SEARCH

SEARCH BY CITATION

Keywords:

  • social protection;
  • millennium development goals;
  • PRSPs;
  • scorecards;
  • vulnerability;
  • equity

Abstract

  1. Top of page
  2. Abstract
  3. 1 INTRODUCTION
  4. 2 DEFINING SOCIAL PROTECTION, VULNERABILITY AND RISK
  5. 3 LITERATURE REVIEW
  6. 4 DATA AND METHODOLOGY
  7. 5 RESULTS AND DISCUSSION
  8. 6 CONCLUSION AND POLICY RECOMMENDATIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. APPENDIX A
  12. APPENDIX B
  13. APPENDIX C

Social protection and equity are missing elements of the Millennium Development Goals. Both are essential to redress the situation of the most vulnerable. This study provides a comprehensive review of social protection packages in Poverty Reduction Strategy Papers and devises an index to evaluate the extent of a country's level of social protection in its development strategy. The findings show that the social protection agenda in Poverty Reduction Strategy Papers is more pronounced in countries with higher levels of income and lower levels of ethnic diversity and that social assistance packages are more aligned to higher levels of official development assistance and governance. Copyright © 2013 John Wiley & Sons, Ltd.

1 INTRODUCTION

  1. Top of page
  2. Abstract
  3. 1 INTRODUCTION
  4. 2 DEFINING SOCIAL PROTECTION, VULNERABILITY AND RISK
  5. 3 LITERATURE REVIEW
  6. 4 DATA AND METHODOLOGY
  7. 5 RESULTS AND DISCUSSION
  8. 6 CONCLUSION AND POLICY RECOMMENDATIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. APPENDIX A
  12. APPENDIX B
  13. APPENDIX C

Social protection is a set of public actions designed to reduce levels of vulnerability, risk and deprivation. It is an important mechanism in directly addressing the issues of inequality and vulnerability (Guillaumont, 2001; Gaiha & Imai, 2008; Barrientos & Hulme, 2009; Barrientos, 2011). The Millennium Development Goals (MDGS) are a set of time-bound global targets designed to reduce poverty and improve health, gender equality and environmental sustainability. With the 2015 deadline for the MDGs looming, increasing equality has emerged as a missing component in the agreed targets (Fukuda-Parr, 2010; Vandemoortele, 2011). Social protection is a direct way of filling the gaps that the MDG agenda has failed to capture, and the World Bank's Poverty Reduction Strategy Papers (PRSPs) represent a practical means of implementing a broader agenda. PRSPs are often considered as the back-end of the MDGS. Although they do not galvanise the same level of support, they represent greater potential for country-specific strategic plans to build more fluid and responsive policies. This study examines the adoption rates of social protection within PRSPs to identify which policies are more prevalent, and then examines possible factors that could be associated to developing countries adopting a social protection agenda.

The world has become an uncertain place, and risk aversion has become a central feature of the global economy. Natural disasters have more than doubled in the last thirty years1, energy prices have increased two-fold, and food prices have markedly increased in the last twenty years.2 The global economic crisis is continuing to influence the world's financial and labour markets, and only a few countries have escaped its impacts (World Bank, 2012). Climate change is also contributing to this uncertainty, with changes in weather patterns and temperature. Social protection is the mechanism to protect against economic other shocks, and increasingly, individuals must devise means to manage and overcome risk.

This study contributes to the literature by addressing the following research question: what are the factors which explain the variation in countries adopting social protection in their PRSPs? As PRSPs are designed to be locally led, it also assesses the donor influences on policies, and whether the intentions to implement social protection in developing countries are realised at the country level. It uses a unique scorecard system to assess the degree of adoption of social protection programmes in the 87 PRSPs in 59 countries from 2000–2011. The study builds upon the Social Protection Index (SPI) created by Baulch and Wood (2008) for the Asian Development Bank (ADB). Cross-tabulations are used to determine the common country characteristics of those PRSPs deemed as high-scoring social protection adopters. The study found PRSP recipients with a high income per capita, and low levels of ethnic diversity were more likely to be aligned to the SPI. Countries adopting social assistance programmes are best explained by common determinants of development. Of all the components of social protection, labour standards were the least implemented programmes in the PRSPs.

The second section of this paper defines the terms social protection, vulnerability and discusses the absence of inequality and vulnerability in the MDGs and how the PRSPs can fill this gap. The third section reviews the literature on the impacts and content of social protection. The fourth section discusses the data and methodology. The fifth section presents and interprets the results and the final section reviews the findings and provides policy options for future planning.

2 DEFINING SOCIAL PROTECTION, VULNERABILITY AND RISK

  1. Top of page
  2. Abstract
  3. 1 INTRODUCTION
  4. 2 DEFINING SOCIAL PROTECTION, VULNERABILITY AND RISK
  5. 3 LITERATURE REVIEW
  6. 4 DATA AND METHODOLOGY
  7. 5 RESULTS AND DISCUSSION
  8. 6 CONCLUSION AND POLICY RECOMMENDATIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. APPENDIX A
  12. APPENDIX B
  13. APPENDIX C

Conceptually, social protection, vulnerability and risk are all intertwined. Social protection's role is to help households deal with economic, natural and social shocks. It is the household's exposure to risk and their capacity to cope with such shocks that determine their level of vulnerability (Holzmann, 2000).

2.1 What is Social Protection?

This study uses the ‘social protection’ definition as suggested by Baulch and Wood (2008): as ‘a subset of public actions that address risk, vulnerability and chronic poverty…. a set of social policies and programmes designed to reduce poverty and vulnerability by providing efficient labour markets, diminishing exposure to risks and enhancing their capacity to protect themselves against hazards and the interruption and loss of incomes’ (Baulch & Wood, 2008 p.7).

Although there is a consensus on the need for social protection, this does not extend to defining social protection (Gentilini & Omama, 2011). Definitions fall under three categories: intentions, programmes or a combination of the two. Definitions that are operational can be narrow in focus, concentrating only on risk, whereas definitions based on the intentions of social protection serve as a set of aspirational goals of empowerment and opportunity rather than prescriptive inclusions.

Clear-cut definitions based on social protection components are contentious because of disagreements regarding which programmes fall under which components as potential cross over exists within the components. Barrientos (2011) outlines three categories: social assistance, social insurance and labour market regulations. Baulch et al. (2006) includes the two additional components of micro and area-based schemes (community-based), and child protection. Gentilini (2011) has a separate category for safety nets (inclusive of transfers and subsidies) and groups labour market programmes and insurance (health, pensions and minimum wages) in the same category; it also includes a new category for social services related to policies for health, education, nutrition and agriculture. Other authors have included a fourth component for rights and legislation.

Many researchers have explored intention's definitions (Holzmann, 2000; Baulch et al., 2006; Barrientos, 2011) and these broad definitions focus on the roles social protection has in public policy and political and social transformation. Norton, Conway and Foster (2001) define social protection as ‘public actions taken in response to levels of vulnerability and risk and deprivation which seem socially unacceptable within a given polity or society’ (p.7). Whereas Holzmann (2000) takes a much more micro approach by defining social protection as a set of public interventions to increase individuals, households and communities' ability to manage risk and provide support to the poorest of the poor. The three conventional components of social protection are examined: social assistance, social insurance and labour market regulation.

