Join our community of smart investors

Asset Allocation for 2018

The past year has seen a widespread yet shallow economic recovery, and central banks have taken this as a signal to tighten monetary policy. James Norrington adjusts his tactical asset allocation portfolios to mark the end of the Goldilocks markets in 2018
December 15, 2017

Even before his untimely demise on 25 December last year, George Michael’s anthem ‘Last Christmas’ ensured his association with the festive period. Granted, the investment link is tenuous, but the lyrics “I gave you my heart, but the very next day you gave it away” are a seasonal metaphor not to take yesterday’s apparently low-risk gains for granted tomorrow. Applied to the excellent performance of our tactical asset portfolios from last Christmas, the message is not to dismiss risk in 2018 on the basis of 2017’s gains, which were achieved with remarkably low volatility.

Our two models at Christmas 2016 rebalanced portfolios from features that year and both were tactically rejigged once more in April 2017. Dating overall performance from our 16 December 2016 issue, the portfolio originally conceived as a response to the rise of populism has made total returns of over 12.2 per cent. Taking the gross redemption yield on 10-year gilts from 16 December 2016 (1.438 per cent), that’s 10.8 per cent in excess of the ‘risk-free’ rate. The more general tactical asset allocation (TAA) portfolio didn’t do quite as well, chalking up total returns of 9.8 per cent (8.4 per cent excess).

This figure for general TAA is comparable with the iShares FTSE 100 ETF (ISF), which made 9.6 per cent. Our general TAA portfolio invested in several ETFs, so allowing for these additional costs, just buying the FTSE 100 would have done better. The justification of course is that by investing more broadly we diversify some of the risks of individual asset classes such as UK shares.

Modern portfolio theory (MPT) takes volatility (the standard deviation of returns) as its proxy for risk and the capital asset pricing model (CAPM), which expanded MPT, introduced the notion of a risk-free rate of return. Nobel Laureate William Sharpe made an integral contribution to the CAPM and the ratio that bears his name shows excess returns relative to volatility. Our populism portfolio not only performed well, it did so with low standard deviation of returns, which the exceptionally good Sharpe ratio of 2.74 reflects. Our general TAA portfolio was far more risk-efficient, with a superior Sharpe ratio of 1.74 versus 1.01 for the standalone investment in UK blue-chips.

It would be easy to rest on our laurels and pat ourselves on the back with these numbers, but there are reasons to be cautious in case the story goes sour. Firstly, while volatility has been low for 11 months, we should question whether our investments are low-risk. In 2017 monetary policy remained accommodative which, coupled with a benign set of macro circumstances, including the return of organic GDP growth in the eurozone and robust company earnings figures in the US, meant there was very little disagreement among market participants. Hence asset prices continued to tick higher with few significant pullbacks and volatility remained conspicuous by its absence.   

Along with realised volatility, there has been little negative emotion in financial markets in 2017. Implied volatility – as measured by the rate of change in options prices (the most famous index is the VIX, showing the rate of change of S&P 500 options prices) – has also been at record lows. As argued earlier this year (‘Nevermind the volatility’, 2 June 2017) this doesn’t mean risks have disappeared. One interpretation is that in a ‘Goldilocks’ environment of positive economic data and easy money policy, investors (rightly in hindsight) felt the positive impetus for the market overrode other risks. Basically, market beta was the only risk investors were concerned about and there was massive consensus it would be positive.

For our portfolios, we can say therefore that we achieved good returns relative to low-beta-driven volatility. What our past performance doesn’t show us is how the diversification of beta exposures that we had would protect us when other factors start to have more of an effect on asset prices.

What are these factors? Well, at the most basic level we can split portfolio returns by those that come as a result of correlation with an overall market or benchmark – beta returns – and those that are thanks to active decisions – alpha returns. Drilling down further, academics Eugene Fama and Kenneth French (1990) identified three types of beta. Looking at US equities, they concluded that some of a stock’s beta is due to the impetus of the market, some is due to the relative size of the stock by market capitalisation (ie, a risk premium exists for small caps) and some is due to the market re-rating the economic usefulness of a company’s assets (ie, a value premium for unloved stocks coming good). Later, a momentum factor was added by Mark Carhart (1997), who wanted to quantify the well-attested effect of the best-performing stocks often continuing to see positive price movements in immediately subsequent periods.

Ignoring alpha for the moment, we need to consider how our asset allocation exposes us to different strands of beta. Sticking with the example of equities, if we hold passive instruments where stocks are weighted according to market capitalisation it is possible bigger dividend-paying companies could relatively underperform, as these are already the most expensive. Say organic economic growth continues and monetary policy is gently tightened, it would be reasonable to expect value stocks – where there has been less multiple expansion already – to outperform as the market cottons on to their assets becoming potentially more productive.

