The new state of Telangana was carved out of Andhra Pradesh in June 2014, after a prolonged movement by the people of Telangana region for a separate state. Based on field experiments among citizens in the two successor states, Lata Gangadharan, Tarun Jain, Pushkar Maitra and Joseph Vecci find greater trust in politicians in Andhra relative to Telangana, which may facilitate effective functioning of the State and signal citizens’ expectations from the government.
The state of Andhra Pradesh (AP), formed in 1953, was bifurcated into two successor states of AP and Telangana in March 2014. The separation was the result of numerous agitations that most observers directly related to the assertion of regional identity, and differences over allocation of economic resources (Pingle 2010, Srinivasulu 2002). A particularly important grievance was the imbalance in political power – while the Telangana region contained over 40% of the state’s population, only two chief ministers had ever been from Telangana districts. Instead, political and economic power was concentrated among those from Coastal Andhra and Rayalseema.
The economic and social disparity between AP and Telangana is significant. Reddy and Bantilan (2013) identified that in five of 10 Telangana districts, more than 10% of the rural population lived in poverty. In contrast, just two out of nine districts fell into this category in coastal Andhra. Literacy rates in Telangana are lower than Seemandhra and Coastal Andhra. In particular, education rates for the most disadvantaged are alarmingly low in Telangana – the proportion of Scheduled Tribe (ST) students with a completed secondary and post-secondary education is 27% and 14% respectively (Reddy 2014, Reddy and Bantilan 2014).
Mutual trust and cooperation between politicians and citizens is crucial for the effective functioning of the state as it improves compliance with rules and regulations and reduces the cost of enforcement (Knack and Keefer 1997). Public governance is inherently discretionary, and so citizens’ trust in public officials is an important measure of whether citizens expect transparency in decision-making (Dalton 2005). Economic interactions between individuals are not only governed by contractual relationships but also by trust between individuals, which often plays a crucial role in facilitating interaction and trade. These factors make citizens’ trust in governments and other citizens conducive to economic and social development.
So could the historical lack of political representation of Telangana and the large differences in social and economic characteristics between the states, be related to differences in perceptions and attitudes towards the political elite? Lacking trust in politicians, citizens may vote for and follow people they personally know such as members of their own community, rather than competent politicians. This perpetuates a system of cronyism and potentially generates discrimination between groups. Conversely, citizens’ trust in politicians might generate a virtuous cycle of positive governance and good economic and social outcomes.
Measuring trust and trustworthiness among politicians and citizens
Do citizens in Telangana and AP trust local political elites? And are local politicians perceived to be trustworthy? Since few sources of secondary data report behavioural variables such as trust, we investigated these questions using an artefactual field experiment (see Harrison and List 2004). Using 2011 census data, we randomly select 40 villages in the border districts between the two states, ensuring that villages are observationally equivalent in terms of demographics and infrastructure.
A total of 960 men and women residing in these 40 villages participated in a modified ‘Trust game’ (Berg et al. 1995). The Trust game is a two-person game where both players are endowed Rs. 200 each. The first player (the Trustor or the Sender) can choose to send some money to a second player (the Trustee or the Receiver). The experimenter triples these funds, and the Receiver has the option of returning some money. Since the Receiver can choose to return nothing, any initial contributions by the Sender is a measure of trust in the Receiver while the amount returned is a measure of trustworthiness.
In all games, the players who were recruited from the village population were randomly assigned an identity either of “Local Politician” or “Landless Labourer”. Our experimental design modifies this standard game by varying players’ knowledge of the identity of the person they are playing with. Hence, by comparing contributions when the other player is a politician versus a labourer, we are able to determine trust in the political elites relative to that in landless labourers and also how this varies across Telangana and Andhra districts. Similarly, by comparing the amount returned by players’ assigned the identity of a local politician versus a landless labourer we can measure the perceived trustworthiness of political elites relative to that of the landless labourer, and draw comparisons between the two states.
