Incubating policy for economic transformation: Lessons from Nepal

Booth, David A. Research Report Incubating policy for economic transformation: Lessons from Nepal ODI Report Provided in Cooperation with: ODI Global, London Suggested Citation: Booth, David A. (2018) : Incubating policy for economic transformation: Lessons from Nepal, ODI Report, Overseas Development Institute (ODI), London This Version is available at: https://hdl.handle.net/10419/190849 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc/4.0/ Incubating policy for economic transformation Lessons from Nepal David Booth April 2018 Report Overseas Development Institute 203 Blackfriars Road London SE1 8NJ Tel: +44 (0) 20 7922 0300 Fax: +44 (0) 20 7922 0399 Email: [email protected] www.odi.org www.odi.org/facebook www.odi.org/twitter Readers are encouraged to reproduce material from ODI publications for their own outputs, as long as they are not being sold commercially. As copyright holder, ODI requests due acknowledgement and a copy of the publication. For online use, we ask readers to link to the original resource on the ODI website. The views presented in this paper are those of the author(s) and do not necessarily represent the views of ODI. © Overseas Development Institute 2018. This work is licensed under a Creative Commons Attribution-NonCommercial Licence (CC BY-NC 4.0). Cover photo: A young Nepali man cutting a marble slate in Pokhara, Nepal © Igor Ovsyannykov 2017. 3 Acknowledgements This paper has been prepared with funding from the Economic Policy Incubator, supported by the UK Department for International Development in Nepal. However, responsibility for its content lies with the author. He is grateful to all those in Nepal who provided him with information, insights and opinions on the issues discussed, not least various members of the EPI programme team itself. Useful comments on drafts were provided by Fabiola Esposito, Harry Jones, Ram Khadka, Shankar Sharma and Danielle Stein; Edward Laws kindly acted as peer reviewer, making several helpful suggestions; and Roo Griffiths edited the copy with her usual skill. 4 Contents Acknowledgements 3 Acronyms 5 Executive summary 6 1 Introduction 8 1.1 The practice deficit 8 1.2 Why EPI 8 1.3 Scope and aims 9 2 Smart economic transformation: from theory to practice 10 2.1 Jump-starting economic development 10 2.2 Getting politically smarter 11 2.3 Doing development aid differently 11 3 Nepal’s Economic Policy Incubator 13 3.1 Chronic problems, complex changes 13 3.2 EPI’s set-up and approach 14 3.3 Early successes 15 3.4 Consolidating rela tionships 16 3.5 Increasing scope 16 3.6 Moving towards impact 17 4 Enabling conditions 18 4.1 DFID Nepal 18 4.2 Programme design and management 18 4.3 Monitoring and reporting 20 5 Looking forward 22 5.1 Keeping focused 22 5.2 The new demands of sub-national programming 22 5.3 Engaging with firms 23 5.4 Following up EPI post-2020 23 6 Transferable lessons 25 6.1 Going all the way on ‘local leadership’ 25 6.2 A breakthrough on monitoring? 25 References 26 5 Acronyms AiiN Accelerating Investment and Infrastructure in Nepal APPIIC Accelerating Private and Public Investment in Infrastructure CIG Centre for Inclusive Growth CITCW Committee on Industry, Trade and Consumer Welfare CPN–MC Communist Party of Nepal – Maoist Centre CPN–UML Communist Party of Nepal – Unified Marxist-Leninist DRC Democratic Republic of Congo DFID UK Department for International Development EPI Economic Policy Incubator IBN Investment Board of Nepal IEA Industrial Enterprises Act IFC International Finance Corporation M&E Monitoring and evaluation MoI Ministry of Industry MoU Memorandum of understanding MP Member of parliament ODI Overseas Development Institute PDIA Problem-driven iterative adaptation PPP Public–private partnership SAWTEE South Asia Watch on Trade, Economics and Environment SEZ Special economic zone SEZB Special economic zone at Bhairahawa TAF The Asia Foundation ToC Theory of change UK United Kingdom USA United States of America USAID United States Agency for International Development 6 Executive summary A substantial body of theory now exists on economically sound and politically smart ways of jump-starting progress in poor developing countries. At several levels, however, the practice is lagging behind the theory, meaning there is much to learn from any new experiences suggesting precisely how to advance this agenda. This paper reports some early successes from a UK Department for International Development (DFID)-funded programme in Nepal, the Economic Policy Incubator (EPI). Although this programme is at a relatively early stage, it has some highly transferable features and has already generated valuable lessons. The paper begins by expanding on the proposition that practice is not yet matching theory. This is true, it is argued, at three levels: (1) economic policy priorities in a world of global value chains; (2) smart ways of navigating the political obstacles and reaping the political rewards of wise policy choices; and (3) approaches to development assistance that support the ambitious pragmatism that is at the root of successful economic transformation. At the third level, the needed shift is from solution-driven to problem-driven aid, and from ‘blueprint’ to adaptive programme designs. While the reasons why this is a difficult transition are well known, how to overcome the main hurdles is much less clear. EPI is an innovative response to chronic deficiencies in Nepal’s economic policy processes that threaten to hinder the recovery from political instability and economic stagnation that the country now needs. It is distinguished by being led by a well-known and widely respected team of senior Nepali economists, and by a smart working method. The method rests on targeting policy problems that are both important and tractable, in the sense that powerful stakeholders or a potential coalition wish to see them resolved; making ‘small bets’ on a range of such problems; and revisiting strategies and assumptions at regular intervals to revise, scale up or abandon approaches on the basis of anticipated effectiveness. The team applies a toolkit including brokering, convening and technical assistance. Research evidence is brought to bear in response to problem definition and demand, rather than generated in advance in expectation of policy uptake. Within a few months of completing its inception phase, EPI contributed to a significant breakthrough in fast-tracking and improving the legislation on special economic zones (SEZs). It went on to facilitate important amendments to other Acts that are critical to the investment climate in Nepal. These results were the product of a learning process that led the team first to abandon an initial ‘bet’ on improving the bureaucratic procedures for the entry of new businesses and then to discover ways of addressing the political concerns that were holding up the passage of key laws in parliament. This early success has been followed by further convening activity and technical assistance designed to enable Nepal’s first SEZ to become operational. Problem identification and otherwise absent coordination efforts have been undertaken with mayors and administrators in the newly established municipal tiers of government in the economic corridor where the first SEZ is located. A range of other economic policy weaknesses and gaps combining the qualities of importance and tractability are being tackled in a similar way. EPI has been able to apply the principles of problem-driven adaptive working or ‘development entrepreneurship’ in a relatively