Bringing the citizen back in

Bringing the citizen back in

From last week From last week 
Since the items cannot form a reliable composite scale, as they are weakly correlated, I estimate separate ordered logit regression models  with each of them as a dependent variable. I expect the results from each of the models to be similar since they represent the same phenomenon — ‘demand for accountability’. I begin the modeling process with the following five control variables: age (mean cantered), the quadratic term of age, gender, urban-rural division and ecological zone (see Table A1 in appendix). There are no statistically significant age, gender and rural-urban differences in support for vertical accountability. However, Basotho females, older people and rural residents are less likely to express support for democracy.

Surprisingly, for a small  and culturally homogenous country, the propensity to demand accountability varies with region (i.e. ecological zone). For example, residents of the Maluti Mountains’ ecological zone are eleven percentage points less likely than lowlands dwellers to approve the statement that democracy is always preferable to any other form of government. The fully specified models include the main explanatory variables of internal political efficacy, news media use, political interest, ethic of democratic citizenship, education attainment, perceptions of corruption, a composite index of self-reported poverty (lived poverty) and relative living condition. Table A2 (in appendix) provides the exact wording of the items used to measure these variables. Since some of the variables violate the parallel odds assumption underpinning the ordered logit model, I switch to the generalised ordered logit technique.

This preserves the ‘ordinal nature’ of the dependent variable while relaxing the parallel odds assumption for those variables that violate it and imposing it where it holds (Williams, 2016). The generalised ordered logit technique can reveal variations in the dependent variable that cannot be captured when the proportional odds assumption is imposed where it doesn’t hold. To simplify the model, I merge the answering options relating to statement 1 of the demand for accountability measure (i.e. I combine options ‘agree very strongly agree’ with ‘agree’). The results of the two models (models 1 and 2) are shown in Table 1. Consistent with the theoretical expectations outlined in section two, increasing political interest increases the probability of approving the statement that it is more important for Basotho to be able to hold government accountable and that democracy is always preferable.

However, respondents’ ability to define the concept of democracy, a measure of political knowledge, increases support for democracy, but decreases support for vertical accountability. This inconsistency can also be observed in the effect of the positive change in education attainment, which increases support for democracy but has no effect on support for vertical accountability. This implies that, even though ‘cognitively sophisticated’ individuals profess support for ‘democracy’, they are unprepared to offer an unconditional support for citizen’s involvement in the decision-making process of the state.This is an interesting finding, given the premium put on formal education as the driver of democratic accountability, as illustrated in this chapter’s second section. But one would point out that much of the literature summarised in this section is based on studies conducted in western countries.

Mattes and Mughogho (2010) found, contrary to such studies, that educated Africans were not any likelier than their less educated counterparts to participate in governance.The ethic of democratic citizenship has a strong and statistically significant effect on support for democracy and support for vertical accountability. On an all other things being equal basis, respondents who feel that a good citizen in a democracy should always complain about poor public services are likely to choose higher scores on both measures of demand for accountability as shown in Figure 2.

The probability to select the middle and lowest scores declines steadily as the ethic of democratic citizenship changes from the lowest to the highest values.On the one hand, as the ethic of citizenship changes from its minimum to maximum values, the probability of choosing to agree strongly with the statement that it is more important for citizens to hold government accountable increases by 0.121. On the other hand, the predicted probability to agree and strongly agree that it is more important to have a government that can get things done, even if people have no influence decreases by 0.115 as the ethic of citizenship changes from its minimum to the maximumSimilarly, the probability of saying that democracy is always preferable increases by 0.112 as the ethic of citizenship changes from minimum to maximum values.

The predicted probability to identify with the lowest category, ‘democracy doesn’t matter’ decreases by 0.084 as the ethic of citizenship changes from its minimum to maximum values.A unit increase in the perception that corruption is increasing reduces the propensity to demand accountability. As shown in panel 1 of Figure 3, the probability of approving the statement that ‘it is more important for citizens to be able to hold government accountable’ is 57% when a respondent thinks that corruption has decreased somewhat, and 44% when he thinks it has increased a lot. Similarly, there is a 12 percentage point decrease in the probability to approve the statement that democracy is always preferable when the perception of corruption changes from ‘decreased somewhat’ to ‘increased a lot’. These results are reasonable if we consider with Della Porta (2000) and Bratton and Mattes (2001) that corruption perceptions can erode citizens’ satisfaction with democracy and weaken citizen’s belief that their vote makes a difference (Bohn 2012, Warren 2015; Seligson, 2002). In developing countries, like Lesotho, where governments fail to provide many of the basic public goods and services, the perception that corruption is increasing can make citizens “wish for a powerful authority — a dictator or saviour — that will cut through the blather of politicians and actually make things work” (Fukuyama, 2014: 84).

