Improving paratransit service: Lessons from inter-city matatu cooperatives in Kenya

Roger Behrensa (corresponding author), Dorothy McCormickb, Risper Oreroc and Marilyn Ommehb

a
Centre for Transport Studies, Department of Civil Engineering, University of Cape Town, Private Bag X3, Rondebosch, 7701, Cape
Town, South Africa; email: roger.behrens@uct.ac.za
b
Institute for Development Studies, University of Nairobi, PO Box 30197-00100, Nairobi, Kenya; email: dmccormick@uonbi.ac.ke;
saramarilyno@gmail.com
c
Department of Business Administration, Kenya Methodist University, PO Box 45240- 00100, Nairobi, Kenya; email:
risorero@yahoo.com

© 2016. This manuscript version is made available under the Elsevier user license
http://www.elsevier.com/open-access/userlicense/1.0/
Abstract

While providing essential access for large portions of city populations, the quality of paratransit services in
Sub-Saharan African cities is poor. Poor quality of service can be attributed to two features of the paratransit
business operating model: driver remuneration on the basis of a daily ‘target system’; and cash-based business
management in which vehicle depreciation is ignored as an operating expense. In Kenya, the voluntary
organisation of fragmented inter-city matatu businesses into Savings and Credit Cooperatives (SACCOs) has
resulted in improved service, regulatory compliance and technology adoption, but little is known of how they
operate. The aim of this paper is to gain insight into how the Kenyan inter-city matatu SACCOs are organised
and have improved services, and to explore the transferability of this experience and the lessons it offers. The
exploratory nature of the study, and constrained resources, necessitated that the research adopt a qualitative
case study method. It was found that most of the case study SACCOs have addressed the root causes of poor
service quality by shifting the remuneration of drivers from daily cash targets to salaries, and by requiring
vehicle depreciation costing through compulsory contributions to the cooperative’s capital savings from which
vehicle acquisition and repair loans can be derived. Due to the particular shuttle-like nature of inter-city
services and the considerable institutional support that exists in Kenya for cooperatives, the direct transfer of
successes to other contexts is likely to prove difficult. Identifying the features of the inter-city SACCO model
that have led to paratransit service improvements, and attempting to replicate these, may therefore be more
effective than attempting to replicate the model in its entirety. These features are argued to be operator
consolidation, accompanied by salaried drivers, systematic vehicle monitoring and compulsory vehicle
depreciation costing. They can be adopted in other forms of paratransit organisation and regulation, but will
require considerable adaptation to context.

Keywords:

Paratransit
Service quality
Kenya
Cooperatives
Driver employment
Operations management
1. Introduction

The decline of scheduled large bus services in most Sub-Saharan African cities in the latter half of the
previous century led to the establishment of extensive paratransit1 operations. These offerings typically took
the form of mini- and midi-bus services provided by small informal businesses, subjected to varying degrees
of market entry regulation. While providing essential access for large portions of city populations, the quality
of paratransit services in Sub-Saharan African cities is widely regarded by passengers to be unsatisfactory
(see, for instance, Agyemang, 2013, Behrens and Schalekamp, 2010, Govender, 2014, Koimur et al, 2014).

While the reasons for poor service quality are no doubt complex and contextually variant, a common
underlying cause is the nature of the paratransit business operating model that is found in many, if not all,
cities on the sub-continent. It is posited that many problems can be attributed to two main features of this
operating model. The first is the basis for driver remuneration. Vehicle drivers either keep cash fare revenue
less fuel expenses and a daily ‘target’ rental payment to the vehicle owner, or are paid a commission by the
vehicle owner in the form of an agreed portion of weekly fare revenue (McCormick et al, 2016, Joubert,
2013). These driver remuneration practices create strong structural incentives for drivers to compete
aggressively for passengers in the road space, overload vehicles, speed as they attempt to increase the number
of service trips during peak periods, and delay departures from termini until the vehicle is full during off-peak
periods. The second feature of the operating model to which problems might be attributed is the day-to-day,
cash-based nature of business management. Operating costs are typically perceived to be routine labour, fuel
and vehicle maintenance expenses. Vehicle depreciation is seldom included as an operating expense, and
consequently when vehicles reach an optimal age for replacement, reserves with which to purchase a new
vehicle or secure affordable credit are not available, and vehicles remain in use in an increasingly
unroadworthy and unsafe condition.

It was against this particular framing of the public transport problem that a research project was undertaken to
study public transport cooperatives2 operating out of Nairobi (McCormick et al, 2015a). As in other parts of
Sub-Saharan Africa, the general quality of paratransit services in Kenya is notoriously poor, and in need of
improvement. These services are known colloquially as matatus, from the Kikuyu term (mangotore matatu)
for the 30 cents flat fare that was charged when the mode first became popular. In around 2010 the Kenyan
public transport authorities observed that the voluntary organisation of some fragmented inter-city matatu
businesses into Savings and Credit Cooperatives (SACCOs) in the 1990s and 2000s had resulted in improved
service quality, technology uptake and regulatory compliance. As will be demonstrated in the following
section, a review of the literature in 2014 indicated that, while there might be a considerable literature on the
cooperative movement and matatus in general, there is comparatively little on how public transport SACCOs
work and the benefits they offer. The aim of the research undertaken was therefore to investigate how the
Kenyan SACCO model works with respect to public transport, and to explore its value as a means of
managing paratransit operations and of improving service quality.

This paper reports upon the findings of the inter-city matatu SACCOs study, and has two main aims. The first
is to expand what is known of matatu SACCOs with respect to how they are organised, how drivers are
employed, how SACCO members benefit (particularly with respect to vehicle acquisition and maintenance),
and how service operations management has been improved. The second aim is to explore the transferability
of the Kenyan inter-city matatu SACCO experience, and the lessons it offers for paratransit service quality
improvement in South Africa and in other contexts where similar paratransit systems prevail.

The paper is divided into seven main sections. In the following section the relevant literature on paratransit
service improvement is briefly reviewed. Section 3 goes on to describe the cooperative movement and the key
policies that have been introduced to regulate matatu operators in Kenya. Section 4 explains the method of the
study. Section 5 presents the main findings of the study, focusing on SACCO organisation and membership,
1
For the purposes of this paper ‘paratransit’ is defined as flexible public transport services that do not follow fixed schedules
or adhere to a specified duration of service.
2
For the purposes of this paper a public transport ‘cooperative’ is defined as an autonomous, voluntary and democratically
controlled organisation within which members pool savings, and in return are provided with credit facilities and other benefits to
facilitate the provision of services.

2
driver employment, member benefits, and advances in operations management. Section 6 discusses the
transferability of the successes that can be observed in the Kenyan inter-city SACCO experience. Section 7
concludes with reflection on the lessons that can be learned, and with a discussion of further research needs.

2. Literature review

Scholarly publications on paratransit services in developing world cities first emerged in the 1980s and 1990s,
with South-east Asian cities receiving the greatest attention (e.g. Rimmer, 1984, Roth, 1987, Cervero, 1991,
2000). The main arguments advanced in the literature from this period were essentially that paratransit modes
perform an important role in passenger transport systems, often providing niche services under conditions in
which conventional scheduled services cannot be sustained. In perhaps the most comprehensive publication of
this period, Cervero (2000) presented potential strategy options for rationalising and improving paratransit
services, ranging from operator organisation, to regulation, financial support, infrastructure improvement,
traffic management and operator training. From a spectrum of policy responses to ‘informal transport’ ranging
from ‘acceptance’, to ‘recognition’, ‘regulation’ and ‘prohibition’, he argued that ‘recognition’ policies (in
which rules and minimum standards focussed largely on safety and insurance are enforced without public
sector mediation of levels of supply) and ‘regulation’ policies (in which market entry and exit is publically
controlled) are most commonly appropriate. Some authors from this period argued that the developing world
offered commercially successful models of demand-responsive service provision that, with modification,
could be transferred to developed world cities (e.g. Silcock ,1981, Roth and Shephard, 1984, Chujoh, 1989,
Cervero 1997).

