdiction. Each office is headed by a president, vice president and secretary appointed by the CEP. These three officials are assisted by two legal advisors charged with resolving electoral disputes at the local level. The departmental electoral offices are present in the following cities: North (Cap Haitien), Nord-West (Port-de-Paix), Nord-East (Fort Liberte), Artibonite (Gonaives), Grande-Anse (Jeremie), Nippes (Miragoane) and South-East (Jacmel), Plateau Central (Hinche), South (Les Cayes), West I and West II (Port-au-Prince).

The role of the voting bureaus is to manage election day operations at the polling stations during the 2006 elections.

THE VOTING CENTERS

Voting centres are accessible to the people living in the area and are designed to be easily accessed. They are situated by preference in schools, government buildings or other public facilities. Each voting centre has between 1 and 50 voting bureaus. Each voting bureau can receive a maximum 400 voters. The geographical identification of the voting centres and their physical description, the distance and travelling time from the BEC, accessibility, risk, etc., was conducted by MINUSTAH. All voting centres were approved by the CEP. A total of 9,214 voting centres have been established for 2006 elections.

THE NEW STRUCTURE OF CEP

In October 2005, the CEP will be transformed into a Conseil d’administration and will have a new mission, namely: to plan, organize and realize elections and take strategic decisions. In the meantime, a new Director General is in charge of all executive and operational aspects of the process. The general direction is divided into other components: electoral operations, administrative, legal, communication, security, etc.

THE REGISTRATION PROCESS

The voter registration operation for 2005 will be conducted with technical assistance from the Organization of American States (OAS). Of a total electoral budget of nearly $49 million, some $9 million will be allocated to the task of voter registration, which ideally will include the possibility of creating the basis for a civil register, still non-existent in the country. The overall voter registration process will be automated, instead of the current manual registration. Registration period, originally scheduled to take place from 9 August to 25 April, has been extended until 9 October 2005 in order to reach all the country. Presidential and legislative elections will take place on 7 February 2006 for the first round and on 21 April 2006 for the second round. Local elections are scheduled to take place on 18 June 2006.

All voters will receive a digital card that will gradually replace the national identification card. The OAS has established 615 offices for registration activities (400 registration centres, 185 automated registration offices, and 30 automated registration mobile units). The registration centres have been located in rural areas; the automated offices in urban areas; and the mobile units in the semi-urban areas. The number of automated centres will be adjusted as required to reach a larger number of potential voters. The registration process will take six months (April-October 2005).

As part of the operations, 3,200 polling centres will be established.. All automated registration offices will also be considered as polling centres. Voters will be advised about their voting locations during the registration process. The CEP and the international donors have approved this project to be implemented by the OAS.  Some 3,535,025 identification cards are expected to be distributed by OAS to voters around all the country.

ELECTIONS CONSTITUTIONAL FRAMEWORK

From 1801 to 1987, Haiti had a total of three Fundamental Acts, 22 Constitutions (two of which were imperial), 14 amendments, and one accord. Regarding electoral provisions, there have been 25 electoral acts.  The current electoral law, the 3 July 2005 Electoral Act, is regulated by the 29 March 1987 Constitution as well as by the Electoral Acts of August 1990 and February 1995. There is also the Political Party Functional Law, dated 31 July 1986.

The president is elected for a five-year term (consecutive terms are forbidden, although one can hold the office a second time after a five-year hiatus). Since the reintroduction of a bicameral Parliament in 1987, deputies are elected for a four-year term and senators for a six-year term. One third of Senate seats are up for election every two years.

Under regulations for the 2006 elections, which will fill every elected seat in the country, the Senate will be composed as follow: winners by majority in the first round of the election will be elected for a six-year term. The second round will oppose the candidates who have the same quantity of votes for a four-year or a two-year term (articles 76 to 78 of the electoral decree). The main elements of the electoral system are presidential elections (articles 86 to 90), elections to Senate (articles 80 to 85), elections for Deputies (articles 74 to 79) and local elections (articles 91 to 111).

Winners at local elections are elected for a four-year term (articles 98, 102,108). Local elections will take place after the second round. They are very important, because they will permit the establishment of the Conseil Electoral Permanent called for by the current Constitution of 1987.

