Inovação operacional: uma proposta de medida para mercados com baixa qualidade da informação em P&D

Abstract

Studies on innovation demonstrate its importance in the contemporary world, particularly in the development of companies and countries. Among the different types of innovation, operational innovation stands out for its significant influence, both in the ability to respond to market needs and in the results and benefits achieved. The competitive edge of innovation has driven the growth of academic research, which, however, faces conceptual and measurement challenges. Traditional metrics such as R&D and patents capture some aspects of innovation but are not effective for specific innovations like operational innovation. Additionally, obtaining such metrics is not straightforward in markets like the BRICS, where most member countries provide limited innovation data, reflecting the low quality of accounting information. This lack of data makes it difficult to determine whether these countries do not invest in innovation or simply do not disclose such investments, thereby limiting research efforts. This thesis proposes an alternative measure for assessing operational innovation in companies. The proposal involves using the error term of a multiple linear regression (OLS) as a proxy for innovation, similar to the approach used in studies on investment decisions and earnings management. Unlike these models, which use residuals to quantify investment or earnings management, in this thesis, the residuals represent innovation. The econometric model has operating profit as the dependent variable, as innovation, specifically operational innovation is expected to improve a firm's efficiency and consequently increase its operating profit. The independent variables are characteristics identified in the literature as linked to innovation in firms. According to the theoretical framework, these variables are size, age, CAPEX investment, equity capital, and industry sector. Beyond its main objective of proposing an alternative innovation measure, this thesis aims to apply it. First, it will be compared with the Research and Development (R&D) variable, a metric widely recognized in the literature as indicative of technological innovation. The intention is not to replace R&D or any other innovation measures but to contribute to measuring aspects of innovation not captured by current metrics. In a second step, the measure will be used to analyze operational innovation in BRICS countries. For this purpose, a quantitative method is proposed, divided into at least three stages: a) constructing the regression model to obtain the residual variable as a proxy for operational innovation, referred to as INOVAOP; b) conducting comparative tests between INOVAOP and the R&D variable; c) applying the proposed measure to the BRICS economic bloc countries. The dataset for this study comes from the LSEG (London Stock Exchange Group), formerly known as Refinitiv Eikon. The sample will consist of companies from BRICS member countries (Brazil, Russia, India, China, South Africa, Saudi Arabia, the United Arab Emirates, and Egypt) covering the period from 2019 to 2023. The results indicate that the proposed measure is promising as an operational innovation metric. Comparative tests between the proposed variable and the R&D variable suggest that operational innovation is more explanatory for ROA (Return on Assets), which measures a firm's ability to generate profit from its total assets, and ROE (Return on Equity), which measures a firm's ability to generate profit using shareholders' equity. In contrast, the R&D variable, indicative of technological innovation, appears to be more explanatory for variations in Market Value, which reflects the external perception of the company. For the BRICS group, the alternative variable highlights the importance of operational innovation, showing positive coefficients for the study variables ROA, ROE, and Market Value, although the significance varies across countries.