On the 18th of October, WisdomTree (finally!!) listed a UCITS version of their ETF NTSX in Europe.
If you are wondering what’s NTSX, you must be new around here.
“The ETF is designed to deliver a 90% exposure to large-cap US equities and 60% to US Treasury futures, effectively providing a leveraged, more capital-efficient alternative to the traditional 60/40 portfolio.” It is a US-based 60/40 levered 1.5 times. It is also the core of The Italian Leather Sofa Model Portfolio.
WisdomTree has also two other similar ETFs that invest in International (NTSI) and Emerging Markets (NTSE) stocks, plus US Treasuries, but they are not (yet?) available on this side of the pond.
What can you do with NTSX?
The long version would be: read this paper. The short version is two things:
- You can replace your portfolio allocation to SPY, or any other ETF linked to US large caps, with NTSX and get higher absolute and risk-adjusted returns.
See it? It works for International and EM too.
If you do not believe the paper, you can run the same backtest on PortfolioVisualizer:
(I used PGOVX simply because it allowed a longer backtest, with IEF you get similar results on a shorter horizon)
I never cared about tracking errors…until my friend Mr RIP compared NTSX ag various combinations of stocks and bonds, starting from when it was listed in the US. And since it started trading quite recently, that comparison horizon is not that long (c5 years). Any portfolio with 14% vol can do pretty much anything in 5 years.
Especially if these 5 years include this:
Annual real returns of the 60/40 from PortfolioCharts
If we leverage the 60/40 during one of its worst periods ever, no wonder we end up with a non-stellar result 🙂 I guess it is worth mentioning this part of the paper then:
“However, it’s worth noting that the 90/60 strategies exhibited annualized tracking error of 3.2-3.7% and frequent relative drawdowns versus their respective equity markets. For example, 90/60 strategies significantly underperformed during the dot come bubble and in 2022. Despite long-term historical outperformance, and prospective outperformance, not all investors are able to stomach (and stick with) an equity substitute that underperforms the market for years at a time.“
So here it is: NTSX should outperform SPY but it should not outperform every year. Remember ergodicity? 😉 It is true that this type of tracking error is not free, meaning it requires additional fortitude to stick to the plan (this week, I was not sure if I wanted to write this piece or one inspired by this post on diversification. It explains what I am trying to say better than I could ever do).
- 2. 90/60 strategies can be used as replacements for traditional stock and bond allocations to implement a return stacking strategy
This is how I use NTSX in the Model Portfolio, to be able to allocate to alternatives without sacrificing the standard 60/40 returns (at the cost of leverage, obv). From the paper:
“Reiterating this outcome – the capital efficient strategy enabled a portfolio diversifier that underperformed equities to push the total portfolio return above the equity market. More importantly, strategies that clearly underperformed diversified portfolios, like hedge funds and managed futures, were also accretive to investor returns since they produced returns in excess of the alpha return hurdle. Capital efficiency lowers the bar for good diversification decisions by enabling low-returning diversifiers to play a role in the portfolio. Despite the additional complexity inherent in capital efficiency, many investors might find it behaviorally beneficial versus a more traditional approach“
Maybe this passage is a little too complex. I found another great example here. Do you remember gold? That “asset” with lousy returns but high volatility? Look what it can do in a capital efficient context, i.e. in a portfolio with NTSX and its cousin GDE (another ETF from WisdomTree that offers a 90/90 exposure, 90 to US large cap stocks and 90 to gold):
NTSX (and GDE) allows the investor to basically dial up returns to their desired level without going crazy on the volatility side, as it would happen when dealing with equities alone. Or, if you want to see it from the other angle, the investor can start from a target return and reduce the volatility necessary to get there. The Capital Efficient Golden Butterfly has the same CAGR as VTSAX with half the max drawdown; or, the 90/40/30 offers the same risks as VTSAX but with 2% of additional CAGR.
Big, BIG, aside from the PWRs you see in the table above: this PortfolioVisualizer analysis consider only the period 12/04 – 10/23. You get extraordinarily high PWRs simply because that period was very good for equities; PWRs for longer periods should be lower across all the portfolios (maybe not for the VBMFX one but…who cares eheheh). That said, if we consider that this period is sufficiently representative of the correlations between the assets in those portfolios, then the conclusion should directionally stand. I am not sure the premise is 100% correct but I would expect capital-efficient portfolios to offer higher PWRs.