PRESIDENTIAL ELECTIONS

After the introduction of the direct presidential election, a plurality system was established for the 1950 and 1957 elections. However, only one candidate campaigned in the 1950 election (Paul Eugene Magloire); in 1957 there were two candidates (Francois Duvalier and Louis Déjoie), so the system of decision was actually the majority system. No presidential elections were held during the two Duvalier governments (1957-1986). After the fall of the Duvaliers, there have been four presidential elections: 1988 (Leslie Manigat elected), 1990 (Jean-Bertrand Aristide elected), 1995 (Rene Preval elected), and 2000 (Aristide elected again). The president is elected via absolute majority. If no candidate achieves this majority in the first electoral round, a second round is held between the two candidates with the highest number of votes.

ELECTIONS TO THE SENATE

• The country is divided in 10 departments, 99 circumscriptions, 140 communes and 570 rural sections.
• Each of the 10 departments constitutes a three-member constituency. However, since one third of the Senate will be renewed at each election in the future, the country will be divided into single-member constituencies in each election.
• All 30 Senators are elected via a majority system.

ELECTIONS TO THE CHAMBER OF DEPUTIES

• The country has been divided according to population density into 99 single-member constituencies (electoral decree of 3 February 2005).
• The 99 deputies, one by circumscription, are elected through a majority system during the first round.
• If any candidate wins, a second round is necessary.
• Seats are allocated by a first-past-the-post system, in which the candidate who has secured the most votes (not necessarily an absolute majority of the votes) is the winner.

LOCAL ELECTIONS

• Local elections will take place in 140 cities and 570 rural sections on 18 June 2006. Seats are distributed on a basis of who has more votes.

AMENDEMENTS TO THE ELECTORAL SYSTEM

The new electoral decree of February 2005 was modified many times. Appeal was made at the Cour de Cassation of the contentious decisions: requirement for candidates to the Senate, rules of majority for president, requirement process for financing parties and candidates.

Various decrees are expected to be enacted by the government in the second half of 2005 about various aspects of the electoral process like the electoral lists, financing of political parties, nationality of candidates, and identification card.

ELECTORAL ASSISTANCE SECTION SUPPORT

The Chief Electoral Officer (CEO) joined MINUSTAH on 18 August 2004. This was the official starting point of Electoral Assistance Section (EAS) within MINUSTAH. The mission carried out many important tasks. It assisted the CEP in revising the electoral law by preparing working papers on the main issues involved in the drafting of the new legislation and internal by-laws. It also provided support to the CEP in formulating a chronogram of activities with sets of options. The main areas of support are: training sessions for members of CEP, political parties, civil society, voter registration plan proposal, assisting CEP staff, registration process, communication systems, budget management, establishing civic education programs, tabulation centre, electoral calendar, logistics plan, security system, procurement guidelines, procedures for the elections, establishing voting centres, legal aspects, and regional coordination.

One of the critical phases of electoral processes are the procedures around voting and counting, complaints and appeals processing, tabulation and announcement of the final results.

The greatest challenges were for the EAS to plan, work with a CEP, sometimes divided, execute complex electoral operations, solve budget problems, and participate in creating good environment to attain the objective of successful, peaceful, nationally and internationally accepted elections.

OAS ELECTORAL TECHNICAL ASSISTANCE

The Organization of American States (OAS) voters plan provides for the establishment of approximately 615 offices for registration activities (400 registration centres, 185 automated registration mobile units). The registration will be located in the rural areas, the fixed automated in the urban area and the mobile in the semi-urban areas.

The overall voters’ registration will be automated and voters will all receive a digital card (Carte d’identification nationale) that will gradually replace the “numéro d’identification fiscal” (NIF).

OAS provided support to the CEP in creating a reliable electoral registry for the elections, issuing a secure and adequate identification card to the voters, providing the operational framework for a plan of strengthening and modernizing the civil registry and contributing to the training of electoral technicians. A high-tech registration system has been conceived.  The electoral partial list (LEP) and electoral general list (LEG) have been prepared and are available at the time of the writing of this document. The registration system is fundamental for the success of the vote and the future of a strong civil register.  

STRUCTURE AND CONCEPT OF OPERATIONS

The concept of operations provides for establishment and operations of the BEDs and BECs, elaborate technical strategy for registration, staffing BEDs, logistical support, purchase and deployment of transport vehicles, communication facilities, security and coordination between the MINUSTAH Electoral Unit and other institutions.

