Geraldo Costa Junior

Hi, welcome.

I am an assistant professor of finance. Currently, I hold a position at Universidade Federal Fluminense, department of Entrepreneurship and Managment.

My background includes a B.A. in Economics from Universidade Federal de Viçosa and a MSc and a PhD in Applied Economics from the Universidade de São Paulo (Brazil).

Between 2016 and 2017, I was a visiting scholar at Wageningen University (Netherlands) and at the School of Economics and Finance - Queen Mary University of London (UK).

My research interests are mainly related to agricultural economics and financial economics. I have also worked with applied microeconomics.

CV Lattes

Contact: geraldocjr@gmail.com

Selected Publications

Concentration and Liquidity Costs in Emerging Commodity Exchanges

(with Andres Trujillo-Barrera and Joost Pennings)

We analyze the relationships among liquidity costs, volume, and volatility in the Brazilian agricultural futures market, along with the role of market concentration. We estimate a structural three-equation IV–GMM model using data from Bolsa, Brasil, Balcão corn and live cattle contracts from March 2014 to February 2016. Results show a negative association between liquidity costs and volume and a positive association between liquidity costs and volatility. Market concentration impacts corn and live cattle differently. Concentration contributes to volume reduction for live cattle and to liquidity costs reduction for corn. Our findings shed light on the microstructure of emerging markets.

Journal of Agricultural and Resouce Economics, 43 (3): 441-456, 2018

"An analysis of the dynamic relationship between the price and quantity of mineral fertilizers and the purchasing power of Brazilian farmers"

(with Cristiane Ogino, Nataliya Popova, and João Martines)

Mineral fertilizers have been some of the major contributors to the increase in agricultural productivity in Brazil since the 1980s. The dependence on scarce natural resources and the extensive use of energy for their synthesis, however, have made agricultural production sensitive to fluctuations in the supply and demand for this type of fertilizers. In this sense, this paper analyzes the dynamics between fertilizers consumed quantity, price and producers’ purchasing power weighted by productivity in Brazil’s Central-East region. This region is responsible for the highest consumption of mineral fertilizers and the largest agricultural production in the country. Using the Structural Vector Autoregressive model (SVAR), three equations for each of the major groups of mineral fertilizers were estimated: nitrogen, phosphate and potassium. The results show that quantity is the most endogenous variable, while the most exogenous variable is price. The estimates of the impulse-response functions show that a shock in fertilizer price contributes to a bigger reduction in quantity consumed than the purchasing power of the producers, with the greatest for the group of potassium fertilizers.

Brazilian Journal of Rural Economics and Sociology, 59 (1): 1-19, 2021

Other Publications

Tonin, J; Costa Junior, G.; Martines Filho, J. Liquidity Costs in Emerging Corn Futures Markets. Revista de Administração Mackenzie (RAM), v. 18, p. 201-223, 2017.

Costa Junior, G; Bacha, C.J.C . Análise do Pessoal Ocupado no Meio Rural do Paraná de 2000 a 2010. Pesquisa & Debate, v. 27, p. 91-107, 2016.

Costa Junior, G.; Popova, N.; TANUS, T.; Martines Filho, J.G. Elasticidades de transmissão espacial de preço do milho no Brasil: Um estudo para o Mato Grosso. Economia & Região, v. 4, p. 63, 2016.

Working Papers

"Modelling and Forecasting Realized Volatility in Emerging Agricultural Futures Markets"

(with João G. Martines Filho)

This paper focuses on modelling and forecasting realized volatility in emerging commodities markets using intraday data from the corn and live cattle contracts traded at BM&F-Bovespa. For this purpose, four different models are considered in the analysis: the heterogeneous autoregressive model (HAR-RV) developed by Corsi (2009) and its extensions adapted to include jumps (Andersen et al., 2007) and leverage components (Corsi and Reno, 2012). Results for the in-sample modelling show that the best way to harness all volatility features in the corn and live cattle future markets is to use models that account for both jump components and leverage effects. However, the out-of-sample forecast analysis show significant differences between the above mentioned models.

"Liquidity Provision, Concentration, and the Bid-Ask Spread in Thinly Traded Markets"

(with Marcelo Fernandes)

This paper examines the dynamic relationship between dealers’ activity and market microstructure in the live cattle inter-dealer market at BM&F-Bovespa. Firstly, a descriptive analysis of the live cattle inter-dealer market structure is carried out in order to answer questions such as: Is the live cattle market competitive in the sense of many dealers active in the price discovery process, or is there a dominant dealer? How important is the dealers’ activity? Does market concentration affect dealers’ profits and activity? The next step is an investigation about the dynamic of dealers’ activity and its determinants. The behavior of dealers’ activity is modelled using an instrumental variable probit model, giving value 1 when the dealer is active on any given day and 0 otherwise. Results indicate that the live cattle inter-dealer market is not competitive and that dealers’ activity is positively related to market concentration, quoted bid-ask spread, number of active dealers and the dealer’s traded quantity. Besides, results also point that the impact of the concentration and number of dealers on the bid-ask spread observed in the live cattle market.

Teaching

At UFF:

Principles of Economics; Brazilian Economy; Monetary Economics and Financial Markets

At UFT:

Principles of Economics; Microeconomics; Financial Markets; Statistics II; Calculus.

Invited Talks

"Child labor across Brazilian states and economic sectors" - University of Rome/Tor Vergata, 2014 (with Ana Kassouf and Roselaine Almeida)