The literature on paratransit in developing world cities from the 2000s onwards, however, raised concerns
around the potential dangers of unbridled ‘in the market’ competition (e.g. Sohail et al, 2006, Cervero and
Golub 2007, Golub et al, 2009). The paratransit industry was not advanced by authors like Golub et al (2009)
as an alternative to mass public transport system improvement, but as an important auxiliary upon which
many poor households will continue to depend. Sohail et al, (2006) noted the importance of the capacity of
the industry to self-regulate in conditions of weak government regulation and enforcement capacity, and
argued for regulatory frameworks that strike a balance between avoiding the negative externalities of
paratransit, while avoiding overly detailed or ‘heavy’ regulations that increase the potential for non-
compliance and corruption.

Scholarly publications on paratransit quality of service improvement in developed world cities predate those
on developing world contexts. The first publications emerged in the early 1970s in the United States in which
various authors argued that the eradication of ‘jitneys’ had robbed American cities of an important
complementary public transport service, and that a variety of unscheduled, demand-responsive shared services
should be allowed to re-establish (e.g. Saltzman, 1973, Kirby et al, 1974, Orski, 1975). The main arguments
advanced in this literature were essentially that existing regulations presented a major obstacle to the
emergence of innovative paratransit services, and that allowing freer market entry and fare deregulation would
enable a rich mix of new services to emerge as well as fare structures that better reflected actual operating
costs. Reducing the barriers to market entry was argued to increase the supply of, and competition between,
services, and thereby reduce fares, incentivise service quality improvement and eliminate illegal operations.

More recently, research into technology-supported ‘demand-responsive transit’ in developed world city
contexts has grown fairly rapidly, as these cities sought alternatives to private car dependency in lower density
environments in which conventional public transport systems are not viable, and sought to cope with
mandates to provide services for the disabled (e.g. Mageean and Nelson, 2003, Enoch et al, 2004, Palmer et
al, 2004, Brake and Nelson, 2007). Despite the significant innovations in this research, flexible transport
services have not been widely implemented, and there remains a service offering gap between conventional
public transport and the private car. Echoing earlier authors from the 1980s, Finn (2012) suggests that
paratransit operations in the developing world may hold lessons on how some of the barriers to their
implementation (see Mulley et al, 2012) might be overcome.

3
In both the developing and developed worlds, the paratransit field, and how paratransit services might be
improved, remains under-researched at present relative to other public transport modes. This is particularly
true in the Sub-Saharan African context, despite the heavy dependence of contemporary city public transport
systems upon this mode, and the centrality of paratransit to current public transport reform processes. The
limited literature focussed on paratransit in Sub-Saharan Africa has focussed on the emergence, violent
competition and labour relations associated with paratransit in Southern Africa (e.g. McCaul 1990, Khosa,
1992, 1995, Dugard, 2001, Woolf and Joubert, 2013, Joubert, 2013), and on the organisation and service
quality of paratransit in East and West Africa (e.g. Adeniji, 1987, Khayesi, 1999, Rizzo, 2002, Kanyama et al,
2004, Mutongi, 2006, Wa Mungai and Samper, 2006). With respect to the scope of the study reported upon in
this paper more specifically, while there might be a considerable literature on the cooperative movement and
matatus in general, there is comparatively little on how public transport SACCOs work and the benefits they
offer (limited to Gicheru et al, 2011, Kimani, 2007, Muriungi, 2013, and Orero and McCormick, 2011, 2013).

3. Contextual background

Nairobi, the capital city of Kenya, has a population measured at 3.14 million in the 2009 census (KNBS,
2010). Its paratransit vehicle fleet was observed to be 9 554 in 2012 (Envag Associates, 2012), and paratransit
services have been estimated to hold ~87% of the public transport passenger market share (Kumar and Barrett,
2008). Apart from entry restrictions placed on the smaller 14-seater paratransit vehicles into the central
business district (Ommeh et al, 2013), this dominant passenger market share translates into a service network
that spans the majority of the city (Williams et al, 2015), and serves a diversity of passenger types and trip
purposes. The major radial paratransit service routes leading into the city (Thika, Juja, Jogoo, Uhuru and
Ngong roads) have been observed to carry in excess of 70 000 passengers/day (JICA, 2006). These routes
experience heavy congestion during peak periods, with mean commercial speeds ranging between five and 10
kilometre/hour in central areas, and trip lengths of up to 15 kilometres in the morning peak experiencing a
mean trip times of 58 minutes (Bruun et al, 2016).3

To situate the research undertaken further, and to provide a necessary backdrop for later discussion on the
transferability of policy successes, this section describes the key policies that have been introduced to regulate
matatu operators, and the Kenyan cooperative movement.

3.1 Kenyan public transport policy

Scheduled public transport operations in Nairobi date back to 1934 when the Overseas Motor Transport
Company of London was granted monopoly rights to operate a network of services in the city, and established
Kenya Bus Services (KBS). Large-bus services were provided by KBS and later market entrants (Nyayo Bus
and Stagecoach-Kenya Bus) for almost half a century, but a variety of factors in the 1970s, 1980s and 1990s –
including fuel price spikes, decayed road infrastructure, currency devaluation and a no standing passenger rule
(Klopp and Mitullah, 2016, Opiyo, 2002) – made viable operations difficult. The quality and quantity of
formal, scheduled bus service declined as a consequence.

Following independence in 1963 urbanisation restrictions were removed and the demand for public transport
services grew. Matatus service provision increased to satisfy this demand. These informal and unregulated
services were tolerated by the authorities because they filled gaps in the formal bus system, especially in
Nairobi’s Eastlands and the rural areas (Bruun et al, 2016, Aduwo, 1990). In 1973 Jomo Kenyatta issued a
presidential decree declaring matatus legal, enabling them to carry fare-paying passengers without obtaining
Public Service Vehicle (PSV) licenses from the Transport Licensing Board (TLB) or paying any taxes to the
city or central government. This resulted in a quadrupling of the city’s paratransit vehicle fleet between 1973
(373) and 1979 (1 567) (Otachi, 2013). Table 1 presents a chronology of key events in matatu regulation and
organisation (drawing from Klopp and Mitullah 2016).

3
For a more elaborate description of the public transport system in Nairobi, see Bruun et al (2016) and McCormick et al
(2016).

4
Table 1
Chronology of key events in matatu regulation and organisation

Key event
1973  Presidential Decree issued, which legalised matatus and allowed them to operate without a PSV license
1979  Matatu Vehicle Owners Association formed
1984  Traffic Amendment Act (10 of 1984) passed, which required all matatus to obtain PSV licenses, undergo
annual vehicle inspection, and carry passenger insurance
1988  matatu associations banned for multi-party political activism
1999  TLB rules introduced, which required matatus to display and remain on the routes for which they are licensed
2001  Matatu Welfare Association formed
2003  Matatu Owners Association formed
2003-2005  ‘Michuki rules’ (Legal notices 161 of 2003, 83 of 2004, 97 of 2004 and 65 of 2005) implemented, which
required:
o display of route details and painted yellow band on matatus;
o fitting speed governors in all PSVs, with speed limited to 80 km/h (with vehicle owners held liable for any
device tampering);
o fitting seat belts in all vehicles;
o employment of PSV vehicle crews on a permanent salaried basis;
o display of PSV driver’s photograph in vehicles
o wearing badges and uniforms by PSV vehicle crews (navy blue for drivers and maroon for conductors);
and
o biennial re-testing of drivers
2010  TLB directive introduced, which required all PSV owners to consolidate into either SACCOs or management
companies in order to be licensed
2012  National Transport and Safety Authority Act (33 of 2012) passed, which established a new regulatory body that
assumed the functions of the TLB
 National Transport and Safety Authority (NTSA) directive introduced, which required:
o minimum SACCO fleets of 30 serviceable vehicles;
o installing digital speed governors;
o (for those travelling at night) installing a fleet monitoring system;
o implementing a cashless fare collection payment system by 1 July 2014
2014  NTSA extended the deadline for cashless fare collection to 31 December 2014, and announced that no
operating license will be renewed unless the vehicle has a cashless (or ‘cash-lite’) fare system
2015  NTSA further delayed the deadline for cashless fare collection as a requirement for the renewal of operating
licenses

Various associations were formed over the following two decades to promote the emergent matatu industry’s
interests. The Matatu Vehicle Owners Association (MVOA) was formed in 1979 as an umbrella body with
local level route association affiliations. MVOA, together with a number of smaller unaffiliated matatu
associations, were however banned in 1988 because of their activism in support of a multi-party political
system (Klopp and Mitullah, 2016). This ban on matatu owner organisation was subsequently lifted, and two
new national owners associations were registered: the Matatu Welfare Association in 2001; and the Matatu
Owners Association in 2003.

