Capital markets function as a barometer of a country’s economic health and reflect investors’ collective expectations about future conditions. Stock prices are highly sensitive to new information, including global political developments and policy shocks that can transmit quickly across borders. The uncertainty caused by geopolitical events not only impacts local markets but can also impact international financial markets (Buigut & Kapar, 2020). Recent geopolitical events, particularly changes in U.S. trade policy, have attracted considerable attention from investors and policymakers in emerging markets, such as ASEAN.

In 2025, President Donald J. Trump, the 47th President of the United States, declared the introduction of a new “reciprocal” tariff regime as a national “Liberation Day” for the American people (Ozili, 2025). The policy, formalized in Executive Order 14257, consists of a universal 10 percent tariff on all imports into the United States combined with additional reciprocal tariffs targeting countries with large trade surpluses against the U.S. (Aldaba, 2025). This move represents a significant escalation of the protectionist stance first seen during Trump’s first term, when the U.S.–China trade war in 2018–2019 triggered global financial turmoil, higher inflation expectations, and disruptions to supply chains (Kapar et al., 2025).

Starting in mid-2018, the U.S. imposed a 25% tariff on $50 billion worth of Chinese imports as an initial measure and expanded it to cover nearly all Chinese goods in 2019. These tariff policies have triggered significant changes in global trade patterns and supply chains. U.S. imports from China plummeted, falling from 22% of total U.S. imports in 2017 to approximately 18% in 2022. In response, many companies have shifted their supply sources to countries such as Vietnam and Mexico, which are not subject to the same tariffs. U.S. imports from other countries increased by approximately 38% compared with pre-trade war trends, indicating substantial trade diversion (Bown, 2022) (see Table 1).

Table 1. Reciprocal tariff by the United States of America

Source: (Aldaba, 2025)

The 2025 reciprocal tariffs extend this logic to a broader set of trading partners, including more than 60 countries, such as those in ASEAN. Within ASEAN, only Singapore faces a baseline tariff of 10% tariff, while Vietnam is subject to much higher rates. Indonesia, Malaysia, the Philippines, and Thailand have relatively lower but still non‑trivial tariff levels. The variation in tariff intensity is closely related to each country’s bilateral trade balance with the U.S. and its previous tariff stance toward American products. Consequently, the new tariff structure simultaneously creates threats and opportunities: highly taxed countries, such as Vietnam, risk losing their manufacturing advantage, whereas lower-tariff‑ countries may attract trade and investment diversion. Empirical studies on earlier episodes document sharp and often asymmetric abnormal returns in many equity markets following tariff announcements, with trade-deficit countries showing muted or mixed reactions and trade-surplus countries experiencing significant cumulative abnormal returns (Rao et al., 2025) (see Figure 1).

Figure 1. U.S. trade balance with selected ASEAN countries (Indonesia, Malaysia, Singapore, Thailand, Philippines, and Vietnam)

Source: United States Census Bureau

These trade shocks are likely to affect macroeconomic conditions through exports, investment, employment, and exchange rates, and in turn influence stock markets. According to Mankiw (2012), stock prices incorporate expectations about future economic performance because investors are willing to pay higher prices when they anticipate strong corporate earnings. Positive expectations tend to generate bullish markets, whereas heightened uncertainty and fears of slower growth trigger bearish reactions. External policy shocks such as U.S. tariffs can therefore generate significant revaluations of equity indices, especially in sectors and countries that depend heavily on U.S. demand

Event study methodology provides an empirical framework to measure such reactions by quantifying abnormal returns, which are defined as deviations of actual returns from their expected normal levels, around the announcement of new information (Kothari & Warner, 2007). By examining abnormal returns over a short event window, researchers can infer whether an event contains value-relevant‑ information and how quickly markets incorporate that information into prices. Prior work on tariff announcements finds that markets typically register sharp negative abnormal returns on or shortly after announcement dates, followed in some cases by partial reversals when clarifying or de-escalating‑ signals are released. Building on this literature, the present study applies an event study approach to analyze the impact of U.S. reciprocal tariff announcements in 2025 on six major ASEAN stock indices, namely IHSG (Indonesia), STI (Singapore), FTSE KLCI (Malaysia), SET (Thailand), PSEi (Philippines), and VN Index (Vietnam), by measuring abnormal returns around three key announcement dates to assess how these external shocks are transmitted to regional equity markets.

