The cost of elections is a concern to governments and election management bodies alike, regardless of the country's level of development. A prohibitively high cost electoral process that the country cannot afford can jeopardize the entire democratic process.

Traditionally, to minimize the cost of an election, the results management system, processes and procedures chosen should make use of the existing infrastructure, such as transport and communication networks, as much as practicable, and take into account literacy levels, infrastructure challenges, and the sources for the recruitment of officials. As of late, EMBs have increasingly learned that introducing ICT-solutions can expedite results reporting, but depending on IT-solutions chosen, that could also significantly increase costs. Sustainability over more than one electoral cycles of such an investment sometimes has become paramount for the actual decision to upgrade a results system, as much as transparency, speed and reduced fraud risks in the actual counting and tabulation processes.

What is affordable, expensive or feasible can vary greatly between countries. Here are some generic guidelines regarding costs:

• the simpler the system, the lower the vote counting costs;
• the greater the use of existing infrastructure, the lower the cost;
• the more permanence and continuity between elections, the higher the savings in materials and training costs;
• the lower the political stability, the higher the cost in terms of added security measures and the more transparent the processes and EMB must be.
• the higher the risks of fraud and electoral malpractices are the more efforts the EMB must put in place to effectively deter, detect and mitigate such vulnerabilities. This can require both additional financial assets and time. 

Alternative technology-focussed questions include:

• Is it difficult to recruit qualified voting station and/or counting staff?
• Is it difficult to recruit qualified IT support personnel, particularly in rural and remote areas, where any electronic solutions will be deployed?
• Have there been problems with irregular vote counts? 
• Is there a culture of trust of high technology used by the EMB and will the extension of its use to electronic voting benefit or suffer from this?
• Is there a need to reduce the number of election workers?
• Will the reduction in election workers be offset by the increased cost of recruiting and deploying highly-skilled ICT workers, or the outsourced cost if these functions are fulfilled by a contractor?
• Is the ballot becoming more complex?
• Will the voter education function be able to ensure that voters can cope with and properly use any technology being considered?
• Is the vote count coming in too slowly? 
• Is the telecommunications infrastructure adequate?
• Do voters have to wait too long to cast their ballot?
• Will the throughput in polling stations be increased by the technology being proposed? 

Hidden Costs Frequently Overlooked or Underestimated in Elections Technology Procurement

• Vendor Lock-in.
• Obsolete technology - absent the capacity to implement minor updates, entire systems can be abandoned.
• Storage – New technology (computers or similar machines) may need different storage condition than manual voting materials (air conditioned, for example). Since manual-voting materials should also be kept, switching methods may require more storage space for EMBs.[1]
• Excess use of proprietary products - rather than standards-based, modular approach. This is a separate issue from the open source (code) software issue and relates primarily to hardware.
• Depreciation - more rapid in harsh environments.
• Inventory wastage due to high rates of theft or pilferage
• Sustainability does not simply happen - it must be set as a requirement of the procured solution and, like energy conservation, you will have to spend money to save money - higher specification, hardware that offers longer duty- and life-cycles. 
• Procurement Integrity Costs - fees paid where donors require procurement by international or other organisations in order to overcome real or perceived deficiencies with national procurement mechanisms.
• Contingency - for example, air, rather than sea freight because of inadequate time available for procurement.
• Inadequate testing is dangerous. But effective testing will inevitably results in changes to hardware, software and procedures. These have cost implications and are often overlooked.
• Third party certification of software or systems, sometimes necessary to create, sustain or rebuild stakeholder confidence in high technology is expensive and time-consuming. 

Do not reinvent the wheel

While Results Management Systems cannot reasonably be considered COTS (Commercial-Off-The-Shelf) technologies, by and large, the hardware and software components of such systems are typically familiar, mature and cheaply available. Laptop computers, their smaller and more mobile replacements, software such as operating systems, database management systems, data capture, services such as mobile data communications - these are all widely available even in developing countries. However, the expertise required to integrate these familiar components into effective results management systems is far from commonplace and comes at a premium cost. Even the expertise necessary to properly analyse the electoral voting and counting processes of the EMB and turn this into sufficiently detailed specifications. 
 

Specific Needs for Vote Counting and Aggregation of Results

The local environment and political situation may add to specific needs for vote counting at voting stations, or at central locations. 
Such local factors might include the need for additional security measures during counting, transportation requirements, or specific communications technology that is not readily available.

Higher than expected turnout may affect the counting duration, and additional staff may be required to cope with the additional ballots to be counted. Expected budgets may be exceeded if such contingencies have not been planned for. Worst (or best) case scenarios (for example, expecting 100 percent of registered voters to turn out to vote) are frequently used when election budgets are planned in order to build in sufficient capacity.

The selection of a specific counting, tabulation and communication technologies may also increase costs or reduce them, depending on the choices made. However, if not planned properly and when sustainability and local ownership are considered, the cost saving argument used to justify an ICT-upgrade could prove faulty. Before properly tested, the old manual system often needs to be in place as a fall-back and thereby add additional transitional costs.

Transmitting results by telephone is often cost-effective, depending on distance and availability, but could negatively affect the accuracy given increased error rate. As a result, EMBs are beginning to introduce various SMS solutions to convey interim results from voting stations. SMS is a relatively inexpensive communications solution, both in terms of software development and hardware requirements. Having said that, given its extremely important functionality – transmitting provisional results – extensive testing of the system, targeted training of users and a proactive EMB command and control capability are necessary additional components adding costs.

Where conventional fixed telephones or facsimiles are not available, mobile phones or radios may be used. Results may also be sent by electronic mail where access is available. Where no electronic options are available, results may have to be delivered by courier.

Human Resources versus Cost of Technology

Use of technology for counting votes may reduce the overall cost of an election and be more effective under certain conditions.

However, depending on local factors, the process of manually counting paper ballots may be comparatively less expensive. Whether a technological solution is cost-effective will depend on a range of local circumstances, including:

• the complexity of the voting system,
• the number of ballots being counted,
• the relative cost of labour compared to technology,
• whether counting is centralized or decentralized.
• the availability and cost of skilled IT professionals for development and support, and
• the differing recruitment profiles; are suitable persons available to hire and train to operate the technology? 

In general, a mechanical or computerized method of counting paper ballots (as opposed to systems where votes are cast electronically or mechanically) may be cost-effective, where large numbers of ballots are being counted centrally, where a voting system is used that lends itself to mechanical counting, and where labour costs are relatively high.

Cost considerations aside, election management bodies should be aware that mechanical or computerized counting methods might be faster and more accurate than manual methods. However, some systems have proven insufficient transparency as lacking a paper trail and thereby the actual results stemming from some voting machines cannot be verified in a recount.

While the use of technology might appear a costly option, if the technology acquisition costs can be spread over several elections, the investment may be worthwhile. Experience has shown, however, that many systems do not deliver the hoped-for lifespan that would make the original investment worthwhile.

On the other hand, rapid advances in technology tend to cause electoral technology to age quickly and for technology to become defunct between one election and the next. Availability of technical support during the electoral period is another important aspect that should be considered.

A detailed cost-benefit analysis, accompanied by a study of the legal impacts and management risks associated with the adoption of a particular technological solution, may be worthwhile before any final decisions are made to adopt technological vote counting methods.