What does Raghuram Rajan’s exit tell us about the Modi sarkar?

In the last couple of months, we have witnessed the spectacle of the intellectual hero that is the Bharatiya Janata Party’s Rajya Sabha MP, Subramanian Swamy, take on and vanquish the intellectual Raghuram Rajan, the governor of the Reserve Bank of India. While this will go down as yet another feather in Swamy’s giant-killing cap, for Rajan, this should be an occasion to reflect on the vagaries of public policymaking in India.

If you stop taking sides, you would easily see that what Swamy pulled off is quite spectacular. He took on the head of an aspiring economy’s central bank, a man with impeccable academic credentials and a stellar record in the economics profession (having got more predictions right than wrong). Also, critically, Rajan enjoyed that all-important factor in the fate of economies these days—positive perception among the pundits. Add to that the man’s charm which prompted the coinage of terms such as ‘dosanomics’ by an adoring media.

Put together, Rajan was intellectually sharp, charming and outspoken. As Shashi Tharoor learnt earlier, that is a fatal combination in India, especially in electoral politics. This episode tells us that whether in electoral politics, or outside it, you cannot pull off wearing a personality that makes you stand out in a crowd.

A mere mortal may have been deterred by all of this—but not Swamy. His attacks were sharp, measured and went largely unchallenged except for a perfunctory word here and there from Prime Minister Narendra Modi and Finance Minister Jaitley (not dissimilar to those that have been uttered before in response to, say, attacks on minorities). Swamy’s Twitter-fandom cheered him on, and in the end, when victory was his, he issued pithy statements that fully reflected the enormity of his success. The Art of War could not have scripted this battle better. The outcome suits the worthies in the Modi Sarkar as well, and it tells us that with the right tools, you can get your work done without sullying your own hands.

Also read: India will miss Raghuram Rajan as teacher more than as banker

So, where does this leave our economy now? There are real fears that with a job as unfinished as it is with our economy—for instance, the push towards cleaning up the balance sheets of public sector banks—this development could see the economy slide. It is important to remember that almost every international analyst who has stoked the story of ‘India being a bright spot in the global gloom’ has also pointed out the dangers of the spiralling non-performing assets.

As evident from Rajan’s quest to clean up our banking system, this is not only a technical challenge, but also deeply political one. One can only hope that the next RBI governor will be equal to the task, but Rajan’s fate ensures that his successor would tread extremely cautiously on this matter. This outcome suits all the status quo-ists—among them, several key players in the Indian economy, including the Modi Sarkar.

Also read: The lessons from the Raghuram Rajan affair

Finally, op-eds from those sympathetic to the government and Swamy’s efforts have been quick to seize on Rajan’s outspokenness as the cause for this exit. In their view, the RBI governor and other such key positions must be an ally to the government, and stay within their brief. At best, this would give us Arvind Subramanian (chief economic aadvisor), and at worst, Pahlaj Nihlani (Central Board of Film Certification chairman). Rajan on the other hand, is being accused of not being part of the euphoria over the now-once-again-shining India, and instead, making cautionary statements such as “In the land of the blind, the one-eyed man is king”—or for expressing his views on political corruption and ‘intolerance’. When civil society and individual artists can be vilified the way they often are, what hope did the RBI governor–whose salary is paid by the government of India–have?

Also read: Swamy hails Raghuram Rajan’s exit as liberation for India economy

India is a country ravaged by a never-ending stream of institutional failures. Our Parliament does not function, law and order operates according to political whims, the top judiciary seems to veer out of control, and so on. The list is depressingly long and endemic to our daily lives. So, we take our heroes where we can find them. The Election Commission remains a redeeming feature of our democracy, although political parties seem to have figured out how to stay within the letter of the law while violating its spirit with impunity. The Reserve Bank of India, largely due to its independent-thinking governor, has been another shining light–but one that the government now has seen fit to dim.

To come back to my original question: after everything that has passed in the last two years, what does Raghuram Rajan’s exit tell us about the Modi Sarkar? Nothing new at all!

Reading ICAI’s review of DFID WASH results

The Independent Commission for Aid Impact (ICAI), UK’s aid watch dog, today, released its review of DFID’s programming and results in water sanitation and hygiene (WASH). – In this impact review, they take a close look at the results DFID reported in its 2015 Annual Report; results that cost £ 713 million between 2010 – 2014.

