These results gain importance as crises become more frequent and damaging, increasing the marginal utility of my framework for understanding adaptation in thin-institution contexts.

They formalise how collective behaviour equilibria can be shifted temporarily or durably by manipulating costs and observability, offering general principles for rapid norm change in urban environments.

Empirically, the results confirm the model’s comparative statics and extend theory to non-spatial urban networks, where compliance is mediated not by geography but by visibility, attention, and network salience.

This is the first paper to use an electricity distribution company’s data and infrastructure to randomise at the sub-division level in Lahore and link electricity utility billing records to treatment roll-out -leveraging administrative systems as a delivery mechanism for behavioural change.

My Lahore masks RCT increased proper mask use at a fraction of comparable rural interventions, illustrating economies of scale in dense networks.

In 2021, I tested a threshold model of behavioural adoption through a randomised experiment across Lahore. Building on Abaluck et al. (2021), the design isolates how lowering adoption costs and raising salience interact to shift public-goods provision in dense urban networks.

Together, these projects form a unified research program on thresholds, adaptation, and misallocation across networks, disasters, and digital systems. Looking forward to sharing the new papers furthering this agenda in the coming weeks.

I formalise adaptive misallocation: when firms expect limited state support, survival-driven reallocations entrench inefficiency. Evidence from Pakistan’s 2005 earthquake shows how these dynamics distort capital allocation and generate persistent production scarring.

In a second RCT I extend my threshold model of adoption to vaccine uptake in high-density urban networks,. Uptake was free but stalled. Community mobilisation raised vaccinations at local health units, but intensification yielded no further gains. Time-stamped links on OSF.

I develop a threshold model of adoption where lowering adoption costs and increasing salience of the message drives compliance at high centrality nodes in dense urban networks. In 2021, I tested this framework through a mask experiment in Lahore. Time-Stamped: Drafts & protocols: osf.io/bckja/

Together, these two RCTs advance my broader agenda: mapping how dense urban networks mediate collective behavior. They complement my current work on adaptation under weak state credibility.

Drafts, protocols, and acknowledgements will be made public on OSF upon publication.

osf.io/bckja/

Building on that foundation, my 2021 Lahore experiment develops a threshold model of adoption and tests how salience and cohesion shape equilibrium compliance in dense urban networks. (2/6)

A second RCT turned to vaccines. Uptake was free but stalled amid mistrust, misinformation, and logistical barriers. Community mobilisation raised vaccinations at local health units, but intensification yielded no further gains and the value of adaptive engagement. (5/6)

The intervention confirmed the intuition: distributing free masks and reinforcing salience at mosques and markets increased proper use, with gains strongest in cohesive neighborhoods. Marginal cost per additional adopter was very low, showing economies of scale.

(4/6)

The project develops a threshold framework where both private incentives and network effects matter. In a metropolis of 15 million, we tested whether reducing costs and reinforcing salience at central public nodes could shift collective behavior.

(3/6)

Building on that foundation, my 2021 Lahore experiment develops a threshold model of adoption and tests how salience and cohesion shape equilibrium compliance in dense urban networks. (2/6)

Link to the Twitter thread: x.com/MahaRehman1/...

Link to published paper, ungated version, and blog: maharehman.github.io/research/