uncompromising way, which has not often been possible in programmes funded on a substantial scale by bilateral donors such as DFID. The paper attributes this to a layered set of enabling factors, including some distinctive features of DFID Nepal; the country track record and intellectual contributions of the design and support team provided by the lead contractor, Palladium; and an exemplary approach to reconciling adaptive working with rigorous and transparent reporting to the funder. Looking forward over the remaining two to three years of EPI’s expected lifetime, the key challenge that will be faced is avoiding any dilution of the following principles: •focusing on problems that stakeholders are motivated to solve; •maintaining a range of tactical options with respect to entry points and coalitions; •combining high-grade technical advice with formal and informal brokering, convening and persuasion. The paper suggests that the upcoming dangers include pressure to provide increasing amounts of conventional technical assistance, especially in the new sub-national tiers of government, and failing to adapt the approach to the somewhat different contexts of provincial and municipal economic governance. It recommends a substantial increase in the team’s capacity for brokering and facilitation work with firms and businesses. When EPI reaches its scheduled end in 2020, what should be the next step? The paper makes a plea for realism about the timescale on which the dysfunctions of Nepal’s policy system might be tackled by some form of institutionalisation of EPI’s practice. The effectiveness of 7 EPI depends crucially on the team’s not being part of the civil service and subject to its rules and informal norms. The need for the kind of ‘policy incubation’ it practises is going to continue for some time. Any suggestion that this is an insufficiently ‘transformative’ agenda should be rejected. World-wide historical experience suggests that selective removal of critical blockages to economic progress almost always precedes widespread institutional modernisation. Palladium and EPI should disseminate actively the lessons learnt from their experience to date. Particularly important and widely relevant lessons can be drawn about: •the benefits and the feasibility of implementing a strong form of the ‘local leadership’ principle in problem-driven programming by appointing a distinguished national of the country to the programme leader position; •the gains from, and practicality of, an approach to programme monitoring and reporting that sets out quantitative indicators of performance against a ‘menu’ of desirable outcomes and against outputs that constitute the typical steps in an agreed form of problem-driven adaptive practice. 8 1 Introduction 1 For more on the background, see Harford (2011) and Sims (2011). After 50 years of unprecedented progress in building economies and raising living standards across the globe, a great deal is now known about how to make development happen. This knowledge includes a solid understanding of three topics: 1. what it takes to set in motion cumulative processes of economic transformation in initially poor countries by implementing well-chosen policies, especially in the field of globally oriented manufacturing; 2. how, in a great variety of country contexts and periods, leaders have found ways around apparently overwhelming political and institutional barriers to put policies to work and set in motion cumulative change; and 3. how foreign aid can be reformed to make it more effective in supporting change in complex politicaleconomic systems, by following and promoting a problem-driven and adaptive approach. In each of these three areas, however, the theory is running well ahead of the practice. 1.1 The practice deficit To begin with, while jump-starting economic development with well-designed industrial policies presents a massive opportunity for nearly all less developed countries, only a minority are currently grasping this opportunity firmly. Evidence on how to do so is therefore more historical than current. Second, leaders in many countries find it difficult to perceive how to navigate the immediate political obstacles in order to harvest the potential political pay-offs from mass job creation in manufacturing. There are few formulas for doing this that seem replicable across countries and political regime types. Finally, debates about the effectiveness of foreign aid have generated broad consensus on the value of problem-driven adaptive approaches, not least when seeking to support economic development. However, aid-funded programmes that are fully committed to this type of approach are fewer than the rhetoric might suggest. Those successfully applying the principles are probably fewer still. Implementation of adaptive working is challenging for several reasons, and there is a severe shortage of successful models. It follows that early dissemination of good or promising experience in any one of these fields of lagging practice is an important task. Leaders of countries’ development efforts, international agencies, development think-tanks and implementing organisations alike have reasons to share and reflect on the latest experience of relevance to what they are collectively seeking to achieve. This paper aims to contribute to such a shared learning process by reporting on the experience to date of Nepal’s Economic Policy Incubator (EPI). This is a five-year programme led by senior Nepali economists and funded by the UK Department for International Development (DFID) through the commercial service provider Palladium. It is only two years into implementation but is already showing promise, first as an enabler of the manufacturing revival that Nepal needs; second, as a pioneer of politically smart problem-driven policy reform; and third, as a case of adaptive programme management succeeding under the constraints large-scale bilateral donor funding imposes. 1.2 Why EPI The view of this paper is that EPI is at the leading edge of current practice in the field of problem-driven adaptive support to economic development. It exemplifies a successful application of the principles of ‘development entrepreneurship’ set out by Jaime Faustino, based on his experience with The Asia Foundation (TAF) in the Philippines. Development entrepreneurs are politically smart indigenous reform leaders who, by making ‘small bets’ and ‘failing quickly to learn fast’, replicate the adaptive approach that has proven successful in business start-ups and other fields (Faustino, 2012; Faustino and Booth, 2014).1 The results achieved with this approach in the Philippines have been described previously in a series of books and papers (TAF, 2011; Booth, 2014; Fabella et al., 2014). They include the liberalisation of air and maritime transport, substantial improvements in urban land tenure and the creation of new revenue streams that have financed widened access to primary health care. Each reform had a measurable impact, and the cumulative effects on the process of national economic development have been very significant. The results attributable to development entrepreneurship in the Philippines are not yet matched 15 and joined up actor that can address economic policy limitations across the board, but (2) there are particular organisations and individuals that are motivated to address particular issues, especially if they are brought together with like-minded others and stimulated to think in fresh ways about potential solutions. The method consists of convening those actors in a sensitive way and prompting them to work together to identify and remove the key blockages. The element of technical assistance is essential for two reasons. It provides a convenient way of ‘opening doors’ and starting conversations. And it is important for arriving at modalities of collective problem-solving that are technically sound (that is, liable to make a significant difference to economic development performance) as well as politically feasible. 