Changing correlations prompt adiversification rethink

MPT rests on the principle that negative correlations ensure losses for one investment are mitigated by the rest of a diverse portfolio. A consequence of ultra-low interest rates and central banks’ quantitative easing (QE) programmes has been that some of the relationships portfolio managers have historically relied on have broken down and/or reversed. The most obvious example is the prices of equities and bonds.

Thanks to central banks’ bond-buying programmes, price increases have accounted for the majority of bond returns in recent years. This massively inflated bull market means that yields, which move inversely with prices, are at rock bottom. The effect of low yields on government bonds is that there is no risk-free rate of return, once inflation is taken into account, and investors have flocked to riskier assets such as shares. With electronically ‘printed’ money in abundant supply, we have a situation where both asset classes have become expensive and their prices keep moving in the same direction. Encouraged by low interest rates, global property prices have risen too and we are left with what The Economist has dubbed “The bull market in everything”.  

The assumption has to be that if everything can rise together, it can fall together, too. It has often been said that “in a downturn, the only thing that goes up is correlations”. Against such a backdrop, we cannot guarantee avoiding losses if there is a crash but, by taking a more nuanced approach within asset classes, hopefully it is possible to diversify more of the risks that could be most prevalent in 2018.

 

Taking a thematic approach to risk

Diversifying asset classes doesn’t always mean risk is being spread. For example, investments in high-duration bonds, buy-to-let property and UK mid-cap equities give a mix of assets, but all of them may be adversely effected if interest rates rise significantly. Duration is the measure of a bond’s price sensitivity to interest rates and longer duration bonds will suffer significant capital loss when rates rise. Buy-to-let investments also become more costly to finance as mortgage rates increase at a time when tenants’ capacity to pay rent may be reduced. There could be a dip in the value of the property too as higher rates slacken demand in the housing market. Finally for mid-cap UK shares, the increased cost of debt financing and tighter credit conditions for customers in the domestic economy drag on earnings.

It is essential to take a top-down view of the risks that exist, how they affect different asset classes and, at the next level of granularity, which securities are most exposed. In updating our portfolios for 2018, we must consider what are the macro risks, how each major asset class could be affected and which investing styles might do best. 

In portfolio management, unless we have a very low required rate of return, our objective is not necessarily to diversify away all risk. We are instead making active decisions about the risks we think will be rewarded and looking to tilt weightings towards these, while reducing the exposure to the risks we think will be unrewarded or realised in the form of negative returns. Populism and political risk remains one strand of the challenges faced by investors, but the asset allocation we initially suggested has rather morphed into a more general multi-factor risk portfolio (this system makes greater use of factor ETFs), so we will rename it accordingly.

Positioning for 2018

One of the main risk themes for 2018 is of course the tightening of interest rates which, together with the tapering (ECB) and unwinding (US Federal Reserve) of QE, is likely to create headwinds for bonds. For this reason, we are making alterations to our fixed-income positions. We are reducing the exposure to longer -duration government bonds and reducing bonds overall in the portfolios. In the multi-factor portfolio, we drop INXG and IGLO completely due to the proportion of long-duration issues that will be highly sensitive to interest rate changes. We also take IS15 corporate bond ETF down to just 5 per cent of the total allocation.

The fixed-income holdings in the general TAA portfolio were lower duration, with a 7.5 per cent position in the SPDR 1-5 Year Gilts ETF (GLT5). This has done badly over the past few months, dropping as yields on UK gilts rose in response to the Bank of England raising rates by 25 basis points to 0.5 per cent. This highlights the capital risk of tightening and we reduce the position to 5 per cent as well as getting out of IGLO, reducing our corporate bond position to 5 per cent and also backing out of high-yield bonds. With fixed income a tricky area to get right, we are adding a managed fund to both portfolios. The MI Twenty Four Asset Backed Fund invests in asset-backed securities and has a strong focus on capital preservation. With inflation rising, this is a better option than just holding more cash, although an increasing cash cushion is important too as we take proportionately more equity risk.

Commercial property allocations in the portfolios are rebalanced to their April levels as these investments offer an alternative play on global economic growth. We keep gold at 5 per cent of general TAA and bring the multi-factor portfolio into line on the same allocation. 

 

The only way is equities…

For many analysts, the only way is equities and we are beefing upallocations in both portfolios,although the increase is chiefly made by investing in listed infrastructure. This is a sub-class of equities that focuses on companies with solid, growing cash flows and barriers to competition. We’ve made a 10 per cent investment for each portfolio in the new M&G Global Listed Infrastructure fund (GB00BF00R928). This is an internationally diverse fund, which focuses on the ability of investments to grow dividends. As well as traditional economic infrastructure such as oil pipelines and toll roads, the fund invests in growing digital infrastructure, such as data warehouses, and is not overly exposed to the UK or the sort of private finance initiative (PFI) contracts that carry political risk in this country.

