The Impact of US and UN Economic Sanctions on Income Inequality of the Target State: Evidence from a Bayesian Approach

Abstract

This study examines the impact of US and UN economic sanctions on income inequality in African- and Asian-targeted states from 1980–2019, using the Bayesian Generalized Method of Moment’s (BGMM) technique. One essential weapon of international politics for promoting peaceful international cohabitation is the imposition of political and economic sanctions, despite ongoing debates on the effectiveness of punishments in generating disutility. The BGMM was chosen due to its ability to address the dynamics of several entities in the data. The study reveals a positive relationship between economic sanctions and income inequality, except for pre-tax income held by the top 10%, indicating a significant contribution to income inequality. The study further finds that African countries seem to suffer the most from the execution of these sanctions compared to Asian countries. From a policy perspective, the study suggests implementing targeted assistance programs such as financial support and job training programs for vulnerable groups, such as low-income workers and small businesses, affected by sanctions. Additionally, policymakers should prioritize investments in less-impacted sectors, creating alternative job opportunities, and reducing income disparities. This approach can mitigate the negative effects of economic sanctions on income inequality by addressing specific needs and diversifying the economy.

1. Introduction

Worldwide economic sanctions remain a foreign policy tool of statecraft used by many countries to demand a change in the actions of the targeted region [1]. In the literature on economic sanctions, economic sanctions are viewed as a peaceful, more human alternative to military intervention. The imposition of the sanction is frequently encountered with harsh criticism, which is based on the unfriendly reality that, even though these measures are directed against governments, more often than not, it is the target state’s public that bears the costs. These results can be particularly unfair when the regime against which sanctions are directed lacks democratic legitimacy.

Scholars defined economic sanction in different ways, such as Galtung [2], who defined economic sanction as the action initiated by international senders against the targeted state. The author further stated that these actions may be used to punish the receivers by depriving them of something valuable or to force the receivers to comply with certain norms that the senders deem important. While other studies review economic sanctions as an indication of a serious threat to the targeted region from a political stability perspective, this may increase the chances of uncertainty in the political system. Therefore, it has been argued that political uncertainty might have a negative effect on the economy of the targeted region due to a decline in inflows of import-export, foreign direct investment, and financial aid. Which in turn might increase the level of inequality in the targeted states. Hence, the empirical literature confirms that sanctions episodes indeed lead to political turmoil [3, 4]. Then, this is expected to decline economic growth, influenced by a decline in savings and investment if there are political instabilities [5]; therefore, it might lead to a decline in income equality.

Kaempfer and Lowenberg [6] argued that economic sanctions contain sanctions on investment, such as constraints on the inflows of capital to the receiver (the target country) and even disinvestment, as well as trade sanctions constraints on exporting to and importing from the receiver (the target country). Hence, this has shown that sanctions episodes lead to political turmoil [3, 4]. Moreover, Pepe [7] argued that economic sanctions are used as instruments that seek to lower the welfare of the receiver in a way that lowers its international trade, with the aim of persuading or coercing the receiver country to change their political turmoil.

There is a massive literature on the effect of economic sanctions on the targeted states’ humanitarian situations. Economic sanctions have been shown to have a high negative impact on public servants by reducing access to clean water and food [1] and access to health care services [8], both of which have a negative impact on infant mortality and life expectancy [9]. However, these studies focused on a single country (taking a qualitative method), where they mainly focused on the impact of sanctions on various measures of human rights [10], the level of democracy in the targeted states [4], political stability within the target state [3], and their success in terms of meeting the desired objectives [11]. Their results are dispiriting. Peksen [10] finds that economic sanctions deteriorate the targeted government’s respect for human rights; Peksen and Drury [4] report that economic sanctions have a harmful impact on the level of democracy. Moreover, economic sanctions fail to achieve their aims in 65–95% of the cases in which they are imposed [11]. Apart from these studies on the litaratue, there are limited studies that have explored the impact of economic sanctions on income inequality in Africa. After scrutinizing the literature, we found three studies: Porter [12], Khan [13], and Afesorgbor and Mahadevan [14] that have used African data. Porter [12] used a linear programming model to estimate the impact of trade and investment sanctions on white and non-white incomes in South Africa. Khan [13] used multipliers from the Social Accounting Matrix, and Afesorgbor and Mahadevan [14] conducted cross-country analyses of 68 target states from 1960 to 2008. Therefore, the current study seeks to utilize the panel data of 22 African countries and compare the results with those of 19 Asian countries covering the period 1980–2019. See Appendix A for the list of countries.

The history of US and UN economic sanctions implemented in African countries and Asia and their impact on income inequality is complex and multifaceted. Both regions have experienced various interventions aimed at influencing political situations, protecting human rights, and addressing economic concerns. These sanctions have had both positive and negative consequences for income inequality, depending on the specific context and the effectiveness of the measures taken.

In Africa, economic sanctions have been frequently employed as a tool to condemn oppressive regimes and human rights abuses. For instance, the apartheid regime in South Africa faced substantial international sanctions during the 1980s, which aimed to isolate the government and pressure it to dismantle the discriminatory system. These sanctions exerted significant economic pressure on the country, resulting in reduced foreign investment and hampered economic growth. However, this negatively impacted the income of ordinary citizens and led to increased poverty and inequality.

Similarly, in Sudan, the US and the UN imposed economic sanctions due to human rights violations in regions like Darfur. While these sanctions aimed to pressure the government to address the crisis, they also had broader economic consequences. The restrictions on trade and financial transactions undermined the country’s economy, leading to a decline in income levels and exacerbating inequality among the already marginalized populations.

Conversely, economic sanctions have occasionally shown positive effects on income inequality. In Zimbabwe, international sanctions imposed on the government of Robert Mugabe due to electoral fraud and human rights abuses caused significant economic strain. However, these measures contributed to the removal of an authoritarian regime and opened up the potential for more inclusive economic policies that could benefit marginalized groups in the long run.

In Asia, economic sanctions have been implemented in countries like North Korea and Myanmar to address governmental policies, nuclear weapons development, and human rights violations. However, these sanctions have often had a limited impact on income inequality due to the highly centralized nature of economies. In North Korea, for example, the ruling elite and the military continue to retain significant control over the economy, ensuring that the effects of sanctions are largely felt by the most vulnerable populations, further deepening income inequality. Solt [15] pointed out that the gap between the wealthy and poor is becoming wider and wider. This has a straightforward effect on sustainable economic growth as it goes against the principle of inclusive growth in relation to lower-income groups, which are more likely to lead to political and civil unrest.

Economic sanctions aim at triggering political reforms or even overthrowing the target’s political regime. Moreover, economic agents may view sanctions as a sort of early-warning signal that political or societal conflicts in the target state have the potential to escalate. The literature itself shows that although there is an intensive literature that focuses on the impact of economic sanctions on macro-economic objectives and international trade, there is little focus on the impact of economic sanctions on income inequality. Therefore, this gives strong evidence that more empirical literature is required to address the impact of economic sanctions on income inequality in the targeted states. This current paper seeks to contribute to the existing literature by empirically examining the three main hypotheses:

1. Hypothesis 1: The imposition of economic sanctions is detrimental to the income inequality of the African target states.

The sanctions literature has also examined the mechanics of adjusting the effect of longer-term restrictions when political leaders do not cooperate early on. According to the Dizaji and van Bergeijk [16] theoretical study, sanctions may be more destructive in the early stages rather than later, as the targets find ways to alter their economies and so minimize the impact of the sanctions over time. However, their claim contradicts Kaempfer and Lowenberg’s [6] contention that sanction harm grows over time.

