Theoretical Framework

Green bonds and their impact on stock market performance rest on several core financial and sustainability theories, each generating distinct testable predictions for this study.

Signalling theory (Spence, 1973) posits that green bond issuance serves as a credible signal of a firm’s commitment to sustainability, reducing information asymmetry between managers and investors. This should manifest as positive abnormal returns concentrated around the announcement date, as markets reward firms perceived to be genuinely pursuing environmental goals.

Stakeholder theory suggests that addressing environmental concerns strengthens relationships with diverse stakeholders investors, customers, regulators and communities yielding reputational capital and operational efficiencies over time. This predicts longer-term performance improvements and sustained reputational benefits beyond the immediate announcement effect.

Finally, the efficient market hypothesis (EMH) in its semi-strong form underpins the event study methodology, assuming that stock prices rapidly incorporate all publicly available information, including green bond announcements. Consequently, any abnormal returns should be short-lived and concentrated around the event date, with prices reverting to fundamentals thereafter.

Together, these theories provide a cohesive framework for the hypotheses. Signalling theory predicts immediate positive market reactions to green bond announcements, while stakeholder theory anticipates longer-term benefits from enhanced reputation and stakeholder alignment. The EMH ensures that any observed abnormal returns will be concentrated around the event window, guiding the event study methodology and window selection in this analysis. This integrated theoretical lens motivates testing whether green bond issuance generates significant short-term abnormal returns (H₁) in the Indian context, with potential implications for sustained value creation.

Empirical Insights and Global Evidence

Empirical research on green bond announcements offers evidence of positive reactions in many developed markets. Flammer (2021) documents significantly positive abnormal stock returns around corporate green bond announcements in the USA, attributing this to improved environmental reputation and investor confidence. Tang and Zhang (2020) similarly show that shareholders often interpret green bond issuances as credible signals of a firm’s long-term sustainability orientation, leading to short-term wealth gains. Roslen et al. (2017) also report positive shareholder wealth effects across several countries, particularly where disclosure standards and green finance ecosystems are relatively mature.

However, a growing body of work points to mixed or conditional reactions. Mocanu et al. (2021) find that while short-term stock price responses to sustainability bond issuances are generally positive, longer-term effects are more muted and highly sensitive to market maturity and regulatory context. Antoniuk and Leirvik (2021) show that transition climate risks can dampen or offset the advantages of green bonds, especially for firms with substantial exposure to carbon-intensive activities. Kazokas (2024) reports heterogeneity in market reactions among European issuers, with responses varying systematically by firm size, industry and prevailing investor sentiment, suggesting that not all green bond announcements are rewarded equally.

A third stream emphasizes the role of certification and credibility in shaping market responses. Wu (2022) argues that green bonds are not necessarily priced below traditional bonds in primary or secondary markets, challenging the assumption of a universal ‘greenium’ in bond yields and indicating that investors scrutinise issue quality and issuer fundamentals rather than paying an automatic price premium. This has implications for equity reactions as well: If investors do not systematically reward green bonds through lower yields, any stock price response is more likely to depend on perceived credibility, transparency of use-of-proceeds and alignment with broader ESG strategies rather than the label alone. Boschetti and Baldini (2017) further suggest that the success of a green bond, in both bond and equity markets, hinges on the perceived genuineness of the issuer’s environmental claims and the strength of external verification or second-party opinions. Together, this literature indicates that while positive abnormal returns are common, they are neither universal nor uniform and depend strongly on sector, certification and the institutional quality of the underlying market.

