Return on Equity and the Global Compact: Its Effect on Colombian Companies that Have Adhered to It

This document analyzes whether belonging to the Global Compact (GC) affected the Return on Equity (ROE) in 1,000 of the largest Colombian companies that reported their financial information to the Superintendency of Corporations in 2017. The impact was calculated using the Ordinary Least Squares (OLS) with standard errors and Feasible and Generalized Least Squares (FGLS) models; the variables ROE and GC were linked to each other by considering control variables such as financial variables (size and leverage) and non-financial variables (economic sector). Statistically, no significant relationship was found between the return on equity and these companies being members of the Global Compact; which is consistent with the neutrality hypothesis, which affirms that socially responsible companies do not experience changes in economic benefits or in the wealth of shareholders, hence showing that socially responsible investments have no effect. This research contributes to the empirical research on the relationships between finance and corporate social responsibility, which contributes to decision making in the field of sustainability by companies.


Introduction
Companies engage in Social Responsibility (CSR) for multiple reasons, including improving their financial performance. Engaging CSR is a choice and not an imposition by some agent within the economic system, which means the entrepreneur needs to be sensitive towards society and the environment.
When a company joins the Global Compact (GC), it does so because it wants to show its commitment to the environment and society. The GC is an initiative by the United Nations (UN) that implies fulfilling the ten Millennium Development Goals and the seventeen Sustainable Development Goals.
Having described above, the objective hereof is to analyze whether or not belonging to the Global Compact (GC) affected the return on equity (ROE) of the one thousand largest Colombian companies that reported their financial information to the Superintendence of Corporations in 2017. The effect was calculated using the Ordinary Least Squares with standard errors and Feasible and Generalized Least Squares models; the variables ROE and GC were related, the size and leverage (financial variables) and the economic sector in which each company performs (non-financial variable) were taken as control variables.
This document sets off with this introduction, moves on to the methodology, the results, and ends with the conclusions. Figure 1 shows the literature review about relationships between social performance and financial performance.

Methodology
The companies included in the sample numbered one thousand and were classified into six different sectors: agriculture, trade, construction, manufacturing, mining and services (categorical control variable). The observations of each company belong to the financial balance sheets reported in 2017, from which the variables profitability of equity (ROE) (dependent variable), size of the company (logarithm of assets), and level of leverage (quotient between liabilities and assets) were constructed. The last two are financial control variables.
Taking as a reference the companies included in the GC, a dummy variable was constructed which took the value of 1 when the company was affiliated to GC and 0 if not. The descriptive statistics of these variables are shown in table 1. In order to determine the effect of belonging to the GC on the member companies, the function proposed in the document by Ortas, Álvarez, and Garayar (2015) was used, with the difference that the dependent variable herein is the ROE (equation 1). This equation was estimated using the Ordinary Least Squares (OLS) with standard errors and Feasible and Generalized Least Squares (FGLS) models.
Ɛ i is the error term.
Electronic copy available at: https://ssrn.com/abstract=3303331 Results Table 2 shows the results of regressions by OLS and FGLS. The returns on equity were found to be statistically not different between those companies that belong to the Global Compact and those that do not. Furthermore, no strong correlation between the explanatory variables and the ROE was demonstrated.

Conclusions
The non-existence of a statistically significant relationship between return on equity and companies' membership to the Global Compact, which is consistent with the hypothesis of neutrality, which states that socially responsible companies do not experience changes in economic benefits or in shareholders' wealth, thereby showing that socially responsible investments have no effect. For this specific case, it is suggested that there may be very different factors from those defined herein that affect the ROE, many of them probably discretionary and subject to the internal management of each company. This research contributes to empirical research on the relationships between finance and corporate social responsibility which contribute to corporate decision-making in the area of sustainability.

Funding
Product derived from the research project INV-ECO 2582 financed by the Vice Presidency for Research of the Universidad Militar Nueva Granada -2018 term.