Inherent constraints and creative possibilities : employee participation in Kenya

79418 AUTHOR ACCEPTED MANUSCRIPT FINAL PUBLICATION INFORMATION Inherent Constraints and Creative Possibilities : Employee Participation in Kenya The definitive version of the text was subsequently published in The International Journal of Human Resource Management, 24(10), 2012-09-27 Published by Taylor and Francis THE FINAL PUBLISHED VERSION OF THIS ARTICLE IS AVAILABLE ON THE PUBLISHER’S PLATFORM This Author Accepted Manuscript is copyrighted by the World Bank and published by Taylor and Francis. It is posted here by agreement between them. Changes resulting from the publishing process—such as editing, corrections, structural formatting, and other quality control mechanisms—may not be reflected in this version of the text. You may download, copy, and distribute this Author Accepted Manuscript for noncommercial purposes. Your license is limited by the following restrictions: (1) You may use this Author Accepted Manuscript for noncommercial purposes only under a CC BY-NC-ND 3.0 Unported license http://creativecommons.org/licenses/by-nc-nd/3.0/. (2) The integrity of the work and identification of the author, copyright owner, and publisher must be preserved in any copy. (3) You must attribute this Author Accepted Manuscript in the following format: This is an Author Accepted Manuscript of an Article by McCourt, Willy; Brunt, Carol Inherent Constraints and Creative Possibilities : Employee Participation in Kenya © World Bank, published in the The International Journal of Human Resource Management24(10) 2012-09-27 http://creativecommons.org/licenses/by-nc-nd/3.0/ © 2013 The World Bank  INHERENT CONSTRAINTS AND CREATIVE POSSIBILITIES: EMPLOYEE PARTICIPATION IN KENYA1 Willy McCourt (World Bank, Washington DC, and University of Manchester, UK) Carol Brunt (University of Manchester) 1 We gratefully acknowledge the good nature of the INGO employees and others whom we interviewed, and also a grant from the British Academy which part-funded the field work. Views expressed here are the authors’ own, and should not be attributed to any official agency. The authors can be contacted on carol.brunt@postgrad.manchester.ac.uk and willy.mccourt@man.ac.uk. 1  INHERENT CONSTRAINTS AND CREATIVE POSSIBILITIES: EMPLOYEE PARTICIPATION IN KENYA Introduction In 2009, the UK Secretary of State for Trade and Industry, Lord Mandelson declared in a report commissioned by his department that 'We all know intuitively ... that only organisations that truly engage and inspire their employees produce world-class levels of innovation, productivity and performance' (quoted in MacLeod and Clarke, 2009: 1). And when Baringa Partners, a management consultancy company, won the UK Best Workplace Award for the second year running in 2011, their managing partner highlighted their 'cultural energy rooted in a sense of pride, fairness and involvement. Despite our rapid growth (35% in the previous year), people still feel that they can influence the direction of the company' (Mansour, quoted in Finn, 2011: 8). Yet such powerful convictions fly in the face of longstanding academic scepticism about employee involvement. For Strauss (2006), worker participation is a volatile phenomenon which, having experienced three waves of popularity since World War Two, is currently in a trough (see also Cox et al., 2006).i For Argyris (1998: 105), another veteran observer, employee empowerment programmes (a 'third-wave' initiative in Strauss's terms) are 'the emperor's new clothes' because 'the change programs that could create high levels of ... empowerment do not yet exist.' Yet like Mandelson and Mansour, Strauss and Argyris continue to advocate participation. Despite the ebbs and flows, the belief stubbornly persists. This paper advances our understanding of employee participation in two ways. First, it reports empirical findings from a sector which has a major impact on global wellbeing: the development sector and in particular the INGOs which have their bases in industrialized countries but operate mostly in developing countries. We shall see that their distinctive approach offers instructive comparisons with the mainly for-profit private companies in industrialized countries which have been the venue for most of the participation studies in the HR literature. 1 Second, our paper brings together two widely separated discourses for the first time: in HR and international development, the latter dealing with the participation of development programme 'beneficiaries' (the communities and individuals for which development agencies provide services or act as advocates). Wilkinson and Fay (2011: 66) have observed that 'scholars (of participation) from diverse traditions often know relatively little of the research that has been done in other areas.' It will surprise some readers of IJHRM to learn that we owe the term 'voice', which in recent years has largely overtaken 'participation' in the HR literature, to the international development literature, through Hirschman's (1970) analysis of goods transport on Nigerian railways (Dundon et al., 2004: 1151). HR scholars of participation can repay the debt they already owe to international development by bringing our expertise to bear on development questions such as those which are treated in this article. Our study questions are: 1. What are the styles of employee participation in INGOs operating in Kenya? 2. How should we understand them? 3. What do we learn from the case study about the potential and constraints of employee participation? INGOs in international development Increasingly over the past 30 years, INGOs have facilitated aid distribution through delivery of personal services in poor countries. From the 1980s onwards, they have been favoured by official donor agencies such as the World Bank and the UK's Department for International Development (DFID), who have seen them as flexible, cost-effective and reliable (Chambers, 1987; Hyden, 1983). In 2009 17.3%, or US 2.1 billion, of international humanitarian assistance was channeled through NGOs (Global Humanitarian Assistance, 2011). 2 Kenya's capital, Nairobi, as a transportation hub hosts the regional offices of some of the world’s largest INGOs, such as Oxfam International, CARE International and World Vision. By 2007, there were roughly 4500 national and international NGOs operating in Kenya, four times as many as in 1997. The sector makes a significant contribution to the Kenyan economy, employing 291,000 full-time equivalent employees compared with roughly 677,000 for the entire Kenyan public sector. (Kanyinga et al., 2007). For our case study we sampled at least one of each of the three major NGO types: ‘relief and welfare agencies’, whose raison d’être is emergency relief; ‘public service contractors’, which are funded by donors to implement their programmes; and ‘popular development agencies’, which work with Southern agencies to promote social development and grassroots democracy (Clark, 1991). All the NGOs sampled are international in scope, with headquarters in either North America or Europe (one agency had moved its operating base from Europe to Nairobi). Employee participation In this section we provide a framework for making sense of the research findings we will present later on. We review the participation debate: definitions of participation, its degrees of penetration in different organizations and its practical forms; the main rationales that have been advanced for it; two inherent constraints on its scope; and a number of contingencies that affect it in different places. Definition, degrees and forms ‘Hirschman was exercised by the problem of how to improve the performance of an organization when the economic stimulus of customer exit was unavailable or ineffective. ‘Voice’ was his solution. ‘Voice’ refers to feedback from service users or customers on the quality of the service they have received. Hirschman’s notion became influential in international development, notably in the development across all three continents of the South of ‘citizen report cards’ for the 3 performance of public services (Paul, 1992; World Bank, 2004). It was imported into the industrial relations literature by Freeman (1980). Identifying it with collective bargaining between unions and employers, Freeman found that it reduced staff turnover, or ‘exit’ in Hirschman’s terms. It was the ‘exit-voice tradeoff’ which prompted him to adopt Hirschman’s neologism. Freeman was true to Hirschman’s original usage in that he was interested in workers being able to influence, and not merely express an opinion on (as is sometimes the case in the HR discourse), the direction of their organizations (see also Strauss, 2006; and G. Wood, 2010). Hirschman and Freeman’s usage gives us our definition of participation. The distinction which we have just implied between degrees of participation (the term we will mostly use in preference to voice) is important. Wilkinson et al.’s (2010) ‘escalator of participation’ is a suitable working model (see Figure One below). We note that it has a managerial flavour in comparison with the six forms of participation which Poole (1992) identified in an industrial relations (IR) context: shop-floor programmes, trade union action, works councils, co-determination, producer co-operatives and workers’ self-management. There is also an important