Indian and Emerging Market Context

Research on green bonds in India and other emerging markets remains limited but is steadily growing, highlighting both opportunities and unique challenges. Verma and Bansal (2023) conducted one of the earliest event studies specific to the Indian market, analysing six green bond issuers over a 21-day event window and reporting overall positive cumulative abnormal returns (CAR). For instance, Yes Bank experienced a 5.3% CAR, reflecting favourable investor sentiment toward credible sustainability commitments (exact figures confirmed from published study). Prakash and Sethi (2021) emphasise the importance of supportive policy frameworks and favourable macroeconomic conditions in enhancing the effectiveness of green bonds across Asia, while Bansal et al. (2023) identify persistent challenges in India, including high coupon rates, limited investor awareness, and insufficient sectoral diversification. Verma and Agarwal (2020) criticise the heavy concentration of green bond proceeds in renewable energy and advocate for broader allocation to underserved areas, such as water infrastructure and waste management. Glavas (2018, 2020) demonstrates that regulatory clarity significantly bolsters market confidence, underscoring the need for standardised green bond guidelines. Ahmed et al. (2024) argue that green bonds can effectively bridge corporate finance with the UN Sustainable Development Goals, provided transparency and third-party certifications are robustly implemented.

This study extends Verma and Bansal (2023) in four key ways: (a) a larger sample of 35 green bonds from 14 listed firms (vs. their 6 firms), (b) a longer time horizon spanning 2015-2024 to capture market evolution, (c) multiple robustness models (MAR, MKAR, RAR) beyond a single specification and (d) wider sectoral coverage including banking, energy, and infrastructure for more comprehensive insights.

Emerging markets like India present distinct dynamics compared to developed markets: weaker disclosure standards, higher funding costs due to sovereign risk premiums and thinner market liquidity can lead to more volatile and delayed investor reactions in event studies. These factors may amplify the role of signalling through green bonds while potentially muting long-term effects due to macroeconomic noise and limited ESG data integration.

Green Bonds During Crises and Market Spillovers

The resilience of green bonds under economic stress and their integration into global portfolios have also attracted research attention. Sisodia et al. (2022) and Yi et al. (2021) demonstrate that green bonds exhibited relative stability during the COVID-19 crisis, displaying hedging qualities against equity market volatility. Guo and Zhou (2021) confirm this pattern in US and Chinese markets, while Dong et al. (2023) note that green bonds show lower correlation with conventional equities, enhancing portfolio diversification benefits. Mensi et al. (2023) and Tiwari et al. (2023) document increasing financial spillovers between green bonds, global equity markets, and even Islamic financial assets, signalling their growing systemic integration.

These findings matter for equity event studies because green bonds’ demonstrated diversification and resilience qualities may influence investors’ stock price responses at issuance, particularly for firms targeting ESG-oriented portfolios that value stability during uncertainty.

Research Gaps and Hypotheses

Current literature reveals several critical gaps that this study addresses, particularly in the Indian context. While global studies often report positive short-term market responses, analyses of long-term financial and ESG performance among green bond issuers remain limited. Sectoral distinctions such as differences between banking, energy and infrastructure are rarely explored, despite varying investor perceptions across industries. Additionally, the influence of third-party certification and ESG disclosures on investor sentiment has not been fully examined for emerging markets like India, where disclosure standards and market liquidity differ from developed economies.

To address these gaps, this study tests the following hypotheses using an event study approach with Average Abnormal Returns (AAR) and Cumulative Average Abnormal Returns (CAAR) across multiple event windows:

• H₀

  (Null Hypothesis): Announcement of green bond issuance does not lead to significant abnormal returns for issuing firms in the Indian stock market.

• H₁

  (Alternative Hypothesis): Announcement of green bond issuance leads to significant abnormal returns for issuing firms in the Indian stock market.

Subsidiary hypotheses:

• H_1a:

  Green bond announcements generate positive abnormal returns in the short-term announcement window [0, +2].

• H_1b:

  There are no significant long-term abnormal returns in post-announcement windows [2, +10] and [10, +20].

• H_1c:

  Abnormal returns differ significantly across sectors (banking vs. energy vs. infrastructure).

These hypotheses directly test the signalling effect predicted by theory, the short-lived nature expected under EMH, and the heterogeneity suggested by sectoral and certification literature, using India’s comprehensive green bond dataset from 2015-2024.