2.1. Efficient Market Hypothesis

According to Fama (1970), an efficient capital market is one in which security prices at any point in time fully reflect all available information. In such a market, prices adjust rapidly to new information, such that no investor can consistently earn abnormal returns using information that is already known. The key aspect in assessing market efficiency is how quickly and accurately new information is incorporated into equilibrium prices, which reflects the speed of market adjustment to relevant news.

Based on the type of information reflected in prices, market efficiency within the efficient market hypothesis (EMH) framework is classified into three main forms: weak, semi-strong‑, and strong. The weak form of EMH argues that all historical information, including past price and trading volume data, is already fully incorporated into current stock prices; therefore, technical analysis that relies on historical patterns and trends becomes ineffective for earning returns above the market average, a view closely associated with the random walk hypothesis (Malkiel, 1989; Naseer & Yasir, 2015). The semi-strong‑ form states that security prices adjust rapidly to incorporate not only historical data but also all publicly available information, such as dividend announcements, stock splits, earnings reports, and broader political or economic events, implying that fundamental analysis cannot consistently generate abnormal returns because prices already reflect this public information (Malkiel, 1989; Naseer & Yasir, 2015). The strong form of EMH holds that prices reflect not only public information but also private or insider information; therefore, even individuals with privileged access to non-public‑ information, such as corporate executives or insiders, cannot systematically outperform the market, implying a state of near-perfect‑ transparency with no information asymmetry and prices acting as a comprehensive aggregator of all available knowledge (Malkiel, 1989; Naseer & Yasir, 2015).

2.2. Event Study

According to MacKinlay (1997), the event study methodology allows researchers to measure the impact of a specific event on the value or performance of a firm by examining price reactions around the event date. Event studies start from the assumption that markets behave rationally; therefore, the financial impact of events, such as corporate actions or macroeconomic policy changes, is quickly incorporated into asset prices, enabling the observation of abnormal returns in a narrow window around the event. Compared with traditional performance measures that often require long observation periods, the event study approach is relatively efficient because it captures the market’s immediate reaction.

MacKinlay (1997) notes that the use of event studies is very broad, extending beyond finance and accounting into areas such as law and economics, including applications to mergers and acquisitions, earnings announcements, regulatory changes, and macroeconomic policy shifts. The methodology is flexible and can be adapted to various types of securities, including debt instruments, although common stocks remain the most frequently analyzed assets. This framework is closely linked to the concept of an efficient market, in which current stock prices reflect all available information, and the key characteristic of efficiency is the speed with which new information is absorbed and manifested in price changes that establish a new equilibrium.

2.3. Abnormal Return

An abnormal return is defined as the deviation between the actual return observed around the time of an event announcement and the expected return that would have occurred in the absence of that event; this deviation can be either positive or negative (Tandelilin, 2017). Under normal conditions, when there is no significant new information, actual returns tend not to differ meaningfully from expected returns; therefore, abnormal returns are generally small. However, when new information is released that is believed to affect a firm’s future cash flows, the market reacts and this reaction is reflected in the difference between actual and expected returns, creating abnormal returns

In this context, the abnormal return calculation requires two main components: the actual return and the expected return (Suganda, 2018). The actual return is the return that occurs at time t, typically measured as the percentage change in price from the previous period to the current period (Halim, 2015).

The expected return is the return that investors would normally anticipate in the absence of a specific event; one common way to estimate it is the mean-adjusted‑ returns model, which uses the average historical return over an estimation window as a proxy for normal performance (Brown & Warner, 1980).

Abnormal returns are obtained by comparing what actually happened in the market with what would normally be expected. By calculating the difference between the actual and expected returns for a given day, the resulting value represents the abnormal return for that period (Brown & Warner, 1980).

