When we look at the pension expenditure of the states, so far, there has been no gain from the NPS reform.
Pension expenses as a share of total state revenue expenditure rose from 4% in 1991-1992 to 10% at the time of the NPS reform, and have risen further to about 12%, observes Ajay Shah.
The Old Pension System (OPS) was an important threat to India's fiscal stability.
At first, the National Pension System (NPS) had the promise of becoming the comprehensive solution to this problem.
The NPS is now missing with military personnel, West Bengal, Rajasthan, and a few other states.
This is starting to look like an important problem in the fiscal outlook.
There has been a loss of institutional memory on fiscal stress, the role of rising pension payments in this stress, and the strategy for solving this through the NPS.
The NPS was born of the fiscal crisis of the late 1990s. Pension payments -- both civil and military -- were growing at an explosive pace.
In some states, pensioners were experiencing delays in pension payments.
This was a uniquely unnerving experience for the people who had been promised a guaranteed pension.
The writing was on the wall, that the promises of stressed governments were not entirely reliable.
Projections for the future showed that things were going to get worse.
The old-style pension promises of the Indian State are not affordable with increased longevity.
The promises that an employer can afford, when a pensioner will die on average by age 65, are expensive when the expected life span goes up by 10 to 20 years.
This forces difficult choices upon policy makers. It is particularly stark in defence.
India's gross domestic product is a given and we can spend about 2.5 per cent of GDP on defence.
This gives a fixed pool of resources.
Within this, we face a choice between a small sophisticated military where soldiers have a better chance of surviving and winning battles, versus a weak army with generous pensions.
The National Democratic Alliance-1 and the United Progressive Alliance leadership teams looked hard at these difficult choices and chose the politically difficult path of pension reform.
From the viewpoint of a civil servant, it was better to get a block of personal wealth at retirement, and decouple one's own financial planning in old age from the financial problems of the employer at dates deep in the future.
These decisions were shaped by the then prevalent fiscal stress.
The fiscal stress of the Union government is best measured as the share of interest payments in net tax revenue. This peaked to 80 per cent in 2001-2002.
It improved dramatically to 39 per cent in 2007-2008.
But chronic fiscal stress is not behind us.
This value is back up to 48 per cent in 2020-2021 even though interest rates are at low values.
With weak growth and difficulties in the fiscal system, we are in for a difficult few decades.
This is not a good time to revert to more profligate decisions such as going back to the defined benefit (DB) pension.
Another problem to contend with is growing longevity.
The OPS became unaffordable by 2002, when the Vajpayee Cabinet made the decision to introduce the NPS.
Over the following 20 years, longevity has gone up.
Over the next 40 years, longevity will go up further.
Calculations about the OPS need to reckon with the improvements of health in India that take place every year, which drives up its cost.
Pension reforms are a difficult field in that there is a long delay between the hard work of doing the reform, and reaping the gains.
For one generation, starting from January 1, 2004, the NPS reform is actually fiscally expensive.
It involves paying the old generation the DB pension and the young generation the NPS contribution.
The real fiscal gains from the NPS reform lie further out in the future, where elements of the Indian State which manage to do the reform -- and hang in there against attempts at reversing the reform -- will reap substantially greater fiscal health.
Hence, when we look at the pension expenditure of the states, so far, there has naturally been no gain from the NPS reform.
Pension expenses as a share of total state revenue expenditure rose from 4 per cent in 1991-1992 to 10 per cent at the time of the NPS reform, and have risen further to about 12 per cent.
The nominal year-on-year growth of state pension payments has been consistently higher than that of nominal GDP.
Without the NPS, like in West Bengal or Rajasthan, or with defence pensions of the Union government, this share will just keep growing for decades.
In contrast, for the Union civil pension expenses, and for states that are still in the NPS, there is redemption at the horizon.
In about 17 years time, the gains from NPS will kick in: Pension expenses as a share of total expenses will first stop growing and then gradually tail off to zero.
That is how big the stakes are: We are looking at 12 per cent of the present revenue expenditure of the state governments.
This will grow a lot (and we have no estimates about how far this will go) under the OPS, or it will go to zero for the government organisations that have the nerve to go through with the NPS reform.
These difficulties are a cautionary tale on the process of reforms.
Like other aspects of improvement of the republic, the work of pension reforms is determined by the quality of the pension reforms community and its discourse.
In the 1990s and 2000s, there was a remarkable pension reforms community.
Many people thought about the subject, many papers were written and there was a bubbling discussion on the subject.
This community developed the NPS and carried it through into implementation.
In pensions, as in many other elements of policy, the dispersion of the erstwhile policy community removed the intellectual foundation of the policy process.
This has adversely affected the policy outcomes of the following years.
Policy reform is not an alpine style assault.
Establishing a good policy framework means never saying that you are finished.
Each field requires a sophisticated policy community, that is analysing and thinking, that is bubbling with the policy discourse and that carries the institutional memory from one generation to the next.
Ajay Shah is a researcher at the Pune International Centre.
Feature Presentation: Aslam Hunani/Rediff.com