Paying the Public or Caring for Constituents? Popular Expectations and Parliamentarians’Expenditure at the Constituency Level in Kenya Transparency International- Kenya June 2003 Paying the Public or Caring for Constituents?

Published by: Friedrich Ebert Stiftung(FES), P.O.Box 14932, Nairobi, Kenya. Peponi Road, Peponi Plaza. Telefax:+254-2 –3748338/9 E-mail: kenia@fes.de and Transparency International- Kenya P.O.Box 198 00200,City Square Nairobi, Kenya. Tel:+254-2 –2730324/5 Fax: 2729530 E-mail: transparency@tikenya.org © Friedrich Ebert Stiftung(FES) © Transparency International- Kenya ISBN: 9966-957-74-X June 2003 ii Paying the Public or Caring for Constituents?

Table of Contents Table of Contents iii Preface iv Foreword v Section One: Introduction 1 General Introduction 1 Objectives 2 Subsidiary Question 2 Guiding Assumptions 3 Justification for the Study 4 Methodology: The Samples 5 Conceptual Approaches 6 Section Two: Findings 10 A. The Two Samples 10 B. Findings: The Qualitative Sample 13 C. Findings: The Quantitative Survey 19 Section Three: Concluding Considerations 27 Paying the Public or Caring for Constituents? iii

Preface Everywhere, efforts to combat corruption in public life have had to recognize the fact that this ill is multi-faceted. Involving upward and downward-linkages of debts and rewards, promises and obligations,formal and informal individual and collective networks,measures to combat the acquisition and use of public resources for private gain must be equally intricate and complex. This is so even when the moral impulse upon which such measures rest is unequivocal. In this connection, it has also been recognised that areas of public life apparently quite far removed from such corrupt practices may, nevertheless, both feed off and support them. This is so even if establishing hard evidence of such behaviour may often constitute a formidable research challenge. Members of Kenya’s Eighth Parliament who participated in this Transparency International-Kenya project did so voluntarily. Regarding such participation, it was clear from the outset – and is demonstrated in the interviews that were conducted with them – that they regret the significant financial strains placed upon them by virtue of their highly visible public roles. This is so not because they object in principle to such‘forced generosity’, but because they acknowledge the deleterious impact of such pressures on their performance as elected officials,both inside the National Assembly in Nairobi and on the ground in their constituencies. And this is so not just before an election, but throughout the five-year period of incumbency. This report,then,raises what are deemed to be key questions regarding such pressures and practices, presents the data obtained that illustrates them, and concludes by returning to the themes presented at the outset in light of these findings. The final conclusions as to how great a public ill such practices constitute,and what measures,if any, should be taken to address them, are left to you, the reader, and indeed, to the Kenyan public at large, to consider.As such,if this report helps to provoke a more informed debate about this aspect of public life, it will have served its purpose. Transparency International – Kenya iv Paying the Public or Caring for Constituents?

Foreword “If you vote for me,I will bring development;I will make sure that all the roads in this constituency are tarmacked, our children get jobs, our dispensaries get working again and that clean water will flow from the taps again.” The above is a typical speech of an aspiring parliamentary candidate on the campaign trail. It raises constituents’hopes and expectations of the ability of the candidate to perform and deliver on his/her promises.On winning the election,all look to him/her to fulfill what he/she promised. Certain questions arise:What is the role of a member of parliament(MP)? Is an MP an implementer or a facilitator? Is he/she a person who“brings”development or one who mobilises resources for development? Is he/she a provider – paying hospital bills,school fees and buying uniforms,etc? Does the MP understand his/her role? Is he/she obliged to be a provider? If he/she is one,where do they get the money from? Does the constituent, on the other hand, understand the MP’s role? If they did, would they be demanding the kind of things they do from them? This book explores expenditure patterns of some members of the eighth parliament and in its findings teases the above questions. The study covered the period just four months before the general elections were held. Of course, the nearness to the election must have influenced their expenditures upward. The findings reflect what the MPs spend their money on as well as how much. We at the Friedrich Ebert Stiftung hope that this book will trigger useful debate on the question of the real and perceived roles of the MP as well as those of their constituents in the development of the nation. Wanjiku Mbugua Programme Manager Friedrich Ebert Stiftung August, 2003 Paying the Public or Caring for Constituents? v

Section ONE: Introduction The legislative function of parliament is under threat from the public’s expectations on[sic] members of parliament. Invariably, MPs are expected to have enough money to contribute to a series of fund raisers[sic] or communal projects, school fees, medical bills, weddings, funerals, etc.They are expected to build roads, hospitals, boreholes, water dam[s], among others[sic].To get elected, most parliamentary aspirants actually use these societal expectations as their campaign platforms.The begging question then is, what should be the role of the government?1 General Introduction This report,‘Paying the Public or Caring for Constituents?’ is a Transparency International-Kenya project initiative, part of a larger endeavour that examines various aspects of public affairs that engender and are impacted upon by corruption. The study takes a look at one aspect of political life that appears particularly conducive to corruption in various forms: how MPs respond to actual and perceived demands placed upon them for individual and collective financial assistance at the constituency level. It should be stressed here that addressing such needs is not deemed to be corrupt in itself. Nor can it be assumed that any of the funds MPs use in doing so necessarily come from corrupt sources. Rather, a more general pair of questions is being asked:(1) If MPs cannot satisfy such demands from their own salaries and savings, where is all this money to come from, and;(2) What would have to occur in order that constituents no longer constantly bombard their elected representatives with such personalised and collective requests in the first place? 1 Kenya’s Parliament- Members’ Participation(1998-2001)’, Institute for Civic Affairs and Development, Nairobi, Print Displays Ltd., 2002. p. iii. However, the actual question here, it would appear, is what is the role of parliament, rather than of“the government” as a whole. 1 Paying the Public or Caring for Constituents?

