Elite Capture and Political Economy | Parallel Session VIII | Path2Dev 2023

This panel discussion was held on Dec 20, 2023 as Parallel Session VIII during the Pathways to Development Conference 2023. #Path2Dev is a multidisciplinary conference that brings together experts from within and outside Pakistan to present the latest development research on Pakistan.

Paper titles and presenter names are listed below:

– Unpacking Rural-Urban Clientelist Networks by Mahvish Shami (London School of Economics)

– Demanding the Last Mile: Foreign Aid and Political Participation in Pakistan by Syeda Shahbano Ijaz (Occidental College)

-Spillovers in State Capacity Building: Evidence from the Digitization of Land Records in Pakistan by Shan Aman Rana (University of Virginia)

Constructing Electricity as Entitlement—Energy Politics in Lahore, Pakistan
Erum Haider (The College of Wooster)

To learn more about #Path2Dev, please visit www.ideasdev.org/conference

Okay good morning everyone thank you so much for uh joining our panel we have four uh super papers uh in front of us um my name is AAL Naki I will be the chair and moderator of this panel which means you have to listen to me very little and I

Will just be sitting there and giving you a 15 minute uh well 15 minutes is what we start with but I’ll give you a five minute marker and a one minute marker to guide you to the end and we’ll have five minutes of discussion after each paper open to uh Q&A uh following

That so with that we can start with the first paper right so oh this is really loud um my name is mai shami uh and the paper I’m presenting today is a co-authored piece with had Majid um and I’d like to start by saying I was a

Student here and it’s always been a dream of mine to present at s Sego so I was very excited when I saw where I was presenting so just to give a little bit of background um the paper looks at clientalism and um I’m assuming a lot of

People in the crowd are familiar with clientalism it’s essentially how in the presence of a dysfunctional State you have people approaching um people of resources people of power to get access to resources and that could be simple things like getting a job getting a house getting um getting access to food

Or even more complicated things like getting access to public goods that the state is not providing to you um and this is a you know it’s a it’s a very vastly studied field which has been going on for a very long time um but it tends to focus on Urban or rural areas

So our paper sort of is one of the second there’s one other paper out there that uses uh actual household level data to compare urban and rural uh outcomes in uh in terms of clientalism so the question we’re essentially investigating in this paper is how do who do citizens believe give them access

To public goods and there are two public goods we look at uh one is your classic public goods of drainage and paved streets which in Pakistan are sort of your very basic public goods uh particularly in villages that that Villages demand and the second is dispute resolution so as most of you

Must be aware it’s very difficult to access the police so in that case how do you how do you get justice as an average citizen um and we’re focusing on household perception of who is providing it to them right so it’s not on who actually provided it but who do they

Believe gave them access to these Goods um so you know I’ve sort of covered most of this but historically clientalism is seen as this face-to-face exchange relationship between groups of unequals and the sort of catchphrase hair was groups of unequals and it was seen as an exploitative relationship

Where the landlord would exploit The Peasants for as much as he could get out of them and you know the peasant had nothing else and no other option so therefore he was stuck with going along with this relationship and as I said most of this tended to or so the

Literature started off in the rural context with you had your landlords and your peasants and your surfs and now how that relationship evolved um but in recent decades you know we’ve come to realize actually that sort of style model does not hold anymore these networks are actually very adaptive and

They’ve evolved based on the environment in which they’re based um so you know as as poverty has migrated to the city so has clientalism we’ve come to realize that now the urban poor also engage in these relationships because you know they don’t have access to State resources and the extractive powers of

These landlords are actually curtailed by the presence of exit options um and clients have have much more bargaining power that was than was initially sort of stated in these uh early models um but there’s a lot of fluidity in how these networks function they they’re not

Homogeneous it’s not this one model of a broker or a patron at the top and your clients and presents at the bottom it’s actually the the level of bargaining power varies considerably and part of it is affected by the geography which is what we argue in urban areas there there

Are much more fluid as you would expect being in a big city as opposed to in in villages which tend to be much more isolated right um so building on that what we did was we looked at um uh slums in Pakistan and Villages and we chose Pakistan largely also because of its

Checkered history with democracy which makes Democratic institutions fairly weak and the Reliance on patronage politics has has sort of persisted in its entire history uh till today um and we look at lore uh slums in lore so just around around here um and hsad which is essentially the rice producing area of

Of Pakistan and it’s it’s a fairly rural District um the biggest city is half Saad City and when I went to visit it uh I had to tell the driver no no we’re going to the city and he’s like yes yes this is the city this is not the village

So it’s it’s a very rural District um so in terms of the methodology we had 2 by two variation in both the urban areas and the and the and the rural ones in slums uh we looked at both registered and unregistered slums registered slums are those which the

State has acknowledged as being legal settlements and therefore they cannot be abolished unregistered slums continue to be illegal squatter colonies and then they vary based on their location so half of our slums are categorized as being in the center of laor and half of them are in the periphery right um

Similarly in rural areas we also have a 2 x two variation half of our Villages are dominated by um large landlords which you would call sort of monopolist landlords who control all the land and all the resources so what what sort of historically you would think of as your

Landlord based Villages under the British and then half of them are what we term egalitarian Villages that does not mean that there are no landlords it’s just that they’re multiple landlords so in an average Village there’d be about 15 to 20 landlords which means there’s competition amongst them for clientlist following so that

Clearly changes the nature of that relationship moreover half of our Villages are on the motorway Link Road so they’re extremely connected to hsad city but also to other cities um but even if peasants are not traveling outside of the village what they do is they bring traffic to the city to the to

The Village so there’s sort of thorough Fair going through the village and these villagers are now very um used to Outsiders and they can use these as opportunities for themselves so you know they can set up little tea shops they can set up car repair shops and stuff

Like that so there’s sort of more opportunities outside of the landlord in these Villages and half of them are far enough that you would classify them as being isolated so when we visited these isolated Villages all these um you know some of the the the students who went

With me are sitting here they can they can vouch for this kids would come to sort of look at us because they had not seen Outsiders in ages and who were these people coming in their fancy joggers and their funny clothes uh asking all these questions whereas in

The connected Villages nobody cared um so moving on so just looking at basic public goods provisioning across these Villages as you as one would expect uh urban areas receive Urban slums receive much higher public goods provisioning than than rural areas and this is both um drains and paved streets right but

The question we were interested was not so much on the level of provisioning but who was providing it so we would ask that you know who was instrumental in getting you these public goods and the answer was very interesting um by far the majority in villages would name the

Local landlord or their local Patron or broker whatever you want to call it so it was their clientlist Network that was giving them access to public goods from the state not the politician um and when you when you went to uh how do I oh there when you went

Went to an isolated Village dominated by a large landlord this number increased significantly more because this is where the landlord had complete control and he was essentially the owner of this Village and he dictated what went on so here almost everybody believed that he was the person giving you access to

State provision and I’d like to point out that this was these were also The Villages where you had the least amount of provisioning right um in slums it was a very very different story in slums politicians actively come and claim uh credit for provisioning lore is a very important political Capital politicians

Do not want to take the risk of not getting credit for provisioning but also um the nature of calist networks are very different in slums than they are in villages and this is what the paper sort of brings out much more in villages you have a very stable wot block system you

Have a very stable clientlist Network even in egalitarian Villages which allows um Brokers to act as Gatekeepers that is not the case uh in the slums in slums it’s much more fluid these relationships are constantly renegotiated uh slum residents are constantly changing networks some of them only join it during electoral

Cycles so it’s a very different type of clientlist network due to which politicians do not uh sort of secede um decision- making to to local patrons they they take credit they they send their someone from their office or they come themselves to make sure that people know

That they provided this drain and they provideed this streat and of course this was all done around electoral Cycles um so yeah exactly so in slums also what you find is these uh clientlist networks tend to have much lower levels of enrollment as opposed to in The Villages so just to give you

Numbers in The Villages about almost close to 90% of households were part of a clientlist network whereas in slums it went to about 50 Max and again when you sort of dig deeper you realize that some of them were only joining during electoral Cycles not all year round all

Sort of uh yeah all year round so the question becomes what do Urban Brokers do right the only way these networks survive is if they provide something people are joining it to get something out of it and what is that because it’s clearly not public goods which is sort

Of the big good that uh or sort of drains and streets which is the big Goods pakistanis tend to to to ask for so what is it that uh he is providing them uh and the other good we looked at was dispute resolution so as I said most

Of you are here are Pakistani so you’d be aware that um the police is problematic getting access to justice is is a big problem in Pakistan particularly for poor people so how do they gain access to Justice um and what we found was that this was the big

Service uh Urban Brokers were providing to their clients they would assist them with dispute resolution either through their informal Network so their informal dispute resolution or by going along with them to the police to provide extra support to ensure the minimizing of harassment by the police itself um and

This service as you can imagine was provided predominantly to their own clients so this is the big thing that a a broker would provide and this was the main reason Urban uh residents would join these clientless Networks Works um and this was very very interesting to us because this is very

Different from what we found in rural households and it all sort of comes down to a function of exit options in slums you have greater exit options these these slums tend to be more fluid people are coming and going uh which also impacts how these networks function

Which means that um slum residents have more options not just from outside the slum but also in terms of the Brokers they can have similarly an egalitarian Villages because there were multiple landlords it meant that there were multiple clientlist networks you could pick from which then increased your

Bargaining power if one broker did not provide you what you what you needed you would just you would switch networks that was not an option in your landlord dominated Villages due to which there were higher levels of exploitation also in slums you are more in you’re closer to the state you’re

Closer to the politician politicians in rural areas very seldom went down to the The Village they would just contract with the broker because the broker would guarantee you votes right um so increasing exit options then essentially reduces the hold of this broker and allows citizens to get more out of uh

Out of their network but also you know it is heartening to see that when you weaken these clientlist networks and you water them down we are seeing higher levels of provisioning so think back to the graph I showed you slums do receive much higher levels of public goods provisioning which could potentially be

A function of these watered down networks um but I have to sort of um throw a sort of caution to the wind because this provision I should point out is by no means programmatic this is very targeted provisioning that comes close to an electoral cycle and they are

All aware of this they very sort of openly say you know oh you know my trade is breaking down but don’t worry an election is coming very soon so it’s going to get fixed so and it didn’t bother them that it was it was uh lowquality targeted blatant vote

Capturing but they were like you know at least we’re getting something out of it um but it’s a step in the right direction so it’s it’s something to think about um how we can sort of wean off clientlist networks and get the state to provide more to these to these

Peasants thank you very much oh there we go I’m gonna come down stand you can stand there am I allow to okay so so thanks a lot um I have the hard task of giving comments on such an amazing paper um I have to admit that I have promised Mish that I’m going

To send in more detailed comments after fully reading the paper but I was able to go through bits and pieces so please free disposal um the paper is very fascinating it gives a lot of detail trying to understand how you know net these kind of networks and how people

Believe that they’re getting public goods and from whom are they getting public goods and how that differs by the environment in which these networks are located right and so a few things that I’m just wondering and I don’t know if that’s even like maybe it’s beyond the

Scope of your study so I think what I was wondering was since you did like these primary you know you did primary data collection you went and you had surveys I think it would be so useful for the paper to have detailed descriptives on who these people are so

