Vulnerable populations in developing countries are highly susceptible to economic shocks. Social safety nets (SSNs), encompassing cash or in-kind transfers, aim to mitigate poverty and protect households from such shocks. The study focuses on evaluating the comparative effectiveness of public (government-provided) and private (NGO-provided) SSNs in rural Pakistan. This is crucial because existing evidence-based assessments on the relative effectiveness of these two SSN types in rural Pakistan are lacking. The 2008 food price hike and the subsequent launch of the Benazir Income Support Program (BISP) highlighted the need for such an analysis. While programs like Zakat and BISP have been studied, other SSNs, and especially a comparison of public and private provision, remain under-researched. The study uses the first round of the Pakistan Rural Panel Household Survey (PRPHS) 2011-12, collected after nationwide floods, providing a valuable dataset for analyzing the impact of SSNs on shock coping mechanisms. The primary objective is to determine the impact of both public and private SSNs on shock coping strategies and to compare their relative effectiveness.

The literature emphasizes the importance of resilience – the ability to absorb shocks without resorting to detrimental coping strategies – as a crucial aspect of living standards. Social safety nets are seen as mechanisms to redistribute wealth and ensure resilience, especially in developing countries where private markets may fail to provide adequate protection. The Social Protection Floor (SPF) concept, advocating basic human rights, highlights SSNs as a key component. While developed countries have established formal social security systems, developing countries often rely on informal systems or community support. Impact evaluations of SSNs in developing countries show mixed results; well-designed, consistently implemented programs show greater positive impacts. Previous studies in Bangladesh and Ethiopia show mixed findings on the effectiveness of SSNs in reducing shock coping strategies, with some emphasizing the positive effects of multiple SSNs or well-designed and consistently provided SSNs. In Pakistan, studies on Zakat and BISP show mixed results, often showing insignificant positive effects. The present study aims to add to this literature by focusing on a comparison of public and private SSNs and using the PRPHS dataset, which allows for a more comprehensive analysis of the issue.

This study utilizes a quasi-experimental design employing propensity score matching (PSM) to assess the impact of SSNs on shock coping strategies. The study focuses on three detrimental coping mechanisms: reducing food consumption, switching to cheaper food, and distress sale of assets. Households receiving SSNs form the treatment group, while a control group of observationally similar households that did not receive assistance is constructed using PSM. This method addresses the selection bias inherent in non-random program implementation. The PRPHS dataset, comprising 2124 rural households from Punjab, Sindh, and Khyber Pakhtunkhwa, provides data on shocks, coping strategies, and SSNs received. SSNs are categorized as public (government) or private (NGO). Two binary logistic models were estimated, one for each type of SSN, to predict the probability of receiving assistance based on various socioeconomic covariates. Nearest Neighbor Matching (NNM) with replacement was used to match treatment and control units with similar propensity scores. The Average Treatment Effect on the Treated (ATT) was calculated to measure the impact of each SSN type on the three outcome indicators (coping strategies). The analysis ensures no overlap between the public and private SSN groups; households receiving both types are excluded, as are those not reporting shocks or relying on informal safety nets.

The logistic regression models for both public and private SSNs identify several factors influencing the likelihood of receiving assistance. For public SSNs, house ownership, loan attempts, and age of the household head (with an inverted U-shape relationship) significantly increased the probability of receiving assistance, while belonging to Punjab decreased it. For private SSNs, attempting to obtain a loan and household head education increased the likelihood, while having a larger number of large animals decreased it. After propensity score matching, the ATTs for the shock coping strategies were calculated. Overall, the results show insignificant impacts of both public and private SSNs on reducing food consumption and distress asset sales. However, public SSNs displayed a statistically significant negative impact on switching to cheaper food, indicating that households receiving public assistance were significantly less likely to resort to this coping strategy. This positive impact of public SSNs on food security may be attributable to food-based assistance provided in many public programs.

The findings suggest that while neither public nor private SSNs fully protect households from all detrimental coping strategies, public SSNs offer greater protection against compromising food quality. The insignificant effects across other strategies could be due to various factors, including inefficient delivery mechanisms, leakages, and small transfer sizes. While private SSNs, particularly microfinance initiatives, may have longer-term positive impacts on livelihoods not captured in this short-term analysis, the lack of significant positive impact in this short-term analysis is noteworthy. These results highlight the need for improved SSN design and implementation in Pakistan. The relatively stronger performance of public SSNs in reducing reliance on cheaper food suggests that in-kind transfers may be more effective than cash transfers in ensuring food security.

This study offers a comprehensive comparative analysis of public and private SSNs in rural Pakistan. Public SSNs exhibit stronger effects in protecting food quality. It highlights the need for improved targeting, program coordination, monitoring, and evaluation of existing SSNs. Future research could explore the longer-term impacts of private SSNs, particularly microfinance, on livelihood diversification and resilience. The findings offer crucial insights for policy makers in designing and implementing more effective SSNs, particularly within the context of recent devastating floods in Pakistan.

The study's limitations include the cross-sectional nature of the data, restricting the analysis to short-term impacts. Reliance on self-reported data on shocks and coping mechanisms introduces potential recall bias. Informal social safety nets were excluded from the analysis due to difficulties in identifying appropriate covariates. The shocks module in the survey asked about shocks in the last five years, introducing potential ambiguity concerning the timing of shocks relative to SSN receipt. The study did not include Balochistan due to security concerns during data collection.