3.3 Early successes EPI will eventually be judged by its cumulative contribution to easing the constraints and harnessing the opportunities affecting investment, production, job creation and exports in Nepal. In its first two years, it has already registered some striking successes in a couple of policy areas. At this stage, these contributions are not certain to have a major impact, given the complexity of the processes involved. However, feed-through to impact is considered likely on the basis of a robust assessment of the plausibility of different pathways of change. More important, these early successes exemplify a formula for ‘incubating’ economic policies that seems suited to producing a cumulative flow of such gains over the remainder of the programme period and perhaps beyond. Within a few months of its inception, EPI had already made one notable contribution to Nepal’s economic policy environment. The centrepiece is the legal and regulatory regime for SEZs and its implementation in Nepal’s first SEZ, inaugurated in 2014 and located close to the Indian border at Bhairahawa (SEZB). Addressing the principal infrastructural and institutional constraints on manufacturing by the SEZ method (that is, through geographical concentration of effort) emerged as a focus soon after inception along with targeted interest in public and private investments in the Economic Corridor from Pokhara to Sunauli, in which SEZB is located. The main issue was that the governing legislation, in the form of an SEZ bill, had been stuck in parliament for eight years. Without its passage, SEZB could not be made operational. Initially, work on the SEZ issue was accompanied by interest in improving Nepal’s poor performance against the World Bank’s Doing Business criteria, which focus on ease of entry for new investors. This would have rested on work with the Ministry of Industry (MoI) to simplify business registration and compliance procedures. Consultants supervised by EPI Senior Advisor Hiramani Ghimire prepared an Agenda for Simplifying Business Regulation in Nepal, and the MoI validated and accepted this (EPI, 2017a). However, early consultations with the leadership of the MoI and other stakeholders detected little serious interest in pushing through the required changes in procedures. It was also confirmed that these could be expected to incur significant bureaucratic resistance. Meanwhile, discussions in parliament, especially with the Committee on Industry, Trade and Consumer Welfare (CITCW), suggested an unexpected level of interest among parliamentarians in unblocking the SEZ legislation. Expressions of appreciation by the CITCW for the technical advice offered by Dr Sharma and his team on the SEZ bill encouraged EPI to revise its initial focus on business entry procedures and redirect its efforts to legislation affecting broader aspects of the investment climate, including amendments to the Industrial Enterprises Act (IEA) and the Companies Act. This shift involved EPI in revising its initial assumption that, given Nepal’s fractious party politics and shifting parliamentary coalitions, legislation would be a harder route to effective policy improvement than changing bureaucratic procedures. While first impressions and standard ex ante political economy analysis might have supported this assumption, engagement with CITCW members on the detail of the SEZ bill revealed a more complicated picture. Not only were the Minister of Industry and his secretary keen on getting the SEZ law passed, but also there were champions in the relevant parliamentary committee. A principal reason for the blockage was found to be the inclusion in the bill of a clause limiting labour strikes in SEZs, which had been opposed by most political parties and their affiliated trade unions. In informal meetings with the parliamentarians, however, the EPI team found some support for its inclusion, even among left-leaning MPs. Some of them were concerned about the impact of factory closures caused by strikes in their constituencies. Others had been influenced by their knowledge of the background to notorious inefficiencies in government-owned firms. This provided fertile ground for acceptance by committee members of broader arguments about the impact of Nepal’s reputation for unruly labour relations on future foreign and domestic investment, especially in SEZs. The EPI team worked at first quite informally, including through bilateral conversations with individual MPs. Later they brokered and facilitated more formalised evidence sessions with the whole committee and external stakeholders. Dr Sharma’s presentations on experience in other countries were backed up with study tours to southern China and Thailand. Influential members were persuaded to sponsor the bill in committee. Having been persuaded of the potential political gains from increased manufacturing employment, the politicians were apparently able to get sufficient assent from the party-linked trade union leaders. A secondary blockage to the passage of the SEZ law was also addressed in the committee, thanks in part to politically smart technical guidance from the EPI team. The draft bill required investors benefiting from SEZ 16 conditions to export 100% of their output. The final text lowered the requirement to 75% to meet objections from private sector representatives that could have prevented its approval. With the endorsement of the bill, including the controversial clauses, at the committee stage, the Act was passed by parliament on 30 August 2016 without further opposition. This breakthrough opened up a stream of work to operationalise the provisions of the Act, with particular reference to SEZB. The principal counterpart for this work is the new and relatively inexperienced SEZ Authority. The work takes the form of high-grade, politically attuned, technical assistance to the preparation of implementation guidelines that will ensure sound implementation of the principles set out in the law. This includes active problemsolving on such issues as the legal framework for electricity supply to the SEZB site, and leveraging additional public investments in the site’s infrastructure. 