Objectives Based upon such concerns, the specific objectives of the study were as follows: 1) To establish the amount of money that MPs spend on constituency affairs during the period covered, on a monthly basis, and to identify the principal expenditure-categories; 2) To probe MPs’ own perceptions of the political as well as the financial costs and benefits of such expenditure, and conversely, of failing to make them; 3) To determine the general sources of the funds used for such purposes; 4) To suggest(or at least question) the relationship between such demands and expenditure and the evolving democratic transition in Kenya. Subsidiary Questions To keep in context the data obtained with regard to the above objectives, attention was also paid to several subsidiary questions, even if it was assumed that firm answers to them would likely prove elusive: 1) How large and representative a sample would be necessary to draw any firm conclusions about such issues for MPs in general, at least for those of the Eighth Parliament? 2) To what extent did the timing of the data-gathering affect the results obtained? Would such an exercise undertaken at a different point in the electoral cycle have yielded significantly different data? 3) Given popular(and apparently escalating) demands for such individual and collective financial assistance, can MPs’ official monthly remuneration(of whatever level) suffice? Paying the Public or Caring for Constituents? 2

4) If not, is it reasonable to expect that they would have a personal interest in stamping out the corrupt sources of money, at least those related to their role as legislators? 5) To the extent that such public demands for assistance foster corruption, what would have to change in order that they are eliminated, or at least reduced? Further, regarding possible corrupt sources of money, what specific measures would contribute most to their curtailment or elimination? 6) Do election results suggest any clear correlation between such expenditure and success at the polls? All of these(and undoubtedly other) issues should be kept in mind while reviewing the results generated by this research. Guiding Assumptions While it is recognised that the Kenyan political landscape exhibits considerable structural and cultural variation, the study was guided by the following practical if general assumptions about electoral politics in Kenya that relate to its subject matter: • While personalised expenditure does not guarantee re-election, the refusal to provide such disbursements is likely to significantly lower one’s electoral chances; this applies both at the nomination and general election stage. • While such expenditures are incurred throughout the five-year period of incumbency, they are likely to be greatest in the months prior to elections. • Other factors, such as perceived political proximity to a party/ethnic leader, may have far greater impact on voters’ choices than‘generosity’ 3 Paying the Public or Caring for Constituents?

with regard to constituents’ or constituency needs; again, depending upon the level of party-electoral competitiveness in the constituency, this may apply far more to the nomination stage. • Achieving prominence in terms of critical national issues can rarely, on its own, guarantee electoral success. • Public awareness of the recent(2001) increase in parliamentarians’ remuneration, along with endemic if not rising per capita poverty, is likely to have raised demands for such hand-outs; future increases in MPs’ pay and benefits would probably have the same effect. Justification for the Study This survey was designed as part of TI-Kenya’s Transition Programme. The Programme rested on the assumption that given the impending departure of President Daniel arap Moi after more than two decades of highly personalised rule, a rather unique opportunity might arise in which critical aspects of political culture, as well as public institutions themselves, could be re-shaped. Among these was the relationship between MPs and their constituents as expressed in the personal assistance and donations MPs are expected to provide, especially in the period leading up to an election. For TI-Kenya, attention to public expectations of such largesse(as evidenced by its previous work in this area) 1 relates to its larger concern with the level and nature of corruption in public life. This is so even if the ‘generosity’ captured in this survey does not, in and of itself, suggest corruption in terms of either the sources of the funds provided or the use to which such monies are put. Rather, more generally, such data underscore the demands to which MPs are subjected, and the levels at which they attempt to respond. In light of this concern, and also taking into account the present political situation in the country, it was felt that drawing attention to this aspect of the relationship between parliamentarians and their constituents could contribute Paying the Public or Caring for Constituents? 4

to a more general national debate about how ethnic standards in public life might be raised. Moreover, it was hoped that such a debate might acquire particular urgency if a regime took power that was energetically committed to just such an agenda. In this regard, TI-Kenya looked forward to concrete measures being taken that would constitute a break with the past, whichever part was victorious in the December 2000 polls. Methodology: The Samples The study employed two samples, examined through the use of two distinct research devices. The design and use of both were motivated by a general assumption that the insatiable demand for financial and other material assistance that bombards MPs may encourage – or even necessitate- the use of office to gain wealth through corrupt practices, or at least constitute such an imposition that doing so distracts Members from other, more publicoriented tasks. This assumption takes into account, however, that those MPs most willing to participate in such a study would probably be among those least likely to engage in any such corrupt practices. Using one research instrument, the personal assistants of participating MPs were to keep weekly accounts of all expenditures related to the performance of their duties at the constituency level. Seven Members agreed to this procedure, and records were kept(with one exception) for periods of about one to four months, during the latter quarter of 2002, several(though not all) continuing right through the election process at the end of December. 2 In the other survey, a questionnaire was administered to twenty MPs(three of whom were also participating in the account-expenditure survey). In addition to recording basic data about the MPs’ personal background (constituency, education level, years in office, party affiliation, etc.), interviewers recorded responses to particular factual and attitudinal questions 2 No records were kept, however, of requests that were denied, and why, which perhaps might have been of at least equal interest to the data that were obtained. 5 Paying the Public or Caring for Constituents?