Do read the full report here.

Some thoughts on the areas of concern in the report:

  • The focus on ‘leaving no one behind’ is spot on. It is easy to stack up impressive WASH numbers if one ignores the poorest and the most vulnerable in communities. Safe sanitation and hygiene need to be universal for health benefits to accrue to communities. Within WASH, sanitation is specifically complex, sometimes also called a ‘wicked problem’ – a challenge foremost, of inducing lasting behaviour change. The very nature of careful social engineering required to bring about this behaviour change seems to run contrary to some of the factors that make an intervention scalable – an ability to standardise inputs and break programme components down to easily replicable bits.
  • Within the broad basket of ‘service delivery’ interventions, WASH is one of the trickier sectors when it comes to measuring sustained impact, especially at scale. Naturally then, ICAI find that while DFID’s claims of having reached 62.9 million people are broadly correct, it is very hard to establish if the benefits are sustained. Therefore, the results reported remain at the ‘output’ level and that is what ICAI ends up assessing, even though what they set out to do is an ‘impact’ review. While the report speculates on sustaining benefits beyond the 2011-15 period, I wonder whether those that accessed the programme in 2011-12 continued to experience any benefits in 2015.
  • The link with government systems, in terms of implementation, monitoring and sustenance remains unclear: another typical WASH issue. Barring say, India, (and this is true especially in sub-Saharan countries, government WASH budgets are highly inadequate. A lot of the work that happens is funded by donors and this implies that monitoring and maintenance happens outside the official system. Achieving local ownership in such a context is a challenge.
  • ICAI finds it difficult to assess value for money (VfM) in DFID’s WASH programmes. On one hand, it finds that there isn’t enough competitive procurement, but also there is a lack of established metrics and benchmarks to analyse VfM. Following DFID’s own 3Es framework, an Economy and Efficiency analysis should be possile across the portfolio, and as far as I can tell, is rapidly being developed in the sector, and within DFID. However, partly as a consequence of the lack of ‘outcome/impact’ data, cost-effectiveness studies are likely to remain a challenge. This work by an OPM-led consortium should be particularly relevant in improving VfM analysis across the sector

How could this review have been better?

  • We could do with a better feel for DFID’s global portfolio – what the variationa in context-specific experiences, challenges and results are. But probably fair to concede that these boundaries to scope must have been considered when the methodology was being finalised.
  • Use administrative data: The review states that several DFID programmes contributed towards larger national programmes, or attempted to strengthen national/local data management systems. It would be great to hear more about experiences from different countries – with varying state capacities – on the extent to which it was feasible to use administrative data as an effective monitoring tool.
  • While in absolute terms, 62.9 million is an impressive figure, we are looking at it against the backdrop of a staggering 2.4 billion who lack access to reliable WASH services. In the countries where DFID made significant investments, was there significant progress made towards reaching the MDG targets? Yes, the MDGs were output-based too, but it would be great to get a sense of the scale of impact achieved at the national-level.
  • Finally, on learning, the report is critical of DFID, saying that not enough is being done to share lessons across programmes, or from related sectors such as health and education. But surely, there are lessons programmes have learnt over the five years (2010-15)? Further, in the absence of good impact data, was there at least good quality process data? Have we learnt how to implement a WASH programme better?

As I have written previously, ICAI reports are a great addition to the sector, both in terms their scrutiny function of one of the biggest donors in play, as well as because of the questions they ask. So do keep an eye on this space.

Does the BJP need to rein in Narendra Modi?

Since May 2014, the Bharatiya Janata Party (BJP) government has been running in fits and starts. Some stalwarts keep themselves busy spinning tales and lawyerly arguments to defend the government and defame the opposition, quite forgetting that it is this reliance on moralistic rhetoric that partly led to the fall of the United Progressive Alliance-II government. The misadventures with the land acquisition bill, corporate taxation policies, pension funds and most recently, Uttarakhand, betray a mix of incompetence and arrogance.

At the same time, even as a handful of senior and junior ministers—Sushma Swaraj, Nitin Gadkari, Jayant Sinha, Suresh Prabhu and Piyush Goyal—continue to impress, every piece of public communication by the government is curated to carefully credit only Narendra Modi for these successes. Needless to say, the failures have been glossed over and, one can be sure, will be airbrushed from the annals of history by a set of complicit cheerleaders in the media and the new political establishment.

