Syndicate content

Disrupting Low-level Political Equilibria

Shanta Devarajan's picture

Absentee teachers, negligent doctors, high transport costs, missing fertilizers, and elite-captured industrial policy all stand in the way of poor people’s escaping poverty.  While the proximate reason for these obstacles may be a lack of resources or an erroneous policy, the underlying reason is politics. Lawmakers meet during a session of Parliament in Accra

- In many developing countries, teachers run the political campaigns of local politicians, in return for which they are given jobs from which they can be absent.  The situation can be described as an equilibrium, where the candidate gets elected and re-elected, and teachers continue to be absent.  The losers are the poor children who aren’t getting an education.  The equilibrium has no intrinsic force for change, especially if, as in Uttar Pradesh, India, 17 percent of the legislature are teachers.

 - High transport costs in Africa are due not to poor-quality roads (vehicle operating costs are comparable to those in France) but to high prices charged by trucking companies, who enjoy monopoly power thanks to regulations that prohibit entry into the trucking industry.  High transport prices and monopoly trucking profits are an equilibrium. In one country, the President’s brother owns the trucking company, so prospects for deregulation there are grim.

- Several countries subsidize fertilizer, sometimes to the tune of several percentage points of GDP, only to find that it fails to reach poor farmers.  Thinking that the problem is the public distribution system, some governments have tried to use the market to allocate fertilizer, by giving farmers vouchers that they can redeem with private sellers.  A scheme in Tanzania found that 60 percent of the vouchers went to households of elected officials. When subsidies are captured to this extent by political elites, their reform will be resisted—another equilibrium.

- Invoking the experience of East Asian countries, governments selectively protect certain industries to achieve “structural transformation”.  Unfortunately, instead of picking winners, they pick family and friends.  In a forthcoming paper, my colleagues Bob Rijkers, Caroline Freund and Antonio Nucifora show that the 122 firms associated with former President Ben Ali’s family in Tunisia, while producing 3.2 percent of output, earned 21.3 percent of net profits in the economy.  The sectors in which these firms operated received the most favorable regulatory treatment.  Worse still, some of these sectors produced non-tradable inputs (such as transport and telecommunications), whose inflated prices undermined competitiveness (and job creation) in Tunisia’s tradable sector.  Inasmuch as Ben Ali was in power for 23 years, this situation was an equilibrium, one that was disrupted in December 2010.

If these obstacles to poverty reduction are the result of a political equilibrium, how can they be removed?  Money alone is unlikely to dislodge the equilibrium.  In fact, it may increase the rents (building roads enhances trucking profits without lowering prices).  Even money conditioned on policy reforms may not do the trick.  For if there is a political benefit to the distortion—be it a fertilizer subsidy or protective tariff—why would a politician agree to remove it, even for some financial assistance?  As one politician said to me, “If you have a choice between a $100 million loan and winning the next election, which would you choose?” 

That leaves the last refuge of scoundrels economists, namely, knowledge.  Perhaps undertaking a study of the costs of resource misallocation from industrial policy, or the beneficiaries of fertilizer subsidies, or the extent of teacher absenteeism, will convince the government to reform.  But if the distortion is a political equilibrium, with people in government, including politicians, benefiting from it, what is the incentive for government to follow the recommendations of the study and upset the equilibrium?

The only way these equilibria will shift is if the incentives facing politicians change.  That will happen only if politicians perceive that there is a shift in public opinion, which could be reflected in future elections or protests in the street. Public opinion may shift if the public is better informed about how much they are gaining and losing from current policies.  So there is a role for knowledge, but it should be knowledge to inform the public—not just the government—about the evidence, so that they can bring pressure to bear on politicians and, possibly, move to a new equilibrium.

This shift in approach to knowledge is not innocent.  It may involve undertaking studies that the government opposes. But a case can be (and has been) made that if the purpose of the study is to collect evidence on an important policy problem, it should be undertaken despite a government’s objection.  This approach is also not easy. Disseminating the study to a group of scholars in the capital city is not enough.  There should be ways of getting the message and the findings to poor people in rural areas.  After all, the study is about their lives. 