about the sources, amounts, and uses of money in the performance of the MPs’ duties with specific regard to addressing needs and demands of constituents. 3 As this was a pilot exercise, in retrospect it is recognised that certain questions might have been more precisely phrased, and others asked that were left out altogether. Before looking in detail at the make-up of these two samples, followed by a presentation and discussion of the results, a general conceptual approach to the role of money in parliamentary leadership at the constituency is offered. Conceptual Approaches While there is much that is symbolic(if not charismatic) in Kenyan politics, 4 analysts have argued that patron–client and other relations linking constituency representatives to the centre of State power appear to be more important in sustaining electoral success over the longer term. 5 While such a basis for success may have been displaced somewhat in the more competitive context of multi-party politics(where personal courage and suffering as well as commitment to a particular policy agenda have gained 3 For both instruments, it was agreed that the identity of participating MPs would remain anonymous. 4 Okumu took up this issue in an early yet still highly relevant fashion;(Okumu, John J., Charisma and Politics in Kenya: notes and comments on the problems of Kenya’s party leadership’, East Africa Journal, February, 1968, pp. 9-16). 5 Barkan explicitly analyses this aspect of relations between Kenyan MPs and their constituents(‘Legislators, Elections, and Political Linkage’, in Barkan, Joel D., ed., Politics and Public Policy in Kenya and Tanzania (Revised Edition), New York, Praeger, 1984, pp. 71-101); Widner and Throup, among others, do so in the context of considerations of other issues(Widner, Jennifer, The Rise of a Party-State in Kenya: From Harambee! To Nyayo!, Berkeley, University of California Press, 1992; Throup, David,‘Elections and Political Legitimacy in Kenya’, Africa, Vol. 63, No. 3, 1993, pp. 371-396). Paying the Public or Caring for Constituents? 6

salience, as have various forms of‘rigging’ 6 ), bestowing material favours and rewards remains a necessary if not sufficient requirement for political survival in most Kenyan localities. Such a mode of support-recruitment and retention requires a capacity to reward of two types. One is the ability – actual or perceived – to promote development by directing State resources from various ministries and other departments to the constituency. In this respect, it is generally argued that the closer one is to State House(usually, though certainly not exclusively, through cabinet position) the better one’s chances of gaining access to such resources in the first place. The other consists of individual financial capacity- and generosity- in addressing individual problems and contributing to local projects through donations that serve primarily to encourage others both within and outside the constituency to do likewise. Much of the latter comes in the form of harambee fund-raisings, an institution that, whatever its historical contribution to national development, has begun to receive its own welldeserved attention as contributing to escalating levels of corruption in recent years. 7 6 For the 1992 election, see especially,‘Why KANU Won’, in Throup, David W. and Hornsby, Charles, Multi-Party Politics in Kenya: the Kenyatta and Moi States and the Triumph of the System in the 1992 Election, Oxford, James, Currey, 1998, pp. 453-532; their description of the contest and outcome of the race in Mombasa’s Mvita constituency is particularly illustrative here. For the 1997 contests, see the collection by Rutten, Marcel, Mazrui, Alamin, and Grignon, Francois, eds., Out For The Count: The 1997 General Elections and Prospects for Democracy in Kenya, Kampala, Fountain, 2001. Of particular interest here perhaps is this comment:“Serious incidents were reported in about 15 constituencies. They varied from outright rigging to intimidation and included attempts to add ballot papers to boxes or spoil correct ones, introduction after closing hours of unsealed ballot boxes, kidnapping of returning officers, and improper handling of ballot papers by election officials”(Brakhuis, Norbert, ‘International Election Observation during the 1997 Kenya Elections’), pp. 224-241. The net effect was to give KANU majority control of parliament with 107 seats, leaving the collective Opposition with 103, before gradually widening this margin with a string of defections followed by victorious by-elections. 7 See,‘Harambee: Pooling Together or Pulling Apart?’, Transparency International-Kenya, Nairobi, 2001. 7 Paying the Public or Caring for Constituents?

