Big Saturday Read: Rural bias and voter registration in Zimbabwe
For a lot of people, the mention of “vote rigging” conjures up images of shady characters in dark glasses busy stuffing ballot boxes and disposing of ballot boxes in rivers, replacing them with pre-loaded ballot boxes. To be sure, vote rigging is a broad and amorphous term which includes a lot of things in its boundaries. It is not a single act and oft-times it involves multiple activities at different times before, during and after the electoral process. The primary purpose of those who rig elections is to create the maximum advantage possible for their preferred candidate while reducing the chances of their main opponents in all cases while using unfair means. The difficulty with identifying a lot of the vote rigging is that it is often done using means that have a veneer of legality. This false appearance of legality fools many people, including election observers and courts of law into believing that everything is free and fair. This BSR looks at one of the more subtle but very effective, albeit understated ways of manipulating electoral outcomes. This area is the control the voter registration system and the voters’ roll.
Determining the political community
Voter registration is critical because it regulates access to the electoral process. Those who control voter registration are the gatekeepers of the electoral process. They determine who votes and who can’t vote. Voter registration determines the composition, quality and size of the political community that is eligible to participate in elections. It therefore presents the first opportunity of controlling and manipulating the electoral process. While the principle of universal adult suffrage is recognised by the Constitution and while the right to vote is protected in the Declaration of Rights this right is qualified by the requirement that only persons who are registered to vote and appear on the voters roll are eligible to vote. The right to vote is therefore dependent upon the realisation of the right to register as a voter. It follows therefore, that if you want to control and manipulate the outcome of an election, your starting point is controlling the population that is eligible to vote. You make sure that those who are likely to support you are registered while excluding those that are likely to favour your opponents. In that way you enhance your chances while diminishing your rival’s opportunities. This explains why voter registration is an important battleground in the electoral process.
In assessing an electoral system, it is therefore important to consider whether it provides for a fair, open and inclusive voter registration system. A voter registration system that is exclusionary and therefore leaves out other sections of society cannot facilitate a free and fair election. Voter registration is an important part of the bargain that is struck between society and the individual. The individual has a right to vote while society has an interest in ensuring the preservation of its political community. As one of the great political philosophers, John Stuart Mill wrote in his treatise Considerations on Representative Government, “it is personal injustice to withhold from any one, unless for the prevention of greater evils, the ordinary privilege of having his voice reckoned in the disposal of affairs in which he has the same interests as other people”. The underlying belief in the idea of preservation of a political community is that only those people with sufficient interest in a particular society should be allowed to vote in that society.
Historically, the criteria for determining the adequacy of those interests has varied according to race, gender, wealth and residence. Even for John Stuart Mill, his preferred political community was limited and excluded those who did not have property or those who depended on the welfare of the state. Indeed, for a long period in Europe only white males with a certain level of property ownership constituted the political community which was entitled to vote. It was not until 1918 that women got the right to vote in Great Britain. In the African colonies, black people were not allowed to vote although over time a few became eligible if they were able to meet certain educational and property qualifications. One of the major reasons for the rejection of the 1961 Constitution by the African nationalists in Southern Rhodesia was that it imposed restrictive voting qualifications on Africans. The colonial state maintained its power partly by controlling the pool of voters, ensuring that it included most of those likely to favour the preservation of the political community upon which it was based. The idea of black majority rule a threat to the political community of the colonial state.
Mirage at independence
At independence, while the principle of universal adult suffrage was adopted, it did not completely transform the political community. Instead, a political bargain was struck at the Lancaster House Constitutional Conference in order to protect the interests of the white minority population and therefore preserve residual elements of the old political community. This was achieved by creating two voting rolls: the White Roll, for the white community and the Common Roll for the rest of the population. It ensured that, at least for that limited period, the white minority still had a notable stake and representation in the new political community that came with the advent of independence and black majority rule.
Nevertheless, the new electoral system, open though it seemed retained certain qualifications which had the potential to be exclusionary. The major one was the residence requirement, which still remains on the books even under the new Constitution. In terms of Schedule 4 to the 2013 Constitution, to be eligible to register as a voter, a person must be at least 18 years old and a law may make an additional requirement for proof of residence. Although it is not a compulsory constitutional requirement, the Electoral Law requires applicants to produce proof of residence. This could be in the form of title deeds, utility bills, letter from the landlord, letter from the village head, etc. This proof of residence requirement has class implications, with potential to discriminate not only the homeless but also those who have no access to the official documents that are demanded.
