Decentralization and Development: Making Local Governance Deliver

it is my great great pleasure to to have today three three excellent speakers on a topic that is of great interest the issue of local governance the issue of local governance and democracy the issue of local governments and development and what we have learnt in in in this arena and particularly if we want to focus today may be on what is it that we we can do to make local governance contribute more to both development and democracy we have three excellent speakers we have from the World Bank Ghazi Aslam and she'll be saying a bit more about herself when she speaks but apart from other things she has been working along with others at the World Bank on a major study on accountability and local governance and she'll be saying some of those things and tell us more about about that too and it's a great player to to also welcome two colleagues from Boston University both of whom have been have associations with and and friends of the party center the Left Mukherjee from the Department of Economics who has done a lot of work in the area of local governance and we'll be talking about lessons from that and John Goering who is also a fellow faculty fellow here at the party center and a professor in political science here at Boston University who has been running a whole sort of series of projects and initiatives on this issue at political science and also some with with with the party center so it's a great great pleasure to have the three of them one last one last note you might see at the back a whole bunch of our publications please do have a look and if any are of interest to you please grab a copy we have the two these two series that we've been published in the last two years and there's now a number of number of important publications that might be of interest you'll also find there this little little flyer that looks like a may see advertisement but is actually a party center advertisement the back side is more important it just announces our our next big conference which is on April the 16th and the topic of the conference is the good news from Africa and we have actually a very very good set of people coming from Africa and from all across the Europe the US and and other places outside Africa to think about various dimensions of Africa's future and this is part of a larger project called Africa 2060 so that on that little note of advertisement I'm going to hand it over to my three guests the way we are going to do it is I would ask the three of them starting with John and then maybe going to Garcia and finally the left to maybe start us off with a few remarks maybe seven eight minutes of thoughts on the topic in general and then we are going to open it up for a discussion amongst the three of them and with you essentially on what have we learned and what can we learn on grow on local governance and how it contributes to development and democracy thank you again everyone and John my colleagues on the right and left are in this area of local governance but I have been thinking about it and doing a little work that that is kind of related so I appreciate the opportunity to to share some thoughts with you in political science some of you may be aware that local governance was traditionally understood as urban politics and traditionally understood as American politics American urban politics machines and all of that it's a subfield that sort of died out and I think now we're starting to think more seriously about problems of governance at the local level in the international arena but it's still a largely unstudied area and so I guess I say that by way of justification for my own ignorance in this in this area but having said that there is enormous interest in the in the question of decentralization both in policy circles and in academic circles and a lot of kind of a priori theorizing and a lot of very messy empirics the problem is that we don't really have enough nation-states we don't have good conceptualization and measurement of the the question of decentralization we have nothing or very few instances that could be called quasi experimental so although we have some many hypotheses about how decentralization might affect the quality of governance I would say that the evidence is is pretty ambiguous pretty pretty ambivalent and even worse when you scope down to the level of local governance it's not clear whether we should conceptualize local governments as national governments writ small in which case everything we think we know about the nation-state applies to cities and municipalities and villages or whether there's something fundamentally different going on once you scope down to a very low level and of course you know I guess you could say large cities mega cities approximately are the size of many nation states larger than many nation states so if it's a sheer question of population you might have a strong analogy but it could be that something different is working in a situation where the political unit is really only semi sovereign and shares sovereignty with a national government in some cases very very clearly subordinate but what's remarkable about decentralization is the extension which both on the left and the right there's enthusiasm for it I suppose academics have thrown some cold water on that but you know during the Reagan era that Reagan Thatcher era it really they picked up a theme from the new left and that is the dhih centralizing power is good it's more democratic at least the better governance and so forth and I think that's largely what's been driving the policy agenda although perhaps now we've reached a more sober moment having said that a lot of the governance that exists at any point in time is going to be local regardless of what level of centralization and decentralization national governments work out and so it's an extremely important topic and the question that that I'll raise is about how do we improve local governments is extremely pertinent to to policy circles right now so I guess what I'm going to talk about is one part of that larger question and that is that part concern with democracy I'm presuming that the quality of local governments is going to be partly determined by the quality of democracy at the local level so the question then becomes well what is the quality of democracy at local levels and again do we measure that in the same way we measure democracy at the level of the nation-state or do different different parameters apply and although you know we complain all the time about the indicators of democracy that we have at national levels like Freedom House and polity and so forth at the local level it's pretty clear we've got virtually nothing so we don't really even have bad empirical measures of the quality of democracy at local levels and I I'm happy to learn of any indicators that are out there I haven't done a thorough survey but that's that's my sense and yet there is I guess some some anecdotal evidence to suggest that the quality of democracy at local levels is not very strong and that the degree of accountability and certainly of turnover and competitiveness is higher at national levels than at local level so I'm thinking particularly of countries like I don't know maybe Brazil or Thailand or the Philippines where local levels are often often pretty much political monopolies controlled by a clan or a family or or clique and so I think there's reason to be concerned in particular with the with the the policy injunction to decentralize power you know what is this going to mean in terms of democracy on the ground it may enhance democracy insofar as greater power greater Authority sometimes leads to greater investment as it were by citizens and elites and the quality of local governments but it may or it may not and I think this is something to be taken pretty seriously now just off the bat I guess I should I should point out the obvious and that is in a very authoritarian polity like China you know if there's an opening for democracy at local levels well that's going to probably improve the quality of governance and certainly improve the quality of democracy nationwide so my comments are addressed more at democratic and quasi democratic countries the question is why might it be difficult oh and I'll just say that that I'm kind of led to think about this subject by by dint of some work I've been doing on at the national level looking at the relationship between population and democracy and what I found very provisionally is that there's a correlation between larger states that