A major problem associated with agency conflicts is managerial short-termism, i.e., an excessive focus of executives on short-term results at the expense of long-term growth (Edmans et al. 2022). Surveys show that 78% of executives is willing to sacrifice long-term firm value to smooth earnings (Graham et al. 2005), while archival studies show that executives tend to cut R&D investments near the end of their tenure, increasing short-term profits to presumably secure bonuses (Dechow and Sloan 1991). Furthermore, Jensen (2005) shows that executives might have an interest in keeping firm stock overvalued, which in turn is detrimental to the core value of the firm. While it may be optimal for (incumbent) shareholders to incentivize managerial short-termism in speculative stock markets (Bolton et al. 2006), the detrimental long-term effects of these actions on firm value are generally regarded as problematic from a societal point of view. Following Marinovic and Varas (2019), two views on managerial short-termism can be distinguished in the literature: one studying executive behavior under existing incentives and another studying the optimal design of compensation contracts. In this article, we follow the first perspective by empirically assessing how different remuneration structures affect executive behavior.

Next to managerial short-termism, the literature recognizes an additional risk associated with equity-based executive compensation that is relevant to our study: the concept of gaming the system. Gaming the system describes the tendency of executives to focus on increasing the chances that performance targets are met and stock options will be paid out to maximize their personal compensation (Fahlenbrach and Stulz 2011). Following financial theory, the higher the volatility of the underlying, the more likely it is for an option to be in the money when vested. In practice, this could be achieved by executives (excessively) increasing leverage or, following our argumentation, seeking out short-term value gains through strategically engaging in regulatory noncompliance (‘strategic noncompliance’). However, strategic noncompliance bears the risk of incurring regulatory sanctions in the medium or long-term.

Strategic noncompliance, defined as the deliberate or tacitly sanctioned breach of regulatory requirements, has been widely examined in the literature. While previous research examined both causes and consequences of strategic noncompliance, our investigation particularly builds on studies addressing the ways in which executives personally affect regulatory violations. Prior research in this field has shown that, for example, higher frequency of visits by top executives can reduce the occurrence and scope of facility-level misconduct (Heese and Pérez-Cavazos 2020). In addition, evidence suggests that executives’ personal liability for wage theft affects firm regulatory noncompliance (Raghunandan 2021).

Most closely related to our own approach are recent studies by Chircop et al. (2025), Gencer (2021), and Park et al. (2025), which, too, investigate the relationship between executive compensation and regulatory noncompliance in the US. Chircop et al. (2025) find that equity-based incentives positively affect workplace misconduct; Gencer (2021) notes that compensation structures that exclude certain cost savings from executive rewards mitigate corporate violations across various regulations; and the results by Park et al. (2025) indicate that option-loaded compensation plans incentivize managers in the restaurant industry to engage in more regulatory noncompliance than their industry average. In addition, Hass et al. (2016) show that equity incentives increase the likelihood that executives of Chinese listed firms commit regulatory violations. By using a broad set of payment components and regulatory violations, we extend these previous studies. In addition, our approach differs from that of, e.g., Chircop et al. (2025), as we only consider companies that were fined for regulatory violations at least once during the sample period. This approach improves comparability between firms in the sample by eliminating differences between violating and non-violating firms that may otherwise impact our results.

For the purpose of our investigation, we assume a degree of transitivity in the behavior of executives and the actors that engage in behavior that is ultimately in violation of regulations and is fined accordingly. Often executives do not directly violate regulations, and do not personally receive fines for their behavior. These regulatory infractions often occur in lower levels of decision-making in the firms, or even in subsidiaries. In particular, the dataset used consists mainly of employment-related, environment-related, and safety-related fines, which are traditionally associated with lower levels of decision-making. Executives are, however, able to influence firm decision-making on multiple levels, and that decision-making influences strategic noncompliance across the firm. In particular, Mei et al. (2023) found that executive-level ideology can significantly shape firm behavior by affecting the tone and culture at the top and cascading through organizational decision-making structures. Accordingly, our theoretical framework assumes that executives have significant influence over both the frequency and extent of regulatory infractions committed by the firms they lead.

The chosen approach does come with limitations. Regulatory violations may not always reflect deliberate noncompliance, as they can result from limited knowledge or accidental incidents, such as technical malfunctions causing environmental fines. However, given the large sample and prevalence of repeat offenders, unintentional violations are likely rare and are disregarded in this analysis (Duong et al. 2024).

