Thursday, February 05, 2015

Understanding one-rank-one-pension

This is a post that should have been written a long time ago. It seems impossible to reverse one-rank-one-pension in India now, and it seems hopeless to fix the military pension. However, it's worth understanding what the issues are, and the mess that we've landed in.

Nominal annuities in a zero growth environment

Let's start with a country with no GDP growth. So a worker earns some income all his life. We want him to tuck some money away every year into a pension account that will buy an annuity at the age of retirement. He builds up this pension wealth, and at age 60, he buys an annuity. It's common to target a "50% benefit rate", i.e. the magnitude of the annuity should be half his last wage.

To fix intuition, let's assume the number A is the price of an annuity which yields a flow of income of Rs.1 per month. In this case, the pension wealth to get to half the last wage is Aw/2.

This is the challenge of the ordinary pensions discussion. If you want an unfunded, i.e. a `defined benefit' pension, then you want the taxpayer to pay Aw/2 for each person. There is no other difference; the basic story is the same.

In India today, A is roughly Rs.133. That is, if you buy an annuity at age 60 which pays Rs.1 per month until death, the price is Rs.133.

So far, we have asked the annuity provider a simple question: We have said: I want a fixed cash flow of Rs.1 per month until I die. What would you charge for this? The annuity market says: I will charge A for this contract. This is the lowest price of an annuity; this is a simple unindexed nominal annuity.

Now we can modify the terms of this annuity in many ways.

Real annuities in a zero growth environment

You could say: Instead of giving me a nominal Rs.1 per month, give me an inflation indexed Rs.1 per month. This is an inflation indexed annuity. This will of course cost a lot more than A. To produce a nominal annuity, the annuity provider invests in nominal bonds which produce a stream of cash. But to produce a real annuity, the annuity provider has to invest in inflation-indexed bonds, which yield a lower stream of cash. Hence, it needs much more than A to produce an inflation indexed stream of Rs.1 per month. Suppose the price is B, and we know B >> A.

A government that promises an unfunded inflation-indexed annuity is placing an expense on the tax payer of Bw/2.

The problem of GDP growth

Into this environment, let's inject high GDP growth. Let's go to the higher side by assuming per capita GDP growth of 7%, which means that per capita GDP doubles every decade.

When a person is 60, he was at half the wage of persons who are 59. But when he reaches 70, his pension has stood still, but the persons who are at age 59 have (roughly) got a doubling of their wage. The pensioner has lost ground compared with the worker.

That pensioners lose ground when compared with workers is a fact of life of all pension systems. In the West, where pensions were invented, this was not a big deal, as they have had a slow growth environment. But when there is high GDP growth, this can yield glaring gaps. A pensioner who is at the 90th percentile of the Indian income distribution at age 60 will endup at perhaps the 70th percentile of the Indian income distribution at age 70.

This is just a fact of life and you can't do anything about it. Anyone who builds up wealth in India over the working years 1980 to 2020 will seem prosperous in relative terms in 2021 but will seem less prosperous in relative terms in 2031 and in 2041. That's the inexorable logic of high GDP growth.

Suppose we go to the annuity provider and say: Sell me an annuity which is not just inflation indexed, but wage indexed. The payment per month will go up to reflect the average wage growth of the economy. This is a pension which will keep up with the Joneses.

In my knowledge, there is no private insurance company which will produce such an annuity. It's a very expensive annuity to produce. Let's use the symbol C for the price of this annuity. C >> B >> A.

A government that promises someone a wage indexed pension is asking the taxpayer to put up Cw/2 which is much bigger than Bw/2 or Aw/2.

That's one-rank-one-pension

This is the costliest pension imaginable. The Indian government seems to be on the trajectory of offering this for all military personnel. It is a dramatic escalation of the implicit pension debt for the government on account of military personnel.

Once this entitlement is in place, it will be hard for the government to go through with the NPS reform, where the second stage was supposed to be integrating uniformed personnel into the NPS.

It is a disappointment that we did not have adequate thinking on these issues in time. The delays and sloppiness of implementing the NPS have been extremely costly for India.