is to say more populated countries and a more greater democracy as judged by these cross-national indicators like polity and Freedom House and that's a little counterintuitive since I think you know at least in the Western tradition we've associated democracy with direct participation with the MA the model of the city-state the polis and so forth and yet insofar as there's a story that can be teased out of these cross national indicators it could be it could be the opposite and if that also applies you know at local semi sovereign units like municipalities there may be some cause for concern and so I guess my final thoughts are about why reasons why it might be difficult to create and to sustain democracy at local levels as opposed to national levels you know just to put it into context why might it be in some sense easier for the United States at national levels that is to say and you know in Congress and the presidency and the bureaucracies associated with the national government to be more democratic than the local elected officials here in Boston or in Suffolk County or localities in the Philippines and Thailand and India and elsewhere and I thought of five things but I'm sure there are more elements of this story that one could want to talk about the first is that participation is often lower at in local elections as opposed to national elections and so insofar as participation is lower it's often more more skewed and it's easier for local elites to maintain control without without a serious opposition the second is that there is often a dearth of educated and highly trained personnel who would serve either as elected officials as bureaucrats at local levels and who would kind of sustain the the social capital that might be important for the the vibrancy of democracy at that level the third is that in some sense the homogeneity of many local areas may lead to again a dearth of viable opposition or the ability of smaller groups to contain political conflict to pay off opponents rather than leading to to turn over in political rule we generally see and not only more people but more a more heterogeneous set of people at national levels as opposed to local levels fourth the media tends to be weaker and more consolidated at local levels than at national levels so insofar as the media provides a voice for you know for accountability for opposition and so forth it's unlikely you're going to have multiple sources of news that are truly independent of whatever ruling elite there is at the local level even though you often find this at national levels and finally I think that scrutiny from the outside may be a lot lower at local levels than at national levels nowadays an election even in a small country in the developing world gains a certain amount of attention from the global media from you know I don't know regional groups and international groups donors for sure where is it local levels these things pass more or less under under the radar and so since dilip has asked us to provide some some policy recommendations I guess I'm thinking of a principle that a friend of mine our Conn Fung has called double accountability where in order to make local governance work there has to be accountability both to the grassroots some I guess there's some electoral mechanism and to a higher level of government whether it be regional or national and that higher level of government has to ensure that rules of whatever however there they are define the rules of accountability and democracy are our maintains and sustained and if necessary the the monopolies of local elites are broken up so I guess that's all I want to say but again I want to stress the preliminary nature of my own research and and and these thoughts that I've put together really very quickly this morning ok so I am here to basically discuss actually which is in the process of being finalized so hopefully this discussion today would contribute towards that just a little bit of introduction of the what kind of report it is it is it it forms the bank's knowledge base and informs the policy dialogue with external clients and also the lending policies the objective of this specific report was to provide impaired the initial objective I should say was to provide empirical illustration of the theoretical framework that was established in the report of the previous years so the you know the idea was to give very precise checklists for the task team leaders who are actually conducting these and decentralization operations in the bank or to you know or to help other stakeholders actually implement or assess decentralization reforms the other objective was to understand the social accountability mechanisms specifically in addition to public accountability what John mentions that you know there has to be accountability to the citizens so social accountability is it takes that face and you know it's a new way of doing business in the bank overall as well there are different initiatives like demand for good governance or participate rebadging these are more specific sort of interventions but we can talk about them and the in the discussion but one thing that I will actually like to focus on in my you know 7 minutes talk is the as soon as is empirical reports started coming in it became very obvious that you know the design of the decentralization reform itself does not conform to you know what well-designed decentralization system should be according to the theoretical literature and I would you know acknowledge here that there are some areas where even the theoretical literature and I would not even go into the empirical side of it it's not very clear and you know we can we can discuss that as well but I would not go into very much details right now so this is very different from what from you know from what the literature at the moment stresses and with and that is the implementation side of the reforms at how political factors affect actually implementations of the reform but you know what you find is that even while articulation even at the stage of articulation of the reforms most of the rules do not conform to what they should be from what we have learned from the theory at the moment so very briefly just to set the stage for the you know the later discussion it's the framework of decentralization that this report at Hertz toward which the last report was the last SW report advocated for it it has three as this different aspects of decentralization it has three different fiscal political and administrative and you know mostly in the literature public finance economist deal with the allocated by efficiency side of the reforms on the other hand for political scientists deal with the democracy or you know citizen participation side of the dissent of the of the decentralized systems but what this report try to do was to consolidate these three aspects and also to fight try to find the interactions between these three aspects a trying to stress that all these three aspects are important in in trying to get the kind of results that the theory actually advocates for so the and the other important characteristic of this framework is that there is a reciprocal relationship between discretion and accountability it's very intuitive but you know you would find very surprisingly as we did that in most of the countries you can find governments where they have where there are a lot of accountability measures are in place where there is not much discretion so you can't you can hold a government or an entity responsible for something if you don't give them discretion for a certain thing so these are these are the two main aspects that this framework highlights and the report actually has very detailed questions for each aspect of decentralization system which I would not discuss here because that's but we can discuss those specific aspects late you know in the discussion if you're interested right now I would just I would just very briefly highlight some of the important decisions in terms of the politics of decentralization and that came out of the report the first is of course the electoral very especially I would like to emphasize where politics the areas of decentralization system where politics can become very important and distort the results first is of course that electoral competition you know they can be restricted entry in political space or there is usually intermediation between the states and the citizen in different ways voters are sometimes obliged to make their votes public or they if voting is usually not an individual decision it's usually done on the communal