3.4 Consolidating relationships The SEZ work forged a partnership between EPI and the CITCW that seems likely to survive the current transition from a three-way governing coalition (Nepali Congress plus two leftist parties) to a two-party leftist one. A memorandum of understanding (MoU) between EPI and the committee commits EPI to providing ongoing advice and capacity-building, including the preparation of a procedural handbook for new members. Under the new constitution, parliament is smaller and there will be fewer parliamentary committees, but it is expected that the agreement will be picked up by the successor committee whose remit includes trade and industry. The relationships consolidated in this way permitted EPI to play an influential part in steering the IEA and Companies Act amendments through the committee stages. Improvements in the content of these laws are expected to make a significant difference to the climate for investment. In several respects, the changes are attributable to wise committee decisions based on arguments put forward by EPI. For example, the IEA as amended entitles firms to submit required documentation electronically and requires ministries to recognise firms’ digital signatures. This is expected to relax at least some of the bureaucratic delays involved in processing paperwork in the traditional manner. Environmental and social impact assessments are now required after land for factories has been acquired and before company operations start, rather than, as previously, at the initial investment proposal stage. The IEA also restores previously abolished fiscal incentives and includes some limits on wildcat strikes. Some seven amendments of these types were made to the Companies Act, drawing on proposed improvements suggested years previously in a study by the International Finance Corporation (IFC, World Bank Group) but never acted upon. EPI’s early successes, most of which took place within 18 months of the programme start-up, have confirmed the team’s belief in selective issue identification and agile coalition-building. The principle that powerful stakeholders must be interested in, and willing to assume ownership of, a reform has been powerfully endorsed. Experience has also shown that a senior team combining technical expertise and a reputation for political impartiality can use offers of technical assistance to open doors and influence people in ways that sidestep the notorious dysfunctions of the Nepali policy process. 3.5 Increasing scope It is on this basis that the team has expanded its agenda of ‘small bets’. On the investment climate, a Foreign Investments Bill improved with EPI help is due to go to the newly elected parliament in early 2018. Thanks to EPI, the MoI has a vision document giving a central place to reinvigorating manufacturing and creating employment opportunities over the decade 2017-27. The SEZ follow-up has included the approval of necessary bye-laws and regulations by cabinet, securing agreement on a roadmap for the operationalisation of SEZB with key stakeholders, removal of uncertainties around the SEZB electricity supply and an additional budgetary commitment to implementing the SEZ policy. In addition, new projects have been started on (1) sub-national economic growth (support to investment policies and procedures at the new municipal level of elected government, focusing on the Pokhara– Sunauli corridor), and (2) facilitating investments in Solving problems in the Bhairahawa SEZ. © Harry Jones 17 e-commerce. Thirdly, responding to a request from a Ministry of Finance official, who was the first chair of the programme’s Steering Committee, EPI has been providing training in economic analysis for civil servants in a sub-programme called Economic Brain. Of these recent and current initiatives, the most important is probably the move to sub-national work. The creation of two new tiers of elected government, at provincial and local (municipal) levels, is one of the most significant features of the constitution now being implemented. What happens at these levels will be a major determinant of the prospects for inclusive growth and economic transformation during the coming decades. Since the required technical, administrative and decisionmaking capabilities are going to need to be built almost from scratch, the demands for routine technical support and training are going to be large. For all these reasons, sub-national work represents a great opportunity but a dangerous one from the perspective of EPI’s distinctive method and contribution. The work has begun in a promising fashion. The team has had conversations with the mayors and secretaries of four of the new municipalities in the Pokhara–Sunaili corridor (Pokhara-Lekhnath, Butwal, Tilottama and Siddharthanagar) to identify investment projects and other initiatives of priority interest to them. EPI also supported wider consultations and evidence-based prioritisation processes in some municipalities. MoUs have been signed with the four mayors, under which EPI is committed to advisory and facilitation support. The investment priorities include bus stations, leisure parks, multi-storey car parks and other infrastructure projects that are attractive both politically, as symbols that the mayors are having an impact, and as a potential source of local tax revenue. Since the capital investment required is more than municipal budgets are likely to permit, there is interest in PPPs. Under the new constitution, both the provincial and the municipal councils have legislative powers, and these powers can be used to translate federal legislation into appropriate local versions. EPI, using specialist consultants, has been helping with this task as it applies to localising the federal law on PPPs. They are also working on longstanding problems in the business registration and tax systems at local level. EPI’s sub-national workstream is geographically targeted, partly with a view to creating demonstration effects that can be emulated within the corridor and elsewhere. However, it has set out to ensure that any learning on what works within the decentralised structures gets reflected in national policy. It helped the government establish an inter-ministerial task force on the implementation of decentralisation, and the MoUs 4 Although significant at £6 million over four and a quarter years, EPI’s budget is small by many standards. It is a modest share of the £35 million awarded to the Accelerating Investment and Infrastructure in Nepal (AiiN) programme of which it is a part. Over the four years corresponding most closely to EPI’s programme period, DFID Nepal’s total project budget is projected to be £426 million (source: DFID Development Tracker). with the selected municipalities are anchored by regular reporting to the National Planning Commission. Little of this effort would be necessary if Nepal’s government system were well coordinated and steered consistently through time by a coherent vision for national development. However, as we have seen, the actual state of affairs is that these qualities are for the most part absent, and the challenges are compounded by the decision to undertake an ambitious decentralisation programme. Under these circumstances, the type of brokering of agreements, facilitation of change processes and expert steering that EPI is qualified to provide may make a very substantial, and ultimately measurable, difference to the outcomes that matter for Nepal. 3.6 Moving towards impact There is a real sense in which EPI’s real challenges still lie ahead. The progress that has been made in improving the general investment climate and the SEZ policy framework in particular through intelligent legislation will count for something important only when the Bhairahawa SEZ is operational and investments begin to flow. Various anticipated snags are being dealt with now and others may well emerge over the next couple of years. Nevertheless, there are already signs that the expected impacts will be realised. For example, the IEA provisions on online registration of businesses and the timing of impact assessments are up and running, with the necessary software and hardware in place. On the basis of the legislative changes, Nepal has crept up a couple of places in the Doing Business rankings. As of March 2018, there has been a 14% increase in registration of industries at MoI. Confidence among businesses that SEZB will soon be operational is suggested by the fact that thirteen investors have applied for permission from SEZA to register their industries, foreseeing the creation of 2,963 direct jobs. The