1. Hypothesis 2: The detrimental effect of economic sanctions on income inequality is more severe in African countries than in Asian countries.

Senders of sanctions employ different instruments of sanctions. Hufbauer et al. [17] explain that different types of sanctions would have different effects in several ways. For instance, they state that financial sanctions are more likely to hit the personal pockets of political elites. However, this could also hit hard on the poor, especially if financial sanctions disrupt financial flows such as remittances (see [18]). Trade sanctions may produce limited damage compared to financial sanctions since the disruption of financial Flows may also disrupt international trade even without an explicit trade sanction [17]. In addition, trade controls may be applied to selective products because the Geneva Convention prohibits the ban of essential goods such as food and medicine. Unlike financial sanctions, trade sanctions may be difficult to enforce and thus enable the target states to circumvent the ban [17]. Thus, the normative assumption here is that financial sanctions would have a more adverse impact on economic outcomes. Such as poverty or income inequality relative to trade sanctions [19]. Financial sanctions include the interruption of commercial finance, the transfer of remittances, and access to SWIFT, AID, and other official financial flows while exporting. Import sanctions refer to the interruption of exports (imports) from the sender to the target. On the basis of these different sanction types, we test the hypothesis of whether there exist any differential effects on income.

1. Hypothesis 3: The impact of trade sanctions on income inequality is more severe than financial sanctions.

To achieve the objectives of the study, we made use of the Bayesian Generalized Method of Moment (BGMM) model over the period 1980–2019 due to data unavailability. Moreover, according to literature ([20, 21]; among others), Bayesian GMM has a comparative advantage over standard GMM as the BGMM it employed prior to effectively reducing the dimensionality of the coefficient. Prior restrictions are superior to the dogmatic restrictions used by the Stardand GMM. The findings could be of substantial importance for policymakers to get further insight into the economic benefits and consequences of global sanctions.

The remaining portion of the paper is organized as follows: Section 2 briefly surveys the related literature. Section 3 presents an overview of the model. Section 4 discusses the results of the BGMM models. Section 5 provides concluding remarks and discusses policy implications.

2. Literature review

2.1 Theoretical debate on sanctions and income inequality

The impact of economic sanctions on income inequality was first discussed by Cooper [22] using the Stolper-Samuelson Theorem in the trade framework, followed by Kaempfer and Lowenberg [23] focusing on the public choice approach. The argument was further taken by Wang [24] using the Harris-Todaro model, the last of which is the microfoundations approach, which was studied by Kirshner [25]. These studies are the ones that underpin the relationship between economic sanction and income inequality.

Going as far as the study by Metzler [26] and Bhagwati [27], explains how international trade may affect the division of income within the country in general by relating to the rent earned by the various factors of production. While the impact of economic sanctions and inequality was first studied by Cooper [22] within the theoretical trade model, the study by Cooper [22] argues that economic theory predicts that the position of capital is likely to be strengthened, not weakened, by the imposition of sanctions against target states.

Based on the Stolper-Samuelson theorem, it states that when sanctions are imposed on imports, their normal favors factor is used intensively in the import competing sector as the domestic demand for domestic production of importable goods upsurges. Using the Edgeworth box representing capital and labor inputs versus imports and exports, together with the related production possibility curve of imports and exports, Cooper [22] illustrates this particular case, which leads to an increase in the return to capital, thereby favoring capitalists; and if politicians are manipulated by capitalists, the effect of sanctions will be to slow down the pace of political change while making the income distribution more unequal.

More specifically, while an import embargo restricting imports to a target state allows domestic producers of import-competing goods in target countries to gain compared to producers of exports, the consumers of imports in the target state are, however, adversely affected. But Black and Cooper [28] highlight the fact that the losses suffered by producers of exports may be partly offset by benefits derived from their role as consumers of exports. At the same time, labor may spend a larger part of its income on exports and capital owners may operate in both export-and import-competing industries in the target countries, thereby making the final effect on the income of the various groups of people unclear (ibid.). Wang [24], on the other hand, uses the Harris-Todaro model, comprising a two-sector model with the production functions and factor-price frontiers of the agricultural and manufacturing sectors, to show that export and import embargoes have asymmetrical effects on national income (through the impact of demand on wages and employment) and income distribution. All these analyses point to the fact that the impact of sanctions could differ depending on a targeted state’s level of trade openness and also on the intensity of labor or capital in the economy. For example, Black and Cooper [28] argue that if domestic exporters use more labour-intensive production processes than capital-intensive ones, labor is expected to suffer more from sanctions than capital owners.

The public choice approach of Kaempfer and Lowenberg [23] examines economic sanctions from a different angle, whereby sanctions may be imposed to serve the interests of certain pressure groups within the sender state. These interest groups have different motives as they may enjoy some pecuniary benefits from the imposition of sanctions, which are essentially specific instruments of protection that regulate goods or factor flows. For example, an embargo on exports of a target country would benefit producers of import-competing goods in the sanctioning country but harm producers of the sanctioning state that use imports from the target state as intermediate inputs. This means sanctions may affect domestic constituents in the target (and sender) states differently in terms of varying degrees of income loss or gain. This may skew the income distribution favorably or unfavorably towards one segment of the target population.

Lastly, the micro-foundations approach argues that sanctions work because they weaken the government directly as well as motivate the most influential groups (such as the military, the middle class, agricultural laborers, big business, etc.) to pressurize the government into protecting their own interests [25]. The governments of target states will need to respond domestically because this pressure may destabilize their rule by creating political costs. In so doing, Escribe-Folch [29] explains that if the rulers’ budget is not strictly constrained, they tend to increase spending towards the core of their political support groups. Thus, sanctions can have dramatic differential effects on various groups within society [25].

2.1.1 Single-rational actor and game theory approaches to sanctions

The interest group theory of the political process examines sanctioning behavior and its political effects from the perspective of domestic interest group politics within sender and target nations. National governments are seen as impartial arbiters of competing pressures, with no independent policy preferences or agendas. This approach aligns with neoclassical economics’ methodological individualism, where interest group behavior is based on individual group members’ utility maximization, ensuring that sanctions are influenced by domestic interest group politics.

International relations scholars and economists often use a single-rational actor methodology to study sanctions, focusing on entire nationstates rather than individual interest group members, voters, or politicians. This approach emphasizes that states are the main players on the international stage, making decisions about sanctions and compliance. While single-rational actor theories may not necessarily contradict interest group theories, they focus on different questions. Interest group models aim to show how national policy choices reflect constituency groups’ interests, while single-rational actor models aim to show how one country’s international policy decisions affect and are influenced by other governments’ decisions. Game theory is often used to analyze state behavior.

Drezner [30, 31] proposes a game-theoretic model of economic coercion, where both senders and targets of sanctions consider future conflict expectations and short-run opportunity costs. He identifies a “sanctions paradox” attributed to conflict expectations, where sender states that anticipate frequent conflicts with the target state are more likely to initiate sanctions, while target states that anticipate future conflicts are less likely to comply with the sender’s demands due to fear of future threats to their security. This leads to senders imposing sanctions against adversaries more than against allies but receiving more significant concessions from allies due to a low expected likelihood of future conflict. Drezner [30, 32] supports the conflict expectations model and dismisses domestic politics as a cause of sanctions initiation. Evidence supports his position, as sanctions events are strongly correlated with crises where states’ interests are directly threatened, and sanctions events are not randomly distributed across all international crises.

