Electoral Corruption and Implications for Development in Kenya
Author: Jon Helfers, 2018.
“Looking at the Afrobarometer data and the Gallup polls, this paper shows that Kenyans lost significant trust in the courts and electoral commission after the disputed 2007 elections.”
Image from Gallup Poll.
The general relationship between corruption and economic development has been a focus of both the international development community as well as academics for some time. However, less attention has been paid to the specific links between electoral corruption and its impact on national economies. Utilizing Afrobarometer data and Gallup polls, this paper seeks to determine if the disputed 2007 elections corresponded with a dip in the Kenyan economy. Analysis shows that Kenyans lost trust in key democratic institutions after the election and while longer economic trends appear positive, electoral corruption corresponds with a short-term negative economic impact. The findings provide initial evidence that electoral corruption can have a pernicious effect on short-term economic growth; and while more research is needed, if political corruption is not addressed, long term economic trends could be negatively affected.
The academic literature is somewhat divided on the implications of corruption on economic development. Much of the micro level data supports theories that suggest corruption leads to inefficiencies with the allocation of resources and production (Svenssen 31-32; Murphy 503). However, Samuel Huntington and others have argued the macro effects of corruption allow firms to circumvent bureaucratic systems that suppress growth (Leff 8; Huntington 579). Jakob Svenssen suggests that some of the disagreement may be due to the difficulty of untangling cross-country effects of corruption on growth when different forms of corruption plausibly have varying impacts on growth, depending on the context (21-22). Thus, it is important to expand our knowledge about through what mechanisms particular types of corruption may impact development in particular contexts.
In the classic Capitalism, Socialism, and Democracy, Joseph Schumpeter defines democracy as a political regime that is characterized by free and contested elections in which rulers are selected, and incumbents leave office after they lose (269-283). Incumbents have significant advantages in elections and leaders are rarely voted out of office. Gandhi and Przeworski found that only one in five incumbents are defeated in national elections (2). Of course, this advantage can be for several reasons. It can be legitimate, in that constituents favor a leader who is performing well. Or, an incumbent can use the tools and resources of office to ensure s/he remains in power even if it is against the will of the people. When leaders manipulate the democratic process in this manner, it is considered political corruption (Fackler and Lin 3).
Bianca Clausen and her co-authors find in a review of the Gallup World Poll that there is a significant negative correlation between corruption and citizens’ trust in public institutions (212). In reviewing 2008 Afrobarometer data, Emmanuelle Lavallée and her co-authors provide evidence to reject the “efficient grease” argument, which posits that corruption increases trust by providing easier access to public services (16). Their analysis concludes that corruption lowers trust in political institutions but that it varies by institution (16). Clausen finds that lower trust in public institutions hinders economic development but does not expound on through which causal mechanisms this might occur; however, other strains of development literature have investigated this more thoroughly (212).
A well-established assumption in development literature is that heterogeneity within communities can hinder development (Alesina et al. 1243; Banerjee et al. 639; Easterly and Levin 1203; Gören 275). There is some evidence showing that fractionalized communities have challenges overcoming collective action problems and funding public services compared to communities that are more homogenous (Miguel and Gugerty 2326-2326; Habyarimana et al. 709). At the same time, there is some debate on when ethnic identity triggers negative economic effects. Edward Miguel and his co-authors investigated this subject in a lab experiment and found that coethnics exhibit no greater altruism and favoritism toward each other than non-coethnics (Berge et al.). However, Jonas Hjort found in a naturally randomized setting that workers discriminated against non-coethnics and this discrimination increased in times of high ethnic political tension (1).
