About Target Maturity Debt Funds

This week will see the NFO of a target maturity debt fund, that has also been billed as India’s first debt index fund.  Open-end target maturity debt funds are a product category that merits serious consideration, and one that we should see a lot more of.  However, based on some of the commentary that I have seen in the media (mainstream and social), as well as what I’ve been hearing from advisors and investors, I feel that some of the wider understanding could be enhanced, and expectations tempered. In this post, I want to touch upon a couple of less-talked-about points.

It might be worth mentioning that open-end target maturity debt funds were first launched in India over two decades ago.  Back in the day, they used to be referred to as ‘serial plans’.  Unfortunately, this category (like some others) didn’t quite take off.  Perhaps it was an idea whose time had yet to come.

The first point that I want to bring up is that, conceptually speaking, such a fund doesn’t have to be an index fund or an ETF and even if it is structured as such, it may not be purely passively managed.  From what I have gathered, unlike equity index funds and ETFs, it is quite common for bond index funds and ETFs worldwide to not replicate the complete set of index constituents.  Instead, they follow what is referred to as a ‘sampling’ approach whereby the fund manager seeks to match the more fundamental characteristics of the index such as credit profile, duration, and yield.  To be fair, in India, SEBI has put in guidelines that limit the deviation from index constituents.  But it still leaves room for some degree of active bond selection. 

As a point of contrast, take the original ‘serial plans’. These were typically single-security gilt funds whose maturity coincided with the maturity of the underlying security.  Consequently, despite not being index funds, they were perfectly passively managed.  The same can also be said for any fund that follows a hard-coded list of issuers, and a pre-defined credit and maturity profile.  Thus, a part of me wonders if the index structure currently in vogue, might just be a way to work around the inflexibility of SEBI’s open-end scheme categorization.  Regardless, with the growing interest from investors, advisors, as well as fund houses, it may be worthwhile for SEBI to expand the existing scheme classification to include this category. 

The second point that I want to talk about is the predictability of return in these funds, as represented by the yields that are reported.  No doubt, this is a prominent reason to invest in these funds.  However, investors need to be careful in relying upon the disclosed yield.  For one, there is the question of how the yield might be impacted if there are large flows into or out of the fund.   For another, there is the extent of the gap between the maturity of the fund’s underlying securities and the actual maturity of the fund.  As an example, in the case of one existing fund, I noticed that around 10% of the portfolio is scheduled to mature 7 months or more, before the actual maturity of the fund.  As and when those bonds mature, it is a moot question as to whether the fund will be able to earn the same yield as what it is getting on those bonds today. 

In this context, it might be worth looking at the evidence shared by BlackRock in the US, with respect to its target maturity iBonds ETFs that have matured.  Among other things, it shows the difference between the initial net yield and the final return that investors got.  So far, the difference across all matured funds has ranged from +0.19% pa to -0.30% pa.  In absolute terms, that may or may not seem much but when seen relative to the yields (and the time horizon), that is certainly not a small difference.  For example, one of its ETFs started off in December 2014 with a net yield of 2.64% pa.  When it matured 6 years later, the total return to investors was 2.35% pa.

As a side note, I don’t know what to make of the fact that SEBI doesn’t allow fund houses to offer any indicative yield on FMPs yet it seemingly has no issue with the disclosure of yields on open-end target maturity funds.  All things equal, FMPs offer a more predictable return than these funds.

Yield vs. Return

An investor reached out to me with questions related to a certain floating rate fund.  The fund had reported its May-end net portfolio yield (i.e. after expenses) to be ~6.5% p.a..  How was it then, he asked me, that over the month of June, it had shown an annualized return of ~23% p.a.?  If it was, as he suspected, somewhat linked to the last round of rate cuts by RBI, then why was it that other funds that had relatively higher yields as well as higher duration, hadn’t gained as much as this fund?

His suspicion was not without merit.  Such a difference in return over yield is usually seen at times when RBI cuts interest rates.  It’s just that, this time around, bond yields have been taking their time to adjust to the cuts.  While RBI cut rates in March and May, some of the expected fall in yields played out only over June. 

However, there are a couple of things that the investor overlooked.

Firstly, the fall in yields in June wasn’t uniformly across the yield curve, or issuers, and there were some striking differences in returns across funds and fund categories.  For instance, the category that gave the best return was that of corporate bond funds yet the best performing fund across categories was a credit risk fund.  While the categories of short and medium duration funds did better than ultra short and low duration funds, gilt funds, despite their much longer duration, gave lesser return than all these categories.  In a nutshell, it boiled down to the individual securities that a fund was holding. 