  1. Why should an annuity paying Rs 1 monthly cost ~4000 ?? 12 Rs annual yield on 4000? Or did you mean 1 Rs daily?

    1. I had a mistake in the earlier version of this post. Rs.4000 for an annuity maps to Rs.1 per day until death. But the text of the article has been written in units of rupees per month. Rs.133 for an annuity maps to Rs.1 per month until death. Thanks for pointing out.

  2. Very well articulated. Does this hold true even for those who take premature retirement and, if yes, wouldn't that make a material difference to the overall payout by the armed forces?

    1. The problem is the worst for people who take early retirement.

      For a person at age 35 who will live till 85, there are 50 years to go. Over this period, the gap between the NPV of a nominal annuity vs. a real annuity vs. a wage indexed annuity is MASSIVE.

  3. A Premature retirement or a Voluntary Retirement in the civil services (which is not a superannuation) is treated in the same manner in which a Superannuation is treated. There is no difference between these categories of pensions on the issues of Dearness Relief (DR) which is inflation indexation and the OROP which is wage indexed as it emanate from various pay commissions.
    However, as there is no OROP in the Armed Services, the only common element that they have is the DR. which is declared by the GoI from time to time and gets effective every year from 1st January and 1st July, the current being 107% wef 1st July 2014. The DR does not discriminate between a Voluntary, Premature and Superannuation Pension. Even Family Pensioners receive the same percentage of DR, though for them from the IT Act point of view it (both pension + DR) is covered under the head 'Other Income' and Not under the head 'Salary'.

  4. Ajay Shahji, has not mentioned about two points, from employers side obligation. i.e.,employer has to contribute an equal amount, to pension fund as employee contributes per month, Second point is Supreme Court given direction in many judgements stating that pension is an extended salary of an employee. This point should not be forgotten. Hence, pensions should not be shown in negative perspective. If there is a will, then there will be way. I appreciate his accounting methods. At last Tax Payers money is to be spent somewhere, in the nation on some head, by government.
    In addition, the employees contributed capital is lying with Pension Fund, after death of pensioner and his wife. That amount is never returned to anyone.

  5. But this does not take into account the fact that the corpus (accumulated over working years) used to purchase the annuity/pension is never returned on the death of the pensioner when pension ceases. This accumulates to a huge amount over a period of time and provides an additional income stream to the pension fund. If this is factored in then the corpus requirement will go down substatially and will be much less than 133..

    1. The fair value of an annuity reflects an averaging between people who die early and the people who die late.

  6. Are you suggesting that we pay our soldiers corporate salaries instead of the low wages they get. Further as they are not in cities and in remote areas may be taxpayers pay for setting up a corporate house to invest. Also given the risk to life and grievous injury no fund will agree for assuring they pensions like a regular corporate worker. Please be reasonable in your writings and bring out the complete aspect.

    I recommend you read Samuel Huntingdon's Soldier and the State. You will understand what is covenant. Also back closer home in arthasastra it is said if you do not look after your soldiers we will be a nation of cowards.

    What is the price a nation will pay in case it's army is filled with cowards? So we must not go down that path.

    Thus OROP is a great step.

    1. The government is obliged to buy inputs at the lowest possible price (L1 price) for the required technical quality.

      When recruitment into the armed forces takes place, there is an oversupply of recruits. This proves that for the risks, the compensation being offered is not grossly off.

      There may be an inequality problem -- senior staff are likely paid too little. This requires adjustment to increase inequality while holding the overall wage bill intact.

  7. Leave ur Re 1/- or Rs. 133/- pd song. We Armed forces have contributed to Indian Nation an area of 147,570 km² for which we paid with flesh and blood, which you later gifted to another govt. Yes I am speaking about the nation called Bangladesh. Where you can include this mammoth amount worth of area? We too gifted 90000 uniformed Pakistani Soldiers also to Indian Nation. We are not asking anything for free. At least pay a paltry OROP to the poor septuagenarian soldiers. We are not interested in your mathematics and arithmetic. Ask the north BLOCK Babus not to BLOCK at least this paltry amount of 8500 crore. We are not asking lfor 1,76 lakh crores of rupees

    1. Emotions cloud clear thinking.

      The government pays L1 price for steel, regardless of whether the steel is used to build guns or bridges.