In parallel to the industry’s endeavours to organise itself, the government introduced or amended a group of
acts (chapter 404 of the Transport Licensing Act, chapter 403 of the Traffic Act, and chapter 229 of the
Regulation of Wages and Conditions of Employment Act) which, together with various associated legal
notices, provided a framework for the regulation of the industry. Early attempts at regulation were introduced
through the provisions of the Traffic Amendment Act of 1984, which became known as the ‘matatu bill’. This
act required matatu business owners to obtain PSV licenses from the TLB, undergo annual vehicle
inspections, convey no more than 25 passengers, and carry passenger insurance cover. In 1999 TLB rules
were introduced that linked PSV licenses to routes, and required matatu owners to display these routes on
their vehicles. The provisions of the acts and legal notices that comprised (and continue to make up) the
paratransit regulatory framework were not fully implemented, however, resulting in a variety of service
quality and safety problems persisting.

5
To address poor service quality, a set of legal notices were introduced (or at least reaffirmed) between 2003
and 2005 (known as the ‘Michuki rules’ after the responsible Minister for Transport and Communication,
John Michuki) which required PSV owners to, inter alia: display route details; fit speed governors; install seat
belts; employ drivers and conductors on a permanent salaried basis; issue badges and uniforms to PSV crews,
and re-test drivers biennially (Klopp and Mitullah, 2016). These legal notices were the most successful in
improving regulatory compliance, but their effects were short-lived. After Minister Michuki left office most
matatu operators reverted to their old practices, and by 2009 any improvements had effectively been eroded.

As noted earlier, renewed attempts at matatu service regulation followed observations by the authorities that
service quality was better, and regulatory compliance higher, amongst inter-city matatu businesses that had
organised themselves into SACCOs. The perceived success of the inter-city SACCOs had led some intra-city
matatu business owners to voluntarily attempt to replicate the SACCO model (e.g. Embassava SACCO and
Nawaku SACCO) (Migwi, 2011), but this practice was not widespread. A legal notice was therefore
introduced by the Ministry of Transport and the Transport Licensing Board in 2010, which required all inter-
city and intra-city matatu businesses to consolidate into either a SACCO or a transport management company
in order for their PSV licenses to be renewed (by 2011 no individual matatu owners were to be registered).4
This was followed by a further notice issued in 2012 by the National Transport and Safety Authority (NTSA)
(which replaced the TLB in that year) which required that SACCOs have a minimum fleet of 30 serviceable
vehicles, and that they install digital speed governors and, for those travelling at night, a fleet monitoring
system. It further stipulated that all PSV operators should implement a cashless (or ‘cash-lite’) fare payment
system by 1 July 2014, a deadline that was later extended to 31 December 2014. This deadline was
subsequently delayed even further, with a view to conducting trails of a revised system in the second half of
2016.

3.2 Kenyan cooperatives

Cooperatives were introduced into Kenya by the colonial British government, and embraced as a cornerstone
of economic development policy by the post-1963 independence government. While initially focussed on
agricultural production and marketing, cooperatives were proactively spread to other sectors of the Kenyan
economy as well (Wanyama, 2009). The Kenyan government played a leading role in this diffusion. A
Ministry of Cooperative Development and Marketing was established (currently a department within the
Ministry of Industrialization and Enterprise Development), which oversaw the provision of a guiding
legislative framework, policies and codes of conduct. A Cooperative Bank of Kenya was created in 1968 to
offer bespoke financial services, and a Cooperative University College of Kenya was established in 1972 to
facilitate training and research in support of the movement. Cooperatives in Kenya therefore have a long and
established history. The International Cooperative Alliance estimates that they account for a remarkable ~45%
of the national gross domestic product and ~31% of total national savings and deposits (ICA, 2006), while
Wanyama (2009) suggests that ~63% of the national population (38.61 million in 2009, KNBS, 2010)
participates directly or indirectly in cooperative-based enterprises.

Cooperatives emerged in the public transport sector relatively late. The first matatu SACCO, 2NK, was
formed in 1993 by competing businesses operating inter-city service routes between Nairobi and Karatina
(located ~130 km apart). These small, fragmented businesses perceived that ruinous competition was eroding
their individual profit margins (Gicheru, 2011). On the formation of 2NK, responsibility for the management
of vehicles and drivers was gradually shifted from owners to SACCO staff, and as a result profit margins
improved. From the mid-1990s onwards, this success inspired the formation of numerous other inter-city
shuttle services connecting all major towns. Over time, and especially following the 2010 TLB directive
requiring matatus to be part of SACCOs or companies, many of these shuttle service providers chose to form
SACCOs. The inventory available at the time of the study indicated that in 2013 there were 509 registered
public transport SACCOs in the country.

4
In the subsequent period, the formation of SACCOs has proven more popular than transport management companies. An
inventory compiled in 2013 indicated that there were 509 registered SACCOs compared to 190 registered transport management
companies.

6
4. Case study method

The exploratory nature of the research, and constrained resources, necessitated that the research conducted on
the inter-city matatu SACCOs adopt a qualitative case study method. This method is described in terms of:
case selection; and data collection.

4.1 Case selection

The unit of analysis for the study was the SACCO as an organisation, and the sampling frame was a list of 509
SACCOs registered with the Department of Cooperative Development and Marketing. Ten SACCOs (2% of
the sampling frame) were selected on a purposive basis. Originally 13 SACCO cases were selected, but this
was reduced to 10 when it became evident during fieldwork that one of the SACCOs selected had merged
with another which was also part of the original group, and a further two were unresponsive to interview
requests.

Three criteria for non-random case selection were applied. The first, for pragmatic fieldwork reasons, was that
the SACCO should have a service route destination in Nairobi. The second selection criterion was to cover a
diversity of geographical regions. The third selection criterion was to compare SACCOs that were registered
over different time periods and are thus more or less established. The delimitation of these periods was
derived from shifts in paratransit regulatory requirements and the growth of the Kenyan cooperative
movement. The fact that the first SACCO is reputed to have been founded in 1993 suggested that the first
period should extend from the 1990s. At the other end of the time spectrum, it was clear that the 2010 legal
notice regarding mandatory SACCO membership was a critical divider. The definition of the middle period is
associated with a period of expansion for cooperatives in the 2000s. Thus three periods were defined: 1990-
1999; 2000-2009; and 2010-2014. Table 2 indicates the geographical region and year of establishment of each
of the selected case study SACCOs.

4.2 Data collection

Interview guides were developed for the purposes of administering semi-structured personal interviews with
two groups of respondents. The first group were any combination of available senior officials from the
selected case study SACCOs, including elected chairpersons and vice-chairpersons, and operations managers
and chief executive officers in the case of the SACCOs with larger vehicle fleets and non-elected professional
management. A standard interview guide was used for this group to guide discussion, with questions relating
to: the formation of the SACCO; organisational structure; vehicle ownership and management; the benefits of
SACCO membership; interactions with other SACCOs who share routes and termini; and the impact of
government regulations. The second group of respondents were key industry stakeholder informants from: the
Department of Transport within the Ministry of Transport and Infrastructure; the Department of Cooperative
Development and Marketing within the Ministry of Industrialisation and Enterprise Development, the SACCO
Societies Regulatory Authority; Frotcom East Africa, a provider a vehicle tracking systems; and the
Cooperative Insurance Company Insurance Group, a provider of insurance cover specifically for cooperatives.
Each key informant interview guide was composed of three general questions concerning SACCO registration
requirements and the quality of SACCO management and services, and a further set of specific questions on
their involvement with SACCOs tailored to the nature of their organisation.

SACCO official interviews were conducted between April and May 2014, and industry stakeholder informant
interviews were conducted between February and May 2014. Interviews were conducted in the offices of the
respondents concerned, with one or two researchers present and a field assistant responsible for transcribing
the interview. In the case of the SACCO official interviews, the interview record was complemented with the
researchers’ site observations with respect to vehicle branding, the location and condition of the SACCO’s
terminus, the staffing of the SACCO office, and the use of computers and telephones in operations
management. The mean duration of interviews with SACCO officials was in the region of two to three hours,
and in most instances included an accompanied tour of the relevant SACCO’s offices and terminus. A

7
stakeholder workshop was held on 7 August 2014 at which participants in the study were given an opportunity
to discuss and provide feedback on preliminary findings.