Recognising this potential limitation and exclusionary effect, the authorities have expanded the definition of proof of residence to include a sworn affidavit by the applicant declaring a particular address as his residence. Use of this affidavit can help to overcome the proof of residence bottleneck. It’s a measure that can enhance inclusivity in the electoral process where proof of residence would otherwise be exclusionary. However, it could also be a complete charade if the numbers of affidavit forms and the Commissioner of Oaths before whom the affidavits are sworn are inadequate. It would give the misleading impression of inclusivity, while actually delivering very little in practice. This was a real problem before the 2013 elections leading to the exclusion of many potential applicants.
Apart from the issue of proof of residence, those who seek to control the political community that votes can do so through various means which reflect multiple biases. These biases can be based on regionalism, ethnicity, class or the rural-urban divide. Some regions, ethnic groups or classes may be favoured more than others. This would mean allocating more voter registration facilities in some regions or ethnic communities compared to others. Electoral patterns in the 1980s showed that of the two major political parties, Mugabe’s ZANU PF drew the bulk of its support from the Shona ethnic groups while Nkomo’s PF ZAPU had a near monopoly in the Ndebele ethnic groups. Zimbabwe of the 1980s was divided along ethnic lines. The Gukurahundi policy pursued by the ZANU PF in the Matebeleland and Midlands regions in the 1980s affected the local population’s access to birth and death registration facilities. The effect is that there are still many young men and women who to this day remain without registration documents which are necessary to register as voters. Chronic and systematic under-development in those regions since independence means young people traditionally trek across the border into South Africa, thereby reducing the voting population.
Of interest in this BSR is the phenomenon of rural bias, whereby the electoral system is biased towards and tends to systematically favour rural areas compared to urban communities. This normally manifests in the disproportionate allocation of electoral facilities to rural communities compared to urban areas. However, it also shows in the policies championed by the ruling establishment, particularly towards elections. This rural bias is evident in the voter registration system. Although it may not immediately appear as such, rural bias is part of the election rigging infrastructure because it is designed to manipulate the outcome of elections.
In their 2015 work on rural bias, Boone and Wahman point out that in the African countries they studied, “the electoral game has been systematically stacked against the urban areas.” The reason for this, as Robin Harding writes is that, “across sub-Saharan Africa support for incumbent governments is significantly higher among rural residents than urbanites, although the magnitude of this difference varies across countries.” Boone and Wahman refer to “the problem of electoral malapportionment” whereby there is an unequal allocations of parliamentary seats between rural and urban areas, with the result that some votes will end up carrying greater weight than others thereby undermining the ‘one person one vote’ principle which requires all votes to carry the same weight. When rural areas have a disproportionate number of parliamentary seats compared to urban areas, this distorts the weight of votes since rural votes carry greater weight. This is why in some cases, delimitation of constituency boundaries has resulted in the creation of more seats in rural areas which enhances the levels of malapportionment. In other cases, peri-urban constituencies are merged with predominantly rural constituencies as a way of diluting the urban votes. Although Boone and Wahman’s paper is in relation to malapportionment of parliamentary seats, creating a rural bias, the argument can be applied with necessary modifications in relation to malapportionment of voter registration centres between urban and rural areas.
Diluting urban vote
As Boone and Wahman point out, rural bias in electoral systems matters a great deal because its effect of diluting the weight of the urban vote which usually form the core support base of opposition parties. As scholars like Mahmood Mamdani have pointed out, civil society activism has largely been an urban feature. Across Africa, opposition support has largely been drawn from urban areas and this has been a pattern even in the anti-colonial struggles where nationalist politics emerged with the new educated urban intellectuals and working class movements in the 1940s and 50s. Ruling parties tend to rely on rural support to counter-balance the urban-based opposition movements. They refer to the work of Huntingdon who argued in the 1960s that in developing countries rural support is used by authoritarian regimes as a counter-weight against pro-opposition urban support.
The liberation struggle bequeathed a highly politicised and politically active rural population in the new Zimbabwe. Guerilla warfare meant that much of the war was fought on rural terrain. Using Maoist war strategies, the guerillas referred to themselves as the fish while the rural masses were the water that sustained them. Guerilla warfare involved political education and morale boosting all-night sessions called pungwes where the fighters interacted with the rural masses. In the 1980 elections, the rural areas were a key electoral battleground with ZANLA in particular deploying guerillas to do the campaigning for political candidates. ZANU PF structures were formed in that era and they have endured to this day. Naturally, rural areas especially in the Mashonaland provinces have been known to be strongholds for ZANU PF. It is in ZANU PF’s interests to ensure that residual loyalty in the rural areas is properly and adequately registered.