level the other very important aspect is the inter government intergovernmental transfers and there you find that since local governments usually do not have capacity or discretion to hold their own - for to collect their own revenues they have to depend on intergovernmental transfers heavily to finance their activities and interestingly in most places not all but in most places these transfers are not rule-based but they are arbitrary and that leaves a lot of room for political manipulation and there are some studies that are going on some more detailed studies on some of the countries that were part of the report which actually have are coming up with empirical data which actually shows that there there's a lot of political manipulation going on it the transfers depend on a lot of on you know the affiliation of the local elected officials with the central government for example or with the other the main party of the government and the third area is of course civil service and employment policies now civil service is very it's very important for the local government to control because you know it has fiscal implications and it also has if you you should be able to control or manage your own personnel if if you if you want them to perform a certain certain task but in none of the countries in none of the study countries we found that this kind of discretion is actually available to the government not even in the courts implementation is another issue but not even in the courts they're allowed to hire their own personnel or their their even part of the recruitment or especially in performance management most of the local civil service is still under provincial or the central government but there's one very there's one very important observation which is that usually it's the intersection of the political landscape or political characteristics and the structures of decentralization of the rules that distort the situation I'll give you two examples first is that infra indirect elections in Pakistan in local governments in Pakistan the district knows at the district mayor which is a unit of the local government a is elected indirectly by the union councillors which are elected which are directly elected at a lower level it is a it is not a particularly bad form of a bad rule because then the you know the idea is that the district mayor will be accountable to the union councillors which are directly elected from the public so there is a there is a chain of accountability however in Pakistan the district mares are very influential in politics I mean most of them are related to either through families or through informal political relationships to to politicians at the center and since we had the decentralization reforms were implemented in a an authoritarian center directly to the dictator as well so what happens is that there is a reverse accountability going on the instead of Union councillor the district mayor being accountable to the Union counselors now union councillors actually try to toe district mayor's line so that you know they would be able so that they can collect some political influence that they can use for later elections so that creates really bad dynamic and then you have district mayor who was very who was very powerful as well but you know I mean the point of the lesson from this example is that structures are not bad you know sometimes they're they're not really they're not really bad form of government but it's just that when they intersect with the political landscape that's when it becomes very problematic the other example that I would give you is from Ethiopia the partisanship of local elections I mean there there is no literature to basically tell us if you know if you should have parties at the local level or not it's very unclear at the moment but in a topi as it is it is local elections are party based so however the problem is that that's I mean that's only that's a dominant single party so they control the electoral space that is why partisanship of local elections is criticized on the other hand in Pakistan local elections are nonpartisan but you know that characteristics of the local government is criticized because you know the idea is that as the authoritarian center is trying to diffuse the political forces in the country and that is why it's that is why I mean the local elections are non party based so now we mean depending on the political context the rule becomes very important so the very you know the one lesson from these observations is that you have to be very careful of the political context before you before you design decentralization reforms and you know that's that's that we need to take a step back from actually observing the implementation of the reforms and the other important lesson to understand is that it does not mean the distorted effect of the decentralization reforms do not mean that we do not implement these these reforms in the countries with these kind of political contexts where we know that implementation is going to be affected that only means that you need to have a different structure for for political context for different kinds of political context and you can actually design decentralized you know those kind of reforms which would ameliorate those distorted effects because in a lot of debates this does come up that decentralization you know in certain political context decentralization just does not work so maybe that's not the right form of government for example in Pakistan that that debate comes up again and again every few years but but this is so I think that you know what during this debate this is a point to understand so I guess I am I mean that's that's what I would talk about the other things in whew if you have any questions about forms of government like specific rules or social accountability especially that would be more than happy to discuss that Peter thank you okay so let me let me actually given the limited time focus on what I called the pitfalls or the seven deadly seas of decentralization but let me let me preface my remarks by saying that I think the word even thinking about decentralization is heterogeneity it the heterogeneity of experience with decentralization across countries and even within countries is something that one ought to be very sensitive to and understanding the dimensions of that heterogeneity makes it a very complex exercise then I think this I'm completely in agreement with with the remarks made earlier and so what I'll be saying is you can see is an elaboration of some of some of those points but let me mention that you know decentralization has created huge promise and excitement because you know it's been ongoing in Asia our Latin America and Africa since the 1980s and the excitement and the promise is in reversing decades of top-down over centralized development patterns and the main advantages are that it's going to create a development pattern which is more responsive to local need it's going to enhance accountability and reduce corruption and government and it's going to enlarge democratic goals of empowering citizens and enhancing the legitimacy of the state and in fact many contexts you have seen very impressive results just give some examples Porto Alegre in Brazil has this participative budgeting system where something like eight to twelve percent of city residents turnout neighborhood residents turnout and and discuss priorities for the budget and those are aggregated up to the city level and it has been associated with substantial improvements in road sanitation and primary Caroline India 40% of all plan expenditure in the state is now devolved these local governments and local citizen councils are deciding on development priorities in Bolivia a substantial reduction in inequality between nine districts on the one hand and 300 other districts the remainder of the country it's not Africa the dismantling of apartheid this was an essential ingredient in improving social services Indonesia as well after the collapse of Suharto and yet come back to the seven at the substantial heterogeneity and the seven deadly seas so very quickly the first the first pitfall is captured on corruption okay and John referred to local elites and the problem is that there may be distortions in the functioning of local democracy and I can give you lots of examples but what it means is that the score for El for local elite capture depends on the local context particularly the extent of inequality and ownership of assets of land in a rural context education and there are numerous studies now which document