We’ve rebalanced our portfolios

[gmachado]

Part of our role as portfolio managers is to make changes to our clients’ investments when we think it’s necessary or when opportunities arise.

Given the broad market sell-off we’ve seen over the past 12 months, coupled with the increase in yields from fixed income assets, we’ve made some changes to our growth portfolios to better position them for the current environment.

This short post outlines the changes agreed at the most recent Investment Committee meeting. We’ll be giving further information on the changes made, as well as a look into how we decide what goes into our portfolios, in our upcoming ‘Investment Philosophy’ document.

Asset allocation changes

The InvestEngine Investment Committee has decided to reduce the risk in our portfolios, by lowering our exposure to equities and increasing allocations to bonds.

Although this change will be most significant for clients with lower-risk portfolios, there will also be an increase in bond allocations for clients investing in higher-risk portfolios. This reflects a preference for bonds over equities throughout portfolios in general. These changes will allow clients, particularly on the lower end of the risk spectrum, to benefit from bonds’ more stable returns, which are considered safer than equities.

Recently, long-dated bonds have shown increased correlations with equities – i.e. their performances have been more similar. Short-dated bonds, however, tend to be better protected against large drawdowns given their shorter time to maturity and lower sensitivity to interest rate changes (also known as “duration”).

While they will not rise as much as long-dated bonds when interest rates fall, they are also unlikely to fall as far when rates rise. For example, the iShares Core UK Gilts UCITS ETF has a duration of about 10 years – in the 12 months to the end of November 2022, it fell 23%. Compare this with the iShares UK Gilts 0-5yr UCITS ETF, which has a duration of only 2 years and fell only 4%, and the difference becomes clear.

This high interest rate environment has caused short-dated bonds, in particular, to offer a more attractive risk/return profile given the higher yields on offer. As the yield curve is currently relatively flat (i.e. you only see a minor increase in yields for investing in riskier, longer-dated bonds), the risk/return opportunities are better at the shorter end of the spectrum.

Equity changes

As for the equities in our portfolios, we’ve increased the allocation to factor-based ETFs. Factor-based funds (also known as “smart beta” funds) are purposely tilted towards certain characteristics like lower stock prices, smaller companies, or higher quality companies, to achieve specific risk and return objectives.

Bringing more factor-based funds into our portfolios increases the potential for higher risk-adjusted returns. By diversifying risk factors, investors also reduce the risk of a single part of the portfolio underperforming for extended periods of time, which reduces the potential long-term variations in the portfolio’s value.

More information on factor investing will be released in our upcoming ‘Investment Philosophy’ article.

Fixed income changes

Clients may also notice a larger allocation to government bonds than corporate bonds. With yields having risen so dramatically this year, government bonds now offer more attractive returns, without some of the risks associated with corporate bonds. They also come with lower correlations to equities, and are more likely to provide superior protection in the event of a market downturn.

Other changes

We have removed the small allocation to gold across portfolios. Despite its reputation, gold has shown very little real-world evidence of being a reliable hedge against inflation. This has been clear during the current period of high inflation, in which gold has provided disappointing returns.

For portfolios requiring a higher level of protection against unexpected spikes in inflation, inflation-linked bonds are likely to be a more reliable hedge (we have these in our portfolios). While gold has provided some cushion during past market crashes, high credit-quality bonds have also offered protection, and crucially they come with higher long-term expected returns.

[Highview]

Thank you for the informative article.
As a ‘newbie’ to InvestEngine I must say I am impressed with the client contact both in its content and information regarding managed accounts.

I have always been a favorite of rebalancing over my years of self investing having due regard to circumstances prevailing at the time.
The days of ‘buy & hold’ are in my opinion often a recipe for disappointment particularly when bear markets prevail.
Cash is now fashionable once again with interest rates rising but is still on the wrong side of inflation.
Here is hoping Bonds are a safe bet. !

[shrimper]

If you diversify the factors in your allocation to smart beta - aren’t you in danger of just replicating the general market but in higher TER ETF’s?

[chungf]

An (over-)simplistic way to tell is by backtesting both approaches.