In a study of road construction in Kenya, Robin Burgess and his co-authors find strong evidence of ethnic favoritism in the construction of paved roads, whereby regions loyal to the president receive twice the expenditure as other regions; however, this favoritism reduces in times of increased democracy (1817). Burgess’ findings imply that while ethnic heterogeneity within society may have negative economic impacts – particularly in times of ethnic political tension, some of these effects can be mitigated through an inclusive democratic process. However, the impact of electoral corruption may be a hindrance to positive force of democracy by undermining unity within a nation, whereby resources are not divided on the bases of efficiency to increase development but along ethnic lines to maintain control. If this is true, the literature suggests that one would expect to see electoral corruption followed by a loss of trust in institutions and increased emphasis on ethnic identity and decreased economic growth.
In the following case study, suggestive evidence is provided, using Afrobarometer and the Gallup poll data, that electoral corruption reduces trust in democratic institutions in Kenya. This analysis strengthens the argument that corruption – and more specifically electoral corruption – impedes the trust of citizens in a nation’s government, weakening its legitimacy. The evidence also suggests a clear short-term impact on the economic activity after citizens perceived that the 2007 election was not free and fair. In combination, the low trust and direct negative impact on short-term economic activity should raise concerns for longer term development.
A Case Study: Kenya
Prior to its independence in 1936, Kenya held its first election open to nonwhites in which Jomo Kenyatta, leader of the Kenya African National Union (KANU), swept the Legislative Council, forming the state’s first government. However, it was not until 1964 that Kenya was declared a republic and Kenyatta became the first president. After independence Kenyatta consolidated power and Kenya became a one-party state (“The World Fact Book: Kenya”). Following the first election, politics were largely segregated along ethnic lines. Under one-party rule, ethnic divisions were utilized to keep KANU in power (Branch). Following Kenyatta’s death in 1978, his Vice President Daniel Arap Moi transitioned into power, which he retained for another twenty-four years—fourteen of which were under single-party rule. Moi did win multi-party elections in 1992 and 1997, but both were marred by election violence and electoral fraud (“The World Fact Book: Kenya”).
Arguably, in 2002, following Moi’s retirement, the first genuinely free and fair democratic elections were held. Mwai Kibaki won under a multi-ethnic coalition. By 2007, the coalition had splintered, and former coalition partner Raila Odinga challenged Kibaki for the presidency in 2007. The Independent newspaper, along with others, concluded that “systematic elector fraud including vote-rigging in a third of all constituencies, stuffed ballot boxes, and election officials changing results had decisive impact on the outcome”. Nonetheless, the Election Commission of Kenya announced the incumbent Mwai Kibaki the winner (Bloomfield; “Elections in Kenya in 2007.” 2-3; Kriegler et al. ix). Following the announcement of the results, election violence broke out in several regions of the country. Over 1,200 people lost their lives, and more than 350,000 people were displaced (“Elections in Kenya in 2007” 3).
Perceptions of Electoral Corruption
Similar to the conclusion of observers and the Independent Electoral Review Commission, many Kenyans suspected the incumbent, Mwai Kibaki, had calculated that it was better to tip the scales in his favor than to leave office. This paper does not explore the evidence for electoral corruption but rather examines the impact of the public’s perception that the election was not free and fair. In June-July 2008, the Gallup poll found that 70% of Kenyans reported they did not believe the election was fair (Rheault & Tortora). This is a dramatic drop from just two years prior when 63% of Kenyans reported to have confidence in the fairness of elections (Rheault & Tortora). As Figure 1 from the Gallup survey shows, Kenyans’ confidence in the integrity of elections took a precipitous drop after the 2007 disputed elections, as well as a parallel decrease in confidence among respondents that the Electoral Commission of Kenya was autonomous. This suggests that Kenyans did not believe the Commission was independent of the Kibaki administration and thus able to be an independent arbiter of the will of the people.
*Source: Gallup Poll
When the results of the survey were disaggregated by ethnicity, the contrast between those who saw the election as honest compared to those who did not split along ethnic lines. Sixty-seven percent of Kikuyu, Kabaki’s ethnic support base, believed the elections to have been fair while the other major ethnic groups did not. As seen in Figure 2, it is also notable that those ethnic groups who traditionally supported the opposition candidate Raila Odinga, a Luo, more likely to respond that it was not honest (95% of Luo and 92% of Luhya).