Social assistance covers non-contributory cash or in-kind transfers and welfare assistance (such as conditional and non-conditional cash transfers). Social insurance refers to contributory schemes that reduce the risks associated with old age, disability, unemployment and sickness. These programmes represent policies for better-off individuals to prevent them dropping below the poverty line (Farrington & Slater, 2006). Labour market regulation guarantees basic employment standards for workers and legitimate rights for labour organisations. The number of components and programmes included in these categories is still widely debated in the literature (Gentilini & Omama, 2011).

2.2 What is Vulnerability? Why Differentiate Between Vulnerability and Poverty?

Vulnerability is a dynamic concept directly relating to the likelihood of risk in the future (Imai et al., 2010). In contrast, poverty is a static concept addressing a situation at a specific point in time. Technically, vulnerability refers to the probability of a household's exposure to shocks and how that household will cope; it equates to the probability of remaining poor or becoming poor in the future (Holzmann, 2000; Alwang et al., 2001).

Vulnerability and protection are opposing concepts. A lack of social protection for households at risk of falling below the poverty line makes it difficult for these households to develop coping strategies. The ability to cope with change involves being able to ‘bounce back’ to where they were before. Ideally, households develop resilience (i.e. the ability to cope with risk). Globalisation and market integration have opened up opportunities for developing nations, but there are elements of globalisation that leave some households exposed (World Bank, 2012). An equitable society has an obligation to protect those who are unable to protect themselves, and social protection is an effective tool with which to address the dynamic nature of vulnerability (OECD, 2009).

2.3 MDGs and the Missing Components

The MDGs are a set of global time-bound targets and Sindicators with the focus of halving poverty in 25 years. The MDGs culminated in September 2000 at the Millennium Summit, where after a decade of meetings formulating a new development agenda, eventually 189 countries and 143 leaders agreed to scale up aid to meet a series of global development goals.3 They concentrate on the areas of health, education, water provision, gender equity and the environment. The targets have been relatively static, with the exception of four MDG targets that were included in 2007. The expanded targets are related to reproductive health and opportunities for decent work. As the 2015 deadline looms,4 questions arise about what will replace the current set of targets and indicators.

Fukuda-Parr (2010) found inequality and vulnerability to be missing elements in the MDG broad list of development goals. Social protection and inequality played a secondary role to goals of poverty reduction when the MDGs were developed in the year 2000. Since this time, there has been a shift in development thinking as represented by the genesis of the Social Risk Management (SRM) framework at the World Bank in 2000/2001(Holzmann, 2000). The Social Protection Unit oversaw this framework to integrate risk and vulnerability into the World Bank's approach to development.

It is possible for countries to achieve MDG targets and, at the same time, fail to raise the living standards for the poorest of the poor. Without equity, the vulnerable will be left behind as targets can be achieved between countries but not necessarily within countries. As flagged earlier, social protection provides the most direct way of overcoming the inequality gap. Vandemoortele (2011) is critical of within-country disparities, where countries achieve MDG targets while neglecting the bottom quintile. Most notably, governments are able to achieve MDG targets by focusing on urban areas, where the provision of services is more accessible to a majority of the population while neglecting those remote rural areas in need of these services.

2.4 PRSPs: What are they? Can they Fill the MDG Target Gaps?

As development strategies, the PRSP's purposes are two-fold. In the first instance, they represent country-specific plans to reflect each country's development path and priorities to tackle poverty reduction. In the second instance, collectively, they represent the global influence and development priorities of the World Bank and International Monetary Fund (Tarabini & Jacokvis, 2012). The PRSPs framework is intended to be locally owned and generated and developed through a wide participatory dialogue with civil society.

The PRSPs have been the vehicle to translate the MDGs into country plans (Hulme, 2010); however, neither social protection nor equity is included in the initial MDG targets outlined in the year 2000. PRSPs are of the same genesis,5 these national plans that are designed to evolve and respond to the changing development landscape, as by design, a new PRSP lasts for a period of three years. The United Nations provides the role of the goal setter via the MDGs, and the PRSPs role is to devise the plan of how to achieve these targets. Although decent work opportunities are new MDG targets, equity, social assistance and social insurance targets are missing from the MDGs, and the absence of these goals detracts attention from core concerns that could dethrone the original intentions of the Millennium Summit.

3 LITERATURE REVIEW

  1. Top of page
  2. Abstract
  3. 1 INTRODUCTION
  4. 2 DEFINING SOCIAL PROTECTION, VULNERABILITY AND RISK
  5. 3 LITERATURE REVIEW
  6. 4 DATA AND METHODOLOGY
  7. 5 RESULTS AND DISCUSSION
  8. 6 CONCLUSION AND POLICY RECOMMENDATIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. APPENDIX A
  12. APPENDIX B
  13. APPENDIX C

The literature on social protection can be divided into three categories: definitional and operational aspects; impacts of social protection; and quantitative outcomes of social protection. There is a fourth category included in this study: social protection and PRSPs.

3.1 Policy Content

The definitional and operational aspects of social protection are covered extensively by the multilateral and bilateral donors, international NGOs and Gentilini (2011) and Barrientos (2011). Recent reports from ILO (2011), World Bank (2012), and DFID (DFID, 2004; DFID, 2011; DFID, 2011; DFID, 2005) the United Nations discuss recent developments and the evolution of social protection in developing nations. The impact of social protection on poverty and inequality is the subject of a number of studies (Mares, 2005; Farrington & Slater, 2006; Barrientos & Hulme, 2009; Barrientos, 2011; Ulriksen, 2012). Recent developments in the efficacy of conditional cash transfers in Latin America, in particular, have demonstrated clear links to poverty reduction and decreasing inequality (De Brauw & Hoddinott, 2011; DFID, 2011).

Quantitative studies into the impacts of social protection are relatively sparse. Barrientos and Hulme (2009) and Gentilini and Omama (2009) advocate that the policy makers in developing countries should design social protection programmes specific to the stage of development and institutional capacities. Both find that countries without the capacity should concentrate on social assistance programmes, whereas those with higher capacities should concentrate on social insurance. Gentilini (2009) creates a scale of absence of social protection systems to consolidated social protection systems for developing nations.

3.1.1 Social protection index

Three studies devised an SPI: Baulch and Wood (Baulch et al., 2006; Baulch & Wood, 2008) and Mares (2005). The first index was devised with the support of the ADB and covers four components comprehensively: social protection expenditure, social protection coverage, poverty targeting and social protection impacts. Initial SPI by Baulch et al. (2006) allows countries to compare the relative social protection performance of six Asian nations, and Baulch and Wood (2008) second study reviews 31 Asian nations with the same SPI. A lack of data in developing countries makes it difficult to replicate an SPI for each developing nation as social protection expenditure and coverage are not easily identifiable in national accounts as there is no uniform agreement as to the components considered being part of the social protection expenditure. Mares (2005) created an SPI that ranks countries from 0 to 10 depending on the level of coverage each government provides for disability, unemployment benefits, family and pensions. Definitions of social protection would find this index wanting as it only sparsely covers the core aspects of social protection.

