Medical corpus

Health insurance in India pays for hospital stays. It does not pay for most of what actually happens to an ageing body.

The workbook is explicit: health insurance policies in India cover only hospitalisation and related expenses. So it says it is important to build a medical corpus to take care of the other expenses.

Those other expenses have names. Medicines. Doctors' visits. Tests. And long-term care costs, which can run for years and are nobody's idea of a hospital admission.

A medical corpus is money kept for exactly that, separate from the general retirement corpus so it cannot quietly be spent on something else.

At retirement it changes shape. The workbook says at least three years of these expenses must sit in liquid assets, with the rest in products that earn better returns though not very risky.

Why a separate pool at all. Two facts in section 8.1.3 combine to force it. Insurance covers only hospitalisation. And most people are covered by employer medical benefits while employed, but these cease on retirement — unless the individual works with the government and related agencies, in which case medical cover is provided in retirement too. So the person most exposed is a private-sector retiree on the day after they retire.

How to size it. The workbook gives four stages of health cost planning:

1. Understand the health risks and assess current health.
2. Consider the impact of hereditary factors.
3. Take the average of the immediately preceding 3 to 5 years' health costs as the starting point, and adjust it for expectations on health.
4. Estimate the cost in retirement after considering the years involved and the rate of inflation — noting that inflation applicable to health expenses is typically much higher than that for other consumer costs.

It also concedes the limits of the exercise. The expenses depend on the state of a person's health, which depends on gender, lifestyle choices and hereditary factors, so it may not be possible to have a definite plan — but even a rough plan makes sure health costs do not catch a retiree unprepared.

What it must hold at retirement. The instruction is specific: on retirement, the medical corpus created to meet health expenses not covered by insurance has to be moved to liquid assets for at least 3 years of expenses, while the rest can be held in products with better returns though not very risky.

Notice the shape that produces. It is the same logic as bucket 1 in the bucket strategy — near-term money in low-risk, accessible form — applied to a category of spending that arrives without notice.

What it sits beside, not instead of. A medical corpus does not replace health insurance. The workbook pairs it with a policy taken early, carrying lifelong renewability, with the sum assured raised at each renewal. Insurance handles the large, lumpy hospitalisation risk; the corpus handles the steady out-of-pocket drip. And PFRDA's own answer to the same problem is NPS Swasthya, introduced because out-of-pocket expenditure during old age serves as a major outflow of retirement savings.

Mr Wagle is 58 and retires at 60. He follows the workbook's four stages.

Stages 1 and 2. He is diabetic and his father had cardiac disease — a hereditary factor he must plan for rather than hope about.

Stage 3 — the base figure. He adds up the household's out-of-pocket health spending, excluding anything the insurer paid, for the last five years.

Five-year average = 2,50,000 / 5 = Rs 50,000 a year

He adjusts upward for his diabetes and the family history, to Rs 70,000 a year at today's prices.

Stage 4 — carry it to retirement. He uses 10 percent health inflation, higher than the 6 percent he uses for general expenses, exactly as the workbook instructs.

At 60, two years away:  70,000 x (1.10)^2  = Rs 84,700 a year
At 70, twelve years away: 70,000 x (1.10)^12 = Rs 2,19,700 a year

The general-inflation figure would have said Rs 1,40,800 at 70. Using the wrong inflation rate would have understated his need by nearly Rs 79,000 a year.

What he holds on the day he retires. Three years of expenses in liquid form:

3 x Rs 84,700 = about Rs 2,54,000 -> liquid funds and a sweep-in deposit

The balance of his medical corpus, Rs 9,00,000, goes into short-duration debt funds and bank deposits — better returns than a liquid fund, and not very risky, which is the workbook's wording.

The year it earns its keep. At 67 he needs a knee replacement. The insurer pays the Rs 3,40,000 hospitalisation bill. The medical corpus pays for what the insurer does not: eleven months of physiotherapy, imaging, mobility aids and follow-up consultations totalling Rs 1,45,000. Without the corpus that Rs 1,45,000 comes out of the general retirement corpus — money that was sized to fund twenty more years of ordinary living.

Chapter 8, section 8.1.3 (Planning for Medical Costs), is the source for all of this: the reason a corpus is needed, the four planning stages, the higher health inflation, and the three-years-in-liquid-assets rule at retirement.

Expect a direct question on why a medical corpus is needed when the retiree already has health insurance (because policies cover only hospitalisation and related expenses), on the starting point for estimating it (the average of the preceding 3 to 5 years' health costs), and on how much must sit in liquid assets at retirement (at least 3 years of expenses). The point that health inflation is typically much higher than general consumer inflation is a reliable one-liner. A case-study question may note that employer medical cover ceases on retirement except for government employees.

• A medical corpus is not health insurance, and neither replaces the other. Insurance pays hospitalisation; the corpus pays medicines, consultations, tests and long-term care.
• Three years in liquid assets, and the rest in low-risk products with better returns. Both halves of the instruction are examinable, and "the entire medical corpus in liquid assets" is wrong.
• 3 to 5 years is the look-back for estimating, 3 years is the liquidity holding. Two different uses of similar-looking numbers in the same section.
• Use a higher inflation rate for health costs. The workbook says health inflation is typically much higher than other consumer inflation; applying the general figure understates the need.
• Employer cover ceases on retirement — except for government and related agencies. The exception is stated in the workbook and options exploit it.
• Long-term care is part of this, not part of hospitalisation. It is named separately alongside medicines, doctors' visits and tests.
• A rough plan is the workbook's own recommendation. It concedes a definite plan may be impossible, so an answer that says health costs cannot be planned for at all is not the position the workbook takes.