But several recent events suggest that it is important to ask: does the BJP need to rein in Modi?

Take the recent #PoMoneModi fallout. This was a familiar script. The principal actor was a familiar one. As Modi belted out yet another aggressive election speech, this time attempting to shame the voters of Kerala by comparing the state’s human development indicators with that of Somalia, there was a feeling of déjà vu.

Yes, Somalia is that insignificant country where our prime minister is unlikely to make an unscheduled stopover. It is that country which will never host a grand Modi stage-show. And Kerala is that state where the BJP will struggle to establish a significant political footprint for a few more years. Therefore, Modi’s speech-writers must have decided that none of this really matters. But what it did do was make both the prime minister and his party an object of ridicule everywhere.

Moving away from Kerala, one only needs to be reminded of the manner in which Modi seemed to auction—in an election campaign speech—Bihar’s poverty in exchange for a notional central package. The vocabulary during the Delhi assembly polls prior to that was vicious as well, with the prime minister himself cautioning voters against thebazaru media. All of this has now become the new normal—the bombastic speeches and the personal attacks— and one wonders how this affects the ‘cooperative federalism’ that Modi so loves to talk about.

It is one thing for states to go to elections and witness bitter contests. It is an entirely different matter when the prime minister of the country turns up and converts state elections into a race of competitive populism, or worse, lowers the discourse further by indulging in name-calling and polarization. Of course, he has had able assistance from an excellent supporting cast. And this is not helping, as Delhi, Bihar and Kerala show.

Let’s quickly look back at 2013-14. The UPA government’s handling of Parliament, the economy, the national auditor, investigating agencies and even our police and internal security forces had left us with institutions that we could not trust. As he started his Lok Sabha campaign, Modi promised change, and an escape route from all that was rotten with the previous regime.

A careful analysis of his track record in Gujarat would have revealed every single problem we now see with the central government: the tendency to centralize power; over-reliance on bureaucrats; misuse of the police and investigation agencies; penchant for showmanship over deliberation; and systematic suppression of opposition—both political and non-political.

The wider political implications have been serious. For one, the BJP has settled upon a strategy of keeping minority communities and student campuses in a state of simmering tension. When it comes to the Congress, corruption cases are another part of the strategy—those that are juicier as allegations rather than conclusive investigations. This ensures that political opponents (and more significantly, non-BJP state governments) continue to engage with the BJP purely in retaliatory terms. This, too, is a strategy that has a clear Modi imprint, one that was perfected over a decade of political machinations in Gujarat.

And this is now eating away at the core of the BJP. The galaxy of superstar chief ministers that the BJP had just over two years ago has been dismantled and replaced with a set of pliant politicians who are too eager to please the prime minister. The personality cult that Modi has built around himself and the substitution of cerebral foreign policy with event management has resulted in utter fiascos with Nepal and Pakistan. Analysts have pointed out how Indian interests have suffered in trade deals with the US and the Dassault deal in France due to the over-eagerness of the prime minister to claim quick glory.

In sum, we often obsess over the statements made by “fringe elements” within the BJP and its spiritual parent, the Rashtriya Swayamsevak Sangh. Arguably, many of those instances tend to blow over quite quickly, not in the least because of the frequency with which they come, as well as their bizarreness quotient. But when the prime minister starts making too many gaffes, it is a completely different matter.

There is no disputing that Narendra Modi led the BJP to power and remains its biggest vote-catcher. That his stature is unparalleled within the BJP is not in any doubt. But do not be mistaken by the obeisance that his party workers and Internet fans shower upon him. The lessons from setbacks at home and abroad are quite stark and should worry serious strategists in the BJP. Is there any chance of a course correction? Let’s wait and see!


This is my latest livemint column

The logframe in an ‘iterative & adaptive’ world

Often, the stark reality facing the seekers-of-the-brave-new-PDIA-world is this:

But the Donor needs a Logframe! 

This would bee case, even in programmes that are committed to adopting the problem-driven-iterative-adaptation (PDIA) approach. Let’s assume that high-level commitments from the donor is in place. Even so, where the rubber meets the road, a logframe is an item in the checklist of requirements for those funds to flow.

Well, fear no more, and watch this Matt Andrews video

Broadly, this video explains how you can adapt your logframe to the PDIA-frame. Matt fittingly calls it a search-frame, reflecting a spirit of iterative learning as we move closer to our goal.