In this regard, it should be noted that at the time the surveys were conducted, a Kenyan MP was receiving a gross salary of Shs. 395,000, comprising “commuted mileage”(over Shs. 330,000), and constituency, house, sitting, and“responsibility” allowances. However, a typical member in the study’s sample ended up with less than Shs. 50,000 as“net salary”, by the time deductions were made for various mandatory and optional expenses, the former including pension scheme, health and accident insurance, and cooperative society contributions, and the latter, car-loan and insurance payments and political party contributions. This being the case, it is evident that only in exceptional circumstances do MPs possess the wealth to satisfy such demands themselves, at least those of a collective nature. More commonly, an MP(or prospective candidate) seeking to display generosity must supplement his(or her) personal wherewithal with material support from elsewhere. It is this latter requirement that may contribute to the rising level of corruption(as well as even more opprobrious forms of criminality 8 ) that Kenya witnessed in recent years, even if other motivations, including simple private-accumulation/greed, undoubtedly have also been at work. 9 More specifically, a number of factors that may relate to but are distinct from corruption per se have combined to raise the salience of such individual largesse in mobilising support, or simply in being able to perform one’s duties as a leader. One has been the recent precipitous decline in the economy that, combined with population growth and the relentless AIDS scourge, continues 8 Beyond the largely State-instigated and more tragic overt violence, much of it evidently paid for(Killing the Vote: State-Sponsored Violence and Flawed Elections in Kenya,(Kenya Human Rights Commission, Nairobi, 1998), MPs(or their challengers), too, may use money for far more nefarious purposes than just buying support, and votes. This has included intimidation, bribery and thuggery, particularly during the election season but also more generally, reportedly backed by State machinery; the latter has often involved financial and legal harassment as well. And such tactics have also been employed at the level of local government electoral politics(Wolf, Thomas P.,‘Paying the Price: The Personal Gains and Losses of Local Government Contestants in the 1992 Kenya General Election’, Paper Delivered at the African Studies Association Annual Meetings, Toronto, 1994). 9 See,‘Public Resources, Private Purposes’, Transparency International-Kenya, Nairobi, 2002. Paying the Public or Caring for Constituents? 8

to drive down per capita income. Another has been the crumbling of national infrastructure and the falling quality and availability of primary social services, combined with the cost-sharing implications of SAPs(“structural adjustment policies”), together placing further strains on individual and family pocket-books to meet basic needs. Yet another may have been the substantial increase in MPs’ remuneration that took effect in 2001 and which probably became known to most constituents, so that however well-endowed their MPs were perceived to be prior to the increase, their‘problem-solving’ capacity was substantially enhanced thereafter, at least in the public eye. Keeping the above considerations and issues in mind, the data obtained from the two survey instruments may now be presented. 9 Paying the Public or Caring for Constituents?

Section TWO: Findings A. The Two Samples The Quantitative Sample As noted above, only seven MPs participated in the quantitative survey. Of these seven MP’s, three participated in the qualitative survey. Their distribution in provincial terms is as follows: Province Rift Western Nyanza Eastern Central Nairobi Coast Northeastern Number Participating MPs 3 2 1 1 0 0 0 0 Note here also that only one MP in the sample of seven was from KANU, and had been actively associated with the Opposition for some time as a so-called‘rebel’. All seven MPs were elected Members. As already mentioned, due to the small size of the sample, the data is best aggregated, even if personal variations might suggest patterns that would apply to the membership of the Eighth Parliament as a whole, and several of these are pointed out. The Qualitative Sample As also noted above, the qualitative data was obtained from interviews conducted with twenty MPs of the Eighth Parliament. This is also clearly a very small number, comprising fewer than 10 percent of its membership. Paying the Public or Caring for Constituents? 10

Moreover, three of those who agreed to participate were nominated MPs, whose particular relations with constituents that the survey attempted to explore may be somewhat different from those who entered parliament via the ballot(even if two had been candidates in 1997, and the other was an elected MP in the Seventh Parliament.) Several limitations stem from this. The process of self-selection may have excluded MPs with more representative characteristics than those who agreed to participate, thus distorting the findings; such a small sample magnifies such distortions. For example, only three members of KANU appear, and only one of them from the basically mainstream or core element of the party. At the same time, the sample size makes any significant intra-sample comparisons or correlation impossible. This could otherwise have been done in terms of particular variables, such as party membership, gender, the location of the constituency, and so on. Consequently, as in the case of the quantitative sample, there is no attempt to disaggregate the results along such lines. At the same time, however, it could reasonably be argued that the vast majority of MPs would have responded in a largely similar fashion to most of the questions contained in the questionnaire. 10 Table 1 presents several descriptive characteristics of the 20 MPs who participated in this survey. From left to right, it indicates(1) whether the constituency is in an urban or a rural area,(2) how far it is from Nairobi in terms of four zones indicating the same, 11 (3) the MP’s gender,(4) the MP ’ s party(here, KANU or Opposition, with“N” representing those in the ruling party who had already identified themselves with the NARC Opposition coalition at the time the survey was conducted, 12 (6) whether the MP had 10 More broadly, it appears that no general comparison of Kenya’s MPs has been undertaken since Hornsby did so more than three elections ago(‘The Social Structure of the National Assembly in Kenya, 1963-83’, Journal of Modern African Studies, Vol. 27, No. 2, 1989, pp. 275-296). 11 Roughly, these zones represent the following distance-areas in terms of vehicle driving-time: A: within an hour or so of Nairobi city centre, B: 1-3 hours from the city, C: 4-6 hours away, and D: more than six hours distant. 12 Indeed, one of these KANU MPs had been a‘rebel’ for nearly the entire life of the Eighth Parliament. 11 Paying the Public or Caring for Constituents?

