Actionable insights on equities, fixed-income, macros and personal finance Start 14-Days Free Trial
Actionable investing insights Get Free Trial
Mutual Funds

How Debt Funds Bluff Regulations and What SEBI Can Do To Fix It


We wrote about the Franklin saga in debt funds (Read post) where six of their funds have been shut down since. The problem was:

  • There have been a lot of redemptions, with some of these funds losing over half their AUM
  • The funds have to borrow money to meet redemptions, if they can’t sell enough bonds
  • The interest on the borrowed money is paid partly by the fund itself (meaning by all unitholders eventually) but only until the yield of the bonds of the fund
  • Anything more is paid by the AMC
  • So if a fund has to pay 8% to borrow 1000 cr. but the yield of the fund is 6%, then Rs. 60 cr. is charged to the fund, but Rs. 20 cr. (the extra interest above the yield) is paid by the AMC from its own pocket. (Assuming a year of borrowing – it would typically be a few months, so the impact is lower)
  • The Franklin AMC, through its funds, has borrowed significant amounts to meet redemptions, so they will be seeing interest payouts.
  • This is why they’ve shuttered redemptions. They’ll not need to borrow more- but won’t get any money either.

In the process of doing so, they’ve created fear in unitholders of a lot of mutual funds, many of who have decided to sell some of their funds. The run exposes a lot of other practices that seem to be doing the rounds. And it’s not just Franklin.  Nearly every fund house with debt schemes are using complex mechanisms to skirt regulations. And, while legal, these are not really in the investors’ favour.

In this post, we won’t single out Franklin. But we’ll show you things we feel aren’t right. And also give suggestions for what can be done to fix it.

A quick summary of the issues:

  • Floating rate bonds are used by Mutual Funds to make long term bonds appear as short term ones.
  • Additional Tier 1 Perpetual bonds are bought as if they are short term debt instruments but there is no guarantee of any repayment of principal
  • Unlisted bonds that cannot be traded
  • Zero coupon bonds from companies with no income (only collateral)

Note: SEBI has set up some regulation as mentioned above. See our post. (No more AT1 Bonds, says SEBI)

The Jugaad Duration: Using Floating Rate Bonds?

We’ll start with Franklin. We got a lot of queries looking at this image in our last post:

Franklin maturity


How come we have so many bonds maturing in 2021, 2022 and even 2029?

Remember an ultra short term fund should have a Macaulay duration of less than 6 months. The Macaulay duration is simply put, the average time to maturity for all the bonds in the portfolio.

It didn’t cross our mind much. But how does this fund show a duration of less than 6 months?

The answer seems to lie in a look at the 2029 bond.  How do you get a 2029 bond – which has 9 years left to maturity – into an ultra short term fund that should have an average maturity of 6 months?

There’s a way. Some bonds in the fund are floating-rate. The 2029 bond too is one of them.  Meaning: the interest rate is not fixed.

How Debt Funds Bluff Regulations and What SEBI Can Do To Fix It


Now this sounds like it meets the letter of the regulation but not the spirit. What is the “time to maturity” of a floating rate bond?

The answer, muddied by a lot of unncessarily complex mathematics, is apparently “time to the next reset”. This is because, people who designed the math, were using Macaulay Duration to find a calculation of the  price of the bond, where this definition makes sense. But if you really want to know when you’ll get your money back, you should not use this definition.

The real thing that matters to you, in terms of getting your money back, is the actual time to maturity.

But mutual funds will take the Macaulay Duration definition to their best, and the 2029 bond can now be looked at as a 6 month bond!

This is a wrong way to look at regulations. If you can’t get your money back in 9 more years, the calculation should use 9 years as the number.

(Note that Franklin now maintains there is a put or call option on this bond – read here – but we found no such option on the bond’s rating documents or in the NSDL bond details. Given that such documents are public, there is no such put or call option, in our opinion)

The Problem, Simplified

Mutual funds need to invest in specific average maturity instruments to be called short term or ultra short term etc. They instead repackage a long term bond as a short term one, using a floating rate bond which resets interest rate payable, say, every six months, and that will meet the regulatory norms.

