Passive investing: ETF vs Mutual Funds. Do we have a new winner? (Data backed research)

[shammysdad]

TL;DR – For >1 year investments, lower annual costs for ETFs cover up for the higher transaction costs due to price-iNav mismatch.

Update: as /@zashmaster pointed out, Tracking error includes TER. So while calculating the annual costs - have only considered the tracking error now.

Hi everyone,

Basis last 5 yr market returns, and personal preference I have decided to move to passive investing. Now in passive investing, I had 2 options Index funds and ETFs. Have gone through this forum as well as numerous reports mentioning Index funds are better primarily due to liquidity available + lower transaction charges so decided to test it out.

In the last week, have monitored the iNav and prices for the top ETF (by volume) and the variance ranged from 0.05% (rest) to 0.25% (Nifty Next 50, IT, Bank) + the volumes were on par with a nifty 20 stock. This got me thinking, IF in fact ETFs could be the better choice. [iNAV to price difference was done manually, in case of better method, do share]

Have drawn up the summary of the Annual + Transaction charges for ETFs vs Index funds. DO let me know what you guys think about this and if I am doing this right

Assumptions:

1. Only the top ETF, Index fund considered – primarily by AUM + (for ETFs) daily txn volume
2. Annual costs include Total expense ratio (TER) + tracking error.
3. Tracking error would be negative for MFs – as they need to hold cash for redemption, and could be positive or negative for ETFs. Have assumed the worst case for ETFs with tracking error leading to lower annual returns – thus have added tracking error to the annual costs.
4. Transaction costs include STT + Stamp for MF, and STT + SEBI + Stamp + Transaction + Brokerage charges for ETFs.

a) Additionally, for ETFs, have added the spread as transaction cost – 0.25%*2 = 0.5% for JuniorBees, NETFIT, Bankbees and 0.05%*2 (buy&sell) = 0.1% for rest

b) For Mutual funds, given the recent changes, the NAV for SIPs would be at best T+2 (One click mandate T+2 settlement), thus we are losing out the returns for atleast those 2 days. Have added the impact of 2 lost days (assuming simple growth rate)

c) STT = 0.001%, Stamp = 0.005%, Brokerage = 0 (discount brokers), SEBI = 0.015%, Transaction = 0.0069%+GST

Analysis:

Index
ETF
Annual costs (tracking error only)
Transaction costs
Index Fund
Annual costs
Transaction costs

NIFTY 50
NIFTYBEES
0.18%
0.12%
UTI
0.30%
0.10%

Nifty Next 50
JUNIORBEES
0.22%
0.52%
ICICI Prud
0.45%
0.10%

Nifty Midcap 150
NETFMID150
0.63%
0.12%
MOSwal M150
1.10%
0.12%

Nifty Small Cap 150
N/A
-
-
MOSwal S250
1.07%
0.09%

Nifty IT
NETFIT
0.13%
0.32%
-
-
-

Nifty Bank
BANKBEES
0.86%
0.32%
MOSwal
2.20%
0.12%

Nasdaq 100
N100
0.21%
0.12%
MOSwal N100 FOF
0.64%


S&P 500
N/A
-
-
MOSwal S&P
0.49%


China
N/A
-
-
-
-
-

This clearly shows ETFs will make up for the higher transaction costs post 1 year of investments due to lower annual costs. And, for a 20 year period, this can add up to 5-15% in additional returns. Have I missed anything here?​

 

[mgreene05]

@shammysdad In my opinion we are splitting hairs here... In the end just one day of market volatility could swing this either way. I guess in the end more important is discipline and time in the market. In my personal opinion i prefer a uti nifty direct index fund from uti directly or mfutility just for the convenience to make it truly passive... My 2 annas...

 

[shammysdad]

@mgreene05 Fair enough. Just felt with SIP in ETFs coming in, and with time the ETF market is only going to improve so maybe starting retirement planning in ETFs is the right thing to do.

 

[mgreene05]

@shammysdad Well over 20 to thirty years or longer i will try to just pump in as much as possiple and then leave it alone. Buying etfs is just like you are trying to time the market which beats the whole point of being passive, doesnt it? I would focus more on earning and spending time on more useful things and pump the extra cash in without worrying about dips... Because in the decades to come these days or hours wont matter as all the studies have shown so far...

 

[shammysdad]

@mgreene05

Buying etfs is just like you are trying to time the market which beats the whole point of being passive

Understand your reasons, but why do you say buying etfs is trying to time the market? They can be scheduled monthly as well. It has an additional feature where we can buy in times of large dips - but that is just an additional option

 

[nickel]

@mgreene05 Lumpsump beats SIP according to the Vanguard study.

 

[mgreene05]

@nickel Exactly lumpsum beats sips and that was my point...

 

[aksumite]

@shammysdad

Not sure why the annual cost for UTI's Nifty 50 Index fund (Direct) is 0.40%, believe it is 0.10%. Similarly for ICICI's Next Nifty (Direct) it is 0.39% not 0.84%.

Perhaps, you have taken the expense ratio for regular funds, instead of direct. If so, Index funds at least for the two most popular passive funds still cost less than their corresponding ETFs.

 

[shammysdad]

@aksumite Have added the Tracking error as part of annual costs, as it leads to lower returns (same as expense ration)

 

[aksumite]

@shammysdad Ok. How much have you assumed tracking error/cost for UTI N50 & ICICI NN50 to be?

 

[shammysdad]

@aksumite Tracking error:

UTI N50 - 0.30%

ICICI NN50 - 0.45%

 

[aksumite]

@shammysdad This is a very good effort. Suggest, re-do the exercise for one - Nifty 50, lay out all your calculations/assumptions.
As for the Tracking Errors #s - not sure of them, suggest you show your sources for them too. Similarly what is Transaction costs, why is it 0.10% for N50, NN50? But, 0.12% for N Midcap?

 

[yosoyguapo]

@shammysdad I think if we buy etfs, the fund house buy shares that matching with index. That's include of tracking error.

 

[cupressi]

@shammysdad ETFs can declare dividends which are taxed at the slab rate (SBI Nifty 50 ETF has last week) whereas Index funds with the growth option will never declare dividends meaning gains are taxed at 15%/10% - STCG/LTCG.
This would really come down to a person's slab rate. Could snowball at the higher slab rates.

 

[wilddaisy]

@cupressi Wow. This is news to me.

 

[shammysdad]

@cupressi I didn't know about this. In fact, there is nothing in the news around this - are you sure? I was of the opinion the dividends are re-invested

 

[cupressi]

@shammysdad It's on sbimf.com, on the SBI - ETF NIFTY 50 page under the Historical NAV tab.

There is a note under the NAV.

Historic dividends are also shown at the bottom of the page.

 

[shammysdad]

@cupressi These dividends gets reinvested. No impact on taxes for investor

 

[cupressi]

@shammysdad The dividend announced last week was paid out today, can confirm that I received the dividend.

 

[cupressi]

@shammysdad Dividends received from the portfolio held by the ETF are reinvested, not dividends announced by the ETF itself.