Does better service delivery result in more votes for the ruling party?

The recently released community survey presents a profile of a country where households have, in 2016, better access to services than at in time in the past. For example the percentage of households with mains electricity increased from 32% in 1996 to 90% in 2016.

In 1996  80% of households had access to piped water in or near their dwelling. By 2016  this figure had increased to 90% of households.

In 1996 52% of households had their household refuse removed at least once a week by local authorities. That percentage rose to 61% in 2016.

Now 79% of households live in formal dwellings. In 1996 only 65% of household did so.

Given that most of  these improvements took place under the auspices of ANC-led national, provincial and local government the improved service levels should bode well for the ANC in the upcoming local government elections.

The essence of electoral democracy is giving citizens an opportunity to select representatives who they think best serve their interests. If the current administration has served the electorate it will presumably return that administration to office – possible with an increased majority.  If the electorate does not so reward the administration (or punish poorly performing administrations)  there is little point in using elections to hold representatives to account.

However, a recent study comparing changes in service delivery levels to the proportion of votes won by the ANC shows that better performance does not necessarily result in that party winning a higher share of votes. (de Kadt and Lieberman. Do citizens reward good service? Voter responses to basic service provision in southern Africa.

Afrobarometer Working Paper 61. October 2015.) de Kadt and Lieberman show, with statistical rigour, that voters do not systematically reward those responsible for good performance with an increased share of the votes.

In fact the study by de Kadt and Liebermann shows the opposite trend – poorer performance results in the ruling party getting a higher share of the votes cast. Conversely the better the improvement in services levels the smaller the share of votes the incumbent (now the ANC) is likely to get at the polls.

de kadt and Liebermann state it as e..

“Voters who receive services may in fact be more likely to punish, and those who receive fewer services are more likely to stick with, the incumbent… By performing poorly the incumbent party is likely to get a greater share of the votes cast.”

This indicates that the improved service levels indicated by the Community Survey of 2016 may contribute to the ANC getting a lower proportion of votes cast in the August election.

While the impact referred to by de Kadt and Liebermann is real and will impact on the election outcome it seems that the electorates’ behaviour is not as perverse as is implied. It seems that de kadt and Liebermann is measuring the impact of disillusionment on election outcomes rather than the impact of service delivery on the incumbents share of votes.

One impact of poor performance is more widespread disillusionment among the electorate and their increased reluctance to vote in elections. It seems that the primary impact of disillusionment is for voters to abstain from voting rather voting for an opposition party.

Disillusionment thus results in a lower turnout and those who do vote are likely to be dedicated supporters of a particular party. This, in general, results in an increasing proportion of votes for the majority party.

When the electorate is more hopeful the participation rate increases. But, in comparison to the dedicated voters, a higher proportion of the voters support parties other than the incumbent.

A higher turnout is thus likely to result in an increase in the total number of votes cast for the incumbent and for other parties. The increased participation rate routinely results in the incumbent getting a lower share of the votes cast.

Factors like improved service delivery and better state performance result in an more engaged electorate and in an increased participation rate. This gives rise to a ‘Catch 22’ situation for the incumbent. A higher turnout arising from better services is also likely to result in a lower share of the votes cast.

There is thus no simple relationship in voting patterns where better service levels result in an increased proportion of votes for the incumbent party. Adult citizens will, on August 3, be confronted with three main options: 1) vote for the incumbent, 2) vote for an opposition party or 3) not vote at all. Generally the ANC (the incumbent party) is likely to fare better if the turnout rate is low.

However a low turnout is also indicative of a disillusioned citizenry and should be taken as a warning sign by the incumbent.  Perhaps the country wins most when improved service delivery result in an engaged electorate and we should hope for this rather than dwell on the share of votes the incumbent gets.


SABC © 2016