Social policies in democratic and authoritarian regimes: comparing the introduction and implementation of the National Rural Employment Guarantee in India and Dibao in China

Abstract Research on the welfare state often examines social policies in democratic regimes separately from social policies in authoritarian regimes. Two bodies of research have emerged, as the extant literature views these political systems as sufficiently distinct to merit the division of analysis. In this article, we challenge the existing approach by showing that differing regime types can indeed be analysed together. By looking for patterns of similarities, rather than differences, we bring the two literatures into conversation and show how a common factor can trigger social policy expansion in both regime types. Using case studies of India and China, the two most populous democratic and authoritarian regimes, this article illustrates how the expansion of policies that serve low-income groups – India’s Mahatma Gandhi National Rural Employment Guarantee (NREGA) and China’s Minimum Livelihood Guarantee Scheme (dibao) – were both prompted by social mobilisation.


Introduction
Conventional wisdom suggests that democracies are more likely to introduce and implement social policies than autocracies. Yet social policies can be found in both democratic and authoritarian regimes (Mares & Carnes, 2009). Indeed, there are notable examples of social policy expansion in democracies (Diaz-Cayeros, Estevez, & Magaloni, 2016;Jenkins & Manor, 2017) in addition to authoritarian regimes (Albertus, 2015;Pan, 2020;Croke, 2020).
While empirical evidence suggests that both regime types have embarked on social policy expansion, the welfare state literature approaches the analysis of social policies by emphasising differences in political systems. This has resulted in two bodies of literature: the first consisting of research that focuses exclusively on democracies (Bersch, 2019;Brown & Mobarak, 2009;Dorlach 2021;Wong, 2006) and the second composed of studies investigating autocracies (Albertus, 2015;Chen, Pan, & Xu, 2016). Studies that examine both together do so to determine whether one regime type produces better outcomes than the other (Blaydes & Kayser, 2011;Donno, Fox, & Kaasik, 2022;Haggard & Kaufman, 2008;Harding, 2020;Harding & Stasavage, 2014;Huber & Stephens, 2012;Kudamatsu, 2012;McGuire, 2010;Wang, Mechkova, & Andersson, 2019), maintaining the analytical focus on differences between the two regimes.
By contrast, this article looks for common drivers of policy change. While this may seem like a subtle shift in the conversation, it is in fact a significant reframing that allows us to look for patterns of similarities, rather than differences. In doing so, we advance the literature by bringing two important, but often separated, bodies of research into conversation. We show that the introduction of social policies in both regimes can be triggered by social mobilisation, which we define as a bottom-up activity where ordinary citizens try to elicit political change through mass action. While the underlying strategies for managing unrest are shaped by political regimein democratic systems, governments introduce social policies that respond to civil society with the goal of achieving electoral victory; in authoritarian regimes, where civil liberties and electoral competition are weak or absent, governments use social policies to maintain social control and pre-empt future unrestgovernments responses reveal patterns of equifinality.
To illustrate, we compare two politically significant social assistance programs that are intended to serve the poor: India's Mahatma Gandhi National Rural Employment Guarantee, commonly abbreviated as NREGA, which provides wage employment for the poor, and China's Minimum Livelihood Guarantee Scheme, colloquially known as dibao, which provides unconditional cash transfers to lowincome households. We also consider the implementation of these policies and the extent to which they have been able to achieve their pro-poor goals.
In India, severe droughts led to social unrest in the late 1990s and early 2000s. Grassroots organisations began mobilising, claiming that the government was failing to provide adequate food and employment. The introduction of NREGA was part of the Congress Party's platform, used to respond to social unrest and achieve an electoral victory. However, the same democratic forces that contributed to the introduction of NREGAelectoral and party politicsundermined policy implementation. In their bid to maintain political power, there were instances where local governments awarded work contracts based on party affiliation rather than need. This contributed to the underprovision of government assistance for the poor.
