Two million people unregistered


HARARE – The Research and Advocacy Unit, RAU’s analysis of the voters’ roll has been facilitated by the release of the preliminary results of the 2012 census.

However, these results do not disclose the statistic of prime importance here, the number of adults in Zimbabwe as indicated by ages of 18 years or more.

The analysis has thus proceeded on the basis of the percentages of the population in each age band supplied by Zimstat (the body which conducts the census) and used by Zimstat to calculate data in a 2012 survey on Health and Demographics in Zimbabwe.

Zimstat is of the view that these percentages have not changed significantly in the few years since that survey, a view supported by the fact that the census shows that Zimbabwe has a low growth rate of 1,1 percent.

The number of adults (and thus potentially eligible voters) in the 15 to 19 age band has been determined by calculating two-fifths of the total. It is believed that the number thus arrived at is sufficiently accurate for present purposes.

The number of adults in each age band according to the census, compared with the number of  people in those age bands registered as voters. In an ideal situation, all those who ought to be registered as voters are, and those who ought not to be, such as  those deceased, are not.

Although the Zimbabwe Electoral Commission took special steps to facilitate registration before the production of the June roll analysed here, and although nearly 300 000 deceased people were reported to have been removed from the roll in the months immediately preceding its compilation, showing a large disjuncture between the ideal and the actual.

Very few adults aged less than 30 years are registered.

This is most marked in the 18-19 age band, where only eight percent are registered. In numerical terms, this means that a total of 1 920 424 people under the age of 30 ought to be registered as voters but are not.

This is almost 29 percent of the total adult population of 6 647 779.

Since there are unregistered people in the other age bands, the total percentage of the entire adult population who ought to be registered as voters but are not, is considerably higher than 29 percent.

The “under-registration” in these other age bands is concealed by “over-registration” in all of the age bands from 30 years and above.

If the under-registration in these bands were to be taken into account, as it ought to be,  the  percentages of over-registration would increase.

It should also be noted that the over-registration figures assume a 100 percent registration rate, which is improbable.

The percentages of over-registration will increase in direct proportion to the extent that the registration rate is less than 100 percent.

Even without considering these factors, however, it indicates that the number of registered voters exceeds the population of Zimbabwe, in all age bands of 30 and above, by a considerable margin.

This is most notable in the 40 to 44 age group (162 percent) and the 80+ age group (219 percent).

In the former, some of the over-registration may be  explained by people who are registered as voters,  but  were  absent  from  the  country  during  the  census  count,  that  is,  are  part  of Zimbabwe’s Diaspora.

The latter group most probably represents a large number of  deceased persons who names still appear on the voters’ roll.

In numerical terms the “over-registration” of people aged 30+ against the actual population is 1 146 760.

This conclusion is further supported by the fact that there is an unlikely 116 195 people aged over 100 still on the roll. Constituency “Over-Registration”

The problem of over-registration can also be considered by constituency. The table indicates data for each province  with the number of constituencies in each province where the number of registered voters exceeds the population in those constituencies.

It thus indicates that in 63 (30 percent) of Zimbabwe’s 210 constituencies, the number of registered voters exceeds the total population there, as indicated by the census.


These  problems  of  “over”  and  “under”  registration  obviously  affect  the  delimitation  of constituencies,  since this is based upon the number of registered voters in each constituency.

Delimitation will be inaccurate to the extent that the voters’ roll incorrectly reflects the number of voters in each constituency. Constituencies were last delimited ahead of the 2008 elections.

There is to be no new delimitation ahead of the 2013 elections  despite extensive demographic changes and changes to the voters’ roll since 2008.

Both the old and the new Constitution provide that the number of registered voters in each constituency should, as far as is possible, be equal, thus ensuring that no vote is of greater or less value than another in elections  for the National Assembly.

The number of registered voters should thus be spread evenly across the 210 constituencies, making a mean of 27 972 voters per constituency.

Both constitutions, however, allow a variation of up to 20 percent more or less than the average number of voters registered in each constituency.

The permissible (though undesirable) variation of 20 percent has the result that the maximum and minimum number of voters in  each constituency should not exceed 33 566 registered voters per constituency or be less than 22 378.

The distribution of voters on the June 2013 roll is such that 23 constituencies exceed the permissible maximum and 17 are below the permissible minimum. Mbare has a voter population 154 percent greater than the mean (43 169 voters) and Chipinge East is has a voter population of only 46 percent of the mean (12 940 voters).


The voters’ roll shows no gender bias when considered as a whole. The picture may, however, change when  gender is considered across each age band, though this analysis has yet to be undertaken.


The compilation of the voters’ roll is always a moving target. Old voters die and new voters are added. The voters’ roll can thus never meet the ideal indicated at the outset. However, the gap between the ideal and the actual should be kept within reasonable limits if the electoral process is to be deemed fair and credible. The statistics presented here show:

1.  That there are nearly 2 000 000 potential voters aged under 30 who are unregistered.
2.  That there are well over 1 000 000 people on the roll who are either deceased or departed.
3.  That 63 constituencies have more registered voters than inhabitants.
4.  That 40 constituencies deviate from the average number of voters per constituency by more than the permitted 20 percent.

Such statistics suggest that the gap between the ideal and the actual impinges upon the integrity of Zimbabwe’s electoral process.

Comments are closed.