You have people who are connected to Brokers you have people who are not connected to Brokers and then there are all these other dimensions of heterogenity I think it would add so much richness to the T to the paper if you could give us like more details

About you know who these people are what their backgrounds are Etc um so one thing I was wondering about with the urban rural you know the bar graph you showed about public goods and you know the fact that there is more public good provision in urban locations

Is what I understood is that by per capita or you know it would be useful to see if that’s by per capita or not because once it’s you know normalized you’ll probably see a difference right you’re going to have more in the rural and less in the urban then um I mean I’m

Going to leave out the econometrics part because but I don’t think this is not an econometrics paper it’s beautifully descriptive and I it’s fascinating so then we come to the mechanisms for my understanding these are stated beliefs of people about who is giving who are those set of people who are giving them

Public goods and who are their main sort of Brokers with the state or you know people who can provide them access to the state and its services so are there um given the large literature and this is probably beyond the scope of study but given the large lit that’s out there

On you know biased beliefs Etc it seems to me like what you’re capturing is that first stage that beliefs get changed but is it really the case that people were getting these public goods from these actors that they thought that they were getting them from or they wanted to get

From so that final linkage of that belief translating into a reality can also be useful maybe you know a in a complimentary paper then the last thing I want to say is about so you have these and my understanding was that the communities were stratified based on registered unregistered egalitarian and

And landlord dominated um I mean of course these are the most natural I would have done the same thing it’s these are the most natural sort of Divisions I think given that you’re making a comparison between the two groups right so rural urban is your comparison um do you think you know

Maybe we can have some base which is common between the two right so you’re making a comparison but then you stratified communities on on different dimensions so in the in the urban it’s more State registering and you know State sort of accepting that the slum exists whereas in the rural it’s more

Like landlord versus not maybe that also is a state acceptance you know in some way I don’t know but I think that common there has to be a commonality and then you do a comparison and then we can see oh these things are very different but because we sampled from different you

Know different communities maybe what we’re saying is just a result of that but overall a phenomenal paper thank you I look forward to reading it and I’ll provide you detailed comments as promised thank you of course am I I thought I’m the third speaker I’m you sorry okay happy

To go next if you so uh from the audience we can take a round of questions uh there’s uh SAR over there thank you um thank you so much Mavish this is really really fascinating and perhaps there’s more detail on this in the paper but I had a question about

Measurement of networks and kind of how you identify whether an individual is in the network is that based on simply knowing the person or is there some more intensive bar for um being uh considered part of a network um and then the second question is about dispute resolution

Which I found really fascinating that that is something that brokers in the urban context are doing and I wondered what that looks like in the rural context and in part is it also the case that um residents in your rural sample are turning to informal dispute resolution and whether that is sort of

Driving the difference in the function that Brokers are are playing right so in in urban areas there’s more of um this need for intermediaries because there’s engagement with formal dispute resolution thank yes of fasinating work and um you know to kind of answer one of the questions the that Sean brought up about

Kind of completing the link so that’s the question I had because I find in my work that if you ask people and this is looking at local politicians so not looking at Brokers but very um uh counselor level politicians and if you ask them do you think that this good

Will be provided to you whether you were to vote for them or not and they they always respond that we will get this good during electoral Cycles irrespective of whether we actually bit our vote for them or not which kind of makes it in in in my view it’s a

Different kind of clientalism in the sense that the transaction is not complete so they people know that we may or may not capture the vote but they’re still willing to provide at that time so I just wanted to I I in my work kind of frame it as this is potentially leading

To some form of accountability and so I wanted to hear your thoughts on that um and the other thing just what are who are the politicians that you are looking at particularly in the rural setting so are those local level politicians or are you looking at provincial because you

Said it’s the landlords who are doing most of the provisioning um and finally when you say that there is non-programmatic provision in the urban slums is that because there getting this dispute resolution good and not the programmatic sort of paved streets and all of those which I would

Think are more programmatic Goods is is that why you’re saying it’s non-programmatic provisioning or is it because it’s being done during the Electoral Cycles thank you so much one more question over there and then come back so thanks a lot I learned already a lot by listening um so uh for me the

Story makes sense with like the different es in exploit or extraction and you’ve provided this evidence that there’s more public good provision right that’s one side of the coin and the other is like how much do the like um yeah the people have to provide to the

Broker like what is kind of what do the how much do the Brokers get out is there some evidence on that there’s the differential and kind of yeah how much is they exploited in a way let’s just take one last question it’s fascinating paper My Wish uh one

Thing you didn’t talk about and I’m curious about is the difference between connected and isolated rural areas because uh are the connected Villages closer to the urban results or not because we know that in Punjab something like 80% 80% of the population lives in area which are very

Well connected and does so that would be interesting to thank you over to you Mish okay great thank you um sorry let me just write this down so I don’t forget it um let me start with Sean’s comments um yes so descriptive statistics is it was a very

Tight space so it’s a lot that had to be fit in with your question about the common base that’s an easy one to answer these were two very separate data sets connected completely independently of each other and there was never a plan to put them together it was only recently

We read uh a paper by Adam orbec from American University like oh we have rural Nal data too so we could have a look uh so this is sort of a first step um and then but that’s a good point to flag and I think we’ll raise that in the

Paper that you know these vers independent data sets collected for different purposes so there it sort of lends itself to collect more data in the future with the intention of having commonality um the part about stated belief is it’s a great point that you know people have biased beliefs and

That’s fine but this is more a question of you know who do you think is getting you the good who are you who’s getting credit for it and in rural areas they don’t it’s not the politician they’re very clear our Patron went to the politician got it for us whereas here

The politician got it for us the the broker in the middle didn’t do anything so that’s sort of the the way it’s uh phrased um so they are wrong it is the state given it and it is a belief and that’s why it says it’s a function of

What households believe um and you might be right it might be that a as more awareness is spreading when you’re in an urban area you’re you’re more aware of it but the politician is coming down himself also to make sure people know it is him that’s providing it so he’s also

Dispelling uh false beliefs um I never thought of doing it per capita but I I should point out that the village sizes are very similar to the slum size sizes so I would think it would still be much higher um and it is it’s I think it’s

Because the the sort of starting ground in slums is higher so they they have to provide more to to keep building on it whereas The Villages had nothing um is that everything yeah I think that’s about most of what you’ve um with the Sara with measuring um cental networks

Uh there was a series of questions we asked we start out by asking who is your H who’s the influential your household is aligned with uh and what’s the basis of his authority why do you go to him but then throughout the survey at different points we would ask oh so you

Have this problem who do you go for it oh so your land got taken away who did you go you had a dispute problem who do you go to and then we sort of check to make sure it’s the same person and that allowed us to create so that’s when you

Have a network when you go to multiple to this same person for multiple problems um and then so dispute so I have a paper that I actually presented here last time uh which was on dispute resolution in rural areas so dist resolution it’s not a rural or Urban

Problem it’s a Pakistani problem right it’s even well off people avoid going to the police uh because it it results in harassment and you have to pay bribes and all sorts of other problems and it’s particularly pronounced for the poor uh and that’s true in rural areas as well

As Urban um so for sort of petty crimes a lot of times they would just get solved in house so to speak so the broker would just help solve um sort of disputes between neighbors or if it’s a small sort of theft or something but for

Bigger crimes you do have to go to the the police right so if it’s a it’s a I don’t know big fight with people getting beaten up or with murders you have to go um and in that case you know as I said it would involve him either going along

Or him uh sending someone with you but what I found is that when Brokers had extreme control in rural areas they would actually block you from going alone so it wasn’t even an option you had to go through uh the broker but we can talk more about it because I had all

Sorts of other questions to tease out your level of access to Justice um that I don’t think we have too much time for here um so Shaban you’re right at some level this is democracy doing what it’s doing right politicians provide targeting constituents to get what they should get and you’re absolutely right

These politicians do not know whether these people will vote for them the broker does right right he has complete control which is why in rural areas they go through him um but the reason I say it’s non-programmatic is it’s it’s not based on need so in the paper I go in

More detail on that it’s the the ones in the periphery have much lower levels of provisioning that’s where they should be targeting if they want to sort of have an egalitarian provision level but they continue to Target the center because they’re more visible and everyone can see that look they’re doing something um

So it’s not being given to who needs it most that’s the case in rural areas also uh where were we um so what do Brokers get out of it that’s a that’s a great question and I think that’s a whole other paper uh sadly a journal only

Gives you about 8 to 10,000 words um but in the rural areas pretty clear right these are landlords and they have them where the big service peasants provide is agricultural Labor uh labor is very important particularly during sewing and harvesting season where it tends to even get in short supply so having a

Guaranteed supply of labor is very valuable uh and most of these peasants don’t realize how valuable they are uh so as a block they could actually put a lot of pressure on the on the landlord in urban areas it gets a little more watered down so it’s more of a political

Service they get so they get social benefits because they can sort of show that they have all these followings and they’re the leader uh but also politically even though the politician doesn’t give give up control over public goods he does still contract with these Brokers so there is some sort of rent

Seeking going on in exchange for the votes of their clients but it is as I said it’s a much more watered down system uh sorry Network um I have written a range of papers going into great detail between connected and isolated Villages there’s connectivity has done wonders um and I was telling

Somebody else this recently how on the surface nothing has Chang they still work for the landlord they still go to the landlord everything works exactly as every other Village but but it’s when you start talking to people you realize attitudes and reasons for joining the network have changed so um in Connected

Villages when I would ask that you know why does he help you he’s like they’re very clear because he wants us our votes he wants us to work for him that is why if he doesn’t do what we want him to we will leave um in in isolated Villages

When I would ask that they would say but how will we feed our children if we defy him sure I’ll get a drain but then what you know and if he kicks out of the k kick us out of the village we’re doomed so there’s a there is a stark difference

Connectivity makes a huge difference due to which you know Shabah sharif’s roads are a great idea so I’m the poster child for his uh for his politics um but yeah it’s sort of changing people’s opportunity it’s raising their opportunity cost from zero to to Market values yeah I think that’s about

Everything thank you very much these were great comments thank you Mish and thank you to the audience for a great set of questions to start us out with our next speaker shabo from ocidental college yes welcome thank you so much and like Mish I was also a

Student here so you know this is indeed it’s good to be on this side but also I wanted to acknowledge that I’m part of an all female panel where all other Scholars are more senior than me and and you know and and and doing great work and so you know

It’s exciting I’m humbled to be in this in this company so what I’m presenting today is part of my dissertation and what I hope to be a book project it’s called demanding The Last Mile foreign aid and political preferences in Pakistan I just want to share a little

Bit of motivation the reason um I started this project is because a lot of research on foreign aid in developing countries focuses on how Elites can coopt foreign aid so it’s about them using foreign aid to Target it towards their constituents or you know there’s a lot of corruption there’s

A lot of leakage but there is very little work done on how much voters actually know about the fact that foreign a exists or that benefits exists that they can actually uh receive and therefore this work is kind of focusing on the voter re side of the story um to

Kind of show that they can do things to get the aid benefits that others might be trying to keep from them so the question that the research primarily asks is can foreign aid change political preferences as well as participation in recipient States the argument that I make briefly is that the

Last mile of Aid delivery so this is right when the benefits have to get to the people is interrupted by bureaucratic and information hurdles and voters recognize these hurdles so they know that there is this benefit that exists but there is something that is keeping me from getting to this benefit and therefore