5. Case study findings

Selected findings of the thematic analysis undertaken of the interview records are reported in terms of:
SACCO organisation, size and resources; driver employment and management; member benefits, particularly
with respect to vehicle acquisition and maintenance; and service operations management.

5.1 Organisation and membership

SACCO organisational structure is largely dictated by the Cooperative Societies Act, which stipulates basic
operating rules and procedures. Important amongst these are that SACCOs prepare annual estimates of income
and expenditure, and keep accounts which must be audited. The governing body of every SACCO is a
Committee, which is subject to an Annual General Meeting (AGM). The Committee has between five and
nine members, from which the members elect a chairperson and a vice-chairperson. According to the Act,
members must be over 18 years of age, and resident within the society’s area of operation. No member may
hold more than 20% of the ‘share capital’ (i.e. the SACCO’s accumulated capital savings reserve), and
members must make regular payments to the SACCO.

With one exception (S5), all the case study SACCOs reported that members make contributions to the
SACCO’s ‘share capital’ on a per trip, daily, weekly or monthly basis. ‘Share capital’ is typically invested in
the Cooperatives Bank of Kenya, from which loans to members are advanced, and from which annual
dividends are sometimes paid. The ‘share’ contribution is set during the AGM, and was found to range
between KES 2 000/vehicle/month (S10) and KES 11 700/veh/m (S1) (USD 20.60/veh/m and
USD 120.51/veh/m at the mean April-May 2014 rate of exchange respectively). Two SACCOs (S6 and S9)
reported that a member could voluntarily increase his or her contribution in order to qualify for bigger loans.
In the case of S5, member contributions are used solely for the purposes of employing SACCO staff, but it
was reported that plans are in place to initiate ‘share capital’ contributions at a later date.

The financial strength of SACCOs can be measured by the size of their ‘share capital’. From the financial
information provided it appears that capital reserves range considerably, from KES 189 million (S1) to
KES 150 000 (S8) (USD 1 946 700 and USD 1 545 respectively). While the sample was only 2% of registered
SACCOs, and statistical representativeness therefore cannot be claimed, figure 1 suggests an unsurprising
positive correlation (r=-0.913) trend between SACCO age and the size of ‘share capital’.

KES 200 100

S1
KES 175
Kenyan Shillings (millions)

80
KES 150
KES 125
index score

60
KES 100
KES 75 40
S9
S4
KES 50 share capital
20
KES 25 service operations improvement index (1)
S3 S6
trendline (share capital reserve) S8 S5
KES 0 0
1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

2014

year of SACCO establishment

Notes:

8
1.
The ‘service operations improvement index’ is calculated on the (unweighted) basis of whether the SACCO has adopted the
following practices or technologies: vehicle tracking; speed governors; cashless fare collection; collective fare management;
salaried drivers; collective driver employment; and driver monitoring. Each practice or technology adopted was assigned a value
of one, and an index out of 100 was calculated by dividing score achieved by the maximum possible score, and multiplying by
100.
2.
mean April-May 2014 rate of exchange = KES 1.0000:USD 0.0103)

Fig. 1. SACCO case study ‘share capital’ reserves and service operations improvement, by year of
establishment

It was found that, in addition to vehicle crews, the SACCOs employed managers, office staff and terminus
staff. In the more established SACCOs (e.g. S1 and S9) managers are employed full-time and hold
postgraduate degrees in business administration. Office staff in the larger SACCOs take the form of
secretaries, accountants, information technology support technicians and clerks responsible for the booking
office, while in the smaller SACCOs one or two people perform many or all of these tasks. Terminus staff,
typically employed on a casual basis, were found to manage the queuing and departure of vehicles, load
luggage and in some cases handle packages sent via side-business courier services.

Table 2
Summary of SACCO case study organisational and membership characteristics

S1 S2 S3 S4 S5 S6 S7 S8 S9 S10
Year of establishment 1993 2003 2004 2000 2011 2011 2010 2010 2003 2011
1
Geographical region C E E C W E C C E W
No. of SACCO office staff 38 2 n.i. 23 30 n.i. 27 n.i. 13 25
No. of routes operated 10 1 1 1 2 1 2 3 2 1
No. of members 700 26 40 150 29 15 35 16 125 3
No. of vehicle-owning members 350 26 40 150 29 15 35 16 125 3
Total number of vehicles 600 63 100 204 55 30 56 31 200 40
No. of member-owned vehicles 598 60 94 200 55 30 56 31 200 n.i.
No. of SACCO-owned vehicles 2 3 6 4 0 0 0 0 0 n.i.
Limit of member vehicle fleet 6 n.i. 3 4 none none none none n.i. n.i.
'Share capital' (KES millions) 2 189.0 n.i. 3.0 60.0 0.0 0.7 n.i. 0.2 62.0 n.i.
'Share' payment 11.70 5.20 n.i. 3.00 0 2.25 n.i. 7.80 n.i. 2.00
(KES thousands/veh./month) 3
Notes:
1.
C=central region (incorporating the towns of Embu, Murang’a, Nyeri, Ruiru and Thika); W=western region (incorporating the
towns of Eldoret, Kakamega, Kisumu and Nakuru); E=eastern region (incorporating the towns of Malindi and Mombasa)
2.
mean April-May 2014 rate of exchange = KES 1.0000:USD 0.0103)
3.
‘share’ payment comparison estimation assumptions: 26 vehicle operating days in 1 month; 6 vehicle operating days in 1 week;
1.5 vehicle trips in 1 vehicle operating day

SACCO membership and vehicle fleets were found to vary widely, from 700 (S1) to three (S10) in the case of
members, and from 600 (S1) to 30 (S6) in the case of vehicles. As in the case of the indicative trend observed
in ‘share capital’, unsurprisingly, the older SACCOs were found to have more vehicle-owning members (r=-
0.893) and larger vehicle fleets (r=-0.888) (see figure 2). Some SACCOs reported limiting the number of
vehicles a single member can own to between three (S3) and six (S1), in order to prevent larger businesses
from dominating the cooperative. A strong correlation (r=0.992) was therefore observed between the number
of vehicle-owning members and the size of vehicle fleets (see table 2). Outliers are S10 which has only three
members and a member to vehicle ratio of 1:13, and S1 which reported that half of their 700 members are no
longer active in the public transport business, but maintain their membership in order to continue saving and
qualifying for loans.

9
 600 S1 total number of vehicle-owning members
total number of vehicles
500
trendline (vehicle-owning members)
400 trendline (vehicles)
count

300
S9
200 S4
S7 S5
S3
100 S8
S10
S2
S6
0
1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

2014
year of SACCO establishment

Fig. 2. SACCO case study membership and fleet, by year of establishment

5.2 Driver employment

Despite the legal notice in 2004 requiring vehicle crews to be employed on a permanent salaried basis (as part
of the ‘Michuki rules’), most matatu drivers in Kenya have continued to be remunerated on the basis of the
‘target system’, in which drivers keep whatever fare box revenue remains after paying fuel costs and the
vehicle owner a fixed daily rental. As posited earlier, this system underlies many service quality problems,
and moving to a different driver remuneration model is a likely precondition for service quality improvement.
The appeal of SACCOs, for the Kenyan regulatory authorities, was that they potentially facilitate a shift away
from the ‘target system’, through collective vehicle and driver rostering and collective fare collection. It is
therefore instructive to explore how the inter-city SACCOs remunerate and manage their drivers.

The case studies indicate that the inter-city SACCOs are indeed moving away from the ‘target system’ (see
table 3): 60% of the case SACCOs reported that drivers are remunerated on a salaried basis (S1, S2, S3, S5,
S7 and S10); and two of the five SACCOs that are still paying drivers on a daily target basis reported that they
are making preparations to introduce salaried remuneration (S6 and S9). Three SACCOs reported that drivers
are recruited and employed by the SACCO (S2, S3 and S10), while six reported vetting driver recruitment by
individual members (S1, S2, S4, S5, S6 and S7).