Bottom line: If we don’t do things differently, we risk leaving poor people stuck in the low-level political equilibrium they’ve been caught in for decades.


Submitted by James Abola on

Come on, if teachers were so politically powerful they would be the highest paid profession.
For many of the nations that got independence in the period 1950s to 1970s at the time of independence teachers were the educated people in the populace and the profession also gave them communication skills which is useful in politics.
In my country Uganda many teachers became politicians and even nowadays there are many politicians that were formerly teachers.
Uganda suffers from the problem of absentee teachers and doctors; I believe many of the countries you reviewed/studied have those twin absentee problems. The cause for teacher absenteeism is more due to the pathetic low pay given to them. Ugandan teachers are paid about $100 a month, it is impossible for a person live alone a family to survive on that kind of pay with supplementary income.
In the case of Uganda the sitting government with the help of World Bank and IMF restricted investment in education to building school blocks and furnishing and equipping them while freezing the pay of teachers. If teachers where so politically powerful as you suggest then we would be seeing them becoming the most highly paid profession. The reality though is different

Submitted by Shanta on

James: Thanks for your comment. I'm afraid the evidence does not support your statement that "the cause of teacher absenteeism is more due to the pathetic low pay given to them." All the statistical work I've seen finds no relationship between pay and absence rates. [In fact, in some cases, you find a positive relationship--higher paid teachers are absent more often! The reason is that these are the senior teachers or the principal, who is monitored even less than the junior teachers]. Another piece of evidence is that public-school teachers (which is where absenteeism is highest) get paid about two or three times as much as teachers in low-cost private schools. I'm sure in Uganda, as is the case in every country I've studied, when there is an opening for a public school teacher, there is a long queue of people applying for the post. It's hard to sustain this fact with the statement that teachers are underpaid.

The underlying cause of teacher absenteeism is the way in which teachers get paid: their pay is independent of whether they are present or not. There is some evidence that if teachers are paid a bonus for being present, then absenteeism goes down. So the solution may be to change the way teachers get paid, and make presence in the classroom a condition.
Finally, teachers are not the highest paid profession because there are other professions (including doctors) who are even more powerful politically.

Submitted by Thabo on

James does make a relevant point- the actual pay does also matter. The fact that there are queues for the job, simply indicates the scarcity of work and has nothing to do with whether the pay is decent or not. Most rational decision is to get the job and then to go do other work to make up the low pay.
The argument is another example of shortsighted one variable arguments so beloved of economists. This one is a complex problem - look at the pay, look at way to incentivise teachers, look at way to improve their engagement - the stick approach you recommend may get a quick blip up but will not result in improvement in the education system

Submitted by Shanta on

Thabo: Thanks for your comment. I agree that it's not a one-variable problem. In fact, I was questioning James' one-variable solution, namely, to increase teachers' pay. There is no evidence that that will reduce absenteeism because absenteeism has multiple causes. I would also agree that simple solutions, such as tying teachers' pay to attendance may work in the small, but are not likely by themselves to transform the whole system. I was using that example to illustrate that the solution of simply raising teachers' wages will not work.

The reason teachers are absent is that they work in a system where being absent is acceptable, possibly due to the low pay they receive, but also due to the lack of accountability to the students and their parents. To improve student learning outcomes, we need to transform the whole system, so that teachers get adequately compensated, but it is no longer acceptable for them to be absent 25 percent of the time. In short, we need to disrupt the low-level equilibrium that was the subject of the original blog post.
This is turning into an interesting discussion.

This is really interesting. Knowledge is power and it is the only way to hit politician's weakest point i.e. the need to get re-elected.
An interesting point you mentioned is how to disseminate that knowledge, I have always thought radio is best but what other suggestions would you have.

Submitted by Shanta on

Samuel, you are right. Radio is a powerful medium (and one that is probably under-used). Other ways include disseminating messages by text message. People receive a text message with a quiz about the relevant subject (e.g., guess the level of teacher absenteeism in your district). If they get the right answer, they get an additional 25 cents on their cellphone. Wrong answers earn 10 cents.