Abnormal return at a single point in time only shows the market reaction for one index on one specific day, so it needs to be aggregated to capture the overall effect of an event more clearly. Cross‑sectional aggregation combines abnormal returns across all indices at a given time   to test whether, on average, the market as a whole reacts significantly on that day. Time‑series aggregation accumulates abnormal returns for a given index over several days in the event window to test whether the total impact of the event on that market is statistically significant over time.

The cross-sectional‑ aggregation (average abnormal return, AAR) is defined as follows:

Time‑series aggregation (Cumulative Abnormal Return, CAR) for index   over an event window from   to   is defined as: 

2.4. Research Hypothesis

In an event study framework, the starting point is to specify whether a particular announcement generates stock returns that deviate from their normal pattern. Kothari and Warner (2007) state that the central objective of an event study is to test whether average abnormal returns around the event window are statistically different from zero. Building on this view, the hypotheses in this study are formulated as follows:

H0  : There is no abnormal return on ASEAN stock indices around the announcement period of the initial U.S. reciprocal tariff.

Ha  : There is abnormal return on ASEAN stock indices around the announcement period of the initial U.S. reciprocal tariff.

H0  : There is no abnormal return on ASEAN stock indices around the announcement period of the pause of the U.S. reciprocal tariff.

Ha : There is abnormal return on ASEAN stock indices around the announcement period of the pause of the U.S. reciprocal tariff.

H0  : There is no abnormal return on ASEAN stock indices around the announcement period of the adjustment of the U.S. reciprocal tariff.

 Ha : There is abnormal return on ASEAN stock indices around the announcement period of the adjustment of the U.S. reciprocal tariff.

This study employs a quantitative research design using an event study approach. The main objective is to examine whether United States tariff announcements generate abnormal returns in six ASEAN stock indices, namely PSEi (Philippines), IDX Composite (Indonesia), FTSE Bursa Malaysia KLCI (Malaysia), STI (Singapore), SET Index (Thailand), and VN Index (Vietnam). The events analyzed are three policy announcements related to the U.S. reciprocal tariff regime in 2025: the initial tariff announcement on April 2, 2025 (Event 1), the pause announcement on April 9, 2025 (Event 2), and the tariff adjustment announcement on July 31, 2025 (Event 3).

3.1. Event Study Design

As suggested by Peterson (1989), there is no universal rule regarding the length of the estimation period and event window in event studies; therefore, these choices should be guided by prior literature and the nature of the event. This study adopts an estimation window of −120 to −11 trading days before the first event date, following Rao et al. (2025), to estimate normal returns in a period considered free from the influence of tariff announcements. The event window is set from t0 to t+3, that is, four trading days starting from the event date and extending three days after, consistent with Kaczmarek et al. (2025), to capture the immediate market reaction around each announcement. Owing to time zone differences between the United States and ASEAN markets, the effective event date for each market is defined as one trading day after the calendar date of the U.S. announcement, when the information becomes tradable on ASEAN stock exchanges (see Table 2).

Table 2. Time Zone difference between USA and ASEAN Region

Source: Processed by the author (2025)

3.2. Data Analysis Techniques

Data analysis in this study is conducted using the Statistical Product and Service Solutions (SPSS) version 27 and Microsoft Excel for data processing and computation. The analysis focuses on identifying whether there are significant abnormal returns in each stock index around the three tariff announcement events. Actual returns are computed from daily closing prices, whereas expected returns are estimated using a mean-adjusted‑ model over the estimation window. Abnormal returns are then obtained as the difference between actual and expected returns for each index and trading day within the event window.

The analytical procedure consists of three main stages. First, a descriptive analysis is used to summarize the behavior of abnormal returns across indices and events. Second, normality tests are applied to abnormal return and cumulative abnormal return data to determine the appropriate statistical procedure. Third, hypothesis testing is conducted using a one-sample t-test‑‑ for normally distributed data and the Wilcoxon signed-rank‑ test for non-normal‑ data to assess whether abnormal returns and cumulative abnormal returns differ significantly from zero around each event date.

4.1. Result

For more clarity, see Table 3.