In an earlier post, writing about the ‘results agenda’, I had this to say about logframes:

Have logframes become the blunt hammer in our pursuit of results? Probably. I personally don’t mind logframes as a concept, but detest the way they are usually used. Ideally, a logframe development exercise should reveal the level of complexity of a project design. Those working on it realise during the process that the outcomes that really matter are often misfits in a logframe, because there aren’t clear baselines, nor are there objective and accurate measurement techniques. If a project logframe is then drawn up primarily to satisfy a contractual obligation, it usually sacrifices this complexity and settles for outputs and outcomes that can be easily counted. Donors must resist the temptation to do this.

That may have been all gloom and despair, but I have of course not given up on the logframe. Instead, working with different agencies – donors upstream, as well as implementers downstream – have suggested ways in which we might work a way out of this problem. Some of these are quite simple, and appear obvious. Yet, they require a relationship of trust between different actors in the implementation chain. Here are a few initial ideas:

  1. Start from the beginning: Ensure that you have a sight of your results framework as you go about designing your intervention. Remember, there is no escaping measurement – every successive iteration of the programme is based on lessons learnt from the previous one. So we are no longer in a world where a programme is fully designed, and then we draw up a results framework/logframe that is ‘fit for purpose’.
  2. Rethink indicators: Outputs in a PDIA programme are ‘resolved problems’ – for each of the deconstructed small problems. While in a standard logframe, each output has multiple indicators that get scored on the basis of whether the targets have been achieved or not, and the output then gets scored on the basis of the number of indicators that are met. I am suggesting we change the way this works – and have indicators represent the multiple pathways, establishing from the very beginning that not all of these pathways would work. Meaning that as project implementation proceeds, targets for some of the indicators would be met, while some others would not. But we now score outputs on the basis of how far we have moved in terms of tackling our ‘problem’, and not based on the score that each indicator receives.
  3. Capture progress on processes:  Most logframes ignore processes – they concern themselves with only outputs and above. But at the risk of increasing the amount of work involved, there is a case for bringing process indicators into the logframe. Some of these would look like a check-box on activities to be carried out; yet others might focus on key steps that are of importance to the process of change the intervention is attempting – say, a joint meeting of parties involved, the production of a joint manifesto, and so on…
  4. Logical revision: Matt’s video basically focuses on this: have a clear window and a mechanism for logframe revision at periodic intervals. This is often already the practice, but it is probably fair to say that the overall formulation of the problem (and thereby, the outcomes and outputs designated) remain broadly untouched in this process. (At this point, watch Matt’s video again) It would be great progress if we start tinkering with that, even at a small scale, and accept at the beginning that a programme’s logframe at the end will look nothing like how it did at the beginning. It is important that the rationale for these revisions are honest, and captured accurately. With iterations to your approach as a result of evidence gathered on each ‘bet’, change is inevitable. If the logframe has to keep up with the reality on the ground, this change needs to be captured and represented systematically.
  5. Empower your team: Make it clear to everyone from the very beginning that data collection and analysis is not the sole responsibility of an M&E officer. Monitoring progress, and results is essentially project management. Therefore, each workstream should have its own decentralised data management system that not only gathers data, but participates in generating lessons from the programme, as well as in iterating subsequent versions of the logframe.

Finally, to borrow a line from the PDIA-crew: All of this might sound like common sense. If so, it might be great to see this all happening in practice too!

Are we being too quick to judge Liberia’s ‘partnership schools’ pilot?

Liberia has provoked serious outrage from some quarters with its decision to enter into an agreement with private providers to run its primary schools. An official release on Liberia’s Ministry of Education website sets out the problem, saying that

42 percent of primary age children remain out of school. And most of those who are enrolled are simply not receiving the quality of education they deserve and need”.

One must commend George Kronnisanyon Werner, the Minister of Education for focusing his attention on what is really critical – not only are many children out of school, but even the ones that are part of the system are not receiving quality education. Children enrolled in Liberian schools are not learning, and when asked about the state of the education sector, are quite critical.

Meanwhile, the United Nations Special Rapporteur on the right to education, Kishore Singh is scathing in his criticism of the Liberian government’s decision. He asserted that

“Provision of public education of good quality is a core function of the State. Abandoning this to the commercial benefit of a private company constitutes a gross violation of the right to education,”, and “it is ironic that Liberia does not have resources to meet its core obligations to provide a free primary education to every child, but it can find huge sums of money to subcontract a private company to do so on its behalf,”.