Studies into the social protection content in PRSPs found inequity and vulnerability missing (Marcus & Wilkinson, 2001; Fukuda-Parr, 2010). Marcus' (2001) analysis of six full and 17 interim PRSPs6 found two-thirds of PRSPs explicitly addressed social protection issues. However, only a small number of PRSPs addressed issues of equity and distribution.

4 DATA AND METHODOLOGY

  1. Top of page
  2. Abstract
  3. 1 INTRODUCTION
  4. 2 DEFINING SOCIAL PROTECTION, VULNERABILITY AND RISK
  5. 3 LITERATURE REVIEW
  6. 4 DATA AND METHODOLOGY
  7. 5 RESULTS AND DISCUSSION
  8. 6 CONCLUSION AND POLICY RECOMMENDATIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. APPENDIX A
  12. APPENDIX B
  13. APPENDIX C

This devises an SPI based on 87 PRSPs to measure the degree of social protection uptake in the PRSP documents. It is the most comprehensive assessment of social protection in PRSPs to date covering all PRSP from 2000 until 2011,7 from 59 countries. Scores measuring the degree to which social protection policies are present within PRSPs are obtained using the criteria in Table 1. Higher scores indicate a stronger alignment/adoption to a social protection framework.

Table 1. Social protection framework for Social Protection Index
Social Assistance IndexSocial Insurance IndexLabour market programmes IndexAdditional components to overall Social Protection Index
  • *

    Baulch et al. (2006).

  • **

    Farrington (2006).

  • ***

    Sabates–Wheeler & Devereux (2004).

1. Cash-in-kind transfers agricultural inputs, shelter, non-food items**1. Unemployment insurance*1. Labour market legislation to protect labour rights***1. Priority or pillar for social protection in the Poverty Reduction Strategy Paper
2. Subsidies for housing, energy, SSS and food**2. Unconditional unemployment payments**2. Child labour protection – labour code***2. Microfinance*
3. Educational *assistance Scholarships3. Health/sickness insurance*3. Minimum Wage*** 
4. Fee waivers for essential services.**4. Non-contributory pension schemes**4. employment promotion, matching people to jobs*** 
5. Disaster relief programmes—funds for emergency relief or post-emergency transitions.**5. Contributory pension schemes 
6. 7 Targeted conditional cash transfers for service delivery **6. Disability pensions* 
7. Programmes for vulnerable groups: the elderly, disabled widows and, orphans.*7. Maternity allowances* 
8. Programmes for the internally displaced: migrants and refugees**8. Industrial injury payments* 
 9. Family payments** 

The index of social protection draws upon the social protection programmes included in the ADB's 2006 index. There are three categories of programmes that inform the overall index, and they follow Barrientos' (2011) definition.

The social assistance programmes take reference from the Baulch et al.'s (2006) SPI, which ‘targeted at the sick, poor, orphans, disabled and other vulnerable groups, cash-in-kind transfer and temporary subsidies for utilities and food’ (p.7). There are additional elements of a social assistance package, drawn from Farrington and Slater (2006) research into the validity of cash transfer payments, which identify more specific programmes to concentrate on the needs of the most vulnerable. These programmes include disaster relief, fee waivers, conditional cash transfers and packages for displaced persons and refugees.

The social insurance component is derived from Baulch et al. (2006) and includes programmes to cover the risks associated with unemployment, sickness and maternity, disability, industrial injury and old age. The study by Baulch et al. (2006) had family payments in a child protection component but, in this study, they fall under the category of social insurance. Additionally, pensions and employment insurance are separated into contributory and non-contributory schemes, which aligns with Farrington and Slater (2006) version of social insurance.

The labour market component is a combination of Baulch and Wood (2008) employment matching component and the concept of transformative social protection is particularly with reference to labour market legislation and minimum wages and the child labour (Deveraux & Sabates–Wheeler, 2004; Sabates–Wheeler & Deveraux, 2007).

Table 1 is a summary of each of the three categories and the additional components that are included in the overall index. The SPI is created by summing the social assistance index, the social insurance index and the labour standards index, and including the additional components of microfinance and priority or pillar of social protection in the PRSP.

The second stage of the analysis involves devising a scorecard that grades the degree of policy adoption. Each PRSP receives a score for each social protection policy; there are 24 individual policies that form the SPI. The scorecard method is a quantitative evaluation of PRSP policy content, previously used by Bojo and Reddy (2002) Eggen and Bezemer (2008) and Sapkota (2011). Although the method involves subjective judgements with clear specifications, it is considered sufficiently sound to provide reliable and transparent judgments to evaluate policies that determine paradigm alignment in the PRSP, by condensing information in a practical way that allows for clear interpretation (Bojo & Reddy, 2002). The scorecard system is provided in Table 2.

Table 2. Table 1: Poverty Reduction Strategy Paper scorecard system
ScoreCriteria
0No mention of the policy
1Policy is mentioned
2Policy is elaborated upon
3Policy is extensively elaborated upon

In the next stage, all individual PRSP social protection scores are summed and normalised to create an overall index value. The index is normalised by taking highest score given to a PRSP and dividing this by the individual country's score, creating a maximum value.

  • display math

A full list of the countries and their social protection alignment scores is provided in Table A1 of the Appendix.

This model is a cross-country two-way tabulation, where the high and low scores of PRSP SPI and the indices for the three components of social protection are tested against possible common explanations for these scores. The growth and development literature suggests several key explanatory factors are present in countries able to change their human and economic circumstances. The following economic indicators were seen as possible explanations for social protection policy alignment: the impact of official development aid as a percentage of GDP; GDP per capita (by PPP), ethnic diversity; the (summed) World Bank governance indicators; and regional factors. These cross tabulations highlight prime facie evidence of common characteristics between the indices and can be used to identify any further associations that need to be investigated. Further econometric estimations are investigated using ordinary least squares regression estimations, and the results are presented in the APPENDIX B.

In an attempt to show causal links between the continuous explanatory variables and the SPI, the study uses data lagged for one year prior to the PRSP implementation. GDP per capita, ODA and the summed World Bank governance index data from the previous year. These variables are, therefore, predetermined. Regional dummies and ethnicity are assumed to be exogenous variables.8

5 RESULTS AND DISCUSSION

  1. Top of page
  2. Abstract
  3. 1 INTRODUCTION
  4. 2 DEFINING SOCIAL PROTECTION, VULNERABILITY AND RISK
  5. 3 LITERATURE REVIEW
  6. 4 DATA AND METHODOLOGY
  7. 5 RESULTS AND DISCUSSION
  8. 6 CONCLUSION AND POLICY RECOMMENDATIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. APPENDIX A
  12. APPENDIX B
  13. APPENDIX C

This section presents the results of overall programme rankings in the study and cross tabulations and discusses the implication of these results.