Eaton and Engers’ game-theoretic [33, 34] treatment of sanctions, based on a theory of bargaining under incomplete information, suggests that success is more likely when the cost of a threatened sanction to the sender country is low relative to the gain to the sender from changing the target’s behavior, while the cost of the sanction to the target is high relative to the cost to the target of complying with the sanctioner’s demands. This model has interesting implications, as in a world of perfect information, sanctions would never be implemented. If a threatened sanction is effective, the target would comply immediately, obviating the need for the sanction. If sanctions are observed, it means the sanctioner underestimates the target’s cost of compliance or the target underestimates the sanctioner’s resolve.

Sanctions can be effective even if they are not used, as they can be imposed when they are not likely to succeed. This is because sanctions can be effective even if they are not actually used. An observer of actually applied sanctions would likely conclude that sanctions do not work, even though most are successful at the threat stage. Empirical studies based only on observed sanctions would be biased against sanctions success. However, a resolute sanctioner might impose sanctions repeatedly to demonstrate its resolve, thereby initiating a pattern of compliance on the part of targets over time. This would lead an observer to conclude that sanctions are extremely effective. Empirical analyses based on observations of actually implemented sanctions would be biased in favor of sanctions success. In general, sanctions that are actually imposed constitute a very unrepresentative tip of an iceberg [34].

Eaton and Engers [34] found that the success of sanctions in ongoing interactions between sanctioner and target depends on the sender’s commitment to impose sanctions if the target balks. By committing to always use sanctions, the sender removes the incentive for the target to balk, hoping to increase the probability of sanctions being lifted in the future. The degree of compliance depends on the cost of the sanction to each party and the target’s patience. Eaton and Engers [33] developed a measure of “toughness” based on a party’s willingness to incur the cost of sanctions. The greater the target’s impatience and the lower the cost to the sender, the more likely the sanction is to be successful. However, a patient sender and high cost to the target could hinder compliance, as the threat of implementing such a sanction every period is not credible. In such cases, a sanction that imposes less harm on the target and is more credible in repeated implementation can sometimes be more effective.

2.2 Empirical review

2.2.1 Emprical liturature on economics sanction and income inequality

The literature on economic sanctions has confirmed that sanctions may adversely affect the targeted state through several channels, namely: contraction of international capital flows, the decline in imports and exports, the loss of bargaining power on international markets, the withdrawal of financial grants, foreign direct investment, and foreign aid [11]. The funny part is that this effect is intended to occur even though trade suspensions, capital flows, and international aid are not imposed. However, much attention has been paid to how economic sanctions affect economic growth, national currency, employment, poverty, and government consumption [16, 23, 35, 36, 37]. While other studies analyzed the effectiveness of economic sanctions by empirically investigating their failures or successes [17, 38, 39].

After an intensive empirical evaluation of this subject matter, it became obvious that the relationship between economic sanctions and income inequality has received sparse attention to date, and several economic concerns remain unsolved to this day. Given the outdated and scarcity of empirical research on the impact of economic sanctions on income inequality and the inconsistency of the available results in the existing literature, a fresh investigation using modern data and economic models is required. Furthermore, there has been a small batch of studies whose focus has been on the impact of sanctions on specific segments of the target state population.

After critically evaluating the empirical literature on this subject matter, we found that there is controversy in the literature as the existing studies created a strong paradox on the relationship between economic sanctions and income inequality. Other studies find positive relationships [13, 14, 36, 40, 41, 42, 43, 44, 45, 46], others find negative relationships [47, 48, 49, 50, 51, 52, 53]. While others believed that, there is no clear relationship between economic sanctions and income inequality [54].

Going far back as the study by Porter [41] and Khan [13] conducted studies on the impact of trade and investment sanctions on incomes, employment, and economic growth in South Africa. Porter’s study used a linear programming tool to estimate output and input relations in eight sectors. Khan’s study used the Social Accounting Matrix (SAM) to analyze inequality in South Africa, showing that inequality among whites widened more than among blacks. However, the SAM is static and needs to be continuously updated to be relevant. Hufbauer et al. [40] analyzed the diverted change when the US imposed unilateral sanctions on China and Cuba using OECD and United States trade data from 1985, 1990, and 1995. They found that Cuba trades with Italy, Germany, Spain, China, the Netherlands, and Belgium more than expected, despite the model’s predictions. The authors also criticized the model for incorrectly predicting that Germany, Australia, and Canada export more to China.

Economic sanctions have a negative impact on the distribution of income in target countries, according to a study by Marinov [49]. The study argues that the costs of sanctions are divided between political leaders and the population. Ordinal voters and political elites are affected differently by the distribution of costs, resulting in a heterogeneous effect on income distribution. Kaempfer and Lowenberg [6] argue that political or international connections can minimize the income-reducing effects of sanctions damage. Therefore, the distribution of income can be affected inconsistently from the perspective of the target countries. Saghafian’s [50] study uses the OLS model to examine the impact of economic sanctions, natural disasters, and wars on income inequality, finding that these factors will decrease inequality. Alvaredo and Gasparini [47] study suggests that income distribution in emerging states is more unequal, and economic sanctions could potentially link this to income inequality due to the Stolper-Samson theorem’s major re-distribution predispositions [22].

Recent literature on economic sanctions has shown that trade and financial sanctions have different impacts on income inequality, while economic sanctions have a harmful effect. Afesorgbor and Mahadevan [14] found that trade and financial sanctions have different effects on income inequality, while economic sanctions have a harmful effect. Saghafian [50] investigated the impact of declining trade openness on income inequality in 113 countries between 1982 and 2001 using fixed-effect and ordinary least squares models. The empirical findings confirmed that economic sanctions would lower income inequality in targeted states. O’Driscoll [44] found that sanctions have a more severe impact on income inequality and gross domestic product.

Neuenkirch and Neumeier [36] found that US sanctions negatively impact poverty in target countries from 1982 to 2011, with a 3.8 percentage point larger poverty gap in sanctioned countries compared to a control group. Jin [43] also found that economic sanctions have a discernible effect on income inequality, but this effect varies across sanctions instruments and economic conditions of sanctioned countries. Neuenkirch and Neumeier [36], findings contradict Jin, [43] and Savin et al.’s [51] findings, which found that economic and financial sanctions are detrimental to the low-income population and inappropriate for policies aimed at reducing income inequality. The study suggests that the effects of economic sanctions on income inequality vary across sanctions instruments and economic conditions of sanctioned countries.

Which then contradicts those studies who believe in the negative relationship, as the study by Pajak and Keuangan found that economic sanctions have a significant positive impact on income inequality. However, other studies, such as Mariev et al. [52] and Pahlavani et al. [53], found that economic and financial sanctions are detrimental to the low-income population and inappropriate for policies aimed at reducing income inequality. The study also confirmed the negative relationship between economic or financial sanctions and income inequality, contradicting previous studies that documented a positive relationship. These studies highlight the need for further research to understand the complex relationship between economic and financial sanctions and income inequality [14, 36, 42, 44, 48].

Further support for those studies that documented a positive relationship was further strengthened by the study documents by Saeed [46] and Eslamloueyan and Kahromi [45]. Saeed [46] conducted the same subject matter in developed and developing countries and found a positive relationship between trade and financial sanctions on income inequality. Eslamloueyan and Kahromi [45] investigate the interaction between institutions and sanctions. The estimation results show that good institution decrease but sanctions and income inequality increase the poverty gap.

2.2.2 Emprical on the political institutions and sanctions

The literature on sanctions focuses on the role of domestic institutions and politics in determining the likelihood and political outcome of sanctions. The nature of the political regime in both the target and sanctioner, characterized as democratic or non-democratic, is a key aspect of domestic institutions. This interest stems from the international relations literature on democratic peace, which suggests that democratic dyads are less likely to enter into military conflict than non-democratic or mixed dyads. Democratic political competition reveals a country’s level of resolve, preventing escalation of disputes into violent conflict. Accountability of democratic politicians to large constituencies also gives them an incentive to conduct successful foreign policies and protect citizens from war costs. Autocrats, on the other hand, are less concerned with public welfare and are more likely to lead their nations into military conflict ([55], p. 647).