Submitted by CAA on

Samuel and Shanta, Radio and mobile phones are an excellent idea. However, how would you ensure that the very same politicians do not use this media to spread their own 'propaganda' so to speak?

Submitted by Shanta on

Samuel: The text messaging idea was used (by us) in South Africa a few years ago. I'm sure the idea has been refined and improved upon since. As for the problem of politicians' using it to spread propaganda, we have to rely on the wisdom of crowds. As the number of people with access to cellphones increases, it becomes harder for one voice to dominate. Shanta

Submitted by Ramon Ynaraja on

Excellent entry! Yes, we are approaching the bottom line of development practise; funding is a necessary element, conditionality does not necessarily work, and breaking the equilibrium is resisted domestically in direct relationship with the volume of the implicit benefit that the elite gets out of the equilibrium. And yes, the only exit to this conundrum is transparency in the form of producing and disseminating the assessment of the equilibrium in terms of “who gets what”. But if you think that in Mozambique the President will allow information to be published (let alone disseminate across the country) on e.g. why unitary maritime transport costs are twice as high as in Asia (he owns one of the two companies operating the domestic market and he is a shareholder in the other one) you are either mad or… and Economist.

Submitted by Shanta on

Ramon: Thanks for your comment. My question is: How can the President (or anyone else) stop information from being published and disseminated in this modern age of social media? In fact, you just disseminated the information about maritime transport costs on my blog. Regards, Shanta (the mad economist)

Submitted by Gituro Wainaina on

Well thought article. If I was a politician, I would find a way of getting the $100 million loan as well as winning the next elections - that is how they are. They want it all to themselves - poor people are not in their ladder.

Submitted by amouzou on

I like this blog. Very interesting. Thank you very much Mr. Shanta for your courage and for your sharing. Yes, the “public distribution system” that explains the marginalization of local poorest people is a greatest challenge in local and international sustainable development arena.

Submitted by Arvind Nair on

Great blog post! Hopefully the WB will not be averse doing more such information work, with the view that the clients are ultimately the people and not the government.

Another way to try to shift the political equilibrium (not relevant for the World Bank) is to do the hard yards and enter directly into the political fray.

This is risky but this is what the anti-corruption movement in India, that led the street protests in Delhi last year, is trying to do by competing in upcoming state elections. This is an interesting experiment in engaging directly with voters, especially in candidate selection, as a recent piece in Tehelka ( and a piece by Pratap Bhanu Mehta ( highlights.

Hopefully it won't be the case of old elites being replaced by new ones (if this party comes close to power), and it may disrupt India's cozy elite equilibrium and start moving it towards a genuine change in the way politicians engage with their constituents!

Submitted by Kwabena Otoo on

I am happy that people like Shanta are coming home to the crust of the problem in developing countries especially in Africa. The problem, as has been stated is POLITICS. Unfortunately, past approaches to dealing with the African challenge have tended to rely solely on economics and economic ideas. I agree that the politics has to change before anything else. I also agree that knowledge is key in breaking the dysfunctional political equilibrium that keep large numbers of people in a state of perpetual poverty. But what will be the results when a knowledgable population comes up against BAD leaders who are bent on perpetuating the status quo through manipulating the electoral system? I'm of the view that for nearly all the countries in Africa we are unlucky to have bad leaders whose behaviour cannot be changed simply because the people have become "knowledgeable" about their modus operandi and about the policies and practices that keep them in poverty. 'Good' people have strangely been unsuccessful at politics probably because the tend to play the game overly too fairly. Before I propose my own sets of things to be done as part of the way forward, let me continue with my reflections. I will come back!

