G²LM|LIC

Latest News

Economic Development and the Organisation of Labour: Evidence from the Jobs of the World Project

Labour is the sole endowment of the poor and the main factor of production in all economies. Therefore, understanding whether labour is employed efficiently is key to understanding poverty at the micro level and differences in national income at the macro level. In this working paper, the authors use the Jobs of the World database to discuss evidence to show how the nature of jobs and their allocation vary within countries by wealth and gender and across countries in stages of development.

The key contribution of the paper lies in documenting broad transformations in the organisation of labour- that is the nature of jobs and their allocation. Authors outline the transformation in three stages: 1) Marketisation of work; 2) Replacement of self-employed work by wage work; 3) Increasing variety of available occupations. This alludes to the fact that the nature of jobs changes over the course of development from subsistence work to self-employed work to increasingly specialised wage work, but wealth and gender shape the allocation of these changing jobs in the same way throughout. Furthermore, the emergence of more specialised occupations coincides with occupational segregation of the labour force by gender which results in misallocation of talent. This is clear evidence that social norms play a crucial role in organisation of labour. Participation in market work varies enormously even within countries with very similar levels of income. For example, the interquartile range of the share of women who hold paid jobs is around 20 percentage points within every decile of GDP. The dispersion is highest in the lowest decile of GDP at 25 percentage points and only 10 percentage points in the highest decile. To study the role of norms the authors present a systematic meta-analysis on active labour market intervention for female labour market participation. Read the detailed Working Paper here.

The (Very) Long-Run Impacts of Cash Grants during a Crisis

It is repeatedly argued that the potential effects of giving cash grants to poor people are constrained by the fact that this income group is trapped in poverty, generating little permanent income effect. However, the literature points to mixed results, not only within short- and longer-term evaluations, but also between them. Therefore, the timing of a long-term study and the potential heterogeneity, specifically by gender, may play an important role in the effects of cash grants.

Motivated by the above, this working paper’s researchers set themselves to experimentally assess the long-run impacts of a randomized cash grant in Uganda on three pre-specified outcomes (after 4, 9, and 12 years), including a heterogeneity analysis by gender. By randomly assigning beneficiaries of the Ugandan YOP (Young Opportunities Program) to their treatment and control groups, the investigators find that although the positive income effects recognized after 4 years had vanished after 9 years, they were again found to be positive in this study during the Covid-19 pandemic. Also, considerably stronger positive effects on men should be highlighted, even though these income effects do not translate into higher food security. Read the detailed Working Paper here.

Childcare, Labor Supply, and Business Development: Evidence from Uganda

It is almost common knowledge that access to childcare is and has been fundamental for increasing women’s labor force participation in High-Income countries. However, it is not fully clear if the lessons drawn from such countries can be applied to contexts in low-income ones. Since in such settings evidence of the effects of childcare on women’s labor market outcomes is scarce and that of the effects it has on child development mixed, there is a research gap that needs to be filled.

Motivated by this, researchers of this Working Paper set themselves to experimentally assess this research question in Uganda. Their research design is not only intended to analyse the effect of childcare on to the labor market outcomes of women, but also on the ones of husbands and that of the household in general. Researchers therefore selected Women that had children between the age of 3 and 5 at the time of the experiment and either gave a cash grant, a childcare subsidy or nothing. Consequently and with this setting, investigators find that childcare leads to a 44 per cent increase in household income and that the childcare subsidy improves child development while the cash grant does not. Read the detailed Working Paper here.

A small selection of our Projects

Our newest publications

Impacts of Local Labor Market Information on Search and Employment: Evidence from India

The job search process is complex and rife with information frictions. We partner with an online job portal to provide over 6,000 job seekers in India with customized and timely information about labor market conditions for their preferred city and occupation on the portal. Specifically, we randomly provide information regarding vacancies (demand), searchers (supply), or both (tightness); a control group receives only general job search guidance. We find that treated respondents are 10% more likely to be employed than control respondents six months after treatment. This positive effect is driven by those with less optimistic beliefs about finding new employment at baseline; these individuals adjust their search to find new jobs if unemployed and stay in the same jobs if employed at baseline. Our experiment shows how access to information about local labor market conditions can reduce frictional unemployment.

