📄 Abstract
Indian IT services has removed idle capacity without improving deployment. Bench strength has fallen from roughly twenty-two per cent to under eleven per cent in four years, yet net utilisation of deployed staff has declined over the same period from 87.6 to 85.0 per cent. The two movements cannot both be explained by capacity discipline; together they indicate that the binding constraint has shifted from having too many people to matching the people retained against the tasks that remain. This study treats that shift as a problem of allocative efficiency and asks what human resource architecture resolves it. Redeployment policy is operationalised as an eight-component Redeployment Policy Index (RPI), and allocative efficiency through a composite Allocative Efficiency Index together with the within-unit dispersion of the log marginal revenue product of labour, the standard misallocation measure. Evidence is drawn from 540 redeployment episodes nested in 62 business units across 24 organisations, supplemented by interviews with 75 workforce-planning leaders. Three estimators are used: a multilevel model of allocative efficiency, a Cox proportional hazards model of time to productive reassignment, and a complementarity test of policy bundles. Redeployment policy raises allocative efficiency strongly (γ = 0.412, p < 0.001) and compresses misallocation, with dispersion falling from 0.62 to 0.28 across policy quartiles and an implied gain of 11.4 per cent in output per employee-hour. The central result is an interaction: automation intensity on its own reduces allocative efficiency (γ = −0.186, p = 0.002), but the interaction with redeployment policy is positive (γ = 0.014, p < 0.001), so that above an index value of approximately 65 deeper automation becomes allocatively productive rather than disruptive. Policy components behave as complements rather than substitutes, with observed efficiency exceeding the additive prediction only above a critical mass of five of eight components. Matching infrastructure, not income protection, drives reassignment speed: the talent marketplace carries a hazard ratio of 1.94 while the redeployment guarantee window is not significant. Employees above forty-five and those with more than ten years of tenure are reassigned significantly more slowly after full controls. The study closes with a structured seven-study research agenda extending the HRM-variable-to-economic-outcome programme of which it forms the second instalment.
🏷️ Keywords
📚 How to Cite:
Dr. N Subbukrishna Sastry, Dr. Manjula Mallya M , TALENT REDEPLOYMENT PRACTICE AND TASK REALLOCATION EFFICIENCY UNDER AUTOMATION , Volume 14 , Issue 8, September 2026, International Journal of Indian Economic Light(JIEL) , Pages: 1 - 21 ,