    2. Please dont debase yourself by comparing human lives to steel. This is the problem with us today: human lives are less valuable to many people than material goods. Its sickening

    3. My friend, you are actually asking for 1.76 lakh crore if you do the mathematics properly.
      Govt would happily give 20,000 crore extra per annum. But it is not enough for OROP.

      Alternately the govt should say we will give 20,000 crore extra annually increasing it by 5% every year. You take it and divide it among yourselves. Are you ok with it??

      The cost of OROP will be over 1 lakh crores, not 8000 crores as you are stating.

    4. I respect the armed forces retiered but as per my survey I found that doctors from defence come and join medical colleges and blocking post of those candidate not having simple pension post. These reteired defence doctors enjoy pension and subsideze goods from defence canteens still demand pensions?

  8. It is all very well to argue from a finance and GDP perspective, but Brig Sarathy's argument has alot more weight: the defence personnel are in a totally different category - they CANNOT continue serving even when they want to and can! the OPROP for defence is justified and I hope the Civil Services esp the IPS dont start demanding it too. Time for the Civil Services to treat their comrades in arms a little more civilly and the PM to stand by his commitment

    1. If the PM stands by his commitment, the country will be bankrupted. Do some basic mathematics. This is not a question of respect to the army. It is about common sense.

    2. Presume the commitment was made after the math was done. If not, it means it's just political posturing for votes, which I refuse to believe of this PM. Commitment by definition means something you believe in, that's just common sense.

  9. The article may be correct economics wise. But in this case we are dealing with the lives of human beings who have at the age of about 20 or less given off their best for the defence of the motherland. Many have lost their lives in the bargain. About 80 percent are compulsorily sent home by the time they are just 35 years or thereabouts. Whereas those in civil jobs or even the police continue to serve till they reach the age of 60 year. If the Government assures the personnel of the Armed Forces continuous employment till the age of 60 years - like for other employees - then Ajay Shah's arguments may be worthwhile. In the absence of assured employment till the soldier reaches 60 years, and as he has to leave around 35 years of age, I do believe that OROP is more than justified. The maoney that the Government has "saved" in not paying 'higher wages' due to promotions beyond 35 years possibly accounts to 25 lakhs per person, which can be utilised for payment of the OROP!!

    1. Just emotional arguments. Do the maths and you will realize that the cost of OROP is at least 40% of the defence budget and more than the education, health and agriculture budget put together.

  10. "When recruitment into the armed forces takes place, there is an oversupply of recruits. This proves that for the risks, the compensation being offered is not grossly off."
    ...tell me, Sir, that you want to retract the above statement... and that you don't really believe that compensation to the Forces is merely a supply-demand thing...
    This is exactly why our Nation desperately needs some form of conscription; so citizens may appreciate the costs paid to ensure national security and the reason why the men ensuring the same need to be compensated honourably.

  11. A very educative Article. Wish the author would also do a similar study on the effects of introducing NFU by the 6th.CPC. NFU is applicable to all Govt. employes except the Military. It has effects on the pension entitlements of all those who entered service before NPS was introduced., still a larger section than the Military. Please bear in mind that a soldier’s life expectancy is 61 to 62 years while that of his equal in the civil services is 71 to73 years.

  12. Subramanyam V A has said in his comments on 14th.June "The article may be correct economics wise." I am wondering if that is entirely true. Ranjana in the post of the same date has alluded to a "finance & GDP perspective". This seems closer to the mark.
    In all discussions, there is another side to every story. Even Economics-wise. Readers of this blog may like to read another viewpoint, captioned as "Per Capita Soldiers in India" at
    These views are also based on finance and GDP.
    Would like to know how these viewpoints dovetail into the thesis of this blog.

  13. Let us see the whole problem from the perspective of the Armed Forces while keeping in mind the resource constraints of the Govt.

