Voorhees-Large Prize for Leeds Alumna
Alumna Samira Marx Pinheiro wins Voorhees-Large Prize, March 2018.
Samira studied Public Administration at the Fundação João Pinheiro School of Government of Minas Gerais State, graduating in 2009; she also completed a programme in Statistics at the University of Minas Gerais. This is the fourth largest state in Brazil, covering an area similar to the size of France, with a population of 21 million, making it Brazil’s second most populated state.
Samira’s professional career started at the Minas Gerais' State Secretariat of Transportation and Public Works where she worked on transport infrastructure regulation and transport services. By 2015, she was the Chief-Advisor of Planning, a post she left in 2016 to start her MSc Transport Economics at Leeds, for which she had been awarded a Chevening Scholarship. Chevening Scholarships are a UK Government international awards scheme aimed at developing global leaders, providing Scholars with the opportunity to develop professionally and academically, to network extensively, to experience UK culture and to build lasting positive relationships with the UK. They are funded by the Foreign and Commonwealth Office and its partner organisations.
Having returned to Brazil from Leeds, Samira is now the Director of Concessions in the Minas Gerais' State Secretariat of Transportation and Public Works, responsible for the regulation of PPPs and other long-term public contracts in the transport sector.
Samira believes that ‘the professionalisation of the public service is fundamental in developing countries in order to improve their quality. This is the purpose of the Fundação João Pinheiro School of Government of which I am very proud to be an alumnus. The scenario of poor human capital in public transport generates the perverse binomial of “weak regulator and strong corporations” which jeopardises the quality of services. I believe that helping disseminate principles of transport economics in public policy is my major contribution to change for the good.’ Samira sees winning the Voorhees-Large Prize ‘as recognition for the institutions which have supported my academic and professional life. My School of Government studies provided my initial insights and guidance in the public service and I had the opportunity to expand those ideas and gain practical knowledge working for the Minas Gerais' Government. The Chevening Scholarship gave me very high-quality international experience and at the University of Leeds, Institute for Transport Studies I had access to a global community of practitioners and experts in transport’.
Samira’s dissertation focussed on quantifying the sensitivity of rail demand with respect to the different fares available for a given market applying Bayesian statistics. She explains that ‘for a given route there might be different options for rail tickets with different fares, differing from each other by their service conditions. When there is a change in the price of one fare, the demand for its competitors might also be affected. For example, a passenger who usually buys a standard ticket might find it convenient to choose first class when it becomes cheaper and the standard fare is kept the same. The question is, therefore, how a change in one fare impacts rail demand more widely – everything else being constant?’
‘In economic terms, this is the problem of estimating the price elasticity of the demand for competitors’ goods (own and cross elasticities). When standard theory is applied, the estimation often results in wrong-sign coefficients - a problem acknowledged in the econometric world: own-price elasticities, supposedly negative since the higher the price of fare A, the lower its demand, are estimated as positive and cross-elasticities, supposedly positive since the higher the price A, the higher the demand of the competitor B, as negative.
To overcome this problem, I applied a Bayesian method to estimate the coefficients of demand sensitivity - the elasticities. The great advantage of the Bayesian's was the straightforward framework to incorporating prior information in order to constrain the estimation into a feasible interval - in this case, positive or negative. Therefore, the wrong-sign coefficients are banned from the estimation possibilities, and coherent solutions are provided. The overall conclusion was that, as a proof-of-concept study, the work has demonstrated that Bayesian methods potentially have practical application for price elasticities estimation and rail demand forecast.’
For more information on the Voorhees-Large Prize and the Brian Large Bursaries visit www.blbf.co.uk
