Informações sobre commodities agrícolas: efeitos dos relatórios de safra periódicos em contexto espacial (internacional, nacional e regional), surpresas e portfólios

Abstract

Achieving higher performance is an important strategic objective for investors and portfolio managers (Yoon, Al Mamun, Uddin, & Kang, 2019). To improve performance and reduce the level of risk, many investors and portfolio managers look to other asset classes, such as commodities, in addition to stocks and government bonds. Information available to the public is shaped by actions of decision makers in the economy, who can choose to highlight its positive or negative side. Consequently, prices tend to be influenced by the dynamics of narrative presented. Given that there are different ways in which investors can have exposure to assets linked to agricultural commodities, it is essential to understand how this market behaves in the face of new information made available, especially when issued by government entities regarding commodity production forecasts. In a predominantly agricultural country, this is thesis focus, which is divided into three studies involving commodities coffee, sugar cane, corn and soybeans. To achieve complexity in return on investments in commodities given the information dispersed in spatial context, the first study (Chapter 2) proposes to analyze the tone effect of government harvest reports published periodically on agricultural commodities volatility produced in Triângulo Mineiro and Alto Paranaíba region. Between January/2017 and February/2023, an autoregressive model of moving averages was used with a generalized autoregressive conditional heteroscedasticity approach, in a national, international and regional analysis. The results indicate that there is an unexpected element in reports moving agricultural commodity prices, absorbing language tone used in a higher volatility. However, results were not conclusive regarding the predominance direction tone. In the second study (Chapter 3), objective is to understand the relationship between tone information released by government harvest reports, in extreme market conditions (rising and falling), and the behavior of agricultural commodity prices. In period January/2017 and February/2023, an autoregressive model of moving averages was used with a generalized autoregressive conditional heteroscedasticity approach. The findings point to some signs of the relationship between tone information released by government harvest reports, in extreme market conditions (rising and falling), and the behavior of agricultural commodity prices. The third study (Chapter 4), objective is to analyze the performance of Brazilian funds that diversify via assets linked to agribusiness. In this analysis, tone of crop reports is also considered. The analysis covered the period between January/2017 and February/2023, using regressions with panel data (random effects). Among the main results, funds that diversify via agricultural commodities can obtain a different performance from other funds (multimarket and shares), depending on type asset linked to agricultural commodities. It is also evident that optimistic content related to the field does not necessarily promote better funds portfolio performance that allocate assets related to agribusiness. In general, the literature is consistent with the possibility of divergent findings for narrative tone because in some moments negative content appears to be more important for markets while, in others, good news is more relevant. This indicates that investors may deviate from interpretations, predominating numerical analyzes for their decision making.