I plotted the following in backtest.curvo.eu:

• MSCI World – using SPDR MSCI World (SWLD) because it’s an accumulative fund with the lowest TER;
• MSCI World Discrete Factors – using the following in equal weightings, same as what InvestEngine’s managed funds would do:
  • iShares Edge MSCI World Minimum Volatility (MINV)
  • Xtrackers MSCI World Value (XDEV)
  • Xtrackers MSCI World Quality (XDEQ)
  • iShares MSCI World Small Cap (WLDS)
  • Xtrackers MSCI World Momentum (XDEM)

The date range is Dec 2000 to Oct 2022 because that’s the longest common date range in which the backtest tool has data for all the funds.

The result:

The numbers:

	MSCI World	MSCI World Discrete Factors
Compound Annual Growth Rate	5.49%	7.13%
Standard deviation	14.42%	13.47%

This would indicate that, historically within the various sorts of investment environment from Dec 2000 to Oct 2022 at least, the “smart beta” approach has added extra return on top of the general market.

But wait, there’s more! The backtesting tool also helps you optimise your portfolio by efficient frontier analysis. Just for the heck of it, I hit the button and out came 2 more alternative weightings:

	MSCI World + Factors, max Sharpe ratio	MSCI World + Factors, max return
Min. Vol. (MINV)	50%	0%
Value (XDEV)	0%	0%
Quality (XDEQ)	0%	0%
Small Caps (WLDS)	15%	40%
Momentum (XDEM)	35%	60%

And their risk vs. return would be like:

	MSCI World + Factors, equal weightings	MSCI World + Factors, max Sharpe ratio	MSCI World + Factors, max return
Return	7.13%	7.16%	7.87%
Risk	13.47%	12.33%	14.6%
Sharpe ratio	0.48	0.52	0.50

The “max Sharpe ratio” allocation suggests that, in the historical date range of Dec 2000 to Oct 2022, you could have eked out a little more performance without compromising on risk by dropping the value and quality factors and redistributing to the other 3 factors.

Whether this is going to still hold valid going forward, particularly in the inflationary environment and value rotation we are in, is for InvestEngine’s specialists to consider, of course.

Some caveats:

• The backtest tool lets you test with different investment patterns (lump sum with/without regular contribution) and rebalancing strategy. I chose a lump sum of €10,000, no regular contribution, and yearly rebalancing;
• The tool uses backdated data for some of the funds calculated by unknown methodology, as the younger funds surely haven’t debuted in year 2000;
• The tool uses Euro as its base currency so there is some forex conversion involved

(Now if InvestEngine would develop its own backtesting tool for clients’ use, based in GBP, that would give them an enormous edge over its rivals… *hint* *hint* 😉😉)

[Tykethat]

Terrific “Chungf” thank you. Looks like IE should offer you an analyst job!

[shrimper]

Thanks for that - do the alternative weightings suggest that the outperformance is coming from essentially overweighting small cap and momentum versus the ‘normal’ MSCI world?

[chungf]

Yes, as an after-the-fact explanation of what happened within Dec 2000 to Oct 2022.

Looking forward, though, I’ll probably worry a bit more about fluctuations, and therefore give credence to the max Sharpe ratio weightings too. In that case, the most influential factors were minimum volatility, momentum and small cap (in that order).

Among the 5 factors, small cap is a problematic one. I say this because it has rather unfortunate performance characteristics. Here is the Dec 2000–Oct 2022 performance of the 5 factors from backtest.curvo.eu (you can see the details for yourself):

	Min Vol (MINV)	Value (XDEV)	Quality (XDEQ)	Sm Cap (WLDS)	Momentum (XDEM)
Compound annual growth rate	6.32%	6.71%	6.25%	7.88%	7.59%
Standard deviation	11.29%	15.88%	14.01%	16.87%	14.16%
Sharpe ratio	0.48	0.41	0.41	0.46	0.50

Small cap has the highest growth rate (benefit), but also the highest standard deviation (risk). On balance, its Sharpe ratio (benefit over risk) is not so great.

So factor investing experts do have a point when they say small cap by itself is a weak factor, and has to be integrated with other factors to get the benefit of amplified returns while managing volatility.

Unfortunately we don’t have “small cap + (some other factor)” ETFs that will do this. (Buying a small cap ETF + a value ETF is not the same as buying a small cap value ETF, for instance.) So for those of us under the managed portfolio, small cap as an ETF by itself will remain that outperformer with a mood swing.