distinction between 'indirect participation', where workers are represented at their own initiative by an independent intermediary (typically a trade union, though other representative structures are possible); and 'direct participation', where management initiates its own communication with employees. When we turn to the forms of participation, we find a considerable yet unrecognized overlap between organizational practice and international development. Both domains host practices loosely grouped under the headings of empowerment (Argyris, 1998; Luttrell et al 2009: 21) and partnership (Ackers and Payne, 1998; Fowler, 2000). These, and also some distinctive forms in the two sectors, are respectively reviewed in Wilkinson et al. (2010) and World Bank (1996). Rationales 4 Efficiency has been the predominant rationale for participation in the organizational literature. It also figures in the international development literature. ‘Staff perform best when they are involved in decisions that affect them,’ declares the Code of Practice produced by People in Aid (2003: 14), an agency with a remit for HR in the INGO sector. Participation is argued to promote efficiency in two ways: first, by giving employers access to crucial knowledge that employees hold: that is a central tenet of so-called ‘high performance work systems’ (Wood and de Menezes, 2008); and second, through its effect on motivation, tapping into 'organizational citizenship behaviour' (OCB). There are reports of participation improving performance in both the organizational and the international development literatures (Finsterbusch and Van Wicklin, 1987; Isham et al., 1995). But the organizational literature has notoriously failed to establish a secure causal relationship, right up to the present (S. Wood, 2010). Even the strong evidence of an association between OCB and organizational performance (Podsakoff et al., 2009) contains an element of tautology: it is hardly surprising that behaviour defined by its impact on organizational performance does indeed improve performance. In consequence, much of the impetus behind participation initiatives has come from normative rather than efficiency concerns. Building on Wilkinson and Fay's (2011) four theoretical paradigms of participation, we identify three normative rationales in addition to the efficiency rationale (Table One).ii INSERT TABLE ONE HERE An ethical view of employees as autonomous Kantian subjectsiii is the first of those normative rationales. It entails ‘a vast democratization of the workplace,’ and puts participation on a par with efficiency as an organizational objective (Bowie, 1999: 102; Budd, 2004). A political view (the second normative rationale) accords workers a democratic right to participate: ‘If democracy is justified in governing the state, then it must also be justified in governing economic enterprises’ (Dahl, 1985: 111). 5 These rationales have had a practical impact in the creation of employee-owned trusts like the UK’s John Lewis Partnership (Bradley and Estrin, 1992), and in the emergence of works councils in Germany. It was works councils which led in the end to the European Information and Consultation of Employees Regulations (2004), which in turn require employers in the EU to set up an elected participation structure when their workers request it (Davies, 2009; Gollan and Wilkinson, 2007; Müller- Jentsch, 2008). Works councils are also an expression of the classic industrial relations rationale: participation designed to benefit workers’ pay and conditions. Reflecting that rationale, IR writers like Freeman and Strauss have explicitly viewed collective bargaining as participation’s only valid form. These ethical and political rationales are even stronger in the international development and related literatures – ‘very few think … participation is a bad thing’ (Birch, 2007: 145) - with the major difference that they mostly apply to beneficiaries, not employees, about whose participation the international development literature has relatively little to say (though see Fowler 1997; and Suzuki 1998). For the Nobel Prize-winning development economist Amartya Sen (1999), participation is nothing less than the process aspect of freedom. Since Sen equates freedom with development itself, participation in this view does not merely enhance development but becomes part of what constitutes it (see also Cortina, 2007: 11). With the ethical rationale augmented by a political one, in which development beneficiaries are conceived as citizens who should be ''makers and shapers' rather than 'users and choosers of interventions … designed by others’ (Cornwall, 2000; Gaventa, 2004: 29), we find participation framed as a human right in the UN's 1986 Declaration of the Right to Development, and in so-called rights-based approaches to development (Hamm, 2001). 6 The converging normative orientation of the HR and international development literatures is reflected in the remarkably similar way they present participation visually (Figure One).iv INSERT FIGURE ONE HERE Gareth Morgan (1986) taught us to take such metaphors seriously. Notice how employee control is the escalator's implied destination, while all the ladder rungs below ‘partnership’ are dismissed as ‘tokenism’ or worse. The top of the escalator/ladder is implicitly the destination we should be trying to reach. Constraints Yet we know that most organizations do not get there. In this and the following section we review some relevant constraints and contingencies which help to explain why. For clarity's sake we have summarized them in Table Two. INSERT TABLE TWO HERE We suggest that two inherent constraints operate in a quite uniform way across organizations. The first is hierarchy. In principle, organizations can exist without a hierarchy, but very few contrive to do so. Hierarchies are integral to Weber’s 'ideal type' bureaucracy. A more modern justification is that hierarchy explains why organizations exist in the first place, as a superior alternative for allocation of resources to a market in which every individual contracts freely with everyone else (this was the economist Ronald Coase's Nobel-prizewinning insight). Organizations which forego a formal hierarchy must find another way of minimizing transaction costs which ensue. They must also find a formal way of supplying the leadership function, failing which they are likely to acquire an informal leadership - an organizational application of Michels' 'iron law of oligarchy'. And informal leadership may be insidious, being 7 outside the procedural checks which unions prize as a bulwark against arbitrary management authority. The 'iron law' applies every bit as much to employee-owned companies as to conventional ones. The UK's celebrated John Lewis Partnership has a chief executive who appoints managers, and also his successor, in the regular way. Managers in Spain's equally celebrated and equally successful Mondragon Co- operative are elected rather than appointed, but they still constitute a recognizable hierarchy and exercise conventional authority over their staff (Bradley and Estrin, 1992; Lutz, 1997; see also Prasnikar, 1996). The second inherent constraint is the need to balance competing stakeholder interests. An organization's stakeholders are conceptualized as individuals or groups who can affect or are affected by what the organization does. They can be categorized in terms of power, legitimacy and urgency (Freeman, 1984; Mitchell et al., 1997). As such, they clearly include owners, managers and workers, and, additionally and significantly, some groups outside the organization about which the HR and IR literatures have had relatively little to say. We can appreciate the stakeholder model’s radical implication for worker participation by relating it to Fox's (1974) well-known 'frames of reference'. Fox's 'pluralist' frame of reference attempted to legitimize unions, and other groups within organizations, in contrast with what Fox called the 'unitary' frame of reference in which management is the sole legitimate guardian of the organization's interests. The stakeholder model effectively accepts Fox's view - employees are clearly stakeholders - but it goes further by requiring employees to make room for stakeholders outside the organization. To be sure, the stakeholder model stipulated that management as well as employees should make room for the external stakeholders. But doing so – and let us concede that management may try to ignore its stakeholders, however short-sighted or unethical that might be - strengthens management's position in one important respect. Consider the external stakeholders' position, outside the organization with no formal 8 purchase upon it. They are not owners or managers, nor a trade union with bargaining rights. Hence in the stakeholder model it falls to management, by default, to represent their interests, and balance them against those of the internal stakeholders, including employees. This aspect of stakeholder management is unfavourable to the employee interest. For as Kerr and Caimando (2004: 86) point out, 'Management retains its authority … by representing the interests of stakeholders in general, not the interests of organization members in particular. From the stakeholder perspective … too much democracy - i.e. too much representation of employee interests - must inevitably come at the expense of the organization's other stakeholders.' In short, while the stakeholder model is 'pluralist' in Fox’s terms in legitimizing the employee interest, it also legitimizes new external interests which it becomes management's job to interpret and represent. It dilutes management's authority in one way, but strengthens it in another: management can use external stakeholders' interests as an argument (or an excuse) for rejecting employees' preferences. Contingencies Case studies like ours, being rich in detail, provide an insight into the contingencies which mark participation in different workplaces. At the international level, indirect forms of participation – ‘extensive participatory rights ... linked closely to strong overarching unions’ – are more prevalent in 'co-ordinated market economies' like Germany and Sweden than in 'liberal market economies' like Australia and the US (Brewster et al., 2007; Thelen, 2001: 102). The special case of multinational corporations (MNCs) is also relevant to us. Müller- Carmen et al. (2001) find a 'home country effect' when an MNC operates in a country with weaker economic institutions and the home country employment norms prevail. A 'host country effect' occurs when things are the other way round. 