In China, citizens witnessed rising levels of unemployment and income inequality in the 1980s and 1990s. This led to a growing number of protests, especially amongst laid-off workers. Like India, the government in China moved to address social unrest. The Chinese government introduced dibao in response to mobilising workers. Over time, it began to use dibao to maintain social control, targeting specific groups and individuals it believed would protest in the future. As the government could use the program to manage social unrest, dibao was rapidly scaled up. However, the government targeted individuals who were potential threats to the regime, rather than the poor, undermining the policy's benefits to the poor.
The cases of India and China show how countries with varying political regimes responded to social mobilisation, expanding pro-poor policies in the late 1990s and 2000s. While the countries' political regimes shaped the pathways they travelled down, they nevertheless arrived at similar policy outcomes. We elaborate on the impact of democracy and authoritarianism by beginning this article with an overview of the welfare state literature on social policies and political regime. Following this, we show how a common variable, such as social mobilisation, can appear in both regime types and trigger policy change, suggesting that the literature on democratic and authoritarian regimes can be brought into a fruitful conversation. After discussing our methodological approach, we illustrate our argument in the India and China cases. In the concluding section, we touch on a research agenda that describes how further case studies, complemented by large-n studies, can continue to bridge the divide between literatures.
Social policies and political regime Social policies are government efforts to improve population welfare and well-being. Governments have the final say on these policies; however, they do not make policy decisions in a vacuum. All governments operate within a broader political regime. The political regimethe set of rules that govern power relations between the state and society (Przeworski, 2000) is defined by political rules relating to civil liberties and electoral competition (Teorell et al., 2019).
Civil liberties include the freedom of association and the freedom of speech and expression (McGuire, 2020). The freedom of association ensures that individuals are free to form and participate in groups, either formally or informally, then organise to achieve collective goals (Kapiszewski, Levitsky, & Yashar, 2021). Individuals have the right to join up with others, present their demands, then take action to pursue their demands free from government interference. Relatedly, the freedom of speech exists when individuals can express their beliefs in public and private spaces. This includes beliefs that criticise the government.
Electoral competition encompasses an electoral process where there are multiple political parties, providing choice to the voting populace. Adult citizens enjoy universal and equal suffrage and can vote for the candidate or political party of their choice. Elected candidates are duly installed in office and form a functioning government within a reasonable period.
While few, if any, countries are purely democratic or authoritarian (Dahl, 1971), countries can be characterised as leaning in one direction or another. Democratic regimes are characterized by expanded civil liberties and electoral competition. By contrast, under authoritarian rule, civil liberties and electoral competition are heavily curtailed or absent.

How political regime affects pro-poor social policies
Research links democratic political regimes with expanded social policies (Dorlach 2021;Lavers & Hickey, 2021). Indeed, scholars contend that "democratisation processes have strengthened the political voice and influence of low-income groups" and "shaped effective demand for policies addressing poverty and vulnerability" (Barrientos, 2013, p. 11). In democratic regimes, the freedom of association enables the poor, and groups representing the poor, to organise for change that favours their interests (Ganz, 2009;Pribble, 2011). The freedom of speech allows these groups to express their grievances and exert pressure on political leaders. They use a variety of strategies, including the framing of demands (Díez, 2013) and protests such as roadblocks, demonstrations, and occupations of buildings (Garay, 2007) to legally voice their discontent and elicit government action. These freedoms strengthen the political influence of low-income groups, increasing the likelihood that a government introduces policies that meet their needs.
At the same time, electoral competition creates incentives for political elites to respond to the demands of the poor. Politicians have sought to achieve electoral gains by strategically introducing social policies, such as universal healthcare coverage, that serve the needs of low-income groups (Aspinall, 2014).
