BSR: Dealing in Numbers – a critique of voting figures in Manicaland Province


In the last installment of the BSR, we looked at the election results from Mashonaland West province and the lessons that could be drawn from them. This installment focuses on Manicaland province. We will observe that there are similar themes. At the end of this exercise, after examining all 10 electoral provinces we shall identify common themes and some that are unique to each province. All these are relevant to political strategists as the country heads towards the next general elections in 2023.

A tale of two races

Manicaland has 26 constituencies consisting of urban, peri-urban, and rural constituencies. At the parliamentary level, ZANU PF won 19 seats while the MDC Alliance won the remaining 7. However, although none of the other parties and independents won any seat in parliament, as we shall see, in some cases they had an influential role in the outcome of the constituency race. In the presidential race, ZANU PF’s candidate, Emmerson Mnangagwa won 17 constituencies, two less than his party’s tally, while the MDC Alliance’s Nelson Chamisa won 9, 2 more than his party’s total.

The MDC Alliance did not do as well as its presidential candidate because it lost seats to split voting in the parliamentary race. Some voters were prepared to vote for Chamisa in the presidential race but they preferred other opposition candidates in the parliamentary race. For example, Chamisa won Mutasa North with a 1380 majority. However, the ZANU PF candidate narrowly won the parliamentary race with 11913 votes against the MDC Alliance’s candidate who got 11635 votes. This happened because some of the opposition votes went to an MDC-T candidate who took 1060 votes. A combined effort by the opposition (12,695) would have beaten ZANU PF.

Musikavanhu in Chipinge District is another seat that was lost by the opposition due to split voting. Chamisa beat Mnangagwa in the presidential race with a 2433 majority. However, the ZANU PF candidate won the seat with 6599 votes in the parliamentary race narrowly beating the MDC Alliance’s candidate that had 6162 votes. But 2 other opposition candidates got 1855 and 1029 votes respectively. A combined effort by opposition candidates would have defeated the ZANU PF candidate with a majority of 2447, very similar to the majority that Chamisa had ahead of Mnangagwa.    

Mutare North could have been won by the opposition if the opposition parties had agreed to field a single candidate against ZANU PF. Candidates of the MDC Alliance and the NPF had 3962 and 3922 respectively, a total of 7261 votes between them. By contrast, the ZANU PF candidate who won had 6511 votes.

Red Flag in Mutare North

However, Mutare North provides the first sign of chronic problems that affect Zimbabwe’s electoral system. There is a glaring divergence in the number of voters who participated in the presidential and parliamentary races. The ZANU PF candidate got 6511 votes in the parliamentary race, while Mnangagwa is credited with 18999 votes in the presidential race. Parallel to this, while the MDC Alliance candidate got 3962 votes in the parliamentary race, Chamisa is credited with 12014 votes in the presidential race. The difference in both cases is both extraordinary and exceptional.  

In total, according to ZEC’s official figures, the number of people who voted in the parliamentary race in Mutare North is 14063 whereas the total number of voters in the presidential race in the same constituency is 32880. According to these figures, 18817 people in Mutare North chose to vote in the presidential race but not in the parliamentary race.

Table A: Mutare North Constituency – Presidential and Parliamentary Election Results

Number of voters in the Presidential ElectionNumber of voters in the Parliamentary ElectionDifference between voters in presidential and parliamentary races
Mnangagwa: 18999
Chamisa: 12014
Other candidates: 1217
Total: 32880

ZANU PF: 6511
MDC Alliance: 3962
NPF: 3229
ZIPP: 154
Total: 14063
Presidential race: 32880
Parliamentary race: 14063
Difference: 18817
Source: Zimbabwe Electoral Commission website

In other words, the ZEC would have us believe that most voters in Mutare North did not bother to vote in the parliamentary race but chose to focus only on the presidential race. This figure is far more than the total number of people who voted in the parliamentary election. Analysis of the polling station registration data shows that 36806 voters are registered to vote in Mutare North. This would mean the voter turnout in the presidential race was 89% while in the parliamentary race it was just 38.2%. This is hard to believe.

Something truly smells fishy in the Mutare North election numbers. Therefore, it is important to carry out a more systematic analysis of the voting figures, comparing the number of voters that supposedly voted in the parliamentary and presidential races. While there are bound to be some differences, they must be within reasonable limits. The differences in Mutare North fall outside the boundary of what might reasonably be expected. The big question is how many more Mutare North scenarios are there across the country?

Mind the Gap

Although both Mnangagwa and ZANU PF won most of the constituencies in Manicaland in both the presidential and parliamentary races, Chamisa won the majority vote in the presidential race Chamisa got 296249 votes while Mnangagwa got 292938 votes. However, although Chamisa won the province, the majority of just 3311 votes was very small. Therefore, questions regarding the voting figures in Mutare North deserve serious attention. One hypothesis is that figures in the presidential race can be deliberately fudged to minimize or enhance the gap between the candidates.  

