Worker absenteeism in labour-intensive industries causes productivity losses for firms, and reduced potential for productivity-based incentives for workers. Based on a study in Karnataka, this article investigates how factory line-managers trade workers amongst themselves when facing particularly low attendance on their lines, and estimates the financial gains that firms can make when they improve the distribution of workers across lines.
Absenteeism is a well-known issue in the public sector in developing countries, particularly affecting the provision of education and healthcare services. A study by Kremer et al. (2005) finds that during unannounced visits to a nationally representative sample of government schools in India, 25% of teachers were absent and only about half were teaching. An analysis by Muralidharan et al. (2011) of attendance data collected during a 2003 survey of over 1,400 public health centres across 19 major Indian states, finds that nearly 40% of doctors and medical service providers are absent from work on a typical day.
High attrition rates and worker absenteeism are also widespread in labour-intensive industries such as garment manufacturing, in low-income settings. Yet, little is known about how managers and firms manage production through absenteeism and how well their methods work.
In a recent study (Adhvaryu et al. 2020), we investigate how factory line managers trade workers amongst themselves when facing particularly low attendance on their lines, and estimate the financial gains that firms can make when they improve the distribution of workers across lines.
Using six consecutive months of worker-level productivity data across four garment factories operated by Shahi Exports in Karnataka, we find that 10-11% of workers were absent on a typical day with nearly all absenteeism being ‘unauthorised’. The reasons ranged from unexpected domestic work, religious or cultural festivals that require travel to the worker’s native places which are often in rural areas, and take-up of temporary economic opportunities that seem more lucrative.
Worker absenteeism, productivity, and earnings
Garment manufacturing firms employ an assembly-line process of production, where the material is first cut, sewn into a garment, and then sent for finishing. The sewing section of the process, which is organised into sewing lines consisting of 50-60 sewing machine workstations, accounts for most of the workforce and involves the most time-consuming and value-added tasks. Every line has a fixed line manager who oversees daily activities, and is responsible for staffing their lines and meeting daily production targets. Due to the sequential nature of production, when a line experiences high worker absenteeism, segmental delays become bottlenecks that ultimately impede the line’s productivity – which in the garment industry is measured as ‘efficiency’ (the fraction of the target quantity of a particular operation that is completed per unit time).
We find that absenteeism shocks have little impact on productivity until average absenteeism reaches 10%. After this point, every additional percentage point increase in absenteeism reduces productivity by 0.25 percentage points.
Since bonuses are awarded to all workers in a line that meets its daily production target, lower line-level productivity due to absenteeism can negatively affect workers’ daily pay on average. A one standard deviation1 increase in line-level absenteeism (13 percentage points) can amount to a reduction of Rs. 13 in the average daily worker bonus of Rs. 8.73, and a decline in the average probability of receiving a bonus (which is about 25% on a given day) by 28.6 percentage points.
Overcoming absenteeism shocks through relational contracts
When faced with high worker absenteeism at the line level, managers leverage informal agreements called ‘relational contracts’, to borrow workers from other line managers with the shared understanding that they will return the favour when partner lines experience worker shortages in the future. However, we find that managers tend to form, on average, at most five trading partnerships, foregoing approximately 15 to 17 partnerships that could be formed with other managers in the factory. Additionally, 40% of all workers borrowed by a manager come from their primary exchange partner. In the event of a large absenteeism shock, where a line manager needs to borrow as many as five workers to attain peak efficiency – which no one manager could lend – the small number of trading partnerships constrains the pool of workers from which they could borrow. Upon further analysis of the characteristics of trading partners, we find that 72% of workers are exchanged between lines that are at most 20 feet apart, with 66% of trades between managers with similar educational levels, and 71% of trades between managers of the same gender.
Using production data for the same four factories, we estimate a production function, which maps the number of workers present in a line to its efficiency on a given day. Counterfactual simulations show that if the firm can generate even three more active trading relationships per manager, it can unlock profits of up to US$ 247,000 per year through improved line and factory productivity. While this would undoubtedly be difficult to accomplish, simulations like these emphasise that relational contracts among managers are highly valuable for the firm.
Intermediary solutions to improve worker allocation
Given the highly unpredictable nature and scale at which worker absenteeism occurs in the garment industry, it is almost impossible for firms to effectively manage worker allocation centrally on a daily basis. To circumvent the demographic and physical distance barriers that leave many potential productivity enhancing trades between line managers unrealised, our research suggests the need for firms to explore ‘intermediary’ solutions to facilitate trading and improve worker allocation.
This solution could take the form of a technology – for instance, a low-cost application that matches worker shortages to spare workers as reported by managers at the start of the day. Such interventions which are scalable, and agnostic to social obstacles which hinder relational contract-building between managers, could make production more resilient to the shocks of worker absenteeism. This would benefit firms through improved consistency of production, and line workers and managers through a better chance to boost their earnings by meeting daily production targets.
I4I is now on Telegram. Please click here (@Ideas4India) to subscribe to our channel for quick updates on our content
- Standard deviation is a measure that is used to quantify the amount of variation or dispersion of a set of values from the average of that set.
- Adhvaryu A, JF Gauthier, A Nyshadham and J Tamayo (2021), ‘Absenteeism, Productivity, and Relational Contracts Inside the Firm’, Working Paper 21-109, Harvard Business School.
- Kremer, Michael, Nazmul Chaudhury, F Halsey Rogers, Karthik Muralidharan and Jeffrey Hammer (2005), “Teacher Absence in India: A Snapshot”, Journal of the European Economic Association, 3(2/3): 658-667.
- Muralidharan, K, N Chaudhury, J Hammer, M Kremer and FH Rogers (2011), ‘Is There a Doctor in the House? Medical Worker Absence in India’, Harvard University.