The results From figure 1 that investigated overall SPI indicate that programmes for the most vulnerable and public works programmes constitute the most implemented policy in PRSPs. The least adopted policies in this sample are the labour standards. The International Labour Organization is advocating a minimum standards security floor across all nations. In an uncertain world, employment conditions and rights are the hardest to guarantee (ILO, 2011). Social assistance and insurance are necessary government mechanisms to compensate for a globally insecure job market. The least adopted social insurance policies were maternity payments and industrial injury payments. Given that these policies are more closely aligned with developed countries and, even then, only with countries with a strong welfare state, these results are not unexpected.

image

Figure 1. Overall Social Protection in Poverty Reduction Strategy Papers in 2000–2011

Download figure to PowerPoint

5.1 Factors Explaining the Variation in Social Protection Alignment Index Scores

This paper provides the first examination of country characteristics that explain the variation in PRSP social protection alignment scores. Cross-tabulations are used to examine whether there are statistically significant relationships between high and low index scores and possible drivers influencing social protection adoption. These drivers include income levels, dependence on foreign aid, the level of ethnic diversity, the level of governance and regional influences. Further econometric estimations are also conducted to test these possible associations.

5.2 Income

Table 3 presents results from cross-tabulations of social protection alignment index scores with the GDP per capita lagged for one year as a proxy for income .9 High index scores are defined as those above the mean. Low index scores are defined as those below the mean. This study reveals that developing countries with higher adoption scores are more likely to be associated with higher per capita wealth levels. The Pearson chi-squared statistic indicates that income differences are significant for the overall SPI and for each of the subdivisions of the SPIs. These results, although intuitive, are also reflected in the social protection expenditures of regions with comparatively higher incomes per capita, where Europe and Central Asia have 10 per cent of GDP dedicated to social assistance and social insurance compared to 4.5 per cent of GDP in Sub-Saharan Africa.10

Table 3. Cross tabulation of Social Protection Alignment scores and lagged one year gross domestic product per capita income
 High GDP income > $1890 USlow GDP income > $1890 USNumber
  1. GDP, gross domestic product; SPI, Social Protection Index.

Social Protection Alignment Index score (%)
Low SPI score < 0.38863.3328.0735
High SPI score > 0.38836.6771.9352
Total100.00100.0087
Pearson χ2(1)10.1645  
(p-value)(0.001)  
Social Insurance Alignment Index score (%)
Low ins score > 0.2566.6721.0532
High ins score < 0.2533.3378.9555
Total100.00100.0087
Pearson χ2(1)17.5873  
(p-value)0.000  
Social Assistance Alignment Index score (%)
Low ass score < 0.39170.0042.1145
High ass score > 0.39130.0057.8942
Total100.00100.0087
Pearson χ2(1)6.1249  
(p-value)0.013  
Labour Standards Alignment Index score (%)
Low labour score < 0.39173.3333.3341
High labour score > 0.39126.6766.6746
Total100.00100.0087
Pearson χ2(1)12.6210  
(p-value)0.000  

Results from APPENDIX B also reveal that lagged GDP per capita is positive and statistically significant across all of the indices at a 5 per cent confidence interval level. However, as GDP per capita and regions are strongly correlated the relationship between GDP per capita and SPI are not significant when regions are included in the regression results.

5.2.1 Official development assistance

Cross-tabulations of social protection alignment scores and one year lagged Official Development Assistance11 are found in Table 4. Official Development Assistance (ODA) has shown to be more effective in countries with political stability, democracy, climatic and external conditions12 (Burnside & Dollar, 2000). High scores are associated with low ODAs, signifying that countries receiving below the mean of 11.3 as a percentage of GDP annually have PRSPs more inclined towards a social protection agenda.13 Cross-tabulations reveal a positive and statistically significant association between ODA and social assistance programmes. Regression results in APPENDIX B confirm these associations with social assistance. Barrientos (2011) found social assistance programmes have been proven more effective at decreasing poverty and increasing human development in developing nations than labour standards and social insurance policies. Further research by Barrientos (2011) found that most of the social transfers occur in low-income countries supported by external donors. High labour standard index scores, on the other hand, were negatively associated with a low percentage of ODA to GDP and low labour standards index being associated with a high percentage of ODA to GDP, which implies labour standards may be more difficult to implement.

Table 4. Cross tabulation of Paradigm Alignment Index Scores with Official Development Assistance
 Low ratio of GDP to ODA < 11.5High ratio of GDP to ODA >11.5Number
  1. GDP, gross domestic product; ODA, Overseas Development Assistance; SPI, Social Protection Index.

Social Protection Alignment Index score (%)
Low SPI score < 0.38834.0946.5135
High SPI score > 0.38865.9153.4942
Total100.00100.0087
Pearson χ2(1)1.3953  
(p-value)0.238  
Social Insurance Alignment Index score (%)
Low ins score > 0.2529.5544.1932
High ins score < 0.2570.4555.8155
Total100.00100.0087
Pearson χ2(1)2.4647  
(p-value)(0.116)  
Social Assistance Alignment Index score (%)
Low ass score < 0.39161.3641.8645
High ass score > 0.39138.6458.1442
Total100.00100.0087
Pearson χ2(1)3.0313  
(p-value)0.069  
Labour Standards Alignment Index score (%)
Low labour score < 0.39139.0258.1441
High labour score > 0.39163.6441.8646
Total100.00100.0087
Pearson χ2(1)4.1386  
(p-value)0.042  

5.3 Ethnic Diversity

Table 5 presents the results of the cross-tabulations between ethnic diversity and the social protection adoption indicators. A statistically significant negative relationship exists between a high index score for social assistance and a low level of ethnic diversity.14 Easterly and Levine (1997) found that homogeneous populations are more likely to have political stability and higher public investment than the ethnically diverse countries. A low SPI and an ethnically diverse population could increase the likelihood of vulnerability for ethnic minority groups, because of influxes of displaced persons, which can be associated with greater resource repercussions. These groups would benefit from higher levels of social assistance. In this study, programmes for the displaced and refugees were categorised under social assistance programmes, which was negatively significant at a 90 per cent confidence interval, as was the overall SPI. Again, this would indicate that populations with high ethnic diversity are not the beneficiaries of the same level of protection as their more homogeneous counterparts.

Table 5. Cross tabulation of Paradigm Alignment Index scores with summed World Bank governance indicators
 Low WB governance indicator < −3.17High WB governance indicator > −3.17Number
  1. SPI, Social Protection Index; WB, World Bank.