This criticism is worrying on two counts – the unyielding stance on engagement of private providers in schooling systems; and more importantly, the statement seems to limit the definition of the state’s implementation capacity to its availability of financial resources. More funds in the hands of governments (or for that matter, donors and NGOs) do not automatically translate into better outcomes.

This is particularly true of primary education. Over the last decade, governments and donors have poured in enormous sums of money into education systems, into securing higher levels of access for children in poor countries. In the process, globally, we have taken our eye off quality and learning outcomes. Year after year now, globally recognised learning surveys tell us that the vast majority of children might be in school, but they aren’t learning. This is true in Liberia’s case too. The debate over Liberia’s policy decision fits right into the global debate over the role of private providers, and to do justice to the enormous education challenge facing us, the least we should insist on, is a nuanced discussion.

In recent years, low-cost private schools have become an important feature on the policy agenda to promote education for all. These schools are not free, but they are low-cost in comparison to the standard understanding of what ‘private’ schools charge. Even so, the $5 – 7 per month that is charged can be quite a burden on the pockets of poor families. Others have countered this by pointing out that education expenditures make up a significant proportion of what families with children normally spend (ranking just behind food and health). Therefore, private schools are springing up, especially in rural and peri-urban areas of low- and middle-income countries. Parents appear to be voting with their feet and private school enrolment is increasing in absolute numbers and as a percentage of gross enrolment figures in countries such as India, Pakistan, Kenya.

Evidence suggests that private providers have a role to play. They deliver learning outcomes that are as good, or slightly better than in public schools at lower per-capita costs, according to evaluations in India and Kenya. Of course, the absolute levels of learning in both types of schools seem abysmally poor, so there is really very little to choose from there. And the costs are kept low by paying teachers lower, which is yet another hot potato. For one, teacher absenteeism and time on task in public schools is very poor in comparison to private schools. More importantly, there is little evidence to show that having higher paid, or qualified permanent civil service teachers yield results in terms of learning outcomes for children.

Other independent studies have found that management systems that are able to hold teachers to account make a huge difference, and many private providers follow models that are more accountable than public systems.  But this debate is far from over. We often complain about the lack of sufficient evidence-informed policymaking—but it is also true that policymakers do not always have the best evidence possible at a given decision-point, where best means both rigorously collected and analysed as well as relevant to the context. This is where the ‘partnership schools’ experiment will provide Liberia with an excellent opportunity – to generate evidence, and make their own decisions based on clear, transparent decision criteria. The official statement says as much: “an independent body will be commissioned to evaluate the outcomes of the pilot program.

Many critics are up in arms over the fact that Liberia is partnering with Bridge International Academies (BIA), an American firm that has investors such as Mark Zuckerberg, Bill Gates, and the World Bank Group’s International Finance Corporation (IFC). Engaging a single provider in such a large experiment is risky. But, at this point, it is not even clear that is the case. Liberia says they will run a competitive procurement process to identify partners, so one would hope that there would be multiple partners in the mix, and an opportunity to test out more than one approach. The eco-system of private providers is a large one. There are the likes of BIA, which are large multinational chains, and then, there are the single-entrepreneur schools, community/NGO schools, church-run schools, etc. One could argue that local private players are better rooted in the communities they work in and therefore, are more responsive and accountable. But these are assumptions that this pilot should help test out.

When Liberia decides to initiate a pilot of up to 120 schools (3% of the all public schools in the country), it is not abdicating its responsibility towards its children. By starting this pilot, the Liberian government is neither off the hook for improving quality of the remaining public schools in the country, nor can it give up its oversight of private schools that would function under its watch. The criticism of Liberia on the lines that it is “outsourcing its entire education system” just isn’t right. As a country in the middle of a post-conflict recovery process, Liberia needed to do something different to revive its education sector. This is a bold gamble, and promises to be an interesting public policy experiment. On this occasion, we must not be too quick to judge Liberia.


