THE debate on retirement funds investing in National Savings Schemes (NSS) remains unsettled.

Many people have proposed an outright ban on such investments. Another recommendation seeks to forcibly redeem

all outstanding NSS securities issued to retirement funds by 2022. The voice of the salaried class has not been heard so far in this debate.

The proponents of the ban give four reasons. One, the NSS have a ‘disruptive effect’ because their profit rate is not linked to market rates. Two, the funds tied in these investments are hindering the development of capital markets. Three, the early redemption feature in the NSS is like a free lunch for investors. Four, the Central Directorate of National Savings (CDNS) requires supervision and the Debt Management Office should be given this responsibility. Let’s look at each of these reasons one by one.

Disruptive effect

Market interest rates in Pakistan are based on the sovereign yield curve that, in turn, depends on the monetary policy stance and supply-demand forces. The State Bank of Pakistan (SBP) has the capacity and tools to ensure that the yield curve trades within its acceptable range. The demand and supply of treasury bills and investment bonds predominantly come from banks and insurance companies and set the daily yield curve.

The CDNS simply follows the yield curve and revises its profit rates after monetary policy decisions. It is worth pointing out that the outstanding stock of the treasury bills and investment bonds is Rs17.46 trillion and that of NSS is Rs1.46tr i.e. 8.36 per cent of treasury bills and investment bonds. Only a portion of NSS is held by retirement funds.

The Central Directorate of National Savings can serve the dual purpose of delivering retirement income and raising long-term debt at a lower cost

Empirical data since December 2000 does not support the ‘disruptive effect’ argument. We can assess this in two ways. The first approach is to compare auction yields of Pakistan Investment Bonds (PIBs) of the equivalent tenor with the prevailing profit rates of the NSS. There have been 387 successful auctions of PIBs of the equivalent tenor. The second approach is to plot the daily yields of PIBs of the equivalent tenor and compare them to the prevalent NSS profit rates. Both methods show that the NSS profit rates were generally lower than the yield curve and fall and rise in tandem with the PIB yields.

Development of capital markets

The capital market development argument is also based on a flawed premise. It assumes that a ban on retirement funds investing in the NSS will increase their investment in shares and term finance certificates (TFCs). This is not going to happen because retirement funds will not take additional risks. They will simply redirect their funds into treasury bills and PIBs. The government will then have to issue more treasury bills and PIBs to replace the decline in borrowing through the NSS.

Free lunch

The free-lunch argument is based on the early redemption feature — the embedded put option — which provides a floor on the value of NSS securities. This embedded put option is characterised as free lunch for retirement funds. What is perhaps being ignored is that the value of NSS securities does not increase unlike PIBs when interest rates fall.

There is a general presumption that it is undeserved and it has a huge cost to the federal government. Unfortunately, the cost has not been quantified on any public forum so far.

Retirement funds are non-profit entities where monthly savings of workers and employees are pooled together. All profit-making corporations have already been banned from such investments since 2003. A ban on retirement funds investing in the NSS effectively means a ban on the salaried class. The vast majority of such people do not have enough disposable income to separately save and individually invest in these securities.

We also need to consider the overall context. Our government does not provide — and will not be able to provide — a decent pension to its citizens. Almost all developed countries provide a reasonable pension where 100pc of the investment, inflation, reinvestment and longevity risks are absorbed by the government. What is wrong if the government provides this redemption feature to the salaried class?

If a rigorous and transparent analysis does in fact show that the embedded put option is unreasonably expensive, then we can pursue less disorderly solutions. One would be to increase the existing penalty rates on early redemptions. Another one would be to ban early redemption but at the same time allow the transfer of NSS securities to other retirement funds.

To frame this analysis and debate as a zero-sum conflict between the government and retirement funds is nonsensical. The gain of public at large is not a loss to the government.

Role of the CDNS

The major societal benefit of the CDNS is that the public can capture 100pc of the real return on sovereign debt securities. An ordinary investor cannot directly buy treasury bills and PIBs. The alternative of going through mutual funds to capture the real return on five-year and 10-year PIBs will wipe out 40-50pc of the real return.

The looming challenge of delivering safe income to retirees is growing globally by the day. New solutions are emerging. Only the CDNS has the necessary infrastructure and distribution capability to deliver these critical solutions.

For example, the CDNS can issue a 25-year retirement bond to retirees at 60 years of age: 4pc of the original principal can be returned annually adjusted for inflation (or deflation) plus 1pc real return. There will be a lifetime maximum investment cap of Rs30 million. The retirement bond can be inheritable, but not redeemable or transferable. It can be issued at the start of each quarter to simplify operational hassles.

The benefits to the retirees and the government of such a product are obvious. The retirees can avoid investment blunders — dubious real estate projects, speculative shares and scams. They will get a secure cost of living–adjusted income in the most vulnerable part of their life. The government will be able to offer a retirement solution on a pre-funded basis. Its cost of long-term debt will fall significantly.

Historically, 10-year PIBs and Defence Savings Certificate have provided a median real yield of 2.71pc and 2.66pc, respectively. It stands to reason that the retirement bond as outlined above can be issued at a significantly lower real yield. Another advantage is that the government will be able to continuously raise long-term debt without the hindrance of prevailing high inflation rate and PIB yields.

The CDNS is a unique public institution of strategic value whose importance will grow over time. It can serve the dual purpose of delivering retirement income and raising long-term debt at a lower cost. Its potential should not be clipped under a different mandate of the newly constituted Debt Management Office.

The writer is CEO of Magnus Investment Advisors Ltd. He and the company have no remunerative relationship or family ties with the CDNS.

Published in Dawn, The Business and Finance Weekly, June 29th, 2020