Leverage
I wrote countless posts about leverage, what it is and its costs and implications; go read them. Here I just want to highlight two facts about NTSX:
- the cost of leverage is not included in the ETF’s TER. In the NTSX case, the cost of leverage is embedded in the price of the futures used to replicate the bond sleeve of the fund. Futures are leveraged products and their price reflects the fact that buyers/sellers get access to leverage, it is just another arbitrage mechanism. Ok, I just realised there is no simple way to explain this…If you want to know what the cost of leverage is today, check the Fed Funds rate: it is a very good proxy. Why would you expect this to be reflected in the ETF’s TER…I honestly do not know but you might have a valid point! Sorry for you, it will never be though. Same for currency-hedged ETFs, the cost/benefit of hedging will never be reflected in the TER, get over it.
- the presence of leverage, and the fact that it has a cost, is a relevant contributor to the fund tracking error. I understand that this further complicates how you explain to yourself how your investment is performing but…that’s part of the bargain. I find Corey Hoffstein’s trick “It is all long/short portfolios all the way down” helpful: look how I built the backtest in PortfolioVisualizer. Leverage and its costs are represented by the short position in CASHX. Plot the ingredients, SPY IEF and CASHX, separately and you will better understand who’s doing what (performance attribution dawgs).
How to track it in your portfolio
Did I convince you? Should I? For sure, WisdomTree nor any of the other capital-efficient ETF providers are paying me to do so (I wish!).
Anyway, Risk Parity Chronicles has a nice post on how to track capital-efficient funds in a portfolio. That’s a great content idea since I had a bit of a struggle myself. Honestly, I am not 100% sure I understood his methodology but I think we arrived at a similar conclusion (considering we might be interested in tracking different things). Here is what I do:
My portfolio currently operates with a leverage of 145%. I engage in investments ranging from unleveraged funds, such as DBMF, to those leveraged threefold, like TYA. The market frequently sees the introduction of new funds nowadays. I closely track this leverage ratio as part of my ongoing experimentation, aiming to cautiously increase it whenever it aligns with my strategic considerations.
The section on the right-hand side shows my target asset allocation vs current allocation. For example, I would like to have 58% invested in stocks and I am currently at 60.7%. Every $1 invested in NTSX contributes $0.9 to “stocks” and $0.6 to “bonds”: this is how I move from “dollar invested” to “dollar exposure”. This way, every time I want to make a change to the portfolio, I can immediately see the relationship between adding/reducing an ETF and what happens to the portfolio allocation and leverage level.
The section on the left-hand side shows the portfolio divided per dollar invested. ETFs are grouped and ordered according to a logic: mine. Sorry. The theory is to put together products that are similar (COM and GDE have both exposure to commodities) and list, for example, RSST (trend+ stocks) between the group that has exposure to trend and the one to stocks. As you can see, my logic is not always respected by yours truly. For sure, I never got a job for my prowess in creating beautiful Excel sheets. Anyway, the main goal of this section is to show me that the exposure to different trend models (the Return Stacking family, CTA and DBMF) is balanced. Or that TYA, ACWV and BTAL do not exceed 5%.
Why 5%? To me, it feels like the sweet spot between “if something goes wrong, I’ll not be overly screwed” and “this moves the dial”. I am still experimenting. One day, I’ll have to take a decision on what to keep, and increase its allocation to a level that really matters, and what to exclude. Do not read this section as a guide but more as a live journal: when (if ever?) I will be 100% sure of what I am doing, I will do a proper manual (don’t hold your breath).
I am not even sure anymore about the starting premise: to enhance a 60/40! Many of the ETF providers started from there because they have to sell to financial advisors who are married to the 60/40. The 60/40 is not the optimal starting point, it is the point that maximises their chances to sell their product. That said, would I ever stomach a lower than 60 58% allocation to stocks? I do not think so. Would I ever manage to allocate 10% or more to gold? Not sure either.
Few, last considerations:
- COM might be unnecessary considering the other Trend funds can also take positions on commodities. On the other side, it would make sense to up the portfolio exposure to gold from its current 6% (GDE)
- RSBT should be 0 cause RSST works better along TYA… but I discovered TYA after I allocated to RSBT and I do not want to sell (the cost of the mistake is to be overexposed to the Return Stacking team Trend model compared to DBMF and CTA. I can live with risk, knowing that it will go down as the portfolio size goes up)
- BTAL is a new idea, will allocate up to 5%
- TYA will also go up to 5%
- Stocks exposure to US (that 64.7%) should be more 60%
Ultimately, leverage is just an extension of what you can normally get going from 0 to 1. It fills further and extends the efficient frontier. Picture it in your head:
No one is telling you to go to the extreme right (especially OUTSIDE of finance). It is just an instrument that provides retail investors more choices; lads that have access to futures have used these concepts for ages and the Earth is still spinning 😉
What I am reading now:
Follow me on Twitter @nprotasoni