9 At the national level, developing countries like Kenya where our case study is situated have a dualistic economic structure, with large informal and subsistence agriculture sectors and a small formal economy, and the latter further segmented between multinational corporations and indigenous firms which are often family- based (Siddique, 1989; G. Wood, 2010). In Kenya’s formal economy, the government's stance has been to 'choreograph' the union movement to facilitate employers (Kamoche et al., 2004: 95). With the economy growing at less than 5% per annum over the last decade even after two previous decades of stagnation, while population growth has been at roughly 2.6%, and with a consequent unemployment rate estimated at over 30% in April 2011, workers' bargaining power is weak. Union membership was at 4.3% in 1999, and employers have been under little pressure from their workers to concede the right to participate; for Nyambegera et al. (2001: 133), its existence is merely theoretical. This may be reinforced in African countries like Kenya by high ‘power distance’, that is, a cultural tendency to defer to authority, and an expectation that leaders will adopt a paternal management style (Blunt and Jones, 1997; Hofstede, 2010; Jackson, 2002). Employees in Kenya’s agricultural sector who experienced little opportunity to participate felt little frustration because they never expected to participate in the first place (Mulinge and Mueller, 1998); in marked contrast with their counterparts in 12 UK organizations, who were demotivated by lack of influence over company decisions (Purcell et al., 2003). At the organizational level, participation may be squeezed out by financial pressure on the organization (Cunningham and Hyman, 1999; Humphreys and Hoque, 2007). Argyris (1998) and Arogyaswamy et al. (1995) have argued that management should drive through any necessary retrenchment before they embark on strategizing and development activities such as participation and ‘empowerment.’ Participation is also affected by management style. Hirschman himself (1976) observed that exercising voice may result in either reprisals or rewards: punishment of whistle-blowers and rewards for organizational citizenship behaviour are examples 10 (Near and Miceli, 1986; Podsakoff et al., 2009; Van Dyne, 1998). Since workers are more likely to speak up when managers show they are listening, it follows that employee voice is a delicate plant that managers need to take some trouble to nurture (Detert and Burris, 2007; Morrison and Milliken, 2003). Methodology We will now report a case study of INGOs operating in Kenya. They are anonymized as Agencies A to G. As Strauss (2006: 796) notes, case studies have the potential to get into the 'little black box, the intervening variables between participation and its outcomes.’ However, the case method requires care in generalizing from the case; and thus in answering our study question 3 (Eisenhardt, 1989). Data collection and analysis Data was collected in the first half of 2011. As well as reviewing agency documents, we recorded and transcribed 39 semi-structured individual interviews: 28 with senior managers at the case agencies, including country directors and assistant directors and HR managers; five with operational officers and elected staff representatives; and six with staff of other organizations which interact with the case agencies. We have borrowed the process tracing method from political science, where it is used to understand political events in the round when they are marked by multiple interacting variables (George and Bennett, 2005). As far as we know, process tracing has been used in organization studies up to now only to assess how individuals make decisions, not organizations (Ford et al., 1989). We use it to throw light on agencies’ actual, as opposed to formal, priorities by asking interviewees to recall recent important management decisions, the process by which they were made and the actors involved in them. Although we agree with Strauss (2006) that adding survey data strengthens a case research design, we were unable to administer the employee survey instrument which 11 we developed, due to pressure of work in the INGOs at the time of our research (caused partly by a tragic influx of refugees from neighbouring Somalia). We draw instead on existing employee surveys furnished by three of the agencies, and on interviews with line staff in three agencies, elected staff representatives in two agencies, and with an officer in an external body which promotes HR practice in INGOs in Kenya. We also draw on existing beneficiary survey data for two agencies. Employee participation: findings The funding chain and the internal hierarchy In this section we begin with a feature which in different degrees is common ground for all the agencies. As is increasingly the case in this sector (Wallace, 2006), Kenya's INGOs typicallyv do business through a lengthy chain. INSERT FIGURE TWO HERE They receive a great deal of their money from official donor agencies, as we noted at the start of the article. Their institutional imperative to maximize funding has been argued to militate against beneficiary participation (Anderson, 2001; Fowler and Biekart, 1996). Moreover, in keeping with current development thinking, six of the INGOs used local 'partners' (indigenous NGOs which may have a better understanding than INGOs of complex local relations) to implement some or all of their programmes: (Marcussen 1996; see also Kanyinga, 1995 for Kenya). In addition, INGOs’ hierarchies constitute an internal chain, so to speak. They are governed by Boards of trustees, a standard governance structure which is reflected in Kenya’s official Non-Governmental Organizations Co-ordination Regulations (1992). In C, staff 'can’t take on grants without … clearance because at the end of the day it’s the international trustees that are responsible.'vi Trustees are usually co- opted, although they are elected by the members in one INGO which is a membership organization. 12 Summary data All the INGOs had a conventional hierarchical structure at the time of our study (similar to John Lewis and Mondragon in this respect), their activities informed by strategic plans and procedures for monitoring and evaluating them which in the international development sector are relatively elaborate. Likewise, all of them had a grievance procedure and a designated HR manager (all were Kenyan nationals, making them the most senior Kenyan staff in at least three of the agencies), and they held regular staff meetings. Table Three summarizes other relevant features. INSERT TABLE THREE HERE None of the agencies was unionized in Kenya, although four of them recognized a trade union in their home countries. Agencies B and G had formal elected structures. Apart from conventional staff meetings, only employee surveys among the popular forms of direct participation in industrialized countries had taken root. Employee participation practice Due to pressure of space, we now provide a narrative description of the style of participation in only four of our case agencies. We will draw on data from the remaining three agencies in later sections. Strategic planning in Agency A crystallized in its Five-Year Plan, on which staff were formally consulted. Yet one agency manager dismissed it in colourful language, adding that 'staff aren't going to be able to tell you … what is in (it).' I reality, much of A’s strategic energy was channeled into piecemeal ‘initiatives’ which cascaded from HQ: 'all (HQ) advisers have initiatives. At one point we came to 14 initiatives, (only) four of them were in the country strategic plan.' And it was HQ advisers who initiated one of the biggest recent management decisions, the shift from a congeries of discrete projects at national level to a coherent programme based on an analysis of national development priorities, a ‘programme mode’ of operation which has its origin in current international development theory and practice. One 13 manager who was personally committed to this initiative still cannily observed that 'it makes sense to piggyback on a process that is being mandatorily implemented.' However, HQ and its 'academic' advisors – they tended to confront country staff with the perceived deficiency of their conceptual frameworks – were just one of the interests which managers had to balance. Once again there were the donors who provided much of A’s funding (‘we need to watch the money … (and) some of our donors also have ‘academics’’). On the other hand, staff appeared not to be a significant interest group. They were not consulted on the shift to programme mode. And when one of A’s regional teams in a majority Muslim area, a significant sub-set of the employee stakeholder interest, ‘completely rejected’ work with women and girls, a new programme priority, a manager bluntly remarked: ‘They’re going to lose … Frankly, HQ will come in at some point (and say) … we don’t agree … We respect your position, so maybe the time has come … when your contract finishes and we have to change staff. That’s the power relations … Some people will never make the shift and they’ll be left by the roadside.’ When one bears in mind that most of A's staff were on short-term contracts, field managers’ insistence on talking about pay and job security when a senior manager tried to engage with them on the shift to programme mode is perhaps understandable. Managers also had discretion over how they involved beneficiaries, the other main stakeholder. ‘You are formally required to say that you have met (beneficiaries). Often this is demonstrated through little human interest stories, or community member X said blah blah blah. Staff from … head office … won’t come and check, there is no annexing, no naming of names … If I say it’s happening, generally they won’t question it.’ 14 In fact the Agency’s HQ generally ‘allows a lot of local decision making.’ However, according to two managers this was largely confined to ‘a few decision-makers at the top;’ A was ‘pseudo-participatory,’ said one of the two. A third manager, making an oblique comment on one of his senior colleagues, said that ‘The … senior leadership really does make a big difference in how open leadership is to listening to people … The culture in Kenya is that you don’t raise your voice to staff. If I do it in front of other people, than I’m really causing a problem. … You begin to get a culture of … people not wanting to communicate.’ Thus when the Agency’s drivers were unhappy about the sudden withdrawal of a benefit, their only recourse was to their immediate supervisors and the HR manager. With informal communication weak, and with staff meetings the only formal mechanism available for staff to express their views, it is perhaps not surprising that only 41% of respondents in the Agency’s own employee survey felt that they could influence decisions that affect them. Agency C, having decided that it should ‘empower people in the South,’ had moved its global HQ from Europe to Nairobi some years before our study. It had resolved at the same time to ‘move away from the command principle to a much more participatory and information-coming-from-the-bottom (approach).’ So when country-based staff pointed out the ironic fact that his new ‘empowerment’ approach had been decided over their heads, A made reparations by giving the job of fleshing out the new organization structure to its eight African country managers. Alas, little by little the job got bogged down, with restructuring becoming an end in itself. ‘(In a) funding environment (that) was becoming increasingly results-based … (the) donors ran riot. They basically said: what is going on? This is total 15  madness! We have given you our money to give us results and you are telling us about empowering people. We have had enough of this.’ Intensely aware that ‘(donor) governments want short-term results that they can show to their electorates,’ C’s Board responded by dismissed first its director, and then all its country managers, and reverted in its new African base to more conventional strategic planning and management. It introduced performance management, expecting its new country managers, in classic unitarist style, to 'operate within common policies' and ensure 'minimum standards' of work. However, since C retained partnership working as a 'fundamental value,' it invited beneficiaries and also donors to what was intended to be C's first annual review and planning workshop in late 2010, just before our data-gathering. The agency's own staff were in a minority at the workshop, so to the extent that the workshop informed agency policy (it concluded with 'propositions' rather than recommendations), the effect of involving other stakeholders was to dilute the staff input. Readers may note that this was very much in line with Kerr and Caimando’s argument. The workshop was C’s only formal mechanism for staff input into policy. For although C had a staff association, it operated as a welfare body. Staff as members had to contribute, and their dues were used for modest purposes such as buying baskets of fruit for staff who fell ill. With communication between staff and management that existed only ‘to some extent, to be honest, to some extent: people are human beings’ - as in A, HR became a default conduit for staff concerns, and also sometimes for management to inform staff: a 'man (sic) in the middle' HR role which older readers may recognize (Thomason, 1976). Agency D was little different from A in its formal structures, which in the eyes of two of its managers were weaker than at other aid agencies where they had worked. There was no staff association, and a union only at headquarters. However, tracing the process used to deal with what was by common consent the biggest recent management issue, the sharp decline in D’s funding which followed 16 the global financial crisis of 2007/8 and which resulted in some compulsory redundancies, disclosed vigorous informal participation, albeit within strict limits. Retrenchment was initiated by the Agency’s HQ, which indicated the savings that would be needed. In relation to local discretion, ‘the red line … is in budgetary control more than anything else’ (the ‘financial pressure’ contingency which we noted earlier). But contradicting the view in the literature that retrenchment should be driven through over employees’ heads, management allowed a staff representative group to produce its own retrenchment priorities. Results were mixed. On one hand, HQ overrode a request from the Kenya country office to protect the staff benefits package in order to avoid a disproportionate impact on African staff, who tend to have larger families; on the other, the Chief Executive backed down on a post that he personally wanted to keep. Management rejected the request for an increase in annual leave to offset the impending pay cut (‘we didn’t think the Governing Body would buy it, and we didn’t think the public would buy it’); but it did concede a half-day increase for one year only. (Notice how the general public is a stakeholder for D, since it relied on them to drop money into the collection tins which they rattled on street corners.) At country level, D's Kenya country director asked the HR manager to consult staff on a proposal to reduce costs by abolishing a ‘thirteenth month’ bonus (the ‘man in the middle’ role again). The HR manager duly fed back staff’s objections in some detail (for instance, the impact on staff’s ability to pay their children’s school fees). But the country director went ahead anyway, arguing that the bonus was an anachronism, as it had been introduced to compensate for the absence of a pension scheme, something which D now had. In addition, staff had two representatives of their own on the ad hoc country strategy steering committee. The country director in turn sat on the global strategy steering committee – appointed, not elected, but seeing it as ‘my job … to get the views of the other country directors, to keep them informed as to what was happening.’ 17 This mixture of direct and indirect participation (indirect insofar as staff had some autonomy to choose representatives, albeit on ad hoc groups) created a style of participation which had elements of ‘codetermination’ in terms of Wilkinson’s escalator, but with management still keeping the last word. That style applied equally to D’s relationship with its local NGO partners: ‘We are reasonably clear that our job is to get programmes done. The means to get it done is the partners … We debated this … Is our role to build capacity of partners, to enable them … or is it to make vulnerable people less vulnerable? … ‘We are in an organization that is about vulnerability’ … is the argument that won in the end and that is what we do.’ Managers appreciated the Agency’s informal approach – ‘I feel structures are there to be smashed or worked around!’ - but conceded that its viability depended on the personality of the country director and ‘how they open up the playing field to everybody.’ Nevertheless, the informal approach was supported by both staff and partners in externally conducted surveys. 76% of employees felt that they could influence decisions (as against 41% in A), and the Agency had the fifth-highest satisfaction rating from its partners out of 25 INGOs. Our interviewees justified D’s distinctive participation style in several ways. Programme staff as well as management needed ‘enough room to operate ... if there are good ideas they will be given the capacity to go and work on (them).’ A programme officer corroborated this: ‘management … always act on my recommendation.’ Despite Kenya’s ‘colonial hangover … (local) staff have been empowered to demand to participate,’ as a Kenyan manager remarked. On the other hand, ‘If you get stuck in participation, people see it as fumbling. Senior management is leadership ... There’s a transaction cost in participation ... paralysis by analysis.’ Most fundamentally, senior staff - ‘very passionate people’ – justified it in terms of the role they saw for themselves as the guardians of the Agency’s mission to help vulnerable people, and the final arbiters of their different stakeholders’ views: ‘We listen to staff, but we have to triangulate. We're there to get people out of extreme poverty.’ 18 Although we have characterized Agency F as a ‘public service contractor’, it also has a major relief and welfare role in Kenya. Our Agency interviewees identified three major issues: a management restructuring, and the two linked problems of high turnover and absenteeism. The restructuring was initiated by the country director as a response to the growth of the country programme in 2009/10. The country director had taken it on himself to adopt G’s generic restructuring blueprint for ‘operational fragile states’ following discussion with HQ: ‘It was pretty unilateral.’ Having done that, however, the country director convened an ad hoc meeting of ‘every staff member, all the cleaners and everyone. … (The CD) explained … the model … (and) … more or less said, ‘This is what we are doing, any comments, got a problem with it?’ And everyone was like, no.’ So the staff stakeholder interest was represented to some extent. However, the Agency’s partners’ role in the restructuring was ‘probably zero,’ and as for beneficiaries, ‘No, no way in the world.’ At the opposite end of the hierarchy, even the HQ input was on the country office’s terms: ‘Deciding how to structure the area offices … we ignored (HQ), we totally pushed back with them.’ As well as ignoring HQ - and turning now to the absenteeism issue - the senior management team, ‘reasonably strong managers at this level,’ did not hesitate to reduce field staff’s entitlement to 'Rest and Recreation' (R&R) unilaterally from every eight to every ten weeks, even though ‘There was murder about that in the field: we were not consulted, this just came out of nowhere, dah dah dah dah dah. And you just think: hmmm, did we make a good decision? Was that well thought through? Perhaps it wasn’t: one manager said that on reflection, the absenteeism problem was caused not by the frequency of R&R entitlement but by local managers’ failure to prevent its abuse. Thus reducing its frequency would punish compliant staff for 19 abuses by their deviant colleagues. Moreover, with R&R after eight weeks the standard entitlement among INGOs working in similar relief environments, there was the danger that staff would express their dissatisfaction ‘in terms of resignations,’ exacerbating turnover, G’s other perceived problem. In doing so, they would inadvertently have provided a graphic illustration of Freeman’s ‘exit-voice trade-off’. With no ‘voice’ input to the R&R decision, employees only had ‘exit’ as a way of showing their displeasure. F had, indeed, no formal employee participation channel outside its line management structure and staff meetings. Thus just as with her counterpart in A, a Nairobi-based manager who tried to engage field staff on global strategy found the meeting swamped by staff venting more immediate concerns. Granted, there was evidence that the informal channels were open. For example, the senior management team acted on representations from both staff and beneficiaries by dismissing one area manager who had had a previous warning, and felt vindicated when they received positive feedback about the manager’s replacement. However, as in D, placing the emphasis on informal communication also placed a weight on the shoulders of management (including the HR ‘man in the middle’, through whom staff channeled their complaints, for instance about the outcome of a job evaluation exercise). One interviewee admired a previous country director’s ‘amazing communication skills.’ But that country director had moved on; and ‘we are constantly getting the feeling that staff don’t have a clue what is going on because the area managers are not (communicating).’ Staff councils and representatives We now draw on data from two of the remaining three INGOs in our sample, since they were the only agencies which had formal representation structures, as noted in Table 3. The structures lie somewhere between direct and indirect representation, in that they were management initiatives in both cases, but employee participation is somewhat autonomous, since employee representatives were elected - or at least, were supposed to be. 20 B had a very well established structure of global, regional and national councils, although its status had been downgraded from executive to advisory in 2007, when the Agency severed the institutional link which it had had up to then with its national government. B’s council successfully persuaded management to introduce a Cost of Living Allowance after a five-year war of attrition, although its attempt to introduce separation pay for employees who had had a series of short-term contracts ‘failed completely because … we didn’t have a solid legislative background … Where the laws are weak, then the bargaining power is weak of the country council.’ Despite the firm formal structure, an operational officer who had worked for B in two other countries said that the council ‘depends on … individual leadership style’: more participatory or, in the case of ‘a top-down (country director who came) from the private sector,’ less; and also on ‘culture difference’: when he asked a senior manager why the management style in Kenya was less participatory, the manager replied in an unconscious corroboration of Hofstede's 'power distance' findings, ‘This is Kenya.’ Still, the staff representative whom we interviewed said that ‘the consultative open process encourages me to stay with the organization … If you look at the people that originally started working with (B), we are looking at those retiring from (it).’ G’s HR manual mandated two elected staff representatives for every country. However, we were told that across the globe, it was more honoured in the breach than the observance, and making it happen in Kenya was the country director’s personal initiative. One of the representatives reported increasing the payment to drivers for mobile phone airtime and getting management to contribute to a staff outing as achievements, both the result of lobbying the country director. Although these were modest achievements, the representative was still positive about the initiative, and contrasted G favourably with another INGO where he had previously worked: 'Here is good, management listens.’ Discussion: explaining participation styles 21 We now step back from our findings to discuss their significance for our understanding of participation. The styles of participation (study question 1) Our research methods cannot reveal significant statistical correlations, both because we have used a qualitative design and because in quantitative language, ours is an N=7 study. However, there is some suggestive patterning in our findings, which we depict in terms of Wilkinson’s ‘escalator’ in Figure 3. The three ‘before and after’ plots reflect B’s global downgrading of its country councils from executive to advisory in 2007; C’s reversion to conventional top-down strategic planning, albeit with some consultation, after dismissing its country managers; and G's recent activation of a staff representative mechanism. INSERT FIGURE THREE HERE What Figure Three shows is firstly, that all the agencies fell short of the participation literature’s nirvana of employee or citizen/beneficiary control, with some convergence on 'consultation' (B from higher up the 'escalator', G from lower down). B and C had had forms of codetermination, as we saw. But both had retreated. Let us recall that our agencies provided a very favourable environment for participation. They were not run for profit - no shareholders clamouring for a dividend - and they were shot through with a development rhetoric which venerates participation. Despite that, none of the agencies had found it appropriate to settle on either the ‘control’ or ‘codetermination’ participation style. They came to rest instead on the lower steps of the ‘escalator’, with consultation as the mode position. Why was that? Explaining the styles (study question 2) 22 We suggest that the two constraints we identified earlier, hierarchy and stakeholders, explain why the agencies ascended no higher than the consultation step/rung on the escalator/ladder. C tested the theory of employee control almost to destruction. Putting the country managers in charge of the pivotal restructuring led to an inward- looking emphasis on process at the expense of results that C’s donors felt they had to squash. In reaction, C used performance management to reassert conventional hierarchical control. In doing so, C was merely falling in line with what the other INGOs had always done. The consistent pattern was that while staff might propose (cf. the ‘propositions’ from C’s review and planning workshop), management would dispose. This was in part because management saw itself as the guardian of the organization's mission, a unitarist view that Fox would have recognized. We saw how the manager at D put management’s interpretation of the needs of 'the vulnerable' above the views of D's partners, even though in international development doctrine, partners should have primacy because they are closer to beneficiaries than INGOs can ever be. The INGOs’ conventional management control was also reflected in the influence of their stakeholders. With ‘the red line … in budgetary control,' the donor influence was transmitted along the funding chain, as interviewees in two of the agencies pointed out. ‘What (beneficiaries) expect from us, we may not be able to achieve because we might not get something from the donors.’ ‘Over time, the country office will grow and contract depending on the donor environment.’ Donors, Boards of Trustees and HQ at A (‘because (HQ) has the resources’) were stakeholders who had ‘power’ in terms of Mitchell et al.’s stakeholder model, constraining management and employees alike. However, their power was not uniform. Unlike A, G’s Kenya office (and also F’s: see above) was able to make HQ dance to its tune: ‘Headquarters … did not want a country programme in Kenya. It was the last Country Director who … said: "The needs are there … I am going to go and expand and I am going to find the money for it." For three years (it) was not 23  recognized by our Headquarters, they never visited … It is like, once you push the idea forward, they say: you guys raise the money, we are not going to have anything to do with it. They leave it to you.’ Across the agencies, it was up to management to guard the interests of stakeholders who had ‘legitimacy’ rather than ‘power,’ just as Kerr and Caimando argued: in particular, beneficiaries, partners and the Government of Kenya. Agency F invited a government ministry to take part in project planning, monitoring and evaluation, and C invited stakeholders to its review and planning workshop. All this explains why participation went no higher than the consultation step/rung: ‘decision-forming, not decision-making’, in the words of a manager at D, the same manager who went on to say, ‘We listen to staff, but we have to triangulate.’ Cornwall (2000: 35) observes of this 'triangulation' of different stakeholder interests that 'it remains entirely unclear how the participation of this rather unwieldy collection of (stakeholders) might actually be managed.' But an interviewee described in a matter-of-fact and winningly humble way how E went about it when it produced its Five-Year Plan. E began by reviewing Kenya government policies, its own global strategy and its performance in the previous Plan period. Then it sounded out donors and its own staff in meetings and workshops. Finally, ‘(The Senior Management Team) did a synthesis … the four of us analyzed, and when we needed to ‘wordsmith’ it, they all … said: who is the only native English speaker here? It was me!’ However, hierarchy and stakeholder constraints do not explain why three of the agencies stopped short even of the consultation step/rung. Why was that? We suggest three additional reasons. 1. The relationship between the style of participation and the CME/LME dimension (Table Three). Both agencies with headquarters in a CME had an indirect participation structure; none of the four agencies based in an LME 24  did. This 'home country effect' is consistent with Thelen's picture of relatively strong indirect participation in the CMEs. 2. A 'host country effect' in relation to trade unions. In Kenya, none of the agencies recognized a union. Clearly, the four agencies which recognized a union at home were not hostile to unions in principle, and Kenya‘s Labour Relations Act (2007) requires employers to recognize a union where a majority of employees request it. But in a high-power distance country where only 4.3% of workers are unionized, employees appeared not to see the need, with one of them commenting that ‘what INGOs offer is vastly superior to what ordinary trade union members (in other organizations) have.vii 3. Management style, whose influence was most marked when formal procedures such as strategic planning and staff councils were loose enough to give managers room to manoeuvre. That was the case in A, whose strategic plan was derided by managers at country level, who operated from the top down, and in D, whose country director chose to manage more consultatively, using ad hoc structures. These different discretionary styles had the motivational effects that the literature suggests: chilling at A, where ‘you begin to get a culture of … people not wanting to communicate,’ and nurturing at D: ‘people have been empowered to demand … to have a say … Actually most people do actively give feedback.’ In terms of access to employee knowledge, the other leg of participation’s efficiency rationale, we saw how F’s arbitrary decision on R&R misread the situation and risked exacerbating the problem of high turnover which already existed. Conclusion: inherent constraints and creative possibilities Our case design and process tracing methodology have enabled us to view participation in the round. In an almost uniquely propitious setting, none of the seven agencies rose any higher on the escalator/ladder than 'consultation'. This was dictated by the INGOs’ conventional hierarchies and the need to 'triangulate' the views of their stakeholders (namely their Boards, management and staff; donors; partners; 25 beneficiaries; and the Government of Kenya). To put it in the homely terms of the manager at E, participation processes, however elaborate, inevitably culminated in a 'native English speaker' or some other delegate wielding the strategic pen on behalf of the management team. Consultation can indeed degenerate into Arnstein's tokenism, as the 'little human interest stories' at A demonstrate. The motivation and knowledge advantages of participation, and its positive relationship with low turnover, which Freeman identified and which F's experience corroborates, show why tokenism is worth avoiding. Our findings show that indirect representation and management style can be antidotes. Their limitations are well known: management can subvert representative structures if employees are not vigilant or determined, and management style is liable to change when a top manager moves on (as we saw at F). Nor are they in free variation with agencies' other characteristics. We have seen that CMEs provide a more favourable base for indirect representation, and Brewster et al. have indicated that that is likely to remain the case for some time to come. With indirect representation, the onus is on employees to demand it, including union recognition if they wish (Kenyan employees did not so wish, though other indirect participation structures were a partial substitute in two agencies), and on managers to recognize the advantages of a staff council or similar structure, as G's country director did. With management style, managers may opt for participation because it offers possible efficiency gains (the evidence is inconclusive, as we saw); a collective channel for staff concerns so that managers are not swamped by those concerns when they try to engage staff on programme issues; and the ethical advantage of practising what they preach about 'empowerment' etc., and harmonizing the way they relate to their stakeholders, and to staff and beneficiaries most of all. The latter remarks apply specifically to INGOs, for whom participation is an almost universally espoused value. What have we learnt about employee participation in general (our third and final study question)? Our findings suggest that full co- determination and employee control, the implied common destination of Wilkinson’s, Arnstein’s and Poole’s participation models, are unrealistic. If INGOs have not 26 achieved them in Kenya, it seems utopian to expect to find them elsewhere. More to the point, they are not even desirable. Our findings suggest, contrary to the implications of both the HR and international development normative literatures, that the top of the escalator/ladder is not the place to aim for.viii However, our findings equally suggest that consultation, at the mid-point in all three models is both realistic and desirable (works councils in Poole’s model roughly corresponds to consultation in our case agencies’ practice). The consequent transaction cost - which we should not overlook - is modest as long as organizations avoid the trap of getting 'stuck in participation' or 'lost in processes' which a manager at D warned against, but which C fell into. D's experience suggests that consultation can be appropriate even when the policy under discussion is the need to make cuts, despite the opposing view of the turnaround literature. The remarkable confluence of the HR and international development participation discourses perhaps represents an incoming tide of participation in modern society as a whole. Ethical and political as well as efficiency rationales combine to dictate that organizations must pay attention to their staff and their other stakeholders. We conclude that consultation is a middle way towards which organizations can realistically aspire. The benefits of participation and engagement which Lord Mandelson asserted at the start of this article are potentially available to organizations across the whole spectrum of organizations; they are not confined to INGOs operating in Kenya. Among those INGOs, we venture to hold up agencies B and D as models of formal and informal participation. D is particularly noteworthy because it shows that organizations can creatively buck the LME trend and offer meaningful participation to their employees. Although we were unable to survey employees directly, there is positive evidence from our interviews and D's own employee survey that staff in both INGOs responded positively to the consultative approach. 