roadmap for the development of e-commerce, supported by EPI, has been agreed by the Minister of Commerce. Two commercial banks have installed payment gateways, and the central bank has authorised 28 banks and three other organisations for online banking. Despite being at an early stage, the partnership agreements on local economic growth signed with the four municipalities should begin to yield concrete results within the next year. If these trends continue – especially in the SEZ policy area – EPI will have made a very significant difference to Nepal’s economic transformation prospects. It will have done so at relatively modest cost to the UK taxpayer.4 18 4 Enabling conditions EPI is already influencing outcomes and showing real promise of continuing to do so. It is working on a broad front and with a significant budget, provided by a bilateral donor agency that is strongly committed to Nepal but also under political pressure to exercise tight control and stay focused on easily communicated results. Ordinarily, such pressures are discouraging, if not prohibitive, of the kind of problem-driven adaptive contributions that EPI is delivering. As noted in the Introduction, the best examples until now of effective development entrepreneurship have benefited from more modest donor funding, provided on much more relaxed terms than those currently governing UK aid. What has made EPI possible is a layered set of enabling conditions. Some of these are specific to DFID’s Nepal office and may not be fully replicable in other countries, although they include some sound general principles. Others have to do with the way the programme has been designed and managed, several features of which seem highly transferable – one of the reasons it seems worth reporting about them at this relatively early stage in the programme’s operational life. 4.1 DFID Nepal The Nepal office of DFID has some track record of supporting innovative programming, especially in the area of politically smart support to economic development. The Centre for Inclusive Growth (CIG), the predecessor of the umbrella programme Accelerating Investment and Infrastructure in Nepal (AiiN), under which EPI sits, was a pioneering example of an advisory facility staffed mainly by Nepalis and intended to provide targeted technical support to the then most dynamic unit of the national government, the Investment Board of Nepal (IBN). The CIG/IBN link-up had a large positive impact, especially on the agreements needed for Nepal to exploit its considerable potential in hydroelectric power generation (Thapa and Basnett, 2015; Laric and Waddell, 2016). The CIG experience encouraged the growth team in DFID Nepal to include in the Business Case and Invitation to Tender for AiiN an ‘anchor component’ intended to address selected institutional problems in economic policy using a flexible and learning-oriented approach (DFID, 2015). The impact of CIG had been written up by DFID advisers alongside other cases of politically smart support to economic growth in a challenging policy context – in the Democratic Republic of Congo (DRC) and Nigeria (Ramalingam et al., 2014; Booth, 2016). This kind of thinking by DFID may come most easily in countries whose economic performance is markedly below potential and where the reasons appear to lie in chronic but potentially tractable institutional dysfunctions. Arguably, there are few DFID partner countries that do not fit that description. However, the inadequacy of ‘business as usual’ is particularly stark for aid donors in places like DRC, Nigeria and Nepal. As a result, DFID Heads of Office in those countries have been unusually supportive of innovation in the country portfolio, which has facilitated the adoption of politically smart adaptive programme designs (Wild et al., 2017). In the Nepal case, the DFID invitation to tender for the implementation of EPI signalled strongly the need for a politically smart, problem-driven approach. The ‘incubator’ concept included the idea of identifying practical problems that Nepali organisations are willing to address. It called for a flexible results framework and ‘a sophisticated approach to dynamic Theory of Change formulation, analysis and updating’ (2015: 3). The critical measure of success would be the number of policies significantly improved, based on a menu of areas needing attention. The implementers would need problem-solving, facilitation and influencing skills, and a team culture including the ability to learn from failure (2015: 5). In line with this concept, DFID Nepal has adopted a relatively ‘light touch’ management style, focusing on outcomes and avoiding undue intrusion into the approach to output-delivery developed by the implementing partner, Palladium. Not all of the same signals and protections have been given to two other components of AiiN, both of which are larger. However, this may have worked in EPI’s favour. Being under an umbrella with more conventionally run programmes has given it more freedom to operate tactically than it might have had as a stand-alone programme with a high individual profile. 4.2 Programme design and management While DFID signalled strongly its interest in flexible policy ‘incubation’, the detail of the EPI approach was worked out in Palladium’s bid document. It was then refined in the programme’s Inception Report, prepared during the three months following signature of the contract. This detail drew heavily on the country experience and networks of members of Palladium’s bid 19 team, of whom a number were well versed in the theory and practice of adaptive programming. One of the things this experience suggested strongly was the importance of assigning programme leadership to a senior Nepali with government experience. Another was the importance of supporting that person with a weighty group of other team members with both the technical expertise to advise on economic policy and inside knowledge of key policy systems and personnel. CIG was reckoned to have been less effective than it might have been because, although mainly staffed by expatriate and locally hired Nepalis, it was led by a foreigner. It also relied on the special, semi-autonomous status of IBN, which insulated it from normal civilservice reporting lines and procedures. This status eventually proved unsustainable. It was also inconsistent with Palladium’s preferred approach of working selectively with the main government bodies to facilitate incremental improvements. The Palladium team, supported in this respect by DFID’s senior economic advisers, must be credited with insisting on the Nepali leadership point and overcoming the typical donor tendency to favour programme leaders that know the aid business and in this respect offer a ‘safe pair of hands’. Thanks to its solid networking, Palladium was also in a position to bring into the proposed EPI team one of the most distinguished and able of Nepali economists: Shankar Sharma has proven to be an exceptionally competent and inspiring team manager as well as an unmatched opener of doors to influence for EPI. His deputy programme leader and the two senior advisers also appear to have been well selected, bringing the necessary combination of expertise, independence, government experience and professional networks, including in a couple of instances DFID and/or World Bank postings. The interview evidence for this paper suggests that the levels of distinction and skill brought together by Dr Sharma and his colleagues constitute on their own one of the most salient enabling conditions of what EPI can do. A less highly reputed and capable team