any post-secondary education, and(7) whether this was his/her first term in the National Assembly, and if not, for how many terms had he/she served, the current one included. Table 1: General Sample Characteristics Location Nairobi Rift Nyanza Western Central Eastern Central Nyanza Eastern Nyanza Coast Central Rift Central Nyanza Central Western Coast Nyanza Rift (R/U) Zone Gender UA M RC M RC M RC M RB M RD M RA M RC M RB M RC M RC M RB F RC F RB M RC M RA M RC M RC M UC M RC M Party Opp Kanu-N Opp Opp Opp Opp Opp Opp Opp Opp Kanu Opp Opp Opp Opp Opp Kanu-N Opp Opp Opp Post-Sec.Ed. Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes Yes No Yes Yes No Yes Yes Yes Yes First Term/No. Terms No 2 No 3 Yes No 5 No 2 No 2 No 2 No 2 Yes No 2 Yes No 2 No 2 Yes Yes No 4 Yes Yes Yes No 2 Total Provincial Distribution Central 5 Nyanza 5 Rift 3 Western 3 Eastern 2 Coast 2 Nairobi 1 Distance A B C D Zones 3 4 12 1 Paying the Public or Caring for Constituents? 12

B. Findings: The Qualitative Sample From the twenty MPs who participated in the qualitative sample, the following data were obtained. The first several questions aimed at characterizing their constituencies, and illuminating their personal backgrounds. Constituency Character Rural: 18 Urban: 2 Post-Secondary Education Yes: 17 No: 3 Party Affiliation Opposition: 17 KANU-NARC: 2 KANU: 1 Gender Male: 18 Female: 2 Period of Incumbency First Term: 8 Second Term: 9 More Than Two Terms: 3 The main section of the questionnaire addressed expenditure. Questions covered such aspects as the MPs’ motivation for such disbursements, the level and sources of the funds utilized, and the objects of such payments. While not all MPs answered every question, others gave more than one answer to those that were more open-ended; as such, in some cases the total responses exceed 20, and this should be apparent. It began with several questions about constituency offices, and then moved on to other issues specifically related to financial assistance to projects and people. 13 Paying the Public or Caring for Constituents?

Do you have an office in the constituency? • Yes: 17 • No: 2 • No Answer: 1 What are the main reasons for having such an office? • to facilitate an exchange of information: 17 • to help people with personal problems: 2 • to help solve local problems: 1 What are the main constituency-level development issues that require your attention and that you discuss with constituents? • harambees: 13 • individual welfare problems: 13 • collective welfare issues: 3 • education: 2 • development planning: 1 • security: 1 Does the electorate expect MPs to address personal problems that require financial assistance? • Yes: 19 • No reply: 1 Do such expectations affect MPs’ use of personal finances at the constituency level? • Yes: 19 • No reply: 1 What is your usual monthly level of expenditure? • Shs. 150,000-200,000: 4 • Shs. 200,000-299,000: 13 • Shs. 300,000 and over: 2 • No Reply: 1 Paying the Public or Caring for Constituents? 14

What are the main sources of the funds you use? • personal salary: 18 • personal business/savings: 13 • contributions from friends: 7 • loans: 1 Note here that MPs were not asked about the content or size of their personal financial-commercial portfolios, let alone how or when such assets were acquired. Nor were they asked whether they had received any money from the political party, or any patron within the party, or from elsewhere(aside from“friends”). Moreover, none expected to get any of this expendituremoney back. What are the main categories of your constituency-level expenditure? 13 • individual problems: 18 • harambee collections: 15 • youth and women’s groups: 4 • educational establishments: 1 • funeral contributions: 1 All MPs attested to the fact that the demand, and actual expenditure they make, varies, and is greatest in the period prior to elections. Specific election-related expenditure was directed at the following, and at the frequencies shown: • harambee fund-raising: 15 • mobilization of supporters: 15 • transport for agents, supporters: 10 • donations to projects, groups: 6 13 More than one answer to the question was allowed. 15 Paying the Public or Caring for Constituents?

Regarding the reasons for such expenditure, MPs’ responses were distributed as follows: • to have a chance of electoral success: 15 • to help people with their problems: 6 • to share with constituents what is partly theirs to begin with: 1 Asked whether they were happy or not about having to engage in such expenditure, MPs responded as follows: • Yes: 13 • No: 3 • Both yes and no: 2 • No reply: 2 From Their Own Mouths: How MPs See The Need For Local Expenditure The qualitative questionnaire concluded by inviting MPs to offer reasons for the answers they had given to the specific questions noted above. A selection of their responses follows. One MP, with a distinctively negative perspective on such expenditure, stated: I have realized that most of our people are thankless. Most of them take you for a ride. However, it is the price you have to pay to educate them. Otherwise, we shall perish. Not many are willing to sacrifice for the cause of democracy. Before elections all projects are revived. Everybody you meet wants some chai , if not food. Every group wants to be visited (and assisted)…To be a leader, one must meet the needs of his/her constituents. Another articulated the same negative view, in these terms: Harambees should not be conducted by MPs, DCs or PCs, PSs or members of the judiciary. The government ought to produce policies to fight poverty rather than having people depend upon hand-outs. Corruption thrives from[sic] harambee. Paying the Public or Caring for Constituents? 16