27 It is striking that B and D's practice on the African continent is closely compatible with the European Information and Consultation Regulations. Those regulations are the culmination of a historical development whose roots are in Germany’s system of works councils and are not easily transferrable. Yet our findings suggest that organizations in other economic sectors and legislators in other continents can realistically emulate them. It was the creative initiative of D's managers that made participation meaningful. But Ramsay (1977) may be right to conclude that much more often, employers allow their staff to participate only when their staff make them. In countries where workers are more assertive than in Kenya with its high power distance, employees might nudge their managers to take the path of virtue and go one or two steps higher on the participation escalator. 28 REFERENCES Ackers, P. and Payne, J. (1998) ‘British trade unions and social partnership: Rhetoric, reality and strategy’, International Journal of Human Resource Management 9: 529-50. Anderson, M. (2001) ‘Aid: a mixed blessing’ in Eade, D. and Ligteringen, E. (eds), Debating Development: NGOs and the Future, Oxford: Stylus Publishing, 292-99. Argyris, C. (1998) 'Empowerment: The emperor's new clothes', Harvard Business Review, May-June: 98-105. Arnstein, S. (1969) 'A ladder of citizen participation', Journal of the American Institute of Planners, 35, 4: 216-24. Arogyaswamy, K., Barker, V. and Yasai-Ardekani, M. (1995) ‘Firm turnarounds: An integrative two-stage model’, Journal of Management Studies, 32: 493-525. Birch, A. (2007) The concepts and theories of modern democracy, Abingdon: Routledge. Blunt, P. and Jones, M. (1997) 'Exploring the limits of Western leadership theory in East Asia and Africa', Personnel Review, 26, 1/2: 6-23. Bowie, N. (1999) Business ethics: A Kantian perspective, Malden, Mass: Blackwell. Bradley, K. and S. Estrin (1992) ‘Profit sharing in the British retail trade sector: The relative performance of the John Lewis Partnership’, Journal of Industrial Economics, 40, 3: 291-304. Bratton, M. (1989) ‘The Politics of Government-NGO Relations in Africa’, World Development, 17: 569-87. 29 Brewster, C., Croucher, R., Wood, G. and Brookes, M. (2007) ‘Collective and individual voice: Convergence in Europe’, International Journal of Human Resource Management, 18: 1246-62. Budd, J. (2004) Employment with a human face: Balancing effiency, equity, and voice, Ithaca, NY: Cornell University Press. Clark, J. (1991) Democratizing Development: The Role of Voluntary Organizations, West Hartford, Conn: Kumarian. Cortina, A. 2007. Development Ethics: A Road to Peace. Kellogg Institute for International Studies Working Paper No. 339. http://kellogg.nd.edu/publications/workingpapers/WPS/339.pdf. Cornwall, A. (2000) Beneficiary, consumer, citizen: Perspectives on participation for poverty reduction, Stockholm: Sida. Cox, A., Zagelmeyer, S. and Marchington, M. (2006) ‘Embedding employee involvement and participation at work’, Human Resource Management Journal, 16, 3: 250–67. Cunningham, I. and Hyman, J. (1999) 'The poverty of empowerment? A critical case study', Personnel Review, 28, 3: 192-207. Dahl, R. (1985) A preface to economic democracy, Cambridge: Polity. Dahl, R. (1961) Who governs? Democracy and power in an American city, New Haven: Yale University Press. Davies, W. (2009) Reinventing the firm, London: Demos. Detert, J. and Burris, E. (2007) 'Leadership behavior and employee voice: Is the door really open?', Academy of Management Journal, 50: 869-84. 30 Detert, J. and Edmondson, A. (2011) ‘Implicit voice theories: Taken-for-granted rules of self-censorship at work’, Academy of Management Journal, 54: 461-88. Dundon, T., Wilkinson, A., Marchington, M. and Ackers, P. (2004) 'The meanings and purpose of employee voice', International Journal of Human Resource Management, 15: 1149-70. Edwards, M. and Hulme, D. (eds) (1996), Non-Governmental Organizations: Performance and accountability – Beyond the magic bullet, London: Save the Children. Eisenhardt, M. (1989) ‘Building theories from case study research’, Academy of Management Review, 14: 532-50. Farndale, E., Van Ruiten, J., Kelliher, C. and Hope-Hailey, V. (2011) 'The influence of perceived employee voice on organizational commitment: An exchange perspective', Human Resource Management, 50, 1: 113-29. Finn, W. (2011) 'The winner: Baringa Partners', in Great Place to Work Best workplaces 2011, London: Redactive, 8-9. Finsterbusch, K. and Van Wicklin III, W. (1987) 'Contribution of beneficiary participation to development project effectiveness', Public Administration and Development, 7: l-23. Ford, J., Schmitt, N., Schechtman, S., Hults, B. and Doherty, M. (1989) 'Process tracing methods: Contributions, problems, and neglected research questions', Organizational Behavior and Human Decision Processes, 43, 1: 75-117. Fowler, A. (2000) ‘Beyond Partnership: Getting real about NGO Relationships in the aid system’ IDS Bulletin, 31, 3, 1-11. 31 Fowler, A. (1997) Striking a balance: A guide to enhancing the effectiveness of non- governmental organizations in international development, London: Earthscan. Fowler, A. and K. Biekart (1996) ‘Do Private Agencies Really Make a Difference?’, in Sogge, D., Biekart, K. and Saxby, J. (eds), Compassion and calculation – The business of private foreign aid, London: Pluto. Fox, A. (1974) Beyond contract: Work, power and trust relations, London: Faber. Freeman, R. (1980) ‘The exit-voice tradeoff in the labor market: Unionism, job tenure, quits, and separations', Quarterly Journal of Economics, 94: 643-73. Freeman, R.E. (1984) Strategic management: A stakeholder approach, Boston: Pitman. Frontera Group (2007) Motivating staff and volunteers working in NGOs in the South, www.peopleinaid.org/pool/files/pubs/motivating-staff-and-volunteers- working-in-ngos-in-the-south.pdf. Gaventa, J. (2004) 'Towards participatory governance: Assessing the transformative possibilities', in Hickey, S. and Mohan, G. (eds) Participation: From tyranny to transformation?, London: Zed, 25-41. George, A. and Bennett, A. (2005) Case studies and theory development in the social sciences, Cambridge Mass: MIT Press. Global Humanitarian Assistance (2011) GHA report, Wells, Somerset: Global Humanitarian Assistance. Gollan, P. and Wilkinson, A. (2007) ‘Implication of the EU Information and Consultation Directive and the Regulations in the UK – prospects for the future of employee representation’, International Journal of Human Resource Management, 18: 1145-58. 32 Hall, P. and Soskice, D. (2001) ‘Introduction’, in Hall and Soskice (eds) Varieties of capitalism: The institutional foundations of competitive advantage, Oxford: Oxford University Press, 1-68. Hamm, B. (2001) 'A human rights approach to development', Human Rights Quarterly, 23: 1005–31. Hirschman, A. (1976) 'Some uses of the exit-voice approach - Discussion', American Economic Review, 66: 386-91. Hirschman, A. (1970) Exit, voice and loyalty, Cambridge MA: Harvard University Press. Humphreys, M. and K. Hoque (2007) 'Have the lecturers lost their voice? Involvement and participation in the devolved Further Education sector', International Journal of Human Resource Management, 18: 1199-1213. Hyden, G. (1983) No shortcuts to progress: African development management in perspective, London: Heinemann. Isham, J., D. Narayan and L. Pritchett (1995) 'Does participation improve performance? Establishing causality with subjective data', World Bank Economic Review, 9: 175-200. Jackson, T. (2002) ‘Reframing human resource management in Africa: A cross- cultural perspective’, International Journal of Human Resource Management, 13: 998-1018. Kamoche, K., Nyambegera, S. and Mulinge, M. (2004) ‘HRM in Kenya’ in Kamoche, K. et al. (eds) Managing human resources in Africa, London: Routledge, 87-102. 33 Kanyinga, K. (1995) ‘The politics of development space in Kenya’ in Semboja, J. and Therkildsen, O. (eds) Service provision under stress in East Africa, Copenhagen: Centre for Development Research. Kanyinga, K., Mitullah, W. and Njagi, S. (2007) The non-profit sector in Kenya: Size, scope and financing, Nairobi: University of Nairobi Institute for Development Studies. Kerr, J. and Caimano, V. (2004) ‘The limits of organizational democracy’, Academy of Management Executive, 18, 3: 81-97. Luttrell, C., Quiroz, S. et al. (2009) ‘Understanding and operationalizing empowerment’, ODI Working Paper No. 308, www.odi.org.uk/resources/download/4525.pdf. Lutz, M. (1997) 'The Mondragon co-operative complex: An application of Kantian ethics to social economics', International Journal of Social Economics, 24: 1404- 21. MacLeod, D. and N. Clarke (2009) Engaging for success: Enhancing performance through employee engagement - A report to Government, London: HMSO. Marchington, M., Wilkinson, A., Ackers, P. and Goodman, J. (1993) ‘The influence of managerial relations on waves of employee involvement’, British Journal of Industrial Relations, 31: 553-76. Marcussen, H. (1996) ‘NGOs, the state and civil society’ Review of African Political Economy, 69: 405-23. Mitchell, R., Agle, B. and Wood, D. (1997) ‘Toward a theory of stakeholder identification and salience: Defining the principle of who and what really counts’, Academy of Management Review, 22: 853-86. 