Social Protection Alignment Index Score (%)
Low SPI score < 0.38839.5340.9135
High SPI score > 0.38861.4759.0952
Total100.00100.0087
Pearson χ2(1)0.0171  
(p-value)0.896  
Social Insurance Alignment Index Score (%)
Low ins score > 0.2537.2136.3645
High ins score < 0.2562.7963.6442
Total100.00100.0087
Pearson χ2(1)0.0603  
(p-value)0.806  
Social Assistance Alignment Index Score (%)
Low ass score < 0.36162.7940.9145
High ass score > 0.36137.2159.0942
Total100.00100.0087
Pearson χ2(1)4.170  
(p-value)0.041  
Labour Standards Alignment Index Score (%)
Low labour score < 0.3046.5147.7341
High labour score > 0.3053.4952.2746
Total100.00100.0087
Pearson χ2(1)0.0129  
(p-value)0.910  
5.3.1 World bank governance indicators

Results from Table 6 analyse cross-tabulations with the social protection adoption indicators and one year lagged World Bank summed governance indicator.15 There is a positive statistical significance for social assistance and World Bank summed governance indicators at the 95 per cent confidence interval. Social assistance packages are the primary focus of an overall social protection policy in developing nations, and as these packages are able to ensure minimum standards of consumption to address the worst of effects of deprivation (Farrington & Slater, 2006; Barrientos & Hulme, 2009; Barrientos, 2011; Gentilini & Omama, 2011). Positive associations with higher governance scores should translate into better implementation of these programmes in developing nations as the summed index includes measures for government effectiveness, political stability and control of corruption. Regression estimations in APPENDIX B, confirm these results, the study finds a positive and significant association between the social assistance index and the summed World Bank governances scores at the 90 per cent confidence interval.

Table 6. Cross tabulation of Paradigm Alignment Index scores with ethnic diversity
 Low ethnic diversity > 0.692High ethnic diversity < 0.692Number
  1. SPI, Social Protection Index.

Social Protection Alignment Index Score (%)
Low SPI score < 0.38832.0852.9435
High SPI score < 0.38867.9247.0652
Total100.00100.0087
Pearson χ2(1)3.7503  
(p-value)(0.053)  
Social Insurance Alignment Index score (%)
Low ins score > 0.2530.1947.0632
High ins score < 0.2569.8152.9455
Total100.00100.0087
Pearson χ2(1)2.5351  
(p-value)(0.111)  
Social Assistance Alignment Index score (%)
Low ass score < 0.36141.5166.6745
High ass score > 0.36158.4932.3542
Total100.00100.0087
Pearson χ2(1)5.6669  
(p-value)0.017  
Labour Standards Alignment Index score (%)
Low labour score < 0.3039.6258.8241
High labour score > 0.3060.3841.1846
Total100.00100.0087
Pearson χ2(1)3.0646  
(p-value)0.080  
5.3.2 Regional analysis

Table 7 reviews the regional variation of social protection adoption scores. The regions are categorised according to the World Bank's regional classification. Results indicate that the social protection within PRSPs reveals regional differences. Three of the four indices have statistically significant relationships at the 99 per cent confidence interval, the exception being labour standards. The SPI index reveals a statistically strong positive relationship between Europe and Central Asia (commonly seen as the transition economies), South Asia and East Asia, and the Pacific regions. This could also be attributed to the strong correlation between income per capita and regions. Lower scores were more prevalent in Sub-Saharan Africa than any other region. Social insurance was found to be more prevalent in Europe and Central Asia, and South Asia. Regression results in APPENDIX B find a positive association between Europe and Central Asia, East Asia and Sub-Saharan Africa.

Table 7. Cross tabulation of Paradigm Alignment Index scores with region
 Sub-Saharan (n = 49)Europe and Central Asia (n = 13)Latin American and Caribbean (n = 8)South Asia (n = 7)East Asia and Pacific (n = 8)Middle and North Africa (n = 2)
  1. SPI, Social Protection Index.

Social Protection Overall Alignment index score (%)
High SPI score > 0.33822.4592.3150.0042.8650.0050.0
Low SPI score < 0.33877.557.6950.0057.1450.0050.0
Pearson χ2 (5) (p-value)21.8402 (0.001)     
Social Insurance Alignment Index score (%)
High ins score > 0.2514.2910050.057.1462.500.000
Low ins score < 0.2585.710.0050.042.8637.50100.00
Pearson χ2 (5) (p-value)37.6971 (0.000)     
Social Assistance Alignment Index score (%)
High ass core > 0.36136.7376.92100.0057.1450.0050.0
Low ass score < 0.36163.2723.080.0042.8650.0050.0
Pearson χ2 (5) (p-value)15.278 (0.009)     
Labour Standards Alignment Index score (%)
High ins score > 0.28634.6976.9262.5057.1450.0050.00
Low ins score < 0.28665.3123.0837.5042.8650.0050.00
Pearson χ2(1) (p-value)8.7455     
0.120

Countries with higher social insurance index alignment scores had a higher GDP per capita, lower levels of ethnic diversity, and with distinct regional differences, World Bank governance indicators were found to be not significant for social insurance. Traditionally, social insurance protects the upper quintiles of the population, and these results suggest that the most vulnerable do not receive the benefits of such programmes.16 However, the role of social insurance has increasingly had been found to have taken a lesser role in the overall social protection framework in developing nations because of the impacts of market liberalisation (Barrientos, 2011)17

Social assistance programmes are more likely to be associated with development frameworks in low-income countries because of their capacity to reduce poverty and vulnerability, more so than social insurance or labour programmes (Barrientos, 2011). Intriguingly, results from this study reveal a strong relationship between the common determinants of development and PRSP recipients aligning to a social protection agenda. Higher social assistance scores were found in nations with higher GDP per capita, significantly higher levels of ODA, higher governance scores and higher levels of ethnic diversity. They also demonstrated strong regional influences for Latin America and the Caribbean, Europe and Central Asia, East Asia and even Sub-Saharan Africa.

Labour standards reflect a transformational concept of social protection, with a greater emphasis on labour legislation, and labour opportunities. Countries with a higher alignment to labour standards had higher GDP per capita, lower levels of aid and lower levels of ethnic diversity. Labour standards were found to positively associated with Europe and Central Asia, Latin America and South Asia. The results for labour standards must be taken with caution because of the limited number of policies investigated, only four labour standards were investigated in this study of PRSPs.

Comparisons between the ADB SPI and the PRSP SPI revealed a consistency in ranking between countries at the bottom of the sample, but inconsistencies with the upper rankings. The ADB SPI (which is comprise social protection expenditure, social protection coverage, poverty targeting and social protection impacts) provides an analysis of 25 Asian nations' social protection programmes. There are 14 countries common to both studies. The PRSP SPI reviewed policy content exclusively, and, therefore, any variation in rankings could be explained by the failure of some countries to deliver what was outlined in the PRSPs. APPENDIX C outlines the ranking results and finds that Mongolia ranks as highest for both ADB SPI and PRSP SPI, whereas Cambodia and Lao scored in the bottom two for both indices.

Bangladesh presented as an anomaly in the rankings: the PRSP SPI ranked Bangladesh as equal first for the 15 countries examined, and in the ADB SPI, it ranked in equal eighth position. Disaggregating the components of both SPI's reveals one of the highest social assistance scores for the PRSP index, but the ADB's SPI was ranked sixth out of the 15 countries. The failure of Bangladesh to deliver the breadth and depth of policies promised may be driving the disparity between the two indices. Even though programmes were in place in Bangladesh, there may be a gap between the programmes' coverage, expenditure and the intended beneficiaries.