They change their political preferences to demanding the kind of goods and services that will help them leap that last Mile and get to those benefits and politicians and these are local level politicians they often respond to these voter demands so these are resource constrained politicians who

Have little else to offer but can offer for consolidative Services by supplying these broadly to the people who need them and therefore the argument I’m making is that the creation of public programs through some kind of foreign a outlay shifts political preferences towards asking for these last mild goods and also improves political

Participation the way I test this is through a natural experiment so I’ll talk more about this in detail but most of you probably know about the bay AER Income Support Program so I’m looking specifically at that program and um the inclusion into the program is

Based on a poverty score and I look at households that are distributed very closely around that poverty score um I do a survey of households this is a telephone based survey distributed close to the discontinuity and then I tease out differences in their political demands their levels of political participation based on whether

Or not they’re eligible for the program so what exactly is last mile service delivery this is a term that borrowed from Logistics um and I apply it to the public goods literature and Define it as publicly provided services that help eligible benefits uh eligible citizens citizens who are eligible for a

Particular good obtain the benefits that they’re entitled to receive so in the life cycle of any uh public good there is a budget allocation here I make the claim that part of it is coming through foreign aid maybe it is a form of conditionality maybe it is some kind of

Support um some kind of monitoring and evaluation that leads to the establishment of the program and then the program has certain eligibility criteria so who is going to get included into the program who is going to get excluded from the program and then there is an enrollment period so people get

Enrolled in the program and finally they get the benefits and it’s that movement from eligibility to actually obtaining the benefit that I Define as um The Last Mile of Service delivery and those of us who have lived in other countries are familiar that this is the case in all publicly provided Services

Across the boat so for example um Medicare in the US between being eligible for the program and actually going to your first doctor’s visit that’s what I term as The Last Mile and we know that there are bureaucratic hurdles there right so you have to uh Supply certain documentation you have to

Go to a County officer give an interview make sure that you’re meeting all the eligibility criteria then get an insurance program find a you know a primary care provider and then make the doctor’s visit so all of that uh those sort of informational and bureaucratic hurdles exist in almost all states but

The ones that exist in developing countries specifically Pakistan are very different and these are two examples that came out in my interviews with councelors in Punjab one of them has to do with fingerprinting so some I think a lot of us are familiar that there is biometric fingerprinting required to

Obtain uh the best funding and this came up again and again that women often have blurred fingerprints because they are washing clothes and dishes and so they need to do an extra thing they have to go to their local counselor or a gazetted officer and get an affidavit

Saying that they cannot be fingerprinted and so they should be able to retrieve the money without getting biometric verification or just the mobility problem so a lot of women said uh I also in interviews with women who said that you know it is difficult for us to just

Make that visit and actually obtain the money so these are the kind of Last Mile hurdles that I’m talking about the kind of facilitation services that voters are looking to receive so just a brief look at the theory um foreign aid the role of foreign aid here is that it is very

Pivotal to the creation of public programs what happens is that potential beneficiaries who I find in Pakistan are very aware that these programs exist and that they can benefit from them they experience feelings of ownership so there’s literature that shows that citizens know these benefits are intended for them and that they’re

Willing to go um and make demands in order to access these benefits however in doing this they face barriers to access so they know this program exists but I have to get registered I have to get a national ID card then I have to make sure I know

What the poverty score is etc etc they’re also aware of program quality and again this is this is where forign a is important because there’s a lot of monitoring and evaluation that is done by Third parties so they know that once they get into the program they will be

Able to get access to it this leads to increased demand for Last Mile Service delivery so they seek these services from their local elected politicians they want them to help them uh help facilitate them to actually get to these benefits and I argue that this allows them in an otherwise information

Constrained environment to actually judge their politician’s quality and just increase their political knowledge so now they know whether or not this politician is going to provide them with something because they’ve actually made an ask and overall this increases political participation more broadly so my theory lends itself to

Three testable hypothesis I’m going to focus on the two at the bottom today the first one is that exposure to foreign aid should increase demand for these last mile Services as opposed to other public goods and that it should allow uh voters to be able to evaluate their

Politicians in the paper I also look at incumbent support that those who are actually providing them last Mile Services should get rewarded at the ballot so to rephrase again the first hypothesis is that voters from households that are eligible for foreign a so these are bis eligible households

Are more likely to demand last month Services than voters from ineligible households and voters from these households are also more likely to evaluate local politicians to be able to say um how their politicians perform on certain characteristics than voters from ineligible households so I’m not going to spend a

Lot of time on this but we know that the B your income support program was created in 2008 it was created largely in response to an IMF conditionality that required Pakistan to have a social safety net it is one of the largest social safety nets in Pakistan Target it

Primarily towards women although the eligibility the poverty score and the fiscal flow is at the household level so the poverty score gets determined according to household level characteristics uh this poverty score is collected bya scorecards in a household survey this is done at a point in time

So it was done in 2010 more recently in 2022 and therefore there is is some room to manipulate the eligibility in the sense that you can go back and ask to be Revisited or reassessed or if circumstances has changed but it is more or less the case that you will get it if

You are eligible for the program and not if you’re not eligible and this is at the time when I did the survey the payments were equivalent to about $9 per month this is a distribution of the poverty score in the population from the last pslm so the initially the the

Eligibility score used to be around 16 but after the last survey in 2020 sorry 2021 I believe it was moved to 32 so there’s a large population of Pakistan that’s actually eligible to receive bis transfers now according to the new poverty score so how do I measure last mile service

Delivery this is again this is a survey that is done between um fielded to households that are distributed just around the eligibility threshold I give them a choice so I ask them whether they would like to go for ask their local politician for public transportation

Which I see as a good that will help them with Last Mile facilitation or ask for local uh Law and Order and the second choice is whether they would like to have an ID card Camp so this is necessary in order to get inclusion into the program versus a vaccination

Camp and then I asked them to rank five options again two of these are what I categorize as Last Mile Goods document verification and having a cell phone signal Tower because a lot of the eligibility verification happens through sms’s so you have to SMS your ID card

Number and then you get told whether or not you’re eligible um against three more conventional public goods so I’m going to go quickly over the empirical strategy it’s a regular uh fuzzy regression discontinuity design I asked them so what the outcome of interest is obviously demand for last M

Services um d is an indicator for whether or not the household is eligible according to the poverty score which I have for each household and then I also asked them whether or not they’re actually receiving bisp and I use that to instrument um and you all know that this

Allows me to compare households that are very alike except that they’re different in terms of Eligibility so these are the results I find that households that are eligible before bisp are more likely to ask for public transportation and for ID card camps versus households that are ineligible um similarly I find that

They’re slightly more likely to ask for document verification but less likely to ask for conventional Goods like free food distribution and this is not I I claim that this is not necessarily a well effect because a lot of these households again this poverty score eligibility threshold shifted very

Recently so these are households that are have just been included so this is I’m looking at PMT scores 29 to 32 29 to 33 and 32 is the cut off then I also asked them if they can rate their local counselor in terms of capacity approachability and honesty

This is not a ranking this is just can you rate them or not a yes or no question and I find that households that are eligible are more likely to say that they know about their politician quality um do they rank them higher if they’re able to get access to these last

Mile Services uh I find that in terms of capacity and approachability they do rank them higher not necessarily in terms of honesty so there are no results in honesty future turnout so does this increase political participation it there is no movement on political participation but in my sample

I also found that more than 80% of the respondents claimed they will turn out in the next elections to vote so this is a very high uh sort of bar to begin with looking at not electoral participation so again those who are eligible are more likely to rank their

Politicians higher they’re also more likely to know the names of their politicians so they have more contact with the state now moving on to the foreign aid aspect do people believe that foreign aid belongs to the people at large if they’re more likely to agree with this

Statement they’re more likely to ask for document verification uh and less likely to ask for conventional public goods like grocery rations and free food distribution does uh program quality matter not as much for the choice variables but again in terms of ranking it moves them up on document

Verification so because those who think that the program is actually good are more likely to ask for document verification camps uh just kind of concluding what this work basically does is it shows that voters know that they have access to these programs and they have the agent to change their political

Behavior change their political demands to kind of bridge the problems that arise because and this is I I flesh this out more in the paper because of donor agencies so when donor agencies say we want you to use biometric verification so we can collect data we want to have

Strong inclusion criteria so that there is no leakage it creates hurdles for voters but they are able to F to kind of overcome these hurdles through this creative use of political agency in a way way it recreates the citizen state contract at the local level around this exchange of services

In the very last mile uh some of the questions that I get is does this apply only to programs that are funded through foreign aid and I think that it applies this kind of ability to overcome Last Mile problems will apply to programs that are non-discretionary so anywhere where you

Have this eligibility criteria for inclusion exclusion might create hurdles and voters are able to step up and overcome these hurdles however foreign aid has reputational spillovers particularly in the case of Pakistan so when programs are funded by the World Bank when people know this is money that is coming from external sources they’re

More likely to trust the program will persist and so therefore they’re more likely to actually engage in asking for the program thank you so much I look forward to your [Applause] feedback I think my okay thank you uh my name is uh Aram Heather I am uh assistant professor of political

Science and Environmental Studies at the College of Worcester and I have the pleasure of being a discussant on shabo’s incredible paper um so shabo I followed your work for some time and I and I have to commend you on a really fabulous sort of I think this is going

To be the core of your book project um so I can’t commend you enough on you know I know personally how hard it is to put these pieces of data together uh and I think we don’t give enough credit to just how terribly difficult this is

Right so at some point you had to not only convince somebody in government to release this data to you but also give it to you in precise enough detail that then you could match it to households in the field um I’d love to know more about

Sort of your field workor and the places that you went to I think that’s terribly fascinating but I want to say that overall your your ability to match eligible populations and beneficiaries to government data sets it’s extremely difficult um and perhaps as part of a

Note in an early chapter or to this I would love for you to emphasize that more right I would love for you to emphasize just how incredibly difficult this is and I think this is part of uh the paper itself right I think this is

Not sort of a sidebar oh I did went and did all this work um to me the motivation is great I really like the the piece that you give us on foreign aid and how we sort of assume that it leads Co-op foreign aid but this is a

Really great example of how really poor and non-literate voters are actually aware that there’s something coming in from outside that they are eligible for so this is tremendous I always think we can be more ambitious with our motivation and I want you to think a little bit about how foreign aid can

Impose costs on the state and also on beneficiaries right and I know you treat this to some extent in your paper and you say yes it can impose costs but then voters are able to come around these costs to me it’s really interesting because part of what you’re looking at

Is whether the distribution or dispersement of foreign aid also creates dependencies right so this idea that we want cash transfers wherever they’re done in the world whether it’s Latin America whether it’s Africa we want cash transfers primarily to be a seamless frictionless way for the poorest of the

Poor to get money in their pocket to to get themselves basic goods and services right and this and this process of having to then go and get verification from your union count and get Biometrics because your hands are too calloused this to me is is crazy right like to me

This is imposing costs on these very poor people uh could there be easier ways to do this yes definitely uh but I want you to think in the context of your work whether if cash transfers are intended to reduce costs on the very poor and hopefully also reduce