The size of the gap matters because if you are going to rig an election in a way that does not raise too many questions, you will let your opponent win some provinces but reduce the gap by which they are winning. And where you win, you win big so that you maintain an unassailable advantage. This means as a player in the electoral process, you must pay special attention to the numbers and not just be drawn to the broad narrative of winning or losing a province. In this election, while Chamisa won Manicaland with a narrow majority of just 3331 votes, in Mashonaland West which we examined last week, Mnangagwa won with a majority of 94430 votes. Therefore, although they each won a province, Mnangagwa’s lead would be 91009 votes. This underlines once again why every vote matters in the presidential race.

Rural-Urban Divide

One of the reasons why Chamisa won the popular vote in Manicaland is that the province has important urban areas that have large population densities, and the opposition traditionally performs better in urban zones. Chamisa drew his biggest vote in Dangamvura/Chikanga, the sprawling high-density zone of Mutare where he got 37370 votes compared to Mnangagwa’s 11750 giving him a majority of 25650. But ZEC registration figures show that this constituency has more than 60000 registered voters, which means there is plenty of room for improvement in what is an opposition stronghold.

In Mutare Central, another opposition stronghold, the gap between Chamisa and Mnangagwa was 9618 votes. The total number of voters was just over 18000 in a constituency of more than 22000 registered voters, which also means there is room to do better in a stronghold. Better performance in Mutare South and Mutare West would also have enhanced the gap. However, as already highlighted, the anomalies in Mutare North raise questions about the authenticity of these figures which helped Mnangagwa to close the gap significantly.

Although Mnangagwa and ZANU PF won most of the rural seats, their advantage was weakened by the fact that both Chamisa and the MDC Alliance put up some commendable performances. In the presidential race, Chamisa managed to snatch Buhera West albeit by a narrow majority of 264 votes, but the MDC Alliance candidate lost the seat. Mnangagwa won the other 3 Buhera constituencies but with an average majority of around 3000 votes.

Chamisa also won Makoni West with a small majority of 227 votes but lost Makoni South with a small deficit of 668 votes. The gaps in the other 2 Makoni seats were wider in favour of Mnangagwa at more than 5000 on average. Mnangagwa posted a large majority in Headlands, probably another reflection of the political economy factor in an area that is dominated by resettled farmers whose insecurity of title makes a captive market for ZANU PF.

Peri-urban swing constituency

Mutasa South, which was recently lost by the opposition in the by-elections had swung to the opposition in the 2018 elections. It is always a closely contested seat in the parliamentary race which makes it a swing constituency. In 2018, the MDC Alliance candidate got 14783 votes while the ZANU PF candidate had 12736 votes. However, in the presidential race, Chamisa was well ahead of Mnangagwa with 17615 votes to 10257 votes. It was one of Chamisa’s largest majorities in the province making it one of his strongholds, although it is clearly a marginal seat at the parliamentary level as the by-election outcome revealed with the seat narrowly swinging to ZANU PF by a tiny majority of just 495 votes.

There is no reason why the parliamentary candidate cannot match the presidential candidate’s performance. With over 33000 registered voters, Mutare South Chamisa has room for improvement. We saw in the recent by-elections that most registered voters are in the urban wards, but they did not turn up in large numbers, which disadvantaged and cost the CCC parliamentary candidate. By contrast, the rural wards, which voted for the ZANU PF candidate, had a better turnout. In short, the opposition has a chance to win back Mutare South come 2023, and Chamisa has room to increase his voting numbers.  

Chipinge – a swing district

Chipinge and Chimanimani districts present an interesting set of results. In the case of Chipinge, the figures suggest that it is a swing political zone. In the 7 constituencies, 73132 voters chose Mnangagwa while 62121 voted for Chamisa. The figures were lower in the parliamentary races where 69999 voted for ZANU PF candidates while 57879 voted for MDC Alliance candidates. The reduced number of voters for MDC Alliance candidates is partly because of more competition from other opposition candidates who took some of the opposition votes. Although they voted for non-MDC Alliance candidates, they seem to have chosen Chamisa in the presidential race.

Although Mnangagwa won 3 of the Chipinge seats 2 of them (Chipinge South and Chipinge East) were very close with only Chipinge Central giving him a bigger margin of just over 5000 votes. In Chipinge East his majority was just 175 votes. At the parliamentary level, Chipinge South is another seat that was lost by the opposition because of a split vote. The ZANU PF candidate won the seat with 9382 votes, but the MDC Alliance candidate had 7870 while the MDC-T candidate got 2366 votes. The two opposition candidates had a combined total of 10236, higher than the ZANU PF candidate’s votes. We have already observed that Musikavanhu is another seat that the opposition lost because of a split vote.