6 CONCLUSION AND POLICY RECOMMENDATIONS

  1. Top of page
  2. Abstract
  3. 1 INTRODUCTION
  4. 2 DEFINING SOCIAL PROTECTION, VULNERABILITY AND RISK
  5. 3 LITERATURE REVIEW
  6. 4 DATA AND METHODOLOGY
  7. 5 RESULTS AND DISCUSSION
  8. 6 CONCLUSION AND POLICY RECOMMENDATIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. APPENDIX A
  12. APPENDIX B
  13. APPENDIX C

Social protection makes an important contribution to the household's capacity to embolden themselves against future shocks and overcome present shocks. Embedded in the developed world, increasingly social protection's relevance is being felt and seen in lower-income countries. The role of PRSPs is to provide governments' national voice of self-determined policies with the backing of the International Financial Institutions, and guided by the overarching goals of the MDGs.

This study finds with regard to the overall SPI within the PRSPs are influenced by the region from which they come from and the level of GDP per capita. Highlighting this relationship, Europe and Central Asia have 63.1 per cent of social protection coverage for their population, compared with 24.4 per cent coverage in Sub-Saharan Africa.18 Labour policies were not as influenced by regions as the other two social protection components. This may be because of policies such as child labour being included in the labour standards framework that are not as critical in the lower-middle income countries as they are in the low-income countries.

Disaggregating the SPI into its component is important. This study reviewed three component parts of the SPI: social insurance; social assistance and labour standards. PRSPs aligning to a social assistance agenda are best explained by the possible determinants of development. High social assistance index were found to have a positive association with higher levels of ODA, higher governance scores, more homogeneous ethnic populations and were higher in the regions of Europe and Central Asia and South Asia than the other two components of social protection. This result infers that whereas social insurance policies may be beyond the administrative capacity of some developing nations, the temporary nature of social assistance policies may allow for greater scope for implementation in the PRSPs. Alignment to a social assistance agenda to higher levels of governance and a higher percentage of ODA to GDP is promising on a number of fronts. Firstly, it means that governments are more influenced by donors to follow directives towards implementing social assistance. Secondly, these packages are being implemented in countries with greater measures of government effectiveness, political stability and regulatory quality, which should reflect positively on the countries' capacity to implement the programmes. However, the PRSPs only represent the plans to implement policies and the ability to follow through on these plans is contingent on the motivation of donors and the national governments' political will and administrative capacity.

Encouragingly, the inclusion of decent employment and work for all in the updated version of the MDGs in 2007, demonstrates a commitment from the United Nations to address the glaring gap of labour standards between developed and developing countries. As labour standards in the PRSPs were found to be the least implemented of the social protection policies and were also found to be negatively associated with ODA flows, this component of social protection is the one area that offers the greatest opportunity to ensure certainty for the most vulnerable.

Social protection has gathered momentum as a critical component of the development framework as it offers the same ladder of opportunity for the vulnerable in the developing world that has long been a part of the developed world's toolkit. PRSPs alongside MDGs play an important role in outlining the future direction of development policies and programmes in the developing world. Reducing inequality and vulnerability have been identified as important elements that are missing from the current set of MDGs, including new MDG targets to increase social protection and reduce income inequality appear obvious additions to a post-2015 development framework that will go some way to create security for the most vulnerable.

ACKNOWLEDGEMENTS

  1. Top of page
  2. Abstract
  3. 1 INTRODUCTION
  4. 2 DEFINING SOCIAL PROTECTION, VULNERABILITY AND RISK
  5. 3 LITERATURE REVIEW
  6. 4 DATA AND METHODOLOGY
  7. 5 RESULTS AND DISCUSSION
  8. 6 CONCLUSION AND POLICY RECOMMENDATIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. APPENDIX A
  12. APPENDIX B
  13. APPENDIX C

The author would like to acknowledge the very helpful comments of the two anonymous referees and additional comments from Simon Feeny, Bob Baulch and John King. The responsibility for the final version of this paper is of course my own. The usual disclaimer applies.

  1. 1

    According to International Disaster Database, the annual number of natural disasters increased from 181 from 1981–1990 to 438 from 2000 to 2011

  2. 2

    World commodity markets show an increase in the energy price index from $37.70 during 1990–2000 to $97.12 during 2000–2010 (constant 2005 US dollars), and the food price index increased from $97.13 1990–2000 to $118.57 2000–2010 (constant 2005 US dollars).

  3. 3

    Hulme (2009) accredits the Development Assistance Committee and for Economic Cooperation and Development for an initial list of development goals, these emerged out of a series of key world summits in the 1990s. For a full chronological account of the evolution of the MDGs, see Hulme (2009).

  4. 4

    Measured by PovcalNet data set (2012), World Bank research team.

  5. 5

    Uganda undertook the first PRSP in 1999.

  6. 6

    As PRSPs are in their relative infancy, with the first PRSP in 1999, the majority of PRSPs were not yet fully implemented.

  7. 7

    As of February 2013, there have been 101 PRSPs; the study uses only data up until 2011. The additional recipients are all 2nd or 3rd generation PRSPs.

  8. 8

    Much of the ethnic distribution was a result of 20th century migration and therefore not subject to significant changes.

  9. 9

    Countries are defined as ‘low-income’ if their GDP per capita falls below US$1,890 (Purchasing Power Parity 2005), and as middle- and high-income countries if it rises above this threshold.

  10. 10

    These results are however, are not reflective of all developing economies within this study, there are only 22 countries common between the two studies. The true results would possibly be lower for the countries within this study.

  11. 11

    Countries are defined as low aid recipients if their ratio of ODA to GDP is assessed at 11.5, and defined as high aid recipients if aid is above ODA to GDP share above 11.5.

  12. 12

    According to Sachs (1997), climate and geography are significant explanatory variables that account for a country's level of development.

  13. 13

    In the past conditionality was seen as a failure of the previous generation of World Bank's policy papers, due to a lack of ownership from national governments Killick (2002) ‘The “Streamlining” of conditionality: Aspirations, Reality and Repercussions.’.

  14. 14

    Ethnic diversity is measured in terms of linguistic diversity in the Ethnologue Lewis (2009). Ethnologue: Languages of the World,. M. P. Lewis. Dallas, Tex, SIL International. Sixteenth edition.

  15. 15

    The World Bank summed Kaufman's governance indicators (regulatory quality; government effectiveness; control of corruption; political stability, rule of law and voice and accountability) to create an overall governance indicator.

  16. 16

    The data available from ASPIRE, a World Bank database covering social protection from developing countries over the period 2005 to 2010, 25 of these countries are common to our study from shows that 93.36 per cent of all social insurance benefits are allocated to the non-poor.

  17. 17

    Social protection data reveals that 1.5 per cent of GDP is spent on social insurance in Sub-Saharan Africa compared with 8 per cent of GDP in Europe and Central Asia, whereas 3 per cent of GDP in Sub-Saharan Africa is spent on social assistance programmes compared with 2 per cent of GDP in Europe (ADePT World Bank)

  18. 18

    According to data available from the ASPIRE database—World Bank SP Atlas of Social Protection. This data covers 56 countries in 2013, 17 countries from low-income countries, 20 from middle-income countries and 15 from upper middle-income countries over the period 2005–2010.