“The owner used to pay the driver but they now take the salary of the driver to the SACCO. The driver will have
to be paid whether the vehicle is on the road or not”. (S2 SACCO manager)

“Drivers will earn KES 15 000/m (USD 154.50) and there will be an agreement form that they will have to sign.
The owner will have to bring KES 500/day for the salary of the driver. He will also have to provide the driver’s
breakfast and lunch”. (S3 SACCO official)

As noted by the S2 SACCO manager above, shifting to salaries means that the driver is paid irrespective of
fare box revenue. Delinking driver income from passenger ridership raises the question of how the SACCOs
ensure that drivers continue to attract and carry passengers in what are often intensely competitive markets.
Table 3 reveals that, with one exception (S3), the six SACCO case studies with salaried drivers reported
tracking vehicle movements in real-time, thus enabling SACCO officials or vehicle owners to detect whether
drivers stop providing service at any time in the day. Also with one exception (S7), these SACCOs proactively
and reactively monitor driver behaviour, with associated penalties in the form of warnings and fines. Proactive
monitoring takes the form of SACCO staff observing driver behaviour on the road, while reactive monitoring
takes the form of a passenger complaints systems. Further, with one exception (S7), the SACCOs with
salaried drivers all collect fares via SACCO staff. Under collective fare collection conditions, shifting drivers

10
to salaries is likely to receive less resistance than in instances where vehicle crews collect cash fares and
underreport daily fare box revenue in order to increase their share.

Table 3
Summary of SACCO case study operations management and member benefit characteristics

S1 S2 S3 S4 S5 S6 S7 S8 S9 S10
Real-time vehicle monitoring Y Y Y Y Y Y Y Y
Speed governors Y Y Y Y Y Y Y Y Y Y
Fares collected by SACCO Y Y Y Y Y Y Y Y
Salaried drivers Y Y Y Y Y Y
Drivers employed by SACCO Y Y Y
Drivers monitored by SACCO Y Y Y Y Y Y Y Y
Max. loan (KES millions) 1 3.0 n.i. 0.8 2.0 none none none none 4.0 n.i.
Loans for vehicle purchase Y n.i. Y Y n.i.
Loans for vehicle repairs Y Y Y Y Y Y Y Y Y
Notes:
1.
mean April-May 2014 rate of currency exchange = KES 1.0000:USD 0.0103)

5.3 Member benefits

Earlier it was posited that the day-to-day, cash-based nature of most paratransit business management fails to
recognise vehicle depreciation as an operating expense, which results in a frequent inability to replace
vehicles when it is optimal or necessary from a roadworthiness perspective. The appeal of SACCOs, for the
Kenyan regulatory authorities, was that they facilitate a form of compulsory savings for vehicle replacement
and repair, through ongoing ‘share’ contributions to the SACCO’s capital savings reserve. In other words,
prior ‘share’ contributions approximate a form of vehicle depreciation costing. It is therefore instructive to
explore how the inter-city SACCOs utilise their capital reserve to renew and maintain their vehicle fleets.

The case studies revealed that the inter-city SACCOs play an important role in advancing loans to acquire (or
at least finance a bank’s down-payment requirements) and to repair vehicles. The SACCOs with larger capital
reserves (>KES 60 million) reported making loans available to members for vehicle purchase (S1, S4, S9),
while the SACCOs with less ‘share capital’ are only capable of providing loans for vehicle repair and
maintenance (see table 3). The amount of a loan was in most cases determined by three factors: the ‘share
capital’ available; the loan ceiling; and the applicant’s accumulated ‘share’ contribution. The SACCOs
typically set the loan ceiling as double the applicant’s accumulated ‘share’ contribution, subject to maximums
which ranged from KES 4 million (S9) to KES 800 000 (S3) (USD 41 200 and USD 8 240 respectively). A
member must typically contribute ‘shares’ for at least 6-12 months before he or she can qualify for a loan.
Loans attract interest at the rate of 1% per month on the reducing balance.

Some SACCOs have invested in collectively owned vehicles (see table 2). In most cases, however, the
number of SACCO-owned vehicles is small relative to the number owned by members (e.g. S2 reported
acquiring three 51-seater buses, but has 60 14-seater vehicles owned by members). In other cases, the number
owned by the SACCO is substantial (e.g. S10 reported that the SACCO from which it split had a fleet of 99
collectively owned vehicles).

Further benefits to SACCO members take the form of the provision of loans to assist members with
unforeseen personal expenses, legal representation when they are charged in the courts, and arrangements for
insurance cover. All the SACCOs reported advancing short-term (<1 year) loans for personal business or
welfare purposes, subject to the extent of the applicant’s prior ‘share’ contributions and various loan ceilings.
Legal assistance comes in the form of either hiring a lawyer for the member in need (S1, S3, S5, S6, S9 and
S10), or extending financial assistance for this purpose (S7 and S8). It was found that third party liability

11
insurance services are either arranged by the SACCO for all the members from one source (S2, S5 and S10),
or by operators individually using the same insurance brokerage specified by the SACCO (S4, S6, S7 and S9).
New vehicles purchased via bank loans generally do not require separate third party insurance via the
SACCO, as the banks insist upon comprehensive insurance while the loan is being repaid.

To minimise income leakage and generate additional revenue, some SACCOs have established insurance
agencies as a service to their members (S1, S4 and S10). Other initiatives reported to diversify SACCO
income included: the ownership of fuel stations (S1, S2, and S9); dealing in vehicle parts (S4); the operation
of a financial service (S1); and the operation of mobile telephone money transfer services (M-Pesa) (S1 and
S4).

5.4 Operations management

Each of the SACCOs studied demonstrated a willingness to adopt innovative service operations practices and
technologies. With respect to vehicle and crew rostering, in most cases the allocation of vehicles to routes is
determined by the PSV licence, and therefore the SACCO does not perform this role.

“We are restricted by the by-law; if you do not have a sticker to terminate at Thika then you cannot go there”.
(S4 manager)

All the SACCOs reported that they manage termini in order to ensure fair vehicle rostering with respect to
departures during peak periods (either on a rotational or ‘first-come first-served’ basis), but SACCO-owned
vehicles are given priority (S1). In instances where drivers are employed by the SACCO (S2, S3 and S10), the
SACCO assigns drivers to vehicles.

“If you have a vehicle in the SACCO, you will not know who will drive the vehicle” (S10 official).

As noted earlier fleet management systems have been a requirement for matatu businesses operating long-
distance night-time services since 2012 (for intra-city services they are optional). Seven of the SACCOs
reported that these systems have been installed in their vehicles (S1, S2, S3, S4, S5, S6, S8 and S10), and one
was in the process of installation at the time of the interview (S9). In some cases vehicles can be monitored by
the SACCO staff, as well as by the respective owners, via laptop or mobile phone (S1, S3, S5, and S10). In
other cases just the owners can track their vehicles via mobile phone (S4 and S8). It was reported that tracking
data are used to ensure that vehicles stick to their designated routes, and adhere to speed limits (S1, S4, S9,
and S10).

All SACCOs reported that vehicles have been fitted with speed governors (as required by a legal notice in
2003 as part of the ‘Michuki rules’), with some reporting that they had purchased speed governors for
members who could not afford them (S4 and S9). Speed governors were reported to have reduced vehicle fuel
consumption costs, and to provide useful evidence in refuting false speeding charges (S1, S4, and S9).
Concerns were raised, however, with the expense and timeframes for compliance.

“The time limit for installing speed governors was not adequate, given that the market did not have enough
devices. The device costs KES 38,000 (USD 391). There are cartels which have over-priced the devices …
There are 20 companies licensed to sell speed governors, most of these are linked to officials in government”.
(S5 SACCO official)

While some matatus in Nairobi have introduced (and subsequently withdrawn) card payment systems (e.g. the
BebaPay and My1963 cards), none of the case study SACCOs reported adopting cashless fare collection. The
SACCOs reported collecting fares in cash, and issuing paper tickets at their ticket offices at route termini.
When a vehicle makes intermediate stops, fares are either collected at ticket offices located at stops, or on-
board by the vehicle crew. Some SACCOs collect fares and deduct from the fare box revenue what is due to
it, in the form of member payments, before remitting the balance to the member. Disbursements to members
are made either via the driver with cash and a delivery note which gives a breakdown of the deductions, or via
electronic funds transfer (S4, S7 and S9).

12
Fare setting practices were found to vary across SACCOs. Some claimed to be pace setters (S1 and S9), some
follow the general trend (S2, S3, S5 and S7), while others reported waiting for competitors to change fares and
then undercutting them (S10). One SACCO has its fares set by an umbrella association of which it is a
member (S4).