Table 3. Descriptive Statistics Event 1 (Tariff Announcement)

Source: Processed by the author (2025)

Based on the descriptive statistics, the average abnormal returns from t0 through t+3 are consistently negative, indicating that ASEAN stock markets reacted adversely to the initial announcement of the U.S. “reciprocal” tariff. The largest decline occurs at t+2 (-5.28%), suggesting that the market reaction was not confined to the announcement day but intensified in the days following the event. The cumulative abnormal return CAR (0,3) of -9.25% confirms that the negative impact was substantial and persistent over the event window. Relatively large standard deviations, particularly at t0, t+3, and CAR (0,3), indicate considerable variations in the market responses across countries. In addition, the wide gap between the minimum and maximum values reflects heterogeneous reactions among the ASEAN markets. Although the magnitude of abnormal returns diminishes at t+3, the cumulative losses remain negative, suggesting that the market was not fully adjusted by the end of the observation period (see Table 4).

Table 4. Descriptive Statistics Event 2 (Tariff Pause Announcement)

Source: Processed by the author (2025)

Based on the descriptive statistics, the average abnormal returns from t0 through t+3 are consistently positive, indicating that ASEAN stock markets reacted favorably to the U.S. “reciprocal” tariff pause announcement. The largest increase occurs at t0 (4.55%), suggesting an immediate positive market response to the policy change. The cumulative abnormal return CAR (0,3) of 6.99% confirms that the positive impact was substantial and sustained over the event window. Relatively large standard deviations, particularly at t0 and t+1, indicate variation in the magnitude of market responses across countries. In addition, the gap between the minimum and maximum values reflects short-term heterogeneity in market reactions. Although minor corrections appear after the announcement day, the consistently positive abnormal returns and cumulative gains suggest that the market largely adjusted to the information by the end of the observation period (see Table 5).

Table 5. Descriptive Statistics Event 3 (Tariff Adjustment Announcement)

Source: Processed by the author (2025)

Based on the descriptive statistics, the average abnormal returns from t0 through t+3 are consistently positive, indicating that ASEAN stock markets reacted moderately positively to the U.S. tariff adjustment announcement. The abnormal return at t0 is close to zero, suggesting that the announcement did not trigger a strong immediate market reaction, while higher mean values at t+1 to t+3 indicate a gradual positive response in the days following the event. The positive cumulative abnormal return CAR (0,3) of 2.24% confirms that the overall market impact was favorable, although smaller in magnitude compared to the tariff pause announcement. The relatively lower standard deviations across the event window suggest more stable and homogeneous market reactions among ASEAN countries. In addition, the narrower range between the minimum and maximum values indicates limited volatility, implying that investors largely perceived the tariff adjustment as neutral to mildly positive information that was absorbed without significant market disruption.

Prior to conducting the parametric hypothesis testing, a normality test was performed to examine whether the abnormal return data were normally distributed. The normality assumption was determined using the one-sample Kolmogorov–Smirnov test to evaluate the appropriateness of using parametric statistical tests in the event study analysis (see Table 6).

Table 6. Normality Test Event 1 (Tariff Announcement)

Source: Processed by the author (2025)

The normality test results for Event 1 indicate that only abnormal returns at t0 are not normally distributed, with a significance value of 0.025 (< 0.05). Meanwhile, abnormal returns at t+1, t+2, t+3, and CAR (0,3) are normally distributed, as indicated by significance values above 0.05 (see Table 7).

Table 7. Normality Test Event 2 (Tariff Pause Announcement)

Source: Processed by the author (2025)

The normality test results for Event 2 show that only abnormal returns at t+1 are not normally distributed, with a significance value of 0.016 (< 0.05). Meanwhile, abnormal returns at t0, t+2, t+3, and CAR (0,3) are normally distributed, as indicated by significance values above 0.05 (see Table 8).

Table 8. Normality Test Event 3 (Tariff Adjustment Announcement)

Source: Processed by the author (2025)

The normality test results for Event 3 show that only abnormal returns at t+3 are not normally distributed, with a significance value of 0.047 (< 0.05). Meanwhile, abnormal returns at t0, t+1, t+2, and CAR (0,3) are normally distributed, as indicated by significance values above 0.05

Different statistical procedures were applied in hypothesis testing based on the results of the normality tests. For observations that are normally distributed, a one-sample t-test was employed to examine whether abnormal returns and cumulative abnormal returns are significantly different from zero. Meanwhile, for observations that do not satisfy the normality assumption, the Wilcoxon signed-rank test was used as a nonparametric alternative. This approach ensures that the hypothesis testing procedure is consistent with the underlying distributional properties of the data (see Table 9).