Bueno de Mesquita and Siverson [56] and McGillivray and Smith [57] argue that war is harmful to the survival of all types of leaders, especially democrats. They argue that domestically accountable politicians lose public support if they fail to cooperate with foreign nations. This makes accountable leaders more trustworthy in foreign eyes and fosters greater international cooperation. Conversely, when replacing leaders is difficult, cooperation becomes less robust, leading to inter-state hostilities. Both studies highlight the importance of international cooperation in maintaining stability and trust among leaders. Bueno de Mesquita et al. [58] argue that authoritarian leaders gain support from narrow constituencies, leading to different interests being represented and policy revisions. Democratic leaders, on the other hand, must appeal to broader constituencies, which do not change significantly with leadership turnover. As a result, policies, including foreign economic policies, are unlikely to change significantly with change in democratic leadership ([59], pp. 346–347). McGillivray and Smith [60] confirm that leadership turnover’s impact on trade between democracies is less pronounced than in autocracies.

The democratic peace theory, which suggests that democracies are more likely to impose economic sanctions against other democracies, has been influential in the sanctions literature. Studies by Lektzian and Souva [55] and Cox and Drury [61] have shown that democracies impose sanctions more often than other regime types, possibly due to the greater variety of interest groups in ruling coalitions. Trade sanctions are also useful for democratic governments to justify domestic industries while maintaining commitment to a liberal trading regime. Cox and Drury [61] suggest that democracies might choose sanctions over military action due to the less public attention and opposition they attract to non-violent measures. These findings suggest that the factors that encourage peace among democracies, such as clear signals of resolve and reliance on successful policies, also play a role in sanctions.

Democracies are more likely to sanction non-democracies than other democracies due to the common reasons for imposing sanctions, promoting democracy and punishing human rights violations, which are largely applied to autocratic targets Lektzian and Souva [55] and Cox and Drury [61]. This is because democratic states are usually not the ones guilty of abusing their citizens’ political or human rights. Democratic leaders are also more motivated to pursue successful foreign policies, preferring non-democratic targets to offset or counter sanctions. Fearon’s notion of audience costs suggests that a democracy, with high domestic audience costs, is less likely to back down in public confrontations during international crises than a non-democracy, which has lower audience costs and greater flexibility to alter its policies in the face of foreign pressure. This leads to a signal of resolve sent by a democratic target of sanctions being more credible than one sent by an autocratic target. Galtung [2] supports the relative resilience of democratic targets by pointing out that democracies have greater legitimacy and are more likely than autocracies to rally their citizens around the flag of resistance to sanctions. However, the economic peace hypothesis is contentious, and the claim that a democratic target is less likely to concede to sanctions than a non-democracy is rejected by many scholars.

Nooruddin [62] argues that democratic political leaders are likely to agree to sanctions to lift the economic burden on their constituents, as they are compelled to consider their public’s preferences. Bolks and Al-Sowayel [63] show that democratic governments typically do not resist sanctions for long due to the domestic political costs imposed upon them. Nossal [64] notes that political leaders in target nations who fail to change their behavior to stop the economic pain caused by sanctions risk being ejected from office. In non-democracies, unpopular ruling elites can often protect themselves and their supporters by shifting the economic burden of sanctions on disenfranchised groups. Bolks and Al-Sowayel [63] argue that when the leadership of a state is concentrated in the hands of a few, the leadership is better able to implement countermeasures that insulate the government from the economic hardships caused by sanctions. Non-democratic and illiberal regimes find it easy to hold out in the face of damaging sanctions because they can pass on the costs of the sanctions to the governed and rely on armed forces to deter political opponents. Damrosch [65] contends that sanctions will almost inevitably benefit an autocratic regime because the regime will always be in a better position to control external transactions and the internal economy.

Bolks and Al-Sowayel [63] and Nooruddin [62] show that sanctions against autocratic targets are less successful than those against democracies. Nooruddin [62] concludes that sanctioners are more likely to sanction democracies because they are more likely to concede. Democracies are also more likely to use nonmilitary coercion, including sanctions, when confronting other democracies in inter-state disputes.

3. Research methods and data adopted for this study

Sanctions have become significant in recent years; however, there are limited studies on the impact of economic sanctions on income inequality in African countries, apart from the three studies by Porter [12], who uses the static input-output relationships of eight sectors in a linear programming model on South Africa to estimate the impact of trade and investment sanctions on the incomes of whites and non-whites. Khan [13], also on South Africa, uses multipliers from the Social Accounting Matrix (SAM), and Afesorgbor and Mahadevan [14] use cross-country analysis of 68 target states from 1960 to 2008.

This study seeks to utilize the data from 22 African countries and compare its findings with those of the Asian countries (19) to empirically investigate which region bears the cost of sanction between the two regions. The main objective of the current study is to analyze the impact of US and UN economic sanctions on the income inequality of the African target state and compare its results with those of the Asian target state. The Bayesian generalized method of Moment’s technique will be utilized to answer the specific obojective of the study covering the period 1980–2019 using the Bayesian generalized method of Moment’s technique.

Following Afesorgbor and Mahadevan [14], we use the Gini coefficient (Gini) extracted from the Standardized World Income Inequality Data (SWIID) of Solt [15] to measure income inequality. Unlike in their analysis, we take the point into mind that the Gini coefficient per nation generally reflects minimal fluctuations throughout time and is regarded as a very stable indicator of inequality. As a result, we employ the World Inequality Database’s pre-tax income held by the top 40% (dIncTII40) and additional pre-tax income held by the top 10% (dIncTII10) and the Palma ratio (incPalmar) as a robustness model [66].

On the other side, economic sanctions (ECO) comprise financial sanctions (FIN), which are defined as a type of sanction that consists of freezing financial transactions in the targeted states. Financial sanctions are becoming an important type of sanction since international trade is increasing and financial transactions are becoming top-notch due to the increase in international trade. While trade sanctions (EX and IM) evoke imports and exports, the European Union, the United States of America, and the United Nations are the three countries that impose sanctions. While trade openness is where they are constraining trade activities, therefore, the dummy variable takes the value one for the years when any of the principal senders imposed sanctions on the target states and zero otherwise.

We then capture for economic development, using the GDP per capita in constant prices (US$) (GDPp), trade openness will be derived for computing export minus import over GDP, government consumptions (GEXP) (captured by government spending as a percentage of GDP), population growth (PG), house price (HP), investment (INV) are extracted for Penn World Table 09 and others on the IMF. This study further controls the democratic pressure (DM) from the dichotomous Democracy and Dictatorship database developed by Cheibub et al. [67]. Lastly, for the dIncTII40 and dIncTII10 models, we included internal conflict (IC) as model sensitivity. The IC variable was adopted since a number of African countries experienced conflict in the past years. Therefore, for internal conflict, the study will use a dummy of one if strikes are incurred in selected countries and zero otherwise.

