PTDoes a high ESG score make money cheaper?
Agencies score from 0 to 100 what firms do for the environment, for people and for good management (ESG). The promise is that a high score brings cheaper investors and credit and a more highly valued firm. The research tested this in 594 firms from Brazil, Russia, India, China and South Africa (BRICS), from 2015 to 2022. Explained in 12 screens.
One column per country: in blue, the average ESG score; in red, the average cost of debt (Tables 2 and 3). Where the score is lowest, debt is also cheapest.
Four promises of ESG put to the test
Studies in rich countries often find that a high ESG score lowers the cost of money and raises the firm's value. In emerging countries, the results are mixed. The thesis tested four hypotheses. Bet on each one: we keep your answers and show the tally on screen 11.
Three yardsticks: two prices of money and the firm's value
Every firm pays to use other people's money. The cost of equity is the return shareholders demand; the cost of debt, the interest lenders charge. Tobin's Q compares what the market pays for the firm with the value of its assets. Play with an invented firm.
illustration Invented firm, in Brazilian reais (R$). Formulas from Box 2: Ke = risk-free rate + beta × premium; Ki = (risk-free rate + credit risk) × (1 − tax); Q = (value of shares + debt) ÷ total assets. In the thesis, all three come ready from Bloomberg; the averages are from Tables 1 and 2.
594 firms from five countries, 2015 to 2022
Listed BRICS firms with an ESG score in the Refinitiv database, from 2015 (the year of the UN 2030 Agenda) to 2022. Banks and insurers were left out because they use debt differently.
- ESG scores from Refinitiv, 0 to 100: a total score and one for each pillar (environmental, social and governance).
- Costs and value from Bloomberg: cost of equity, cost of debt, weighted average cost of capital and Tobin's Q.
- Each country's economy from the World Bank: GDP growth and inflation, plus the interest rate.
Each dot = 1 firm. Counts from section 6.1 (they add up to 594). The panel tables report 4,765 observations (a firm in a year), slightly above the maximum of 594 × 8 = 4,752; the thesis does not explain the difference (our observation).
Chinese firms have the lowest ESG scores
In every pillar, China lags behind. Brazil, South Africa and India are almost tied at the top. Choose a pillar.
Averages from Table 3 (differences between countries with p < 0.001, Kruskal-Wallis test). Below, the correlation between pillars (Table 6): firms that do well in one pillar tend to do well in another, especially environmental and social.
Money costs very different amounts in each country
Before you see the bars, take a guess: in which country is firms' debt most expensive?
Averages from 2015 to 2022, Tables 1 and 2. Differences between countries are statistically significant in every measure (Kruskal-Wallis, p < 0.001).
At first glance, a high ESG score seems to make debt more expensive
Taking all firms together, the higher the ESG score, the higher the cost of debt. It is the opposite of the promise. Tap both cards.
Country averages from Tables 2 and 3; circle proportional to the number of firms. The reading that the correlation reflects differences between countries is ours, not the thesis's.
The model compares each firm with itself over the years
The thesis used fixed-effects panel regression: instead of comparing a Chinese firm with a Brazilian one, it asks whether, when a firm's score rises, its cost changes. It also discounts inflation, interest, GDP, size, debt and profit.
illustration Invented numbers to show the logic. The thesis ran three models (fixed effects, random effects and pooled least squares), and the Hausman test (p < 0.0001) chose fixed effects.
Higher ESG score, slightly more expensive equity
In the chosen model, each extra point of ESG score came with 0.026 percentage points more in the return shareholders demand. It is the opposite of the hypothesis, and small next to what interest and inflation do.
Coefficients from Table 10 (fixed effects: ESG score 0.0261, inflation 0.2842, interest 0.2126, GDP −0.1134). The conversion into percentage points assumes cost, inflation and interest are measured in % (our calculation). Size, debt and profit had no effect.
In market value, the ESG score made no difference
The hypothesis expected a higher Tobin's Q with a higher ESG score. In all three models, the effect of the score stayed close to zero and was not statistically significant. Value followed other things: profit, size and the economy.
Change in Tobin's Q for each change shown, fixed-effects model (Table 12). Multiplications are ours, from the coefficients. For comparison: country averages range from 1.41 to 3.72. Tap a row to see the thesis's number.
Only the social pillar showed up, and the other way round
The thesis tested each pillar against each yardstick. Of the nine combinations, only one had an effect. Tap the blocks.
Fixed-effects models from Table 14. Tall block = statistically significant effect; low striped block = no effect. For the cost of debt we use only this pillar-by-pillar model: the table of the model with the total score (Table 11) repeats the numbers of the cost-of-equity table.
What changes for those who invest, lend or manage
Investors and lenders
In the BRICS, a high ESG score did not guarantee lower risk or lower cost. For the author, the results help to understand how ESG actions affect the cost of capital in each country.
Firms
For the author, ESG practices may require high investment at the start, which can raise costs in the short run. Sustainability strategy needs to fit each country's market.
Governments and regulators
Scores vary widely between countries, and China lagged in every pillar. The author notes that Chinese regulators are expanding ESG disclosure requirements.
Researchers
The thesis calls for new studies on the cost of capital in emerging markets, with other methods and statistical techniques and more countries.
What this research does not say
- The models explain little. The share of cost and value explained within each firm is between 1% and 3%. Cost and value depend on much more. · our observation (Tables 10, 12 and 14)
- Association, not cause. Comparing each firm with itself does not say whether the score moves the cost or the cost moves the score. · our observation
- A single ESG yardstick. The scores come from one agency (Refinitiv), and the thesis itself notes that different agencies give different scores to the same firm. · according to the author and our observation
- Six in ten firms are Chinese. BRICS averages and correlations lean heavily towards China. The thesis does not run the models country by country. · our observation
Credits
SILVA, Claudia Vasconcellos. Análise do impacto de desempenho ESG em relação ao valor e custo capital das empresas listadas no mercado de capitais dos BRICS [Analysis of the impact of ESG performance on the value and cost of capital of companies listed on the BRICS capital markets]. Doctoral thesis (Business Administration), Mackenzie Presbyterian University, São Paulo, 2025. Advisor: Prof. Dr. Wilson Toshiro Nakamura.
Research supported by a full tuition waiver from the Mackenzie Presbyterian Institute. Data from Refinitiv, Bloomberg and the World Bank. This version simplifies the work for the general public; the formulas and full tables are in the original text.