The Determinants of CEO Compensation in the Banking Sector: A Comparison of the Influence of Cross-Listing and Loan Growth in Developed Versus Developing Countries

Abstract

This study explores the determinants of CEO compensation in the banking sector, focusing on cross-listing and loan growth. Using 8800 observations from 45 countries spanning 2004 to 2018, the analysis reveals significant differences in compensation structures between developed and developing economies. The findings show that CEO stock options and restricted stock compensation are positively correlated with cross-listing in the U.S. market, with a stronger effect in developing countries. Loan growth is associated with higher incentive-based pay but lower fixed salaries, aligning CEO compensation with performance-driven growth and risk management. These results underscore the role of regulatory environments and institutional quality in shaping executive pay, offering valuable insights for policymakers, financial institutions, and investors navigating a globalized banking sector.

1. Introduction

Chief executive officer (CEO) compensation is crucial for executive recruitment, retention, and motivation, while also shaping organizational dynamics. In the banking sector, Hagendorff and Vallascas (2011) document that the structure of CEO compensation incentives significantly impacts financial stability. Understanding its determinants is therefore essential (Edmans et al., 2023). Extensive research on CEO compensation has primarily focused on developed economies, such as the United States, the United Kingdom (Conyon & Murphy, 2000; Core et al., 2003, 2008; Guay et al., 2002; Ahmed et al., 2022; Ahmed et al., 2023), and Japan (Abe et al., 2005). In the financial sector, CEO compensation has been subject to considerable media scrutiny and additional disclosure requirements in certain countries, including the US and the UK (Hendriks et al., 2023). Compared to developed nations, developing countries often exhibit weaker governance mechanisms and lower institutional quality, both of which influence CEO compensation structures. However, limited studies have examined CEO compensation across both developed and developing countries. This study addresses this gap by analyzing bank CEO compensation across 45 countries, using a unique dataset over the 15-year span of 2004–2018. Specifically, we investigate how the cross-listing of a bank on multiple stock exchanges influences CEO compensation compared to non-cross-listed banks. Furthermore, we explore the effects of cross-listing between developed and developing economies.

Cross-listed banks generally face greater regulatory scrutiny due to their adherence to international governance frameworks (“bonding”). Such banks also tend to be larger institutions, which often translates to lower stock-based compensation but higher incentive-based pay for CEOs. While previous research has predominantly focused on CEO compensation in less-regulated non-banking firms, banks are subject to greater regulatory oversight, which affects their compensation structures. Studies indicate that bank CEOs receive lower compensation than their non-bank counterparts (John et al., 2010; Conyon & He, 2016), and that their pay is often tied to risk exposure (Conyon et al., 2011; Handorf, 2015). Larger banks generally offer higher CEO compensation (Fernandes et al., 2013; Bennett et al., 2021), with compensation structures varying based on governance mechanisms and monitoring intensity (Fernandes et al., 2013; Hartzell & Starks, 2003; Melis & Rombi, 2021). Additionally, Khayati and Ariail (2015) suggest CEOs who face greater monitoring pressures may negotiate higher compensation packages.

Cross-listing, the practice of listing a firm’s shares on multiple exchanges, often in different countries, imposes additional monitoring pressures and aligns firms with international governance standards (Coffee, 2002). The additional benefits of cross-listing include a reduction in investment barriers, enhancement of transparency, and improvement of liquidity (Coffee, 1999; Stulz, 2022). In developed markets with strict governance norms, cross-listing increases bank visibility and access to capital markets (Coffee, 2002; Doidge et al., 2004). Based on these factors, we hypothesize that the number of exchanges on which a bank is listed significantly influences CEO compensation, with differential effects between developed and developing economies.

Our empirical analysis reveals that CEO stock options and restricted stock compensation are positively correlated with cross-listing in the U.S. market across all sample markets. In developing markets, the results differ. While CEO salaries and bonuses are positively correlated with cross-listing, CEO incentive compensation is negatively correlated with cross-listing. This outcome suggests that banks in developing economies may prefer fixed compensation structures to offset the increased level of monitoring associated with cross-listing. Notably, the impact of cross-listing on CEO compensation is stronger for banks in developing countries than for those in developed economies.

Unlike non-financial firms, banks prioritize loan growth, which can boost short-term profitability but also heighten credit risk (Laeven & Levine, 2009). Our findings show that an increase in loan growth is positively correlated with CEO incentive and bonus compensation but negatively correlated with CEO salary compensation. These results align with Chen et al. (2017), who found that bank CEO compensation increased in response to both external and internal asset growth.

We also find that higher CEO ownership leads to lower incentive compensation, consistent with the notion that greater ownership reduces external monitoring (Khan et al., 2005; Choe et al., 2014; Buigut et al., 2015). As CEO ownership increases, risk aversion rises due to the concentration of personal wealth in the firm (Demsetz & Lehn, 1985), leading firms to compensate with higher fixed pay. Existing studies predominantly focus on CEO ownership in non-banking firms, leaving a gap in research on CEO ownership in financial institutions. Additionally, our results indicate that banks listed on US exchanges allocate a higher proportion of incentive-based compensation to their CEOs compared to banks that are not US-listed. This finding complements prior research (Conyon et al., 2011; Fernandes et al., 2013), though these studies primarily examine non-banking firms.

To the best of our knowledge, this study is the first to comprehensively compare the effects of cross-listing, CEO ownership, and loan growth on bank CEO compensation across developed and developing countries. Bank CEO compensation structures differ significantly from those in other industries, particularly in developed countries, where they are subject to intense monitoring by regulators, institutional investors, and external stakeholders (John et al., 2010). Existing research has largely examined CEO compensation in general or the impact of cross-listing on all firms, without distinguishing the unique regulatory and governance challenges faced by banks.

The remainder of this paper is structured as follows: Section 2 reviews the literature and our hypotheses, Section 3 discusses data sources and variables, Section 4 presents the methodology and empirical results, and Section 5 concludes the study.