3.1 The two-step system dynamic panel data: BGMM

The study uses estimation techniques to support and verify the results of the BPVAR model. The two-step System Dynamic Panel Data through the Bayesian Generalized Method of Moment would be adopted as second baseline model in producing and quantifying the magnitude impact of macroprudential policy instrument on income inequality in a panel of 15 emerging countries. The BGMM equation can be expressed as follows:

3.1.1 Bayesian framework setup

$$Y_i=\sum _{r=1}^{\rho }{\theta }_k{X}_{\mathit{it}}+{ϵ}_i$$E1

where ${\mathit{Y}}_{\mathit{i}}$ is a N · 1 vector denoting the variable of interest, with $\mathbf{i}\mathbf{=}\mathbf{1}\mathbf{,}\mathbf{2}\mathbf{,}\dots \mathbf{,}\mathbf{N}$ , ${\mathit{X}}_{\mathit{it}}\mathit{=}{\mathit{X}}_{\mathit{i}\mathbf{1}}\mathit{,}{\mathit{X}}_{\mathit{i}\mathbf{2}}\mathit{,}\dots \mathit{.}\mathit{.}\mathit{,}{\mathit{X}}_{\mathit{im}}$ is a [$\mathit{N}·\mathit{m}$] matrix including a few, or a large set of continuous and/or discrete covariates, with $\mathit{k}\mathit{=}\mathbf{1}\mathit{,}\mathbf{2}\mathit{,}\dots \mathit{,}\mathit{m}$, ${\mathit{\theta }}_{\mathit{k}}\mathit{=}{\mathit{\theta }}_{\mathbf{1}}\mathit{,}{\mathit{\theta }}_{\mathbf{2}}\mathit{,}\dots \mathit{.}\mathit{.}\mathit{,}{\mathit{\theta }}_{\mathit{m}}$ is a $\mathit{k}·\mathbf{1}$ vector of unknown regression coefficients, and ${\mathit{ǫ}}_{\mathit{i}}\mathit{\sim }\mathit{N}\left(\mathbf{0},{\mathit{\sigma }}^{\mathbf{2}}\mathit{u}\right)$is a $\mathit{N}·\mathbf{1}$ vector of disturbances, with σ to be an unknown positive scalar. Here, for simplicity, the constant term is dropped, and it is assumed that the error component is independent and identically distributed ($\mathit{i}\mathit{.}\mathit{i}\mathit{.}\mathit{d}\mathit{.})$ and homoskedastic.

3.1.2 Dynamic panel data with GMM estimators setup

The baseline where the Bayesian setup is combine with the GMM is presented as follows:

$$Y_{\mathit{it}}={\delta }_i^{\prime }+\sum _{r=1}^{\rho }{\gamma }_r^{\prime }{W}_{\mathit{it}-r}+\sum _{i=0}^{\lambda }\sum _{\xi =1}^k{\theta }_{\mathit{l\xi }}^{\prime }{X}_{\mathit{it}-\mathit{l\xi }}+{\mathrm{u}}_{\mathrm{it}}$$E2

where Yi t is a NT ·1 vector of outcomes, δi is a N·1 heterogeneous intercept,Wi t − r is a NT ·1 vector of predetermined variables, Xi t − l,ξ is a NT ·κ matrix containing continuous/discrete endogenous variables, with l = 0, 1, 2, …, λ, r = 1, 2, …, ρ denotes generic Auto-Regressive (AR) orders for the predetermined variables, γr and θ$\tilde{\mathrm{\lambda }}$ξ are the autoregressive coefficients to be estimated for each i and couple of (i,ξ), with $\tilde{\mathrm{\lambda }}$ = 1, …, λ, and ui t ∼ i. i. d. N(0,σ2 u) is a NT · 1 vector of unpredictable shock (or idiosyncratic error term), with E(ui t) = 0 and E(ui t ·uj s) = σ2 u if i = j and t = s, and E(ui t ·uj s) = 0 otherwise.

Here, some considerations are in order: (i) the predetermined variables contain the lagged values of the outcomes Yi t and lags of heterogeneous individual-specific factors; (ii) the δi’s denote cross-unit heterogeneity affecting the outcomes Yi t; (iii) a correlated random effects approach is adopted in which the δi ‘s are treated as randomvariables and possibly correlated with some of the covariates within the system; (iv) the roots of r (B) = 0 and $\tilde{\mathrm{l}}$ (B) = 0 lie outside the unit circle so that the AR processes implicit in the model (6) are stationarities, with $\tilde{\mathrm{l}}$ = 1, 2, …, $\tilde{\mathrm{\lambda }}$ denoting generic AR orders for the endogenous variables and B referring to the lag operator; and (v) the instruments are fitted values from autoregressive parameters based on all available lags of time varying variables and their causal interactions.

4. Empirical analysis, data analysis, and interpretation of results

4.1 Data analysis

Before running the BGMM model in this study, we conducted several data inspections, such as the unit root test, the descriptive statistics, and others in the background, to better understand our data. Table 1 displays the descriptive statistics for the various variables. According to descriptive statistics, these nations’ average total income inequality is approximately 60.78% for African targeted states and 57.43% for Asian targeted states, while economic sanctions are around 0.58% and 0.43%, respectively.

Variables	African targeted state				Asian-targeted states
	Mean	Std.d	Min	Max	Mean	Std.d	Min	Max
Gini	60.784	17.439	26.350	83.423	57.43	14.546	19.543	79.098
incPalmar	50.972	2.728	20.128	73.338	42.28	7.943	10.398	63.122
dIncTII40	56.097	4.983	23.039	69.100	38.93	2.483	24.849	80.183
dIncTII10	50.202	5.8983	19.928	78.441	45.83	3.508	20.684	74.765
ECOUS	0.580	0.780	0	1	0.437	0.470	0	1
FINUS	0.409	0.987	0	1	0.565	0.382	0	1
EXUS	0.387	0.877	0	1	0.394	0.454	0	1
IMUS	0.105	1.098	0	1	0.245	0.234	0	1
ECOUN	0.398	0.095	0	1	0.604	0.143	0	1
FINUN	0.551	0.105	0	1	0.543	0.345	0	1
EXUN	0.330	0.283	0	1	0.454	0.187	0	1
IMUN	0.178	0.302	0	1	0.210	0.250	0	1
GDPp	7.980	1.402	6.948	10.234	6.039	2.940	5.938	13.983
DM	0.510	1.002	0.159	0.895	0.465	3.984	0.209	60.345
HP	75.089	14.392	30.142	80.100	71.093	9.837	13.983	84.523
OPEN	74.400	13.983	34.934	78.985	77.981	18.456	26.981	83.420
GEXP	78.455	16.023	39.143	80.545	68.393	8.565	11.987	76.095
INVE	35.989	1.203	25.782	45.805	28.289	1.983	10.765	50.384
PG	55.105	2.934	24.345	78.995	59.123	0.944	20.987	70.984
IC	23.112	8.178	12.134	47.393	17.432	5.458	6.876	50.065

Table 1.

Data (descriptive statistics).

To assess the variables used in this study, we evaluated at the partial existence of the for cross-sectional dependence and cointegration, as proposed by Friedman [68], Frees (1995), and Pesaran [69]. The findings of cross-sectional dependency and Pedroni cointegration are shown in Table 2.

Regions (s)	Pedroni tests for cointegration			Tests for cross-sectional independence
Africa	Augmented Dickey-Fuller t	4.99	Pr = 0.008	Friedman’s test	154.94	Pr = 0.00
	Modified Phillips-Perron t	3.20	Pr = 0.003	Frees’ test	2.95	Pr = 0.00
	Phillips Perron t	4.65	Pr = 0.040	Pesaran’s test	16.00	Pr = 0.00
Asia	Augmented Dickey-Fuller t	6.03	Pr = 0.000	Friedman’s test	124.11	Pr = 0.00
	Modified Phillips-Perron t	2.00	Pr = 0.000	Frees’ test	3.56	Pr = 0.00
	Phillips Perron t	5.19	Pr = 0.000	Pesaran’s test	18.40	Pr = 0.00

Table 2.

Cointegration and cross-sectional independence tests.