this so just to give you an example Argentinian school decentralization there's a big study in the Journal of public economics last year which shows that the schools in the in the more backward poorer areas actually became worse after decentralization but the ones in the better areas improved and you know the simple explanation is that you know the whole process is a potential potentially liable to capture by by local elites before number two is capacity lack of administrative and fiscal capacity of local governments okay and that has also been referred to pitfall number three is coordination into jurisdictional externalities do not get internalized in decision-making and of course for this reason a lot of programs have to remain at at a higher level some education and health programs for instance of construction major construction infrastructure projects and so on cannot be fully devolved to the local level and there is also the problem of the soft budget constraint local governments with limited capacity to raise their own resources do rely overwhelmingly on average almost everywhere more than 80% of funds for local government are devolved from higher levels and this gives rise to all the game playing between lower level governments and upper level and lack of responsibility on the one hand of local local governments to fiscal concerns because they passed the buck and lack of sensitivity of upper level governments that are reluctant to devolve money downwards and related pitfall number four is cash often various unfunded mandates in in I think Ronald Reagan's time part of the whole design was to transfer responsibility for social programs to local governments but you know in local governments have less capacity and cash to to disburse those responsibilities so that's one way of shrinking the welfare state but this has become a massive problem all over the world bit 4 number 5 is what I call clientelism or callousness so you often have politically motivated transfers across jurisdictions higher level governments use local government local governments to undermine sort of potential political opposition particularly true in Pakistan but also in in many other countries you find this paradoxical situation where military dictatorships at the high level or the political dictatorships actually are more favorable to local democracy and so with the revival of democracy in Pakistan for instance in fact we had predicted and it happened that local governments were going to lose a lot of their authority in the last year also there's often tendency for high-level governments to use local governments as instruments of client holistic development of their their own power but for number 6 is and these are some of the political the practical issues in implementation there's often a lack of consistency between fact and rhetoric between de facto and de jure a decentralization constitutional authority is often not provided to local governments again true in Uganda Pakistan and China where it's not a democracy at the national level these governments can be dismissed at will by the national government there's often confusion between what the responsibilities are of different level of government Ugandan health is an excellent example of that the experience with decentralization on health immunization and Uganda has been unanimously negative it's partly because of confusion about who is responsible for what there are practical delays in implementation there's failure to oversee local elections she all Rosanna just referred to a political parties not being allowed to contest which in some context also undermines genuine political competition and also not allowing local governments to hire and fire appointed officials and then the seventh pitfall which is irk you can think of either as sort of comprehension or consumer sovereignty local residents even if you have a well-functioning local democracy local residents may not sufficiently appreciate the importance of certain investments for instance long term investments for instance in child health and this was another issue in Uganda the local mayor's were populist and basically the local population did not want money to be spent that much on public health and so it has a tendency to go in for more short-term income enhancing programs so conclusion what are the policy implications first you know it's no magic bullet it's no panacea for development problems there are many pitfalls and the pitfalls are both a combination of fundamental pitfalls that are inherent in underdeveloped countries problems of elite capture non functioning local democracy problems of coordination and problems of capacity but also you have the political problems that was also mentioned earlier and so quite often you have decentralization programs and carried out to satisfy external aid givers or try to appease disgruntled elements within the country but then an attempt to sort of disguise retention of power at the state or at the national level and therefore decentralize the other thing is keep in mind is that decentralization may give rise to increasing disparities across regions and also increasing inequality within regions because of these problems of capture and capacity are quite often more in more backward regions and also prosperous regions are not as willing to subsidize lagging areas so quite often in the centralised regime you had greater redistribution across regions and therefore net I think all of this suggests the importance of necessary safeguards and monitoring of decentralization the process of decentralization itself has to be subject to central supervision and there has to be genuine devolution of what the Indian local government Minister calls the three F's funds functions and functionaries outcomes have to be monitored in terms of actual allocations and how they are targeted to to the poor the Argentine's study suggests that the central government should retain the option of taking over failing schools so that that is another way that one can supplant some of the weaknesses of decentralization and then finally I think what John mentioned the role of the information and media citizen oversight as well as auditing from above all of these institutions are typically quite weak and definitely need to be strengthened thank you thank you all three I'll just just open up for questions in a minute but I think thanks for for three very very interesting presentations though I must confess I was more on the pitfalls here than on the on the promise and maybe we can we can we can encourage you to also talk about that I was particularly fascinated by the seven seas except the seas were about twelve the points were seven seas a fascinating little you know c is for cookie and that's good enough for me and but but she is if as if you if you look at the academic literature how many lists of seats it is quite amazing I have contributed a few myself but but I did want to before we open a scause here maybe just to list because I think you may not have what countries did your study cover African and South Africa mostly South Asia it was a few I'm probably not be able to remember all of them but some Kerala in India than Punjab in Pakistan is Sofia angola burkina faso Guinea and Uganda so these are some of them so before they open up if I can ask all three and I think the loop date did answer this already but you might want to add more but most also to the others is the message we should take from what you are saying that be careful about decentralization because it may not work or is the message that be very careful about how you design it because otherwise it may not reach its promise I mean is there actually danger in decentralization that you are highlighting or is it that the promise is not automatically achieved it's a bit of both I think it's I would like to add also that you know this is a long term experiment if you like if you want to think of it as an experiment but it's once it started it's it's very difficult to reverse but quite often the capacity-building and the growth the growth of local democracy is something that takes time and so we've seen in some instances in India for instance one-third of all local mayor positions have been reserved for women and the studies are showing that the first two times around you know it didn't have much of an impact and in fact particularly with respect to the attitudes of the citizens with respect to women political candidates but the third time around you know the attitudes are changing so in the long run you know it may end up being very different from