Dependencies uh does it fulfill this right so is there a way within your data to see how satisfied are people with this process of of having to go to their counselor or go to their politician to demand these last Mile Services or is this something that they just wish they

Could do without right um you know rahan Jam’s really great work shows that uh as you’re also seeing that voting actually declines after getting uh being eligible for or getting a cash transfer uh and part of the reason he he sort of says is because you no longer need your

Politician right so you’re like okay why should I bother go voting for them and that’s a good thing right you’re no longer locked into these dependencies where someone’s giving you something and you have to trudge across and do this big show of voting for the guy who you

Really don’t like but you just kind of have to right um so I’d like you to think a little bit more about that um I really like your sort of idea of improving participation I do see how uh being included in this program does increase your exposure to the state and

That’s in tremendously interesting I think the really crucial thing is here that they know that they are eligible and that puts pressure on their representatives um I want to see whether this spills over into other Goods so maybe as you’re saying you’re looking at people who are just eligible 5 years on

Or even like two years on is there a a difference in the basket of goods that they’re demanding now that they’ve had this $9 a month come in consistently uh is there is there other basket of goods that comes in um and then if you could

Reflect a little bit on how foreign aid that is distributed through the government and not third party actors uh is is going to have different effects right we had a great conversation in the first on the first day on sort of a lot of the development in Bangladesh for

Example is happened through third party actors and nonprofit bisp is a really great example of this being through the government so how does this create on the one hand dependencies but on the other hand there accountability structure instead of teasing that out maybe through some of your qualitative

Work but overall this is such an impressive project and I look forward to seeing it in a book form thank you so much thank you for those uh excellent comments and for shabo to uh helping us continue this theme of house citizens can get services from the state okay so

Uh over to Crow start with Roco thanks thanks so much really enjoyed this and looking forward to seeing where it goes i’ I’ve got a couple of questions first is you’re doing a phone survey and yet you’re asking questions about cell phone towers so isn’t there bit of selection

There I’m not sure what uh yeah um and then you frame this as like political agency but but it feels like what you’re measuring is more sort of passive preferences for certain things so could you just elaborate a little bit on how to make that jump from preferences that

People uh asked to elicit in a survey to how they actually act and then um and then on the foreign aid angle I wasn’t quite clear on how the source of the funding of this program matters to people uh do do they even know do they even care where the funding comes from

And that seems like quite a sort of Keystone of your of of your framing that I didn’t quite get um yeah thank you so much hi um I’m aan thank you so much for this talk um I was going to Echo what um Rocka just said I think it’s quite

Important to measure who these people think this program is associated with so we have old work from Kon has showing that now they do think that it is more technocratic and less involved in the government ever since the introduction of the PMT but also in follow-up work

That I’m doing with them it does seem like people still associate it very much specifically with Beno so to what extent where do people think that this program is coming from because for all of the last mile actions that you’re thinking of you do need to interact with a bisp

Official or nadra official not your politician necessarily to to get that stuff done but in what yeah is it that they still attributed the politician or is it that these particular technocrats who were instituted specifically so that the dependency on the politicians is eliminated are not delivering this the

Are not doing their job uh so that’s just uh one thing and then second I think it would be really nice to know what are the Baseline characteristics of this sample um because I would have imagined that yeah like maybe many of them already have ID documents or some

Of the services and that’s why they’re not demanding the idence and they’re not that necessary in The Last Mile because they’re very necessary in the first instance of even being surveyed by the NSR question hi shabanu um so bisp is an unconditional cash transfer program

Right uh do you think how do you think uh your results might uh generalize to a conditional cash transfer where the need for political engagement might be a little bit greater I suppose yeah okay um so my question is hi so my name is Muhammad My question is more

About like identification so I I was thinking of so you’re trying to understand the effect of what the source of funding on on the outcome variables how do you distinguish that from just the effect of getting cash because the discontinuity is between eligibility for cash versus not right so where is really

The variation in the source of funding and how how are you making how you basically using that discount to say something about the source of funding was there one more hand at the front Okay then shabo I don’t think you need to feel obliged to answer every

Question in detail but it’s better to include hi so I’m Adil and my question was kind of similar regarding identification from Muhammad in the back um like how do you disent disentangle like just exposure effects to the government and also like you’re kind of when they go to the government to kind

Of you know get biometrically verified you’re priming them on just like knowing about say like verification and so after like doing that perhaps you just like understand the importance of that more so like you know how do you disentangle the priming and the exposure effects from like what you were saying which is

Something like deeper about like you know understanding of Service delivery got okay so quickly responding um field workor this was done mostly during covid which was even more challenging so initially it was I had data on nine districts in Punjab and distributed around the initial discontinuity which was

16.17 and then shift to a representative telephone survey so then I had data I I also did a WhatsApp survey so I have data on 30,000 households the WhatsApp survey does all but the telephone survey is limited to 3,000 so it was in terms of where I went so I spoke to counselors

In Punjab and then I did focus groups with B beneficiaries in the Islamabad office and also in rural m in rural Punjab so close to close to Multan um could there be easier ways to do cash transfers absolutely so I don’t I don’t think that this is what they

Should be asking for I don’t think it’s like what we’ve been discussing throughout this conference right it’s the second best so it’s not the first best they should not be asking for Last Mile facilitation they should be asking for other Goods but it’s just that this is the first step towards getting to

Know that there is a political process and politicians will respond um okay um in terms of foreign a coming through the Govern government so I think that’s important I think that not only should foreign a come through the government I make the case somewhat in the broader dissertation that

Politicians local politicians have very crucial knowledge of where people are and what people’s needs are and so these are very specific politicians most of the counselors I spoke to are people and which is why I was interested in mish’s work on you know when is there an overlap between landlord and politician

Because a lot of these were just slightly more educated people in their communities who decided that they wanted to run for counselor because they wanted to be able to serve their Community better and so they are not distanced from their their uh constituents they told me that you know we count our

Constituents as people and so you know we’ve been there we we eat with them we know who they are we know what their needs are and we are driven by an interest to serve which I thought part of it is obviously social desirability bias them telling me what I

Want to hear but you know I also think that there was this aspect of service at least at the very local level um phone survey versus cell phone towers yes I haven’t actually I haven’t actually thought about that because this was a survey done on um on cell phones but the

Idea of the cell phone tower was basically to find out whether they would get better service because in a lot of these communities they will go to the person who has the cell phone or has the ability to go somewhere uh to actually find out about their eligibility whether

It’s political agency or just passive preferences so the WhatsApp survey actually adds a cost so that’s a um that’s where I move it from preferences to whether they want to Bear a cost to actually make a demand which is can you send an SMS asking for the thing you

Want and so that is forthcoming work I haven’t looked at the analysis yet um source of funding so I get this question every time I present this is not for8 because people don’t perceive it as forign a and in my theory it doesn’t matter whether people perceive it as

Forign a in fact to your question name and a lot of people you’re right initially there was a so I started working on this in 2016 when there was a lot more um sentiment of this is coming from Beno a lot of women thought that this was her inheritance that she had

Left to the women of Pakistan um there is much less of that now as you know in in my phone survey a lot of people knew that this is a like like uh and Biber find that you know this is something which is provided from the state that

They are entitled to receive however first they still have to overcome those hurdles and they go to the counselor the local level politician because they know that they can resolve their problems and often it is easier for them to communicate to the counselor than to the

Bureaucrat um but you know in in future work I want to look at whether there are there is a way that local bureaucrats might be better at doing this but the forign it is really in it’s an indirect effect it’s the fact that forign 8 keeps the program running with the efficiency

With which it runs so this is one of the programs in Pakistan that has been monitored the most evaluated the most there’s Oxford policy management has done like multiple rounds of surveys trying to find out where are the Grievances like what do people like what do people not like and it is fornate

Because of which you have these you know biometric verification things because the government needs data to ask for more money so they say you know we do it so well there’s no leakage there’s no inclusion errors which is why if you fund ASAS which is based on the same PMT

Score infrastructure it’s going to run great because you know this runs great and so I think there is an indirect effect of foreign a in improving program quality and people are aware of it they know that because you know uh in in the survey experiment I ask there’s a survey

Experiment I do as part of the survey where some of them get told it’s a for8 program they rank it higher in terms of quality so I think it’s an indirect effect rather than uh a direct effect um identification yes so ident I am using eligibility as a proxy for whether or

Not they’re getting exposed to this program um whether I’m priming them so I’m not actually telling them that this is a survey that has anything to do with bisp the bisp data I have from administrative records they’re just getting a political survey what do you want like would you

Rather demand X versus y but they don’t know that I’m doing this to find out whether their bisp eligibility and then in order to not prime them the bisp questions are right at the end of the survey do you actually get bis so they do the whole survey and then they get

Asked about whether or not they get the program but I hope that answers most of the questions thank you so [Applause] much thanks shabo and thanks everyone for continuing the excellent questions over to our next speaker Shan Amano joining us from the University of Virginia thank you thank you everyone um

Thank you to the organizers for including this paper I’m I’m very excited to present this uh this is actually the second time I think I’m presenting it uh this is on spillovers in state capacity building this is Joint work with Clum Audia and this is part of

Our larger research agenda in trying to understand you know we know that state capacity is important but where does that animal come from so this this is like you know we’re trying to contribute to that agenda and I look forward to your feedback um so like I said I think finally

Economists are catching up to the political scientists and sociologists you know who’ve been saying this for ages but finally we recognize that state capacity building is important for development the larger sort of open question right now is you know how do you build that state capacity so part of

That those conversations that are happening the biggest push that we’re getting at least developing countries are getting is um from the World Bank and the IMF Etc is that you know maybe you should go for technology because technology will cut through the bureaucracy etc etc and so as a result

Of it for example you know you have 52 countries computerizing their land Registries between you know in a 9-year time period you’ve had such a large roll out of digitization so you know technological reforms the the good thing with them is that they reduce corruption they are able to reduce information asymmetries

You know you can have better targeting of public goods etc etc at least that’s the argument but what goes missing in all of that that you know the one-sided dialogues that we’ve been having is is are these two bullet points which is you can have lower corruption on one angle

But given that people and their incentives haven’t changed that social structure probably you know changes in a way in which you can have a displacement of corruption from one side or one activity onto another or you can have people who are you know it can also be displaced across individuals moreover it

Can and it does disrupt the organization of bureaucracies and what I mean by that is you know whenever you have uh a digital technology coming in you replace some task you can have more specialized bureaucracies on top of that of course technology is going to change the embeddedness it’s going to change social

Interactions between people and it’s actually this side that that the conversation is completely missing so there’s little empirical evidence on how technology has an impact or interacts with this sort of inner working of the bureaucracy right so in this in this paper we’re asking do Technology based State capacity reforms have spillover

Effects on the organization of bureaucracies and their effectiveness and the reform that we’re looking at is the digitization of land records that happened in Punjab Pakistan it was a staggered rollout and we’re we’re looking at the capacity of the state to collect taxes so just to give you a

Snapshot what this these land records did is that they fixed property rights to the extent that it reduced property rights uncertainty and we’re trying to see you know because in some sense it’s a legal capacity reform so we’re trying to see how that legal capacity reform spills over into the fiscal capacity of

The state or not um so what we do is you know we’re able to exploit the staggered roll out of the program and we you know the identification strategies are classic differences and differences the tax that we’re looking at is the agricultural income tax and the tax base there is