*Source: Gallup Poll
Kenyans’ perceptions that the 2007-08 elections were not free and fair also had implications for the institutions who were mandated to conduct the polls and protect the results. Key democratic institutions such as the electoral commission and courts lost the trust of the Kenyan people after the flawed election. One way to see this is to compare the level of trust in those institutions over time. Below is such a comparison using three rounds of public perception data.
The data for the following analysis comes from Afrobarometer: “Let the People Have a Say”. Afrobarometer is a non-partisan research network that conducts public attitude surveys in 35 countries throughout the African continent (Afrobarometer: Let the People Have a Say: About”). The surveys focus on democracy, governance, and economic conditions. Afrobarometer has now collected six rounds of surveys starting in 2000, with the most recent conducted in 2016-17. This analysis makes use of comparing a representative sample over time, specifically the levels of trust Kenyans had in key institutions prior to the 2007-08 disputed election in Kenya and those opinions taken just after the elections in a follow-on survey 2011. Afrobarometer surveys are nationally representative, random and stratified probability samples meaning that results should reflect the attitudes of the population at large (Afrobarometer 2005; Afrobarometer 2008; Afrobarometer 2011).
Data Analysis: Comparing Levels of Trust
For this comparison, this paper make use of a subset of questions Afrobarometer asks respondents about trust. Every respondent is asked: “How much do you trust each of the following, or haven’t you heard enough about them to say: [insert institution]” (Afrobarometer 2005; Afrobarometer 2008; Afrobarometer 2011). The institutions of interest in this analysis are the President, Parliament, the Independent Electoral Commission and courts of law. The respondents are provided with a Leichardt scale of: a lot, somewhat, just a little or not at all. This paper aggregates the responses into two categories for easy comparison of trust and don’t trust.
In a comparison that utilizes a sample of the population over time, it is important to keep in mind that larger trends inevitably impact results; thus, dips may not be permanent but rather reflect the attitudes of individuals at that time. Inevitably larger trends continue unless reversed by the event under analysis. In the Afrobarometer data, this is apparent when looking at the data over a longer time frame, which shows that Kenyans are increasingly more likely to trust their government. This is a positive sign for development. However, the point of this analysis is not to shed light on the larger trends which undoubtedly have to do with macro-economic and social changes that are happening in Kenya; but rather focus on a significant dip in the trust of Kenyans after the disputed elections of 2007-08 and consider its impact of the legitimacy on key democratic institutions.
In Table 1, the comparison is broken down by year and institution. The survey prior to the disputed Kenyan election in 2007-08 is captured in the 2005 data. The 2008 survey was conducted ten months after the disputed election, and the following survey in 2011 was three years later. Several things become apparent upon looking at the data in this manner.
First, it appears the levels of trust in elected representatives did not drop as much as other institutions. For example, the level of trust in the presidency did not decrease significantly from the 2005 to 2008 surveys (4 percentage points); however, those that “don’t trust” the president did increase slightly more (6 percentage points). There could be a couple of explanations for this. One explanation is that in the wake of the violence and by the time of the survey, a unity government had been formed. The incumbent Mwai Kibaki retained the presidency but a new position of Prime Minister was created for the opposition leader Rialia Odinga. The fact that the incumbent compromised and ended the violence by joining this unity government may have engender goodwill in the population. Thus, any dip in the level of trust was not captured due to the timing of the survey. Another explanation could be that given the president’s base of support was largely ethnic, as discussed previously, then the disputed election may entrench pre-existing levels of trust. This seems less likely than the first reason since there is good evidence that the incumbent, Kibaki, lost the election and would thus have had less than 50% of Kenyans supporting him (Bloomfield). The 56% of Kenyans who said they trust the President in the Afrobarometer survey in 2008, eleven months after the disputed election, are thus likely being influenced by the events that resolved the crisis. Either way, the data suggests that citizens’ trust in the President did not as significantly change eleven months after the disputed election in contrast to other institutions discussed below.