According to the last census conducted in 2012, 67% of Zimbabwe’s population resides in the rural areas, which leaves on 33% in the urban areas. If these figures are taken to be representative of the current population, this means the majority of the voting population is likely to be in the rural areas. Taken together with all the other advantages to be found in the rural areas, the size of the population is an important incentive for rural bias in the electoral system. The incumbent will pour more resources and design more pro-rural policies in order to attract the attention and support of rural communities. That is where the majority of voters are. This point is echoed by Harding who writes, “simply stated, because a majority of Africans live in rural areas, competitive elections make African governments more responsive to rural interests, resulting in dissatisfaction on the part of urban voters.” The point is that it is not so much the threat of urban hostility as the opportunities presented by rural populations that drives incumbents to promote rural bias in the electoral system. The point is succinctly summarised by Harding who writes, “In the presence of a rural majority, incumbents who can win in the countryside can afford to ignore urban voters, and risk generating a certain degree of urban dissatisfaction, so long as doing so does not lead to urban unrest that may destabilize the regime. Therefore where a majority of the population is rural, incumbents should distribute sufficient resources to buy-off urban unrest, without needing to ensure that they win urban votes.”
ZANU PF has successfully used the coercive apparatus of the state to put down any signs of urban unrest and considers the threat to be minimal. With the threat of unrest being so negligible, it can afford to lose urban votes while concentrating on the rural population which stands in the majority. It is not surprising therefore, that the electoral system reflects serious rural bias.
In an agrarian economy, the land is an important asset which has been at the centre of contestations for generations. The political economy of rural Zimbabwe revolves around land and agriculture. Boone and Wahman refer to rural areas “captive constituencies” for incumbent regimes. The incumbent’s policies tend to reflect the interests of the rural population. Since, as we have seen the majority of voters reside in rural areas, competitive elections push incumbents to devise public policy decisions that are more responsive to the rural population. ZANU PF’s policies have been tailored to meet the interests of this agrarian political economy. The much-hyped Command Agriculture policy, the Presidential Inputs Scheme, the pro-producer grain pricing policy from the GMB and indeed the entire land reform programme are all policies that are deliberately designed to meet the interests of this large local economy in the rural areas. It has become more important with the land reform programme and the creation of an insecure land holding constituency which is eternally grateful and loyal to ZANU PF. Without secure property rights to the land, these new farmers have very limited choices but to work with ZANU PF. The opposition can expect the hype around Command Agriculture to persist throughout the election season. A lot of noise has been made about Command Agriculture and the good harvest that came courtesy of the good rains mean Command Agriculture or similar will definitely be at the centre of ZANU PF's election campaign. Naturally, it is in ZANU PF’s interests to ensure that people in rural areas have easy access to voter registration.
It is relatively easy to manipulate the traditional leadership systems to work in their favour. Traditional leaders are very malleable. They were co-opted by the colonial regime but after independence they flipped and became a pillar of the ruling ZANU PF party. Likewise, these traditional leaders will work with whatever party comes after ZANU PF. Without power of their own after colonialism dismantled their sources of power, traditional leaders are at the mercy of whoever holds state power. In turn they become willing tools of any incumbent. As Boone and Wahman put it, “In many places, rural strongmen owe their positions and power to rulers at the center, and rural voters are less autonomous from local strongmen, less mobile and more enmeshed in local social networks, generally poorer and less literate, and easier to monitor than their urban counterparts”.
In addition, as Harding points out, rural voters tend to be less autonomous than their urban counterparts in their voting decisions. Election observers and opposition parties have in the past reported how traditional leaders literally corral villagers under their command to comply with ruling party demands at the pain of punishment and other societal sanctions which include exclusion. In an article on the Panopticon Effect in rural areas, I have previously explained how the rural population is forced to comply because of a system that makes them believes that they are constantly being watched. The watchmen for the ruling party are usually the traditional leaders. In this regimented system, it makes sense for ZANU PF to concentrate voter registration facilities in rural areas.