Cultivated area and farm profits and um you know the big heavy lifting came in actually both of these things so this is this data was hand collected and then digitized um and this is agricultureal tax data from 2006 2013 and then to be able to look deeply into the bureaucracy

Itself we did a survey with the local bureaucrats which was a retrospective survey of what happened before digitization what happened after and we’re able to build career trajectories and then connect it with the tax data so we can see individuals performing before and after the reform um of different

Types as well so what’s the big result uh contrary to results that are almost always positive of digitization we actually find that once the state digitized land records it had a negative effect on tax collection right what does that mean it means that the state is fixing one end and instead of there

Being a complimentarity you have a fall in tax collection so you’re negatively impacting fiscal capacity and we look deeply into the mechanisms this is not driven by tax base so it wasn’t the case that fixing of property rights led to a fall in the tax base on the contrary it

Actually improved the tax base and there’s now a paper that’s out in in G which shows that tax base actually went up up land labor market frictions reduced and so this was in essence you know for the market this was a good reform but for the bureaucracy it ended

Up having some negative consequences what we see when we when we combine this survey data with the tax data is that it’s actually the bureaucrats capacity to collect taxes and then we you know we have some results on what the exact mechanism within that can be so this is

The background uh this is obviously not visible at all but there are three phases of the program and and um you know this was supposed to be rolled out starting 20 9 but that did not happen in the paper we go into details about you know how balanced these are these are in

Essence balanced except that phase one had more of a lower population than the rest and so what were the Direct effects of the reform you know the first was it was good for the market land and labor market frictions reduced it strengthened property rights the average transaction

Time went down I’m just checking the time um the average transaction time went it went down from 2 months to 45 minutes which is phenomenal right it’s a it’s a great thing to have and and then 72% of users perceived the new system to be more secure and 60% of users reported

Reduced land disput so in essence you know there is something positive that’s happening in the market the thing that this digitization reform changed was were these three first you know it changed the the scope of task that bureaucrats were doing so before this reform happened the the bureaucrats at

The revenue Circle level were collecting taxes assessing taxes and then they were also in charge of land records which means that they were fixing the land records and also giving out Land Titles which of course is a highfrequency event then after the reform this land record

Business all of this went to a separate bureaucracy people who are from Pakistan I don’t need to convince this audience um it takes me a while in in the US for them to understand what’s happening but the land record business just like you know it went to a different bureaucracy

A new setup was was put in place there were arazi centers and and so this has kind of taken off the thing that you know on top of that what happened is that it reduced the frequency of meetings of the bureaucrat with the tax um payers right and for those of us who

Are familiar with that literature on embeddedness it changed that embeddedness of the bureaucrat in that Society Economist would say oh that’s great it reduced corruption but I have a different way of thinking about it and then it also so while there was a reduction in bribes because now you’re

Not giving out Land Titles to people right so now you don’t have that high frequency event happening but it you know while it changed bribes in that setting you know they do talk about you know I’ll show you results which show that this corruption displaced from Land Titles onto tax collection and

Assessment so what kind of data we have um we have the planned roll out uh roll out of the digitization across different districts this was collected from the project office once it was in place uh now there’s no longer a project office then we also had the actual roll out of

Digitization across villages in Punjab which we got from the pl like I said the heavy lifting comes here um can you click on the picture for me uh and so this is actually you know this help this data helps with you know trying to address this larger ree agenda on on

State capacity and does it not work so um so this I I was able to get this data from the board of revenues uh record room which is in their basement and it took a while and and so there is you know there’s a lot of good data and

This the reason I’m saying this is because there are other researchers in the room and so if anybody wants to collaborate and work on it you’re more than welcome to talk to me then the last set of things we did was we talked to these bureaucrats and we carried out two

Sets of you know we did a repeat sort of service so we did the survey once with one set of bureaucrats and then we went in again because we wanted to rebuild their career trajectory so that we could then combine it with this tax data it

Was a long effort of string matching and and like double checking um so now we are able to look at districts and over time but we’re also able to look at bureaucrats over time which is which is good for us so it’s a classic difference in difference two-way fixed effects you

Have District fixed effect time fixed effect the outcomes we’re looking at are tax collected in a district our outcome is Right skewed so we have inverse hyperbolic sign which is basically nothing more than a log 1 plus the the outcome and then we also report taxes in

In rupe the digitization dummy is it takes phase one and two and then you know it switches on one after digitization and remains zero after so this is what we find so this is showing you each dot is a coefficient and then the bars are 95% confidence intervals so you can see pre-reform

These you know digitized non-d digitize are sort of similar except a little bit of up and down but you know we don’t see a large effect but once you once you reach this side um we’re only able to estimate by the way effects for two

Years so this is a short run effect once you reach this side you see that the magnitude of this fall is pretty large so even if you ignore the significance and whatever the economic magnitude of this negative effect is quite large and this you can see also from so maybe you

Know people can focus on on these percentage changes the First Column is tax collected IHS and then the second one is tax collected in levels and rupees and you can see that the effects range from 49% to 84% once we first saw this you know we

Were like okay this is super large you know this is a little bit too much maybe it’s a short run like there’s too much adjustment going on but actually if you look at it and you place this in the literature it is not a very far removed

Uh you know effect that we are seeing in other studies as well so if you see the study on E filings you know you have e filings and that led to a 40% reduction so the kind of effects that there that are there in the literature ours is sort

Of in the middle of that and then the thing that we take away from this is you know across 36 districts there is a thank you there is a you know we actually lost so we had digitization you know I’m not doing a welfare analysis here but we had digitization fixed labor

Market land market but it led to a loss of 290 million which is almost the same as supporting a you know 15,000 families on on bisp I’m going to ignore this this is you know the recent literature on diff and diff is is rif with like lots

Of uh lots of robustness checks we did all of them happy to talk more about it um but I’m more interested in talking about the mechanism so you know the first you know sort of cut of results is that digitization led to a fall in tax collection then you know we were curious

Whether it’s because of a fall in tax base and so we sabrin B has a paper exactly on this and so we wanted to like retest that within our sample and see what’s going on and so when we looked at data and I’m going to show you in a

Second that table we don’t see any negative effects on tax base if anything it actually went up it’s actually all the action is coming from the bureaucrats capacity to collect taxes which is coming from first the assess less and then on top of that as a fraction of that assessment they’re also

Collecting less so it’s also the effort margin plus the collusion or assessment margin that that’s happening so this is um the effects on tax base we collected satellite data on vegetation cover because recall that the tax is on agricultural um cultivated area and then profits so we have data from household

Surveys that the government collects uh that the government does at its log Farm level profit per acre we combine that with you know we have vegetation cover data which is satellite imagery and then these again are proxies for that cultivated area if you can see I mean

The effects even I mean forget the significance the effects themselves are pretty small as well so you don’t see any negative effect on cultivated area but in fact you see a positive effect on profits so if anything you know you should have had a positive effect on fiscal capacity which you

Don’t so then we looked at this is coming from you know what are the taxes that that were assessed by these bureaucrats so these bureaucrats are in charge of going to the Farms um you know if people who who are familiar with this uh the patwari or the the people we are

Studying are gavar they actually go into the field with the patwari or the patwari goes and so their team is visiting is supposed to be visiting these crops uh these cultivated area and then write down you know this much sewn this much cultivated this is the tax

Levied and so they’re in charge finnally enough of the assessment itself and so you can see you know compare this to the fact that satellite imagery is not showing you a fall in cultivated area right compared to that you’re seeing a fall of tax demand that goes you know we

Have two measures of this assessment this is coming from the board of Revenue itself and you see a a fall in tax on top of that when we look at the the more disaggregated data there is actually a fall in log reported cultivated area by these bureaucrats as

Well and so this is suggesting if you place this again in the literature this is suggesting that collusion between that bureaucrat and that taxpayer has gone us gone up and one of the reasons we think that can happen is because now there are no longer those land

Transactions right so you are not you have to if there is a threshold of income that you want there is a a task that you could get you know rent on that task is away now the only task you’re left with is this and so maybe that increases your the collusion you have

And so that increases your bribes then we you know this right here is a bureaucrat level panel and so you’re looking at within bureaucrat variation in how much taxes they’re collecting and so the first thing you can see so these are the outcomes tax collected as a fraction of tax assessed

So tax Target is actually tax assessed whether at least 50% tax Target was achieved whether at least 70% 5% was achieved and then the share of months with zero tax collection so you can see after tax collection there’s a sizable drop in that bureaucrats performance so notice that because we’re doing we’re

Using a within bureaucrat variation we’re sort of controlling for the ability or the unobserved heterogeneity of that bureaucrat as well so you can see that taxes fell tax base nothing happened but both on tax assessment and on tax um you know collection conditional on that assessment you see a

Fall in performance and so we use survey data and we ask them actually you know do you think digitization has helped you with tax collection um I mean 39% said no change some said yes some said no but then actually a large you know 46% are saying digitization has actually made it

Worse and we said why is it that it has made it made it worse and so the arguments that they put forth is now we have less leverage over people so there’s less influence due to less public dealing um and we can translate that into less information less bribes

Etc but it’s basically that literature on Leverage that is sorry that is um that is what it speaks to and so the key takeaway here is digitization reduced tax collection not due to a fall in tax space it’s coming from lower bureaucratic capacity to collect taxes and so you know change the collusion

Dynamics removing one task you can focus all your energies on the other task and do more collusion and then there’s loss of bargaining power of that so when I did this field work with the with the bureaucrats and I was you know the interviews I’m almost out of time okay

So the interviews I conducted I was told that actually you know earlier we used to uh you know um withhold their Land Titles and then say you have to pay taxes otherwise I’m not going to give you your land title and now we’ve lost that leverage so all in all um implications

For State capacity reform we know that now digitization is widespread like I said there are 52 countries that digitize within 9 years that’s like a huge agenda that that the international organizations are pushing forward and you know this idea that this bureaucrat was doing both land records and tax

Collection it’s not a novel idea this is the fact that bureaucrats have multiple tasks is something that’s like it pervades public sector bureaucracies across the world and so you know our key takeaway is that whenever you do these kind of technological reforms first of all you need to sit back and think about

The set of people that you’re you’re influencing or the organization that you’re going to change and either couple them with some some kind of incentive schemes or social intera you know understand how social interactions are changing or you need I’m done or you need um you know you can’t have peace

Meal reforms where you set up the digitization as like okay now there’s one bureaucracy and you left the other bureaucracy you know with with no changing incentives Etc to to still collect taxes okay thank you I’m going to stop [Applause] here please go ahead so thank you so much for this paper and

You know um echoing irim I think that this was the kind of uh data collection effort that had me flabbergasted because I was like I I don’t even know what has gone in to this kind of data collection effort also because at one point I was trying to collect data in the basement

Of the fpsc and it didn’t work out so I know I know how hard that can be all right so um so the paper is basically about investment in one form of State capacity and how that actually reduces another form of State capacity the objective of the reform is to make

Property rights more secure and that happens because profits go up um however taxes that are collected have actually reduced and the mechanism uh that is talked about is basically a reduction in influence right and as a as a political scientist I have no problem in buying

That result because I think that the way bureaucrats and politicians function in clientelistic societies is very different so you need those kind of personalistic relationships uh to be able to achieve things efficiently um I’m I’m a little I’m I’m not entirely convinced as to why you frame this as a