Second, similar to the President, citizens’ trust in Parliament stayed relatively consistent across all three years (46%, 48% and 49% respectively). It is possible that respondents inferred this question to be asking about trust in their local member of parliament rather than the Parliament as a whole. If representation falls along ethnic lines, then trust in their representative may not change due to strong partisan bias. It could also be that since the level of trust is relatively low to begin with, levels do not fluctuate significantly over time.
Third, in stark contrast to the President and Parliament, the trust in the Independent Electoral Commission dropped significantly (28% decrease). This is not entirely surprising considering the previously-discussed Gallup poll found that Kenyans perceived the Electoral Commission as less autonomous after the disputed election and did not have confidence in the results of the election. Additionally, it is logical that trust would be lost in the institution whose mandate it is to conduct the election if that election is largely seen as to not reflect the genuine will of the people. A positive sign is that the loss of trust was not permanent. This temporary impact was possibly caused by a national referendum for devolution in 2010, which also reformed the electoral commission (“Our Mandate”). The reforms in the new constitution could have engendered trust in the institution or devolution could have been seen as the solution to the disputed election by decentralizing power to new county governments.
Last, it is striking how much trust citizens lost in the courts after the disputed election (13 percentage points). The courts played no active role in the disputed 2007-08 elections or resolution. The opposition candidate Raila Odinga decided not to take the dispute to the courts (Gettleman). One potential explanation is that citizens agreed with the Odinga’s justification and felt the court—similar to the Electoral Commission—was not independent. This lack of independence could be then correlated with citizens’ level of trust in the institution.
*Source: Aggregation of Afrobarometer Data from 2005-2011 utilizing questions about trust.
Data from Gallup polls corroborate the Afrobarometer data. As can be seen in Figure 3, trust in the courts fell after the disputed election to a low of 27% in 2009, until it reversed course in a 2011 survey (Lyons). As mentioned, this is likely due in part to the unity government that was formed after the election violence ended, but also because the court was given new independence in the constitution that passed by referendum in 2010. One additional observation is that the general opinion toward the Kenyan national government tracked very closely to that of the court and decreased from 66% to 24% confidence in all institutions in 2009 (Lyons). This suggests that when Kenyans reflected on all institutions within the government, the disputed election resulted in a significant overall loss of confidence. However, this does not show up in the more specific institutions discussed individually above. One possible explanation is that partisan bias led respondents to retain trust in the President and Parliament, but this did not transpose to the government as a whole.
*Source: Gallup Poll
Discussion & Further Research
The Afrobarometer data and Gallup polls show that Kenyans lost trust in the Electoral Commission and courts following the 2007-08 disputed elections. The Gallup polls further suggest that this loss of faith was more generally ascribed to a loss of confidence in the national government. However, it is also evident that this loss of faith was not permanent. As suggested, fundamental reforms and successful elections reversed the negative trends. In many ways, this is positive. Electoral corruption can weaken trust in key democratic institutions, but those trends are not permanent. Considering this, several questions arise for further deliberation that the Afrobarometer data can shed some light on in conjunction with economic data.
First, more focus should be given to how electoral corruption impacts economic development. As discussed in the literature review, the impact of corruption on economic growth is somewhat disputed. Micro-level data seems to suggest that there is a negative correlation (Svenssen 31-32; Murphy 503). The evidence from economic growth after the 2007-08 disputed elections in Kenya would appear to concur with the micro evidence and dispute Huntington’s macro arguments. As can be seen in Figure 4 below, real GDP growth in Kenya fell substantially in 2008 (Turner). It should be noted that real GDP growth declined in Africa and much of the world due to the global financial crisis; however, the depth and timing shows clearly that the disputed elections had a significant part to play in decline in percent growth.