MINUSTAH UNVs

Many UNVs participate in supervision of electoral process in assigned regions. They assist CEP technicians for registration of voters, identify sites for registration and voting centres, plan training sessions, and provide logistical support to BEDs and BECs. Each BED is headed by a District Coordinator who reports directly to the Chief Electoral Officer. The district coordinator supervises the electoral process down to the communal levels.

CIVIC EDUCATION

Only $1,263,167 has been provided for civic education, despite the importance of this component of the electoral process. Under this heading comes the cost of consultants that the CEP plans to hire to conduct civic education programs over 13 months. The consultants are expected to conduct civic education programs prior to the registration process and through election day. Additional funds have been requested from donors for this line item of the budget. An intensive campaign is expected during the registration process, followed by a massive campaign to promote voter participation after political parties and candidates are registered. So far, there is nothing planned to support political party activities by either civic education NGOs or domestic monitoring groups.

As the voter registration process, public information on the electoral decree and the national identification card began in April 2005. Approximately 200,000 radio spots and 1,500 television spots have been launched. Civic education materials have been produced and distributed nationwide (500,000 posters, 1,000,000 pamphlets and 1,000 banners).

In June 2005, two important national registration days were conducted: one for women and another for civil society. It should be noted that women represent 52% of the Haitian population and this national day has significantly boosted the registration process.

TRAINING SESSIONS

In August 2005, a training of trainers’ course will be conducted for 60 participants, who then in turn will have to train 550 civic education agents who are expected to then train up 3,500 civic education promoters from departmental level down to the communes.

According to the year 2003 census, 4.2 million Haitian citizens will have the right to vote in the 2006 elections. The municipal elections are expected to be held in June 2006, and the first round of the legislative and presidential elections scheduled to take place in February 2006. The second round for these elections will follow on 21 April 2006. In compliance with the Constitution, an elected legislative power will be installed in May after the second round, and the president will be sworn-in in May 2006.

The total resource requirement for the CEP, the provisional electoral council, for the period from 1 October 2004 to 31 December 2005 is planned to amount nearly $49 million, covering personnel and operational costs. The Haitian government will cover the regular costs of the CEP’s approximately 82 headquarters staff. It will also contribute $2.9 million for the extraordinary electoral budget.  Three main international donors (Canada, the European Union, and the United States) have committed to provide $41 million. The financial resources for operational items have been linked to quantitative outputs.

This budget covers all elections (at least two, but possibly three) in 2006. Since the CEP is not a permanent institution and there are no other activities between election years such as updating the voting list or issuing IDs as do most permanent EMBs, there is no ordinary budget.

The substantial financial contribution of the international community, as described below, underpins the entire electoral operation in Haiti. Because of financial constraints, the national contribution is less than 10 percent of the total budget. Table 1 shows the structure of the required funds.

Table 1. Budget projection from electoral division of MINUSTAH

Note:  This budget is a work-in-progress and may very well change as donor negotiations precede.

Some of the main issues and trends regarding election cost structures are detailed below.

a) Transportation

Given the precarious conditions for communications in Haiti, transport represents a major issue. Thus, the CEP will purchase approximately 200 vehicles and have three block coordinators based in Port-au-Prince to provide direct support to the electoral departmental offices in their area of responsibility. The three blocks coordinators are going to be divided in the following way: 1) South, comprising Jacmel, Les Cayes, Jeremie and Miragoane; 2) North, comprising Artibonite and Plateau Central; and 3) West, comprising Port-au-Prince. Each block coordinator will have one vehicle. The departments will be provided with three vehicles each while the communal office will receive one vehicle each. This will bring the total fleet to 191 vehicles; additional vehicles for transport of electoral equipment from seaports to storage facilities and communal sections can be rented locally. There are also ground fuel station companies that can be contracted locally. Air transport is also available in almost every commune for landing helicopters. Several seaports are available and provide facilities for unloading equipment, but storage facilities are rare. In most departments and communes, telephone facilities exist, both fixed lines and mobile services (except in Jeremie, which has no cellular phone service). There are community radio services and TV channels. Radio communication services will be placed in each department and commune to facilitate direct communication with Port-au-Prince and other departments.  

b) Staffing and logistics

Local staff and logistical facilities will be provided. Office space will be provided by the government where possible, otherwise offices will be rented when it is not possible to secure government-owned premises. There will be requirements for the refurbishment of all office spaces and small construction services to ensure that the electoral infrastructures are viable. For local and temporary personnel, Table 2 outlines what is planned for electoral staff costs.

