Why an asset allocation switching ETF might make sense

David Stevenson

a black and brown hat

The vast majority of investors buy ETFs for very specific reasons, usually involving deploying money to one specific investment niche, expressed via an index.

For example, if you want US stocks, buy the S&P 500. Sounds simple?

The conventional advice is then usually to think long-term - sit tight, ignore market volatility and stay invested for the long-term.

If we're honest, though, this sensible advice hides an important behavioral truth. Investors aren't actually patient enough to sit tight. They look at news about the markets and worry about whether an asset class - their asset class - is too expensive. They fret about market volatility. Most importantly, although they tell themselves they can withstand a loss, when market turbulence heats up, they usually head for the exit - and at the last minute along with the rest of the panicking mob.

The signal and the noise

Quantitative investors have long been trying to work around this behavioral challenge by constructing elegant - and not so elegant - signaling systems which warn investors of impending trouble and then 'switch' them into safe cash or cash alternatives. Over at the FT, I've frequently written about technical-based systems which make use of say the 20- and 200-day moving average to prompt an asset allocation switch. These ideas have struggled to make it into mainstream products although in the more alternative world of structured products there have been a few attempts at building 'smart portfolios' that use switching systems.

Thus, it was only a matter of time before these technical-based signaling systems found their way into the racy world of ETFs. In fact, I've been amazed at how long it's taken the ETF strategist brigade to wake up to the possibilities.

Consider this market opportunity: investors can benefit from the full upside of equity markets but switch back into cash when markets look risky and over priced. What's not to like about such a system, especially if it can be cheaply administered?

US issuer VanEck has decided to address this market with a new US ETF launched last week - it's the VanEck VectorsNDR CMG Long/Flat Allocation ETF (NYSE Arca: LFEQ), and represents another product based on the relationship with Ned Davis Research, an ETF strategists firm.

The central insight of this new US tracker is that "since 1928 the S&P 500 has spent 70% of the time either in a bear market or recovering from one". The ETF implements "a systematic approach that seeks to preserve capital by increasing cash when market health is weak, and participate in uptrends with a full allocation to equity."

Here's the basic facts. LFEQ seeks to track the Ned Davis Research CMG US Large Cap Long/Flat Index (ticker: NDRCMGLF or "Index"). The ETF is built a rules-based index "that follows a proprietary model developed by NDR and CMG Capital Management Group, Inc. (CMG). The model produces trade signals that dictate the Index's equity allocation (100%, 80%, 40%, or 0%) and/or cash (U.S. T-bills) allocation."

I think it sound smore complicated than it is. The fund invests in two options.

  • Option 1 - US equities, via the S&P 500 Index

  • Option 2 - Risk free assets. "Cash exposure is provided by the Solactive 13-week U.S. T-bill Index"

According to VanEck "the index may rebalance intra-month based on signals from the model. LFEQ will allocate to S&P 500 equities (through ETFs initially) and/or U.S. T-bills. LFEQ has an estimated gross expense ratio of 0.63%, net expense ratio of 0.59%, and is contractually capped at 0.55%, through February 1, 2019."

Bait and switch

So, what does this switching system look like up close? The core driver seems to be technical measures which look at market breadth and the index direction - which indicates that not only are we looking at measures such as moving averages but also market volumes. The index looks at short, medium and long term "trends" based on momentum as well as secondary indicators. These include "three mean reversion measures, for example, the degree in which a market may be over sold or over bought to help support the primary trend readings."

This system then issues a score "over four zones" which can result in the model moving from 0% exposed to equities, through 40% and 80% and then 100%. According to Ned Davis "this strategy has had 48 trades (i.e changes in allocation percentages) since 1995 based on historical research. The average number of days between trades historically has been about 125".

What about returns? Obviously, the underlying index is very new but for the period between the end of 2016 and September 30 2017, returns were slightly below the S&P 500 (12.31% vs 13%) but with lower standard deviation, a higher sharpe ratio - and the same max drawdown i.e not that different from the index. The obvious comment here is that this historical data is next to worthless and we need to see how the system works over the long term.

If I'm honest the devil is always in the detail with these clever ideas. Won't all the trading back and forth start to jack up costs? What happens when the model gives out false alerts, resulting in investors missing a sudden market uptick? What happens if the model misses a huge sell signal?

Despite all these obvious worries, this is actually a damned smart idea for many fundamentally lazy investors; those who want US large cap exposure but also probably want to be in cash when this enormously liquid market starts to head south?

What this index and accompanying ETF doesn't of course protect you against is overvaluation. Market momentum might carry an overvalued index into frankly expensive territory. That's because momentum-based measures don't really care about whether a PE is looking toppy or not. But that simplicity also contains an important truth - value considerations doesn't always trump momentum.

Do it yourself

Of course, investors could always do this kind of switching themselves. They could buy access to simple technical software such as Sharescope, plug in the ticker for the S&P 500 index and then wait for the usually obvious signals.

I'm sure that Ned Davies will try and suggest that their system is multi-layer and very clever. But in reality, my guess is that its built on trend following 101 stuff, which is no bad stuff. You could use these ideas yourself and then switch back and forth between a US ETF and a cash tracker. Your total expense ratio would probably be below 20 basis points in total, saving you at least 30 to 40 basis points per annum compared to this ETF.

But I also maintain that most investors - including fairly active ones - simply can't be bothered. They have too many things to do on a daily basis, and looking at MA technical measures doesn't sound terrifically exciting. They might also miss key signals simply by not being in front of a computer at the right time.

Thus, an ETF like this, though by no means perfect, could serve an important market: namely those investors who want to access the US equity markets but with some downside protection. Let's hope that similar products comes out in the UK which give access to a broader range of underlying asset classes including the FTSE 100 and the Eurostoxx 50.

Featured in this article


No ETFs to show.