Reduction in Bure like this reduction in tax collection as a reduction in bureaucrats performance and the loss of bureaucratic ability because as I was reading the paper I kept thinking is it just not a reduction in an extortion so perhaps they were collecting more taxes because they could

And now taxes are aligning better with the cultivated area that actually that actually exists and you show that cultivated area has not changed according to satellite imagery but one thing I think you can very easily do is show whether the assessment of the tax collectors now aligns better with the

Satellite imagery so that is I I don’t know if this was in the appendix but I didn’t see a discussion of well now because you know these records have been digitized and you have data from satellite imagery we see that this reassessment even though it is lower it

Is aligning better with the with the cultivated area Um so so in that sense is this really signaling a lack of influence and or and reduced Distortion as a result of that loss and influence or is it signaling higher levels of collusion also in terms of collusion I find that a little contradictory so you say that there’s

Less public dealing but at the same time you’re claiming that there might be collusion which is why the tax targets have dropped and so I was I was just thinking that that might be a little difficult to sell because on the one hand if they’re not meeting as often

Then why would they how would they collude to reduce the tax Target um and finally I think in the paper there is more of a discussion on how there are profit-based taxes so large landlords are paying taxation on profits and smaller landlords are paying taxation on

Cultivated area and so I also wanted to know whether this would in the long-term shift taxation towards higher taxation of profit-based landlords that are larger landlords and away from this um away from cultivated area landlords that are smaller landlords so maybe in the long run it is reducing taxes from these

Farmers but actually raising taxes on another set and so this is actually a progressive reform in that sense uh but thank you so much those are all my comments [Applause] yes wonderful okay so we can start off over here uh great work U this is Imran I um

I did some work in digital transformation earlier in my life so based on that not a direct U comment on the paper but just certain observations for f future possible work uh I mean digitization is great and all of that but it does introduce a lot of burden

And friction and I’ll give you a very simple example um a few years back the government decided that the applications to colleges from students who do FC will be all online so the burden that created on Rural population was that for example we do we do we do some social work in my

Villages and a lot of kids started complaining that Internet is just not available how how do I even file an application right so something that’s supposed to make your life easier it actually made your life difficult because previously you could ask someone who’s going to the city hey you’re going

To the city bring the prospectus or whatever that is fill it out and another person who’s going will deposit it but now you know that Loop was cut off so actually it introduces a burden instead of facilitating it so that’s one comment and my second comment is that um a lot

Of these U projects are financed by World Bank funding and others and they are all nice shining shining and glossy when they come out but UPS batteries die down in 2 years and generators break down after four years and then that actually could create a huge issue

Because now you have systems that are even more dysfunctional than paper and Pen four years down the road so how do you how do you account for maintenance of these systems and that’s something that you know as a society and as political and um economic scientists we

Need to um think about and finally um you know we can talk offline maybe later but at least the uh property uh the land tax bill that my family is still getting is actually with a you know on a pen and paper and is not coming from a computer

And I can actually literally give you a copy of the last one that we received so maybe some of that is playing into you know why text collection is not reconciling and you’re actually seeing a dip because it’s you know paper versus pen and to your point of systems not

Being digitized through and through thank you um hi my name is isra so I had a question on um how whether or not maybe the reduction in tax has something to do with the fact that maybe the state just doesn’t have the capacity anymore because I did a bit of research on this

And some of the senior officials in like the board of Revenue would say oh we just don’t have enough patwaris and stuff anymore to just go and do our work for us anymore enrollments in this have fallen a lot so maybe this is more of a

Capacity issue on that end and the other based on what was just said is that I know the arazi record centers have been digitized but there’s still a fair bit of Reliance on the bwari anyway even now in a lot of these um systems so how how

Have you exactly accounted for the fact that this might not have just completely eliminated because this is still a discussion that is having that we’ve digitized everything but people are still relying on the patwari and stuff for land record management and stuff so can can I respond because then

I I’m not a very good like I won’t remember so so um I think all these points are well taken um the first thing was was coming to this gentleman on this side um was that you know of course digitization creates frictions there are papers on that you have lower targeting

You can have distributional consequences and I think some of the papers here have already already shown that you will have distributional consequences uh but we’re not really you know we’re not speaking to the direct effect of the digitization on you know markets for example on what’s Happening uh to land and labor

Because that that agenda is sort of already fertile and you’re getting a lot of papers there what we are trying to say is that it has an impact on the bureaucracy itself so it’s not about the households that’s not the unit of analysis here it’s about the bureaucracy

Itself that we want to talk about and then this idea that you know that there could be more dysfunction uh I think why on the contrary as far as my understanding goes the paper that studies this directly um I mean from my although wait so that paper doesn’t study the bureaucracy the new

Bureaucracy itself that paper is studying what happens to the markets we don’t see that dysfunction maybe in the long run uh but we we haven’t reached that long run as yet so maybe in the long run there will be more dysfunction if the external environment changes but

This reform at least is not that one of those then the last thing was um people are getting so the effects I showed you were the intention to treat effects which means that you know it’s the intention of the government to digitize and that’s the variation we use not the

Actual variation which basically means that there are still Villages that do not have um you know that don’t have their mozas live on a razi center that doesn’t mean that every single you know land record has been digitized for sure that’s that’s not the case and maybe the

One that they were talking about was more for the urban setup because in the urban setup I I have no idea what what goes on there maybe that’s less computerized doesn’t seem like it um so there were great questions on this side uh the the effects I’m studying are the

Effects for two years after the reform and as far as my understanding goes patwaris were you’re correct in saying that you know you are losing patwaris I think they earlier had around 6,500 they’re now left at 5,800 something like that um but the the effects that I’m

Setting are so short run that you’re not going to see a large drop off of that bwari right in the long run Maybe you’re losing human capital and you’re not getting in the new patwaris you’re not hiring new patwaris uh and that’s another paper to be written but I I

Completely agree with you um the the thing that I think we we I haven’t talked about it here but in the paper we explore this idea that maybe it’s just that transition from one system to the other which means that a lot of Burden fell on the patwaris and and so we did

That this in the survey we asked them about their task their time use their you know what kind of task they’re doing and you’re correct that bwari were involved they were not involved in land records but they were involved in record correction and and trying to fix that

That problem but if we look at their time use there is no statistically significant difference in the time that they’re spending on the job in the digitize versus non- digitize so although the nature of the job has changed the number of hours they’re spending has not changed so you should

Still not see a negative effect right if the tasks are the same and now if there’s time I can go to Amon otherwise we can have another round question Amon I think had no I think ly answered it so I I’ll let someone else go okay the the tax that you mentioned

Which um which has been collected is um is an income tax my understanding is that a large part of the tax in agriculture is collected as a land tax um to what extent is is that distribution between land tax and income tax changing and What proportion is

Current ly being collected in in this form that’s question that’s yours thank you thank you this was very interesting um I’m Anam uh I think this has kind of been hinted at by a few people but maybe if you could speak to it directly so the

Impact on the bureaucracy I guess I can imagine like two different channels two different kinds of impacts and and to me it sounded like what you’re saying is that bureaucrats don’t have this task anymore and because of that they have fewer social interactions with people and hence less leverage but could it

Also be that you know and and that’s where I think some of the points around how well digitization actually works come in um the ability of bureaucrats to to interface with the digitized system if some are digitized and others are not digitized then do they just ignore the

Digitized records all together and then that creates a hurdle for them to go through so just how the quality of digitization may actually affect bureaucratic Behavior Uh I mean I’m also thinking of this you know in the context of I paid a recent visit to nadra for my

Passport and you know nadra has digitized a whole bunch of systems but and and their online forms you can do them from abroad and this is obviously totally different system but you know it was that quality uh question okay great thank you uh one last one last and then

I can quickly answer then um thank you for the presentation and I found it very insightful I have a few questions one being very simple asking whether it was an immediate change to digitization whether as soon as the system was developed they just immediately stuck with that because um

In my experiences working with the public sector in some cases of digitization it’s a very gradual shift where they keep it a hybrid in a sense where they have both um paper records and online records and then eventually uh move to digitization cuz I feel that affects the end ending result in terms

Of um in this case tax collection uh secondly I wanted to ask it um the conclusion here being that it there was a reduction in tax collection because possibly of a reduction in leverage so I wanted to ask that whether this was a the people in the actual bureaucracies

Suspected this and whether this would call for them to investigate this further and perhaps introduce penalties for Under reporting on behalf of the people as to whether that was a possible mechanism I did didn’t quite catch that sorry can you repeat it so um one of the conclusions was that there was an

Increase the sorry redu due to the reduction in leverage that reduced uh tax collection so and I suppose uh implication of that is that that means that the people themselves are Under reporting the land owners so whether that calls for further penalties for them or whether this was something that

Would potentially be further investigated um secondly uh sorry lastly I just wanted to uh ask one last thing that um if how do we if power reduces and that um reduces tax collection and the the solution to that would potentially be to increase leverage how do you increase leverage without again

Increasing the propensity for corruption and misreporting amongst the bureaucrats and I suppose you mentioned a few incentivization strategies potentially so I wanted to if you could touch upon them if you perhaps know them if you have the time obviously so do I just have one minute probably or or I’m happy

To CH offline if that we need to move on okay take minute to okay um I think let me just start with this gentlemen so um we now are digitizing it you know we we did digit I did a first round of digitization where I tried to just get at the final like

What’s the tax collection which included both income tax and the tax on their income as well as cultivated area now what we’re doing is red digitizing so that we can separate out these two things from my understanding in the field is that because income taxation is very self-reported which means that you

Know you have to come and file a tax for your previous year’s agricultural income it’s really not done I mean the only thing that they do is um the government knows that anybody who’s above 50 acres the government would go to them themselves and so there’s an expectation

That larger land owners are the ones that going to file so in essence it Remains the most documented part is cultivated area and that’s the one that gets taxed um happy to chat offline so that I don’t take up too much time thank you thanks everyone okay okay does everything work all right

Wonderful uh so I’m so thrilled to be here not least because um this paper was kind of conceived in lore my work is primarily on Karachi uh I look at the privatization of um Public Service delivery primarily of electricity and so I’ve spent sort of the last six years immersed in uh seven

Immersed in karachi’s privatization effort with k electric and I’m happy to talk about this more but what I’m going to do is actually tell you my comparative case study today which is on laor so really I want you to take this in the context of a comparison between

What was happening in Karachi which for me is a lot more detailed and I can go into a lot more detail on that um but this is really my sort of um sort of a final chapter where I look at you know Pakistan is this state that is in a

Chronic electricity crisis we’re talking about chronic uh debt when it comes to payments for energy uh and lore is the second largest city in the country that has been facing the threat of privatization for very long time but has not been privatized it’s discom its Distribution Company is still a state

Owned company and so I was very curious about how uh equivalent sort of people and specifically sort of middle and upper middle class people but then also sort of the lower sector thinks about energy right and there’s and so I’ll go into detail on that uh I also before

Starting want to thank uh Dr om Javed who was instrumental in helping me train and get this survey off the ground he was a copi on the ground that was provided by igc okay so this is my original fieldwork right I start here as