Table 9. Hypothesis Test Results for Event 1 (Tariff Announcement)

Source: Processed by the author (2025)

Based on the hypothesis testing results for Event 1, the announcement of the initial U.S. “reciprocal” tariff generated statistically significant abnormal returns in ASEAN stock markets on most days within the event window. Abnormal returns at t0 are significant according to the Wilcoxon signed-rank test, indicating an immediate market reaction on the announcement day. In addition, the one-sample t-test results show significant negative abnormal returns at t+1 and t+2, suggesting that the adverse market response intensified on days following the announcement. In contrast, abnormal returns at t+3 are not statistically significant, implying that the immediate shock began to dissipate by the third day after the event. The cumulative abnormal return CAR (0,3) is significantly negative, confirming that the initial tariff announcement had a persistent negative impact on ASEAN stock markets over the event window (see Table 10).

Table 10. Hypothesis Test Results for Event 2 (Tariff Pause Announcement)

Source: Processed by the author (2025)

Based on the hypothesis testing results, abnormal returns at t0 are statistically significant and positive (p = 0.002 < 0.05), indicating an immediate favorable market reaction to the U.S. tariff pause announcement. At t+1, the abnormal return is not significant (p = 0.753), suggesting a short-term correction following the initial response. At t+2, the one-sample t-test shows a highly significant positive abnormal return (p < 0.001), indicating a renewed market response as investors further processed the information. At t+3, the abnormal return is not statistically significant (p = 0.201), implying that the market reaction began to stabilize. The cumulative abnormal return (CAR) (0,3) is positive and statistically significant (p = 0.002), confirming that the tariff pause announcement generated a significant overall positive impact on ASEAN stock markets. Therefore, the null hypothesis (H₀) is rejected, indicating the presence of abnormal returns around the announcement period (see Table 11).

Table 11. Hypothesis Test Results for Event 3 (Tariff Adjustment Announcement)

Source: Processed by the author (2025)

Based on the hypothesis testing results, abnormal returns at t0 and t+1 are not statistically significant (p = 0.897 and p = 0.205), indicating that the tariff adjustment announcement did not trigger an immediate market reaction. At t+2, the abnormal return becomes positive and statistically significant (p = 0.015), suggesting a delayed market response as investors reassessed the implications of the policy adjustment. At t+3, the abnormal return is also significant based on the Wilcoxon signed-rank test (p = 0.046), reinforcing evidence of continued positive adjustment. The cumulative abnormal return (CAR) (0,3) is positive and statistically significant (p = 0.010), confirming that the tariff adjustment announcement generated a favorable overall impact on ASEAN stock markets. Therefore, the null hypothesis (H₀) is rejected, indicating the presence of abnormal returns around the announcement period, albeit with a delayed reaction.

4.2. Discussion

The initial tariff announcement on April 2, 2025, triggered a broad and significantly negative reaction across ASEAN stock markets. The presence of significant negative abnormal returns on t plus one, t plus two, and cumulatively over the event window indicates that the market response intensified after the announcement day rather than being fully absorbed immediately. This delayed reaction suggests that investors required additional time to reassess the economic implications of higher trade barriers, particularly regarding export competitiveness, supply chain disruptions, and potential retaliatory measures. Vietnam and Singapore experienced the largest cumulative losses, reflecting their high exposure to global trade and integration into international supply chains. These results are consistent with Kaczmarek et al. (2025) and Kapar et al. (2025), who document sharp negative stock market reactions following U.S. tariff announcements, especially in economies with strong export linkages.