Most of the SACCOs interviewed reinforced a common view that corruption is widespread in the Kenyan
public transport sector, that operators frequently encounter demands for bribes which they have to pay in order
to stay in business, and that increasing fines simply results in increased bribes (S2, S3, S4, S5, S6, S7, S8,
S10). A motivation for cashless, or ‘cash-lite’, fare collection noted by a key informant in the Ministry of
Transport and Infrastructure was that it may reduce corruption, as traffic police will know that vehicles are not
carrying large amounts of cash.

“We are harassed by traffic police, when our bus goes on the road; police stop it so that they can get a bribe. If
the driver does not give a bribe then the bus can be charged for anything. The magistrate in court follows what
the traffic police have written as the offence. We are forced to give traffic police KES 600 (USD 6.18) on a
weekly basis. … People would rather pay a bribe than go to court to pay a high fine.” (S2 SACCO official)

“The traffic police have taken advantage of the increased fines; they see it as a loophole. Before we used to give
them KES 50 (USD 0.52) every day. They were ready to accept KES 1 000 (USD 10.30) when you commit an
offense. Today no traffic police is taking less than KES 2 000 (USD 20.60). Some even ask for KES 20 000
(USD 206). Since the fines went up the bribe has also gone up. The government is encouraging corruption”. (S7
SACCO official)

6. Discussion

The absence of dedicated rights-of-way make heavily congested Kenyan cities, and Nairobi in particular,
hostile environments within which to operate rapid and reliable public transport services. Analysis of geo-
location data collected for the purposes of constructing a network map of matatu services (Williams et al,
2015), revealed that over most of the network within the Nairobi city centre mean commercial speeds during
the morning peak range between five and 10 kilometres/hour, and drop to less than 5 km/h in some sections
(UONBI et al, 2013). Despite such operating conditions over at least a part of their routes, the formation of
cooperatives has evidently improved the quality of inter-city services. Longitudinal quantitative data on
service attributes not directly determined by congested rights-of-way (e.g. vehicle roadworthiness, vehicle
collisions, passenger comfort, fare fluctuation, etc.) are not readily available to measure the extent of this
improvement, but as noted earlier, the public transport planning and regulating authorities at least have
recognised that this improvement has occurred. The question which arises then is: how transferable might
these successes amongst inter-city services be to other contexts?

There is certainly current policy interest in pursuing the reform and consolidation of paratransit services
through cooperatives, both within Kenya and elsewhere. Within Kenya, as noted earlier, SACCO (or limited
liability company) membership became mandatory for both inter- and intra-city services in 2010. Elsewhere,
the South African Department of Transport has recently expressed renewed interest in cooperatives as a means
of organising parts of its minibus-taxi industry. In 2001, the responsibility for developing cooperatives in
South Africa was transferred from the Department of Agriculture to the Department of Trade and Industry, to
widen the promotion of cooperatives across all economic sectors. The policy framework subsequently
developed includes a new Cooperatives Act (14 of 2005), and a policy, regulations and a 10-year strategy
produced in 2004, 2007 and 2012 respectively (DTI, 2012). In the public transport sector more specifically, a
‘roadmap’ was prepared in 2015 to stimulate struggling cooperatives (DoT, 2015).

The transfer of the innovations and practices that have led to inter-city matatu successes to other contexts
may, however, be difficult.

Within Kenya, transfer of these innovations and practices, particularly with respect to fare collection and
driver remuneration, to intra-city services is likely to be made difficult by different on-route passenger

13
boarding and alighting profiles. Whereas in the inter-city services the bulk of passengers board at the origin
terminus (with only a small amount of boarding occurring on-route at established stops), in the intra-city
services on-route boarding and alighting is far more frequent and erratic. Consequently cash fare collection or
ticket sales by the vehicle crew, as opposed to by terminus staff, is far greater in the latter. Given that the
dominant means of driver remuneration amongst intra-city service providers is the ‘target system’, there has
been, and is likely to continue to be, considerable resistance on the part of vehicle crews to accept a business
operating model in which they lose the ability to collect cash fares directly, and thereby illicitly increase their
income by failing to disclose the full daily fare box revenue to the vehicle owner.

Outside Kenya, transfer of the cooperative business operating model more generally is likely to be made
difficult by considerably less institutional support and a weaker cooperative culture, and the existence of other
solutions developed to mitigate problems of vehicle financing. In South Africa more specifically, public
transport cooperative experiences have achieved mixed results. While there are examples of cooperatives that
have endured (e.g. the Greater Rustenburg Transport Cooperative, established in 2002, which opened a fuel
station in partnership with a petrochemical company in 2013), there are more examples of cooperative
initiatives that have collapsed. Perhaps most notable amongst these are the failed attempts to establish
cooperatives in KwaZulu-Natal and Port Elizabeth. The initiative in KwaZulu-Natal involved the provincial
government establishing 14 paratransit cooperatives in 1999 to facilitate entry into the vehicle manufacturing
industry. By 2001 11 of the 14 cooperatives had collapsed (Jennings et al, 2016). The initiative in Port
Elizabeth in 2009 involved the Nelson Mandela Bay Metropolitan Municipality creating five cooperatives that
consolidated existing paratransit associations, and one overarching cooperative to coordinate their activities,
to facilitate the introduction of a bus rapid transit service network. Progress was impeded by the murder of the
leader of the paratransit negotiating body in 2010, and concerns over the maladministration of start-up funds
provided to the secondary cooperative, ultimately leading to a curtailment of the reform process. (Jennings
et al, 2016).

A further potential difficulty in transfer, both within and outside Kenya, relates to government involvement in
cooperative establishment. In Kenya, the inherent contradiction between the voluntary nature of the SACCO
concept and its mandatory status in public transport regulations has yet to be fully understood, but it is
possible that the experience of voluntary and mandatory SACCOs will ultimately be different (McCormick
et al, 2015b). Certainly the South African Department of Transport has concluded that amongst government-
initiated cooperatives, the recipients of assistance have rarely exhibited the initiative and ownership required
to sustain the enterprise, and this has been a key contributor to the widespread failure of cooperatives in the
paratransit sector (DoT, 2015).

With respect to the interest in developing world paratransit systems serving as a potential model for demand
response transit in developed world city regions, espoused by various authors in the earlier literature review
(Silcock, 1981, Roth and Shephard, 1984, Chujoh, 1989, Cervero 1997, Finn, 2012), the Kenyan SACCO
model is unlikely to offer important lessons. The challenges associated with vehicle roadworthiness, labour
exploitation and illegal driver behaviour prevalent in Kenya are unlikely to be experienced to the same degree,
if at all, in developed world city contexts, and consequently the benefits of cooperative structures with respect
to fleet renewal and salaried driver remuneration are unlikely to accrue in these contexts.

7. Conclusion

This paper set out to gain insight into how the Kenyan inter-city matatu SACCOs are organised and have
improved services, and to explore the transferability of this experience and the lessons it offers. With regard to
the first aim, the study demonstrated that variation in size, financial resources, employment practices, vehicle
management and member benefits is considerable, and that there is therefore no single model. It is concluded
that the organisation of inter-city matatus into SACCOs has been an important step in rationalising service
and improving quality. They provide a means of imposing a form of vehicle depreciation costing through
‘share’ contributions, provide finance for fleet renewal, repair and maintenance, and, when drivers are salaried
and collectively managed, they offer a means of removing the negative driver behaviour incentives associated
with the ‘target system’. SACCO dividends from diversified business interests provide additional income, and
member benefits and services reduce vulnerability to the risks that typically afflict small paratransit

14
businesses. In recent years SACCOs have also demonstrated a readiness to adopt vehicle management and
fare collection technologies, which dovetail government efforts to improve regulation and counter corruption,
provided they make business sense.