REFERENCES

  1. Top of page
  2. Abstract
  3. 1 INTRODUCTION
  4. 2 DEFINING SOCIAL PROTECTION, VULNERABILITY AND RISK
  5. 3 LITERATURE REVIEW
  6. 4 DATA AND METHODOLOGY
  7. 5 RESULTS AND DISCUSSION
  8. 6 CONCLUSION AND POLICY RECOMMENDATIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. APPENDIX A
  12. APPENDIX B
  13. APPENDIX C
  • Alwang J, Siegel PB, Jorgensen SL. 2001. Vulnerability: A View From Different Disciplines. Social Protection Discussion Paper Series.
  • Barrientos A, Hulme D. 2009. Social protection for the poor and poorest in developing countries: reflections on a quiet revolution. Oxford Development Studies 37(4): 439456.
  • Barrientos A. 2011. Social protection and poverty. International Journal of Social Welfare 20: 240249.
  • Baulch B, Wood J. 2008. Social Protection Index for Committed Poverty Reduction. Asian Development Bank: Manila.
  • Baulch B, Wood J, Weber A. 2006. Developing a social protection index for Asia. Development Policy Review 24(1): 529.
  • Bojo J, Reddy RC. 2002 Status and Evolution of Environmental Priorities in the Poverty Reduction Strategies. Environmental Economics Series Paper No. 93: Washington DC.
  • Burnside C, Dollar D. 2000. Aid, policies and growth. American Economic Review 94(3): 774780.
  • De Brauw A, Hoddinott J. 2011. Must conditional cash transfer programs be conditioned to be effective? the impact of conditioning transfers on school enrollment in Mexico. Journal of Development Economics 96(2): 359370.
  • Deveraux S, Sabates-Wheeler R. 2004. Transformative Social Protection. IDS Working Paper 232.
  • DFID. 2004. Policy paper on social protection. DFID paper on Social Protection. Shepard A, Marcus R, Barrientos A. London, Overseas Development Institute.
  • DFID. 2005. Social transfers and chronic poverty: Emerging evidence and the challenge ahead.
  • DFID. 2011. Cash transfer. evidence paper. London, UK Department of International Development, Policy Division.
  • Easterly W, & Levine, R (1997). “Africa's Growth Tragedy: Policies and Ethnic Divisions.” Quarterly Journal of Economics 112(4).
  • Eggen AR, Bezemer DJ. 2008. ‘Do poverty reduction strategy papers help achieve the Millennium Development Goals?’ “Cooperation of International Institutions for Related Global Issues” 7030.
  • Farrington J, Slater R. 2006. Introduction: cash transfers: panacea for poverty reduction or money down the drain? Development Policy Review 24(5): 499624.
  • Fukuda-Parr S. 2010. Reducing Inequality—the missing MDG: a content review of PRSPs and bilateral donor statements. IDS Bulletin 41(1): 2635
  • Gaiha R, Imai K. 2008. Measuring vulnerability and poverty: estimates for rural India. WIDER Research Paper No. 2008/40 40.
  • Gentilini U. 2009. Social protection in the “real world”: issues, models and challenges. Development Policy Review 27(2): 147166.
  • Gentilini U, Omama, SW. 2009. Unveiling Social Safety Nets. World Food Program: Rome, WFP. 20.
  • Gentilini U, Omama SW. 2011. Social protection 2.0: exploring the issues, evidence and debates in a globalizing world. Food Policy 36: 320340.
  • Guillaumont PCL. 2001. Aid and performance: a reassessment. Journal of Development Studies 37(6): 6687.
  • Holzmann RJS. 2000. Social risk management: a new framework for social protection and beyond. Social Protection Discussion Paper Series.
  • Hulme DS. 2010. The Political Economy of the MDGs: Retrospect and Prospect for the World's Biggest Promise. Brooks World Poverty Institute.
  • ILO. 2011. Social Protection Floor For Fsssair and Inclusive Globatisation. International Labour Organisation: Geneva.
  • Imai KS, Giaha R, Kang W. 2010. Vulnerability and poverty dynamics in Vietnam. Applied Economics 1(16): 36063618.
  • Killick T. 2002The “Streamlining” of conditionality: Aspirations, Reality and Repercussions.”.
  • Lewis MP. 2009. Ethnologue: languages of the world,. M. P. Lewis. Dallas, Tex, SIL International. Sixteenth edition.
  • Marcus R, Wilkinson J. 2001. Whose poverty matters? vulnerability, social protection and PRSPs. Working Paper. London, Child Poverty Research and Policy Centre. 1.
  • Mares I. 2005. Social protection around the world: external insecurity, state capacity, and domestic political cleavages. Comparative Political Studies 38(6): 623651.
  • OECD. 2009. “Promoting Pro-poor growth: social protection.”
  • Sabates-Wheeler R, Deveraux S. 2007. Social protection for transformation. IDS Bulletin 38(3): 2328.
  • Sapkota BJ 2011. Mainstreaming globalisation in poverty reduction strategy papers in the Asia-Pacific region. Development in Practice 21(7): 9991012.
  • Tarabini A, Jacokvis J. 2012. The poverty reduction strategy papers: an analysis of a hegomonic link between education and poverty. The Journal of Educational Development 32: 506516.
  • Ulriksen MS. 2012. Questioning the Pro-poor agenda: examining the links between social protection poverty. Development Policy Review 30(3): 261281.
  • Vandemoortele J. 2011. The MDG story: intention denied. Development and Change 1(42): 121.
  • World Bank. 2012. “Resilience, Equity and Opportunity.” The World Bank's Social Protection and Labour Strategy 2012–2022.

APPENDIX A

  1. Top of page
  2. Abstract
  3. 1 INTRODUCTION
  4. 2 DEFINING SOCIAL PROTECTION, VULNERABILITY AND RISK
  5. 3 LITERATURE REVIEW
  6. 4 DATA AND METHODOLOGY
  7. 5 RESULTS AND DISCUSSION
  8. 6 CONCLUSION AND POLICY RECOMMENDATIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. APPENDIX A
  12. APPENDIX B
  13. APPENDIX C

Country evaluation from Poverty Reduction Strategy Paper documents

CountryYearOverall SPISocial insuranceSocial assistanceLabour standards
  1. SPI, Social Protection Index.