Asked why he actually liked spending his money the way he did, one MP replied: I believe that God will reward me. It’s a noble task to help people, and besides, the money is partly theirs. Beyond this, when a government abandons its people, it’s people like the MPs that they ask for help. And no bitterness need be associated with such giving, as one MP made clear: One has to assist the people. In fact, I joined politics in order to help my people get out of poverty. Or, as another said: The demands keep increasing due to the failure of the state to deliver. Many of his colleagues expressed similar views. Typical of these was the following: The work of an MP should be divorced from welfare issues,[yet] there are genuine cases of concern such as helping families to take care of their sick. For at least two others, however, the answer was simpler: One has to sustain the interests of his/her supporters. It’s inevitable to spend, but[this occurs] not because one wants to; it’s the only way to remain an MP in Kenya. Or, according to another(and a view that was rather common): I spend in order to sustain my political career. There’s always the hope that one day a government will be in place that will address the issues that affect peoples’ lives positively. This view was put even more succinctly by one of his colleagues: Such expenditure is based on two realities: social pressure, and political expediency. This burden is said to be especially great for Opposition MPs: The government discriminates in its discharge of duties such as the 17 Paying the Public or Caring for Constituents?

development and maintenance of infrastructure in those constituencies represented by Members of the Opposition as a way to discredit them. Even in a KANU-dominated constituency, however, the demands for such expenditure can never be satisfied. As one then ruling-party Member put it: The people’s poverty compels one to spend on their plight. In fact, the amount spent is never enough to meet their needs. Such dependence on individual largesse was not seen by all as perpetually inevitable, however. According to another MP: There is nothing one can do once one has ventured into politics. During every campaign period, one has the hope of forming the next government and thus initiating change from within. The government need to implement what has been passed by parliament. For example, in the last 10 years no water has been delivered to my constituents despite the pledge that every household would have piped water by 2000. Rampant corruption has contributed to this failure… In the same vein, one MP argued that: At times I am made to perform duties that ought to be handled by the Provincial Administration. The Executive branch must perform. Finally, and looking to the future, several MPs stressed the importance of civic education as a mechanism for reducing such demands for their personal expenditure: The electorate needs to be educated on what to expect from an MP. However, if the government had done its work property, there would be very few problems at the constituency level for us to deal with. And this includes the local authorities as well. Re-Election Recall here that of the 20 MPs who constituted the qualitative sample, three had gained entrance to parliament in 1997 via party nomination. Paying the Public or Caring for Constituents? 18

All three again attempted to enter parliament via the voters’ wishes, and one did.(Neither of the other two made it back via nomination.) The electoral fortunes of the entire group were as follows: • Elected MPs: • Nominated MPs: (17) 12 re-elected; 5 defeated (3) 1 elected; 2 defeated Altogether, 13 of them found their way back into the Ninth Parliament. In conclusion here, regarding the samples, it should again be emphasised that no MP from KANU’s inner-circle participated, a Member, it could be assumed, for whom access to money and other resources should have been no(or a much-diminished) problem. Nevertheless, the perspective from that side of the political fence would have been interesting to have. By way of contrast, after the 1992 return to multi-party politics, President Moi began marketing the slogan,“ siasa mbaya, maisha mabaya ”(i.e.,“bad[opposition] politics leads to poor living conditions and suffering”). The clear implication here was that those parts of the country that rejected KANU would be subjected to a deliberate policy of resource-deprivation by the State. The resulting heightened poverty would serve only to increase the personal financial demands on MPs and would eventually drive them – and/or the voters in the constituency- into KANU’s waiting arms in the subsequent election. If such pressures were resisted, it was made clear, such Opposition stalwarts without substantial personal wealth would likely face bankruptcy, an evaporation of their political support, or both. 14 C. Findings: The Quantitative Survey As noted above, only seven MPs agreed to participate in the daily expenditure survey, in which their Personal Assistants were to keep records 14 Such an argument was offered with equal intensity in encouraging Opposition MPs to defect to the ruling party throughout the parliamentary term, repeating a pattern that had begun shortly after the first multi-party elections in 1992(with induced-defections in Migori and Bonchari constituencies); actual results, however, were mixed. 19 Paying the Public or Caring for Constituents?

of each outlay, indicating both the purpose/category of expenditure and the amount. To present the data obtained, the MPs are identified as A-G. The periods during which records for each of them were kept are as follows: MPs’ Expenditure Records: The Survey Period MP Survey Period A 9 September- 25 December B 7 September- 1 December C 20 September- 18 December D 29 August- 12 January E 19 September- 28 November F 5 October- 18 November G 12 October- 22 October Total Days 96 29 56 89 68 31 10 Note, however, that on some days no entries were made, though this varies for particular MPs. For example, the overall period of expenditurerecords for MP“B” covers nearly three months, but contains entries for only 29 days. Whether the absence of any payment records simply reflects the fact that no payments were made on these days, or rather, a failure to remember to note them down(or even to deliberately exclude them), cannot be ascertained. We can now look at how each MP allocated expenditure between the following broad categories designated in the survey instrument: 1. Transport 2. Harambee 3. Personal 4. Political Party 5. Entertainment Paying the Public or Caring for Constituents? 20

Although the distinction between these categories is not always entirely clear, recorded expenditure has been placed in each of them as follows: 1. vehicle hire, repair, fuel and public transport costs, including air fares 2. group and project donations, assistance to individuals in meeting welfare expenses(health, education, etc.) 3. costs relating to the MPs’ own campaigns, including payments to consultants, campaigners, security teams, supporters-voters, and pollwatchers, as well as for posters, t-shirts and other campaign materials, and for photographic and video equipment hire/use of public address systems and communication costs(phone cards, etc.) 4. party costs including building rent and equipment costs, support to other(local government) candidates including their nomination fees 5. meals and accommodation for the MPs themselves along with aides and supporters, payments for choirs and dancers and sporting events associated with campaign activities This information is captured in the following pie-charts for the MPs, as shown below: 15 MP – A Party 3% Entertainment 17% Harambee 15% Personal 47% Transport 18% 15 In this connection, it should be noted that however impressive is the level of detail, it cannot be assumed that all the aides of the MPs who kept these daily expenditure-records were equally meticulous in doing so. 21 Paying the Public or Caring for Constituents?