Zimbabwean elections since independence have been notoriously violent. The 2008 violence was probably the worst electoral violence since 1980, with voters being punished by the ruling party and its associates after Mugabe was defeated by Tsvangirai in March 2008. Although political violence has been less visible since that dark period, sporadic acts and the legacy left by the scourge of violence still affects rural communities. The election season is often accompanied by an atmosphere of fear and insecurity. While urban areas were affected, the tightly-knit communities in the rural areas were more vulnerable. To insure against violence, villagers have to comply with what they believe to be ruling party demands. As Boone and Wahman point out in their work, “… for opposition parties in Africa, the monetary, transaction, and political costs of campaigning in rural areas are often higher than they are in urban constituencies.”
Opposition limitations in rural areas
Given their control of structures of the state and access to greater resources, incumbents tend to have more presence and influence across the country compared to their opposition counterparts. As Boone and Wahman point out, “Opposition parties, unlike their incumbent counterparts, often lack the resources to build nation-wide operations”. Harding adds that rural populations “are less exposed to opposition party campaigning” largely due to the challenges faced in the rural areas. With the natural advantage of incumbency, ruling parties tend to maximise on their wider presence and reach in the rural areas. Naturally, it makes sense for the ruling party to create an electoral system which is heavily tilted towards the rural areas where in some cases it has a virtual monopoly.
Incumbents systematically focus more on rural areas because urban voters tend to be more hostile to them, often forming the core of opposition support. Zambia’s MMD and Zimbabwe’s MDC both trace their roots from the largely urban-based labour movement. In the 2000 elections, the MDC swept the bulk of seats in the urban constituencies, which have remained their strongholds. There is little incentive for the ZANU PF government to deploy resources to hostile territory, preferring their rural strongholds. This focus on rural areas at the expense of urbanites has the effect of creating hostility among the latter. As Bates and Block have argued in their 2009 work electoral competition and the fact that the rural population is larger means incumbents have a powerful to give more priority to the interests of the rural population which is easily bought by gifts. In their research, Boone & Wahman found that while rural bias pre-dates the arrival of multi-party politics in Sub-Saharan Africa in the 1990s, it has persisted even after opposition parties have assumed power. The persistence of rural bias after a change of the ruling party is an interesting phenomenon, which demonstrates that there are incentives for incumbents to preserve the favours towards rural constituencies. The new ruling party might see the gains in maintaining or enhancing the status quo thereby perpetuating rural bias.
Rural bias by the new incumbents may also be explained by the fact that there is traditionally a frosty relationship between incumbents and urbanites. New ruling parties that have risen on the back of urban support have often found themselves losing that backing once they have assumed power. Boone and Wahman cite the examples of Zambia, Ghana and Senegal where such trends have been evident in recent years. In the case of Senegal, after repeated failure, Abdoulaye Wade eventually won power in 2000 on the back of popular urban support. However, in 2012 he lost power to popular urban-based opposition. In their 2007 paper, Larmer and Fraser show how in the case of Zambia, after winning power on the back of popular urban support, the MMD lost nearly all its urban parliamentary seats and all seats in the urbanised Copperbelt region.
In its observer report on the 2013 elections, ZESN the country’s largest network of observers reported that “the voter registration process was systematically biased against urban voters.” Citing an analysis of a voters’ roll that was made available on 19th June 2013, ZESN reported that there was evidence showing that “urban voters had systematically been denied the opportunity to register to vote.” The difference in registration rates between rural and urban voters was staggering. According to the ZESN report, “a total of 99.97 percent of rural voters were registered while only 67.94 percent of urban voters were registered.” The problems with the registration and rural bias were visible on election-day with clear evidence of disenfranchisement of urban voters. According to ZESN hundreds of thousands of urban voters were turned away at urban polling stations (82% rate) compared to rural polling stations which recorded a 38% rate. It was not possible to come up with a definitive calculation of rural bias because, contrary to the law, the electoral authorities refused to avail the electronic copy of the voters’ roll.
Serious concerns over the rigging of the 2013 elections arose from the fact that the electoral authorities refused and/or failed to avail the electronic copy of the voters’ roll used during that election. By law, ZEC and the Registrar-General were required to make searchable and analysable electronic copies of the voters’ roll available to parties and candidates. By the time polling day arrived, they had failed to make it available. The opposition approached the courts but court orders were not complied with. In one case the court made an order which virtually allowed the electoral authorities to avoid providing the voters roll on the basis that the RG’s computer had broken down. Nearly 5 years later, that electronic copy of the voters roll remains a mystery. The secrecy over that voters roll and the refusal to make it available suggests very strongly that it was the centrepiece of the vote rigging that took place in 2013. The fact that judicial remedies were ineffective demonstrates how the entire system is set up to promote the interests of the incumbent and to facilitate the vote rigging that goes on using the voters roll.