would not have received the hearing or elicited the responses from senior politicians and bureaucrats that are the key ingredients of the EPI recipe. That having been said, the distinctive mode of working that EPI has developed owes much to an excellent working relationship between Palladium and the Nepali programme leadership, with DFID’s generally supportive posture in the background. This relationship has made EPI a powerhouse of bright ideas for smart working on economic policy. Particular ideas that have been useful in giving practical form to the basic approach include: •focusing on demand-led opportunities to improve policy on specific issues, as opposed to trying to build policy-making capacities on a broad front; •using evidence and analytical work to feed this demand, rather than seeking ‘uptake’ for research findings; •working up an integrated package of EPI outputs appropriate to each issue area, rather than following a standard sequence in which dialogue is followed by technical assistance and then by institutional strengthening; •engaging broadly with stakeholders across the policy system, including a number of different stakeholders in government and civil society, thereby avoiding Identifying local development opportunities, Pokhara. © Harry Jones 20 ‘capture’ by any one of them and maximising the chances of a winning combination; •maintaining conversations at different levels of government (Secretary, Joint Secretary, Section Head, etc.), to ensure both high-level authorisation and continuity when governing coalitions change and key contacts get transferred; •casting the net wide to search out a number of high-impact yet tractable issues to work on, thereby managing the uncertainties and resulting attrition rates of interventions (due to the continuing volatility in the political environment and new scenarios arising from the constitutional changes as well as the perennial implications of institutional complexity). These were recognised and endorsed in EPI’s first Annual Review by DFID (August 2017) as wise strategic choices. According to the review, they should continue to govern the programme over the next few years, with the exception of the final bullet point. From Year 2, EPI would focus mainly on delivering against existing issues, while continuing to employ imaginative tactics and accommodate new learning in doing so. 4.3 Monitoring and reporting One of the most innovative and broadly relevant dimensions of the EPI/Palladium approach is performance monitoring and reporting progress to the funder, DFID. As mentioned in Section 2, the apparent conflicts between genuinely problem-driven and adaptive programme management and the requirements and habits of traditional monitoring and evaluation (M&E) have been one of the principal stumbling blocks in this field. To the knowledge of this writer, no current development programme operating on a significant scale with official donor funding anywhere has achieved a really successful marriage of adaptive working and donor progress reporting – that is, where the reporting frame supports the learning process while also providing a high level of results-based accountability. Until now, the most serious efforts to achieve such a marriage (e.g., Derbyshire and Donovan, 2016) have relied on frequent modifications of the outputs in the logframe or results framework to reflect learning or changes in circumstances. This is unsatisfactory because assessors (e.g. those leading DFID Annual Reviews or Mid-Term Reviews) are expected to score a programme’s performance in delivering its outputs as well as its expected contribution to outcomes and impacts. This is not possible if outputs are frequently changing. In practice, some reviewers of programmes that have been set up to be flexible and/or adaptive take the view that they should also score the quality of the learning or the wisdom of the adjustments. But this is also unsatisfactory if there are no proper criteria to guide these judgements. Compromises of this sort also leave implementers and M&E personnel under the sway of traditional ‘blueprint planning’ ideas, under which non-delivery of a pre-defined output feels like evidence of programme failure and managers have few incentives to recognise approaches that are not working. It is clear in principle that adaptive programmes will not show their true potential unless and until their Outcome indicators Number of top 9 crosssectoral constraints to inclusive and transformational growth significantly eased (cumulative) Number of sectoral and regional opportunities for inclusive and transformational growth harnessed (H), and number of these that are harnessed on a significant scale (S) Number of key issues where resources are harnessed from government, donors, private sector and others, and average achievement of targets (%) Number of significant steps taken towards more effective policy by EPI counterparts and partners Number of issues around which institutional structures and capacities have been significantly improved Output indicators Sub-output 1: Problem identification, project design and implementation Sub-output 2: Project adaptation and learning Sub-output 3: EPI engagement and learning Number of significant and tractable policy problems identified with stakeholders Number of quality project plans developed with key actions agreed with partners Average % achievement of planned deliverables % of active EPI initiatives adapted based on implementation experience Number of projects that are scaled up (S) or dropped (D) based on evidence and EPI learning processes Number of ideas and lessons generated or learned to support improved policy and programming Number of processes or mechanisms developed that embed strong relationships with core economic governance bodies to support policy incubation and implementation Table 1 EPI performance indicators (2016–17) 21 M&E systems are geared to recording and rewarding the kinds of learning and adaptation that are important for contributing to outcomes. At the same time, any such systems need to provide acceptable evidence that implementation is ‘on track’ and implementing teams are applying the agreed approach to a high standard of quality and at a reasonable pace. One of the features that appear indispensable in making such evidence acceptable to DFID is quantification. Clever use of appropriate quantification is one of the things that distinguish the EPI results framework, making it an exemplary effort to reconcile effectively the twin demands of problem-driven adaptive working and robust scoring of output performance. A basic understanding of the approach can be taken from Table 1, which shows the indicators for outcomes and outputs in EPI’s initial results framework. Performance against these indicators was the basis of the 2017 Annual Review. The columns of the results framework are in rows in Table 1, for ease of presentation. The milestones (targets) and means of verification corresponding to the indicators are left out for the same reason. What is significant and distinctive about these sets of indicators is that they identify things that can be counted (making them useful internally as well as for external reporting, because they incentivise a rigorous pace of work) but they are also consistent with the open-endedness of the problem-driven approach and do not impose distortions on adaptive working. The things that get counted at the output level are the building-blocks of the way of working to improve policy that EPI has adopted. Therefore, the focus for performance scoring is on how well and how vigorously the