Overall, previous research suggests that executive remuneration structures may incentivize strategic noncompliance, through mechanisms such as short-termism or gaming the system behavior. Thereby, it is important to consider the balance between equity-based and non-equity-based pay components as both can have distinct effects. While results by Chircop et al. (2025) and Park et al. (2025) indicate that higher amounts of equity-based incentives increase regulatory noncompliance, based on theory the opposite might hold true. Higher amounts of equity-based pay could decrease regulatory violations if executives’ fear of losing income when violations are detected prevails. Similarly, higher amounts of ‘safe’, i.e., non-equity-based, compensation may incentivize or mitigate strategic noncompliance. Accordingly, when considering the overall structure of executive remuneration, i.e., the relative importance of both components, the direction of the resulting overall effect remains unclear. Based on these considerations, we formulate our first hypothesis (Ha1) nondirectionally. The second set of hypotheses (Ha2.1 and Ha2.2) aims to clarify the underlying mechanism. Due to the aforementioned unclear relationship of equity-based and non-equity-based remuneration and strategic noncompliance, any result for Ha1, i.e., both indicating a positive and negative relationship, can be caused by changes in either of the two pay components. We illustrate this dual hypothesis problem inherent in Ha1 for the case of a positive relationship between the proportion of equity-based remuneration and strategic noncompliance in Figure 1. As depicted in the figure, the positive association can result from a negative impact of non-equity linked remuneration, a positive impact of equity-linked remuneration, or a combination of both.

Figure 1.

Schematic overview of dual hypothesis problem.

Ha1: The remuneration structure of executives influences the number of regulatory violations committed by the firm they lead.

Ha2.1: Equity-based executive remuneration influences the number of regulatory violations committed by the firm they lead.

Ha2.2: Non-equity-based executive remuneration influences the number of regulatory violations committed by the firm they lead.

Data on regulatory noncompliance is collected from the Violation Tracker of Good Jobs First. The entire database, as of the 20^th of January 2025, is processed, containing a total of 662,591 fines issued by U.S. government agencies for firm regulatory violations during the period from 2000 to 2025. Throughout our main analysis, we operationalize regulatory noncompliance as the number of fines levied on a specific firm in a specific year (IF). Table 1 provides an overview of all variables employed in the analysis, including their definitions, data sources, and corresponding abbreviations.

Data on executive remuneration originates from BoardEx, and controls for important firm characteristics are extracted from the LSEG database. We operationalize executive remuneration structure as the proportion of equity-based remuneration (EBRR). The EBRR is calculated as the total executive pay tied to equity-based incentives relative to the executive’s total compensation. Accordingly, total compensation includes all relevant components such as fixed salary, bonuses and options.¹ To further specify the mechanism underlying Ha1, we separately analyze possible explanatory channels in our tests of Ha2.1 and Ha2.2. Consequently, we use several operationalizations for equity and non-equity-based compensation to increase the robustness of our results. We use two operationalizations for non-equity-based payments. Our first measure is executive salary (ES). Executive salary refers to a fixed, annual payout of cash, without regard to company performance. As the overall salary and changes thereof likely depend on firm size, we additionally use a scaled measure of executive salary relative to firm size (RES). To operationalize equity-based remuneration we use three measures in total: relative bonuses (RB), i.e., performance-based bonuses relative to firm size, the estimated value of awarded options (EVOA), and the wealth delta (WD), which quantifies the dollar sensitivity of executive compensation to firm performance. While all of these measures are well-established in the executive compensation literature (Edmans and Gabaix 2016; Meulbroek 2001; Ofek and Yermack 2000), to the best of our knowledge, no prior research has utilized these specific measures on such a broad sample of firms and violations to examine the association between executive incentives and strategic noncompliance.

Turning to data constraints, this study is subject to three key limitations regarding the availability and structure of the dataset. First, most of the fines are awarded to subsidiaries, while the remuneration data of BoardEx is only available for parent-level executives. A conservative approach, only matching parent-level fines and remuneration data, results in a total of 1917 firm-year observations from 245 unique firms, which is too few for the proposed analysis. Consequently, fines incurred by subsidiaries are attributed to their respective parent firms, while all other variables remain at the level of the parent organization. Given parent executives’ influence on subsidiaries’ strategic decisions, there is theoretical justification for linking subsidiary outcomes to parent-level executive data and similar aggregations have been applied by previous research (Chircop et al. 2025).

Second and third, data on executive remuneration is limited to firms publicly listed in the United States, and the Violation Tracker dataset contains both current and historical firm identifiers (ISINs). Historical ISINs concern the ISIN of firms at the time of the violation. Firms’ ownership changes through mergers and acquisitions result in different historical and current ISINs. To prevent a mismatch between our remuneration and violation data, we allocate subsidiary violations to parent-level executive data based on historical ISINs.

The final sample resulted in a total of 6150 firm-year observations from 684 unique firms. As noted above, this sample only includes firms that have been fined for regulatory violations at least once during the sample period. Summary statistics based on raw data for key variables are shown in Table 2. As can be seen here, numerous variables exhibit elevated standard deviations, attributable to substantial firm heterogeneity within the dataset. The inclusion of both small enterprises and major corporations (e.g. Microsoft, Amazon) enhances analytical scope but amplifies statistical dispersion.

Where necessary, we transform the raw data of our measures for the following analysis to ensure methodological validity. For example, the distribution of Total Firm Assets (TFA) is heavily skewed. We account for this by scaling the variable, a standardization process involving centering the mean at zero. Other variables are transformed using a log1p transformation. In addition, for leverage (DTE), some extreme outliers were found and corrected by excluding the top and bottom 1% of results. Overall, transformed variables are EBRR, EVOA, RB, ROE, SR, PROF, DTE, RES and AAF.