From a market efficiency perspective, the persistence of negative abnormal returns beyond the announcement day challenges the semi-strong form of the efficient market hypothesis, which posits that prices should fully and instantaneously reflect publicly available information. Instead, the observed pattern indicates delayed price adjustment, potentially driven by information processing costs, heterogeneous investor expectations, or heightened uncertainty surrounding the future path of trade policy. Interestingly, Singapore, despite maintaining a trade surplus with the United States, also exhibited significant cumulative losses. This finding contrasts with that of Rao et al. (2025), who argue that surplus-trading countries tend to display more resilient or even positive abnormal returns following tariff announcements. The results suggest that trade-balance status alone does not insulate equity markets from external shocks, particularly in highly open financial centers, where listed firms have extensive global exposure.

In contrast, the tariff-pause announcement on April 9, 2025, generated a strong and largely positive market response across ASEAN. Significant positive abnormal returns on the announcement day and cumulatively over the event window indicate that investors interpreted the pause as favorable news that reduced the likelihood of further trade escalation. The sharp rebound following the severe losses from the initial announcement highlights the sensitivity of ASEAN markets to policy signals from the United States. While minor corrections were observed in some markets on subsequent days, the overall positive cumulative abnormal return suggests that the reduction in policy uncertainty outweighed lingering concerns about trade restrictions. This pattern supports the findings of Wengerek et al. (2025), who emphasize that stock market reactions to tariff-related announcements are driven more by changes in investor expectations and uncertainty than by tariff levels alone.

The rapid shift from negative to positive abnormal returns within a short interval between the first and second events also provides evidence of possible overreaction followed by price correction in certain markets. The immediate surge on the announcement day, followed by weaker or mixed reactions thereafter, suggests that some investors may have initially overestimated the positive impact of the tariff pause before reassessing its long-term implications. This behavior is consistent with behavioral finance explanations and further indicates that ASEAN stock markets do not always conform to strict informational efficiency during periods of heightened policy uncertainty.

The tariff adjustment announcement on July 31, 2025, elicited a more moderate and heterogeneous response than earlier events. While daily abnormal returns were mostly insignificant, the cumulative abnormal return over the event window was positive and statistically significant. This finding implies that the market response unfolded gradually, rather than being concentrated on a single trading day. The absence of strong immediate reactions suggests that much of the information related to tariff adjustments had already been anticipated by investors, resulting in a smoother price adjustment process. This outcome aligns with Cocozza and Gallo (2025), who show that when trade policy changes are largely expected or represent implementation rather than surprise, market reactions tend to be smaller and more dispersed over time.

This study examines stock market reactions in ASEAN to three U.S. reciprocal tariff-related announcements in 2025 using an event study approach. The findings show that U.S. trade policy announcements generated statistically significant abnormal returns in ASEAN stock markets, with the direction and magnitude of the response varying across policy events. The initial tariff announcement produced a broad and persistent negative reaction, the tariff-pause announcement triggered a strong positive rebound, and the tariff adjustment announcement resulted in a more moderate and cumulative response. These results indicate that U.S. trade policy remains a key external shock influencing ASEAN financial markets.

From a market efficiency perspective, the evidence suggests that ASEAN stock markets are not fully consistent with the semi-strong form of the efficient market hypothesis. The presence of delayed reactions, cumulative abnormal returns, and asymmetric responses across events indicates that prices do not always adjust instantaneously to publicly available information. Investor behavior appears to be influenced by uncertainty, expectation revision, and learning effects, particularly during periods of rapid policy changes and heightened geopolitical risk.

The contrasting reactions across the three events also highlight the importance of policy signals rather than tariff levels alone. While the initial announcement heightened uncertainty and triggered sell-offs, the pause announcement was interpreted as favorable news that reduced perceived trade risk, leading to a sharp market recovery. The more muted response to the tariff adjustment announcement suggests that investors had gradually incorporated trade policy risks into asset prices, resulting in a smoother adjustment process.

Overall, this study contributes to the literature on trade policy and financial markets by providing recent evidence from ASEAN during a new episode of U.S. protectionist policy. The findings have practical implications for investors and policymakers, emphasizing the sensitivity of emerging and open markets to external policy shocks and the role of communication in shaping market expectations. Future research may extend this analysis by incorporating firm-level data, alternative expected return models, or longer event windows to further explore the transmission of trade policy uncertainty across financial markets.