On Letting Go: Centralization vs Information Aggregation

By Aarushi Kalra

मैं भी मुँह में ज़बान रखता हूँ 

काश पूछो कि मुद्दआ क्या है  

I too have in my keeping a tongue, a mouth

Only if you asked what the matter is

-Mirza Ghalib

 

Set the stage: a round table and three chairs. Enter: Minister, Economist and Bureaucrat (The words “well-meaning” stamped on their foreheads). Lights. Action.

Bureaucrat: The Paharia tribals of Godda district in Jharkhand don’t have bread.

Minister: Let them have cows!

Economist: A Paharia must always pay his debts[1].

As the Policy Makers from Delhi travelled to Godda with the baton of prosperity, they dreamt of the milk markets that were to emerge. What they didn’t know was that our guys didn’t consume cow milk, but have had a taste of beef (Sainath 1996). Contrary to the expectations of a sustainable stream of income, an unsustainable stream of debt was born.

This is not to doubt the intention of our lead characters; they’re well-meaning individuals by our admission. This illustration only highlights the consequences of having an ill-informed decision maker at the helm of affairs. The question that we would like to explore in this essay is would the Paharias have been better off if they were in-charge of their own concerns.

Crawford and Sobel (1982) brought to light the problem of information asymmetry between the decision maker and her biased adviser, who is privy to the information required to make the right decision. The result is that only coarse information about the state of the world is conveyed, when both the sender and the receiver could be better off if information was truthfully conveyed. Dessein (2002) demonstrates the gains from delegating authority to the agent in possession of information, given that suitable limits are imposed on the agent’s bias. Delegation can be read as MK Gandhi’s ideal of self-sufficient “Village Republics” (Constituent Assembly 1948) which could be achieved by empowering local village councils (Manor 2010). This ideal views the Panchayati Raj system as the most efficient unit of governance, enabling recognition and aggregation of information about preferences over public goods.

Ambedkar’s Suspicion

Dr BR Ambedkar, on the contrary viewed village level hierarchies as an impediment to reliable information aggregation, “The root notions of democracy, on the operation of which alone self-governing India can be safe for the masses, run counter to all the ideas which for thousands of years have formed and do form even today the common stock of their beliefs” (Ambedkar 1930). In other words, if the local leaders’ biases are too large, decentralized decision making may not deliver its promise of efficiency. Capture of local governments by men of dominant castes continues to exclude many voices from decision making processes, delivering public goods that serve towards the “aggrandisement of a class.”

That effectiveness of participatory mechanisms may be clouded by structural inequalities of power (Pai 1998) is plain to the casual observer. My first encounter with a Sarpanch Pati was at Achanakpur-Bhatapara, Rajnandgaon district, Chhattisgarh. Our team of student volunteers never met the Sarpanchni of the village, but was often greeted by her husband, who “obviously knew better.” If the Patis are here, can the Betas be far behind? Rajo Devi’s case study illustrates how patriarchal family ties are manifested in the public and political spheres. Rajo Devi, 60, was an illiterate member of Gram Panchayat- Kutel in Karnal, Haryana. Reportedly, she didn’t care for the position “as long as she was not required to take any responsibilities.” (Tekchandani et al 1997) She was ignorant about the functions and roles of the Gram Panchayat and directed any strenuous questions to her nephew, Prithvi Chand.

The Median Voter Theorem tells us that the local government, elected via voting, would implement the preferred policy platform of the median voter. However, it is puzzling that this theory does not predict the observed clustering of the Panchayat’s preferences towards the preferences of the elite of the village, away from the middle-road. Hirschman (1970) explains this phenomenon through the considerable influence yielded by voters with “extreme” preferences “via voice.” Voice, as defined by Hirschman, is “any attempt…to change, rather than to escape from, an objectionable state of affairs…through various types of actions and protests, including those that are meant to mobilize public opinion.” Here, we look at Elite Capture as a function of Voice: by the identity of individuals who are able to exercise it. The ability of elite voters to inflict actual losses on the candidates if they promise to move “too far to the centre” (Hirschman 1970) explains why Panchayat’s preferences reflect preferences of the elite.

In Banerjee and Somanathan’s (2001) specification of Voice as the decision of a random individual to convey her signal truthfully, agents may choose to suppress their signal depending on the cost of communication. If, however, the cost of communication is different for different groups in the population (eg: men have a lower cost of communication than women), the groups would have different ability to exercise Voice. Women, having high cost of communication owing to their secondary status in domestic relations, will choose to suppress their signal about the state of the world (in the truthful reporting scenario). This may explain Chattopadhyay and Duflo’s (2004) findings in Rajasthan: no statistically significant difference in nature of public goods provided by women led Panchayats.

Ghoonghat or Zabaan?

Granted, that participation is defined by individuals and their ideological and cultural background. Our submission, however, is that participation can in turn strike upon ideologies and cultures. It is worth noting that the cost of communication for women falls if the Sarpanch is a woman, as is indicated by varying attendance of women at Gram Sabhas depending on the gender of the Sarpanch (Das 2014). The 73rd Amendment of the Constitution, by way of reservations for women in the Panchayats, endeavours to encourage participation of ‘voiceless’ women. Additionally, if reservations were to create incentives for women to enter the contest on unreserved seats, these new entrants may cast a permanent dent on the cost of communication for women. Such Reservations enable women to exercise Voice effectively and to influence outcomes through participatory action. This may be supplemented by what Farzana Afridi (2017) called “Nudges and Labelling,” by steadily jostling voters out of their comfort zone, branding their world view redundant and labelling the woman’s opinion  valuable. Additional safeguards against the threat of elite capture can be erected by way of training programmes for Panchayat members. We, therefore, view participatory mechanisms as a means to reallocate Voice and to moderate the influence of the elites over a period of time.