Submitted by Shanta on

Kwabena: Please do come back, because you're raising some important points. We know very little about what attracts certain types of people to run for political office. This is an area that is in need of further research. Let's keep working on it. Regards, Shanta

This is absolutely correct. Thank you Shanta. We cannot do real development -- we cannot go beyond tinkering at the margins of political economy systems that keep large numbers of people poor -- without changing the political incentives that rulers face. Advising them, or paying them, to do the things they should do anyway because these are good for their economies and societies, regardless of external support, is naive. We've known about the importance of primary education, health, sanitation and transport for more than a century, but still countless governments fail to provide them. That must be because politicians have powerful incentives to do otherwise. Understanding and then overcoming such incentives is a tough and possibly dangerous job. But without it, countries will continue to fail to develop.

Submitted by Wolfgang on

One of the fundamental shifts that could change the political equilibrium is demographics (and it is already happening). The logic is as follows: The low-level equilibrium you are describing is reinforcing political stability at the price of economic opportunity. This system created enough rents for the groups connected to the leading group in power (and could switch back to another group after a power change). Today, there are just too many people in each of the groups and the rents typically not enough to satisfy even your followers. This coupled with urbanization and an emerging middle class is the environment where knowledge and data can be more effective.

Submitted by Shanta on

Wolfgang, great point. This may explain the improvement in macroeconomic policy in Africa over the last decade. In fact, I recall a paper by Devarajan and Fengler in Foreign Affairs making just that point! Cheers, Shanta

Submitted by Shanta on

Dilawar: Thanks for your comment. You may want to look at Merilee Grindle's book, "Against the Odds" that documents the role of teachers in the political process in several Latin American countries. There is also a forthcoming paper by DFID/World Bank (Harry Patrinos and Tara Beteille are among the authors) that surveys the evidence around the world. The case is quite compelling! Shanta

Submitted by Munir Ahmad Kakar on

The author has convincingly attempted to explain the unholy alliances formed primarily by the politician for rent-seeking which is the ultimate outcome that every agent is having his sights on. I think the economic policies in developing countries are framed to protect the special interest groups at the cost of neglecting the downtrodden. In Pakistan we are witnessing the worst form of education and health apartheid as the elites have their own educational and health institutes in private sector. The poor children enrolls in govt schools where teacher absenteeism is endemic, curricula is outdated and religion-focused and cheating is rampant. The same is the fate of hospitals where doctors are absent and life-saving drugs and vital test are unavailable. Feudal elites have very successfully thwarted attempts to undertake agricultural reforms. Bank loans in billions have been written off and political leverage has been obscenely used in issuing SROs to benefit politicians and their lackeys. This is the kind of top down imposed political and economic equilibrium that has played havoc with the country. Though millions are bearing the brunt of it, the public opinion here has been very cunningly diverted to inconsequential matters to perpetuate the ascendance of rent-seeking.

Submitted by Shanta on

Munir, thanks for those comments. The only qualification I would make is to your statement that the poor send their children to public schools. It turns out that in Pakistan (and elsewhere), the quality of public schools is so poor that even the poor send their children to low-cost private schools. You might look at the following website: Shanta

Submitted by Ronald Mukasa on

Thank you for blog, I believe that parents are most interested in their children's educations. What are your thoughts about making the parents part of the accountability system?
If teachers were directly reporting to a committee with a significant percentage of parents, I believe they would take there jobs more seriously. I also think that parents should directly earn a right to demand for services by making a direct contribution to the school over and above the government contribution.(This could be as simple as providing money, food, firewood or scholastic materials) to the school. I believe that education is very important to all parents and this is confirmed by the fact that most parents would quickly move their children from public schools to private schools as soon as their incomes increase. In my home country Uganda, free education has been interpreted as removing parents responsibility to provide education to their children and handing it over to government. I think there should be a hybrid system that promotes more accountability.

