WisdomTree 90/60 US Efficient Core Fund [NTSX] (formerly US Balanced Fund)

[FundQuant]

70% NTSX, 15% KMLM, 15% gold, vs NTSX, vs SPY. The real use is to diversify.

Why pay 17 bps extra for NTSX when 70% VTI + 15% GOLDX + 15% KMLM gets you the exact same thing? You don't need NTSX to add the extra diversifiers. It's just a drag on returns because of the added fees. Granted, 17 bps isn't the end of the world, but it's free money for those who understand that NTSX isn't adding any extra.

[Lawyered_]

That's not the proper benchmark at all. It's a high vol fund because of the leverage. You can't compare a 12% vol fund to a 19% vol fund and claim they're remotely similar when the track record clearly shows that NTSX is nothing more than a slightly different performing version of VTI. And that makes absolute sense because the leverage of a 90/60 ETF with this structure means it's going to have similar vol characteristics to a pure equity fund.

But yeah, smart people sometimes get duped by clever marketing, which is clearly the case here.

It's almost as if the track record over a seven year period is going to tell a story of that particular period! It's too bad we don't have any way of checking other periods...

Where are you running that backtest? I am not getting the same result. See below:

https://testfol.io/?s=kDTIMqvU8pD

[FundQuant]

It's almost as if the track record over a seven year period is going to tell a story of that particular period!

It's almost as if you took the most anomalous falling interest rate environment in human history and then extrapolated that out into the future and then convinced yourself that that environment (with the 10 year rate at 9%) was somehow comparable to an environment in which the 10 year interest rate is 4.65%.

Your comp, like your benchmark, is a totally apples and oranges comparison.

Again, smart people, meet clever marketing.

[beardsicles]

It's almost as if the track record over a seven year period is going to tell a story of that particular period!

It's almost as if you took the most anomalous falling interest rate environment in human history and then extrapolated that out into the future and then convinced yourself that that environment (with the 10 year rate at 9%) was somehow comparable to an environment in which the 10 year interest rate is 4.65%.

Your comp, like your benchmark, is a totally apples and oranges comparison.

Again, smart people, meet clever marketing.

Right, but your point initially was this was a poor choice compared to VTI, which has experienced a world-historical bull market over the seven years we're talking about. Which one is it?

[FundQuant]

It IS a poor diversifier when compared to VTI. UNLESS interest rates go to 0% again and you get a seismic bull market in T-Bonds.

That's the whole point here. You need a huge outlier move in the bonds to get a premium over VTI. Otherwise you're just paying 17 bps extra in fees per year to own something that will look a lot like VTI.

If you're super bullish about T-Bonds then buy VTI and add zero coupon bonds for free. Why give away free money owning a fund that sounds clever, but really isn't.

[Lawyered_]

It IS a poor diversifier when compared to VTI. UNLESS interest rates go to 0% again and you get a seismic bull market in T-Bonds.

That's the whole point here. You need a huge outlier move in the bonds to get a premium over VTI. Otherwise you're just paying 17 bps extra in fees per year to own something that will look a lot like VTI.

If you're super bullish about T-Bonds then buy VTI and add zero coupon bonds for free. Why give away free money owning a fund that sounds clever, but really isn't.

You actually do not need a bull market in treasuries for NTSX to outperform. Using Testfolio, you can set the T-Bills CAGR to remain flat at 4.65% CAGR. For example, it would have outperformed due to the rebalance bonus of the uncorrelated treasuries from January 1, 2000 to date. See backtest below.

https://testfol.io/?s=fij4rtUDY7P

[FundQuant]

You actually do not need a bull market in treasuries for NTSX to outperform.

True. You need a substantially positive spread and/or falling rates.

Either way, given the current structure of interest rates you are making a very specific interest rate bet since the yield curve is very minimally in contango. You had 2-3% spreads for most of the period of the 90/00s and falling interest rates. That's why NTSX backtests so well.

The current environment is nothing like that. If interest rates stay where they are you're guaranteed to lose to VTI using NTSX. You need a much wider spread and/or falling rates.