    There are two distinct classes to deal with when considering OROP. Personnel Below Officer Rank who are the largest services community which could number about 13 lakhs at full strength and Officers who number could be somewhat less than 50000 at full strength. Retired personnel number about 25 lakhs and family pensioner number about 3 lakhs. It would thus be seen that the pension pay out number is about twice the full strength of the whole cadre. If pensions are updated every ten years at the starting day of a pay commission the pay out for the first month would be the same as total expected salary bill at full strength of serving personnel minus various allowances except DA/DR which for serving personnel average 40 %. Over the next ten years of the pay commission period other than DA/DR there would be continuous widening of the 50 % pension due to promotions and increments bringing down the total pay out to less than 60% of the total salary bill.

    I agree that one cannot take a life time continually increasing liability of an early retiree who when reverting to civil life at an early age could find alternate employment or find self employment avenues. Also it may be difficult for the Govt. to afford a large continually increasing pension bill. What may be required is to have a re-look at the OROP definition that has been put out till now. It may be possible to tweak the definition to exclude very early forced retirees and voluntary retirees. While forced retires may be given full OROP voluntary retirees at an age below 45 years may be given partial OROP in the form of limited parity on percentage basis related to the difference between forced and voluntary retirement ages. Such a scheme will will reduce premature retirement number and also reduce pension out go to those who retire early for a second career / alternate means of earning a livelihood.

    Also OROP definition should be expanded to One Rank One Service One Cadre One Specialty One Service. This will ensure that parity given is logical and no one will get pension higher than that earned by a retiree after one month of commencement of a pay commission period. This will ensure a fair dispensation as during the currency of a pay commission period OROP is not applicable to those who retire during that period.

    These suggestion are practical and fair to everyone including the Govt.

  14. No Emotions, Even Economics Apart, I have a Very Simple Question to the North Bloc as well as to the author.
    Why is it that the Civil Servants are Entitled for OROP and Armed Forces Not?
    If the answer is confined only to the fact of 'Burden on Exchequer' then the Govt has to find out Additional Resources to fund OROP for Armed Services.
    If the answer is Demographic in nature that Armed Forces retire early, say at 40 and hence payment of pension is for a longer duration then the Govt. should pick up the responsibility to gainfully engage them till 60 and then offer the OROP at parity with civil servants.
    In both the cases the issues are NOT Economic in Nature, but more of a Political Will and Improved Management of Human Resources and Financial Management.
    Lack of Resources for the Armed Forces???? Are we not making a Mockery of the Public Financial Management System in India?? In order to gain political sympathies, we waste millions of US$ in providing a Special Status to certain States, Special Packages to states where the corrupt systems eat up the tax payer's money.
    The ONLY Ostensible Reason for this Differential Treatment is that the Civil Servants Make the Policies that are Pro CS. They are part of the Pay Commissions and the Policy Decisions. If that is the reason, then Why can't we have a separate Pay Commission for the Armed Forces? Is that the importance of Armed Forces realized only during Wars or External Aggression or Armed Rebellions??? Sixty Million Humans Globally are Refugees today due to Conflicts. India has no such issues as we have an Armed Force that we rely on even during the Peace Time.
    Government can not treat Civil Servants and Armed Forces differently on OROP Issue. The Sooner they Accept the More Grace they would display to Honor the Armed Forces.

  15. Very lucidly articulated...but I somehow do not understand how you could have conveniently glossed over some glaring facts unless of course the whole idea is to present a biased perspective against ESM.

    1. Firstly, though you have gone into great detail to explain c > A you fail to mention that the Govt currently pays Cw/2 to all pensioners except later inductees who are covered under NPS. Now if granting OROP will preclude including ESM under NPS at a later date, so be it...its peanuts compared to what they have contributed to the country and way lesser than what liability the Govt is undertaking now for all Govt servants.

    2. Secondly, lets talk numbers...if a Lt col retires at 54 and draws a pension of 50000, the Govt is saving 50000 per month which would have been paid to him as full salary, this amount indexed every year at 3% and about 12% DA if saved at a nominal 8% annual compounding interest (Same as PPF) would amount to 81,35,029. Now at your rate of 133 per Rs 1 pension, the Govt can afford to pay an extra Rs 61166 per month to this officer...that's a whopping 122% over and above his normal this bogey of Cw/2 you are raising is unjustified.