As can be seen in Figure 4, Kenya’s decline in percent real GDP growth was both earlier and more substantial than the African continent’s average. From 2007 to 2008, Kenya’s real GDP growth had fallen from 7% to 1.5% while the average for contentment dropped less than a percent from 6.4% to 5.6%. However, by 2009, Kenya was beginning to recover with a real GDP growth rate of 2.6% while the continental average had fallen to just above 3%. This is likely a result of the global financial crisis. This does raise questions about the direct relationship between electoral corruption and real GDP growth. Further research could explore this relationship in a larger sample and might be able to provide additional insights.
* Source: Euromoney
Second, there should be concern that electoral corruption could lead to further ethnic divisions, thereby hindering service delivery and economic development. As discussed in the literature review, Miguel, Hjort, and others have shown that ethnic fractionalization in a weak democracy can lead to unequal and ethnically partisan use of state resources for development (Miguel, 2325; Habyarimana et al., 709; Hjort 1; Burgess et al. 1817). The Afrobarometer survey provides some evidence to investigate this question, but ultimately, it is inconclusive.
The variance of Kenyans’ level of trust for other ethnic groups has not significantly changed over time when asked the question: “Would you say that most people can be trusted or that you must be very careful in dealing with people?” (Afrobarometer 2005; Afrobarometer 2008; Afrobarometer 2011). Notably, the level of trust has consistently been very low – around 10% trust in others. However, when primed if their trust in other ethnic groups had changed since the election violence, 45% of respondents said it had decreased or decreased a lot. The incongruences could be due to the low starting point of trust in the general question and the priming effect of the second. It is also in contrast to the larger trend shown in Table 2. The results in this table suggest that Kenyans are feeling more nationalistic than ethnic over time.
*Source: Aggregated of Afrobarometer Data from 2005-2011
Another explanation for the difference is that other factors are driving a greater sense of nationalism within Kenya and the disputed election dampened that trend but did not reverse it. Ultimately, this is an area for further research to better untangle causes of the longer positive trends and the shorter negative trends caused by electoral corruption.
Last, the analysis provided may raise some questions. In particular, some may question if it was electoral corruption or violence that led to the loss of trust. Undoubtedly, a combination of factors intertwined with one another is likely to affect citizens’ confidence in the government. The failure of the government to protect its people is likely to be correlated with a loss of faith. However, the connections of trust would logically follow lines of responsibility. Citizens are likely to hold an electoral commission responsible for their mandate to provide free and fair elections. This will then combine with other institutional failures, such as the police, to maintain peace, resulting in the loss of confidence in the government as a whole, as shown in the Gallup poll.
Additionally, some may argue that since the impacts were short-term, trust in key institutions is not a significant concern. While the overall economic trends are positive, as argued earlier, this was likely a result of macro trends and reforms put in place after the disputed election. It is very likely that those positive trends would be even greater and more beneficial if electoral corruption had not hampered them in 2007-08. Further, as the most recent Kenyan elections in 2017 have shown, electoral fraud continues to be a significant problem. The Supreme Court of Kenya overturned the 2017 presidential results in a landmark ruling (Freytas-Tamura). The court ruled that the election results could not stand due to irregularities in the election process and a failure of the electoral commission to comply with the constitution (Freytas-Tamura). The ruling provides additional evidence that the trend of electoral corruption continues and the evidence presented in the papers suggest that the economic impacts will be negative. The outstanding question is, will the continued electoral corruption start to outweigh the positive macro trends Kenya has enjoyed over the last few years?
Corruption has been a focus of development for some time; however, less attention has been paid to links between electoral corruption and economic development. Above it is shown that in the context of Kenya, electoral corruption has weakened trust in two key democratic institutions. Looking at the Afrobarometer data and the Gallup polls, this paper shows that Kenyans lost significant trust in the courts and electoral commission after the disputed 2007 elections. Fundamental reforms and larger trends appear positive, but there is still cause for concern since electoral corruption continues to be a problem in Kenya. Although more research is needed to understand the relationship between ethnic divisions and economic development, the evidence provided suggests that electoral corruption has had a negative effect on ethnic divisions and short-term economic growth.
