Progress and Stagnation in the Livelihood of Informal Workers in an Emerging Economy: Long-Term Evidence from Indonesia


The United Nations University World Institute for Development Economics Research (UNU-WIDER) is assembling a journal special issue and a book volume focused on understanding the constraints to livelihood enhancement of informal enterprises and self-employed and wage workers in the informal sector as part of the Institute’s new project, “Transforming Informal Work and Livelihoods”.

With “premature deindustrialization” and the growth of the informal service sector, it seems likely that the trajectory towards informalization in low- and middle-income countries may be intensified in the future. The challenge for policymakers is then to find ways to encourage the movement of workers from the relatively unproductive lower informal sector to the more productive formal sector and, at the same time, provide opportunities for more dynamic informal firms to grow and for those working in these firms to achieve decent and remunerative work, even while remaining informal. However, in this effort, policymakers are constrained by the lack of available evidence on the causes of informality and the most effective mechanisms for reducing informality and strengthening decent work in the sector.


This research project aims to address these gaps:

  1. To characterize workers’ livelihoods in different informal work segments; provide a better understanding of the drivers that cause workers to engage in these activities; and explore the workers’ elements of choice, compulsion, and contextual factors.
  2. To detect the main challenges and constraints that prevent workers in each segment from transitioning to a higher earning and more stable livelihood stream.
  3. To provide guidance on how to foster these transitions and enhance the livelihoods of those in informal work.

We propose to address the following research questions:

  1. What is the proportion of individuals whose first job is low-tier informal, compared with those whose first job is high-tier informal or formal? What are the characteristics of individuals whose first job is low-tier informal, compared to those whose first job is high-tier informal or formal?
  2. Has (1) changed over two decades, as Indonesia becomes a middle-income country?
  3. Of all the individuals whose first job was low-tier informal, what is the proportion of those who switch to high-tier informal or formal job? Is the switch permanent or transitory? Does the path from low-tier informal to formal always go through high-tier informal? What are the characteristics of individuals who managed to switch?
  4. What is the earnings premium of switching to either high-tier or formal jobs, relative to staying in low-tier informal jobs?

In this study, our focus is on individuals rather than firms. Specifically, we limit our sample to male workers in the non-agricultural sector. We use information on individuals’ work status and work type to create three groups: low-tier informal workers, high-tier informal workers, and formal workers.

The first three research questions form the descriptive section of our study. The first and second research questions are tabulations by age group and cohort using all five waves of the Indonesian Family Life Survey (IFLS). We will include the following correlates: education attainment, urban/rural residence when the sample is young, parental education attainment, and parental occupation. In addition, IFLS collects information that could be used as proxy for human capital: height, health outcomes during childhood (for a subsample), and cognitive skills (for a subsample).

To answer the third research question, our main sample consists of 15-year-old individuals or older from the 1993 wave, the first wave of IFLS. We will first construct the sample’s job type (low-tier informal, high-tier informal, or formal) annually, from 1988 to 2014. We can create outcome indicators, such as employment spells, the number of switches between job types, and an indicator of switching permanently. Afterwards, we will focus on the sample whose first job was as a low-tier informal worker.

We will divide the sample into three groups using their long-term work information: always low-tier informal, switch to high-informal, and switch to formal. We will use the same correlates as above, but only using the 1993 conditions to avoid endogeneity.

To answer the fourth research question, we will estimate an individual level estimation shown in Equation 1:

y_it= α+β_1 〖HI〗_it+β_2 F_it+β_X X_it+ε_it    (1)

We adopt Equation 1 from the baseline Mincerian specification in Levine and Rubinstein (2017). The outcome variable yit is earnings level of individual i in year t, to allow negative earnings for informal workers. HIit is equal to one if individual i in year t is a high-tier informal worker. Fit is equal to one if individual i in year t is a formal worker. The control variables Xit consist of potential experience and education attainment. More importantly, the data allow us to control for individual-specific linear time trend and individual fixed effects. Thus, we remove individual time-invariant unobserved heterogeneity and individual-level trend into the choice of work. Since various selection biases could compromise our estimates, we will follow Levine and Rubinstein (2017) in conducting various robustness checks.

Share this page

Daniel Suryadarma
Asep Suryahadi
Team Member 
Mayang Rizky
Completion Year 
Project Donor 
United Nations University World Institute for Development Economics Research (UNU-WIDER)
Type of Service