A graduate student in 2017 I go to a lot of different constituencies in Karachi and I start talking to them about what life is like under a privatized Service delivery regime okay and I tell them what were you doing before and they tell me well you know uh in muzafarabad in

Kurangi uh the &p would sort of come around some guy would come around and give us these little perches these little like um paper handwritten bills for electricity and we would pay them and we assume that he used to go and then pay k kesc or whatever it was right

Um and now since privatization which sort of in in its latest and its most recent attration took place in 2009 uh K Electric which is the newly privatized bureaucracy treats us as individuals so even if we go collectively and we say look this this area has suffered outages

We’ve paid our bills what’s going on they say show us your individual bill right so show us the bill that comes to your house we’re not interested in this Collective action like we we don’t understand understand it right they don’t have any capacity to sort of absorb this kind of collective action so

We used to just pay A&P for the bill now we’re individuals the finding that I have in my book is that this has had a tremendous effect on trust in others so because K Electric has been extremely successful at giving people this individual uh accountability or telling

Them that look you will get electricity for as much as you pay for we can talk about whether that’s accurate or not you need to First deliver yourself as a consumer and then you will get that benefit right this is a pay for you

System and so what I found in in sort of on my core work is that instead of blaming k electric or blaming the government when they have bad service outage they blame each other right so they’ll say well I paid my bill and this guy didn’t pay his bill and that’s why

Our entire neighborhood is subjected to Collective punishment but it’s his fault right they don’t blame the system of collective punishment so this is the kind of socialization right so I’ve done extensive work on the socialization that takes place when something is privatized lore presents me with a really

Interesting comparison what is the kind of political socialization that takes place when the threat of privatization is present right that the that the system doesn’t privatize but the threat is ever present that oh lore might privatize its electricity and there was sort of pmln does aads against this and

There’s a lot of stuff in the news and whatnot right and so uh this is actually my ra Isa from lams from a few years ago and so she was very kind and sort of took me around and did sort of field work with me uh and we went to chungi

Amar sidhu which had just recently before this had massive protests against uh a loss of electricity right so they had load shedding for days or or outages or that you know Transformer had busted and they were on the streets so if you guys were here during that time you’ll remember just before the

Pandemic they just closed off the road that entire neighborhood right and azra this lady in chongi Amar sidu tells me that you have to break some windows and this I found this so interesting she’s openly admitting to the fact that you know Collective action is not always

Pleasant right it can be violent as well like nobody was heard but you know they they they they see this as something that they’re entitled to nothing happens if you don’t go in a group so this is my this is really the puzzle for me right

Like oh here you have a very similar group of people these are Urban residents of Pakistan having very different attitudes towards uh a common resource that everybody in an urban center needs which is electricity okay these guys are like well we’re individuals we’ll only get electricity

If the other person is paying that bill and here nothing happens if you don’t go in a group so kind of these you know opposite and so the puzzle for me was um you know these are highly competitive political constituencies One Way electricity is privatized the other the

Threat of privatization is ever present I want to know whether do repeated and wellp publicized attempts to privatize electricity shape attitudes towards it as a commodity or a right uh I want to know to the extent to which Lor’s middle and upper middle classes are tolerant of

Forms of non-payment and theft uh and I want to see what is the impact of this on trust between individuals and towards the state okay so I’m going to briefly situate this paper in my larger Book Project I’m going to provide an overview of my larger theoretical framework uh

Descriptively uh I’m going to provide you the context of electricity service in lore and findings from a researching experiment to uh foreground some of these I found the respondents report being in favor of publishing defa punishing defaulters right but a significant number are also not paying

Their bills okay so so it’s not like they are hiding the fact that they don’t pay their bills right and and I’m finding that in Karachi there’s a lot more sort of emphasis on the fact that we are paying our bills whereas here it’s like man you know if if I have a

Hard month if I don’t get the kind of income that I needed I I I can’t skip my bill um second what I’m really finding here is that they are very much more tolerant towards theft by lowincome individuals right I’ll also should have differential results of this as well

Okay um so my introduction to the Book Project is that states no longer exclusively provide public goods many of us sitting here in the audience know this I want to I asked how do political parties enable this process and I’ve sort I’ve done that work in Karachi uh I

Look at how this shapes attitudes towards the sh the state and then towards each other and then the final chapter is a comparative study with other cities in South Asia and I’m happy to talk about any part part of this um during the Q&A so I’m going to be presenting this

Today so there’s really good work on why States Outsource right and so the two directions that I’m interested in is that there’s a supply side so there’s the state itself getting pressure from the outside saying you have to privatize you have to rationalize your uh payments for things like in energy you’re you’re

Spending too much of taxpayer money uh but then there’s also a democratic process through which states engage in reforms right so Martin shefter has written really extensively about about how middle and upper middle income groups can provide a constituency for reform right because they get tired of

Going to bureaucrats and we’ve seen some really great papers on how you want to cut out the middleman if I’m paying my bill I should just get electricity let’s keep it simple then I can move on to you know using my energy for other things um and then Rebecca we Shapiro and others

Have done this as well is that as middle classes grow in in Latin America and Argentina they have started demanding a lot more things as pay for use right and the puzzle here is is well laor is a prime example of a middle class that could be demanding something for pay for

Use and it’s not doing it um utility reforms are expected to engender trust and create transparency and yet many of them create clear losers who are low-income and working-class people so we can see that there’s this Clash right you can see that the vested interest in keeping this as a private SE

As a public State firm is probably from the working class and lower income people whereas the constituency for reform is typically from uh uh uh middle and higher income groups um I suggest the political impacts of privatization are under theorized uh and in particular I’m interested in this idea of the political

Influence in state Service delivery uh which has been characterized as perverse incentives by su Stokes and others um and what you find is that of course where there is this kind of perverse incentive for uh tech for politicians to keep something within the state there is a trickle down effect to the poorest

Voters who are typically their constituents as well okay um the other really important piece that guides my research is a recent um well I guess not you know several years ago but a really famous study by Bess and all in Bihar that shows that there is a higher tolerance for theft where

Something is considered an entitlement right particularly where it’s delivered by the state um and it’s really interesting because I think he correctly points out that it makes Elites tolerant of theft right so we’re not so concerned about whether lower inome groups are tolerant of theft because because they

Are you know presumably they are the ones participant in the theft it’s how people who stand to benefit from a pay for you system are also more tolerant of other people’s stealing um and this is particularly concerning for long-term resource scarcity especially uh water and energy needs and we’ve talked

Extensively about sort of the environmental and climate pressures that Pakistan is under um and privatization and I say this in air quotes because I’ve you know sort of done this very extensive study on privatization offers a potential solution um and in laor the privatization of leco has been debated

Several times okay so I’m going to skip over this um the thing I want to tell you is that we actually don’t know what Lor’s distribution looks like and I I can say that with confidence because the state-of-the-art in the room is here with uh IAL n’s work right and so this

Is a framework that I borrow from him and so what we are you know and hopefully we are sort of going to be collaborating on this one one of the things that we’re interested in is whether neighborhoods face disproportionate out outages because of the pressure that lore has been under to

Rationalize its electricity consumption okay so I have two hypothesis really simply I want to know whether people are tolerant of theft among others and then if you have a favorite political party say I’m a PTI voter and I give you a prime saying PTI is proposing these reforms that are going to make

Electricity pay for use does that reduce or improve your Baseline vote for them okay so really really simple um the data uh is simple 2000 uh household face-to-face survey conducted by Gallup in lore and it’s primarily focused on Urban lore but it is uh uh representative throughout lore which is

Important okay um so I want to show you a couple of descriptive pictures of what we found in our survey it’s a face-to-face survey that happened right before the pandemic so we’re very lucky with the timing of it um and we found that uh you know outages are fairly

Evenly distributed throughout right so I’ll show you sort of different the thing I’m also really interested interested in is how does billing look right so billing over here does tend to go up as you go into higher income categories right there’s some noisiness but really there is this sort of upward

Slope and so here’s the way I look at this in Karachi there’s a lot of noisiness around the middle income sector right whereas in lore it feels like it’s a much tighter confidence interval here this doesn’t really tell us much right these bureaucracies are

Both um you know doing sort of a lot of work in terms of trying to reconcile bills with usage this is an imperfect system but it’s interesting to see that this is a private bureaucracy giving you sort of variable bills even though you should be around sort of this line in

Terms of your your household consumption or electricity whereas the public bureaucracy seems to be doing you know slightly better I don’t know so we can debate whether how the public bureaucracy is performing versus this the really interesting thing and this is something that just came out recently

This is what outages look like in Karachi right so now I’ve done this survey for two rounds now we had one in 2018 we had another one last year with my co-author if you go to a higher household income level you will face fewer outages this is a classic pay for

Use right why because not only are you more capable and willing presumably to pay your bill the people you live amongst are presumably also more willing and able to pay their bills so if I’m in a good cluster that has been identified by K Electric as a good cluster they

Will favorably Target me with fewer outages so at the highest end over here you have like 2 hours even in the heat in the peak of Summer whereas over here an average of like about 7 to 8 hours of outage for the lowest of low does anybody want to guess what lore looks

Like any takers of like this what what what what do you think outages in laor look like a cross income Flatline amazing right so this is to me where I’m like okay this is this is incredible right and for for so many different levels I want to get into this

But this idea that you have a public bureaucracy that in spite of having tremendous pressure to rationalize Service delivery according to who is paying for it you have these vast differences and and I think part of what’s exciting about this work is that you can see the two largest cities in

Pakistan have radically different approaches to energy provision at the household level okay I want to get into this yes yes is not that’s right yeah yeah yeah so throughout a day perod which you don’t have energy available to you for whatever reason no it’s just sort of throughout a 24-hour period how many

Times does the light go in your house how many times is there an outage yeah absolutely yeah low shedding it’s low shedding okay all right uh so very quickly to what extent you agree with the uh leco has higher favorability overall as opposed to k electric and

That’s not surprising at all we can go into more details I do want to do a quick work on the experiment okay how should I do this all right so experiment number one I will read you a story from daily Jung an indiv an individ an individual was interviewed at Lore City

And had the following story my bill amount was very high I checked the bill emerged the metor reader had only shown peak hours reading um but a normal reading I took my complaint to letco so this is a scenario that we’re giving our respondents right I to leco several

Times to resolve my issue nobody resolved my issue so I just didn’t pay my bill okay I disconnected my meter and when the leco meter reader came to my neighborhood he threatened to report me and I paid him a bribe so this scenario of a person acting really badly right

And there’s a series of things that they’re doing which are clearly like not nice right and so we asked to what extent do you agree with this guy right so the Baseline is like to what extent do you agree with this person without knowing anything about them and then

After that we have a prompt we have some more information about who the person was who narrated this incident to the daily Jung and then we show them one of these two pictures so treatment one is this workingclass guy you know the sort of a blue blue color worker and

Treatment two is a SEF po worker right like a white collar person okay so what we find is that there is a lot more tolerance over here for the sort of for the white colar individual strangely no no disagree sorry so there’s a lot more tolerance for the for the blue collar

Individual because uh people are more likely to say they somewhat agree with what he’s doing compared to the group over here right so there’s sort of a significant change here what becomes interesting is does this vary with higher and lower income people um and definitely there is less tolerance