“Afrobarometer: Let the People Have a Say: About.” Afrobarometer, 2017, www.afrobarometer.org/about. Accessed 14 December 2017.
Alesina, Alberto, et al. “Public Goods and Ethnic Divisions.” Quarterly Journal of Economics, No. 4, 1999, pp. 1243-1284. https://www-jstor-org.proxy.library.nyu.edu/stable/2586963, Accessed 5 October 2018.
Bloomfield, Steve. “Kibaki ‘stole’ Kenyan election through vote-rigging and fraud”. Independent. 23 Jan. 2008. www.independent.co.uk/news/world/africa/kibaki-stole- kenyan-election-through-vote-rigging-and-fraud-772349.html, Accessed 5 October 2018.
Banerjee, Abhijit, et al. “History, Social Divisions, and Public Goods in Rural India.” Journal of the European Economic Association, Vol. 3, No. 2/3 2005, pp. 639-647. http://www.jstor.org/stable/40005006, Accessed 5 October 2018.
Branch, Daniel. Kenya: Between Hope and Despair, 1963-2011. Yale University Press, 2011.
Berge, Lars, et al. “Measuring ethnic preferences: New experimental evidence”. Vox CEPR’s Policy Portal. www.voxeu.org/article/measuring-ethnic-preferences-new-experimental- evidence, Accessed 14 December 2017.
Clausen, Bianca, et al. “Corruption and Confidence in Public Institutions: Evidence from a Global Survey”. The World Bank Economic Review, 25(2), 212-249. www.jstor.org/stable/23029751, Accessed 5 October 2018.
Easterly, William and Levine, Ross. “Africa’s Growth Tragedy: Policies and Ethnic Divisions”, The Quarterly Journal of Economics, 112(4), 1203-1250. www.jstor.org/stable/2951270, Accessed 5 October 2018.
“Elections in Kenya in 2007.”, DFID UK AID, www.gov.uk/government/uploads/system/uploads/attachment_data/file/67654/elections- ke-2007.pdf, Accessed 5 October 2018.
Fackler, Tim and Lin Tse-min. “Political Corruption and Presidential Elections, 1929-1992”, The Journal of Politics. 57 (4): 971-993, 1995. www.jstor.org/stable/2960398, Accessed 8 October 2018.
Freytas-Tamura, Kimiko. “Kenya Supreme Court Nullifies Presidential Election.” New York Times. 01 Sep. 2017. www.nytimes.com/2017/09/01/world/africa/kenya-election- kenyatta-odinga.html, Accessed 5 October 2018.
Gandhi, Jennifer and Przeworski, Adam. “Holding onto Power by Any Means? The Origins of Competitive Elections.” Working Paper. http://cowles.yale.edu/sites/default/files/files/conf/2009/rat_przeworski.pdf, Accessed 5 October 2018.
Gettleman, Jeffrey. “Disputed Vote Plunges Kenya Into Bloodshed.” New York Times. 31 Dec. 2017, www.nytimes.com/2007/12/31/world/africa/31kenya.html, Accessed 5 October 2018.
Gören, Erkan. “How Ethnic Diversity Affects Economic Growth.” World Development, 59275- 297. http://dx.doi.org/10.1016/j.worlddev.2014.01.012, Accessed 5 October 2018.
Habyarimana, James, et al. “Why Does Ethnic Diversity Undermine Public Goods Provision?” The American Political Science Review, 101(4), 709-725. www.jstor.org/stable/27644480, Accessed 5 October 2018.
Hjort, Jonas. “Ethnic Divisions and Production in Firms.” Quarterly Journal Of Economics, 129(4), 1899-1946. http://proxy.library.nyu.edu/login?url=http://search.ebscohost.com/login.aspx?direct=true &db=bth&AN=101035898&site=eds-live, Accessed 5 October 2018.
