ALUMNI PRESENT PANEL AT CONFERENCE OF
SOCIETY FOR INSTITUTIONAL AND ORGANIZATIONAL ECONOMICS
July 14, 2026
INSEAD, Fontainebleau, France

Workshop alumni Silviya NItsova, Nico Ravanilla, and Alexander Soldatkin presented papers at a session on elite incentives, networks, and violence at the 2026 SIOE conference. Lee Benham was the chair, and Konstantin Sonini was the discussant.
(SIOE also included an additional paper in this session: Corporate Dark Money in Politics: Evidence from India, by Sukrit Puri, London Business School, and Boris Vallee, INSEAD.)

ABSTRACTS BY THE COASE ALUMNI PANEL

Oligarchs and Geopolitical Realignment:
Evidence from the Russo-Ukrainian War
Silviya Nitsova
University of Manchester

How do powerful economic elites and their political networks respond to geopolitical shocks? This paper leverages Russia's 2022 invasion of Ukraine in a before-and-after design that follows the same individuals over time and combines unique relational, expert, and administrative data to examine whether oligarchic networks of influence in the Ukrainian legislature endure and how oligarch-MPs and oligarchs' loyal legislators change their political behavior under extreme disruption. I find some network shrinkage, as some legislators exit parliament, but substantial behavioral adaptation—both oligarch-MPs and oligarchs' loyalists become significantly more likely to support pro-EU integration legislation after the full-scale invasion, while legislators outside oligarchic networks show little comparable shift. The paper evaluates three potential explanations for this realignment: wartime state centralization, the material weakening of oligarchs, and oligarchs' pre-war economic and political exposure to Russia. 

Banking on Family:
Do Family Onwership Networks Matter for the Survival of Russian Banks?
Alexander Soldatkin
University of Oxford

Do political connections help firms survive? We construct the Factional Graph Index (FGI), a time-varying proxy for political embeddedness from Russian officials’ income declarations and state procurement data, linking them to firm ownership networks via a Neo4j graph database. Estimating Cox time-varying proportional hazards models on firms observed from 2004 to 2020, we find that a connection held in the previous year is associated with a closure hazard not distinguishable from that of unconnected banks (HR = 0.90, n.s.) and a closure hazard 25 per cent lower for commercial companies (HR = 0.75, p < 0.001), net of region, sector, registration cohort and calendar year. Amongst banks, neither the connections that existed when the declaration record begins nor those formed later go with a measurably different hazard. With the company’s size held fixed the company association shrinks, and with size lagged three years, on the years from 2014, it is not distinguishable from one. Counting public institutions and government bodies as companies would turn the lower company hazard into a higher one, and the higher hazard comes from the government bodies. In a landmark design that fixes connection status in 2010, connected companies have a lower closure hazard than unconnected ones before the shocks and lose it after them, a change distinguishable from chance, while for banks neither the connection hazard ratio before the shocks nor its change after them is distinguishable from one. Traced through the final graph without their dates, connections go with a lower hazard, but that lower hazard comes, all of it for banks and much of it for companies, from entity-years in which no connecting path was in force, chiefly the years before the connecting tie was recorded, which only an entity that survived to the record can contribute, a pattern consistent with look-ahead bias. A second design compares, within the same years, connected entities that lose their last connection with those that keep it: neither banks nor companies close more often in the years after the loss. Neither design identifies a mechanism.  

Dodging Electoral Scrutiny:Strategic Timing, Project Choice, Bureaucratic Control, and Favoring Connected Firms in Public Works Procurement 
Nico Ravanilla
University of California San Diego

Elections create competing incentives over public works procurement. Incumbents seek visible public works for electoral credit while steering contracts toward connected firms to extract resources for political ends. Yet elections also heighten scrutiny, making preferential contracting more costly. How do politicians navigate these pressures? We study over 140,000 public works contracts in the Philippines from 2004 to 2019. Procurement expands sharply before elections, but connected firms do not disproportionately benefit. Exploiting close congressional elections, we show that connected firms instead gain an advantage between elections, concentrated in projects whose quality is difficult to monitor. Their projects exhibit poorer workmanship and greater delays, dimensions difficult to conceal in official evaluations. Political continuity also reduces turnover among district engineers who administer procurement. Politicians thus adapt to electoral scrutiny by shifting preferential contracting toward less scrutinized times and projects while preserving the bureaucratic relationships through which influence can be exercised.