2. Literature and Hypotheses

CEO compensation is influenced by agency theory (Jensen & Meckling, 1976; Liu & Sickles, 2021), which highlights conflicts of interest between shareholders (principals) and managers (agents). To align managerial actions with shareholder interests, compensation packages often include performance-based incentives, such as equity, stock options, or bonuses, alongside fixed salaries that provide basic financial security. Incentive structures play a pivotal role in managing risk within banks. Armstrong et al. (2022) contend that the structural features of managerial compensation contracts significantly impact systemic risk. Efing et al. (2023) demonstrate that banks can utilize bonus compensation as a tool to reduce operating leverage, thereby minimizing the need for costly capital during periods of financial distress. Conyon et al. (2011) further reveal that US firms offer higher salaries, stock, and options incentives than their UK counterparts, reflecting differences in risk-adjusted compensation practices.

Governance structures and investor protections vary across economies (Hearn et al., 2017; Anginer et al., 2018). These variations significantly affect CEO compensation practices. Hendriks et al. (2023) found that CEO compensation in the financial sector has been under significant media scrutiny and has additional disclosure requirements in some countries such as the US and UK. Using Towers Perrin’s 1997 Worldwide Total Remuneration report, Murphy (1999) showed that US executives are paid more than their international counterparts. Similarly, Conyon et al. (2011) found that US firms adjust CEO pay based on risk-adjusted metrics, offering higher salary, stock, and options incentives compared to industry-matched UK firms. Fernandes et al. (2013) observed a convergence in CEO pay between US and non-US non-banking firms. They found that US CEOs were paid more due to higher institutional ownership, independent boards, and lower insider ownership, which necessitated higher equity compensation. Conyon and Murphy (2000) using an industry and size match comparison, found that US CEOs earn more salaries, options, and total compensation when compared to UK CEOs due to institutional and cultural differences.

Cross-listing imposes additional monitoring pressures and aligns firms with international governance standards (Coffee, 2002). The Legal Bonding Hypothesis (Coffee, 1999; Stulz, 2022) suggests that cross-listed firms adhere to stricter regulations, enhancing transparency and minority shareholder protection. While research shows that cross-listing improves access to capital markets and fosters institutional monitoring (Khurana et al., 2008; Burns et al., 2007; Doidge et al., 2004; Reddy et al., 2023), the extent of regulatory bonding may vary where institutional frameworks are weaker, such as in developing countries. Aharon et al. (2023) found that cross-listed banks that issue American depository receipts (ADRs) from stable banking sectors experience low stock price volatility.

CEO pay structures differ markedly between developed and developing countries. In developed economies, robust governance frameworks and strong investor protections often align CEO pay with firm performance. Conversely, in developing countries, weaker institutional environments and limited enforcement mechanisms drive a preference for fixed salaries over performance-based incentives. Because banks listed on US exchanges must comply with strict regulations, we hypothesize that CEOs in banks listed on US exchanges receive higher incentive compensation than CEOs in other banks.

Hypothesis 1.

The CEOs of banks listed on US exchanges receive a higher proportion of incentive-based compensation and a lower proportion of fixed salary compensation compared to the CEOs of non-US-listed banks. Additionally, the CEOs of banks from developing countries that are cross-listed in the US receive lower incentive-based compensation and higher fixed compensation relative to their counterparts.

The relationship between the number of exchanges a bank is listed on and CEO compensation reflects the interplay between external monitoring pressures and compensation design strategies. As the number of exchanges on which a bank is listed increases, its regulatory and governance requirements become more stringent. To ensure stability and predictability under heightened scrutiny, banks may prioritize fixed salary compensation over variable, performance-based pay. Fixed salaries offer CEOs a guaranteed income that reduces their exposure to uncertainty, particularly in highly regulated or complex cross-listed environments. This stability can be a strong incentive for executives, especially when their roles demand navigating intricate legal, cultural, and operational challenges across multiple jurisdictions. In contrast, incentive-based compensation, such as stock options or performance bonuses, may diminish as cross-listing increases. The enhanced monitoring associated with cross-listing often curtails excessive risk-taking, a key driver of variable pay in traditional incentive structures. For banks, this effect is magnified by the needs to manage systemic risks and to maintain public confidence. These needs make conservative compensation strategies more attractive. Most developed countries already have stringent regulations when compared to developing countries, and, hence, cross-listing may not affect the compensation structures of banks from developed economies. Banks in developing economies, when cross-listed in developed markets, face a dual dynamic. They must comply with stricter international governance standards while operating within domestic environments that often emphasize stability over risk-taking. In these cases, compensation structures may shift towards higher fixed salaries to attract and retain top talent capable of navigating these complex regulatory landscapes.

Hypothesis 2.

As the number of exchanges on which a bank is listed increases, CEO incentive compensation decreases and bonus compensation increases. This impact is higher for the CEOs of banks from developing countries due to differences in institutional quality and governance standards.

Loan growth, while boosting short-term profitability, can elevate credit risks if not carefully managed (Laeven & Levine, 2009). To balance growth objectives with prudent risk management, banks often reward CEOs with higher incentive compensation linked to performance targets. These banks may also limit fixed salaries to reinforce a pay-for-performance structure. Naili and Lahrichi (2022) reviewed 69 studies on bank credit risk and concluded that a substantial body of research finds a positive relationship between loan growth and non-performing loans, which contributed to bank failures during the 2007–2008 financial crisis. Consequently, compensating CEOs with options-based incentives increases their risk aversion, prompting them to prioritize safer investments to protect their personal wealth. Hilscher et al. (2021) found that a bank’s stockholders determine executive pay in the form of equity compensation and executives choose the risk levels on assets. Prior to the 2008 financial crisis, both compensation and asset risk were excessive. However, after the Dodd–Frank Act, regulators set limits on asset risk and executive compensation to promote more optimal risk-taking practices. Chen et al. (2017) reported that CEO compensation positively correlates with merger and internal asset growth.