What SEBI Can Do To Fix:

  • Bonds with “floating rate” should use only maturity date for the duration calculation.
  • Clarity: Figure out when you will actually get paid – and the average of that should be less than 6 months for ultra short term etc.

The Put/Call Option Type Bonds and AT1 Bonds in Short Term Portfolios

Typically bonds only pay regular interest, but the principal is paid at maturity.

A put option allows the bond holder to say give me my money (principal plus interest) now, and the issuer has to pay.  Think of this as a bank fixed deposit, that you can break without a penalty.

A call option means the issuer can tell the bond holder, here, take your money (principal plus interest). And go away. And the bondholder has to, even if he wants to continue. Think of it like a great Fixed Deposit that the bank breaks on its own and you can’t get back the same interest rate, but maybe rebook at a lower rate.

Put and call options are available only in a few bonds. Some have only calls. Some only puts. And even those are exercisable only on or after specific dates. That way it’s not like get out whenever you want – but only at specified dates.

Put option is good for a buyer in terms of figuring out when you’ll get paid. Call option isn’t.

Now let’s look at this portfolio of Kotak Low Duration Fund which needs its maturity to be between 6 months to a year. Kotak Low Duration Fund AT1 Bonds

They have about 2.85% in Syndicate Bank AT1 bonds.

AT1 are Additional Tier 1 Perpetual bonds. Recently, they were in the news because in Yes Bank’s case, the AT1 bonds were written down to zero. (Read our post)

These bonds are perpetual. The bank need not repay the principal, ever. The bonds have call options  usually after 10 years, which allow the bank to pay back principal and cancel the bonds. The bank doesn’t have to do it.

Given that, why would such a perpetual bond be allowed in a Low Duration fund? It would be a “forever” bond. But the current regulations allow mutual funds to take this as a short term bond!

Again, this is a way to technically comply with the regulation, but it really is a bond which you can’t get paid, so it may not meet the objective of the regulation itself.

How SEBI can Fix This

  • Disallow bonds with only Call options to be used in short term bonds, or use maturities as 100 years ahead
  • Disallow Perpetual or AT1 bonds in any duration based mutual funds, or if they exist, they should be considered as if maturity was 100 years ahead.
  • Allow AT1 bonds in only non-duration based schemes – like Corporate bond, Credit Risk or Dynamic schemes.

Note: On Tuesday, a truckload of AT1 bonds were sold in the market at very low prices (high yields) such as 14% for Punjab National Bank etc. This might have been because of Mutual Funds offloading them.

Unlisted Bonds

A bunch of funds have bought bonds that are not listed on stock exchanges. Unlisted bonds were disallowed by SEBI fo in October 2019, but funds were allowed to continue to hold them till maturity. Just yesterday, SEBI changed this rule slightly – that mutual funds can sell unlisted bonds to each other.

The problem? Unlisted bonds don’t have disclosure requirements, so you can’t find out much about them – but mutual funds have only been prevented from buying any new such bonds. The current bonds in their portfolio remain, and because of the rule, other mutual funds will not want to buy them. Yesterday’s rule change will only ensure that one scheme can transfer it to another one

See, for instance, the Nippon India Low Duration Fund.

Nippon LD Unlisted

The Panchshil Corporate Park debenture is actually maturing in 2024 – a good four years later. This doesn’t sound like a great fit for a low duration fund which has to have average maturity of less than one year. While the fund couldn’t sell this debenture, they might be able to transfer it to one of their other funds.

Also see ICICI Prudential Ultra Short Term Fund:

ICICI PRU UST Unlisted Bonds

See that familiar Pune-Solapur Expressway Bond? That was in Franklin as  well, and it’s a 2029 bond. It’s unlisted. What are these doing in an ultra-short-term fund? And now, because it’s unlisted, getting rid of it is near impossible.

How SEBI can Address This:

  • Unlisted bonds will remain on fund books for a long time without liquidity and if there are exits, they can’t be sold easily, so will increase as a percentage of assets.
  • So, SEBI should allow all funds to segregate unlisted bonds into a segregated portfolio.

