Formal and informal employment shifts linked to distinct health outcomes

The world of work has changed profoundly over the past several decades, and not always in ways that protect the people doing it. Across the globe, labor markets have shifted toward flexibility, giving rise to platform-mediated gig work, labor dispatch arrangements, and a vast array of jobs without formal contracts. According to the International Labour Organization, informal employment now involves nearly 2 billion workers and accounts for 61.2 percent of total employment worldwide. In the Asia-Pacific region—the epicenter of this phenomenon—roughly 66 percent of workers are informally employed, representing close to two-thirds of the global informal workforce. For these workers, the absence of a direct, stable employment relationship with the organization for which they work means weaker regulatory protection, thinner social insurance coverage, and chronic exposure to income insecurity. Public health researchers have increasingly recognized this arrangement as a central social determinant of health, one that structures who gets exposed to stress and who gets protection from it. But a new study from China suggests something more troubling: the health damage caused by slipping into informal work may not be undone by climbing back out.

The research, published in SSM – Population Health by Liwen Chen and Kun Liu, tackles a question that most previous studies have sidestepped. The bulk of the existing literature on informal employment and health relies on cross-sectional comparisons—snapshots of formal and informal workers taken at a single moment in time. These studies are useful, but they cannot answer a more fundamental question: what happens to a person’s health when they actually move between employment statuses? Does entering informal work cause a decline in health? And, crucially, does exiting informal work restore it? The assumption baked into most research is symmetry—that transitions in opposite directions have equal and opposite health effects, mirror images of one another. Recent longitudinal research on related questions has begun to challenge this assumption. Studies of unemployment and reemployment have found that while these transitions have similarly sized but opposite associations with psychological health, other employment changes are demonstrably asymmetric. Transitions into and out of informal employment, some evidence suggests, may carry unequal consequences for life satisfaction and well-being.

There are solid theoretical reasons to expect asymmetry. The stress-process framework, developed by sociologists Leonard Pearlin and colleagues, holds that adverse events affect health not only through direct exposure to stressors but also through the loss of coping resources and the spillover of strain across work, family, and economic domains. Entering informal employment fits this description precisely: it typically involves reduced fringe benefits, weaker protection from labor regulation, and diminished access to employment-based social insurance. It also heightens uncertainty about job continuity and future prospects, generating psychosocial stress that undermines well-being. While informal work can sometimes offer autonomy or a better fit with family responsibilities, those benefits depend heavily on whether the transition was chosen or forced. By contrast, the losses that come with informal work—income predictability, institutional coverage, employment protection—hit a much broader group of workers regardless of preference.

The reverse transition, from informal back into formal employment, might be expected to restore what was lost. But the life-course perspective suggests otherwise. The adverse effects of precarious work are often cumulative and persistent: chronic insecurity and hazardous conditions generate physiological and psychosocial wear that does not quickly dissipate when circumstances improve. Research on adverse employment histories across the working life links prolonged exposure to insecure employment with allostatic load—the cumulative physiological burden of chronic stress—indicating that the body keeps a record of precarity long after the precarity itself ends. This is what researchers call “scarring”: health damage from poor-quality employment that is not fully offset by later improvements in job quality. If informal employment leaves scars, then moving into a formal job may halt further damage without repairing what has already been done.

To test these ideas, Chen and Liu turned to the China Family Panel Studies, a nationally representative longitudinal survey conducted by Peking University’s Institute of Social Science Survey, covering the years 2014 through 2020. China offers an especially informative setting because informal employment is widespread and labor-market stratification plays a central role in shaping workers’ exposure to insecurity and unequal protection. The researchers restricted their analytic sample to non-agricultural workers aged 16 to 59 who were actively working at the time of each survey—excluding retirees, full-time students, and observations with missing data on key variables. The final sample comprised 19,489 person-wave observations drawn from 7,167 individuals. Informal employment was defined broadly, encompassing both workers without a formal labor contract and those employed through labor-dispatch arrangements, a legal structure in China in which the intermediary agency, not the workplace, holds the employment contract—separating the legal employer from the organization that actually supervises the worker’s labor.