For the rest of our equities positions we have tweaked our passive ETF holdings. For the multi-factor portfolio, we maintain 10 per cent in the iShares Global Quality Factor ETF and 10 per cent in the minimum volatility ETF. The value factor ETF is reduced to 10 per cent in line with this. Investing passively in value stocks has worked well since April – total returns from the last rebalance are just under 13 per cent. We make the change, however, to help make room for the listed infrastructure investment which, although partly a fixed-income proxy, is subject to equity market risk. The rationale, therefore, is that we must reduce beta elsewhere.   

The reason we haven’t reduced the quality or minimum volatility allocations instead is because, although there is a case that the biggest American companies are very expensive, the sub-plot of President Trump’s proposed tax reforms is intriguing. We have through our global factor index trackers, stakes in companies such as Microsoft (US:MSFT) and Apple (US:AAPL). Although pricey to buy into, these companies will be major beneficiaries if Congress passes corporate tax breaks. Thanks to the combination of our different factor investing strategies and additional positioning towards Japan, emerging markets and the eurozone, expensive US companies comprise less of our total equity holdings than if we held a global market-cap-weighted tracker. Therefore, we won’t miss out if the US mega-caps enjoy further upside, but we aren’t overexposed if the tax break story doesn’t play out. 

In the general TAA, indices are all market cap weighted but the S&P 500 makes up only 7.5 per cent of the portfolio, so we aren’t trying to time a fall in expensive US stocks, but again, thanks to portfolio weighting, the relative size of our position is moderate. With an equal split between FTSE 100 and FTSE 250 ETFs, a high proportion of equity returns depend on UK-listed companies. This does not necessarily mean our fortunes are tied to the UK economy, so we are rebalancing these holdings at their previous weights. The FTSE 100 is a very international index and company profits are made from economies far and wide. The FTSE 250 is more domestically focused, but there are still plenty of companies with a decent portion of overseas earnings.

For emerging markets, we went for a minimum volatility ETF in the multi-factor portfolio. This has done well, but it has underperformed the general emerging market ETF used in the general TAA. This seems reasonable according to investment theory – lower volatility should be rewarded with lower returns. In hindsight, it was probably a bit daft trying to get the low volatility premium from what is the riskiest class of equities. On the basis that these generally risky emerging market shares will have periods of cyclical over- and underperformance, however, we’ll keep the same ETFs in each portfolio. Doing so probably falls into any number of behavioural investing traps, but some assumptions are unavoidable and at least by having a fairly strict range of allocation sizes, the impact of getting these wrong is limited. In other words, one of the risks we are diversifying is human error in choosing individual funds, as our returns rely more on asset allocation than securities selection.

 

General Tactical Asset Allocation Portfolio

FundApril allocation (%)TR since last review 6.04.2017 to 16.11.2017 (%)New allocation 16.11.2017 (%)
iShares FTSE 100 (ISF)7.54.07.5
Vanguard FTSE 250 (VMID)7.55.17.5
db x-trackers Eurostoxx (XESC)109.37.5
Vanguard S&P 500 (VUSA)7.54.67.5
Vanguard FTSE Japan (VJPN)7.510.75
iShares Emerging Markets (EMIM)7.511.27.5
SPDR Barclays 1-5 yr Gilts (GLTS)7.5-0.75
iShares Global Govt Bonds (IGLO) 53.6Sold
iShares £ Corporate Bond 0-5 yrs (IS15)7.50.75
iShares Global High Yield (GHYS) 7.52.2Sold
HSBC FTSE EPRA Nareit (HPRO)10-0.110
Source Physical Gold (SGLD) 52.45
M&G Listed Infrastructure   10
MI Twenty Four Monument Bond  10
Cash 10 12.5
    
  Total returns since last review4.06
    
  Portfolio returns since 16.12.2016 9.79
  Volatility since 16.12.20164.81
  Sharpe Ratio1.74

 

Multi-factor Asset Allocation Portfolio

FundApril allocation (%)TR since last review 6.04.2017 to 16.11.2017 (%)New allocation 16.11.2017 (%)
iShares World Quality Factor (IWQU)1011.210
iShares World Min Vol (MVOL)108.410
iShares World Value (IWVL)12.512.810
SPDR UK Dividend Aristocrats (UKDV)7.5-3.37.5
db x-trackers Eurostoxx (XESC)7.59.35
Vanguard FTSE Japan (VJPN)510.75
iShares Emerging Markets Min Vol (EMV)7.55.75
iShares £ Index Linked Gilts (INXG)5-4.3Sold
iShares Global Govt Bonds (IGLO) 103.6Sold
iShares £ Corporate Bond 0-5 yrs (IS15)7.50.75
HSBC FTSE EPRA NAREIT (HPRO)5-0.15
Source Physical Gold (SGLD) 7.52.45
M&G Listed Infrastructure  New holding10
MI Twenty Four Monument Bond New holding10
Cash 5 12.5
  Total returns since last review5.31
  Portfolio returns since 16.12.2016 12.22
  Volatility since 16.12.20163.93
  Sharpe Ratio2.74