With regard to the transferability of, and lessons from, the inter-city SACCO experience, it is argued that due
to the particular shuttle-like nature of inter-city services and the considerable institutional support that exists
in Kenya for cooperatives, the direct transfer of successes to other contexts is likely to prove difficult. It is
therefore perhaps more useful to isolate the specific features of the inter-city SACCO model that have led to
paratransit service improvements and attempt to replicate these, rather than necessarily the model in its
entirety. It is posited that these key features are operator consolidation, accompanied by salaried drivers,
systematic vehicle monitoring and compulsory vehicle depreciation costing. The benefits of consolidation
range from the management of destructive ‘in the market’ competition, to bulk purchasing benefits, and
ultimately the ability to enter into some form of formalised service contracting. Properly incentivised and
monitored salaried drivers are key to curbing dangerous driving practices and excessive in-vehicle passenger
waiting times, and depreciation costing is key to maintaining roadworthy vehicle fleets. Many, if not all, of
these features can be adopted in other forms of paratransit organisation and regulation, but will require
considerable adaptation to context. These lessons from the Kenyan SACCO experience are, however, most
likely to be applicable to paratransit in other Sub-Saharan African countries and, possibly to a lesser extent, to
developing countries elsewhere.

It is remarkable that a few voluntarily formed cooperatives have inspired an entirely new way of regulating
public transport in Kenya. Mandatory SACCO formation requirements have coincided, however, with other
directives, such as the installation of digital speed governors, fleet management systems and cashless fare
payment, which some operators have found too expensive or too complex. Perhaps the overriding lesson from
the Kenyan SACCO experience is therefore that operators need some time to absorb new regulatory
requirements, before others are introduced. Indeed, a strength of the SACCO model is perhaps its more
incremental path to reform, compared to the capital and capacity intensive approaches that have dominated
Sub-Saharan public transport policy discourse in recent years (Behrens et al, 2016).

A variety of research needs arise from the study conducted. A first is the need to investigate the willingness of
both intra-city service vehicle crews and owners to abandon the ‘target system’ in favour of salaried driver
employment, exploring the trade-off between greater driver control over collection and more predictable and
secure employment terms, and whether the additional fare box revenue anticipated by vehicle owners would
accrue to themselves or to drivers as a result of less exploitative labour relations. A second research need is to
measure the benefits of the new practices and technologies identified in the study for both operators and
passengers. A limitation of the cross-sectional case study method employed in this study is its inability to
produce quantifiable metrics of benefit and change (e.g. ridership, vehicle productivity, operating costs,
revenue, reliability, overcrowding, etc.). Matatu passenger satisfaction surveys, and longitudinal panel studies
of SACCOs, hold promise in this regard. A third research need is to explore what drives the adoption of new
service operations improvement technologies and practices. Notwithstanding the limits of the small sample of
SACCOs, figure 1 reveals little or no correlation (r=-0.189) between the adoption of technologies and
practices and SACCO age, and a very weak correlation (r=-0.427) with SACCO ‘share capital’. Clearly there
are other factors which influence whether or not SACCOs innovate. What motivates paratransit operators to
adopt service operations improvement technologies and practices, and the implication this has for public
transport policy and regulation, should also therefore be better understood.

Acknowledgements

The research presented in this paper was made possible through funding from Development Workshop-
Angola, originating from the Bill and Melinda Gates Foundation. The authors also acknowledge the generous
contribution of time by officials at each of the case study SACCOs, as well as input from key stakeholder
informants who helped shape the study and interpret its findings.

15
References

Adeniji, K. (1987) Para-transit modes in Nigeria: Problems and prospects, Cities, 4(4), pp. 290-392.
doi:10.1016/0264-2751(87)90094-1
Aduwo, G. (1990) Production, efficiency and quality of service of the matatu mode of public transport in
Nairobi, Kenya: A geographical analysis. Department of Geography, University of Nairobi.
Agyemang, W., (2013) ‘Measurement of service quality of ‘Trotro’ as public transportation in Ghana: A case
study of the city of Kumasi’, 32nd Southern African Transport Conference, Pretoria.
Behrens, R. and Schalekamp, H., (2010) ‘Public transport mode satisfaction in Cape Town: Findings of a
passenger intercept survey’, 29th Southern African Transport Conference, Pretoria.
Behrens, R., McCormick, D. and Mfinanga, D., (2016) ‘An introduction to paratransit in Sub-Saharan African
cities’, in Behrens, R., McCormick, D. and Mfinanga, D. (eds), Paratransit in African cities:
Operations, regulation and reform, Routledge, New York.
Brake, J. and Nelson, J. (2007) A case study of flexible solutions to transport demand in a deregulated
environment, Journal of Transport Geography, 15(4), pp. 262-273.
doi:10.1016/j.jtrangeo.2006.08.006
Bruun, E., Del Mistro, R., Venter, Y. and Mfinanga, D., (2016) ‘The state of public transport systems in three
Su-Sahran African cities’, in Behrens, R., McCormick, D. and Mfinanga, D. (eds), Paratransit in
African cities: Operations, regulation and reform, Routledge, New York.
Cervero, R. (1991) Paratransit in Southeast Asia: A market response to poor roads, Reprint No. 90, Review of
Urban and Regional Development Studies, 3, pp. 3-27. doi: 10.1111/j.1467-940X.1991.tb00076.x
Cervero, R. (1997) Paratransit in America: Redefining mass transportation, Praeger, Westport.
Cervero, R. (2000) Informal transport in the developing world, United Nations Commission on Human
Settlements, Nairobi.
Cervero, R. and Golub, A. (2007) Informal transport: A global perspective, Transport Policy, 14(6), pp. 445-
457. doi:10.1016/j.tranpol.2007.04.011
Chujoh, U. (1989) Learning from medium- and small-sized bus services in developing countries: Is regulation
necessary?, Transportation Research Part A, 23(1), pp. 19-28.
DoT, (2015) ‘Final co-operative and enterprise model for road-based public transport industry’, South African
Department of Transport, Pretoria.
DTI, (2012) ‘Integrated strategy on the development and promotion of co-operatives: Promoting an integrated
co-operative sector in South Africa 2012-2022’, Department of Trade and Industry, Republic of South
Africa, Pretoria.
Dugard, J. (2001) From low intensity war to mafia war: Taxi violence in South Africa, Violence and transition
series, Vol. 4, Centre for the Study of Violence and Reconciliation, Johannesburg.
Enoch, M., Potter, S., Parkhurst, G. and Smith, M., (2004) Intermode: Innovations in demand responsive
transport, Department for Transport and Greater Manchester Passenger Transport Executive.
Envag Associates, (2012) ‘PSV demand, termini capacities and compliance level with TLB regulations in
Nairobi metropolitan area’, Transport Licencing Board, Ministry of Transport, Nairobi.
Finn, B. (2012) Towards large-scale flexible transport services: A practical perspective from the domain of
paratransit, Research in Transportation Business and Management, 3, pp. 39-49.
doi:10.1016/j.rtbm.2012.04.001
Gicheru, E., Migwi, W. and M’Imanyara, K., (2011) ‘An analysis of the socio-economic impact of the new
integrated national transport policy: The case of 14-seat transport SACCOS in Kenya’, International
Cooperative Alliance Global Research Conference, Mikkeli.
Golub, A., Balassiano, R., Araújo, A. and Ferreira, E. (2009) Regulation of the informal transport sector in
Rio de Janeiro, Brazil: welfare impacts and policy analysis, Transportation, 36, pp. 601–616. doi:
10.1007/s11116-009-9215-y
Govender, K., (2014) ‘Public transport service quality in South Africa: A case study of bus and mini bus
services in Johannesburg’, African Journal of Business Management, 8(10), pp. 317-326.
http://dx.doi.org/10.5897/AJBM2014.7416
ICA, (2006) ‘Annual report 2006’, International Co-operative Alliance, Geneva.
Jennings, G., Bruun, E., Orero, R. and Browning, P., (2016) ‘Strategy options for paratransit business
development and service improvement’, in Behrens, R., McCormick, D. and Mfinanga, D. (eds),
Paratransit in African cities: Operations, regulation and reform, Routledge, New York.