We conducted our experiments during the period March–May 2014, with a break during the national elections. Participants in the experiments came from different walks of life – from agricultural labourers to school teachers to Panchayat members – although all were screened for literacy and numeracy skills. None had participated in an economics experiment before. The experiments were conducted in public places such as Panchayat offices, government schools and community halls to ensure that all segments of society could participate. See Figure 2 for a typical experiment in progress.
Our main experimental results relating to trust and trustworthiness by identity and by state are summarised in Figure 3 below. We find significant differences in relative trust in politicians between the two states. Citizens in Telangana send Rs. 20 (or 26%) more when paired with landless labourers than when paired with politicians. Importantly, there is also a difference across states – compared to citizens in Telangana, those in AP send Rs. 12.6 (or 11%) more to a local politician. We also find statistically significant differences in terms of trustworthiness – citizens in AP who were assigned the identity of local politician return Rs. 23.2 more than citizens in Telangana who were assigned the identity of local politician.
Figure 3 suggests lower trust of politicians within Telangana compared to AP. Additionally, citizens in AP perceive politicians to be more trustworthy than in Telangana. If this suspicion stems from the fact that villagers in Telangana feel that politicians in the old regime were unwilling to engage with them, then the bifurcation of the state is a positive outcome as it may herald a new system of decentralised government with local politicians who are more likely to be engaged with the local community.
After the experiments, all participants responded to a survey that elicited their views and opinions regarding attitudes towards governance and community engagement. Participants in Telangana were significantly more likely to report that over the last 12 months, the people in the village or the neighbourhood have come together to jointly petition government officials and political leaders for something benefitting the community. This could be suggestive of (i) greater community engagement in Telangana; (ii) Telangana citizens’ need to petition the government due to the lower levels of government services compared to AP; or (iii) greater civic involvement in the political campaign for the new state.
In contrast, a greater proportion of Andhra residents, compared to Telangana residents, report that local governments take their concerns into account when making decisions. So it is not surprising that we also find that a significantly larger proportion of Telangana residents think that bifurcation will improve the quality of government and public services.
Policy implications
Our experiments uncover significant differences in trust and trustworthiness in local politicians between residents of AP and Telangana. In particular, Telangana residents hold lower opinions of local politicians than Andhra residents, cumulating in their belief that politicians are less effective and consultative. This may be the result of poor governance in Telangana compared to AP cumulating in greater mistrust of local politicians; on the other hand, greater mistrust in politicians may have led to poorer perceptions of governance. Irrespective of the direction of causation, the consequence is lower trust in local politicians in Telangana compared to AP. If the creation of the new state results in greater engagement of both political leaders and elites with the citizens, then the bifurcation of the state could improve governance and boost growth and development. This suggests the need for further research in this area to identify if perceptions towards politicians evolve over time and if this is a harbinger for future success.
A key policy implication of our work is that the new government of Telangana should perpetuate an open and inclusive dialogue with its citizens so that a positive cycle of trust and trustworthiness becomes the institutional norm. The new government could devolve greater power to Mandals and Gram Panchayats, which noted commentators such as C.H. Hanumantha Rao (Rao 2014) argue the old state failed to accomplish. This should be accompanied by greater transparency in government programmes. Residual trust in the new state might deteriorate quickly if residents feel like their voice is not represented, a feeling the people of Telangana know all too well.
This blog was reposted from Ideas for India, an economics and policy portal run by IGC India Central
Note: This article gives the views of the author, and not the position of the South Asia @ LSE blog, nor of the London School of Economics. Please read our comments policy before posting.
About the Authors
Lata Gangadharan is a Professor in the Department of Economics at Monash University. She is an experimental economist and her research encompasses both experiments in markets and auction design and experiments on social preferences.
Tarun Jain is an Assistant Professor of Economics at the Indian School of Business.
Pushkar Maitra is Professor of Economics at Monash University. His specific research interests are analysis of poverty and welfare in developing countries; health, education and human capital accumulation; fertility, mortality and gender preference with special focus on South Africa and South Asia.
Joseph Vecci is a PhD student in the Department of Economics, Monash university. His research interest include development, economics of education and experimental economics.