some of the problems which may be of a short-run nature so I think we have to also keep that in mind yeah you know I would I like to think of this in terms of a theoretical distinction between sovereignty and delegation of power in a federal state there is a constitutional sovereignty is shared between the center and the periphery the sub-national units States territories lender a Regents whatever and I think that leads to a lot of rigidities and kind of dysfunctional behavior because it's locked in and what serves at one time may not serve it at a later time but if power is delegated in a flexible manner and and subject to revision and it for 'ti is always maintained at the center I guess that's the model that I would I would suggest is long term more workable now that doesn't maybe that doesn't say very much but at least conceptually for me it's an important distinction you know so so Sweden raises a huge amount of its revenue at regional levels but it's a central state it can alter revenue raising and spending allocation at any time whereas in the United States and most federal federal countries there is a really a distinction between taxes that are raised at national levels and at regional levels and you know for example in this country we it would be very difficult to establish a consumption tax which most people think would be a great idea but you know the Constitution won't let it so this is just an example of the kind of rigidities that I think can be overcome and so I guess I would say that and this echoes a point that Dylan dilip made at the outset because there's so much heterogeneity in the capacity of local units to really absorb and effectively wield power in a democrat and accountable and effective way there's no recipe but as long as the center maintains authority and is flexible in terms of what's working and what's not I think you know that's that's as far towards the generalization as I suppose I could go I think there are two different issues here the first is that you know it's sort of actually knowing if it is effective or not given you know in different forms of decentralization and I think there is not there is a lot of theoretical literature that sort of you know at tribe you know convinces you that yes decentralization because principal-agent theory or you know in other in different disciplines that decentralization is supposed to work but and then you know it added reinforces inequality in regions as well but there there is not a lot of empirical evidence there is just I mean you know there are there are theoretical explanations for it but you don't have empirical evidence that is why in in this report this we don't use the word effective because we don't know if it is effective or not it's well-designed and you know that's that's something that needs to be kept in mind empirical estimation is difficult because there's just not a very good decentralized system available at the moment so that's that's one thing to keep in mind when you know when you're even designing decentralization reforms that the second thing is that there you know I mean it in most areas you do know that these certain rules at least theoretically would results in these would give rise to these kind of results so so you know the problem is actually getting to getting to actual device to get those rules device and get implemented in those way in exactly that way so one of the things that I mean this report or what I really think that came out of this you know in pay is that as I said that you that they can be the decentralization reforms can be designed in a way that would ameliorate some of the problems when if you know what the good rules are you can you can change the structure like the two examples that I gave you can change the structure so you can't I mean yes there are but you know there are problems but it doesn't mean that they cannot be it doesn't mean that decentralization should not be used given you know what theoretical explanations we have okay okay let's let's open it up by a whole host of interesting ideas here but yes Duplo please also introduce yourself there is a mic there it will do nothing to amplify the sound but you will be recorded for posterity my name is avi I'm from Indonesia and probably as you know that Indonesia recently implemented since realization in 2001 I think mostly is to fulfill IMF conditionality and World Bank's program as well and because it happens so quick the result is I think chaotic because the purpose of the Central Station I think to improve local surface delivery but instead the region's perceive it as right to resources and more power so my question is that which one should go first the central a station or democracy or which is both or is it a precondition of democracy yeah anyway I think that this your comment actually you know sort of highlights a bigger problem which is that you know that when you know even if you know what the in for example in Indonesia there are a lot of different rules that you could change to even you know to get to a well-designed stage but there are political motivations not to do that so so you know this there there is a big if question of sequencing that we don't that you know that is not that we don't address at least I mean there is there I don't think that there's any conclusion to that again that you know so it should political decentralization follow before fiscal or you know administration should happen before political so it what you get from the what you get from the literature at the moment is that all of them has to go at the same time but it's system politically impossible at the moment and there are a lot of different countries who are having the same experience that you basically have like half-baked I can answer in general you know given the the importance of you know containing local elite capture and successful local democracy is important for that so in general in the long run they should go hand in hand but in the context of Indonesia I had two students who wrote their thesis on the decentralization Indonesia and one of the chapters was precisely on the topic that you raised it so happened that in Indonesia the replacement of the national government appointed mayor by locally elected mayor's was was at a time when the nationally appointed mayor was due to retire and there was some exogenous variation in that based on that age and so on and the administrative devolution happened according to a timetable so it was a very nice natural experiment where you could look at the distinction between administrative decentralization and political decentralization what did they find they found that political decentralization was associated relative to administrative decentralization with a decline in child health basically because the local governments once the mayor was popularly elected started spending less on public health and they didn't find any significant effect or anything else so again that's the caution that was my seventh see the consumer sovereignty from again but you can't generalize just based on that that's Indonesia that's what happened then okay and one should be careful about drawing any general influences let me take the next question but if I can throw rider to that in case anyone if you might want to take that later does it matter why decentralization is happening whether it's happening because of an external donor push or because of a because democracy in the country has so taken roots that there is a internal grassroot demand for decentralization or because there's a dictator who wants to create a new political constituency and so on and so forth that does is it simply sequencing or does the the trigger matter but but I let me get you Liz's question too and you can combine I'm from Kenya ah maybe I'll discover bit about Kenyan experience with decentralization for example we have seen several iterations of these we have the usual local governments that you have talked about and they have performed badly over the years and they have been have all that central government capture then back some few years back we had a system where the members of parliament decided to allocate part of the central budget to the members of polymer themselves so that each member of parliament had some money in his own constituency and that actually changed the dynamics of the country because for once given the nature decentralized nature and the ethnic politics that play around you could always blame the government if you're ineffective but for once now MPs had that leads to show what they are doing on the ground and this