16
JICA, (2006) The study on the master plan for urban transport in the Nairobi Metropolitan Area in the
Republic of Kenya. Katahira and Engineers International, Recs International Inc., Japan International
Cooperation Agency.
Joubert, J. (2013) Gauteng: Paratransit – perpetual pain or potent potential, in Institute for Mobility Research
(ed), Megacity mobility culture, Springer, Berlin.
Joubert, J., (2013) ‘Gauteng: Perpetual pain or potent potential?’, Institute for Mobility Research (ed)
Megacity Mobility Culture, Springer-Verlag, Berlin.
Kanyama, A., Carlsson-Kanyama, A., Lindén, A. and Lupala, J. (2004) Public transport in Dar es Salaam,
Tanzania: Institutional challenges and opportunities for a sustainable transportation system,
Institutionen för Miljöstrategiska Studier, Stockholm.
Khayesi, M. (1999) The struggle for regulatory and economic sphere of influence in the matatu means of
transport in Kenya: A stakeholder analysis, 6th Thredbo International Conference on Competition and
Ownership in Land Passenger Transport, Cape Town.
Khosa, M. (1992) Routes, ranks and rebels: Feuding in the taxi revolution, Journal of Southern African
Studies, 18(1), pp. 232-252. doi:10.1080/03057079208708312
Khosa, M. (1995) Transport and popular struggles in South Africa, Antipode, 27(2), pp. 167-188. doi:
10.1111/j.1467-8330.1995.tb00271.x
Kimani, P., (2007) ‘Sources of competitive advantage for transport savings and credit co-operative societies in
the public service vehicle industry: The case of MTN SACCO Society LTD’, ADM. NO.
D/61/P/8573/2001, Faculty of Commerce, University of Nairobi.
Kirby, R., Bhatt, K., Kemp, M., McGillivray, R. and Wohl, M. (1974) Para-transit: Neglected options for
urban mobility, The Urban Institute, Washington DC.
Klopp’ J. and Mitullah, W., (2016) ‘Politics, policy and paratransit: A view from Nairobi’, in Behrens, R.,
McCormick, D. and Mfinanga, D. (eds), Paratransit in African cities: Operations, regulation and
reform, Routledge, New York.
KNBS, (2010) ‘The 2009 Kenyan population and housing census: Volume 1B Population distribution by
political units’, Kenya National Bureau of Statistics, Nairobi.
Koimur, I., Kangogo, L. and Nyaoga, R., (2014) ‘Assessment of commuter preferences of 14-seater public
service vehicles versus alternative modes of public service transport in Nairobi City’, Journal of
Business Economics and Finance, 3(1), pp. 115-132.
Kumar, A. and Barrett, F., (2008) ‘Stuck in traffic: Urban transport in Africa’, Africa Infrastructure Country
Diagnostic, World Bank, Washington, DC.
Mageean, J. and Nelson, J. (2003) The evaluation of demand responsive transport services in Europe, Journal
of Transport Geography, 11(4), pp. 255-270. doi:10.1016/S0966-6923(03)00026-7
McCaul, C. (1990) No easy ride: The rise and future of the black taxi industry, South African Institute of Race
Relations, Johannesburg.
McCormick, D., Orero, R., Behrens, R. and Ommeh, M., (2015a) ‘Savings and credit cooperative
organisations as operational arrangements for paratransit management: Case studies of inter-city
public transport SACCOs in Kenya’, Report commissioned by Development Workshop – Angola,
Luanda.
McCormick, D., Schalekamp, H. and Mfinanga, D., (2016) ‘The nature of paratransit operations’, in
Behrens, R., McCormick, D. and Mfinanga, D. (eds), Paratransit in African cities: Operations,
regulation and reform, Routledge, New York.
McCormick, D., Orero, R., Behrens, R. and Ommeh, M., (2015b) ‘Cooperatives as a means of paratransit
reform: Case studies of inter-city matatu savings and credit cooperatives in Kenya’, 34th Southern
African Transport Conference, Pretoria.
McCormick, D., Schalekamp, H. and Mfinanga, D., (2016) ‘The nature of paratransit operations’, in
Behrens, R., McCormick, D. and Mfinanga, D. (eds), Paratransit in African cities: Operations,
regulation and reform, Routledge, New York.
Migwi, W., (2011) ‘An analysis of socio-economic impact of the new Integrated National Transport Policy:
The case of fourteen-seater transport SACCOs in Kenya’, Transport Workshop, Kampala.
Mulley, C.. Nelson, J., Teal, R., Wright, S. and Daniels, R. (2012) Barriers to implementing flexible transport
services: An international comparison of the experiences in Australia, Europe and USA, Research in
Transportation Business and Management, 3, pp. 3-11. doi:10.1016/j.rtbm.2012.04.001

17
Muriungi, S., (2013) ‘An investigation of the factors influencing the performance of matatu SACCOS in
Kiambu county: The case of selected matatu SACCOS operating in Thika town, Kenya’,
D53/CE/22169/2010, School of Business, Kenyatta University.
Mutongi, K. (2006) Thugs or entrepreneurs? Perceptions of matatu operators in Nairobi, 1970 to the present,
Africa, 76(4), pp. 549-568. http://dx.doi.org/10.3366/afr.2006.0072
Ommeh, M., McCormick, D., Mitullah, W., Orero, R. and Chitere, P. (2013) The politics behind the phasing
out of the 14-seater matatu in Kenya, 13th World Conference on Transport Research, Rio de Janeiro.
Opiyo, T., (2002) ‘The metamorphosis of Kenya Bus Services Limited in the provision of urban transport in
Nairobi’, SSATP/World Bank, Urban Mobility Component, 12th Steering Committee Meeting,
Maputo.
Orero, R. and McCormick, D., (2011) ‘Organisation of paratransit in Nairobi: Voluntary vs. mandatory
savings and credit co-operatives’, Thredbo 12 International Conference on Competition and
Ownership in Land Passenger Transport (Thredbo Series), Durban.
Orero, R. and McCormick, D., (2013) ‘Cooperatives involvement in the paratransit sector: experience and
lessons in Nairobi’, 32nd Southern African Transport Conference, Pretoria.
Orski, C. (1975) Paratransit: The coming of age of a transportation concept, Transportation, 4(4), pp. 329-
334. doi: 10.1007/BF00174734
Otachi, B. (2013) The influence of entrepreneurial personality, human capital and entry barriers on
performance of entrepreneurs in the informal transport business in Nairobi, Kenya. Unpublished PhD
thesis in Business Administration, School of Business, University of Nairobi.
Palmer, K., Dessouky, M. and Abdelmaguid, M. (2004) Impacts of management practices and advanced
technologies on demand responsive transit systems, Transportation Research Part A, 38(7), pp. 495-
509. doi:10.1016/j.tra.2004.05.002
Rimmer, P. (1984) The role of paratransit in Southeast Asian cities, Singapore Journal of Tropical
Geography, 5(1), pp. 45-62. doi: 10.1111/j.1467-9493.1984.tb00145.x
Rizzo, M. (2002) Being taken for a ride: Privatization of the Dar es Salaam transport system 1983-1998,
Journal of Modern African Studies, 40(1), pp. 133-157.
http://dx.doi.org/10.1017/S0022278X01003846
Roth, G. (1987) The private provision of public services in developing countries, EDI Series in Economic
Development, Oxford University Press, New York.
Roth, G. and Shephard, A. (1984) Wheels within cities: New alternatives for passenger transport, Adam
Smith Institute, London.
Saltzman, A. (1973) Para-transit: Taking the mass out of mass transit, Technology Review, 75(8), pp. 46-53.
Silcock, S. (1981) Urban paratransit in the developing world: Foreign summaries, Transport Reviews, 1(2),
pp. 151-168. doi:10.1080/01441648108716456
Sohail, M., Maunder, D. and Cavill S. (2006) Effective regulation for sustainable public transport in
developing countries, Transport Policy, 13(3), pp. 177-190. doi:10.1016/j.tranpol.2005.11.004
UONBI, CU and MIT, (2013) ‘Survey data processed as of 1/13, Nairobi BRT study’, Presentation,
University of Nairobi, Columbia University and Mass Institute of Technology.
Wa Mungai, M. and Samper, D. (2006) "No mercy, no remorse": Personal experience narratives about public
passenger transportation in Nairobi, Kenya, Africa Today, 52(3), pp. 51-81.
Wanyama, F., (2009) ‘Surviving liberalisation: The cooperative movement in Kenya’, CoopAFRICA Working
Paper No 10, International Labour Organisation, Geneva.
Williams, S., White, A., Waiganjo, P., Orwa, D. and Klopp, J., (2015) ‘The digital matatu project: Using cell
phones to create an open source data for Nairobi's semi-formal bus system’, Journal of Transport
Geography, 49, pp. 39-51. http://dx.doi.org/10.1016/j.jtrangeo.2015.10.005
Woolf, S. and Joubert, J. (2013) A people-centred view on paratransit in South Africa, Cities, 35, pp. 284-293.
doi:10.1016/j.cities.2013.04.005

18