Source: Author’s calculation results based on data from [70].

The null hypothesis of no cointegration and cross-sectional reliance on variables is strongly rejected by by both the cross-sectional reliance tests, as well as the Pedroni cointegration test. After assessing the data, the BGMM process will be undertaken. The results of the BGMM are presented in the subsequent sections.

4.2 Bayesian generalized methods of moments results and discussion

To model the impact of US and UN economic sanctions on income inequality in Africa-targeted states and Asian-targeted states, this study adopted the Bayesian GMM, covering the period 1980 to 2019. For African-targeted states, the study adopted 22 countries, while for Asian-targeted states, we adopted 19 countries. The main purpose of adopting the two regions was to compare the impact of US and UN economic sanctions on income inequality in these two regions. Further, to determine the sanctions that are more effective between the US and the UN sanctions in these two regions, We adopted various measures of income inequality in this study, which involve the Gini coefficients obtained from the Standardized World Income Inequality Data (SWIID) version. However, for the coefficient interpretation purposes, we will main focus on the Gini coefficients obtained from the SWIID for interpretation purposes as the main model in the first column. Table 3 reports the results of both the African-targeted states and the Asian-targeted states. The results demonstrate three main, important findings for the study: (1) it signifies that the US economic sanctions are stronger than the UN sanctions across all four types of sanctions adopted in this study; (2) African countries suffered the greater impact of economic sanctions than Asian countries; and (3) when sanctions are implemented, they tend to exacerbate income inequality in the low-income earner (dIncTII40); however, when financial or asset restrictions are imposed, they serve as income redistribution as they have a negative impact on the income of the high-income earner (dIncTII10).

	African targeted state				Asian-targeted states
	Gini	incPalmar	dIncTII40	dIncTII10	Gini	incPalmar	dIncTII40	dIncTII10
ECOUS	4.07 ** (1.00)	3.87** (1.98)	3.20* (0.89)	1.30 ** (0.50)	1.00 ** (0.50)	2.66 (1.11)	0.76 ** (1.40)	1.90 *** (0.50)
FINUS	4.40 ** (0.79)	2.78** (0.12)	1.78** (0.12)	−2.40 (3.00)	1.10 ** (1.00)	3.59 ** (1.40)	4.70 ** (1.29)	−2.70 ** (0.70)
EXUS	−4.00 ** (0.20)	−1.90 ** (0.10)	−1.78 (1.12)	−1.70 ** (0.24)	−2.10 ** (1.24)	−2.38 (3.70)	−2.44** (1.09)	−1.50** (0.50)
IMUS	2.00 ** (1.38)	3.10 ** (0.69)	2.30** (0.22)	2.00 ** (0.90)	1.79 ** (0.20)	2.00 ** (0.80)	2.70 ** (0.29)	1.70 ** (0.60)
ECOUN	1.00 *** (0.09)	2.87 (2.30)	2.20** (0.60)	0.90 ** (0.50)	2.00 ** (0.10)	1.33** (0.11)	0.76 (1.40)	1.30 *** (0.80)
FINUN	2.20 ** (0.79)	1.04*** (0.20)	1.02** (0.32)	−1.40* (0.80)	3.30 ** (1.30)	1.99 ** (0.40)	2.30 ** (1.00)	−0.90 ** (0.12)
EXUN	1.20 *** (0.20)	1.09 ** (0.50)	0.90** (0.12)	1.10 ** (0.34)	4.70 ** (2.00)	2.55** (1.05)	3.20 (1.50)	1.50** (0.33)
IMUN	0.10 ** (1.00)	1.23 ** (0.30)	2.0** (1.02)	1.00 ** (0.20)	3.00 ** (0.20)	1.90 ** (1.06)	2.00 ** (1.00)	1.03 ** (0.50)
GDPp	−2.87 ** (1.20)	−2.09 ** (0.88)	−1.44 ** (0.50)	2.88** (1.08)	−2.88** (1.00)	−2.95 ** (0.63)	−2.00** (0.80)	1.99 ** (0.30)
DM	−1.53 ** (0.13)	−2.0*** (0.70)	−2.33 ** (0.80)	−2.60 ** (0.20)	−1.90 ** (0.20)	−2.50 ** (0.99)	2.40 * (1.0)	−1.50 ** (0.50)
HP	2.96** (0.78)	2.88** (0.98)	1.73** (0.8)	−1.79** (0.18)	1.45 (4.60)	1.79** (0.50)	2.00** (0.20)	−2.08 (3.60)
OPEN	−2.77 ** (1.00)	−2.88** (0.40)	1.56 (1.90)	2.23** (0.10)	−2.70** (1.00)	−2.29** (0.43)	2.09 ** (0.23)	1.89 ** (0.25)
GEXP	−2.07 ** (0.39)	−3.41** (0.38)	−2.11** (1.48)	2.73** (1.00)	−2.00 ** (0.38)	−2.80*** (0.08)	−1.91 ** (0.22)	2.60 ** (0.23)
INVE	−2.88** (0.78)	−1.95*** (0.03)	−3.15** (1.03)	3.56** (1.80)	−1.00** (0.10)	−2.49** (1.18)	−2.44** (1.00)	2.99** (0.48)
PG	2.85** (0.18)	2.77* (0.89)	1.77* (0.80)	2.44** (0.10)	2.20** (0.90)	2.00** (0.80)	2.60** (0.90)	1.00** (0.03)
IC			2.54* (0.93)	2.00** (0.70)			1.93** (0.40)	1.09** (0.32)
AR(1):p-	0.007	0.004	0.008	0.005	0.09	0.097	0.005	0.009
AR(2):p	0.320	0.230	0.450	0.580	0.565	0.655	0.604	0.594
NC	22					19

Table 3.

US and UN sanctions on income inequality of the African and Asian targeted state.

Note: Dependent variable is the income inequality captured by SWIID, incPalmar, dIncTII40 and dIncTII10. The numbers in brackets denote the standard errors in brackets are obtained by using the cluster-robust and heteroskedasticity-consistent covariance estimator, allowing for error dependency within individual countries. (***), (**), (*) reflect the 1%, 5%, 10% level of significance, respectively. The p are the p-values, and NC is the number of countries.

Source: Author’s calculation results based on data from [70].

When all other parameters are held constant, the coefficient for ECOUS in column 2 for Africa-targeted states and in column 6 for Asian-targeted states shows that, on average, a 1% rise in US economic sanctions increases income inequality by 4.07% and 2.00%, respectively. The results are comparable with all measures of income inequality adopted in this study. A further increase in income inequality in both African-targeted states and Asian-targeted countries was witnessed following a 1% increase in US financial sanctions (FINUS), which increased income inequality by 4.40% (African) and 1.10% (Asian). These findings were further supported by the results when we used incPalmar and dIncTII40 as measures of income inequality. However, some interesting results are found when we use pre-tax income held by the top 10% (dIncTII10) to capture income inequality, as the results show the negative impact of financial sanctions on the income inequality of the top 10% earners.

In nutshell, our results show that income inequality among the top 10% earners decreases by 2.40 for African targeted states and by 2.70 for Asian targeted states following a 1% increase in financial sanctions. The logic behind financial sanctions to reduce income inequality is based on the premise that punitive measures can discourage economic practices that perpetuate wealth disparities. Therefore, by imposing penalties on certain activities or individuals, the aim is to incentivize fairer wealth distribution and discourage exploitative behavior. Such sanctions can target tax evasion, illicit financial flows, or unethical business practices. They create a deterrent effect by increasing the cost or risk associated with the perpetuation of income inequality, thereby promoting more equitable economic systems. Additionally, the punitive nature of financial sanctions sends a strong message that societies will not tolerate activities that exacerbate income inequality while also deterring potential wrongdoers from engaging in those practices.