proved to be very very effective actually impede started having competitions at local level they could not talk of central government because they had some money so this was quite effective and it was actually introduced by one MP I don't know he I don't know I got the idea from but it was very very very effective right now following the last clashes that we had in Kenya for the election there has been a push for a new constitution and now we are going full-scale federalism kind of we are going to create I think like this 50 district or so and you're going to have local governments which I elected with their own Parliament something similar to this and that kind of federalism has always been on the in politics in Kenya since independence and it's usually associated with a why to be in a kitchen called my GMO which to the common man the understanding is that if we have our local government and I see some wood with a foreign and a foreign I see some wood grain come from there I can take their business so there has always been that tension that people understanding of that kind of decentralization is that you cannot have a shop if you are not from here and and therefore is a danger that as Kenya moves ahead it may is political tensions or it may entrenched I don't know what my question is like what should Kenya be advice why should we look for examples given the fact that man designing is that how you do it is very very context based so other people's experience I don't know whether we help or not so just your thoughts on Kenya goes forward with its new constitution yeah yeah well I guess I can give the response to which you're now accustomed probably from the three of us that the the there's a long debate over whether federalizing power whether decentralizing it is going to sort of a sway ethnic conflicts political conflicts more generally or whether it will serve to entrench differences and perhaps even lead to greater conflict over issues of secession struggle for resources and so forth you know again it seems like you can you can find a lot of examples on different different sides of that and so I guess I would be reluctant to say anything about the prognosis for for Kenya since then since I'm not an expert on in that area but maybe one thing I'll say is that sometimes conflicts between different groups reach a certain point that no other solution is possible I don't know if Kenya is quite there yet but certainly you know when thinks of of Iraq you know a degree of decentralization of power was necessary in order for for the state to survive not include the state will survive and and and it could be that other countries are in that circumstance I know India's often remarked as an example where the centralizing power was sort of a precondition for survival I don't know if Dilip has a view on that no I don't think it was necessary for survival I think it was a political strategy of you know Rajiv Gandhi who was Prime Minister in the in the late 80s as a way of contending with the declining fortunes of the Congress at the national level and dealing with the rise of various other political parties at the state level and this was a way to circumvent you know the authority of state governments and open a direct line of client holistic transfers to local levels federal solution you know and having different ethnic groups but dealing with border states quite often so sometimes you have central states you know which have a pressure-cooker aspect to it and it's particularly difficult with respect to border states and giving rise to secessionist threats and violent movements and sometimes the only solution is exactly to devolve substantially to those to those areas I think what you're raising is a very very difficult and central issue and I think particularly in Africa but it's also arising in in Iraq and Afghanistan one point of view is that you know when when the state is itself nascent and there is this potential for ethnic conflict that undermines the state perhaps you know a strong central authority is plays an important role in you know legitimizing state institutions example is given for instance in Tanzania of the role that near area played and in fact the linguistic homogeneity that you know was brought about partly to by forced means binaries sort of seen with the benefit of hindsight as something that did succeed with some benefit but on the other hand you also have the argument that the problem in Iraq and Afghanistan is that the you know governance is extremely corrupt and you need some kind of competition so Roger Myerson for instance was Nobel laureate and economics two or three years ago he passionately believes and he has been actively involved in trying to advise civil society groups in Pakistan that encouraging local government is absolutely essential to create a new generation of leaders from below who will challenge you know established people with a substantial monopoly on power so the debate goes on last summer I was at a conference at Columbia with Joe Stiglitz had organized and we had a debate precisely on this issue and it went on for three hours and people arguing on both sides can you tell us where they ended each one stuck to this actually may be logical or you can if I could push on behalf of Julius where it has worked understanding the heterogeneity what might be the conditions that make it work when it works at the moment I mean they're also decentralization after the ethnic conflict has worked has worked pretty well at least in in terms of certain indicators although you know there some reports have come up there's still trouble but you know and that's basic that's basically trying to understand and their social accountability mechanisms are very strong and that's they have used traditional mechanisms to actually you know create social accountability and even sometimes you know the public accountability mechanisms as well so so you know I mean that but trying to understand why that works and why we're what are the circumstances or the local contexts where it works and where it doesn't I I don't think that that's very flattered but I would like to point out the motivations or incentives that you have that's that's actually you know that's one project which is on the table at the moment it's trying to figure out how the structure of decentralization system is different compared like when the motivations are different when you have decentralization and authoritarian regimes or you know due to because of ethnic conflict in other context you cannot have you don't have the central because of ethnic conflicts so you know trying to figure out how decentralization reforms a structure the design of decentralization is reform is there any systematic correlation across these exogenous factors and the structures so that's it's still being formulated at the moment one and two and three well my name is hassan 't and before coming to this talk I did a little bit of homework on this centralization but I was only aware that first web development economist is sitting in front of me but I was not aware that a scholar from Indonesia is sitting behind me so I must be careful making any comment I will not enter into because of time limitations any intellectual discourse of decentralization let me finish intothe sentence and that will have ratified the ideals question what is the issue right everyone talked about in terms of two Gaggia slightly went over there but didn't hit on to the head there are three a decentralization of function and B decentralization of power what did missing here is the decentralization of budget let me give two concrete example volatility I live in Indonesia right Philippine decentralization function but didn't give money it int work Indonesia from Abdul Rahman or heat to Susheela like because of a new democracy or something like that I plan to believe that because of the sincerity of the leadership himself of themselves right they decentralized both power and budget and so despite the criticism that World Bank and IMF influenced my reading is that decentralization in Indonesia is much more successful in comparison to decentralization in Philippines I may be wrong because of my limited study and third one is let me give the example of my experience from Bangladesh grams for Koror panchayat whatever you say during the assad government introduced it and eventually the next comment came and they abolished it it is not because that the rural government is good or bad it is because the previous comment did it and they do not like that people and the government so I must now