Nyanga has two constituencies that were evenly split between Mnangagwa and Chamisa at the presidential level but both went to ZANU PF at the parliamentary level. Mnangagwa won Nyanga North with a majority of 4561 votes while Chamisa’s victory in Nyanga South was narrow at 2581 votes. It is notable that the ZANU PF candidate for Nyanga South, Supa Mandiwanzira (12322), did considerably better than Mnangagwa (10298) whereas Chamisa (12879 votes) did significantly better than his parliamentary candidate (7464). This suggests that voters who chose Mandiwanzira in the parliamentary race also preferred Chamisa in the presidential race, ahead of Mnangagwa.

The Chimanimani Seats both went to ZANU PF and Mnangagwa in the two races although one was closer. Mnangagwa’s majority in Chimanimani West was only 1376 and ZANU PF’s majority at the parliamentary level was also a slim 1558. However, ZANU PF’s majority in Chimanimani East was bigger with the parliamentary candidate getting 13458 more votes than the MDC Alliance’s candidate. The majority at the presidential level was reduced with Chamisa increasing the opposition to more than 8000 votes. However, still, the gap was high at just over 8000 votes.

This is a seat that was formerly in the hands of the MDC during the time of Roy Bennett who was massively popular with locals. The political economy of the area has changed massively following the land reform programme and increased activity in artisanal mining. While there are good chances of snatching Chimanimani West, the strategic hope in Chimanimani East must be to reduce the margin of loss. This is critical for the presidential race where every vote matters.

Manicaland also hosts 3 Mutasa seats all of which were won by Chamisa in the presidential race. We have already examined Mutasa South above. The MDC Alliance lost one of the seats (Mutasa North) because of split voting between opposition candidates. The presidential race in Mutasa North was already finely balanced with Chamisa winning by a thin margin of 1380, so the moment there was split voting at the parliamentary level, the seat was at a higher risk of being lost which is what happened.

Overall, the following points are worth noting in respect of Manicaland:

· The divergence in voting figures in Mutare North raises a significant red flag over the authenticity of ZEC’s election results. While voters do not have to vote in both races, a difference of more than 18800 people voting in the presidential and the parliamentary elections raises serious questions. A further study of polling station and ward voting figures is recommended as is a more systematic study that compares voting figures in all two races across all constituencies.

· A party can lose an election, but for its candidate to win the popular vote in the presidential race as Chamisa did in Manicaland. This is because voters’ choices in harmonized elections are complex and nuanced. There is no necessary correlation between voting for a political party and voting for its presidential candidate.

· In the presidential election, every vote matters, which is why even if a constituency is lost, what matters is not the loss but its size. A candidate should aim to narrow the gap if they are losing and to increase the gap if they are winning.

· Each party and presidential candidate have their areas of strength and zones of weakness. There are, in elections all over the world such things as safe, weak, and marginal seats. As a matter of strategy, each party should do a frank assessment of where it is strong and where it is weak. It should classify constituencies into safe, weak, and marginal/battleground seats.

· These indicators are important for allocating scarce resources. For example, a party might decide to allocate more resources to minimize the gap in its weakest areas to enhance its presidential candidate’s chances while investing in battleground seats to gain an advantage in the parliamentary race. There are two kinds of seats: the Highway and Hinterland seats. The Highway seats are those that are in urban areas or along or near the main roads. They are accessible and citizens have greater access to information. Holding rallies in these Highway seats boosts morale but it can also be deceptive. That is where the other type of seat comes in.

· The Hinterland seats are constituencies that are in places that are far and often hard to access due to poor infrastructure. People in these places have limited access to information and are usually ill-informed. But there are thousands of people there who also vote, even if they only see government officials during election time. But as we shall see in the future analyses. If the opposition is to overcome the ruling party’s advantages, it should balance its presence in the Highway seats with increased presence and activity in the Hinterland seats.

· Since you are not going to win everything, you must concentrate on maximizing the advantages in your strong areas and minimizing the opponent’s advantages in your weaker areas. For example, if you were ZANU PF, you would know that you are unlikely to win Dangamvura/Chikanga, so your best bet is to try to minimize the gap. Likewise, the CCC would realize that Chimanimani East is a weak constituency so the aim here must be to limit the gap.

· However, subject to the qualification over the authenticity of figures, Manicaland also shows that overall, it is a swing constituency. The gap between the major candidates is small and the figures on either side are finely balanced. This makes Manicaland a battleground province, one where each side must strategize carefully to reap maximum benefits.

· Finally, a major theme that is present across the country, the opposition parties tend to gift ZANU PF cheap seats in the parliamentary race due to split voting. We shall do a special BSR dedicated to all seats that have been lost by the opposition just because they fielded double candidates or opposition parties fielded candidates where they could have established an electoral pact. Manicaland had 4 such seats which were gifted to ZANU PF, contributing significantly to its gaining a two-thirds majority which gives it the untrammeled power to amend the constitution, just like it did in 2019.