The study further reported that, on average, a 1% rise in US export sanctions (EXUS) decreases income inequality by 4.00% in African targeted states, while for Asian targeted states, it decreases by 2.00%. The results are comparable when incPalmar, dIncTII40, and dIncTII10 are used as proxies for income inequality. However, for dIncTII40, the results are insignificant in Africa, while in Asia, the incPalmar model is insignificant. The explanation behind the negative impact of export sanctions on reducing income inequality lies in the belief that limiting exports of certain goods, such as high-tech products or natural resources, can protect domestic industries and jobs. This, in turn, aims to address income inequality by fostering economic growth and opportunity within the country. As for import sanctions (IMUS), the estimation results in Table 3 show a similar effect with the impact of EXUN sanctions, where a 1% increase in the imposed IMUS increases income inequality in both the African targeted state and the Asian targeted state by 2.00% and 1.79, respectively, all other factors being equal. Results obtained using the incPalmar, dIncTII40, and dIncTII10 measures of income disparity provided more evidence in favor of this.

Similar to the impact of US economic sanctions except for EXUS, even UN economic sanctions were found to contribute to high income inequality in these two regions. As we found, all the UN sanctions, such as the UN economic sanctions (ECOUN), financial sanctions (FINUN), UN export sanctions (EXUN), and import sanctions (IMUN), as a result of a 1% increase in ECOUN, FINUN, EXUN, and IMUN, result in a 1.00%, 2.20%, 1.20%, and 0.20%, respectively, increase in income inequality for African targeted states, holding the other factors constant. However, for the Asian targeted states, a 1% increase in ECOUN, FINUN, EXUN, and IMUN results in a 2.00%, 3.10%, 4.70%, and 3.00%, respectively, increase in income inequality, holding the other factors constant. The result is comparable when incPalmar, dIncTII40, and dIncTII10 are used as a proxy for income inequality. We found that across all measures of income inequality except for financial sanctions and export sanctions in the high-income group, the models yield similar results to those of our baseline model.

The findings are so interesting, as we find that US sanctions are more severe than UN sanctions in African-targeted states. The possible logic behind these results can be explained as follows: US sanctions are often more severe than UN sanctions in African countries due to a combination of factors. Firstly, the US wields considerable economic and political influence globally, which allows it to impose unilateral sanctions on nations it believes are engaging in activities contrary to its interests. African countries often bear the brunt of these sanctions, as they are vulnerable to the economic repercussions. Secondly, while UN sanctions are typically designed with consensus among member states, US sanctions are imposed independently and can be more punitive. This is because the US has specific national security concerns and geopolitical interests in Africa related to terrorism, regional stability, and human rights. Consequently, the US often applies harsher measures targeting specific sectors, businesses, or government entities in African countries. Lastly, the enforcement mechanisms of US sanctions, such as banning dollar transactions or restricting access to international financial systems, exacerbate the severity of their impact on African nations’ economies, making them more severe than UN sanctions, which may lack similar enforcement mechanisms. This finding is thoerical pluasable and inline with the Stolper-Samuelson theorem, and consistent with previous empirical studies that demonstrated a substantial positive effect of economic sanctions on income inequality, such as [13, 14, 36, 40, 41, 42, 43, 44, 45, 46].

The study then controls for GDP per capita (GDPp), democracy (DM), trade openness (OPEN), investment (INVE), house price (HP), government consumption (GEXP), population growth (PG), and internal conflict (number of coups) (IC). In both African target states and Asian target states, GDP per capita (GDPp) has a statistically negative impact on income inequality across all measures of income inequality except for the pre-tax income held by the top 10% (dIncTII10), where we found that it promotes income inequality. The results demonstrate that following a 1% increase in GDPp results in a decrease in income inequality of 2.87% for the African region, while for Asina, it results in a 2.88% decrease in income inequality. The logic behind this is that if GDP per capita increases, it implies that the overall wealth of the nation is growing, which can potentially benefit everyone, including those with lower incomes. Policies aimed at increasing GDP per capita, such as investing in infrastructure, education, healthcare, and creating employment opportunities, can have a positive impact on reducing income inequality.

However, some interesting results were observed when the study used the dIncTII10 to capture income inequality, as we document that a 1% increase in GDPp results in an increase in income inequality by 2.88% for the African targeted state and by 1.99% for the Asian targeted state. In the context of the dIncTII10, which represents the income held by the top 10% of income earners, an increase in GDP per capita may not necessarily lead to decreased inequality, and in some cases, it can exacerbate it. This is because GDP per capita measures the average income of the entire population, and if the gains from economic growth are not evenly distributed, it can disproportionately benefit those in higher income brackets. For example, if a country experiences significant economic growth through industries that primarily benefit the wealthier segments of society, such as finance or technology, the increase in GDP per capita may reflect a higher average income without necessarily benefiting the lowest-income earners. This can further widen the income gap between the high-income earner and the rest of the population, leading to increased inequality. Therefore, to tackle inequality among high-income earners, it is crucial to have policies and strategies in place that prioritize inclusive growth, ensuring that the benefits of economic development reach all segments of society, rather than focusing solely on GDP per capita as a measure of progress. Our results support the findings documented by Andries and Melnic [71].

When all other parameters are held constant, the coefficient for democracy (DM) in column 2 for Africa-targeted states and in column 6 of the Asian targeted states model shows that, on average, a 1% rise in the improvement in democracy decreases the level of income inequality by 1.53% and 1.90%, respectively. Our results support the finding documented by [14, 72]. These results are comparable even when incPalmar, dIncTII40, and dIncTII10 are used as proxies for income inequality. The logic behind democracy to decrease inequality lies in its fundamental principles of inclusion, representation, and accountability. By providing equal rights and opportunities to all individuals, democracy creates a framework that can bridge the gap between the rich and the poor. Through fair and free elections, citizens can elect leaders who are committed to addressing social and economic inequalities. Democratic systems also promote transparency and accountability, allowing citizens to hold their representatives responsible for their actions and decisions. Additionally, democracy fosters public participation and empowers marginalized communities, ultimately reducing inequality by ensuring that everyone has a voice and can influence the decision-making process.

A further increase in income inequality in these regions was captured following an increase in house prices (HP). However, for the pre-tax income held by the top 10% (dIncTII10), some interesting findings were documented, as we found that it decreased income inequality for the dIncTII10 for both African and Asian targeted states. The results in column 2 for Africa-targeted states and in column 6 of the Asian targeted states model show that on average, a 1% rise in house price increases the level of income inequality by 2.96% and 1.45%, respectively. While with respect to the pre-tax income held by the top 40% (dIncTII40), on average, a 1% rise in house price decreases the level of income inequality by 1.79% for the African target state and 2.08 statistically insignificantly for the Asian target state. Our results support the findings documented by Dewilde and Lancee [73] and Fuller et al. [74].

In our study, we controlled for trade interaction using trade openness (OPEN). The results for model Gini and incPalmar for both the African and Asian targeted states show that, on average, a 1% increase in trade openness the level of income inequality by 2.77% (Gini), 2.88% (incPalmar) for the African targeted state, and by 2.70% (Gini), 2.29% (incPalmar) for the Asian targeted state. However, for dIncTII40 and dIncTII10, the results show that income inequality increases by 1.56% and 2.23%, respectively, for African targeted countries, and by 2.09% and 1.89%, respectively, for Asian targeted states. Trade openness impacts income inequality differently across countries. It disproportionately benefits the very poor in emerging and developing economies, while in most advanced economies, it increases income inequality. Our results support the finding documented by [75].