Amongst the higher income categories what you’re finding right so there’s a lot more tolerance for theft from blue color workers uh from low-income people and that’s sort of an intuitive finding I’m going to sort of skip over experiment two happy to talk about it in Q&A

Um I want to conclude by saying energy costs and access are relatively evenly distributed across income groups in the city uh I want to see that electricity is constructed as a right and an entitlement over here right compared to a commodity in places like Karachi Lor’s poorest residents feel mistreated by

Lesco but this is a good thing that means they feel entitled to better treatment by them right um and they have a strong strong sense of felt citizenship and access to government so before the discussion isal asked me like how are people so favorable of leco I

Said they’re not but I think that’s a good thing right I think that they actually have the sense of how they should be treated by them um and I think that this raises a really provocative question can we think Beyond binaries of State provision and private service provision is there something in the

Middle right given that lore has this higher to Middle income group that might be slightly more tolerant of theft from lower income people depending on their circumstances can we think of Creative Solutions to solving this puzzle of an energy crisis and also how to pay for it without burdening the taxpayer okay

Thank you very much look forward to your [Applause] questions thanks thank you arum and one of you will have to keep me honest regarding time but I would uh rather listen I can time you yeah I would rather listen to and talk about this topic which I’m interested in

Rather than myself but that’s that’s me um so there were several aspects of this uh paper which were near and dear to my heart but um I was particularly interested in and you know one of the things I said just at the end uh was that she went quickly through that data

But there was almost like a 75% favorable rating of leco among uh respondents to the survey which I was uh astonished by did I get that number wrong but it’s like you know it’s like a Pakistani National Pastime it’s like we’re going to watch the cricket complain about the cricket team losing

And then complain about the electricity situation when have you ever heard you know orary Pakistani sitting around saying positive things about uh Lesco and yet you know the other aspect of this which I thought irum showed really well which is that laori privilege is real right you’re getting far be

Electricity provision than the rest so in as much as people may have genuinely felt negative sentiments about the real impact of loadshedding um and their High bills which are also real they may there should also be some kind of awareness I’m not quite sure if you captur this

But like in the implicit rural comparison and this is there were points at which it was so nice to hear this paper coming after the discussions that came earlier particularly mish’s which she’s talking about a rural urban comparison about Villages is like what is that implicit rural urban comparison

Because leco also has a rural service area right and would we see similarly uh positive effects because the uh one thing that uh makes a point of noting in the paper is that it is officially stated policy that the Karachi approach of reward and reprimand I.E that if you

Don’t pay your electricity bill you will get lesser electricity Supply in terms of hours of load shedding in a day is the official policy across Pakistan but it’s also being contradicted by nepra multiple times however it’s less implemented and where it’s more likely to be implemented is a bias towards

Urban areas over rural generally speaking so rural areas will still definitely face uh more more load shedding but it’s not strictly implemented and certainly this socialization has never taken place that that arum talks about and so that part is fascinating because this idea of treating electricity as a right or as an

Entitlement and I think maybe there we need to think very precisely about what exactly we’re calling it because in the field work that I did with people uh particularly kibis and Islamabad so a different city but they were the only people who used electricity as a right they talked

About things as a right I I want an individual meter because it is my right right but nobody else used that language right they didn’t use that language and the the takeaway that I took from it because they felt like it was a waste of their time because that

Is not how you get anything done right how do you get things done this is the story that we heard uh earlier in the day that I’m going to get things done by using the the connections that I have through the Brokers that I’m connected

To to the people that I’m going to vote for and so on and the person that I vote for may or may not have a strong affiliation with the party identity right but that’s how I’m going to resolve the issues including electricity Supply and one one note is that you know

In the literature you mentioned Alicia Holland talking about forbearance but I I think it’s correct to say that the Mayors in those cities actually have responsibility over electricity Supply which is not the case at least in the strict sense it is not the responsibility of any political actor to make decisions concerning electricity

Supply let alone concerning uh um whether or not there should be appropriate load shedding all of that is through this informal set of relationships of influential people uh uh making connections and such but um to follow on with this topic of uh uh you know socialization towards electricity

Theft uh in lore and elsewhere because it is very much Karachi versus the rest of the country uh in that comparison there’s a line from Muhammad hanif’s um Our Lady of Alice butti where uh the the narrator is talking about someone who himself as a gangster and the person

Who’s responding says no no he’s nothing he doesn’t even steal and electricity doesn’t count right you can’t be it it’s not a mark of of theft because it’s so widespread in these attitudes towards electricity I thought that came out very nicely in in in your server results where people actually

They were like yeah well you know that’s what you got to do um and the idea of going in a group uh I think was also very consistent with other things that I’ve seen about like uh to seek a Ive redress is the way to gain attention for

The for the things uh of one’s concern um and uh perhaps lastly this is nice connection again to Karachi where the the Karachi drive to collect arear on electricity builds goes by the name of Sandi right that you hold your head up high literally as a matter of

Self-respect in terms of this individual relationship with the state versus the very different sets of relationships um that we’re uh seeing um elsewhere okay um thank you for uh fantastic paper and uh literally next week I’m going to go to AA to try and see if I can get the

Special data that she needs to be able to connect the survey results to the actual performance uh in those Let’s Go distribution areas but uh now we can open the floor to questions from the uh from the room thank you uh my name is manj from psychology department Punjab University

And you may wonder what I have to do with electricity apart from using it in DHA well I did two trainings for k Electric first for the consumer violence they were experiencing in 2014 just into privatization very recently and then 2016 on sustainable safety culture so why did they need a

Psychology team apart from the engineers and everything they were doing consumer violence was baffling yes so they attacked IBC and the but at the same time Karachi political governance and crime culture and uh peace was also lower than what it was when we went again in 2016 so one thing that I

Learned was that the highly influential and politically connected citizens who had not paid their bills also responded with violence to the IBC teams that turned up to collect bills at their doors whereas the poor kunda people were the whole city and ktic itself was more tolerant of the lowincome consumers um

So the contrast that you were drawing between laor and Karachi as a Lor consumer going into Karachi K Electric is a very professionally run organization highly paid paid upper portion and then the linemen are highly distressed that be although the privatization has happened it has not trickled down to the salary increase of

The the the lower stret up so that discontent was feeding into so any you know uh the contrast between laor consumer culture and Karachi consumer culture uh was uh a very stunning Discovery for me as a social scientist so I would uh sure like you to answer

That this the situation improved as it seems in 2018 and uh since 2016 let’s take one or two more questions uh thank you Professor hether um I had a question about the the social socialization aspect that you were talking about so you were saying that in Karachi since the privatization of K

Into K Electric um if a person is not getting electricity in their house they would be blaming perhaps their neighbor for not paying their bill which is why they’re losing out their electricity I understand that people would be intolerant of electricity theft right but how does this Dynamic work that

Because my neighbor didn’t pay for their bill I’m losing out on my electricity great question all right great thank you um let’s see so you know I want to actually take my discussing questions first thank you that was super helpful as always um I think there is an

Implicit rural comparison here uh I would have liked to sort of I think because we only had 2,000 households to deal with I wanted to make sure I was covering income groups within lore first right and as an extension I think I think covering the rural is also

Extremely important but I think I can tell you that if I just took the case of urban Punjab even the comparison between low-income people in Karachi and low-income people in lore is to me extraordinary I mean look at this right you have 100 degree weather across this

Country in Peak summer and one person who earns 20,000 rupees but lives in Karachi Pakistan is subject to 8 hours of outage in a normal day right by design by the way this is not a flaw this is what the company says they are doing it is on their website right it’s

Not something that oh we don’t know what no they know that they are targeting the poorest of the poor with the most amount of outages and you can talk about you know energy uh you can talk about sort of economic efficiency and things like that there is an environmental justice

Issue here at play right which wasn’t as there as much maybe when I was starting my field work 2023 we need to talk about what is happening here what is life looking like at you know 40° fah inside the house house with 20,000 rupees in

Your pocket okay over here you have a a lowincome person who is relatively sheltered from the worst of international oil price shocks from sort of the the pressures of technocracy like so even if I just kept this to an urban story L is what I’m saying there’s a there’s a really interesting contrast

Here right and my job as a political scientist to tell you if I’m a poor person living in Karachi versus me as a poor person living in lore what is my idea the state and what is my idea of myself how do I feel when my government

Ensures me that in the worst of Summer I still have as much electricity as this guy sitting here at 90,000 rupees versus over here where I can see that contrast this is this is Ian I’m I’m not exact this is grounds for a revolution man I

Mean like the people are going to break down your house one day right and say we’re stealing your ACS so this is to me a really really important comparison and I hope my book can get this out which brings me to SOB’s really great question

Um is that is that you why would I blame my neighbors K Electric has pulled off a brilliant trick it has convince people that they are solely responsible for their electricity this is what the technocrats told them to do this is what our government through the nepra ACT

Told them to do they’ve done it okay I can tell you that guaranteed because you are made to feel like you are personally liable for your electricity your electricity is still provided collectively so lal’s really great book that came out last year shows this as discoms we don’t have the

Ability to shut we don’t have the ability to shut off electricity at the consumer level or even Target electricity at the consumer level what we are doing is creating Aggregates of 200 300 400 households in an urban area that are all provided by the same feeder

Then what you do you take an average you say Okay 40% of the people here pay their bills this is a medium loss Area 3 hours of outages 6 hours of outages right then you make your calculation over here at this send like I said this

Is sort of you know my my mom’s apartment in Clifton she and her neighbors are all you know salaried people they all pay their bills on time two hours of outage her her cleaning lady who comes in she’s paying her bill every single month but two doors down

That person has lost their job their daughter’s in the hospital whatever does not pay the bill the entire neighborhood get shut down this is collective punishment it works extremely well in industrial area so this is where they got the policy from right they got it from uh surveys and studies that said

That if you have 100 factories the way to convince all of them to pay their bill is that even even one person doesn’t pay their bill you shut down everybody and then you get them to police each other you can’t do that with 500 households that is externalizing the

Cost of information onto the end consumer so what does this end consumer do very rightly says it’s my neighbor’s fault right these he or she is working with the information that they have what is this due to politicization it reduces your ability to act collectively there

Are some small sort of areas in Karachi that still do have an anjuman or someone that they collectively act around but it has reduced the ability to act collectively and so uh you know dos up to your question I agree Karachi is a terrible violent place there thugs there

Criminals all sorts of gundas there I’m from that place the problem is if you are not giving if you’re putting people in this situation right and you’re saying okay look you’ve got to pay your bills but but the bureaucracy has been reduced in capacity it’s very professional it’s a third of the force

It used to be when it was a state owned electricity firm by the way I shouldn’t fudge numbers 36% recovery it’s not great right like pre-privatization we were not making any money from karachi’s electricity I think the elite always find a way to make the best of any situation they’re under whether it’s

State whether it’s private it doesn’t matter to me that’s a less interesting question than how do we provide some kind of equitability the other end of the spectrum that’s really interesting these guys are still here they could be here you know the elite in lore could be

A really sweet deal where they get 2 hours of electricity they just need to privatize and yet somehow they’re still at the same level of this person that that’s a mystery to me and and you know I’d love to maybe after this I’m can you

Know you guys can come up and give me your theories of why the elite in laor haven’t actually petitioned or or asked their governments to give them a privatized uh service firm thank you so much this is [Applause] great