Party 10% Entertainment 5% Harambee 21% MP – B Personal 34% Transport 30% Party 13% Entertainment 1% Harambee 20% Transport 11% MP – C Personal 55% Entertainment 12% Harambee 12% Transport 15% Party 0% MP – D Personal 61% Paying the Public or Caring for Constituents? 22

Entertainment 3% Harambee 13% Transport 36% MP – E Party 5% Personal 43% Party 11% Entertainment 1% Harambee 38% MP – F Personal 24% Transport 26% Entertainment 4% MP – G Harambee 73% 23 Paying the Public or Caring for Constituents? Personal 11% Party 0% Transport 12%

The data shown in the above charts can be summarized as follows: Categories of Expenditure: Percentage Distribution M P Personal Transport Harambee Entertainment Party A 47 18 15 17 3 B 36 31 18 5 10 C 55 11 20 1 13 D 61 36 13 3 5 E 43 36 13 3 5 F 26 21 40 1 12 G 11 12 73 4 0 Although the individual variations, even within such a limited sample, are significant, a general pattern holds: relatively low expenditure on Entertainment and Party costs, relatively high expenditure on Personal assistance and Transport, and moderate expenditure in terms of Harambee contributions.(Note here that for only two MPs was this latter category the highest.) Perhaps such a pattern of distribution reflected, at least in part, the pre-electoral season during which the records were kept. Specifically, if the two most extreme figures for each category are excluded (i.e., the highest and lowest), the resulting ranges(as above, expressed in percentages of total expenditure) are quite limited: Expenditure Categories: Range of Distribution Range Personal 26-55 Transport 12-36 Harambee 13-40 Entertainment 1-17 Party 3-12 Let us now turn to average total daily expenditure, recalling the variations noted above in terms of the number of days covered by each MP’s records. Paying the Public or Caring for Constituents? 24

Average Daily and Total Expenditure(in KShs.) M P Average Daily Expenditure Total Expenditure A 13,773 1,308,450 B 50,424 1,462,300 C 9,155 523,320 D 52,436 4,666,760 E 21,654 1,450,790 F 20,217 554,500 G 14,040 140,400 These figures can also be portrayed graphically, as shown below: Most striking in terms of this measure is the gap between two MPs and their five colleagues. For these two individuals, the abundant figures represent, in one case, considerable personal wealth and, in the other, close connections to party‘barons’ who can be assumed to have provided 25 Paying the Public or Caring for Constituents?

at least some financial assistance. Also note here, however, is that only one or two of the seven had any significant challenge at the partynomination stage. Whatever the case(and leaving aside MP“G” who, for reasons that cannot be ascertained, provided expenditure records for only ten days), the total spent by five of the MPs throughout this period far exceeded what their salaries alone would have provided. Paying the Public or Caring for Constituents? 26

Section THREE: Concluding Considerations Whereas it is philanthropic to participate in community development efforts as well as to help people in need, it is wrong to measure MPs’performance based on such activities. These acts perpetuate sluggish development and can not address peoples’ needs substainably[sic]. MPs should concentrate on strengthening the institution of parliament and making credible national development policies and legislations[sic]. The harambee donations are like a drop of water in the ocean. They enhance communities’ dependence on individuals. They also encourage corruption among politicians and people aspiring for elective positions. 16 Kenya MPs, like elected representatives everywhere, are subjected to a number of competing performance criteria. Most publicly, perhaps, they attract the attention of the media for various acts of commission and omission. Likewise, a number of NGOs have turned their sights on MPs, examining everything from the frequency of their parliamentary appearances, to their performance on particular committees and their actual contributions on the floor of the House in terms of offering motions and voting on bills. 17 However, often most critical for their futures are the views of party leaders and key officials, who may have the power to determine political survival at the time of pre-election party nominations(and access to key material resources in the meantime). This is so especially given the fact that much of the country’s political landscape consists of‘zones’ of one party or the other. Consequently, during the last three elections, the level of 16‘Kenya’s Parliament – Members’ Participation(1998-2001)’, Institute for Civic Affairs and Development, Nairobi, Print Displays Ltd., 2002, p. iii. 17 Ibid. 27 Paying the Public or Caring for Constituents?