this government again reintroduced Graham shocker but there is another very important piece unless you go to the ground you will not find the real crux here they have decentralized the budget that means they have given money to list it and union authority but thing is that they have all same colors to buy the leadership of those who are handling the fund to buy the conditioned car from Japan in a duty-free manner so money went to hotel but it didn't go to the people go to the leadership and the two other cases Mali didn't in Philippines money didn't go at all and in Indonesia money power and functional Sun Google went simultaneously so these are the three different areas right with you can I take a couple more and we can thank you Pablo Suarez fellow here the party center and also working with humanitarian organizations I was very intrigued by some of the things mentioned in terms of how if there is existing inequality across jurisdictions before the decentralization process these inequalities can be accessor baited and then there is also the issue of the mismatch both in terms of geographic scale and of time scale between the problem and what the new authorities after is under this intercession can or want to do I'm working a lot with climate risk management and I'm really really concerned that in addition to all the good things that may come from this interrelation we may be creating little time bombs everywhere where there's people at most severe risk I'm thinking in particular of a village in Senegal where it was in December simple story big city facing more and more floods so to solve their flood problem they create a little canal that alleviates their problem but making an enormous problem for a village nearby that now gets flooded from water that they never saw before and this is like two ridiculous levels and no one thought how can we help these people who suffer one of these interjurisdictional externalities if we know that even without a change in landscape without a change in climate we were changing technology and so on there would be incentives for those who are better off to take advantage of this neutralization become stronger by hiring the Argentinian teachers who are good but somewhere else and offering better salaries and so on aren't we at risk of making the most vulnerable even more vulnerable with the centralization and if so is there anything that has been done from the word bank or other entities that promote the centralization to ensure that there's some kind of mechanism that takes care of those who are likely to get the losing end thank you I'll try to be very brief my name is lil Adams I'm a graduate student at the IR department in my thesis I'm looking at community development councils in Afghanistan set up through a World Bank program through their Ministry of development and while it is a program through the central government it's focused on local governments at the village level at the very grassroots level where villagers are elected to these Development councils and one thing that has shown up in my thesis is that there's a difference between local government and local governance a lot of the villagers don't make an association between these councils and their government but they are effective local governance structures for rural development so I was wondering if you could address that anyone on any of these so we have three questions well I was just gonna say I certainly don't know what's going on on the ground in Afghanistan but I do know that the largest I think the largest randomized field experiment to date is occurring in Afghanistan and I do believe it's a bank funded project and perhaps you you know lots about it or perhaps that's the basis for your for your work and I just think more generally that local government is an area where we can plausibly intervene and you know in a in a productive way to find out more about these various questions that we're talking about it's much closer to you know we used to use the term in American politics laboratories of democracy only we you know no one really thought in terms of experiments back then but I do think we have an opportunity to learn a lot about about local governance and maybe if if funders like the bank continued to do this kind of work and if host governments are amount Abul to it we will be able to answer these these sorts of questions some of them with with a greater degree of certainty with respect to your comment you know i'm recalling i think i'm correctly recalling you know the the old was it 1967 work by a Musgrove on on public finance and I think one of the lessons from that is that where you can internalize costs and benefits it makes sense to decentralize and where you can't it makes sense to centralize or at least have some coordinating body that's you know that's an old lesson it's nothing nothing new but I think it goes a long way towards making decisions about what policy arenas it might make sense to to decentralize power towards and which ones that maybe ought to be retained at a central level the three different aspects that you mentioned so yes this is what I mean the framework that decentralization framework is three different aspects as political administrative is a fiscal and yes it is important for all of them to go hand in hand in hand but the problem is that when you're actually implementing these reforms and you're you know when you're suggesting that to the government there are a lot of different things they're just political it's not politically impossible so yes that question is very well taken should administrative decentralization happen before a fiscal I mean if they are not ready to devolve fiscal responsibility to the local government should this dough go ahead with a political a decentralization or not or you know if they're not ready to evolve towards civil service employment you know that kind of authority to a local government is there any you know reason to decentralize is it going to create any benefits there is really I mean I am not aware of any very effective you know studies at the moment there are a lot of different arguments for or against it but I am not sure if it has been studied systematically so you asked your point is well-taken that it has to fiscal you you I mean even if even invent after you give fiscal discretion to the local governments it has to be given in a certain way otherwise it creates a lot of political distortions but you know there is a but it decentralization is hard because it has to be done by the central government it is the reorganisation of the power off the central government so you know either is or there is a lot of reluctance so you know that's it doesn't answer your question but it sort of elaborates on the problem itself the second thing about the vulnerability issue that is why as I mentioned the very beginning that there is a lot of emphasis on social accountability and the demand for good governance I which is actually coming from citizen participation it's I mean there are some there are some part of the there are some policies that have to be part of it you know most projects that go to the lending projects that have that go through so you know social safeguards but again the problem is that when you actually introduce these kind of issues in the in the in the design then it creates opportunities for the central government to manipulate so these are this is a trade-off but yes it has been it you know there are there are efforts being made of course that's a very you know that that is a that is a big problem and yes according to the theory it would make sense to decentralize exactly for the kind of service exactly for that area where you know you can internalize the spell hours but that's just not possible you say there will always be those kind of trade-offs practically go in sequence decentralization of finances obviously is an important part but I don't see it as absolutely critical so within India for instance Karnataka and Kerala two states have devolved financially to a large extent which other states in India have not West Bengal for instance close to Bangladesh deva has devolved very little in terms of finances about 4% of West Bengal local government budgets are coming from upper level grants but the West Bengal governments have been very intrepid and and creative in using all kinds of alternative means of raising finances so it turns out that while 4% is directly funded through upper level grants they supplement that to the order of about twelve to fourteen percent through community you know phishing schemes auctioning