However, democratic regimes are not the only political systems which have expanded pro-poor social policies. Research has investigated the origins of social assistance (Leisering, 2008) and shown that an overwhelming number of social insurance programs were initially adopted by authoritarian governments (Mares & Carnes, 2009). Indeed, scholars have documented how some of the world's most redistributive regimes have been authoritarian (Albertus, Fenner, & Slater, 2018). If the poor do not have the right to organise and express their grievances, and are unable to participate in a competitive electoral process, then why would authoritarian regimes distribute beyond their closest political allies? Put another way, what is the political logic that underpins the expansion of social policies serving the poor?
To remain in power, authoritarian regimes can use strategies of co-optation, providing social benefits to neutralise potential threats and maintain political control. For example, these regimes have responded to the threat of rebellion by expanding social policies to groups that threaten social stability (Chen et al., 2016). Often, this is done in response to the demands of labour, which has historically had the most political and bargaining force (Rudra & Haggard, 2005). Social policies have therefore been used to buy peace and pre-empt social unrest.
Drawing on the literature on historical institutionalism, we argue that social and economic crises can act as a critical juncture (Capoccia and Kelemen, 2007, p. 343). These crises can trigger social mobilisation which, in turn, threatens a government's political survival. For example, social mobilisation can undermine popular confidence in government and threaten its legitimacy. In other contexts, it can cause violence and disorder and subsequently disrupt a host of critical economic activities. Increasingly many countries are characterised by a political climate where "the poor have gained political predominance as the main grassroots source of political threat" (Yoruk, 2022, p. 3). Therefore, governments have strong incentives to manage social unrest amongst the poor, as the failure to do so could threaten their political power.
In democratic systems, governments are obligated to permit social mobilisation, as this is a core civil liberty. Public discontent is therefore funnelled through the electoral process, where incumbent governments must adequately respond to continue in office. Political parties that are viewed as responding to citizens' demands are rewarded with access to office. The goal, then, is to use social policies to win elections. By contrast, authoritarian political systems limit civil liberties and electoral competition. These governments use social policies to maintain political control, targeting those who could potentially cause unrest. Therefore, governments, regardless of regime type, expand social policies in response to social mobilisation or the threat of social mobilisation. Social policies are, in effect, "one tool within the broader political survival strategies employed by ruling elites to maintain power" (Lavers & Hickey, 2021, p. 2).
Once introduced, the implementation of pro-poor policies can be influenced by a government's political calculations, rather than the needs of the poor. For instance, in democratic regimes, governments may allocate social assistance based on party affiliation, rather than financial need, in order to maintain electoral support. Without strong systems of accountability, democracies can fail to effectively implement pro-poor policies. Similarly, in authoritarian regimes, governments may provide social assistance to individuals deemed to be political threats, rather than those solely experiencing poverty, to pre-empt social unrest. In both regimes, then, we see the expansion of pro-poor social policies, but also the underprovision of government assistance for the poor.

Methodology
A case study is an intensive study of a single case (George & Bennett, 2005). Comparative case studies, which examine two or more cases, entail within-case analysis, then a comparison across contexts to identify broader patterns. In this article, we carry out comparative case studies, examining India and China and their social assistance programs. We adopt a "hypothesis-generating" approach and analyze the "cases for the purpose of developing more general theoretical propositions" (Levy, 2008, p. 5). Using process tracing, an intensive analysis of a sequence of events over time, we inductively identify patterns of similarity between political regimes. Our analysis pays close attention to the sequence of events in the introduction and implementation of pro-poor social policies.
We selected India and China for our case studies. Both are middle-income countries with large populations. During the same period, the countries introduced social policies with the stated goal of serving the poor. We focus on the late 1990s/2000s for pragmatic reasons. It is recent enough to access relevant policy information, but enough time has passed to assess policy change. We therefore use published research as our main data source, drawing upon scholarly works from academic books and journals, in addition to policy reports and documents, to develop our analysis. While the countries share similarities in income status, population size, and policy characteristics, India's political regime has democratic features, whilst China's political system has authoritarian features. This presents the opportunity to investigate the role of political regime in shaping countries' trajectories.