Submitted by Shanta on

Ronald: Thank you for those thoughtful comments. You are absolutely right. No one cares more about a child's education than his or her parents. The problem, as you note, is that many parents cannot afford quality education for their children. This is the rationale behind public financing of education (there is also an efficiency argument, based on education having an externality, but the evidence is pretty thin). The problem is that, when government finances and provides education, then accountability to parents breaks down, and the children no longer receive quality education. One possibility is to separate the financing from the provision. Have the government pay for the education, but contract out the provision to NGOs, private sector or the government, with strict performance benchmarks. Another is to give the parents vouchers, so they can decide where to send the children. Finally, as you suggest, having the parents contribute to their children's education is a possibility, although as you know, this is quite controversial (see the debate about "user fees" in my former blog, Africa Can). The bottom line is that achieving any of these reforms is hard because of the power of teachers unions to block reforms that may undermine their rents. So we need more voices like yours, who see clearly the problems created by public financing and provision of education, to lobby for reforms that improve the education of these children. Shanta

Submitted by Nadia F. Piffaretti on

I agree with most of your analysis. I'm not sure however we should be so unsympathetic to low level equilibrium governments. You ask "if this obstacles to poverty reduction are the results of a political equilibrium, how can they be removed?" However, can't this very same sentence be applied to other cases, such as the current impasse on the budget, obamacare, and the the debt ceiling ? We are of course all distraught by this latest showdown and its potential effects. However, would we ask ourselves "how can they be removed"? We likely need a stronger dose of political realism in these countries, letting feelings (often of frustration) out of our analysis, and simply look at incentives.
Shanta, you are dead spot on in pointing to knowledge to the broader society as a key instrument - it drives the DEMAND side of accountability. PETS in Uganda did just that. But when the demand for accountability is lower, it may be mistaken to put exclusive blame on politicians for not offering enough accountability. It liberates our frustration for low level equilibria, but not change the dynamic on the ground.
Knowledge is likely to be more efficient in changing the demand for accountability than the offer.
Thanks for the good blog.

Submitted by Pietronella van den Oever on

When I was in primary school in a small village on a river, (without running water) now 60 years ago, the headmaster had a large chicken farm as a side activity. He used the oldest and strongest boys in the school to do the work on his farm. Needless to say, this headmaster was not very much focused on his pupils, except to use them as non-paid labor, and the pupils who worked for him did not learn a whole lot, except how to take good care of chickens and eggs. The parents were very much aware of the situation, but did not dare to say anything, out of fear of retribution against their children. The situation gradually improved with a land reform program, including a new bridge across the river, and with a strong adult education component - a precursor of community driven development. This contributed significantly to disenclavement of the village and brought people into permanent contact with the outside world. Readers will certainly have thought that I went to school in a developing country? No, I was born and raised in the Netherlands! People often forget that it is not that long ago that there were large pockets of underdevelopment and poverty in the now highly developed countries. The situation in Africa will guaranteed change when people (dare to) hold their government officials accountable, and have the courage and knowledge necessary to enter into a debate with them. After working for 40 years in the African continent I am still very optimistic about its future, because the pent-up energy and intelligence, combined with the rapidly improving opportunities to communicate with the outside world, will accelerate change, and cut through the various equilibria.

Submitted by Shanta on

Pietronella, thank you for that inspiring story. You are right that the situation we describe in developing countries also existed in present-day developed countries not so long ago. The question is: how did the system move from one equilibrium to another? How did Dutch society go from one where teachers' making their students work was accepted to one where such behavior is not tolerated? If you have written or read anything about this subject, I'd be interested in hearing about it. Regards, Shanta

Submitted by Anonymous on

low-level, Tunisia, poor Uttar-Pradesh? I would like to see some of these figures from other levels and countries

Submitted by Shanta on

If you click on the hyperlinks, you will find data on other countries in the phenomena I describe, such as teacher and doctor absenteeism, political interference of teachers' unions, etc. In addition, Ray Fisman has written on crony capitalism in Indonesia, along with the examples of Egypt and Tunisia described in the post. Shanta

What about the World Bank Strategy?

Shanta, you post very rightly identifies the political economy problems that keep services from being delivered. I wondered your view of the new World Bank strategy and its focus on the 'science of delivery'. If your view of the political economy is correct, that mean all the science of "how to" effectively deliver education (for example) will be irrelevant because local political actors (low-level as you call them) will always subvert the design of any solution by gaming the system, undermining accountability, or falsifying records and documents. How do you think the World Bank's work can change to address this systemically?

