The three reasons people buy insurance is:

a) To save tax.

b) As an investment, to make a good return on their money.

c) To feel good that one has some insurance or to get rid of that pesky uncle who keeps mentioning it.

The fourth — and perhaps most important — reason to buy insurance is to let your family be financially secure if you die. This is the only reason anything should be insured. Car insurance gives you money if your car has an accident, and covers costs for people you might injure. Home Fire insurance covers the damages in case there's a fire. You pay every year, and you're happy to not have to claim (because it means you've not had an accident or a fire!); and at the end, you don't get your money back.

Not so with Life insurance. The most policies bought are for the purpose of saving or investing, not for insurance. And that, further, is because Life Insurance is hardly ever bought, it's sold. The sellers get a fatter commission when they sell you a "saving" product, so you don't ever get to see the real insurance. "Pure Term" insurance is the only real deal: where your family gets paid if you die, and your premium is lost when you don't). Anything else, usually called ULIPs, Money-back, Endowment or Savings policies, involve a small amount of insurance and a higher degree of saving.

Even if it sounds like killing two birds with one cheque, you shouldn't mix investment and insurance — because you don't get enough of either. Take a 35 year old with a monthly salary of Rs. 50,000 and expenses of, say, Rs. 30,000. The minimum insurance expected would be about Rs. 1 crore; the idea is that you need your family to live another 40 years off the money, at a current return of around 8% risk-free and expenses rising at an inflation of 6%.

The cost of a "term" policy of Rs. 1 crore could be between Rs. 15,000 and Rs. 30,000 per year — or Rs. 1,500 to Rs. 2,500 per month, easily affordable. But agents find such policies unlikely to give them enough commissions, and they know that if they try, they can get the customer to pay Rs. 10,000 per month. A "ULIP" or an endowment plan with Rs. 10,000 per month as premium might give the buyer just Rs. 10-15 lakhs as insurance cover (typically 10x to 15x annual premium); a vastly inadequate sum compared to the 1 crore the person needs! But the seller persists and gets his way, largely because the customer has no idea how to work the metrics, and gets a feeling of happiness that there is some insurance and investment, when there really isn't.

In the longer term, I expect the tax-benefits of insurance to go away. There are two areas to this — first, insurance proceeds of any sort are tax free, even where the insurance cover is next to nothing and the product was primarily a product to save money. The second is a tax deduction on the amount invested every year, subject to an upper overall limit. Both are under threat in the longer term, as the government tries to find other means of raising revenue to meet increasing deficits. Additionally, it's untenable that long term savings of one nature — insurance or PF — are non-taxable, but buying long dated government bonds or (non-equity) mutual funds makes you pay tax on the gains. Lastly, if the government introduces a tax for inheritance (a proposal under discussion) then life insurance with a large one-time payment becomes an easy way to avoid such a tax; it is quite likely that the government will then plug the loophole by making "insurance as an investment" liable to tax.

In a decade, we are likely to see the tax-free exit status of many schemes vanish or dwindle, or at least force you to invest in low-yielding-annuities if you want to retain a tax advantage. Put another way: To assume that if I buy, I will not be charged a tax on exit even after 20 years is fraught with risk.

The last problem is that of complexity. Insurance products are incredibly complex, despite their heavy regulation. Financial products are typically of two types —high-risk, where the returns cannot be predicted in any reasonable manner, and low-risk, where the return is either guaranteed or specified (the risk is in whether the seller will go bust). Equity is a high-risk proposition, while fixed deposit and other debt options are the second. Insurance products provide a mix-and-match, with some products giving a vague guarantee with an additional potential upside (like 50% minimum guaranteed return or highest NAV in 10 years). Then they give you weird terms — you pay for five years, you can exit only after 10 years, the guarantee applies on the first seven years' NAV, and so on. And then, if you die, the insurance might pay out the guaranteed amount, the "sum assured", the amount that your investment has grown, or the lowest of all three. By the time you understand the terms and are able to calculate your real return, you might find it ridiculously low (if your brain hasn't turned to jelly). A case in point: the real return on that "50% in 10 years guaranteed" cases is just short of 5% per year, which is unacceptably low, even if you consider your taxes saved.

Most people give up before they reach the "real return" calculation — which is why insurers can easily stuff charges into such policies, knowing that if someone is silly enough to invest with a 5% real return, he won't even know that they can take a significant chunk of money as commissions. While we have seen charges that added up to 50% to 60% of the first few years of premium, even the lower 10% charges we see today are massive compared to the 1% to 3% that are charged by, say, mutual funds.

With the problem being that such products are sold — and sold hard — to customers, what we see in the Life insurance industry is more of industry and less of insurance. And as it increasingly sucks the blood out of unwary buyers, less of Life as well.