Huntington, Samuel P. Political Order in Changing Societies. New Haven: Yale University Press, 1968.
Kriegler, Johann, et al. “Report of the Independent Review Commission on the General Elections” Independent Review Commission. 17 Sep. 2008. www.aceproject.org/regions- en/countries-and-territories/KE/reports/independent-review-commission-on-the-general, Accessed 5 October 2018.
Lavallee, Emmanuelle, et al. “Corruption and Trust in Political Institutions in Sub-Saharan Africa.” Afrobarometer Working Papers. 2018. www.afrobarometer.org/sites/default/files/publications/Working%20paper/AfropaperNo1 02.pdf, Accessed 5 October 2018.
Leff, Nathanial H. “Economic Development through Bureaucratic Corruption.” American Behavioral Scientist. 1964. 82:2, pp. 337– 41. http://journals.sagepub.com.proxy.library.nyu.edu/doi/pdf/10.1177/00027642640080030 3, Accessed 5 October 2018.
Lyons, Linda. “Election Crisis Tests Kenyans’ Confidence”. Gallup, 2017. https://news.gallup.com/poll/221036/election-crisis-tests-kenyans-confidence.aspx, Accessed 5 October 2018.
Miguel, Edward. “Tribe or Nation? Nation Building and Public Goods in Kenya versus Tanzania.” World Politics, 56(3), 327-362, 2011. www.jstor.org/stable/25054263, Accessed 5 October 2018.
Miguel, Edward and Gugerty, Mary. K. “Ethnic diversity, social sanctions, and public goods in Kenya”. Journal Of Public Economics, 892325-2368, 2005 www.emiguel.econ.berkeley.edu/assets/miguel_research/42/_Paper__Ethnic_Diversity__ Social_Sanctions__and_Public_Goods_in_Kenya.pdf, Accessed 5 October 2018.
Murphy, Kevin, et al. “The Allocation of Talent: Implications for Growth.” The Quarterly Journal of Economics, 106(2), 503-530, 1991. www.jstor.org/stable/2937945, Accessed 5 October 2018.
“Our Mandate.” Independent Electoral and Boundaries Commission, 2018. www.iebc.or.ke/iebc/?mandate, Accessed 5 October 2018.
Rheault, Magali and Tortora, Bob. “In Kenya, Most Ethnic Groups Distrust 2007 Election.” Gallup. 2008. http://news.gallup.com/poll/111622/kenya-most-ethnic-groups-distrust- 2007-election.aspx, Accessed 5 October 2018.
Schumpeter, Joseph A. Capitalism, Socialism, and Democracy. George Allen & Unwin: London, 1942.
“Summary of Results Round 3: Afrobarometer Survey in Kenya.” Afrobarometer, 2005, www.afrobarometer.org/publications/kenya-round-3-summary-results-2005, Accessed 14 December 2017.
“Summary of Results Round 4: Afrobarometer Survey in Kenya.” Afrobarometer, 2008, www.afrobarometer.org/publications/kenya-round-3-summary-results-2005, Accessed 14 December 2017.
“Summary of Results Round 5: Afrobarometer Survey in Kenya.” Afrobarometer, 2011, www.afrobarometer.org/publications/kenya-round-3-summary-results-2005, Accessed 14 December 2017.
Svensson, Jakob. “Eight Questions about Corruption.” The Journal of Economic Perspectives, 19(3), 19-42, 2005. https://www-jstor-org.proxy.library.nyu.edu/stable/4134971?, Accessed 5 October 20182018.
“The World Factbook: Kenya.”, CIA, https://www.cia.gov/library/publications/the-world- factbook/geos/ke.html, Accessed 5 October 2018.
Turner, Matthew. “Kenya risk outlook improves after election.” Euromoney, 2013. www.euromoney.com/article/b12kjswld4zr77/kenya-risk-outlook-improves-after- election, Accessed 5 October 2018.