Hypothesis 3.

Banks with higher loan growths pay higher incentive compensation and lower salaries to align CEO pay with performance objectives while managing risks effectively.

3. Data and Methodology

3.1. Data

The accounting and cross-listing data for banks were collected from Capital IQ (CIQ) for the period 2004–2018. The sample was restricted to depository institutions by applying the following criteria: (a) the bank’s primary Standard Industrial Classification code must fall within the range of 6000–6099 (institutions classified as “depository institutions”), and (b) the banks must have been classified as either “Diversified Banks” or “Regional Banks” according to CIQ industry descriptions. We excluded institutions with SICs of 6100 (non-depository institutions), 6200 (security and commodity brokers), and 6300 (insurance carriers) because the regulations on these institutions differ from those on depository institutions. We included observations of banks having positive total compensations and excluded banks with negative or missing total compensations. Banks with zero deposits or loans, such as non-depository brokerages and insurance companies, were excluded from the sample.

To mitigate the influence of outliers, all continuous variables except for country-level controls were Winsorized at the 1st and 99th percentiles. The final sample was structured as an unbalanced panel, containing a total of 8800 observations from 45 countries.

3.2. Regression Model

We utilized the following standard OLS regression model:

CEOPAY_i,t = α₀ + α₁CEOWN_i,t + α₂US_DUMMY_i,t + α₃DE_DUMMY_i,t + α₄NUM_i,t + α₅LGRTH_i,t + α₆INSTOWN_i,t + α₇CONTROL_i,t + α₈fe_i,t + ε_i,t

(1)

3.2.1. Dependent Variables

CEOPAY: The dependent variables in our analysis encompassed various components of CEO compensation. TOTALCOMP represented the CEO’s total compensation, calculated as the sum of salary, options compensation, restricted stock grants (RSGs), and cash bonuses. CEOSALIN included both CEO salary and incentive compensation. CEOSALARYCOMP captured only the CEO’s salary compensation, while CEOICOMP represented the CEO’s incentive compensation, encompassing both options and restricted stock grant compensation. CEOOPTIONSCOMP specifically referred to the CEO’s options compensation, whereas CEORSGCOMP denoted restricted stock grant compensation, which constitutes a long-term equity component subject to restrictions on the vesting period. Lastly, CEOBONUSCOMP represented the CEO’s cash bonus compensation, expressed as a percentage of the total compensation.

Figure 1 presents the mean values of the CEO compensation components based on a dataset with 8800 observations from the banking sector across 45 countries over the period 2004–2018. Most of the CEO compensation takes the form of a salary, followed by incentives-based and bonus compensation.

3.2.2. Independent Variables

We incorporated several independent variables in our analysis. The key explanatory variables of interest included a dummy variable indicating whether a bank was listed on U.S. exchanges (US_DUMMY), a dummy variable identifying banks located in developing countries (DE_DUMMY), the number of exchanges on which a bank was listed (NUM), and the loan growth rate from the previous year (LGRWTH).

CONTROL represents a vector of control variables. To account for firm performance, we included return on equity (ROE) as an explanatory variable. Pennacchi and Santos (2021) found that ROE was the most common metric used in bank manager’s compensation contracts, whereas non-banks used EPS. Bennett et al. (2021) found that bank CEOs received lower fixed and equity compensation compared to non-bank CEOs, with CEO pay primarily influenced by return on equity rather than stock returns. Crawford et al. (2021) found that the CEO-to-employee total pay gap was lower in banks when compared to other industries. Other studies have also documented the relationship between executive compensation and firm performance (e.g., Carpenter & Sanders, 2002; Conyon, 2006; Ahamed, 2022). To control for firm risk, we included the standard deviation of monthly returns over the fiscal year (RISK). Additionally, we incorporated a dummy variable (DODD) to indicate data from the post-Dodd–Frank Act period, reflecting regulatory changes that may have influenced CEO compensation structures (Kwak et al., 2021). Following the enactment of the Dodd–Frank Act, all banks listed in the US had to comply with stricter regulations, aligning them with the regulatory framework applied to domestic banks. Similarly, several other countries such as China, the UK, Australia, Canada, and several European countries introduced financial reforms comparable to the Dodd–Frank Act, leading to enhanced regulatory oversight and more stringent compliance requirements. Kleymenova and Tuna (2021) mention that after the US passed the Dodd–Frank act, that regulated, monitored, and modified the level and structure of executive compensation in the financial sector, other countries followed the US (the UK being one of the first to follow) by implementing similar compensation regulations. Handorf (2015) notes that U.S. banks adjusted CEO compensation pre- and post-financial crisis to reward safer loan portfolios and higher capital adequacy. This adjustment underscores the importance of considering such regulatory impacts.

External monitoring is another critical factor in mitigating agency costs. John et al. (2010) show that increased outside monitoring enhances the sensitivity of bank CEO compensation to performance by reducing risk-shifting incentives. Similarly, Hartzell and Starks (2003) and Mangel and Singh (1993) highlight the role of institutional investors as monitors, increasing CEO accountability and equity-based compensation. Khayati and Ariail (2015) further suggest that enhanced monitoring in commercial banks enforces transparency, limits risky managerial behavior, and may encourage CEOs to negotiate higher salaries. Based on this literature, we included the following monitoring-related variables: the percentage of shares owned by the CEO (CEOWN), the percentage of shares owned by institutional investors (INSTOWN), and an index provided by Barth et al. (2013) ranging from 1 to 6 that measures the level of external monitoring in a country (EXTMOR).

Additional control variables included the percentage of non-performing loans (NPL), the percentage of bad loans (BLOANP), and dividend payments scaled by total assets (D_ASSET). Firm size was controlled for using the natural logarithm of total assets (ASSET). Finally, year-fixed effects were incorporated (fe) to account for time-specific influences.