off equipment and so on it's just amazing so about three times as much resources as they're getting from upper level governments local governments are creative enough to raise three times as much okay so there is scope for them to supplement we also mentioned there is one danger of giving too much authority to local governments to raise finances locally and that is again if the local government is captured by local elites they are going to tend to use tax base as local tax bases that are highly regressive and it's particularly apart from property taxes which is usually subject to a lot of corruption in terms of the valuation of properties by and large when local governments have financing capacity it's through indirect taxes and particularly if you look at local government the history of local government in the United States a large part for the motive for devolution in the early part of the 20th century was for white dominated districts to make sure that their own taxes were not going to pay for black schools okay and there's all the evidence there are studies showing that most of taxes that were raised were essentially being borne by the poorest people in those residents because they were essentially of an excise tax nature and the burden was falling on poor workers so you know again all over the world when you I mean all the countries that I have looked at across all continents it is almost always the case that 80% of the funds are coming from higher-level governments but still you see enormous variation of performance between the high performing and the low performing States so it's it's obviously something that helps but it's not absolutely critical and I think the example of Community Development councils in Afghanistan I think the question that has been raised earlier as well the distinction between local government and local governance do you really need political you know democracy of political devolution or administrative devolution itself and can achieve quite a bit and in fact quite often they are not you know they can be in completely different in nature from one another classic study of irrigation bureaucracies by Robert Wade a political scientist in India and Korea Korea was not a democracy but it actually had a very sort of devolved structure of maintenance of of canals whereas in India which is sort of formally democratic and yet it was a very top-down and very corrupt system and similarly China is not really hard nearly as democratic as India's but in China local governments have played an instrumental role in economic development of China in the form of the township and village enterprises in the 1980s and nowhere else in the world have local governments played such an important role in economic development we have one time for one very short question but since you had your hand up I will give the last question to you let me let me add to that and then give you the final word each of you so that question if I can add to that I've been a little struck by some of the countries we are using as example so Rwanda once the Iraq Pakistan when you have qualities where there is already so much broken to then run this experiment and I totally take your point this is a long range experiment and we like everything long range in this building but if that is if those are our cases it seems that sort of that's not a fair fair experiment because you are taking countries where just about everything is messed up and then saying now local government and local governance why don't you perform so so any thoughts on that and the question what does this mean for the u.s. both in the past and the present first or any other final thoughts you might have yeah so I I used I was an American is once upon a time so perhaps I can try to answer this question I always yeah I guess hated the decentralized way in which the United States is structured and I could go on it at great lengths I suppose about that but but maybe the thing to point out with the u.s. is how much less decentralized it is than it appears to be we have a Federal Constitution we have a Tenth Amendment we have a Supreme Court we have you know the Constitution says each state shall be represented by two senators and the Senate is clearly an important equal policymaking body so the upper house sort of in sconces the principles of federalism and yet in some ways we have a relatively a relatively centralized polity I mean if you look at the revenues raised if you look at the way in which the federal government is able to control policies that you wouldn't think it would be able to like I was just chatting with a friend about education policy and the Obama administration has a lot of reforms that they're trying to push and they're doing it with what would seem like a miniscule amount of money I mean the Department of Education is providing do you want this money if so you know we want you to do X Y & Z and and I guess the picture is the states are just knuckling under though oh well whatever we'll do anything for your money you know even if it's a small pittance as a proportion of our education budget most of which is raised locally so you know and this is the long-term trend over the past century and I think there are several things contributing to it one is obviously the ability the federal government to raise revenue and the other is the fact that we have 50 states and no really really huge states I mean in the sense that you know nothing like the disproportion between large and small states that you find in Canada or Brazil or Mexico or whatever and and the sheer number of states and and the fact that they mirror the political fragmentation that we have at the federal level because it's the same basically the same constitutional form stamped on each on each unit means that it's harder for them to coordinate and to you know that oh yeah they have they have meetings you know among the governor's but what does that really accomplish it's nothing like the meters and meetings of the Premier's in Canada and so in some ways we we don't experience as many of the pathologies of federalism as as other countries and of course we don't have that the depth of problems that many other countries have we have this you know perhaps because of our extreme heterogeneity in the population we have and because of the success of our nation state economically militarily there's a very strong allegiance to the state and that overcomes whatever for superest tendencies there might be regionally any last thoughts as you give Phillip again the last word well your question is provocative and I think it brings back the discussion that Julius initiated here so whether in countries like Rwanda and Afghanistan and so on whether it's fair I would pose it that way it's not a question of fairness or you know but I guess the question that you're raising is decentralization something that something that should wait until a certain amount of State Building has already taken place and I am sympathetic to that point of view but you know there are development challenges and I guess the appropriate question is you know at what stage of development is decentralization more appropriate but there are those as I mentioned who are arguing that even in the context of Afghanistan decentralization is going to play a very important role because part of the problem is the legitimacy of state institutions and decentralization can be an important way of legitimizing institutions and promoting social development quite often a problem that you face in in in early stages of development is an overemphasis on sort of more physical economic aspects of development at the expense of social goals and I think by and large you do see decentralized governance is kind of an important instrument of development of social goals so from that standpoint if you think that you know social goals are really important to legitimacy of the state and you know include in that demand for some kind of clean up of the system and reduce corruption it could potentially be it could potentially be an important instrument even at that stage thank you on that happy note and that is a happy note yeah I think that is both a fair and a happy nor defy and large it has the potential but on that happy note let me thank all three of our guests for a wonderful discussion

Channel: Boston University

Views: 9,570

Rating: 4.7377048 out of 5

Keywords: pardee center, governance, world bank, public policy, John Gerring, Dilip Mookherjee, Ghazia Aslam, Nadil Najam

Id: P0NsYY2nK_A

Length: 78min 38sec (4718 seconds)

Published: Tue Apr 13 2010