Should pension assets be invested in equities?

There was a time when the dominant viewpoint in the world was that "pension assets should be completely safe, so they should be entirely invested in government bonds". This view is still found fairly often in India. This view is essentially wrong. However, there are some subtle issues in thinking about equity investments, which we should be careful about.

What are the maximal long-run returns on government bonds? Ultimately, the government can only fund itself by taxes, and tax rates cannot grow. Hence, in the long-run, the revenue of government can only grow at a rate of growth of the wage bill of the country.

This gives us a number in India of the order of five percent in real terms. The long-run growth of the wage bill in India is probably around five percent, which should be our estimate for the long-run return which can be obtained by investing in government securities.

In contrast, if pension assets are invested in productive assets, then the long-run return on pension assets would equal the marginal product of capital. In India, this is probably around ten percent in real terms. Hence, the long-run return on productive assets will handsomely dominate the return on government securities.

When this idea is applied on the scale of decades, the results are quite startling. A one percent improvement in the rate of return implies an improvement in the terminal wealth of the worker of around 30%. Hence, every small improvement in the rate of return is worth worrying about.

So far, we have talked about "the marginal product of capital". In reality, capital is deployed into productive uses through either corporate bonds or equities. Equities involve greater volatility, and earn higher returns as a consequence: this risk premium is called "the equity premium". In India, the equity premium has proved to be roughly eight percentage points. That is, the long-run return on Nifty is around eight percentage points higher than the long-run return on government bonds.

This is an enormous difference from the perspective of the multi-decade horizons involved in pension investment. Hence, thinkers about the pension system all over the world have increasingly focused on the role for equity investment. The equity premium is directly implementable as an investment strategy: using an index fund. Index funds are safer, and involve lower fees. For this reason, the Dave Committee recommended that pension investment in equities should only be done using index funds on the NSE-50 or the BSE-100 indexes.

Criticism. The central criticism of this position is based on the question of equity volatility. Yes, long-run average returns on equities are strongly superior to investing in government bonds. But equity returns are volatile. Hence, equity investment imposes investment risk upon the individual. How should we confront this problem? There are 3 strategies which can be adopted.

Strategy 1 : Do nothing. In a recent paper Investment strategies on multi-decade horizons, Susan Thomas finds that the probability of underperforming while invested in Nifty drops to near-zero levels when we deal with horizons in excess of ten years. Hence, for a person who plans to retire at age 60, and is presently younger than 50 years, it makes sense to constantly invest in a Nifty index fund and do nothing about the volatility.

Strategy 2 : Phase out equities in old age. As retirement approaches, the worker can gradually phase out his equity investment. For example, the worker can move from 100% equities at age 50 to 90% equities at age 51, all the way to 0% equities at age 60. This strategy proves to greatly reduce the variability of wealth at age 60.

Strategy 3 : Financial engineering. A sophisticated answer can be obtained using financial derivatives. Put options are insurance: they give protection against declines below the strike price of the put option. Hence, the portfolio can be insured by purchasing put options. How can this be financed? The money required for buying these puts can be obtained by selling off some of the up-side potential, i.e. by selling call options. This gives the worker a "collar", a range inside which he is guaranteed to be.

All these three strategies can be used depending on the situation. Nifty options do not yet exist in India, but it is expected that trading will commence later this year. They could have a profound impact upon the willingness of many individuals to invest in equities, by allowing such a possibility of insurance.

The main problem with these strategies lies in the fact that life-spans have been elongating. Consider a person who is at age 60 and expects to live to age 80. In this case, the investment horizon at age 60 is twenty years long, which is well in excess of the ten--year rule mentioned above for Nifty. Hence, the traditional idea of moving into riskless assets at age 60, i.e. buying an annuity which invests in government bonds, is actually quite sub-optimal.

What is the alternative? The modern answer appears to be as follows. First, it makes sense to be heavily invested in equities all the way to age 60. Second, at age 60, the worker should not switch into a conventional annuity (which buys government bonds and pays a pension until death). Instead, the concept of a "variable annuity" should be applied. The variable annuity continues to be invested in equities, and gives the worker some of the equity upside, which is quite considerable on the twenty-year horizon.

In all, equity investment is having a profound impact upon old-age planning. The equity premium makes a huge difference to income security in old age. There are millions of workers in India who cannot save more than Rs.3 per day or Rs.5 per day. If their contributions are invested into government securities, they face a certainty of being destitute in old age. Instead, if their contributions are invested into equities, they have a high chance (though not a certainty) of escaping poverty in old age. This is an important, and new, opportunity in poverty alleviation in India.

There is one final catch in this happy picture. Many economists worry about the size of the equity premium. Standard economic models have been unable to rationalise the size of the equity premium; numbers like eight percentage points are simply too high to explain. Some economists fear that the equity premium is so high because millions of households are too ignorant to buy index funds, and that this scarcity generates excessive returns to equity. NSDL now houses almost all the equity in India, and NSDL has only 3 million accounts. Once a modern pension system comes about, and hundreds of millions of households in India indirectly buy Nifty index funds, many economists expect that the equity premium will then decline when compared with its present levels. Hence, we may be the last generation to enjoy an equity premium as high as eight percentage points; 25 years from now it could be as small as half this value. However, as of today, this argument does not apply, so the hardy pioneers in buying the Nifty index fund can enjoy the inexplicably large equity premium.

Back up to Ajay Shah's BS column