Professor Tendler began by citing the work of one of her former PhD students, Anuradha Joshi, now a Research Fellow at IDS, Sussex. Joshi’s research underlined the crucial role played by a lower class Union of Forest Workers in pressing for the introduction of joint forest management approaches into the West Bengal Forestry Department. One reason why this Union was active in lobbying for a decentralised approach was that its front line worker members suffered at first hand the consequences of trying to enforce the law in relation to deforestation and forest encroachment. They thus had their own reasons for searching out innovative and participatory strategies to engage with local communities.

This case study, she argued, underlined one of the key findings of her work Good Government in the Tropics, namely that development agencies have become so conditioned by prevailing thinking and stereotypes within the donor community about the poor performance of third world governments, and so mistrustful of them, that they fail to profit from the opportunities which do exist to work productively with public sector workers. We need to know more about the circumstances in which public sector workers become committed to their work, if we are serious in our desire to improve the performance of public agencies.

Anybody that has done research on the history of pro-poor service and public sector unions [she continued] will know that reform factions within professional associations have been a major contributory factor and very important allies in social innovation. If one reads these histories, then key categories of public servant: social workers, engineers, architects, teacher and nurses - always figure prominently. Not all may perform effectively, but in any such situation, there is at least a minority of these front-line workers that is committed to pro-poor service delivery. These are not the villains of the story but rather the heroes. Important questions for evaluation work are when and why do these workers play this positive role and when and why do they not?

An overlapping issue is the mediation of conflict. Most stories of progressive reform and improved performance of service delivery have their origins in disagreements and conflicts: stalemates in development. They have been involved in conflict but have ultimately managed to reach an agreement. We have teacher’s unions versus the politicians – major disagreements over bread and butter issues. These are no different from private sector conflicts. We need to know how to institutionalise the management of these conflicts and create permanent institutions for their mediation.

An important part of the story of how to support or enhance the positive associations in the public sector and the unions concerns ways to institutionalise these ‘conversations’. This is a topical issue in South America. It is important to look at these conversations, rather than vilifying public sector unions in the way that has become standard in recent documents. As with the case of education, for example. In the educational sector, the thrust of political economy reform is directed at marginalizing the teachers unions. We are not going to make policy reform happen by adopting this approach.

One caveat, Professor Tendler added, was that she was not saying that this requires a new set of techniques – for example, introduction of arbitration or training in dispute resolution, the central issue is more dynamic and conceptual. We need to look carefully at relevant cases and examine how advances were made. In short, we need to adopt a more historical approach.

Going back to the histories of successful reforms, we know that service delivery improves with dedicated service workers. There is a lot of literature in the UK which goes back to the thirties, dealing with worker motivation. This evidence is not entirely unheard in today’s thinking but the latter is rarely grounded in the literature of political science, with its concerns about individual incentives and disincentives and what incentives can be given to bureaucrats to make them perform effectively. What are the collective values that are shared by the professionals? In Professor Tendler’s experience, this literature is unfortunately very passé in public policy schools though interestingly, it is still very much alive in business schools in the USA and elsewhere.

The flipside of development agencies’ over-preoccupation with technical issues is that highly dedicated professionals in the public sector are made to feel ever more isolated, frustrated. This reinforces the importance of encouraging such workers to get together, to learn from each other, develop a collective identity and engage in collective action.

All this may seem obvious, Professor Tendler argued, but why are professional associations and unions missing from the reports of development agencies such as DFID and the World Bank and how can we support them more? We need to be more interested in harmony in this era, and less in polarisation. This is said to be one of the virtues of decentralisation. In the USA important pro-poor reforms have come about from lobbying Congress. Public health workers, physicians, etc., have been major actors in reform. In the US the welfare organisations, social workers, have been a major lobby that questioned legislation, and pushed for progressive change.

In summary, we are missing the significance of the professional associations and the unions and the ability of their reform fractions to bring round significant improvements in pro-poor service delivery if all we do is categorise them as the ‘bad guys’. We need to move the debate away from the mistrust of government and look for more positive experiences from developing countries.

The discussion was then thrown open to the floor:

Some speakers were very much persuaded by these arguments and by the suggestion that we need to look closely at promising aspects of unpromising environments, without methodological blinkers. The problems (said one) come with the observation that there may be some success stories out there but they have been few in number, and those that have been documented have been very heavily used. They don’t go beyond Brazil or India. If there have been cases that have worked well anywhere else – for example, Africa - it would be good to hear of these other success stories.

Also broadly supportive of Professor Tendler’s thesis, other speakers questioned the effectiveness of currently privileged actors, such as NGOs, if they are not working in alliance with government. This relationship can be synergetic – much more so than when an antagonism is created between the two.

However, others were more sceptical of her line - for example, questioning the alleged hostility to professional associations in development thinking. Instances were reported where development agencies were already working with them successfully. Likewise, another speaker pointed out that there is a new interest in re-energising the state. DFID, for instance, has moved back into re-examining how to better support national processes, it doesn’t have the answer yet but it is challenging the paradigm of the eighties and nineties (ie. reducing the role of governments and encouraging the markets).

An example was cited from DFID (ODA)’s experience in the 1980s in Bangladesh. This involved a successful health sector project in the tea estates. Then a strike exploded, and progress became much more difficult to achieve. In this situation, the unions weren’t working, and the type of framework which Professor Tendler is advocating could be seen as a major obstacle to progress. Thus, unions and associations may be good partners where they are progressive, but the risks are great.

Other speakers reinforced these doubts with reference to professional associations. In Africa, for example, there is evidence that they tend to dominate the political process and don’t reflect the needs for the rural communities. Another speaker agreed: the issue is when can one expect to have unions on-side or when can one not?

Summing up, Professor Tendler doubted that there was any such thing as a ‘success story’ – the best one could hope for was to say that one way of doing things was better than another. We need to look at stories that have shown significant improvements over time. The problem with success stories is that we tend only to see them as exceptions, in their entirety, and do not see each and every case as having aspects of ‘success’. You can choose to look at a set of cases that shows a pattern across different sectors. We mustn’t look for ‘recipes’ of success. The issue is more concerned with changing performance over time.

She made clear also that she was not saying that development assistance should be concerned to create or promote any particular style of associations or get involved in politics in whatever form it takes but rather, that we should be attentive to this need to build capacity, and the historical circumstances in which it occurs. The kind of question that needs to be asked is: is there a reform faction in a particular labour union? If there is, then we have to figure out how to support it and turn it into something stronger.