4. Empirical Results

4.1. Descriptive Statistics and Correlation Matrix

Table 1. Summary of statistics.

Variable	Obs.	Mean	Std.	Q1	Median	Q3
CEOSALARYCOMP	8800	0.7188	0.2695	0.5277	0.7589	0.9960
CEOICOMP	8800	0.1547	0.2362	0	0	0.2607
CEOOPTIONSCOMP	8800	0.0352	0.0880	0	0	0
CEORSGCOMP	8800	0.1195	0.2023	0	0	0.1863
CEOBONUSCOMP	8800	0.1265	0.1825	0	0	0.2262
CEOWN	8800	0.0039	0.0122	0	0	0.0015
US_DUMMY	8800	0.8378	0.3686	1	1	1
DE_DUMMY	8800	0.1700	0.3757	0	0	0
NUM	8800	1.7443	2.1673	0	1	2
ROE	8800	0.0750	0.1098	0.0557	0.0896	0.1231
NPL	8800	1.2788	2.0442	0.2567	0.6296	1.4866
BLOANP	8800	2.0393	2.2995	1.0280	1.4133	2.2042
D_ASSET	8800	0.0026	0.0028	0	0.0020	0.0039
LGRTH	8800	0.1141	0.1810	0.0110	0.0783	0.1753
RISK	8800	0.0770	0.0487	0.0456	0.0640	0.0920
INSTOWN	8800	0.3006	0.2646	0.0739	0.2328	0.4766
DODD	8800	0.4964	0.5000	0	0	1
EXTMOR	8800	4.9930	0.9961	4.5	5	6
ASSET	8800	8.4882	2.2536	6.7806	7.9298	9.7748

summarizes the descriptive statistics of the variables used in the paper. All the dependent variables are percentages of total compensation. The proportions of means of CEO salaries (CEOSALARYCOMP), CEO incentives (CEOINCOMP) and CEO bonus compensations (CEOBONUSCOMP) in total CEO compensation are 71.88%, 15.47%, and 12.65%, respectively. These data imply that the majority of CEO compensation is salary. Colonnello et al. (2023) document that banks mitigated the impact of the bonus cap on executive earnings by increasing fixed compensation to maintain overall remuneration levels. By doing so, they ensured that executives did not experience significant reductions in total compensation despite regulatory constraints on variable pay. This adjustment allows banks to retain top talent, maintain competitive compensation structures, and align executive incentives with long-term financial stability while complying with regulatory requirements. Incentive compensation comprises options compensation (CEOOPTIONSCOMP), which accounts for 3.52%, and RSG compensation (CEORSGCOMP), which accounts for 11.95% of total CEO compensation. Panel A of Table 1 also indicates that about 83.78% of the banks in the sample were listed on US exchanges and 17% of the observations were of banks from developing nations. The mean value of the DODD variable indicates that 49.64% of the observations are post-Dodd–Frank Act of 2010, which was passed to reduce risk to bankers and increase transparency and regulation. On average, a bank is listed on 1.74 stock exchanges.

Table 2. T-test of mean compensation.

Variable	Developed Banks	Developing Banks	p-Value
CEOSALARYCOMP	0.7051	0.7859	0.00
CEOICOMP	0.1803	0.0299	0.00
CEOOPTIONSCOMP	0.0405	0.0091	0.00
CEORSGCOMP	0.1397	0.0208	0.00
CEOBONUSCOMP	0.1146	0.1842	0.00

presents the tests that compare the mean values of CEO compensation across two groups: banks in developed countries and banks in developing countries. The last column shows the p-values of the hypothesis if the two mean values are the same. The results obviously show that banks in developing countries pay significantly higher salaries and bonuses but lower options and restricted stock compensations (all p-values are equal to zero).

Table 3. Countries of banks and observations.

		Number of Unique Banks		Number of Observations
	Country	Num. of Banks	Percentage	Num. of Obs.	Percentage
1	Australia	9	0.74	101	1.15
2	Austria	3	0.25	23	0.26
3	Bangladesh	5	0.41	8	0.09
4	Belgium	2	0.17	21	0.24
5	Botswana	2	0.17	6	0.07
6	Canada	10	0.83	132	1.50
7	Chile	1	0.08	9	0.10
8	China	23	1.90	138	1.57
9	Cyprus	3	0.25	10	0.11
10	Czech Republic	1	0.08	9	0.10
11	Denmark	25	2.06	137	1.56
12	Finland	3	0.25	25	0.28
13	France	20	1.65	211	2.40
14	Germany	9	0.74	73	0.83
15	Greece	1	0.08	3	0.03
16	Iceland	1	0.08	1	0.01
17	India	43	3.55	353	4.01
18	Israel	10	0.83	126	1.43
19	Italy	25	2.06	160	1.82
20	Japan	5	0.41	29	0.33
21	Jordan	14	1.16	85	0.97
22	Kenya	8	0.66	18	0.20
23	Liechtenstein	1	0.08	1	0.01
24	Lithuania	1	0.08	1	0.01
25	Malaysia	10	0.83	79	0.90
26	Namibia	2	0.17	8	0.09
27	Netherlands	4	0.33	36	0.41
28	New Zealand	1	0.08	2	0.02
29	Norway	24	1.98	194	2.20
30	Pakistan	14	1.16	36	0.41
31	Palestinian Authority	1	0.08	3	0.03
32	Papua New Guinea	1	0.08	3	0.03
33	Poland	18	1.49	154	1.75
34	Portugal	5	0.41	17	0.19
35	Saudi Arabia	1	0.08	3	0.03
36	Singapore	3	0.25	25	0.28
37	Slovenia	1	0.08	4	0.05
38	South Africa	7	0.58	96	1.09
39	South Korea	4	0.33	11	0.13
40	Spain	11	0.91	79	0.90
41	Sri Lanka	1	0.08	1	0.01
42	Sweden	6	0.50	49	0.56
43	Taiwan	8	0.66	31	0.35
44	United Kingdom	15	1.24	129	1.47
45	United States	850	70.13	6160	70
	Total	1212	100	8800	100

presents the countries under observation. The sample comprised 1212 unique banks from 45 countries. Banks from the US represent the largest proportion, accounting for 70.13% of the total unique banks and 70% of the total observations. The second-largest representation is from India, a developing country, including 43 unique banks, constituting 3.55% of the total unique banks and 4.01% of the total observations.