Table 2. Planning for electoral staff costs

* To be funded by Haitian government

National staff salaries are estimated to cost $4,027,939. Cost estimates are based on the phased recruitment and deployment of 165 staff for the departmental electoral offices and 1,395 staff for the communal electoral offices to be hired for 15 months. The cost estimate for salaries is based on $3,526,250 for departmental electoral offices and $501,689 for communal electoral offices, rates that are derived from the national salary scale rates for public servants. CEP central headquarters operating costs for approximately 82 staff will be directly funded by the government. 

Voter registration officers’ salaries are estimated to cost $2,017,167. According to the electoral law of 1999, each voter registration office should comprise a staff of five: a president, vice president, secretary, and two voter registration officials. The monthly salary of voter registration officers is approximately $111, while voter registration officials receive $97 for a two-month period. A total of $10,080,000 is provided for salaries of 56,000 polling officers for each of the three types of elections: municipal, legislative and presidential (first and second round). The cost estimate for polling officers’ salaries is based on $60 per officer for 56,000 officers covering approximately 14,000 polling stations (300 voters per station on average). In accordance with Haiti’s electoral law, polling stations will be open from 6:00 a.m. to 4:00 p.m.

$501,689 for communal electoral offices, rates that are derived from the national salary scale rates for public servants. CEP central headquarters operating costs for approximately 82 staff will be directly funded by the government. 

Voter registration officers’ salaries are estimated to cost $2,017,167. According to the electoral law of 1999, each voter registration office should comprise a staff of five: a president, vice president, secretary, and two voter registration officials. The monthly salary of voter registration officers is approximately $111, while voter registration officials receive $97 for a two-month period. A total of $10,080,000 is provided for salaries of 56,000 polling officers for each of the three types of elections: municipal, legislative and presidential (first and second round). The cost estimate for polling officers’ salaries is based on $60 per officer for 56,000 officers covering approximately 14,000 polling stations (300 voters per station on average). In accordance with Haiti’s electoral law, polling stations will be open from 6:00 a.m. to 4:00 p.m.

c) International financial assistance

UN Security Council resolution 1542 (2004) gave MINUSTAH the mandate "To assist the Transitional Government of Haiti in its efforts to organize, monitor and carry out free and fair municipal, parliamentary and presidential elections…through the provision of technical, logistical and administrative assistance…"

OAS General Assembly resolution 2058 of June 8, 2004 calls on the OAS special Mission for Strengthening Democracy in Haiti "To assist the country’s Provisional Electoral Council in preparing, organizing and overseeing the elections and the proclamation of the results, in cooperation with MINUSTAH."

Pursuant to these provisions, the OAS and MINUSTAH approved a Memorandum of Understanding to foster cooperation and outline and assign tasks to coordinate their efforts to achieve their stated objectives. The OAS is to be responsible for the registration of voters in the electoral process, which will create a civil register system, while MINUSTAH electoral mission will provide technical assistance to the CEP over the entire electoral process in all areas.

Of the $41 million committed by donors, $9 million has been allocated to the OAS through UNDP. The following countries and regional organizations have pledged or are considering technical assistance: Brazil, Chile, Colombia, France, Japan, Mexico, Spain, Venezuela, the Caribbean Community (CARICOM), and the Organization of French Speaking Countries (OIF). The international community has pledged additional funding for civic education and election observation programs, including political party activities. MINUSTAH is also preparing to establish a Trust Fund aimed at mobilizing additional resources for the elections as may be required.

MINUSTAH is assisting CEP to build internal capacity and to make efficient use of international resources. At the same time, several key organizational issues that may affect CEP staffing, including inter alia, operational planning, technical assistance and logistical support, are currently being discussed at the CEP. This extraordinary budget provides for hiring and training of 57,642 nationals: 165 departmental electoral office staff, 1,395 communal electoral office staff and 56,000 polling station officers. The total resource requirements for 2004–2005 elections have been linked to CEP objectives through a result-based framework.

About the author

Dr. Félix Ulloa is NDI's senior resident director in Haiti, managing the Institute's programs there since 2000. Prior to joining NDI, Dr. Ulloa served as magistrate of the Supreme Electoral Tribunal in his native El Salvador from 1994 to 1999. He has taught at several universities of El Salvador, and has been invited to the Language School of Middlebury College in the United States. He also served as president of the Institute of Law of El Salvador, and has participated in elections programs for the United Nations, OAS, the International Foundation for Election Systems (IFES), the International Institute for Democracy and Electoral Assistance (IDEA), the Center for Elections Assistance (CAPEL), and NDI, providing oversight for electoral bodies, political parties, and civic organizations.