Zero Coupon Bonds Rolled Over Forever, Backed by Shares

Look a little upwards at the ICICI Prudentail Ultra Short Term fund image.

The “Zero coupon bonds” section has this one entry: Adarsh Advisory Services. What’s this?

  • The rating document (click here) says this is a company owned by JSW steel promoters
  • The issue is backed by JSW Steel  and JSW Energy shares
  • The collateral was about 2x the borrowing
  • The company itself – Adarsh – has a negative net worth and is loss making, so it can’t pay.
  • There is no interest payable until maturity!
  • The shares have now fallen by 50% each
  • The company has paid back about half the amount, but the remaining half is still pending

The problem is this: mutual funds lend to a company that has no cash flows, only on the basis of shares as collateral. This was a huge problem in the case of Zee promoter entities, which we warned about first in 2015! This turned sour, horribly, landing losses to investors in a number of funds who held the bonds of the promoter companies – as investors in Aditya Birla funds, to HDFC Funds to ICICI funds, people saw potential losses. Why? Because the share prices fell sharply, and when the funds tried to sell to recover their money, the prices fell even more, and there was no other way to recover any money as the underlying companies weren’t profitable or even had cash flow.

Given that we have seen this recently, should funds not be more careful? SEBI has already increased the cover amount to 4 times now and this should be good enough for now. However given that there is collateral but no cash flows or creditworthiness (negative net worth!) the underlying bond is at the mercy of stock prices, with no knowledge of the end-use of the funds at all.

Read: Using Pledged Shares to Sell Bonds is Risky!

What SEBI can do:

  • Force funds to monitor end-use of the funds thus raised, to follow the money.
  • Such bonds should not be allowed in anything other than credit risk funds.

Funds Borrowing Money, And Then Hitting A Wall

When you redeem fund what happens?

  • Funds may have cash or get more incoming cash (new buys, interest payments)
  • They will use that cash to pay out
  • When that runs out, they will try to sell what can be sold at a decent price
  • If that can’t work, they will try to sell at indecent prices (but not TOO indecent)
  • Then they’ll borrow money against their bonds
  • When those bonds mature – could be a few days or weeks away – the money is used to pay back the loan
  • But there are even more redemptions, so what to do?

The answer, for liquid funds, was to force them to buy a certain number of government securities that are always liquid.

The problem is in other debt funds. Especially where you can exit every single day. You can’t avoid this problem, but maybe there’s a solution.

What SEBI can do:

  • If you have open ended debt funds where you can redeem every day, then 10-15% of the money must be invested in government bonds/T-Bills with less than three months to maturity.
  • Otherwise, debt funds should allow redemptions only once or twice a month, or even at a longer duration. Invest when you like, of course. This can be tweaked further if there is abuse at edge cases.

How This Will End

This will end, in some way or the other. But see this:

  • Debt funds are considered to be “safe”.
  • Mutual fund managers have leeway to buy what they want. They can even structure bonds along with the issuer, so that it meets their regulatory norms
  • People invest in shorter term, or ultra-short-term funds to park money for the short term with slightly better yields than fixed deposits
  • They believe that their funds, too, are investing in the shorter term paper.
  • The above show there is scope for side-stepping regulations, by repackaging longer term bonds as short term through floating rate issuances, by using call options on perpetual bonds etc. which go against the spirit of such funds and the regulation around them
  • Regulators should continue to allow risk, but move it to funds where the risk is apparent, rather than in shorter term funds.

Every crisis is an opportunity to reform and make things better. More regulations won’t help, but at this point we have an entire market that is nearly frozen because of massive redemptions. And those redemptions are coming around because of lack of trust. And the lack of trust is because some funds have tried to get around the rules by using innovative financial machinations. Which have now come back to bite, and so some have shuttered redemptions completely.

Note: SEBI has set up some regulation as mentioned above. See our post. (No more AT1 Bonds, says SEBI)

Thanks to Dheeraj Singh for the inputs on this post and the mutual fund industry. 


Like our content? Join Capitalmind Premium.

  • Equity, fixed income, macro and personal finance research
  • Model equity and fixed-income portfolios
  • Exclusive apps, tutorials, and member community
Subscribe Now Or start with a free-trial