In the end, this fund is just 90% VTI with a portable alpha overlay that relies on bonds performing well. Will that happen? I don't know. But this product is not as clever as some people seem to think it is and I would not expect it to perform vastly different than VTI in the future unless interest rates fall very substantially.

[comeinvest]

Yes, both NTSX and RSSB should now (even after expenses, just barely) be close to having a positive carry on the treasuries. Yield on the treasury future is around 4.625% right now. The effective fed funds rate is 4.33%. Treasury leverage cost on top of that is .1% or so, so really cost of leverage is 4.43%, then you throw in the expense ratio of .2%, and you are basically breakeven on NTSX. RSSB is probably still (slightly) negative, since the expense ratio is .36%.

NTSX regularly publishes statistics on their net yield (https://www.wisdomtree.com/investments/ ... s-ntsx.pdf) - as of Jan. 3 it's currently positive across all ages, and adds a yield of .26%, with another .40% of yield from the cash collateral. (For what it's worth, this total of .66% - if you subtract the NTSX expense ratio relative to VOO - is just slightly less than the yield from simply having 10% in EDV).

Is there any documented definition of "embedded income yield" on the NTSX spec sheet? I'm not sure if the definition is trivial. Futures per se have no yield, and it is not clear to me what yield of the underlying is meant, coupon yield, yield to maturity, etc., and how precisely the implied repo rate is determined, if that is what they are trying to do.

[brightlightstonight]

Is there any documented definition of "embedded income yield" on the NTSX spec sheet? I'm not sure if the definition is trivial.

"Embedded Income Yield represents the annualized rate of return generated by a Fund’s investments in securities and derivatives exclusive of interest rate changes and movement in foreign exchange spot rates."

That's their definition. Not sure it's helpful, but I don't how it could be anything other than the current carry from term premium.

[Kbg]

Yeah, plausible benchmarks here are either:

1. VTI/VOO vs NTSX
2. 60/40 vs 2/3 NTSX and 1/3 "something else", where that "something else" is not just "more stock".

70% NTSX, 15% KMLM, 15% gold, vs NTSX, vs SPY vs 60/40. The real use is to diversify and achieve higher risk adjusted returns.

https://testfol.io/?s=1ceQiAJ5dSr

This is exactly the marketing and the financial point. 2022 was an exceptional year for alts which boosts the comparison favorably. The impact of alts in a portfolio has a decades long profile. Stable economy/markets...alts slowly fall behind a typical stock/bonds blend. Crisis, alts pop and the relative performance catches up with or passes the s/b blend. Rinse, repeat.

[comeinvest]

Is there any documented definition of "embedded income yield" on the NTSX spec sheet? I'm not sure if the definition is trivial.

"Embedded Income Yield represents the annualized rate of return generated by a Fund’s investments in securities and derivatives exclusive of interest rate changes and movement in foreign exchange spot rates."

That's their definition. Not sure it's helpful, but I don't how it could be anything other than the current carry from term premium.

That's where the intricacies start regarding terminology. Do you mean a theoretical annualized return from the futures positions between now and the futures' expiration, under the assumption of a static yield curve? I guess that definition would make sense; but I'm just guessing.
I think it's very hard to tell if it would include the underlying treasuries' yield curve rolldown part of the carry.

[brightlightstonight]

"Embedded Income Yield represents the annualized rate of return generated by a Fund’s investments in securities and derivatives exclusive of interest rate changes and movement in foreign exchange spot rates."

That's their definition. Not sure it's helpful, but I don't how it could be anything other than the current carry from term premium.

That's where the intricacies start regarding terminology. Do you mean a theoretical annualized return from the futures positions between now and the futures' expiration, under the assumption of a static yield curve? I guess that definition would make sense; but I'm just guessing.
I think it's very hard to tell if it would include the underlying treasuries' yield curve rolldown part of the carry.

Yeah, heck if I know! But I don't think I have to know exactly how it's calculated - "current term premium" is close enough for me to approximate how this behaves.