GEXP has a statistically negative impact on income inequality in both regions. However, when the study captures income inequality using the pre-tax income held by the top 10% (dIncTII10), the results show a positive impact of government expenditure on income inequality. This means that, on average, a 1% increase in government spending decreases the level of income inequality by 207% for African targeted states and by 2.00% for Asian targeted states. The result is comparable when incPalmar and dIncTII40is used as the proxy for income inequality. While, with respect to dIncTII10, the results show that a 1% increase in government spending increases income inequality by 2.73% and 2.60%, respectively. This empirical finding is in line with the results reported by Zungu and Greyling [76] in the 10 African emerging economies from 1988 to 2019. According to Tanzi [77], fiscal policy thought government spending alone may do nothing to reduce income inequality but may also make it worse.

Investment (INVE) has a negative and statistically significant effect on income inequality in both regions when the study used the Gini, Inc. Palmar, and Inc.TII40 to capture income inequality. However, it is positive when income is captured by the pre-tax income held by the top 10% (dIncTII10). This shows that, on average, a 1% increase in investment decreases income inequality by 2.88% (Gini), 1.95% (incPalmar), and 3.15% (dIncTII40) for African targeted states and 1.00% (Gini), 2.49% (incPalmar), and 2.44% (dIncTII40) for Asian targeted states. On the other hand, on average, a 1% increase in investment increases income inequality by 3.56% for African-targeted states and 2.99% for Asian-targeted states. The findings confirmed those of Blonigen and Slaughter [78] for the United States, Figini and Görg [79] for 100 developed and developing economies and Zungu et al. [76] for merging Economies.

Population growth (PG) was found to promote income inequality across all measures used to capture income inequality in this study for both regions. The results show that a 1% increase in population results in an increase in income inequality of 2.85% for African-targeted states and 2.20% for Asian-targeted states. These results were supported by the findings yielded from the incPalmar, dIncTII40 and dIncTII10. Our results support the finding documented by [80, 81].

Finally, we included internal conflict (IC) as a model sensitivity to models IncTII40 and IncTII10 in both regions. The results for all the other variables seem to yield similar results as those of Gini and Inc. Palmar, even if we include an addition variable as a control variable in the models Inc.TII40 and Inc.TII10. A further increase in income inequality across all measures used to capture income inequality (Inc.TII40 and Inc.TII10) is documented following an increase in internal conflict (IC) in both regions. The results show that, on average, a 1% increase in IC raises income inequality by 2.54% and 2.00%, respectively, for the African targeted state and 1.94% and 1.09%, respectively, for the Asian targeted state, which is statistically significant. The results confirmed the findings documented by Afesorgbor and Mahadevan [14] and Dylan O’Driscoll [44]. Theoretically, the argument behind the positive impact of internal conflict on income inequality could be primarily fueled by economic systems that prioritize the accumulation of wealth and favor the interests of a privileged few. This conflict arises from the unequal distribution of resources, unequal access to opportunities, and a lack of fair wages and benefits for marginalized groups, resulting in widespread income disparities.

5. Conclusion and policy recommendations

This study aims to contribute to the literature by empirically examining the impact of US and UN sanctions on the income inequality of the targeted African states and comparing the results with those of the targeted Asian states using the Bayesian GMM model. When we adopt the fixed effect as our robustness model, according to the researcher’s knowledge, no studies have compared the impact of US and UN economic sanctions on African-targeted states with those of Asian-targeted states. Moreover, no studies in the literature have been conducted on the current subject. We believe that this study is crucial for policymakers in these countries, as the practice of sanctions has become the norm in many countries. As we further seek to compare the results by tracing which sanctions are more severe between the sanctions executed by the US and the UN in these African and Asian targeted states, if US sanctions are more severe than UN sanctions, what does that mean for these countries?

Our results are so interesting as they show that US sanctions are more severe than UN sanctions for African-targeted states. However, for Asian-targeted states, UN sanctions are more severe than US sanctions in this region. This was analyzed based on the magnitude impact coefficient of the adopted US and UN sanctions in this study, as demonstrated in Table 4.

African targeted state		Asian-targeted states
US sanctions	UN sanction	US sanctions	UN sanction
ECOUS 4.07	ECOUN 1.00	ECOUS 2.00	ECOUN 2.00
FINUS 4.40	FINUN 2.20	FINUS 1.10	FINUN 3.30
EXUS 4.00	EXUN 1.20	EXUS 2.10	EXUN 4.70
IMUS 2.00	IMUN 0.10	IMUS 1.79	IMUN 3.00

Table 4.

The coeffiecient summary of the US and UN sanctions.

The results in Table 4 show that the magnitude impact coefficient of the US sanctions is two times higher in almost all sanctions compared to the UN sanctions for the African region. However, for the Asian-targeted region, it shows that the impact of UN sanctions is more severe than US sanctions. Moreover, the results demonstrate that economic sanctions indeed contribute to the income inequality of the receiver, regardless of the sender, as we find that both US and UN sanctions contribute to the income inequality of these regions (African targeted states and Asian states). The study further controlled fiscal policy using government expenditures. Across all the measures of income inequality except for the pre-tax income held by the top 10% (dIncTII10), we found that government expenditures reduced income inequality in both regions. On the other side, house prices were found to contribute to income inequality in both regions across all measures of income inequality. However, some relief was recorded following some improvement in economic development captured by GDP per capita across all measures of income inequality adopted in this study, including the pre-tax income held by the top 10% (dIncTII10). The estimated results for this study are plausible and consistent with results reported by Neuenkirch and Neumeier [36], Jin [43], Jeong [42], [46], and Eslamloueyan and Kahromi [45]. For policy implications, the implementation of strategic economic, financial, and trade sanctions is crucial. These sanctions can address the root causes of income inequality by encouraging fair and equitable economic practices. Financial sanctions aim to limit access to markets, freeze assets, and restrict the financial transactions of individuals, entities, or countries that engage in unfair wealth accumulation or perpetuate income inequality. By targeting individuals or institutions that exploit loopholes, engage in corrupt practices, or evade taxes, financial sanctions can help level the playing field and promote greater economic fairness. Economic sanctions can be implemented to incentivize countries to adopt more inclusive economic policies that prioritize income redistribution. By imposing restrictions on trade or investment with countries that fail to implement reforms or take steps to address inequality, economic sanctions create pressure for governments to enact policies that protect workers’ rights, promote social safety nets, and ensure fair wages. Trade sanctions can also contribute to addressing income inequality by encouraging fair trade practices. Imposing restrictions or tariffs on countries that engage in unfair competition, such as exploiting workers or evading environmental regulations, can help protect domestic industries and workers from being undermined by lower-cost, unethical practices. This promotes a more level playing field and contributes to reducing income disparities. Therefore, strategic financial, economic, and trade sanctions can serve as effective policy tools to combat income inequality. By targeting and incentivizing fair economic practices, these sanctions can contribute to a more equitable and inclusive economic system, ultimately benefiting the whole society.

Appendix

See Table A1.

Africa (22): Angola, Cameroon, Central African Republic, Democratic Republic Congo, Eritrea, Ethiopia, Gambia, Guinea-Bissau, Kenya, Liberia, Libya, Malawi, Niger, Nigeria, Rwanda, Sierra Leone, Somalia, South Africa, Sudan, Uganda, Zambia, Zimbabwe.; Asia (19): Afghanistan, Cambodia, China, India, Indonesia, Iran, Iraq, Israel, Jordan, Lebanon, Myanmar, North Korea, Pakistan, South Korea, Syria, Thailand, Uzbekistan, Vietnam, Yemen.

Table A1.

List of sample countries.