According to the expectations laid out by democratic theory, India should vastly outperform China in terms of its pro-poor social policies. However, China has rapidly expanded dibao. We investigate this puzzle by examining NREGA and dibao, which are significant social assistance programs in comparative studies of India and China (Roy, Yörük, & Gürel, 2018). In doing so, we contribute to a rich and dynamic body of research that explicitly compares India and China (Frazier, 2019;Singh, 2021).
India: democracy and NREGA NREGA embodies the right to work and obtain access to income by entitling any household living in a rural area to a minimum of 100 days of employment per year. NREGA relies on self-selection rather than a targeting mechanism (Shankar & Gaiha, 2013). A set amount is put aside for NREGA in the federal budget each year, with funds spent depending on how many households avail themselves of the labour opportunities provided through NREGA.
In this analysis, we focus on how the introduction of NREGA was linked to the bottom-up mobilisation of civil society organisations, in conjunction with electoral and party politics.
In the fight for NREGA, civil society worked to put the right to employment onto the agenda of policymakers. Though NREGA is organisationally rooted in Maharashtra's experience with the right to work, which began in various forms in the 1970s, its primary political roots lie in the grassroots activism of Rajasthan's Mazdoor Kisan Shakti Sangathan (MKSS), or Worker and Farmer Power Organisation. MKSS worked in the early 2000s to bring together a broad civil society movement in support of the right to work.
Electoral politics contributed to civil society's ability to get the right to employment onto the agenda of India's policy makers. On the eve of the 2004 general election, civil society found support in the Congress Party, the largest party in the United Progressive Alliance (UPA), which was challenging the incumbent Bharatiya Janata Party (BJP or the Indian People's Party)-led National Democratic Alliance (NDA). Following Congress' electoral victory, civil society continued to use its influence to ensure the legislation of an employment guarantee. After NREGA's passage, Congress claimed credit and was rewarded by the electorate (Gulzar & Pasquale, 2017). However, the implementation of NREGA has been fraught with challenges (Aggarwal, 2016;Dey, Debmalya, & Ram, 2022). In particular, there have been cases of local governments awarding work contracts based on party affiliation rather than need (Das & Maiorano, 2019;Dasgupta, 2016), undercutting the policy's propoor goals.

Mobilising for change: the role of civil society
In the late 1990s and early 2000s, severe droughts affected the Indian state of Rajasthan. Grassroots organisations, such as the MKSS, formed to defend the rights of those affected. They argued that the government was not effectively implementing drought relief programs, pointing to some 50 metric tons of food grains that sat in government warehouses throughout India (Jenkins & Manor, 2017). In 2001, a coalition of activist groups filed a Public Interest Litigation at the Supreme Court, claiming that state governments were violating the Indian Constitution. The government's inability to implement existing food and nutrition programs violated the right to life. They also began to call on the government to uphold a right to employment, which they saw as central to ensuring the right to food.
In May 2001, the Supreme Court issued an interim order requiring state governments to change how they fulfilled the "right to food," which it stated was central to the "right to life" (Jenkins and Manor 2017, p. 38). A set of court-appointed commissioners was created and responsible for supervising all nutritionrelated programs throughout the country. The mobilising efforts of grassroots organisations had therefore succeeded in putting the issue on to the Indian government's agenda.

Introducing NREGA: electoral and party politics
The role of democratic politics can be classified into two categories: non-party and party politics (Khera, 2013). Non-party politicsthe activism of social movements, people's campaigns, and NGOsplayed an important role in drawing initial attention to the right to work. In addition, party politics during the 2004 election cycle was instrumental in ensuring the ultimate passage of NREGA. There are several ways in which political parties in India have developed strategies to achieve electoral success: the use of Hindu nationalism, caste politics, political dynasties, and the emergence of demagogues as leaders (Kohli, 2012). In the case of NREGA, Sonia Gandhi, the national president of the Congress Party, sought to counter the BJP's focus on Hindu nationalism by developing a rural electoral strategy that promised the passing of an unemployment guarantee.