In that light, if I look at the May-end portfolio of the aforementioned floating rating fund, some part of its extraordinary gains in June would have been driven by a ~100 bps fall in yields of two AAA bonds, with 3-4 years left to their maturity, and which made up ~10% of the portfolio.  In addition, some gains would have resulted from a 30 to 55 bps fall in the yields of perpetual bonds that the scheme was holding, and which made up ~18% of the portfolio.  In the aftermath of what happened with Yes Bank, across-the-board, yields of perpetual bonds shot up and have remained high ever since.  Only in June did those yields show signs of softening.

I must point out that all that I said just now assumes that those securities continue to be held in the fund’s portfolio.  We will know for sure, once the June-end portfolio is disclosed.

Secondly, at a more conceptual level, while the portfolio yield of a debt fund is the single most important indicator that an investor should examine and track, it has its own limitations.  Yes, it can give clues as to the level of credit risk in a fund, as well as roughly indicate what might be the immediate return that one could expect.  However, yields change daily whereas fund houses disclose the yields only at the end of each month.  Thus, in a period of sharp changes in yields, the last month-end yield will be of limited use.

In addition, if a fund holds illiquid or thinly traded securities, then the fund’s yield may not be a true measure of how the market may value its portfolio.  Selling those securities can well bring about gains or losses to their portfolio values.

To illustrate this point, I’d like to take the example of a fund whose reported net yield as of May-end was ~12% p.a. but which ended up giving a negative return over June, and not because of any markdown or default.

This scheme held a certain security that alone made up almost a third of the portfolio, and which was valued at an estimated yield of ~16% p.a..  For some reason, the fund manager chose to sell some of that security in June.  As it happens, the sale was made at a yield of 28% p.a..  In other words, the fund got a lot less money than what the security had been valued at.  That led to the NAV falling by ~1.5% that day, and that alone was enough to wipe out whatever gains the reported yield would have brought about over the entire month.

To be fair, this may be an extreme example.  Still, it should help drive home the perils of blindly relying on a  fund’s portfolio yield.

Baffling Trades In ICICI Securities

Why would a fund manager buy shares of ICICI Securities in the IPO and then sell them off at a loss, two months later? 

As many of us would know, since its IPO in March, the stock price of ICICI securities has seen a sustained fall.  At no point, thus far, has the price come back to its IPO price of 520.  If I take the month of May in particular, the price ranged between a high of 421 and a low of 352.  Yet in that same month, fund managers across 10 9 schemes that acquired the stock in the IPO, brought down those holdings or exited them completely.  All put together, these fund managers sold off shares worth 96 91 crore at the time of the IPO, at a loss of somewhere between 19% to 32%.

The table below lists the schemes that took this hit.

Schemes that sold shares of ICICI Securities in May 2018

	No of shares bought in IPO Price: 520	No of shares sold in May   Price: 352-421
HDFC TaxSaver	981,120	508,500
Axis Long Term Equity Fund	480,788	480,788
Reliance Growth Fund	288,456	288,456
Reliance Banking Fund	192,304	192,304
Edelweiss Maiden Opportunities Fund-1	96,180	96,180
UTI Multi Cap Fund	96,152	96,152
Edelweiss Long Term Equity Fund	96,152	86,183
Kotak Bluechip Fund	63,448	63,448
Kotak Equity Savings Fund	30,744	30,744
UTI Banking and Financial Services Fund	288,484	3,957

Data Sources: BSE, NSE, RupeeVest.

The way I see it, a couple of months is just too short a period for a long term investor to have drastically changed one’s view on this stock.  So what else could explain the actions of these fund managers? 

One view is that the fund managers may have been forced to do this on account of scheme reclassification.  That doesn’t make sense to me because the stock fitted comfortably into the portfolios of all of the schemes on the list above.  Another view is that this might have been done to meet redemptions.  But the extent to which most of these fund managers reduced their positions makes me doubt that.  As it happens, one of the schemes on the list is a closed-end scheme while three others are ELSS.  A third view is that the fund managers may have decided to cut their losses.  While that’s not implausible, it strikes me as an approach that a trader would take and not something that a fund manager would do. 

As I took a closer look at the numbers, something else emerged.  This pertains to three two fund houses whose schemes are listed above: Kotak Mahindra MF, Reliance MF and  UTI MF.  It turns out that in the same month, while their schemes listed above reduced or exited their holdings, there were other schemes managed by these fund houses where the exposure to ICICI Securities was increased.  In the case of the latter two fund houses, the shares sold in one scheme were identical to the shares bought in another scheme, suggesting the possibility that these might be inter-scheme transfers.

Fund houses that took contradictory action on ICICI Securities in May 2018

	Action taken	No of shares
Kotak Mahindra MF
Kotak Bluechip Fund	Sale	63,448
Kotak Equity Savings Fund	Sale	30,744
Kotak Emerging Equity Scheme	Purchase	154,828
Reliance MF
Reliance Growth Fund	Sale	288,456
Reliance Banking Fund	Sale	192,304
Reliance Focused Equity Fund	Purchase	192,304
UTI MF
UTI Banking and Financial Services Fund	Sale	3,957
UTI Multi Cap Fund	Sale	96,152
UTI Value Opportunities Fund	Purchase	96,152

Data Source: RupeeVest.