Table 1, Table 2 and Table 3 summarize the statistics and countries of observation. Table 1 summarizes the statistics of the variables used in this paper and the number of observations by country. Please refer to Appendix A for the definitions of the variables.

Table 2 shows the tests for differences in CEO compensation averages between banks in developed countries and banks in developing countries. The last column shows the p-value of the hypothesis that the two means are the same

Table 4. A correlation matrix. This table summarizes the correlation matrix of the explanatory variables for the full sample. The * indicates statistical significance at the five percent level. Refer to Appendix A for the definitions of the variables.

		(a)	(b)	(c)	(d)	(e)	(f)	(g)	(h)	(i)	(j)	(k)	(l)	(m)	(n)	(o)
CEOWN	(a)	1
UB_DUMMY	(b)	0.1411 *	1
US_DUMMY	(c)	0.0659 *	0.6290 *	1
DEV	(d)	−0.0971 *	−0.6913 *	−0.5461 *	1
NUM	(e)	−0.0364 *	−0.4183 *	0.0537 *	0.1642 *	1
ROE	(f)	0.0283 *	−0.1187 *	−0.0361 *	0.1446 *	0.1207 *	1
NPL	(g)	−0.0393 *	−0.2129 *	−0.2029 *	0.1906 *	0.016	−0.3568 *	1
BLOANP	(h)	−0.009	−0.2097 *	−0.2035 *	0.1632 *	0.0393 *	−0.1248 *	0.4198 *	1
D_ASSET	(i)	0.0397 *	0.0336 *	0.0492 *	0.0434 *	−0.0006	0.3059 *	−0.1348 *	0.0924 *	1
LGRTH	(j)	0.0205	0.0519 *	−0.0006	0.0194	−0.0530 *	0.1794 *	−0.2177 *	−0.1112 *	−0.0208	1
RISK	(k)	−0.0162	−0.0396 *	−0.0601 *	0.0723 *	0.003	−0.4292 *	0.3375 *	0.1310 *	−0.2714 *	−0.1233 *	1
INSTOWN	(l)	0.0333 *	0.1638 *	0.2120 *	−0.1365 *	0.2446 *	0.0563 *	−0.1011 *	−0.0350 *	0.0938 *	0.0325 *	0.0014	1
DODD	(m)	0.0875 *	−0.1511 *	−0.0911 *	0.1243 *	0.1577 *	0.0283 *	0.0642 *	0.0383 *	−0.0354 *	−0.1265 *	−0.1635 *	0.2173 *	1
EXTMOR	(n)	0.1453 *	0.5568 *	0.3222 *	−0.3565 *	−0.2316 *	−0.0828 *	−0.0196	−0.0829 *	0.005	−0.0587 *	−0.0817 *	0.2386 *	0.4022 *	1
ASSET	(o)	−0.1344 *	−0.5884 *	−0.1293 *	0.3097 *	0.7610 *	0.1126 *	0.0650 *	0.0410 *	−0.0123	−0.0802 *	0.0389 *	0.3763 *	0.2166 *	−0.2985 *	1

summarizes the correlation coefficients among the variables used in this paper. The correlation between US banks (UB_DUMMY) and banks listed on US exchanges (US_DUMMY) is 0.629, indicating that banks listed on US exchanges are mainly located in the US. Because of this high correlation coefficient, we did not include both variables in the same regression model. The correlation between the banks’ assets and number of exchanges listed is 0.761, indicating that the banks listed in US exchanges are large banks.

4.2. Regression Results

Table 5. The baseline regression results. This table reports the regression results of the bank CEO compensations. The continuous variables are Winsorized at the one percent level. All the models include the robust option to obtain robust standard errors. Numbers in parentheses represent the t-statistics. The ***, **, and * denote the 1%, 5%, and 10% levels of significance, respectively. Please refer to Appendix A for the definitions of the variables.