Our faith in decentralized and participatory decision making is reified by personal encounters with women leaders, functioning in restraining societies. Savitri Devi, proud mother of a woman Sarpanch in Jawaja block, Ajmer district, Rajasthan, recounted how her daughter’s “voice echoed in a room full of men, despite her ghoonghat.” (Personal communication 2014) Santosh, an SC woman was member of Gram Panchayat Samora in Karnal district of Haryana (Tekchandani et al 1997). “Member Saab” had a clear agenda: repair of streets, installation of hand-pumps and a Chaupal for her community. Her husband never represented her at Panchayat meetings and disposed the household chores while she was away. She was a proponent of reservations for women as this way “women will be able to come out of their homes, know the law and see for themselves if women are also [as] capable [as men].” (Tekchandani et al 1997) Santosh saw our argument long before us: it is only participation that gives Voice to dissatisfied voters, for whom Exit from Family and/or State is not an option, thus initiating a virtuous cycle. This is possibly why Chattopadhyay and Duflo (2004) find significant impact of women leaders on the nature of public goods provided in West Bengal, a state endowed with greater experience in decentralized governance. Women leaders, placing a higher weight on the provision of drinking water and sanitation, were able to assert their demands through Gram Panchayats and Gram Sabhas here.

A Compromise

The gains from Decentralized information aggregation may ride over the threat of elite capture, over a period of time. Once exposed to the Voice option (via affirmative action), various groups would be able to influence outcomes through participatory action, further enhancing the quality of aggregated information. In order to fully exploit such economies, the implementation of programmes that require particular information must be left to the local government: treating local diseases and deficiencies, vocational training, distributing and updating (birth, death and wedding of family members) ration cards and creating common assets under NREGA or otherwise.

Given the contradictory forces gnawing at our aggregation mechanism, the ideal comes second to the practical. It must remain the Centre’s prerogative to define what is just and what are the mechanisms to redress grievances when justice is not served. The Centre must keep an eye on Dispute Settlement as traditional biases can play into such sensitive decisions. Additionally, Centre must provide mobility to societies; this is only possible if a Central planner can reallocate resources from regions of plenty, to areas of lack. If each unit were to rely on its own tax revenue (from land or agricultural/manufacturing production) to finance public provision of goods, we can expect some units to fall behind. Therefore, the Centre must be in-charge of tools of redistribution (Taxes and Subsidies) to ensure convergence.

Centralized governance has a role to play, in so far as a bearable amount of uniformity is called for. While the provision of public goods should be guided by the cultural specificities and requirements of the local, the Centre should define a minimum acceptable level of provision in Health, Education and Asset Creation.  The particulars of the provision, above this standard, must be settled by the local government based on local information.

In Closing

Affirmative action for groups with high cost of communication will grant them the ability to exercise Voice. This would ensure that preference of each voter over public goods is reliably aggregated. We may grant some control to the Centre to ensure a common minimum standard of living for all individuals within a federal structure. This way, the Paharias would manage their affairs better and at the same time, may spare our Minister, Bureaucrat and Economist the embarrassment from letting them have cows.

References

Afridi, Farzana. “School Transfer Programs.” Lecture Delivered at Delhi School of Economics. 31 Mar, 2017.

Ambedkar, BR. Swaraj and the Depressed Classes. Critical Quest, 2010. Republished Presidential address at All India Depressed Classes Congress, 1930.

Banerjee, Abhijit, and Rohini Somanathan. “A simple model of voice.” The Quarterly Journal of Economics 116, no. 1 (2001): 189-227.

Chattopadhyay, Raghabendra, and Esther Duflo. “Women as policy makers: Evidence from a randomized policy experiment in India.” Econometrica 72, no. 5 (2004): 1409-1443.

Constituent Assembly. Constituent Assembly of India Debates (Proceedings)-Volume VII (1948). Accessed on Mar 10 2017: http://164.100.47.132/LssNew/constituent/vol7p10.html

Crawford, Vincent P., and Joel Sobel. “Strategic information transmission.” Econometrica: Journal of the Econometric Society (1982): 1431-1451.

Das, Sabyasachi. Town-Hall Meetings in Electoral Democracy: Theory and Evidence from Indian Village Councils. mimeo, Yale University, 2014.

Dessein, Wouter. “Authority and communication in organizations.” The Review of Economic Studies 69, no. 4 (2002): 811-838.

Hirschman, Albert O. Exit, voice, and loyalty: Responses to decline in firms, organizations, and states. Vol. 25. Harvard university press, 1970.

Manor, James. “Local Government.” The Oxford companion to politics in India. Ed. Niraja Gopal Jayal and Pratap Bhanu Mehta. Oxford University Press, 2010.

Pai, Sudha. “Pradhanis in new Panchayats: field notes from Meerut district.” Economic and Political Weekly (1998): 1009-1010.

Sainath, Palagummi. Everybody loves a good drought: stories from India’s poorest districts. Penguin Books India, 1996.

Tekchandani, Bharti, Jyoti Kiran, and Priti Sharma. They Call Me Member Saab: Women in Haryana Panchayati Raj. Multiple Action Research Group, 1997.

[1]George RR Martin.  “A Lannister always pays his debts.” A Song of Ice and Fire: A Game of Thrones. Harper Collins (1996)

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s