Frankly, I can’t think of any good reason why a fund house would have sold its loss-making investments in one scheme only to buy those shares in another scheme. 

In fact, looking at all of this, makes me question the credentials of the concerned fund houses and fund managers to manage long-term investments.  I came upon all of this information, quite by accident.  But now I wonder that beyond ICICI Securities, where else may something like this have happened, and how often it might have happened.  I guess that unless these fund houses/ fund managers decide to open up about this, we may not know.  Personally, I think that investors in these schemes should press hard for answers.

Correction: The original version of the post incorrectly identified UTI Multi Cap as a fund that had sold shares of ICICI Securities.  That scheme was merged into UTI Value Opportunities Fund.  I apologize for the inaccuracy.

Total Return Index Benchmarking

Around six weeks ago, DSP BlackRock MF stated its intention to “compare its funds’ performance” to the total return of their respective benchmark indices.  A few days later, Edelweiss MF made a similar announcement.  While these were noteworthy decisions, I couldn’t quite understand the need for these fund houses to formally proclaim their intentions to the world at large.  To be honest, a part of me felt that these announcements had a holier-than-thou ring about them.  More than that, though, it was the wide media coverage of these announcements that stood out for me.  Sure, it was helpful in spreading awareness about the difference between a Total Return Index (TRI) and a Price Return Index (PRI).  But beyond that, it seemed to me that most of the coverage was hype and lacked clear perspective.  In this post, I’d like to chip in with some scattered thoughts.

TRI benchmarking has little to do with appropriateness
By any reasonable standard, the appropriateness of a scheme’s benchmark index is determined by the similarity of the constituents of the benchmark with the universe of securities that the scheme will invest into.  So, for a scheme that invests in large cap stocks, the Nifty 50 or the BSE Sensex could both be appropriate benchmarks.  It is a fund house’s individual decision as to which index to choose as the formal benchmark.  As for choosing a TRI over a PRI, that has little to do with appropriateness in that sense: it is more like raising the height of a hurdle to jump over. If a fund house chooses a TRI over a PRI, then at best it can be assumed to be signalling its intent to raise the bar for its performance.  Bear in mind, though, that a fund house can raise the bar for its performance in any number of ways, and without making such announcements. 

Where will performance be reported?
Even if a fund house chooses TRI benchmarking, its performance reporting will be restricted to documents that are released by it (e.g. fact sheets, SID, KIM etc.).  So, to investors who prefer to compare scheme performances across multiple fund houses, such reporting will be of little use. 

Investors can choose their own benchmark
No matter what benchmark index a fund house chooses, in analyzing the performance of a scheme, an investor is always free to use whatever benchmark index he or she thinks appropriate.  For example, in all instances of index data used by me on both my blogs, I have used total return numbers only.  It is not as if that is necessarily better than using price return numbers: it is merely an expression of my personal belief.  And I have done so, regardless of what practice the industry or individual fund houses have followed.

The notion of “alpha” varies
There are some who argue that a TRI is a better choice for accurately determining “alpha”.  In my opinion, a lot depends upon how one defines “alpha”.  For those who regard “alpha” as a mathematical formula, it is understandable for them to prefer a TRI over a PRI.  But in my experience, for most people who look for “alpha”, it is just a broad appraisement of the value added by a fund manager.  Sure enough, some of these people assess “alpha” against a TRI but there are many to whom the TRI/ PRI choice may be inconsequential.  There are also those who believe in assessing “alpha” against an appropriate index fund (rather than an index) while there are yet others who, in their own wisdom, assess “alpha” against the returns from a bank deposit.

TRI benchmarking is not new in India
In conversations with investors and advisors, I gathered that quite a few of them were under the impression that these announcements were pioneering/ ground-breaking in the context of the Indian fund industry.  In fact, there was at least one report, in The Economic Times, which credited DSPBR MF as “the first fund house to start with this practice”.  The truth is that Quantum MF has been following TRI benchmarking for years, for both its equity schemes, as well for its Nifty ETF.  Other fund houses following this practice selectively include IDBI MF and Kotak MF (thanks to one scheme taken over from the erstwhile Pinebridge MF).  And then there are some fund houses that have taken a somewhat convoluted approach, if I may call it that.  To take an example, SBI MF compares the returns of its Nifty Index Fund with the Nifty 50 PRI, but measures its tracking error against the Nifty 50 TRI.  As far as I am aware, none of these fund houses trumpeted their decisions and perhaps, for that reason, these haven’t attracted the media coverage that the more recent announcements have.