Variables	CEOSALIN	CEOSALARYCOMP	CEOICOMP	CEOOPTIONSCOMP	CEORSGCOMP	CEOBONUSCOMP
CEOWN	−0.1732	0.4440	−0.6089 **	−0.0761	−0.5051 **	0.1828
	(−0.71)	(1.48)	(−2.48)	(−0.74)	(−2.39)	(0.78)
US_DUMMY	0.0139	−0.0414 ***	0.0517 ***	0.0184 ***	0.0316 ***	−0.0119
	(1.15)	(−2.80)	(4.54)	(4.22)	(3.31)	(−1.01)
DE_DUMMY	−0.0426 ***	0.0935 ***	−0.1320 ***	−0.0240 ***	−0.1039 ***	0.0403 ***
	(−3.32)	(6.01)	(−11.06)	(−5.32)	(−10.41)	(3.24)
NUM	−0.0065 **	0.0026	−0.0080 ***	−0.0027 ***	−0.0054 **	0.0058 **
	(−2.33)	(0.76)	(−3.10)	(−2.73)	(−2.50)	(2.17)
ROE	−0.1814 ***	−0.2288 ***	0.0409 **	−0.0107	0.0476 ***	0.1787 ***
	(−10.09)	(−10.26)	(2.14)	(−1.13)	(2.79)	(10.45)
NPL	0.0052 ***	0.0062 ***	−0.0013	0.0002	−0.0021 **	−0.0047 ***
	(4.62)	(4.51)	(−1.12)	(0.27)	(−2.05)	(−4.41)
BLOANP	−0.0049 ***	−0.0027 **	−0.0022 *	0.0006	−0.0026 ***	0.0049 ***
	(−4.43)	(−1.96)	(−1.90)	(1.23)	(−2.63)	(4.60)
D_ASSET	−1.9646 **	−2.0988 **	0.1396	−0.6901 *	0.7816	1.5360*
	(−2.39)	(−2.06)	(0.16)	(−1.75)	(1.04)	(1.96)
LGRTH	−0.0620 ***	−0.1089 ***	0.0473 ***	0.0268 ***	0.0206 **	0.0593 ***
	(−6.55)	(−9.27)	(4.69)	(5.38)	(2.29)	(6.59)
RISK	0.1113 **	0.0869	0.0116	−0.0315	0.0294	−0.1225 ***
	(2.55)	(1.60)	(0.25)	(−1.37)	(0.71)	(−2.95)
INSTOWN	0.0683 ***	−0.1785 ***	0.2611 ***	0.0385 ***	0.2394 ***	−0.0666 ***
	(5.31)	(−11.30)	(20.23)	(7.02)	(21.48)	(−5.40)
DODD	0.1840 ***	0.0469 ***	0.1303 ***	0.0127 ***	0.1134 ***	−0.1798 ***
	(19.82)	(4.08)	(13.32)	(2.67)	(13.06)	(−20.33)
EXTMOR	−0.0279 ***	−0.0323 ***	0.0072 **	−0.0024	0.0101 ***	0.0273 ***
	(−9.24)	(−8.63)	(2.27)	(−1.63)	(3.63)	(9.46)
ASSET	−0.0258 ***	−0.0517 ***	0.0246 ***	0.0047 ***	0.0189 ***	0.0252 ***
	(−10.04)	(−16.46)	(9.95)	(4.74)	(9.02)	(10.14)
FYr effect	Yes	Yes	Yes	Yes	Yes	Yes
R2	0.1966	0.1037	0.1616	0.0946	0.1952	0.2055
Obs.	8800	8800	8800	8800	8800	8800

reports the baseline regression results. The results in Table 5 indicate that for the entire sample, banks listed on US exchanges pay higher CEO incentive compensations (both stock options and restricted stocks) but lower CEO salary compensations. The coefficients of the US_DUMMY variable are positive and robust in the models of incentive compensation (0.0517 with t-stat = 4.54), stock options (0.0184 with t-stat = 4.22), and restricted stocks (0.0316 with t-stat = 3.31), but negative and significant in the model of salary compensation (−0.0414 with t-stat = −2.80).

Banks in developing countries tend to pay higher CEO salaries and CEO bonuses but lower incentive compensations, including stock options and restricted stocks. Table 5 shows that as the number of exchanges on which a bank is listed increases, the CEO bonus compensation increases (coefficient on NUM is positive for CEO bonus compensation), but CEO incentive compensation decreases (coefficient on NUM is negative for all types of CEO incentives compensation).

The coefficients on LGRTH imply that as bank loan growth increases, CEO salary compensation decreases, while CEO incentive and bonus compensation increases. These results confirm our Hypothesis 3.

Further, higher institutional ownership is associated with lower salaries and bonuses but higher incentive compensations. Larger banks seem to pay higher incentives and bonuses but lower salary compensations. Banks with higher ROE pay lower salaries and higher equity compensations (restricted stock grants) and bonuses. The coefficients on the NPL, BLOANP, and risk variable are mixed. Non-performing loans (NPL) and bad loans percentage (BLOANP) indicate a bank’s poor asset quality and increased credit risk levels. Risk measures the standard deviation of monthly returns and has no significance on the CEO salary compensation and incentive compensation, but does negatively affect CEO bonus (coefficient = −0.1225 with t = −2.95). Our results confirm the findings of Hogan and Burns (2019), who found that following the enactment of Dodd–Frank Act, banks paid higher CEO salaries and incentive compensations. Our results also indicate that post-Dodd–Frank, banks pay lower bonuses. These results imply that higher external monitoring (EXTMOR) is associated with higher incentives and bonus compensations but lower salaries. Higher CEO ownership (CEOOWN) is associated with higher fixed salaries (coefficient positive but insignificant) but lower equity compensations (coefficient = −0.5051 with t = −2.39).

Table 6. The impact of being listed on US exchanges. This table reports the impact of the association between DEV and US. The dependent variables are the CEO compensations. The continuous variables are Winsorized at the one percent level. All the models include the robust option to obtain robust standard errors. The numbers in parentheses represent the t-statistics. The ***, **, and * denote the 1%, 5%, and 10% levels of significance, respectively. Please refer to Appendix A for the definitions of the variables.