For this reason, the voters roll will once again be a key battleground between the opposition parties and ZANU PF. ZANU PF will not want to cede its exclusive hold over the voter registration system and the voters roll because that is where the game is played. The opposition parties have to remain vigilant and it would be foolish to go into an election once again without access to the voters roll. Going into an election without access to the voters roll, as the opposition did in 2013, is the equivalent of a football team playing the game blindfolded while the opponent has complete vision. How the voter registration process is conducted and how the voters’ roll is handled goes to the heart of the election process and affects its legitimacy. All interested parties must create and maintain an audit report of the process which must be made available to election observers who usually arrive just a few weeks before polling day.
In a recent policy brief, ZESN has highlighted these issues demonstrating that while voter registration has been “easy, more accessible and fast in the rural areas, it was more laborious in the urban areas.” It has pointed out that voter registration tends to favour ruling party strongholds. The “systemic rural bias” was clear in the “disproportionate distribution of mobile registration units in the rural areas” compared to the urban areas. The reason is that the electoral authorities tended to give more priority to geographic distance over other factors.
Disproportionate allocation of voter registration centres
Following the proclamation by the President, ZEC was supposed to have started registering voters from 14th September 2017. However, because they were not ready, the actual process was started on 18th September. Even then, ZEC is clearly not ready to roll out the comprehensive voter registration process using the new BVR system. That’s because it does not have the BVR kits that are required for the process. They are due to arrive in mid-October, a full month after the start of the proclamation date. In any event, ZEC was still training its agents when the proclamation was made. More importantly, the distribution of voter registration centres demonstrates a clear bias towards rural areas.
More than just distance
While population differences (67% rural to 33% urban) might be used to justify the deployment of more voter registration centres in rural than urban areas, it cannot justify the grossly disproportionate and inequitable distribution of voter registration centres. According to Zimstat, the national statistics agency, Harare had a total population of 2,1 million people in the 2012 census, which has naturally increased in the last 5 years. However, ZEC allocated only 2 voter registration centres in Mbare and Chitungwiza. By contrast, Mashonaland Central, with 1,2 million people was allocated 8 voter registration centres. In the southern region, Matebeleland South Province with 680,000 people was allocated 7 voter registration centres while by contrast Bulawayo with a comparably close population of 650,000 was allocated just 1 voter registration centre. Other mainly rural based provinces have similarly high numbers of voter registration centres compared to the two metropolitan provinces. It means 23% of the national population residing in Harare and Bulawayo have been allocated just under 5% of the total number of voter registration centres. By contrast, 76% of the population in the mainly rural provinces have over 95% of the centres. This distribution is disproportionate but it reflects the rural bias that has been discussed in this BSR.
The reason that is often cited to justify these allocations which favour rural areas is geography and distance. It is argued that rural provinces cover wide areas and the distances that voters must travel are too long and difficult due to geographical barriers such as mountains and rivers as well as a lack of reliable transport. By contrast, urban areas tend to have more transport facilities and are easier to traverse. Yet, while the issue of distance is important, there has been a risk of over-emphasis of this factor at the expense of other equally important considerations which affect voters in urban areas. Distance is an important consideration but it is by no means the only critical factor.
In this regard, population density is an important factor. Urban areas tend to have higher population concentrations in small spaces, which means more pressure on the resources. 2 voter registration centres for a population of 2,1 million in urban areas is certainly grossly inadequate. It only puts more pressure on the 2 registration centres, leading inevitably to long queues, delays, and fatigue on the part of both voters and staff. A combination of all these factors leads to frustration on the part of voters who are dissuaded from taking part.
Time is an important factor in urban areas where the majority have to work on a day to day basis. This is especially significant in an informal economy where people are working for themselves and cannot afford to take time off for long periods. An urban voter must weigh between queuing for long hours in order to register and plying the streets to trade their wares to earn income. Most of the time, a person will make the rational choice to fend for his family. If, however, the voter registration process is simple, easy and fast, they might sacrifice a little of their time to register, leaving them with time to carry on with their businesses.