    3. Thirdly, the nation has the good fortune of having a highly professional and motivated armed forces for a pittance...Just contrast this...officers posted to UN missions draw emoluments which are as much as 30 times their Indian salary...If you counter that UN salaries may be exorbitant, suffice to note that all aover the world the Armed Forces have a significant edge over Civilians in pay and pensions except in India. So considering this, OROP is a pittance and you should not be making a mountain of a molehill.
    4. Your supply and demand angle...that holds good for only soldier ranks since we have a vast pool of rural youth desperate for employment but that does not justify exploiting them on low for the officer cadre the story is different...there is a huge shortage of officers and every officer has to take on the load of several officers without any extra remuneration. Today, officers children do not join the armed forces as a matter of tradition as before...going ahead, if as you say GDP doubles every 10 years, even soldiers will be in short supply unless we maintain the edge for the Armed Forces...

    I could go on and on...but suffice to say that not everything can be judged by economics...does anyone question the economics of maintaining their children or of looking after parents in old age ?...welfare of ESM is a responsibility of the nation that goes beyond economics.

    1. I am trying to understand the entire concept of OROP, hence ended up in this site.
      Just to notify you that your 2nd Pointer is not justified, as if someone retire, it is not a saving for GOI. The GOI need to backfill that position with almost same salary plus the pension of retired officer as well. This makes the situation more complicated

  16. Ajay Shah, I'm not questioning your logic. But there are more aspects to the issue.

    (1) Notwithstanding inefficiencies, corruption etc, the present economic prosperity of the country is partly due to the work put in by government servants (past and present.) We have every right to share in the economic prosperity of the country.

    (2) All civilian Group A officers have already got OROP via NFU (Non-Functional Upgradation). They all reach the Apex scale 2 years after the IAS does. NFU is bad in concept. But when you give it to all civilian officers, how do you deny it to armed forces officers ?

    (3) The very same civilian officers who get OROP are the ones opposing it for the armed forces. Hypocrisy.

  17. After reading the two judgments Nakara and Vains long back, I always felt that OROP is fully decided by the two SC pronouncements and it was just a question of making this applicable in case of all others and I had written on these lines in Indian Ex-servicemen group earlier. Hence, I always thought and expressed it in writing here in this group that the solution is a court case and not street agitation. Since no one validated my thinking, I thought, may be it was just reading too much into the judgments.

    Payal Chawla's analysis ( vindicates my view which I had written in Indian Ex-servicemen group earlier.

    I again went through both the judgments and there is a very strong case for OROP for all based purely on the judgments alone. An SC pronouncement can not be for the litigants alone but for all similar cases in the future by the principle of stare decisis. Stare decisis is Latin for "to stand by things decided." Stare decisis is essentially the doctrine of precedent. Courts cite to stare decisis when an issue has been previously brought to the court and a ruling already issued. Generally, courts will adhere to the previous ruling, though this is not universally true. Where it is not true is where the previous decision was distinctly flawed due to currently accepted social mores. Slavery is a distinct example.
    "Arguments from precedent and analogy are characteristic of legal reasoning. Legal reasoning differs in a number of ways from the sort of reasoning employed by individuals in their everyday lives. It frequently uses arguments that individuals do not employ, or that individuals employ in different ways. Precedent is a good example of this. In individual reasoning we do not normally regard the fact that we decided one way in the past as raising some presumption that we should decide the same way in the future. Of course there can be special circumstances that have this effect—someone may have relied on what we did before, or may have had their expectations raised that we would do so again—but absent these special considerations we do not regard ourselves as being committed in the future to make the same decision. It is always open to us to reconsider a decision and change our minds if we no longer think our original judgement was correct.
    Law of course is not alone in attributing a special significance to precedent. Many institutional and quasi-institutional practices place weight on what they have done previously in determining what they should do now"

    When you have a case which has the support of judicial pronouncement, it is not a case of just agitational demand but demand on the basis of rule of law and justice. That takes the issue to a different level altogether.


Please note: Comments are moderated. Only civilised conversation is permitted on this blog. Criticising me is perfectly okay; uncivilised language is not. I delete any comment which is spam, has personal attacks against anyone, or uses foul language. I delete any comment which does not contribute to the intellectual discussion about the blog article in question.

Please note: LaTeX mathematics works. This means that if you want to say $10 you have to say \$10.