Afghanistan20080.580.360.870.14
Albania20010.730.600.830.43
Albania20080.630.600.600.29
Armenia20030.750.720.770.36
Armenia20080.670.800.430.64
Azerbaijan20030.810.880.730.57
Bangladesh20050.780.441.000.50
Benin20020.190.080.230.21
Benin20080.270.120.330.14
Bhutan20040.090.000.200.00
Bolivia20010.440.160.530.43
Bosnia-Herzegovina20040.880.920.570.86
Burkina Faso20000.170.000.270.21
Burkina Faso20040.530.040.770.50
Burundi20060.340.080.600.00
Cambodia20020.310.080.430.29
Cambodia20050.250.080.330.29
Cameroon20030.250.120.370.00
Cameroon20100.230.120.130.43
Cape Verde20040.470.320.570.36
Cape Verde20080.380.280.400.36
Chad20030.160.080.230.07
Congo DR20070.110.000.170.14
Cote D'Ivorie20090.270.080.230.36
Djibouti20040.390.080.530.43
Dominica20060.300.040.430.36
Ethiopia20020.190.040.370.00
Gambia20020.140.040.270.00
Gambia20070.110.000.230.00
Georgia20030.590.360.630.64
Ghana20030.220.120.300.14
Ghana20050.420.120.570.29
Guinea20020.230.000.330.21
Guinea20070.090.000.200.00
Guinea-Bissau20110.130.040.170.14
Guyana20020.270.160.370.14
Haiti20080.130.000.270.00
Honduras20010.470.280.570.43
Kenya20040.470.160.400.86
Kenya20100.390.160.330.50
Kyrgyz Republic20020.670.560.670.50
Lao20040.310.280.370.00
Lao20080.420.400.300.50
Lesotho20050.170.080.270.07
Liberia20080.330.000.470.50
Madagascar20030.190.000.300.21
Madagascar20070.340.000.570.14
Malawi20020.470.120.730.29
Malawi20060.330.120.570.07
Maldives20080.360.200.470.14
Mauritania20000.090.000.200.00
Mauritania20060.360.080.470.43
Moldova20040.921.000.730.64
Moldova20080.280.480.130.00
Mongolia20030.780.920.570.50
Mozambique20010.250.040.500.00
Mozambique20060.340.160.470.21
Mozambique20110.200.000.270.36
Nepal20030.330.120.330.50
Nicaragua20010.380.240.570.07
Nicaragua20050.670.240.800.64
Nicaragua20100.630.240.530.64
Niger20080.520.720.370.21
Nigeria20050.280.200.300.21
Pakistan20030.330.160.400.36
Rwanda20020.160.040.300.00
Rwanda20080.270.160.270.29
Sao Tome & Principe20020.110.040.170.00
Senegal20020.390.080.570.36
Senegal20070.720.440.800.57
Serbia Montenegro20041.000.840.871.00
Sierra Leone20050.130.000.270.00
Sri Lanka20020.690.360.800.57
Tajikistan20020.450.280.570.29
Tanzania20000.110.000.200.07
Tanzania20050.380.160.430.50
Tanzania20100.310.160.330.43
Timor Leste20020.310.120.430.29
Uganda20000.110.080.170.00
Uganda20050.310.120.400.36
Uganda20100.500.320.500.57
Uzbekistan20080.610.720.500.36
Vietnam20030.690.640.700.36
Vietnam20060.670.560.600.57
Yemen20020.270.240.370.00
Zambia20020.220.160.300.07
Zambias20060.700.360.930.57

APPENDIX B

  1. Top of page
  2. Abstract
  3. 1 INTRODUCTION
  4. 2 DEFINING SOCIAL PROTECTION, VULNERABILITY AND RISK
  5. 3 LITERATURE REVIEW
  6. 4 DATA AND METHODOLOGY
  7. 5 RESULTS AND DISCUSSION
  8. 6 CONCLUSION AND POLICY RECOMMENDATIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. APPENDIX A
  12. APPENDIX B
  13. APPENDIX C

Social protection 2000–2011 Overseas Development Assistance

 (1)(2)(3)(4)(5)(6)(7)(8)
VariablesSPISPIInsuranceInsuranceAssistanceAssistanceLabourLabour
  • SPI, Social Protection Index.

  • Robust t-statistics in parentheses

  • ***

    p < 0.01,

  • **

    p < 0.05,

  • *

    p < 0.1

  • Latin America and the Caribbean are used as the reference region.

gdp2005_10.059**0.0050.067**0.0100.030*−0.0120.057**0.023
 (2.59)(0.26)(2.41)(0.51)(1.75)(−0.64)(2.37)(0.85)
Ethnic−0.083−0.034−0.085−0.056−0.128*−0.0740.0350.091
 (−1.25)(−0.56)(−1.19)(−1.01)(−1.90)(−1.02)(0.45)(1.16)
govern_1−0.0070.011−0.0150.0050.0010.014*0.0010.012
 (−0.76)(1.34)(−1.48)(0.67)(0.15)(1.72)(0.14)(1.26)
oda_10.0000.000−0.001−0.0010.002*0.002**0.0010.001
 (0.47)(0.55)(−1.49)(−1.30)(1.80)(2.07)(0.46)(0.34)
South Asia 0.104 0.258** 0.046 0.044
  (1.08) (2.63) (0.66) (0.39)
Europe and Central Asia 0.320*** 0.457*** 0.243*** 0.145
  (3.93) (5.42) (3.01) (1.34)
East Asia and South Pacific −0.013 −0.029 0.145** −0.095
  (−0.17) (−0.44) (2.07) (−0.54)
Northern Africa and Middle East 0.094 0.037 0.049 −0.018
  (0.82) (0.79) (0.89) (−0.13)
Sub-Saharan Africa −0.070 0.107 0.240* −0.097
  (−0.92) (1.30) (1.85) (−0.90)
Constant0.295***0.405***0.187**0.246***0.375***0.392***0.167**0.270*
 (3.85)(4.20)(2.11)(3.89)(4.75)(5.33)(2.09)(1.92)
Observations8787878787878787
R-squared0.2050.4490.2270.5770.1210.2430.1390.223

APPENDIX C

  1. Top of page
  2. Abstract
  3. 1 INTRODUCTION
  4. 2 DEFINING SOCIAL PROTECTION, VULNERABILITY AND RISK
  5. 3 LITERATURE REVIEW
  6. 4 DATA AND METHODOLOGY
  7. 5 RESULTS AND DISCUSSION
  8. 6 CONCLUSION AND POLICY RECOMMENDATIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. APPENDIX A
  12. APPENDIX B
  13. APPENDIX C

Comparison table between the Asian Development Bank Social Protection Index (SPI) and the Poverty Reduction Strategy Paper SPI

ADB SPI rankCountry nameADB indexSPI PRSP rankCountry namePRSP index
  • ADB, Asian Development Bank; SPI, Social Protection Index; PRSP, Poverty Reduction Strategy Paper.

  • *

    Average figure for two PRSPs.

1Mongolia0.2851Mongolia0.8
2Uzbekistan0.2351Bangladesh0.8
3Kyrgyz Republic0.2113Armenia0.74*
4Armenia0.1454Kyrgyz Republic0.72
5Sri Lanka0.1125Sri Lanka0.7
6Viet Nam0.0956Viet Nam0.69*
7Nepal0.0487Uzbekistan0.62
8Bangladesh0.0468Tajikistan0.48
8Pakistan0.0469Maldives0.38
10Maldives0.04310Lao0.37*
11Tajikistan0.02111Pakistan0.36
12Lao0.01912Nepal0.33
14Bhutan0.01713Cambodia0.3*
14Cambodia0.01714Bhutan0.11