The strategy at its core is a technocratic exercise which studiously avoids politics. The very interesting interview with your president ( made it very clear that he wants to Bank to steer clear of politics:

"We're going to say 'Okay, so what would you like us to work with you on?'"

That means they aren't going to ask the World Bank to help them upset their own self-enriching equilibrium, as you point out. This makes the Bank's strategy seem to be pointing the institution to continued ineffectiveness.

Submitted by Shanta on

Peter, thanks for your thoughtful question. First, the science of delivery is at least as much about politics as it is about technical solutions, as I have pointed out elsewhere Secondly, as for the World Bank, the only thing the Articles of Agreement prevents us from doing is intervening in the politics of our member countries. There is nothing preventing the Bank from collecting and analyzing evidence on service delivery, and sharing that evidence with the broader development community. Indeed, I would say that we have an obligation to do so. Finally, in terms of the World Bank strategy, while the Bank's financial relationship is with sovereign governments, its knowledge relationship could be with anybody. This is why the Bank's strategy places great emphasis on our knowledge function, as well as citizen engagement with that function. Shanta

Shanta, your original post, I thought, was along the lines of my worries. Namely that the "science of delivery" (or "deliverology" in your older post) explicitly ignores politics. You wrote of deliverology: "Both the diagnosis and proposed solution are compelling. But there are at least two other important elements of service delivery that are particularly salient in education and health." The word "other" suggests that back in June at least you didn't consider that deliverology sufficiently took account of clients or politics. Have you changed your thinking?

I think more data and feedback mechanisms on service delivery will be useful and helpful of course. Particularly for pushing accountability in public systems (which has been the crux of the debate in the UK about DFID spend on education and whether it should go for private schools or boosting accountability in public schools - see for example

Some of the worry about the new Bank strategy seems to be on how it will tilt on that phrase "transformational engagements". There is fear that it could go the wrong way. A ground breaking accountability mechanism that boosts public sector education could be "transformational". But Bank staff might also view a massive export-oriented infrastructure project that benefits multinationals and existing power holders/elites as "transformational". Those two projects would have massively different impacts, and the worry on the strategy is that the Bank's inherent technocratic bias means that challenging the power holders in country governments will be very challenging - meaning the selectivity of the Bank will go in the wrong direction.

Submitted by Shanta on

Peter, I did think that deliverogy, as described by Sir Michael Barber, did not sufficiently incorporate politics. That's why I wrote the post. With the application of the "science of delivery" to development, we are shaping it to incorporate the latest thinking on the subject, which includes politics. Finally, by "transformational interventions" we mean those that will have a sustained impact on the welfare of the poor, so projects that simply enrich multinationals, without say creating jobs or improving service delivery, are not included. Shanta

Submitted by Hodjat on

An excellent and thoughtful post and follow up discussion! To use the power of knowledge and public awareness to gradually shift the political equilibrium in a direction of more accountability may be the only reasonable alternative to costly and bloody revolutions. But the challenge remaining is getting the message to the deprived (impoverished economically or politically). It would be great to hear about ways the new communications technologies have been utilized to take the message to rural areas.

Submitted by Jonah Rexer on

excellent points have been made in this discussion. However, I find that dissemination of information / transparency as a cure-all to these toxic political equilibria have been dramatically oversold, particularly by the Bank. Dissemination of information to the rural poor, even in a cost effective way (which in itself is a huge challenge) is no guarantee of effective political action, and therefore of shifting political incentives. Lots of political economy literature (from both developed and developing country contexts) points to the fact that voter incentives are often unchanged by information. In fact, given the persistence of patronage and ethnic-based voting (my knowledge is of the African context), people will often continue to vote for the same MP even after this person has been outed for corruption, mismanagement etc, the logic being that every vote in a patronage system is a medium of exchange (votes for school fees, medical bills, etc), such that private goods outweigh public goods to the voter, making collective action incredibly difficult. In fact, if given the choice between an MP beholden to powerful interests and one who is uncorrupted, most voters will choose the former because he clearly has the resources to make good on patronage promises! A very tricky situation indeed...

Add new comment