Political efficacy has a positive effect on demand for accountability. Holding constant other variables, a unit increase in the measure of efficacy — the extent to which respondents believe that ordinary people can make a difference in the fight against corruption—increases by 20% the probability of very strongly approving the statement that it is important for citizens to hold government accountable.  Although a unit increase in efficacy increases the probability of approving the statement that non-democratic alternatives are sometimes preferable, it has no significant effect on the probability of supporting the idea that ‘democracy is always preferable’. It would seem therefore that the most efficacious Basotho are likely to prefer the middle category of the support for democracy scale while choosing the highest score on support for vertical accountability scale.

There is a statistically significant interaction between the perception that corruption is increasing and efficacy (interaction not included in Table 1). As corruption perceptions become more negative, those with high levels of political efficacy tend to agree very strongly that citizens should be allowed to hold government accountable. On the other hand, a more negative corruption perception reduces the probability of demanding accountability among individuals with low levels of political efficacy. This means that instead of being cynical, those who possess a strong sense of efficacy are likely to demand accountability when they feel that corruption is becoming more of a problem. Lesotho fares badly when compared with its peers on various measures of popular demand for accountability.

Basotho’s largely negative views about democracy are complemented by the voter turnout that has declined by more than 60 percentage points over the past 20 years of Lesotho’s return to democracy. This very weak demand for accountability underpins the high tendency of Lesotho’s politicians to use non-democratic means of conflict resolution. To the extent that positive attitudes towards democracy are not innate, understanding what drives them is a potentially useful venture for policy making and advocacy. Thus, I examined the effects of several individual-level variables on popular demand for accountability using the recent nationally representative survey data on Lesotho.

Two variables — the ethic of democratic citizenship and perceptions of corruption —emerged as strong predictors of demand for accountability. In light of the strong negative effect of perceived levels of corruption, the recent corruption scandals and the combative reaction of some of the government functionaries (see Mohloboli, 2016; Ntsukunyane, 2016), could portend further decline in the demand for accountability. An increased perception that corruption is rampant is likely to be accompanied by an overwhelming sense that it is pointless to oppose it (see Persson et al., 2013). According to anecdotal evidence, ‘corruption’ is one of the main reasons for an overwhelming sense, among the youth, that ordinary Basotho cannot change the status quo, and that political participation is a futile exercise (see Mosooane, 2015).

This view is consistent with the emerging experimental evidence showing that perceptions of corruption reduce voter turnout in developing countries (see Winters et al., 2015). When coupled with a sense that corruption is a widespread, loss of interest in democratic politics by majority of Basotho people may portend further increases in corruption and violent political conduct. To mitigate the negative effects of corruption perceptions, media houses and civil society organisations could improve their democracy education efforts with a view to instilling the ethic of democratic responsibility, stimulating interest in politics and building a sense of efficacy, especially in the rural parts of the country.

The Independent Electoral Commission should also graduate from their short, temporary and election-focused voter education programme to one that is comprehensive and continuous. This comprehensive ‘democracy education’ programme can foster commitment to democracy as a set of governing principles. Those who are committed to democracy as an ideal are guided by the principle of civic responsibility, and are therefore less likely to be overwhelmed and paralysed by repeated episodes of poor governance.

By: Moletsane Monyake

Previous The nightmare on our roads
Next Our African heritage

About author

You might also like

Insight

Kenya: tough road ahead

THE 1 September decision by Kenya’s Supreme Court to annul the results of the 8 August presidential election and order a fresh vote was at once unexpected, historic, bold and

Insight

Political leadership — a cause of instability?

Continued from last week Military Rule, 1986-1993 As LLA began operations against the government of Chief Jonathan, in the late 1970s, the government transformed the PMU from being a unit

Insight 0 Comments

Legalise abortion

Last week, the minister of health Hon Nkaku Kabi declared that the Queen ‘Mamohato Memorial hospital is overcrowded. One of the reasons put forth by the minister as being behind