Variables	CEOSALIN	CEOSALARYCOMP	CEOICOMP	CEOOPTIONSCOMP	CEORSGCOMP	CEOBONUSCOMP
CEOWN	−0.1885	0.4591	−0.6408 ***	−0.0825	−0.5300 **	0.1970
	(−0.77)	(1.53)	(−2.64)	(−0.81)	(−2.53)	(0.84)
US_DUMMY	0.0507 ***	−0.0733 ***	0.1203 ***	0.0354 ***	0.0832 ***	−0.0464 ***
	(3.35)	(−3.98)	(8.53)	(6.34)	(6.97)	(−3.17)
DE_DUMMY	0.0047	0.0527 **	−0.0443 ***	−0.0028	−0.0385 ***	−0.0042
	(0.27)	(2.51)	(−2.77)	(−0.45)	(−2.86)	(−0.25)
USDEV	−0.1023 ***	0.0882 ***	−0.1850 ***	−0.0425 ***	−0.1366 ***	0.0965 ***
	(−4.07)	(2.88)	(−8.04)	(−4.85)	(−7.07)	(3.95)
NUM	−0.0062 **	0.0024	−0.0078 ***	−0.0027 ***	−0.0053 **	0.0056**
	(−2.26)	(0.71)	(−3.06)	(−2.83)	(−2.49)	(2.09)
ROE	−0.1801 ***	−0.2299 ***	0.0431 **	−0.0098	0.0494 ***	0.1776 ***
	(−10.02)	(−10.32)	(2.25)	(−1.03)	(2.90)	(10.39)
NPL	0.0051 ***	0.0063 ***	−0.0015	0.0001	−0.0023 **	−0.0046 ***
	(4.57)	(4.55)	(−1.29)	(0.19)	(−2.19)	(−4.36)
BLOANP	−0.0045 ***	−0.0031 **	−0.0011	0.0010 *	−0.0017 *	0.0045 ***
	(−3.97)	(−2.27)	(−0.92)	(1.88)	(−1.72)	(4.16)
D_ASSET	−1.9838 **	−2.0701 **	0.0761	−0.7172 *	0.7386	1.5536 **
	(−2.41)	(−2.04)	(0.09)	(−1.83)	(0.99)	(1.98)
LGRTH	−0.0624 ***	−0.1087 ***	0.0471 ***	0.0268 ***	0.0203 **	0.0597 ***
	(−6.60)	(−9.26)	(4.68)	(5.38)	(2.26)	(6.64)
RISK	0.1024 **	0.0948 *	−0.0110	−0.0397 *	0.0108	−0.1145 ***
	(2.34)	(1.75)	(−0.24)	(−1.73)	(0.26)	(−2.75)
INSTOWN	0.0651 ***	−0.1757 ***	0.2560 ***	0.0365 ***	0.2345 ***	−0.0636 ***
	(5.06)	(−11.11)	(20.01)	(6.67)	(21.18)	(−5.16)
DODD	0.1834 ***	0.0473 ***	0.1290 ***	0.0129 ***	0.1127 ***	−0.1792 ***
	(19.77)	(4.12)	(13.23)	(2.71)	(13.02)	(−20.28)
EXTMOR	−0.0288 ***	−0.0315 ***	0.0055 *	−0.0031 **	0.0087 ***	0.0281 ***
	(−9.52)	(−8.39)	(1.74)	(−2.10)	(3.12)	(9.72)
ASSET	−0.0238 ***	−0.0535 ***	0.0284 ***	0.0056 ***	0.0218 ***	0.0233 ***
	(−9.11)	(−16.75)	(11.44)	(5.60)	(10.31)	(9.22)
FYr effect	Yes	Yes	Yes	Yes	Yes	Yes
R2	0.1971	0.1037	0.1623	0.0949	0.1960	0.2059
Obs.	8800	8800	8800	8800	8800	8800

includes USDEV, a cross-section variable between DE_DUMMY, a dummy variable for banks from developing countries, and US_DUMMY, a dummy variable for banks listed on US stock exchanges. The results for the model of CEO salary compensation show the coefficient of DE_DUMMY as positive and robust and that of USDEV to also be positive and significant. This outcome indicates that the CEOs of banks from developing nations and those listed on US exchanges receive higher proportions of salary compensation than their non-listed counterparts from developed nations. Similarly, the coefficient of USDEV is positive for the model of CEO bonus compensation, indicating that bank CEOs from developing countries receive higher bonus compensation if the bank is listed on US exchanges.

The results for the models of CEO incentive compensation (options and restricted stocks) present the opposite picture. The coefficients of DE_DUMMY and coefficients of USDEV are negative and robust, implying that if a bank in a developing country is listed on US exchanges, the bank’s CEO receives a lower proportion of incentive compensation. That same bank CEO would receive a higher portion of salary compensation as compared to a bank CEO from a developed country, as indicated by the coefficient of the variable US_DUMMY being negative and significant for CEO salary while the interactive term (USDEV) coefficient is positive and robust. The coefficients of US_DUMMY are positive while the USDEV coefficients are negative for incentive compensation, indicating that among banks listed on US exchanges, compared with the CEOs of banks from developed countries, the CEOs of banks from developing nations receive a lower proportion of incentive compensation. In summary, banks in developing nations listed on US exchanges pay their CEOs higher salaries and bonus compensations but lower incentive compensations than their counterparts in developed countries. The results in Table 5 and Table 6 validate our Hypothesis 1.

The signs and significance of the variable LGRTH in Table 6 are similar compared to Table 5. These results validate our Hypothesis 3.

In Table 6, the coefficient estimates for NUM are negative and statistically significant for models of CEO incentives compensation and total salary and incentives compensation. The same coefficient estimates for NUM are positive and significant for the model of bonus compensation.

Table 7. The impact of the number of exchanges a bank is listed on. This table shows the test results of the effects of being listed on US exchanges. The dependent variables are the CEO compensations. The continuous variables are Winsorized at the one percent level. All the models include the robust option to obtain robust standard errors. The numbers in parentheses represent the t-statistics. The ***, **, and * denote the 1%, 5%, and 10% levels of significance, respectively. Please refer to Appendix A for the definitions of the variables.