competition in the general election itself was minimal in nearly two-thirds of Kenya’s parliamentary constituencies, with the winners more than doubling the votes of their closest rivals(or even, in several cases, going through unopposed). This has placed additional emphasis on the nomination stage, where in the absence of observers for the most part, far greater deviation from“free and fair” practices may obtain than in the general election itself. Above all, however, MPs can only ensure their political longevity by meeting the expectations and demands of at least a sizable chunk of those to whom they formally owe their positions: their electorate. For the latter, while parliamentary performance is usually not irrelevant, the perception of personal character may count for far more, especially in more rural constituencies where access to information(here, about the conduct of business within the House) via national media is minimal for all but the local elite. Even where the critical decisions about nominations and electoral support are often made by a rather narrow slice of this local elite based on commercial, legal, and other patronage favours that an MP has been able (or is seen to have the potential) to provide, without far more general, popular acceptance, winning is unlikely. 18 And in much of Kenyan(and indeed, of African) society, the most important foundation of such acceptance is the perception of generosity: of using one’s own wealth, and/ or the wealth of others to which one has access, to help those among whom one lives- or from whence he or she comes- to solve their individual problems, and those of the community writ large. Following the substantial increase in Members’ remuneration in 2001, a presidential tribunal has recently(November, 2002) tabled a report recommending a further substantial increase in their pay and benefits. 19 In justifying its position, the Report argued: 18 Of course, in contests with three or more candidates, victory may be grasped with a plurality that amounts to far less than half of the votes cast. 19 See, Parliamentary Service Commission,‘Report of the Tribunal appointed by the Parliamentary Service Commission to review and make recommendations on the terms and conditions of service for Members of Parliament’, Republic of Kenya, November, 2002. This is so even if, for the first time, a portion of it would be subject to tax. Paying the Public or Caring for Constituents? 28

However, the heaviest burden that a Member has to bear throughout his/ her career as a Parliamentarian is in the shape of a voter. The voter may need a contribution to a fund to enable his/her child to go for further studies, or his/her medical treatment in an urban hospital, non-receipt of his/her pension[sic], unfair treatment by a chief or an Administrative Officer in a petty boundary dispute with a neighbour, or cattle rustling, to name but only a few. The needs of a voter are innumerable and generally go with an expectation of some monetary hand out[sic] from the Member to help overcome immediate financial problem[sic]. A Member of Parliament who ignores such personal needs of his/her voters does so at the risk of being ousted in the next election… The above attempt is not, nor is it supposed to be, an exhaustive exposition of duties and work of a Member of Parliament. It is a mere glimpse of some of the factors that cause a Member to undergo untold mental, physical and financial stress throughout his/her political tenure. Clearly, an adequate remuneration is the only way not only to compensate for the stress that honourable Members have to undergo but also for out of pocket deficits incurred by him/her to meet the demands of his/her voter[sic] 2002: 66-7]. Indeed, and as other work by TI-Kenya has shown, it was, in part, this escalating need for MPs’ to seek support from those higher up in the national patronage structure in order to meet such demands that has contributed to rising levels of corruption in various areas of public life. 20 Whether the now-proposed Shs. 20 million constituency development fund(likewise recommended by this Tribunal) will completely obviate (or at least substantially reduce) the need- and the demand- for such contributions, remains to be seen. 20 MPs have sometimes been accused of using their positions on certain committees that have the power to expose wrong-doers of engaging in extortion: demanding payment in exchange for ensuring no(or least only limited) negative information regarding some individual or company is obtained by the media. 29 Paying the Public or Caring for Constituents?

Clearly, then, being a“good person” in Kenya, and in Africa generally, requires that those who are better-off assist the less fortunate, especially if they are socially close. This is so particularly where those who are less well-off can claim direct responsibility for the largesse such an individual is able to enjoy: the fruits of elective office. At the same time, one might argue that when substantial vote-buying has occurred at election time(whether at the nomination or actual election stage), the winner would thereafter be absolved from such transactional transfer in the form of subsequent donations throughout incumbency. 21 However, such payments/assistance appear to be re-cast post-election, as first, the voters must be“thanked”, and then the“fruits of victory” must be shared. However, given that no constituency is small enough for an MP to so share his pay and other perks equally with everyone(even with only those whose votes one received), 22 choices must be made as to who gets, how much, for what, and when. Finally, given such an insatiable demand for financial assistance on the ground, it appears unlikely that any particular increase in MPs’ remuneration would enable them to more fully satisfy such demands (though this remains a subject for empirical investigation). The question thus remains as to whether such personal funds can ever be sufficient to satisfy these kinds of requests- individual or collective- and if not, where the additional resources will come from. 23 The debate on how much MPs should earn, and what they should do with their pay- as well as with other money or material resources they may receive from‘well-wishers’, in addition to their own income from 21 The logic here being that,“Since I’ve already paid you for getting elected, I owe you nothing more.” 22 Registered voters alone in the sample’s seven constituencies ranged from about 40,000 to 90,000, with an average of about 60,000. Clearly, attempting to satisfy constituency“needs” on an individual basis with these kinds of numbers would be impossible. 23 The alternative, of course, and as suggested earlier, is that voters begin to evaluate their MPs’ performance in terms of other, less-immediately material criteria. Paying the Public or Caring for Constituents? 30

other sources- will undoubtedly persist. Nevertheless, the general desire for change in Kenya appears palpable, and may soon allow for initiatives that would begin to alter certain now-engrained aspects of Kenya’s political culture. The question is whether there exists a leadership that will take advantage of this opportunity, and is prepared to propose measures to actually do so. Under the country’s evolving democratic institutions, the public will subsequently have its own opportunity to respond. 31 Paying the Public or Caring for Constituents?