Given the unique characteristics of the sample, this study employs panel data analysis using a negative binomial fixed effects model. This approach controls for differences between firms and over time, preventing bias from unobserved characteristics in the sample that vary across firms and years. The inclusion of entity and time fixed effects is theoretically well-justified (Chircop et al. 2025; Eugster et al. 2024). It is assumed that firm-level heterogeneity is present in the data and the inclusion of time fixed effects captures unobserved temporal variation (Park et al. 2025). For example, unobserved factors such as regulatory reforms, or shifts in enforcement priorities may influence the observation of regulatory violations. The chosen approach is further supported by indicators for the explanatory power of our statistical model. However, it is noteworthy that while firm fixed effects account for much variation in regulatory noncompliance, a model including both firm and year (two-way) fixed effects offers little additional explanatory power based on the Akaike information criteria (AIC), Bayesian information criteria (BIC), and the Log-Likelihood.²

We test our first hypothesis, i.e., whether the structure of executive remuneration affects the number of regulatory violations committed by firms, using the baseline model in formula 1. Given that the EBRR operationalizes executive remuneration structure, we are most interested in estimation results for β₁. TFA, ROE, SR, PROF, and DTE denote the respective firm-level control variables, while FEs denote firm and year fixed effects. To account for the sensitivity of our results to different controls and increase the robustness of our results, we use a stepwise approach for the inclusion of firm-level controls and fixed effects in our analysis.

IF_it = β₀ + β₁ EBRR_it + β₂ TFA_it + β₃ ROE_it + β₄ SR_it + β₅ PROF_it + β₆ DTE_it + FEs + ε_it (1)

Based on our methodological approach so far, the exact channel through which executive remuneration affects regulatory noncompliance remains unclear. Either changes in equity-based, non-equity-based, or both compensation components can result in changes in the EBRR. To identify the underlying mechanisms, we test Ha2.1 and Ha2.2 based on the model presented in formula 1, substituting each of the measures of equity-based (RB, EVOA, WD) and non-equity-based (ES, RES) compensation components for EBRR in separate regressions.

The results of our analysis regarding our first hypothesis are shown in Table 3. Across all model specifications, the proportion of executive equity-based compensation (EBRR) is positively associated with regulatory noncompliance. Specifically, an increase in EBRR of 20 percent (0.4 to 0.48) is associated with an increase in the number of fines of 16.5% based on the estimation results in column 6 of Table 3. The size and statistical significance of the effect remain mostly stable when accounting for different firm-level control variables; they only drop slightly when accounting for firm size (Total Firm Assets).

Nevertheless, when additionally including year fixed effects in the model both the magnitude and statistical significance of the effect decline substantially. This suggests that the baseline association is, at least partly, driven by year-specific factors. However, another possible explanation is that the structure of executive compensation exhibits time-related patterns that overlap with the year fixed effects. A visual examination of the development of the EBRR over time supports this suspicion. The EBRR follows a clear time trend with a steeply increasing proportion of equity-based remuneration in the years preceding the financial crisis followed by small fluctuations around a generally upward-sloping trend. Accordingly, the estimates in column 7 of Table 3 should be interpreted with caution, as the specification may provide limited scope to isolate the effect of relative equity-based compensation (EBRR).

Regarding firm-level characteristics, our results show that especially increasing firm size is positively associated with regulatory noncompliance. An intuitive interpretation would be that larger firms have more opportunities for regulatory violations due to their expansive and widespread operations.

Overall, we conclude that the results affirm our theoretical expectations. Increases in equity-based pay relative to overall compensation appear to incentivize CEOs to engage in strategic noncompliance. To further specify the underlying mechanism, we now turn to our analysis of different equity-based and non-equity-based pay components.

The results of our tests of Ha2.1 and Ha2.2 are shown in Table 4. Regarding non-equity-based remuneration, operationalized as executive salary (ES), no significant effects are found. However, introducing Relative Executive Salary (RES), executive salary scaled by firm size, yields a significant negative association. Large differences exist in the dataset between different firms’ base pay for executives, necessitating this measure. This finding supports the hypothesis that the observed effect operates, at least in part, through the non-equity remuneration channel, i.e., a ‘safer’ remuneration structure can mitigate executives’ likelihood to engage in strategic noncompliance.

With respect to equity-based remuneration, neither performance-based bonuses (RB) nor the estimated value of awarded options (EVOA) has a significant association with regulatory noncompliance. In contrast, the wealth delta (WD), which quantifies the dollar sensitivity of executive compensation to firm performance, exhibits a strong positive relationship with imposed fines. This suggests that WD captures the equity channel through which EBRR interacts with regulatory noncompliance. Both WD and EBRR display positive coefficients with the number of fines, indicating that increasing exposure to equity incentives may encourage executives to engage in or tolerate strategic noncompliance, thereby reinforcing the relevance of the equity-based remuneration mechanism.