Prior to the 2004 general election campaign, the Congress Party, recognising the need to engage with the increasingly vocal social movements in favour of a right to work, met with MKSS and other activist groups. Gandhi was persuaded by the activists' arguments in favour of a right to employment. With Congress deciding to move towards a more 'rural' strategy for the election, and with Gandhi's support, activists were able to convince party leaders to include the right to work within their electoral platform. The Congress-led UPA went on to win the 2004 election. While their success was based largely on urban support, it was interpreted by much of the media as the revolt of the rural poor against the BJP-led NDA (Jenkins & Manor, 2017). Congress chose to embrace this idea of a rural revolt, even if it lacked factual support, seeking to develop a long-term electoral strategy that emphasised the party's pro-poor credentials.
Following the election, Congress established a National Advisory Council (NAC), which served as a social policy development forum and included actors within and outside government, including academics, civil society activists, and former civil servants. The NAC was chaired by Gandhi, whose stature automatically increased the Council's importance (Chopra, 2011). The NAC took as its first major task the passage of legislation on the right to employment, creating an initial draft of the legislation. The draft included extensive transparency mechanisms meant to enable the poor to hold government actors accountable for implementing the program. The Finance Ministry raised concerns, citing the fiscal danger of such an open-ended promise (Chopra, 2014). Together with the Planning Commission, it sought to add an amendment that would allow the Indian government to suspend the program, and a clause that would limit eligibility to families classified as below the poverty line.
These changes were included in the draft National Rural Employment Guarantee (NREG) Bill introduced to Parliament in December 2004. The new watered-down draft Bill was immediately criticised by civil society, who saw it as a betrayal of a promise to a right to work (Chopra, 2014). Members of the NAC and civil society rallied together to place pressure on Parliamentarians to protect the original provisions of NREG, including engaging with BJP leaders in a strategy of embarrassing Congress for abandoning its initial campaign promises. The Bill was referred to a Parliamentary Standing Committee (PSC) on Rural Development, which ultimately opened the Bill to public consultation.
Civil society and members of the NAC continued to engage with the PSC to alleviate their cost concerns. The final recommendations of the PSC restored the promise of guaranteed employment without exceptions. During final rounds of revisions, Congress' reliance on leftist parties to maintain its UPA coalition allowed these parties to extract concessions and to return the final Act to something resembling the drafts initially formulated by the NAC. As a result, both the continuing agitations of civil society and the role of coalition party politics helped to ensure that some of the main promises of the right to work were ultimately retained. Therefore, civil society mobilisation in favour of the right to work, in conjunction with electoral competition, ensured the Act's eventual passage.

Implementing NREGA: electoral and party politics and the limits of redistribution
Democratic politics helps to explain the expansion of social policies like NREGA, but also some of the redistributive limitations. While NREGA has been fraught with several issues (Aggarwal, 2016;Dey et al., 2022), we focus on the role democratic institutions played in the underprovision of services for the poor. In particular, the electoral and party politicswhich brought NREGA into existencecontributed to these implementation issues.
NREGA increased the power of the panchayat raj, the system of elected local government, expanding the resources it managed and its degree of decision-making power. As demand for employment exceeded the work contracts available, local government officials had considerable discretion in deciding who to allocate projects to. "Indeed, much of the time, opportunities to undertake NREGA work are 'rationed' by powerful actors in local arenasusually by the heads of elected local councils, acting in concert with fellow panchayat members and administrative officials" (Jenkins & Manor, 2017, p. 70).