The health outcome was self-rated health, a measure that asks respondents to evaluate their overall condition on a five-point scale. While it may seem subjective, self-rated health is one of the most extensively validated indicators in population health research. Individuals synthesize information about both bodily and psychological states when forming these assessments, and the measure predicts subsequent morbidity and mortality across a wide range of settings, including China and other lower-income countries. Its breadth makes it well suited to a study of employment transitions, which may affect health through overlapping material and psychosocial pathways—psychological distress, physical symptoms, functional capacity, and chronic disease processes—that respond on different timescales.

The methodological core of the study lies in how the researchers handled the panel data. They began with pooled ordinary least squares models, but immediately confronted the problem of endogeneity: individuals in poorer health may be more likely to enter or remain in informal employment because of reduced work capacity, while unobserved traits such as resilience or family support may influence both employment status and health. To address this, they estimated individual fixed-effects models, which use each person as their own control and thereby eliminate bias from time-invariant unobserved characteristics. But the standard fixed-effects specification imposes a symmetry restriction—a single coefficient assumes that entering and exiting informal employment have equal but opposite associations with health. Chen and Liu broke this constraint using an asymmetric fixed-effects approach following statistician Paul Allison’s 2019 framework. The method decomposes within-individual changes in employment status into two non-overlapping transition indicators—one capturing entry into informal employment, the other capturing exit—and accumulates the number of each type of transition over the observation period. A mean-deviation estimation, mathematically equivalent to first differencing but more robust for unbalanced panels, then yields separate coefficients: one for the health association of entering informal work, and one for leaving it. Wald tests formally assessed whether the two coefficients differed in magnitude.

The sample offered substantial variation to work with. Of the 7,167 individuals, 5,281 experienced no observed change in employment status, while 1,886—26.3 percent—transitioned between formal and informal employment at least once. In total, the data contained 2,302 transition episodes: 1,165 moves from formal to informal employment and 1,137 moves in the reverse direction. Descriptively, the sample skewed young and male, with a mean age of 38.3 years and 60.1 percent men; 54.1 percent of person-wave observations were classified as informal employment, and fewer than half of respondents were covered by pension or medical insurance. Beyond the main models, the researchers deployed event-study and segmented-regression analyses to disentangle timing. Because the CFPS interviews occur every two years, event time was measured in roughly two-year intervals, with the wave immediately preceding the transition serving as the reference point. Lead indicators revealed whether health was already changing before the transition occurred—a critical check against reverse causation—while contemporaneous and lag indicators showed whether health changes were concentrated around the transition or continued to unfold across later survey waves.

The asymmetric design matters not just for statistical elegance but for policy. If entering informal employment is associated with a health decline while exiting it produces no comparably clear improvement, then the implications for public health are stark: preventing avoidable or unwanted transitions into informal work may be at least as important as facilitating movement into formal employment. Traditional labor-market policy often focuses on job creation and formalization—helping people out of informality. But a scarring model implies that the damage done during informal spells persists, meaning that the timing of intervention is crucial. Waiting until workers have already cycled through precarious employment may leave much of the harm unaddressed.

The study also speaks to a broader debate in the literature on adaptation. Recent evidence on life-event adaptation suggests that some individuals adjust to changed circumstances while others do not, and qualitative research on precarious work shows that the constant adjustment to unstable workplaces is itself stressful and exhausting. Recovery after returning to secure employment may be incomplete if prior insecurity has already imposed material and psychological costs. Meanwhile, formal employment is not an unalloyed good: it can introduce tighter supervision, lower discretion, and higher job demands that temper the expected benefits of improved employment status. All of these forces point in the same direction—toward health dynamics that are asymmetric, path-dependent, and slower to heal than they are to wound.

By treating informal employment not as a static employment category but as a dynamic social exposure that unfolds over labor-market trajectories, this research reframes how we understand the health consequences of the modern economy. The gig economy’s growth is often described in terms of flexibility and choice, but the epidemiological picture suggests a deeper stratification: a labor market in which the direction of movement matters as much as the destination, and in which the body accumulates the costs of insecurity in ways that formal contracts, however welcome, cannot fully repay. As informal work continues to expand worldwide, the question is shifting from whether precarious employment harms health to how deeply, and for how long, that harm endures.