While transport is more available in urban than in rural areas, there is also the cost factor which is often overlooked. For some people a dollar might seem like small money, but in an economy that is effectively comatose, it is a lot of money which some people cannot even afford. A person will have to make a rational choice between spending a dollar for transport to go and register and using that money to buy food for the family. Oft-times, they will choose the latter. This is made worse when the registration facilities are few, the queues are long, leading to delays and voters might be turned away for one reason or another.
Overall, it is clear that there are multiple factors that must be taken into account in distributing voter registration centres. ZEC has said there will be additional facilities after the arrival of the outstanding 2,600 BVR kits. In Harare for example, ZEC has said there will be an additional 700 mobile BVR units. These mobile registration units will be most welcome and might reduce some of the handicaps in the current system. However, one is always mindful of the precedent set by the initial distribution of centres which was highly skewed in favour of rural areas and has to make sure that it does not reflect future trends.
In addition, civil society groups and the opposition parties have to keep a watchful eye not only on the number of units but also on efficiency at those registration centres and units. There could be more units but there could also be deliberate delays at urban centres. Electoral authorities have form in this regard. In 2013, observers found that thousands of voters in Harare were disenfranchised due to the slow and inefficient registration process which meant those who were queuing were turned away when the gates closed. Indeed, in 2013, ZEC itself acknowledge the problem and ordered a short extension overnight to accommodate queuing persons. Even so, thousands still failed to register. By contrast, there were no such problems in the rural areas, where there were more mobile voter registration units and fewer people per unit or centre. Already, observers are noting the problem of lack of commissioners of oaths who are supposed to play an important role in administering affidavits sworn by applicants declaring their residence. With few or no commissioners of oaths, the process is longer, slow and more frustrating, again reflecting bias against urban areas.
The phenomenon of rural bias is real. It is evident in the demarcation of constituencies where there is a tendency to favour rural areas. It is evident in the policies pursued by incumbents, which tend to prioritise the interests of rural communities. It is also evident in the distribution of voter registration facilities, which tend to be more concentrated in the rural areas than in urban areas. All of this has the effect of promoting rural bias and giving the rural vote greater weight than the urban vote. It also has the effect of disenfranchising urban voters, particularly the young and first-time voters who must seek registration before they can be eligible to vote.
Historically, the one person one vote principle was one of the cardinal demands of the liberation movements. While some of the impediments that impeded this principle were removed in 1980, there remain residual elements that impede its full realisation. The proof of residence requirement is a residual element of the old order which seeks to disqualify others from the political community that votes. The actual operation of the electoral system including the distribution of voter registration centres and the process of registration can also work to disenfranchise persons due to manufactured inefficiencies. While the use of the affidavit to overcome the proof of residence requirement can be helpful, it will be rendered ineffective unless there are sufficient commissioners of oaths to administer the affidavits. Since ZEC is unlikely to provide more commissioners of oaths, the duty falls upon opposition parties and civil society to mobilise the services of lawyers to volunteer their services for this important function. To her credit, Mt Pleasant Independent candidate Fadzayi Mahere is already leading the way in this regard. The Electoral Resource Centre has also tweeted a call for qualified volunteers to act as commissioners of oaths.
This is a time when the spirit of volunteerism must trump the pursuit of profit. Just as lawyers descended upon the Harare Magistrates Court last year to defend Pastor Evan Mawarire, this is one occasion that calls for a similar spirit of volunteerism. For their part, urban voters, especially the young who may have limited reservoirs of patience, must appreciate that for reasons discussed in this BSR, the electoral system will always be designed with a rural bias and to frustrate them. Giving up is not an option because that is precisely what the incumbent regime wants. It would be gutless capitulation. ZANU PF will always prioritise its rural strongholds where it can maximise on its ability to control and manipulate the electorate. Since rural areas have more registration facilities and since one is permitted register at any centre in the country, one strategy will be to go and register whenever one is in a rural area. For parties, if it means transporting urban supporters to rural centres to register, it must be considered, too.
Some useful references used in this article:
Boone and Wahman “Rural bias in African electoral systems: Legacies of unequal representation in African democracies” Electoral Studies Volume 40 December 2015 335
Robin Harding (New York University) 2012 Democracy, Urbanization, and Rural Bias: Explaining Urban/Rural Differences in Incumbent Support Across Africa http://robinharding.org/RuralBias_Draft_June2012.pdf