In certain contexts, the rationing of work contracts was based on party affiliation rather than need. For example, in Rajasthan, "localities which elect a legislator from the state-level ruling party are likely to gain priority in the allocation of scarce NREGA projects since opposition party politicians do not possess the same degree of political influence over the bureaucracy." (Dasgupta, 2016, p. 4). This reflected a broader trend in India, where public works projects were distributed along partisan lines, rather than need (Bohlken, 2018). In the Indian state of Andhra Pradesh, the government used NREGA to distribute government contracts to party-affiliated contractors. In this context: the politicians' main interest is…to provide government contracts to their network of contractors, who funds their political campaigns in return, so that political leaders can distribute cash, liquor and other goods on the eve of the elections in a 'classical'and widespread, in the Indian contextvote buying strategy. (Das & Maiorano, 2019, p. 251) This suggests that electoral and party politics can contribute to the introduction of social policies like NREGA, as electoral victory can be achieved by political parties that strategically respond to social mobilisation. However, the same forces influence policy implementation and resulted in "political practices" that often diverged "from the ideals of democratic equality" (Fischer & Ali, 2019, p. 157).

China: authoritarianism and Dibao
In China, economic dislocation and distress led to skyrocketing levels of unemployment and poverty in the 1980s and 1990s. The labour protests that ensued triggered the introduction of dibao in 1999, an unconditional cash supplement that provides individuals with basic income to purchase necessities. Recipients receive a cash transfer directly from the government.
In this section, we show that the government did not introduce dibao to win elections. Rather it used the social policy as a tool to pre-empt social instability, targeting specific groups and individuals who might protest and petition in the future. As social instability could threaten the government's political survival, it had strong incentives to scale up dibao. Reflecting this, the social assistance program rapidly grew, spreading across the country. However, the program, which "addresses in principle all poor persons or households, based on need," (Leisering, Liu, & ten Brink, 2017, p. 309) has been criticised for failing to serve those facing destitution. The majority of dibao's beneficiaries do not actually qualify for the program (Golan, Sicular, & Umapathi, 2017;Hammond, 2019), leading observers to question whether the program improves the lives of the poor.

Mobilising for change: the consequences of China's market transformation
In the late 1970s, the Chinese government began a process of reform and opening, transitioning from a centrally planned economy to a market-based system (Gao et al, 2013). This was a fundamental change. Before the introduction of a market-based system, Mao Zedong, chairman of the Chinese Communist Party from the establishment of the People's Republic of China (PRC) to his death in 1976, had pursued egalitarian policies with the working class as the bedrock of support. A significant portion of the country's labour force was employed by state-owned enterprises (SOEs). While these organisations were uncompetitive, they ensured long-term employment and welfare benefits (Chen, 2000). As such, pre-reform China was characterised by high levels of poverty, but low levels of income inequality.
After Mao's death, Deng Xiaoping gradually rose to power. He is credited for introducing China's farreaching market-based reforms. Under Deng, capitalism with Chinese characteristics contributed to the rapid economic growth that would eventually lift millions out of poverty. However, the economic transition was marked by the commodification of labour (Wang, 2002). The system produced winners and losers and subsequently, growing income inequality. At the beginning, the government managed the effects of economic growth by protecting workers' interests. For example, it provided dual track pricing and subsidies for groups affected by economic reform to minimise contention over economic reforms (Yang, 2006).
By the early 1980s, however, the government began to prioritise economic growth over workers' interests. Efforts to make China into a global manufacturing hub began and took advantage of the cheap wages and unorganised labour (Hui & Chan, 2011). The effects of this approach became apparent and "disputes over wages and welfare, job responsibilities, labor discipline, promotions, and occasionally women workers' rights arose as market-oriented reformers delegated greater powers to urban enterprise managers to raise labor productivity and competitiveness" (Chan & Selden, 2019, p. 105). The priority of Chinese policies was economic development and rapid GDP growth, at times at the expense of equality (Fisher, Shang, & Blaxland, 2010). At the same time, SOEs began to fail. Statesector employment dropped rapidly, contributing to further tension between the government and labour.