Variables	CEOSALIN	CEOSALARYCOMP	CEOICOMP	CEOOPTIONSCOMP	CEORSGCOMP	CEOBONUSCOMP
CEOWN	−0.1968	0.4545	−0.6465 ***	−0.0846	−0.5330 **	0.2056
	(−0.81)	(1.51)	(−2.66)	(−0.83)	(−2.54)	(0.88)
US_DUMMY	0.0334 ***	−0.0494 ***	0.0788 ***	0.0252 ***	0.0512 ***	−0.0310 **
	(2.62)	(−3.17)	(6.62)	(5.46)	(5.10)	(−2.50)
DE_DUMMY	0.0192	0.0684 ***	−0.0478 ***	−0.0040	−0.0441 ***	−0.0206
	(1.06)	(3.09)	(−2.85)	(−0.62)	(−3.12)	(−1.17)
NUM	−0.0020	0.0008	−0.0020	−0.0013	−0.0012	0.0014
	(−0.68)	(0.21)	(−0.76)	(−1.32)	(−0.54)	(0.51)
NUMDEV	−0.0252 ***	0.0103	−0.0339 ***	−0.0078 ***	−0.0239 ***	0.0249 ***
	(−4.79)	(1.59)	(−7.03)	(−4.33)	(−5.94)	(4.86)
ROE	−0.1812 ***	−0.2289 ***	0.0410 **	−0.0104	0.0478 ***	0.1786 ***
	(−10.09)	(−10.27)	(2.14)	(−1.10)	(2.80)	(10.46)
NPL	0.0051 ***	0.0062 ***	−0.0014	0.0001	−0.0022 **	−0.0046 ***
	(4.57)	(4.53)	(−1.22)	(0.23)	(−2.13)	(−4.36)
BLOANP	−0.0050 ***	−0.0027 **	−0.0022 *	0.0006	−0.0026 ***	0.0049 ***
	(−4.44)	(−1.96)	(−1.91)	(1.19)	(−2.64)	(4.62)
D_ASSET	−2.0358 **	−2.0644 **	0.0033	−0.7350 *	0.6836	1.6042 **
	(−2.48)	(−2.03)	(0.00)	(−1.87)	(0.91)	(2.05)
LGRTH	−0.0615 ***	−0.1092 ***	0.0487 ***	0.0273 ***	0.0216 **	0.0588 ***
	(−6.50)	(−9.30)	(4.84)	(5.48)	(2.41)	(6.54)
RISK	0.1103 **	0.0872	0.0098	−0.0314	0.0284	−0.1215 ***
	(2.52)	(1.61)	(0.21)	(−1.37)	(0.69)	(−2.93)
INSTOWN	0.0655 ***	−0.1773 ***	0.2580 ***	0.0375 ***	0.2366 ***	−0.0639 ***
	(5.11)	(−11.21)	(20.12)	(6.86)	(21.33)	(−5.20)
DODD	0.1854 ***	0.0462 ***	0.1323 ***	0.0136 ***	0.1151 ***	−0.1811 ***
	(19.98)	(4.02)	(13.55)	(2.86)	(13.27)	(−20.49)
EXTMOR	−0.0292 ***	−0.0317 ***	0.0051	−0.0033**	0.0085 ***	0.0285 ***
	(−9.64)	(−8.44)	(1.61)	(−2.19)	(3.02)	(9.85)
ASSET	−0.0259 ***	−0.0517 ***	0.0245 ***	0.0046 ***	0.0188 ***	0.0252 ***
	(−10.11)	(−16.47)	(9.99)	(4.65)	(9.02)	(10.20)
FYr effect	Yes	Yes	Yes	Yes	Yes	Yes
R2	0.1969	0.1037	0.1621	0.0948	0.1958	0.2057
Obs.	8800	8800	8800	8800	8800	8800

introduces NUMDEV, an interaction term between DE_DUMMY (a dummy variable indicating that a bank is from a developing nation) and NUM (the number of exchanges on which a bank is listed). The inclusion of this interaction term renders the coefficient estimates for NUM insignificant, whereas the estimates for NUMDEV remain statistically significant across the three models of total salary and incentives compensation, incentives compensation, and bonus compensation.

These results suggest that the impact of the number of exchanges on which a bank is listed is more pronounced for banks in developing countries compared to those in developed nations. This outcome can be attributed to the heavily regulated nature of the banking sector in developed countries. For banks from developed countries, cross-listing in other markets often introduces regulatory requirements that are comparable to or less demanding than those in their home countries. Conversely, banks from developing countries face increased regulatory challenges when cross-listing in developed markets with stricter governance standards. For the latter group, the managerial effort required to comply with these increased regulations is higher and leads to higher fixed compensation and lower incentive-based compensation. In summary, the differences in management efforts between developing and developed nations’ banks results in different effects on CEO compensation. These findings corroborate Hypothesis 2.

5. Conclusions

This study investigates the relationship between CEO compensation and key factors such as cross-listing, loan growth, and CEO ownership in banks across developed and developing countries. Using a dataset of 8800 observations from 45 countries, we found that CEO compensation structures vary significantly based on the number of exchanges on which a bank is listed. More importantly, the impact of multiple exchange listings is more pronounced in banks located in developing nations than their developed counterparts. Cross-listed banks in developing countries show a preference for higher fixed salaries and bonuses due to increased regulatory scrutiny, while their developed counterparts exhibit less variation due to pre-existing stringent governance standards.

Loan growth is positively associated with higher incentive and bonus compensation but negatively correlated with fixed salaries. This reflects the emphasis on aligning CEO pay with growth-driven performance targets while managing associated risks. Higher CEO ownership is linked to reduced incentive compensation, underscoring the reduced need for external monitoring when ownership concentration is high.

These findings highlight that CEO compensation is profoundly influenced by the industry factors of regulatory environments and governance structures and by firm-specific factors like cross-listing and ownership. The results provide important insights into the relationship between regulatory environments and CEO compensation structures, particularly in the context of cross-listed banks. The significant influence of cross-listing on CEO compensation in developing countries highlights the disparities in regulatory maturity and enforcement between developed and developing markets. In developed countries, advanced regulatory frameworks may minimize the marginal impact of additional cross-listing requirements on compensation structures. In contrast, for banks from developing nations, cross-listing in more stringent markets imposes additional compliance demands and reshapes compensation methods for these elevated governance standards. These findings extend the understanding of how institutional quality and regulatory divergence affect compensation decisions. They also emphasize the nuanced ways in which global integration impacts executive pay, particularly in contexts where firms navigate substantial regulatory and operational complexities.

This study provides practical insights for policymakers and financial institutions in tailoring compensation strategies that balance regulatory compliance and performance incentives. Future research could go further by analyzing the impact of regional variations in institutional quality and exploring post-pandemic shifts in compensation practices. Additionally, future studies could conduct a comparative analysis of CEO compensation across developed, developing, and emerging economies to provide a more nuanced understanding of compensation dynamics in different economic contexts.