Dr. Ulloa has published books, essays and articles in several countries, on topics such as democracy, elections and political parties. He earned a law degree from la Universidad Complutense, Madrid, and from la Universidad de El Salvador. He has also completed other post-graduate studies at the Institut International d'Administration Publique in Paris and at the Hubert H. Humphrey Institute of Public Affairs at the University of Minnesota in the United States.

A standard case study methodology was followed. A panel of authors and election specialists was selected on-site¹. Panel members were guided by a checklist template as well as documentary sources and individual informants for consultation. The case studies were researched in the field and written between September 2003 and January 2005. The case study reports generally include an outline of the current structure of each nation’s EMB, including appointment procedures, terms of office and scope of responsibility, followed by a description of the legal framework for elections—primarily concerning budgeting and funding provisions (i.e., reference to voter registration, polling operations, campaign expenses by political parties and candidates and financing of media access). The reports assess how the legal provisions regarding election funding are actually followed in practice. They also consider the main obstacles— political, financial, administrative or technical—to the application of the law in the proper management of the electoral budget.

The case studies describe the electoral budgets and procedures of EMBs in both non-election and election years. Voter registration costs, especially new registration operations in conflict environments, are considered separately. The cost of polling operations in the most recent general election is scrutinized by disaggregating budget figures by main items and comparing them to previous elections to discover the main reasons for change, if any. The cost of external voting operations is analyzed when it can be separated from the aggregate electoral budget. International funding of electoral budgets is categorized by area of application, i.e. whether awarded to the EMB or to other election-related actors, such as domestic and international observer missions. A number of cutting-edge issues are examined whenever information is available. Among the questions directly addressed are the following:

• Are costs related to political party finance for general operations and campaign funding part of the electoral budget, or considered separately?
• What are the costs of planning and introducing new technologies? And are they envisioned as cost-reducing strategies?
• Is civic education a cost in the budget of the EMB, a cost to political parties or cost to NGOs or other civil society entities? Does the EMB share civic education costs with other national or international actors?

Finally, the case studies offer an overall assessment of current and most recent practices in election budgeting, funding and cost handling, by describing election items that have proved either more susceptible or more resilient to cost-saving measures. Documentary resources most often consulted in the preparation of the case studies included constitutions, electoral laws, laws on political parties, organization charts of the electoral authorities, election bylaws and electoral budget documents. Professional literature and Web sites were also consulted. And finally, informal and semi-structured personal interviews were conducted with electoral authorities and specially qualified informants, academics and practitioners.

Note: 

¹The authors are academics and practitioners with long experience in the field, most of them related to electoral authorities as senior staff or international consultants.

From December 19, 2003 to October 1, 2004, the organizers of the Cost of Registration and Elections (CORE) Project distributed a survey research questionnaire to election management bodies (EMBs) in 178 countries. This Survey Results Report (PDF file) has been written based on these preliminary survey responses.

Cost of Registration and Elections

full text on one page

Cost of Registration and Elections

Working Definitions of Election Costs in a Core-Costing Model

Types of Electoral Costs

Developing a Methodology for Electoral Cost Assessment

Election Budgets and Revenue Sources

Election Funding: Legal Framework and Practice; Election Budget Processes and Cycles

Resource-Sharing Arrangements with Other Public Agencies

How Much Do Elections Cost?

Electoral Budget Histories in the 1990s

Public Funding of Political Parties

Electoral Assistance from the International Community

Voter Registration Costs

General Considerations on the Purpose of Voter Lists in Comparison with Other Registries of People

Voter Registration Systems in a Global Perspective

Quality Standards of Voter Lists

Managing the Cost of Voter/Civil Registries: Country Experiences

Electoral Cost Management Practices

Electoral Procurement Procedures

Inventory Control of Unused Electoral Supplies

Cash Transactions for Electoral-Related Goods and Services

Electoral Cost Overruns

Electoral Budget Audit

Investing in High-technology Electoral Solutions

Cost-Effectiveness in Elections

Annexes

Contributors

Case Studies

Iraq

Haiti

Methodology

Survey of Election Costs in 178 Countires