Data Collection

This study adopts a systematic approach to data collection in order to investigate the impact of announcements of green bond issuance on the stock price performance of Indian companies. Bloomberg was utilised to compile a comprehensive list of all existing and mature green bonds issued by Indian entities. The initial dataset comprised 99 green bonds, with detailed information extracted for each bond, including issuer name, ticker symbol (ISIN), bond serial number, green bond classification, issuance amount (in USD), ESG classification, issuing country, announcement and issue dates, maturity date, coupon rate, currency, security type and market sector description. To facilitate focused analysis, bonds were grouped into three categories: government entities (29 bonds), listed companies (43 bonds) and non-listed companies (27 bonds). The empirical analysis centred on the 43 bonds issued by listed companies, as these allowed for the evaluation of stock price reactions. After excluding eight bonds issued before the respective companies were publicly listed, the final sample comprised 35 green bonds issued by 14 publicly listed companies, spanning sectors such as banking, energy and infrastructure. Table 2 represents the entire dataset under study. Stock price data were retrieved from Google Finance, covering a window of ±120 trading days around each bond’s announcement date. Additionally, daily closing prices of the NIFTY 50 index from 2014 to 2024 were collected to serve as a benchmark for market performance.

Research Methodology

This study employs the event study methodology introduced by Fama et al. (1969) and grounded in the semi-strong form Efficient Market Hypothesis proposed by Fama (1970). It has been extensively used to empirically test corporate events and associated stock market responses (Werner, 2010). This method is based on the semi-strong form of the market efficiency hypothesis, which holds that it is not possible to consistently earn excess returns because all publicly available information is immediately reflected in stock prices. Because of its systematic framework, this method has become more popular among researchers as a valid analytical tool. It involves comparing market performance before and after a significant event and evaluating influential variables such as trading volumes and return volatility (Khotari & Warner, 2006).

As shown in Figure 2, the timeline is divided into an estimation period (days -120 to -21 relative to the green bond issuance date) and an event window (±20 days around the issuance date). Daily stock returns during the estimation period are used to calculate the mean return for each firm using a logarithmic return method. Abnormal returns are then determined within the event window using the mean-adjusted returns model. Abnormal returns after the event may suggest market inefficiency, while abnormal returns before could indicate insider trading:

Figure 2. Timeline of the Study.

Where ARt represents the abnormal return on a specific day, Rij is the mean return on the stock during the estimation period, and Rit denotes the actual log return of the stock. The daily log return was determined using the formula:

where Pit and (Pit - 1) indicate the stock price on days t and (t-1), respectively. The average stock return Rij was calculated using the log return of stock prices.

Further, abnormal returns were calculated using the market-adjusted returns model. Abnormal return $A{R}_{i,t}$ is calculated by adjusting the actual return with the market return.

is the return on the market (NIFTY index return).

Moreover, to check the robustness of the model, a risk-adjusted model was used to support the results obtained through the previous models, using the following formula:

where αi and βi are estimates from a regression of the stock against the market index and ARit and Rit are the same as defined earlier.

The same exercise is done for all companies that have issued green bonds. After that, all abnormal returns are averaged for the same day.

To measure the total impact of announcements of green bond issuance over a specified period, all AARs are added individually to create the CAAR, as represented in the equation below.

The CAAR was calculated by adding the abnormal returns for each window under analysis. An event window of 41 days was used for the study, from day -20 to day +20, where day 0 is the date on which green-bond issuance was announced.

To find the statistical significance, the t-test is considered.

Here, is the standard deviation of the average security return. N is the total number in the sample; here, it is 35. A t-statistic with a p value less than or equal to .05 is significant.

Daily AAR and t-Statistics Analysis

Table 3 presents the AARs and corresponding t-statistics for each day from 20 days before to 20 days after the announcement date.