34 Morgan, G. (1986) Images of organization, London: Sage. Morrison, E. and Milliken, F. (2003) 'Speaking up and remaining silent: The dynamics of voice and silence in organizations', Journal of Management Studies, 40: 1353-8. Mulinge, M. and Mueller, C. (1998) ‘Employee Satisfaction in Developing Countries: The Case of Kenya’, World Development, 26: 2181-99. Müller-Carmen, M., Almond, P., Gunnigle, P., Quintanilla, J. and Tempel, A. (2001) 'Between home and host country: Multinationals and employment relations in Europe', Industrial Relations Journal, 32: 435-48. Müller-Jentsch, W. (2008) ‘Industrial democracy: Historical development and current challenges’, Management Revue, 19, 4: 260-73. Near, J. and Miceli, M. (1986) 'Retaliation against whistle blowers: Predictors and effects', Journal of Applied Psychology, 71: 137-45. Nyambegera S., Daniels, K. and Sparrow, P. (2001) 'Why fit doesn't always matter: The impact of HRM and cultural fit on job involvement of Kenyan employees', Applied Psychology: An International Review, 50: 109-40. Olson-Buchanan, J. and Boswell, W. (2002) 'The role of employee loyalty and formality in voicing discontent', Journal of Applied Psychology, 87: 1167–74. Paul, S. (1992) ‘Accountability in public services: Exit, voice and control’, World Development, 20: 1047-60. People in Aid (2003) People in Aid Code of Good Practice in the management and support of aid personnel, http://www.peopleinaid.org/pool/files/code/code-en.pdf. 35 Podsakoff, N., Whiting, S., Podsakoff, P. and Blume, B. (2009) ‘Individual- and organizational-level consequences of Organizational Citizenship Behaviors: A meta-analysis’, Journal of Applied Psychology, 94: 122-41. Poole, M. (1992) ‘Industrial democracy’, in Széll, G. (Ed.) Concise encyclopedia of participation and co-management, New York: de Gruyter, 429-39. Popper, K. (1989). ‘Kant’s critique and cosmology’, in Popper, K. Conjectures and refutations: The growth of scientific knowledge, London: Routledge, 175-83. Prasnikar, J. (1996) ‘Participation and self-management in developing countries’, in Pagano, U. and Rowthorn, R. (eds), Democracy and efficiency in the economic enterprise, London: Routledge, 269-305. Purcell, J., Kinnie, N., Hutchison, S., Rayton, B. and Swart, J. (2003) Understanding the people and performance link: Unlocking the black box. Bath: School of Management, University of Bath. Ramsay, H. (1977) ‘Cycles of control: Workers’ participation in sociological and historial perspective’, Sociology, 11: 481-506. Sen, A. (1999) Development as freedom, New York: Anchor. Siddique, S. (1989) 'Industrial relations in a Third World setting', Journal of Industrial Relations, 31: 385-401. Strauss, G. (2006) 'Worker participation - some under-considered issues', Industrial Relations, 45: 778-803. Suzuki, N. (1998) Inside NGOs: Learning to manage conflicts between headquarters and field offices, Warwickshire: Intermediate Technology Publications. 36 Thelen, K. (2001) 'Varieties of labor politics in the developed democracies', in Hall, P. and Soskice, D. (eds) Varieties of capitalism: The institutional foundations of competitive advantage, Oxford: Oxford University Press, 71-103. Thomason, G. (1976) A textbook of personnel management, London: IPM. Van Dyne, L. and LePine, J. (1998) 'Helping and Voice extra-role behaviors: Evidence of construct and predictive validity', Academy of Management Journal, 41: 108-19. Wallace, T. (2006) The aid chain: Coercion and commitment in development NGOs, London: Intermediate Technology Publications. Wilkinson, A. and Fay, C. (2011) 'Guest editors’ note: New times for employee voice?', Human Resource Management, 50, 1: 65-74. Wilkinson, A., Gollan, P., Marchington, M. and Lewin, D. (2010) ‘Conceptualizing employee participation in organizations’, in Wilkinson, A., Gollan, P., Marchington, M. and Lewin, D. (eds) The Oxford handbook of participation, Oxford: Oxford University Press, 3-25. Wood, G. (2010) ‘Employee participation in developing and emerging countries’, in Wilkinson et al., op cit., 552-69. Wood, S. (2010) ‘High involvement management and performance’, in Wilkinson et al., op cit., 407-26. Wood, S. and de Menezes, L. (2008) ‘Comparing perspectives on high involvement management and organizational performance across the British economy’, International Journal of Human Resource Management, 19: 639-83. World Bank (2004). World Development Report 2004: Making services work for poor people, New York: Oxford University Press. 37 World Bank (1996) World Bank participation sourcebook, Washington DC: World Bank. 38 NOTES i Marchington et al. (1993) and Ramsay (1977) also identify waves of enthusiasm for participation. ii Wilkinson and Fay build in their turn on Dundon et al.’s (2004: 1152) earlier taxonomy of meanings of employee voice. iii Kant’s relevant maxim, in Popper’s (1989a: 182) paraphrase, is ‘Always regard every man as an end in himself, and never use him merely as a means to your ends.’ iv We speculate that the similarity is because in both cases we are dealing with a collaborative but unequal relationship. Since such a relationship is anomalous in the light of the current norm of fundamental human equality, rationales must be advanced and regulations put in place to contain the anomaly. This may be why even skeptics like Argyris and Strauss continue to advocate participation. v The donor and partner ‘links’ are absent where INGOs use their own funds and (see Table Three) where they implement their programmes directly. vi We follow the convention of italicizing direct quotations from our interviewees. vii There was, however, no straightforward host country effect in terms of national culture in our findings, which revealed a complex interplay of national, sub-national (e.g. Kikuyu) and organizational cultures. Pace Blunt and Jones and others, this perhaps refutes any essentialist view of ‘Kenyan culture’ influencing agencies’ behaviour. viii We remind readers that our scope is confined to management, not ownership. We repeat that even employee-owned companies John Lewis Partnership and Mondragon tend to operate hierarchically. However, we note Ramsay's (1977: 498) final sentence: 'The whole political-economic environment will have ... to be transformed if a genuine industrial democracy is to prevail.' In our case, there is indeed something unsatisfactory about a world in which INGOs operate in developing countries but are accountable to Boards and donors in the North - as more thoughtful INGO staff would be the first to recognize (Edwards and Hulme, 1996). 39 Rationale Wilkinson and View of employees Relevant Fay paradigms literature Efficiency HRM Instrumental to Organization and OB organization objectives management Ethical Kantian subjects Business ethics Political Industrial Democratic citizens Political science democracy Industrial Industrial Collective bargainers Industrial relations relations relations Table 1 Rationales for participation LEVEL CONSTRAINTS Organizational Hierarchy; competing stakeholders CONTINGENCIES International Varieties of capitalism; home and host country effects National Government stance; employee bargaining power; national culture Organizational Organizational policies; financial pressures; management style, inhibiting or facilitating voice Table 2 Constraints and contingencies Agency A B C D E F G NGO type Public Popular Popular Public Relief & Public Relief (Clark, service dev’ment develop service welfare service & 1991) contractor agency ment contractor agency contractor welfare HQ country LME CME (HQ in LME LME LME CME economy* Kenya) Style of Direct and Via Via Via Direct Direct Direct delivery via partners partners partners and via and via partners partners partners Levels in 6 3 4 5 4 5 5 hierarchy Union* X  X  X   Staff X  X Elected X X Elected council (or reps on reps similar) some cttees Employee   X  X  X surveys * CME/LME = co-ordinated/liberal market economy.i ** Home country staff only in all cases. Table 3 Summary data on case agencies i Among the larger OECD nations, co-ordinated market economies as per Hall and Soskice (2001: 19-20) are Austria, Belgium, Denmark, Finland, Germany, Japan, the Netherlands, Norway and Switzerland; liberal market economies are Australia, Canada, Ireland, New Zealand, the UK and the USA.  ___________ ______________ Control _____________ Codetermination __________________ Consultation ___________ Communication Information Figure 1 Wilkinson's escalator and Arnstein's ladder  Donor agency  INGO  implementing 'partner'i beneficiaries Figure 2 The INGO funding chain Agency Information Communication Consultation Codetermination Control A X B X X (time 2) (time 1) C X X (time 2) (time 1) D X E X F X G X X (time 1) (time 2) Figure 3 Degrees of participation at case agencies i We use the term ‘partner’ to refer to the indigenous NGOs which INGOs increasingly use to deliver their programmes.