The transition to a market-based system was accompanied by a process of rapid urbanisation. In the pre-reform era, China was an agriculturally dominated country. However, as China transitioned to a market-based economy, urbanisation was viewed as a mechanism for growth and modernisation. As market reforms were introduced, millions of workers migrated to cities. The Chinese urban population increased by an average of 16 million residents annually between 1978-2015 (Guan et al., 2018). Workers began building infrastructure for the industrial sector including roads, bridges, and high-rise buildings (Wu & Rao, 2017). However, the broader economic transition was not inclusive.
The shift towards a market-based economy was initially characterised by widening income inequality that was especially pronounced in urban centres. While the government had required SOEs to provide living allowances in the 1990s, fewer SOEs provided these payments as the years progressed (Chen, 2000). SOE reforms contributed to massive worker layoffs, particularly among poorly educated and low-skilled workers with limited opportunities for re-employment (Leung, 2006). In addition, poor working conditions and low wages became prevalent and "the downward mobility for much of the working class…created a new crowd of the dispossessed in urban China" (Yang, 2006, p. 151). At the same time, millions of rural-to-urban migrants were not covered by state welfare benefits as the pre-existing hukou system differentiated citizens as either an 'agricultural' or 'non-agricultural' group (Wu & Rao, 2017). The growing income inequality in urban China set the stage for widespread and mounting social discontent.

Introducing Dibao: placating protesting workers
Widespread labour tensions became a significant threat to the Chinese government. As market reforms accelerated, workers' living standards declined. Restructuring entailed delayed payments on wages and the suspension of pensions. In addition, layoffs occurred on a large scale. From 1993 to 1998, the number of laid-off workers increased from 3 million to 17.24 million (Cai, 2002). By 2002, over 60 million urban workers were laid off from state sector jobs (Chan & Selden, 2019). Poverty skyrocketed in urban areas and workers struggled to meet their basic needs, contributing to feelings of injustice. At the same time, information was spreading more easily, facilitated by the close contact inherent to urbanisation. Urban workers increasingly voiced their discontent, and pro-democracy ideas became prevalent in the growing city centres (Gao, Yang, & Li, 2013).
Labour's support of pro-democracy ideas threatened the very foundations of China's authoritarian state. Their protests were not demanding a return to the system of SOEs but rather the respect of workers' rights. Large-scale labour protests multiplied across Chinese cities. In 1995, an estimated 1.1 million workers were involved in protests. A short 3 years later in 1998, this grew to over 3.6 million workers (Chen, 2000). Protest incidents included traffic blockages, collective petitions, and demonstrations in front of government headquarters. Most of the demonstrations were composed of workers affected by SOE reforms.
Protests became endemic across China. Ultimately, government officials recognised that they could not suppress the conditions faced by the working class without creating greater resentment and outrage. Labour unrest was becoming a breaking point for social stability. Political elites believed that settling the growing social unrest was necessary for economic restructuring to progress. The government adopted strategies of co-optation to diminish the threat of labour protests on social and political stability.
The labour protestsresulting from economic dislocation and distresstriggered government action. In 1999, the Chinese government introduced dibao in urban centres (Leisering et al., 2017). On paper, this was introduced "in urban areas to assist people in extreme poverty and vulnerability" (Barrientos, 2013, p. 4). However, it also came at a time when large-scale labour protests were creating political stability and a policy instrument was needed to placate workers affected by the market transformation. In the following years, the government budget allocated to dibao continued to grow. Notably, the government was most responsive to the actors that threatened social stability (Huang, 2013), rather than groups claiming loyalty and membership to the Chinese Communist Party (Chen et al., 2016).
Implementing Dibao: pre-empting dissent through expanded policy Large-scale labour unrest was initially concentrated in urban centres. Reflecting this, dibao was introduced to urban regions. The program grew rapidly, expanding from 2.8 million recipients in 1999, to 19.3 million recipients in 2002, to 23.3 million recipients in 2009 (Solinger & Hu, 2012). Over time, protests spread outward to the rural areas. As wealth was concentrated in urban areas, rural residents were left behind. In addition, agricultural workers were burdened with high taxation and subjected to land appropriation. The wealth gap widened and the average per capita income in rural areas became less than a third of that in urban centres. As a result, the intensity and frequency of protests grew from an estimated 8,700 mass incidents in 1993 to 180,000 mass incidents in 2011 (Yeh, O'Brien, & Ye, 2013).