Pre-Event Window Analysis

In the pre-event window (Day -20 to Day -1), several statistically significant negative MARs and RARs are observed, notably on Day -10 (-0.93%, t = -12.71), Day -11 (-0.67%, t = -9.22), and Day -12 (-0.60%, t = -8.14), indicating investor pessimism or information leakage (Brown & Warner, 1985). However, brief positive spikes on Day -8 (0.52%, t = 7.16***) and Day -4 (0.53%, t = 7.22***) indicate some optimistic speculation ahead of the announcement. MKAR also exhibits mixed behaviour during this period, implying uncertain investor expectations and market speculation prior to the official green bond announcement. The results indicate significant negative abnormal returns in the pre-announcement period under MAR and RAR models, suggesting potential market anticipation, adverse firm-specific conditions, or gradual information diffusion before formal disclosure.

On the announcement date, abnormal returns are modest and largely insignificant, indicating that markets may partially anticipate the information.

Event Day (Day 0)

On the announcement day, a slight positive MAR of 0.12% (t = 1.66) is recorded, while RAR rises modestly to 0.13% (t = 1.82*), indicating a short-term market endorsement of green bond announcements (Flammer, 2021). The insignificant MKAR (-0.14%, t = -1.89) supports the observation that markets may have already absorbed the information or remained cautiously neutral.

Post-Event Window Analysis

The post-event window (Day +1 to Day +20) begins with short-term positive returns, especially on Day +2 (MAR = 0.24%, t = 3.23***), Day +8 (MAR = 0.38%, t = 5.20***), and Day +12 (MAR = 0.26%, t = 3.60***). These findings imply a delayed recognition of the strategic value of green bond issuance by investors. However, this is followed by sharp corrections, such as Day +13 (-0.90%, t = -12.31) and Day +14 (-0.58%, t = -7.88), signalling either profit-taking behaviour or market doubts about the long-term financial implications of such initiatives.

Risk-Adjusted Insights

RAR trends closely align with MAR, confirming that the observed abnormal returns are not substantially influenced by changes in systematic risk. Notably, statistically significant RARs post-event reaffirm that investor sentiment momentarily favours green bond announcements, although this impact is not sustained in the long run.

Cumulative Average Abnormal Returns Analysis

To gain further insights, the AARs were aggregated over selected event windows. Table 4 presents the CAARs and associated t-values for the MAR, MKAR, and RAR models.

Pre-Announcement Periods ([–20, –10] and [–10, –1])

The CAARs for both pre-announcement windows are significantly negative under the MAR and RAR models, suggesting that the market possibly held pessimistic views or experienced uncertainty in the lead-up to the green bond announcements. For instance, the MAR CAAR in the [-20, -10] window is -3.5% with a highly significant t-statistic of -24.02. This may reflect broader market concerns or sector-specific weaknesses during that time.

Announcement Window ([0, 2])

A clear positive reaction is observed in the short-term announcement window. The MAR and RAR models show statistically significant positive CAARs (0.45% and 0.49%, respectively) with t-statistics above 3, indicating that investors react positively to green bond announcements. This is consistent with signalling theory, where the issuance of green bonds sends a positive message about the firm’s commitment to sustainable practices, thus enhancing investor confidence (Flammer, 2021).

Post-Announcement Windows ([2, 10] and [10, 20])

However, the post-event window beyond Day +10 reveals significant negative CAARs, suggesting that initial optimism dissipates and prices revert toward fundamentals. The CAARs for these windows go negative again, particularly under the MAR and RAR specifications. For instance, the MAR CAAR for the [10, 20] window is -2.78% (t = -19.09), implying that the early market optimism is fleeting and might be followed by profit-taking or correction behaviour. This trend is in line with findings from earlier research that environmental cues are likely to make a stronger, shorter-term impact and can lose influence if there is a lack of follow-up or underlying change (Tang & Zhang, 2020).