This is also my livemint column

The ‘decentralisation agenda’ must succeed

Duncan Green’s blog hosted a post by LSE’s Jean-Paul Faguet titled: Is Decentralisation good for Development. Faguet has edited a book by the same name that you can find here. This is a subject very close to my heart, and I believe in decentralisation as a value, just as I believe in democracy. It is often a work in progress, but it is a project worth persisting with, an ideal worth pursuing. Faguet’s research (at least, my interpretation of his work) therefore, really speaks to me. In this post, he makes several interesting and compelling points. For instance:

On the advantage of competitive politics generated by decentralised systems:

Imagine you live in a centralized country, a hurricane is coming, and the government is inept. To whom can you turn? No one – you’re sunk. In a decentralized country, ineptitude in regional government implies nothing about the ability of local government. And even if both regional and local governments are inept, central government is independently constituted, probably run by a different party, and may be able to help. Indeed, the very fact of multiple government levels in a democracy generates a competitive dynamic in which candidates and parties use the far greater number of platforms to outdo each other by showing competence, and project themselves hierarchically upwards.  In a centralized system, by contrast, there is only really one – very big – prize, and not much of a training ground on which to prepare.

Highlighting the importance of careful design:

The genius of a properly designed decentralization system is that it combines technical expertise from above with local time-and-place information from below, in a way that is superior to what either level of government could achieve on its own.

And finally, challenging the decentralisation-sceptic’s belief in democracy:

I have heard colleagues declare their allegiance to both democracy and the centralized state, and I just don’t get it. Citizens must be allowed to vote… but only for national government?  They must not be allowed local governments? Local services with few externalities – like local policing or primary education – whose effects are overwhelmingly concentrated on the residents of a locality… must be provided by a distant central government?

Decentralisation is an ideal, but it needs to be carefully designed. For instance, a local government without funds, functions and functionaries (the three in equal measure), is of no use to citizens. I have been fortunate that in India (and my home state Kerala in particular), decentralised governance is not an abstract idea. It is the reality. In recent years – in spite of all of my own critiques, and that of experts in the field – the central government in India has taken concrete measures that have the potential to deepen democracy, by yielding more power to federal states. Federal states on the other hand, have a mixed record in strengthening local governments, but even then, there is a promising momentum towards a truly decentralised governance system. (Yes, I am generalising, and there are several specific critiques; also important to acknowledge that the pace of reforms is uneven and several cases suggest even a reversal). But what is way forward?

In recent years, many doubts have been raised on the merits of decentralisation itself – on its ability to deliver better quality of public services, on its willingness to raise resources and on its ability to improve local accountability in general. In an old post, I had countered four popular reasons that are often quoted for the impending collapse of the local governance system in India at the lowest levels – the village councils. This is a short recap:

  1. Local governments are corrupt: You hear this all the time, and everywhere. As fund flows to local governments increased, these allegations gained in strength. There is, of course, some truth to this, but this is akin to claiming that the government should not implement any programmes because politicians and bureaucrats are corrupt. Through mechanisms such as social audits local-level corruption is easy to uncover and transparency has proven to be a powerful tool in the hands of communities.
  2. Elite capture of local governments mean that they are therefore not accountable to citizens: Local governments are often headed by elites, much in the same way as our legislators are usually the rich and powerful. The competing pressures of politics and the prospect of re-elections is a strong accountability measure that can reduce the chances of elite capture. Local government elections usually see very high turn-outs.
  3. Local governments lack the skills to draft development plans: This is an accusation thrown in a context where little is done to train elected representatives at the local levels regarding their roles and responsibilities. In many government programmes, local governments are are burdened with not only responsibilities of needs assessment and resource planning, but also monitoring implementation. In the absence of systematic training in the mechanisms of local resource appraisal, it is likely that plans will look more like wish-lists.
  4. Not all development projects can be planned at the village level: This is a classic misconception. The typical error here is that when one thinks of local governments, they only think of village-level, without considering the tiers above them that form the entire complement of local governance structures. Also, the federal states and central governments are supposed to, at their levels, take up projects that require specialised technical expertise, and cover large geographies, and multiple constituencies.

The problem with some of the critics of decentralisation is that they try to look at the work of sub-national/local governments in isolation. Either they express frustrations that these local governments have been unable to transform service delivery, or they reinforce their scepticism of local leaders by pointing to instances of corruption or inaction.

Ultimately, local governments need an enabling environment, where they work collaboratively with bureaucrats in the line departments and service utilities. Without this, and in an environment where local governments have to continually struggle for powers and legitimacy, expectations that they would get work done are highly misplaced. In many developing countries, the ‘decentralisation agenda’ is struggling. But it must succeed.