Fiscal stress emanating from civil service retirement benefits was imposing an annual expenditure of over $30 billion on the central and state governments and had created an implicit pension debt—that is, the aggregate net present value of all pension payments due to government employees in the future—of over $600 billion, equivalent to 64% of India’s gross domestic product in 2006. The steep increase in states’ pension outlays was due mostly to the significant expansion in the number of state government employees, extension of pension coverage to employees of various nongovernment institutions, periodic increases in cost of living allowances, and improvements in life expectancy.
Almost all states were increasingly concerned about the sustainability and fiscal implications of their existing pension schemes and decided to adopt the New Pension Scheme. However, states still face sizable legacy costs for employees and pensioners covered under the old defined benefit scheme, although the magnitude and long-term impact of these costs remain unknown. Furthermore, almost all state governments lack the appropriate arrangements to collect and monitor data relating to current pensioners and future retirees. To quantify the financial impact on expenditure and savings from the reform accurately, state governments require good-quality demographic, income, and service data on their existing employees.
This book contains an account of the processes undertaken during, results achieved by, and the lessons learned from the efforts of five states—Assam, Bihar, Chhattisgarh, Jharkhand, and Madhya Pradesh—to estimate their current pension liabilities, to project annual pension costs over the next 15–25 years, and to explore options for managing their annual costs. These efforts benefited from technical assistance from the Asian Development Bank, which has supported and continues to support pension reform in Asia and the Pacific.
It is our hope that this book proves to be a valuable resource for policy makers, academics, and pension industry stakeholders, not only in India but also in other Asian Development Bank member countries, in instituting the same or similar pension reforms as those undertaken in the five states in India.
Sultan H. Rahman
Director General
South Asia Department
ACKNOWLEDGMENTS
This book records the efforts carried out from 2006 to 2011 to establish a pension and payroll database system for five states in India: Assam, Bihar, Chhattisgarh, Jharkhand, and Madhya Pradesh. The databases were then used to project each state’s pension liabilities and explore options to manage such liabilities. These efforts were supported by a technical assistance project financed by the Japan Special Fund of the Asian Development Bank (ADB). The assistance was headed by Cheolsu Kim (project leader), principal financial sector specialist, ADB, and Gary Hendricks (task manager), a veteran regulator and pension analyst with over 35 years of experience.
The transition from a conventional paper-based to an automated payroll recording system was a huge undertaking. The mind-set of people accustomed to traditional practices had to be changed. Political commitment from each state government—to adopt the proposed reforms and to adapt to these— needed to be secured. Each state’s efforts had to be guided through the arduous task of manually inputting data on up to 300,000 employees and 150,000 pensioners. Technical advice and oversight had to be provided during the formulation and implementation of automated procedures capable of continuously updating data once automated. Throughout these efforts, the project benefited from the strong commitment and generous effort of many institutions and individual stakeholders, which we acknowledge and for which we express our appreciation and gratitude.
First, we wish to acknowledge the Government of India for its active participation and support for the project. In particular, we would like to express our gratitude to K. P. Krishnan, secretary, Economic Advisory Council to the Prime Minister; Tarun Bajaj, joint secretary (pension and insurance), Department of Financial Services; Shashank Saksena, director, Department of Financial Services, Ministry of Finance; and Anuradha Thakur, director, Department of Economic Affairs, Ministry of Finance. Discussions with and comments from them were very useful and provided many insights, which have been used in the implementation of the project and in the writing of this book.
The five states played a pivotal role in the installation of automated employee, payroll, and pensioner databases, with ADB providing the project design and required inspiration and guidance, as well as technical and occasional hands-on assistance. Each state provided the human resources, supervision, and financing (sometimes with donor assistance) necessary to construct the databases, execute the automation designs, update computer hardware and software, and implement new processing procedures. Deep gratitude is therefore offered to the state secretaries of finance, without whose foresight, commitment, and leadership none of the states could have succeeded in installing the data systems. Also crucial were the participation and support of the directors of treasuries and accounts in each state and major district treasuries. They were the primary instruments of project implementation by supervising the many dozens (usually hundreds) of state civil servants who provided the source data and were trained on and executed the new methods of operation. Indebtedness is also owed to the State Bank of India, which provided high-quality, fully automated data on pensioners whose pensions it administers. In some states, it administers upward of 90% of all state pensions. The effort to install automated payroll and pensioner databases in the five states could not have succeeded without its assistance.
We would like to acknowledge the consultants who worked so tirelessly to complete this project. Subhash Garg, an Indian Administrative Service officer who was on sabbatical and, later, partial leave, acted as co-team leader for the 4 years required to construct the databases. Two international information technology experts, Ashish Joshi and Sanjay Sangal, worked assiduously in the field, providing critical technical assistance to the states.
Finally, we also wish to acknowledge Bruno Carrasco, director, South Asia Department, for his support and invaluable advice. Anna Alipio, associate project officer, and Pamela Gutierrez, associate project analyst, South Asia Department, provided administrative and logistic support, and Kimberly Fullerton, consultant, provided editorial support.
Cheolsu Kim
South Asia Department
I. INTRODUCTION
As part of an effort to control escalating civil service pension costs, the Government of India closed its defined benefit scheme (DBS) for pensions to new entrants on 1 January 2004. Civil employees hired on that day or after were and will be enrolled in a defined contribution scheme, the New Pension Scheme (NPS). Under this new scheme, the government and the civil servant each contribute 10% of the employee’s basic pay to a retirement fund, which is invested. At retirement, the balance of the employee’s retirement account, consisting of 20% of wages and all interest that accrued during the employee’s civil service career, is available to support the employee.1
Employees hired before the NPS was included in notification letters of new employees remain in the existing DBS. Hence, the government is still liable for all payments to current pensioners and for the pensions and other DBS benefits of pre-NPS employees. Most such employees will retire within 30 years. However, pensions to survivors and certain disability pensions will continue to be paid well beyond that.
The government has encouraged states to follow its lead by closing their traditional DBSs to new entrants and by adopting defined contribution schemes like the NPS. Many states did move quickly to adopt defined contribution schemes; by the end of 2005, 15 had done so. However, little thought appears to have been given to the full implications of implementing a defined contribution scheme. Records need to be kept centrally and investments centrally managed according to enrollees’ choices; further, states are required