The government responded by introducing the rural dibao program in 2007 (Leisering et al., 2017). Since then, the rural dibao program has grown significantly larger than its urban counterpart (Golan et al., 2017). For instance, from 36 million recipients in 2007, rural dibao increased its coverage to 50 million recipients in 2011, which was then twice the size of urban dibao. Up to mid-2021, rural dibao covered 35.4 million, more than four times as many as urban dibao which covered 7.7 million poor residents (Government of China, 2021).
The government had introduced dibao to quell social unrestand the implementation of the policy followed a similar logic. On paper, dibao was meant to target those affected by poverty. But in practice, it was distributed to individuals the government wished to control (Pan, 2020). Dibao's targeted populations included "individuals whom the regime has insufficient evidence to prosecute but who are placed under intensive surveillance and control" so the regime could "prevent them from committing "crimes" such as protesting and petitioning in the future" (Pan, 2020, p. 3).
As the government could use the social policy to manage social unrest, the program's coverage increased rapidly. However, the social policy did not necessarily serve low-income individuals facing destitution, but rather low-income individuals the regime wanted to manage: …while Dibao was still provided to the poor, local administrators often denied Dibao to 'ordinary' impoverished households but made sure that all impoverished households with individuals who were listed as targeted populations, who were ex-prisoners, and who were undergoing re-education received Dibao. (Pan, 2020, p. 11) Indeed, scholars have noted that the majority of dibao's beneficiaries do not actually qualify for the program (Golan et al., 2017;Hammond, 2019). In the context of authoritarianism, the government introduced the program to maintain the social and political order.

Conclusion
Scholars have long been interested in the relationship between political regime and social policies, examining the ways in which different systems of governance influence social assistance programs. These studies have explored the impact of democracy on social policies, or the impact of authoritarianism on social policies. In this article, we bring these two bodies of literature into dialogue by capturing the experiences of both political regimes. We do so by showing governments introduce social policies in response to social mobilisation. In India, a regime that is broadly characterised as democratic, the government introduced NREGA to respond to civil society mobilisation and achieve an electoral victory. In China, an authoritarian regime, the government introduced dibao in response to labour protests. It subsequently used dibao to maintain social control, targeting groups and individuals believed to have the potential to cause unrest. The India and China cases show how the two literatures on democracy and autocracy can be bridged by looking at a common variable. This serves to advance the welfare state literature, which has overwhelmingly separated the analysis of social policies by regime type.
In addition, the case studies suggest that there are cases where both democratic and authoritarian regimes introduce social policies in response to social mobilisationbut fall short in their implementation goals. Additional research examining the population-level implications is needed. While an investigation of policy impact is outside the scope of this article, further research would advance knowledge on the effect of these social assistance programs on levels of poverty and other socioeconomic indicators. In addition, future research could expand the number of cases considered. For instance, in Latin America, scholars could compare Cuba, an authoritarian regime, with a democratic regime such as Costa Rica. Studies that expand the number of cases would also provide greater insights into the generalizability of the findings. Further research should also examine policy areas beyond social assistance, such as universal health care, as these would help explore how our argument varies across different policy domains.
The key contribution of this article, then, is to point out a separation in the literaturethat scholarly research often separates the analysis of democratic and authoritarian regimes. We bring together work on democracy and authoritarianism by drawing attention to the role of social mobilisation in introducing social policies in both democratic and authoritarian regimes. In addition, we shed light on two empirical cases, India's NREGA and China's dibao, advancing knowledge on programs that affect a significant proportion of humanity.