Analysing gendered impacts of COVID-19 on Job Search Behaviour

COVID-19 has triggered unemployment worldwide; evidence suggests that there will be significant divergence in the recovery between developing and developed countries (ILO 2021). In order to speed up recovery, better data is needed to understand the differential impact of COVID-19. This project builds on a job matching platform, Job Talash, created by researchers at the Centre for Economic Research in Pakistan, which generates data on job postings and applications for over 10,000 jobseekers and 1,200 firms. Using the Job Talash platform, this brief discusses the job search behaviour of job seekers and employers during COVID-19, allowing us to document the impact and recovery from COVID-19.

How Labor Market Tightness and Job Search Activity Changed in the First Year of COVID-19: Evidence from an Indian Job Portal

The COVID-19 pandemic wreaked havoc on the livelihoods of millions of people around the world. In India, the pandemic came at the heels of an already weak labor market with increasing unemployment rates, particularly among the youth. To manage the early stages of the pandemic, the government instituted a severe national lockdown from March to May 2020. To understand labor market impacts in the first year of COVID-19, we use rich administrative data on search and recruitment from a low-wage online job portal. We observe two years—2019 and 2020— of all employer and job seeker activity on the portal. This allows us to compare trends in 2020 with 2019, accounting for seasonality in labor market activity. Consistent with survey-based evidence documented by the State of Working India 2021, the observed labor market activity on the portal suggests an incomplete recovery after the national lockdown and worsening labor market conditions by the end of 2020.

Reducing and Reporting IPV

In Uganda, as in much of Sub-Saharan Africa, intimate partner violence (IPV) is widespread. According to data from the Uganda Bureau of Statistics in 2016, 49.9% of ever-partnered women between 15 and 49 reports experiencing IPV at some point in their life, with 29.9% reporting that it happened at some point in the last 12 months (UBOS and ICF 2018).
While there is increasing evidence showing mixed effects from cash transfers to women on IPV, we know less about how economic empowerment through self-employment affects IPV. Both cash transfers and productive self-employment increase women’s incomes. However, income from self-employment depends on women’s skills and effort. This might affect IPV differently from the economic windfalls of cash transfers.
We study a skills-based entrepreneurship program in Uganda that teaches business management skills (accounting, inventory management, etc) alongside entrepreneurship skills (raising capital, growing a customer base, identifying market opportunities). We assign treated women to one of two mentoring programs: intensive mentoring, in which a coach tries to visit the women at their homes or businesses three times over the six months of the program, or opt-in mentoring, where women choose to show up for mentoring at a centralized venue. Lang and Seither (2022) show that the program has large impact on business ownership and profits. Women re-invest heavily, limiting the benefits that flow back to households.
In a companion paper (Casabianca et al. (2022)), we consider the effect the program has on IPV 12– 18 months after it ends. These effects are not obvious. Women may face pressure to allocate more of their profits to household expenditures, poEntrepreneurshiptentially causing an increase in IPV due to expropriation. Alternatively, having an independent source of income may place women in a stronger position within the household. Identifying and quantifying such effects is critical for the design of programs aimed at increasing women’s economic inclusion.
Measuring the effect of any program on IPV poses challenges because there are two possible effects: true reductions in IPV, and reductions in reported IPV. This concern is particularly relevant for programs that focus on women’s empowerment. Women who participate in such programs may be less willing to report IPV to study enumerators if they know the study is evaluating a program focused on empowerment, potentially leading to inaccurately large, negative effects on IPV.
We follow Dhar, Jain, and Jayachandran (2022) to address this concern. We use the social desirability index developed by Crowne and Marlowe (1960) to measure a respondent’s propensity to provide socially desirable answers. We then estimate heterogeneous effects on IPV based on whether a woman has above versus below-median social desirability bias. Doing so allows us to establish the social norm by observing which direction social desirability biases answers of women in the control group. The heterogeneity analysis allows us to understand whether the treatment is interacting with social desirability bias to generate under-reporting.

Addressing seasonal hunger and rural poverty in Zambia – Testing scalable solutions

Since farmers harvest crops (and therefore earn income) just a few times a year, many agricultural households struggle to save their cash and maintain their consumption in the months between harvests. For this reason, the period before harvest is often termed the “hungry season”. Previous research has shown that offering well-timed, seasonal credit to farmers can help boost agricultural productivity and smooth consumption, but few programs designed to provide seasonal credit to small-scale farmers exist in low-income settings. We work with a private sector partner to test a scale up of seasonal loans. The program increased farmer engagement with the company and had high take up, but led to high levels of default, particularly during the COVID-19 pandemic, which led to its discontinuation.

See all publications