approach is being applied. This has been accepted by DFID as an appropriate monitoring scheme; each of the things counted is seen and approved by DFID, allowing standard objections to output indicators that ‘only’ measure processes to be overcome. As a result, EPI is being monitored as it should be, with a strong focus on desired outcomes and the steps and packages of activities that are necessary in the real world to make them materialise. Two further things are probably essential to making all of this work. One is the contribution of a two-person EPI M&E team, who give constant attention to operationalising the basic indicators and ensuring there is evidence to support the claims made, including ensuring this happens ex ante – in other words, in advance of the required reports (EPI, 2018). The other is a relatively elaborate set of procedures and templates for documenting the rationale, aims, assumptions and plans of action covering each of the selected issue areas. These are based on brainstorming and planning sessions involving the whole team, usually chaired by the team leader. They also involve a substantial recording, drafting and communication effort on the part of the technical and M&E personnel provided by Palladium. This combined effort helps in achieving two important things. On the one hand, it ensures the whole team is of one mind on what EPI is seeking to do and why. On the other hand, it provides a high level of transparency about the process, so that accountability to the funder does not rely exclusively on the necessarily crude machinery of the results framework. Sufficient resourcing of the indicator work and recording functions seems an essential condition for this to work well. The standard items of EPI documentation are: • Issue Theories of Change (ToCs), for each selected problem area. Up to 20 pages long, these include a problem analysis, hypotheses about the economics of the issue, and about scale, sustainability and feasibility; an assessment of potential impact and intermediate outcomes; and a proposed strategy. The Older Issue ToCs have an annex listing any revised assumptions and changes to strategies. • Project Plans, detailing the actions, deliverables and inputs for each of the interventions a given Issue is thought to require. • Case Studies, for reporting to DFID in narrative form on significant EPI contributions to programme outcomes. These materials provide a solid body of documentation that can be drawn on in completing the narrative part of DFID’s annual reporting. To the credit of DFID Nepal, the standard Annual Report format has been customised to reinforce the emphasis on what has been learnt and how the programme has changed over the year. It calls for answers, intervention by intervention, on the assumptions made, what has been learnt, how success is being measured and what will be done differently in future. This too seems exemplary and worthy of emulation. 22 5 Looking forward What are the main challenges facing EPI over the coming years? Answers to this question need to be offered with some humility, as they are based on an intensive but relatively brief exposure to the EPI documentation, its counterparts and some of its ‘live’ processes. If they have merit, compared with the insights of those who have a much better grasp of the detail, it is because they convey something of how EPI looks from the outside, from the perspective of the progress and prospects of problemdriven adaptive programming worldwide. 5.1 Keeping focused If what we have seen in this paper so far is true, the key challenge going forward is to maintain a high level of consistency in applying the principles and methods that have already proven effective. The principal risk is that of allowing the approach to be diluted. Several ways this could happen are associated with the technical assistance aspect. Currently, the responsive technical assistance being provided includes some training of a relatively generic sort (on basic econometrics and spreadsheets under the Economic Brain initiative, and on parliamentary committee procedures). This is being used to consolidate some of the relationships that have proven important to getting EPI recognised and positioned to improve policies. This technical assistance may be described as ‘demand-driven’ but it should be contrasted with the use of that phrase in connection with routine capacity-building of the traditional sort. There is almost always a strong demand for the traditional type of capacity-building, focused on individual capabilities. But international thinking has long given up the assumption that the limitations of developing country bureaucracies and parliaments are to be addressed by transferring individual knowledge and skills. The practical difficulty for EPI is that the demands for technical assistance from its counterparts in government are potentially inexhaustible, and the costs of refusing to scale up this part of the work may be high. Maintaining collaborative relationships may be impossible without at least a constant level of training or advisory inputs. The danger of slipping progressively back into traditional modalities of technical assistance is particularly great at the sub-national level, since the new provinces and municipalities have weak or non-existent capacities for performing many basic governance functions. As more of EPI’s portfolio of issues and projects shifts to sub-national levels, therefore, the demand may be expected to grow. DFID itself is under pressure to respond to the new decentralised configuration of government in Nepal by supplying conventional technical assistance at all levels across the country. On the basis of global experience, any response to this demand on the expected scale will not be problem-driven and may well involve the type of uncritical promotion of ‘international best practices’ in public sector and public financial management that has blunted the capabilities of bureaucracies in many countries (Andrews and Bategeka, 2013; Bridges and Woolcock, 2017). This is not the place to discuss that possibility. However, it is certainly pertinent to urge that, whatever is decided about this issue, it should not burden EPI with any part of the response. 5.2 The new demands of sub-national programming Getting diverted into routine technical assistance is not the only challenge involved in the move to sub-national work. Some additional creativity and learning is going to be required to adjust the method that has been so successful in Kathmandu in a way that maintains its core principles. Two particular issues seem worthy of brief discussion. First, the principle that evidence and technical arguments are brought in solely to help address problems that stakeholders in government wish to solve may be harder to apply in the selected municipalities than it has been at national level. The recently elected mayors and the centrally appointed administrators who are currently supporting them may not yet have clear priorities (and the politicians and administrators may have very different concerns). Policy concepts and practices that are well established in the federal context may be surrounded by an aura of prestige that makes them hard to resist at the local level, even if they do not correspond well to local priorities and implementation capacities. To the extent that this is the case, there will be pressures to revert to top-down, solution-driven remedies to the assumed concerns of municipal leaders. A case in point is the approach to be taken in localising the national PPP guidelines. To date, EPI (and Accelerating Private and Public Investment in Infrastructure, or APPIIC, the other AiiN component that has an interest in this policy area) has turned to the PPP theme for sound reasons, seeing it as a logical implication of the municipal leaders’ expressed priority of undertaking public benefit investment projects that they cannot afford without co-funding. However, something 23 to be watched is whether, in contexts such as the PPP workshop in Pokhara in February 2018, the real and possibly quite uncomplicated needs of the municipalities in mobilising a few quick wins remain the dominant factor. The potentially diversionary factor is the natural tendency of the legal and institutional specialists in this field, as in many others, to set the bar rather high, guided by general notions of sound practice. 5.3 Engaging with firms A second challenge in applying the EPI method subnationally relates to the principle of maintaining a range of tactical options, including entry points and counterparts. EPI has been set up to relate to the principal actors in economic policy processes, including business associations and other elements of civil society but with quite a strong emphasis on the executive and legislative branches of government. This has provided an adequate range of options in rolling out the problem identification, brokering and coalition-building process that constitutes EPI’s working method. Although the new municipal and provincial governments have executive, legislative and indeed judicial functions, the range of official players is expected to be smaller. Local civil societies, even including community organisations, may be expected to be less diverse, especially outside of major cities such as Pokhara. Individual companies and private businesses may be more significant than the local branches of private sector associations. To keep alive the core of its variable tactics approach, therefore, EPI needs the ability to interact productively with firms and businesses as well as with politicians, officials and formal associations. This applies to the municipal investment projects and even more clearly to the work around SEZB and other activities in the Pokhara–Sunauli economic corridor, the success of which will hinge on the brokering of individual investment deals and agreements. EPI’s current staffing reflects the interest in building a team that has both technical expertise and the ability to establish productive relationships with, and make wise judgements about, government stakeholders. An equivalent inside track and set of sensitivities, based on lengthy direct experience, is required for effective work with firms. With the move to sub-national work, EPI should be not only setting up and staffing local offices but also boosting its high-grade capabilities for facilitating multi-stakeholder agreements in ways that make good sense from business as well as public economics and public procurement perspectives. If, as seems likely, there is scope to increase EPI’s resource envelope significantly during the remaining years to 2020, this enhancement of its private sector 5 Just possibly, they may be mitigated by the demonstration effects of solving problems linked to potentially large, tangible outcomes, as suggested by Laric and Waddell (2016). brokering capability should be a priority. In DFID, this should be accepted as a necessary implication of preserving and deepening what is special about EPI. DFID should resist the normal tendency to put its support into distinct packages according to the nature of the country counterpart organisation. A widely observed practice in donor programming is to separate public sector and private sector interventions (as well as government-facing and civil society-facing ones), awarding the implementation contracts to different organisations, with the effect that they become separate silos of activity. There are welcome precedents for moving away from this practice when it comes to investment facilitation, including in DFID’s overarching Economic Development Strategy (2017a) and in recent central initiatives, such as InvestAfrica (DFID, 2017b). 5.4 Following up EPI post-2020 As EPI reaches the middle of its scheduled term, the question naturally arises: what next? The question can be posed in several different ways, and it seems important to pose it in a way that makes realistic sense, given the agreed diagnosis of Nepal’s development prospects and challenges. EPI is proving effective as a means of navigating the dysfunctions of the policy process in Nepal and achieving policy improvements, with potentially large impacts, despite those features. It is not designed to reform the way policy is made except in relation to specific, currently prioritised, outcomes. This is realistic. The literature on improving public sector capabilities in countries at and even above Nepal’s level of per capita income suggests these are very slow, cumulative processes (Fukuyama, 2014; Levy, 2014; Andrews et al., 2017). Seldom do they occur ahead of substantial improvements in the productive capabilities of leading economic sectors. The more common pattern, including in the history of the USA, is that public sector performance improvements lag several decades behind major economic breakthroughs. Studies of the political economy of Nepal do not give us any reasons for thinking that the country is going to be exceptional in this regard (see, notably, Roy and Khan, 2017). The question, therefore, is not whether the follow-up to EPI should be seeking to ‘institutionalise’ problemdriven adaptive approaches to policy improvement. The effectiveness of the EPI method depends crucially on the fact that it brings to bear resources, and, in particular, capabilities for development entrepreneurship, that are not part of the government system or governed by its rules. The reasons why this contribution adds value arise directly from limitations of official processes that are not going to disappear any time soon.5 Therefore, the 24 pertinent question is not whether the country context after 2020 will be significantly different, so that a new and more ambitious approach can be contemplated, but whether there will continue to be plenty of economic policy gaps and blockages that are susceptible to EPI- style problem-solving. The answer to this is a clear yes, suggesting that DFID would be fully justified in investing UK resources in further efforts of the same type. Any suggestion that this reasoning is insufficiently focused on ‘transformation’ should be rejected firmly. Nepal needs economic transformation – productivityenhancing structural change and within-sector accumulation of productive capabilities. But, in the spirit of the discussion of Lin and Monga (2017) with which we began, this is most likely to be achieved not through comprehensive policy or governance reform but through appropriate concentration of effort. The most transformative results will be achieved by jump-starting employment-intensive manufacturing and successfully tackling the policy obstacles of most direct relevance to that aim, which is the EPI thinking. If a breakthrough into inclusive economic growth can be managed by triggering take-off in a few dynamic sectors, it will be possible to tackle the gradual transformation of other important areas of public life and human well-being on a much stronger basis in succeeding years. In Nepal, as in other parts of Asia (Studwell, 2013), economic and social change will bring new groups and classes into the political process, with new agendas and ambitions. The immediate task is to begin to make this happen.