It’s taken four months, but my two employee pensions are now toast and the money has been transferred into a self-managed portfolio. Sipp, Sipp hooray! Then again, no more blaming others for bum returns. We’re on our own from here, readers.
About time too. Stuck in bureaucratic limbo, my portfolio wasn’t changing much month to month. You were losing patience? It’s a frustrated man who has to write 950 words on index-linked bonds. Thank goodness for the distraction of a mini banking crisis.
And sitting on so much cash — while inflation sneered — was excruciating. No more. As you can see from the table below, I’ve been busier than a borrowed mule. Once the money landed, I purchased everything I said I would in previous columns.
For some funds the timing wasn’t great. I’d rather have bought some US government bonds when I wrote about them in December. With banks, on the other hand, the timing feels nice. I pressed buy on a European bank ETF just as the sector was melting down last Friday.
Meanwhile, I’ve added more US equity exposure, as I wrote at the start of the year, and even bought some Treasury inflation-protected bonds. I would have preferred to wait a while on these, but a promise is a promise.
Thus my remaining liquidity is spent and I’m fully invested, as they say in the trade. You can see that my portfolio is three-quarters shares and the rump bonds — with UK equities and short-duration Treasuries the biggest exposures.
Not my best work to be honest. Such was the joy of freedom that I rushed. OK, I roughly aimed for something a tad more aggressive than 60 per cent equities, 40 per cent bonds. But have I optimised my portfolio? Er, no.
Pretty lame for an ex-head of multi-asset research. The main thing I analysed with clients is return and risk targets. Should modern portfolio theory be applied to one’s own savings, then? Like me, I doubt many readers have asked the question.
Which is fine if you have a defined benefit pension or ever ticked a box marked “balanced fund” or “multi-asset fund”. The clever teams managing such products do it for you. Indeed, they often live and breathe modern portfolio theory.
For those of us investing our own savings, however, where to start? A good place is understanding what it is. Conceived in the 1950s by Harry Markowitz, modern portfolio theory assumes we want the best return on our money for the least risk. This is achieved with diversification.
To work out how, securities, asset classes, or funds are observed in relation to each other alongside the two dimensions of return and risk. The aim is to maximise performance, using historical returns as a guide. Risk is then minimised by choosing investments whose volatilities are least correlated.
This is easier said than done. To know how many stocks are needed versus government bonds, or whether adding property and Asian credit will improve a portfolio is a brutal calculation. You need to know the risk and return profile of every hypothetical portfolio given all possible asset allocations.
For example, 60 per cent equities, 30 per cent bonds, 5 per cent property, 5 per cent Asian credit. Then 50 per cent equities, 40 per cent equities, 6 per cent property, 4 per cent property. Or 90 per cent Asian credit, 1 per cent equities, 9 per cent bonds and no property. And so on.
That’s a lot of combinations — more than 150,000 if we keep only to integers. A matrix of all these portfolios is run and usually plotted on a chart, with expected returns on the Y-axis and risk shown on the X-axis.
The boundary formed along the top of the plot area is the so-called “efficient frontier”. It looks like an upside-down Nike swoosh. If you know the return you want, run your finger across and where it hits the frontier, the nearest portfolio will give it to you for the least risk (in theory).
Alternatively, you can choose a level of risk you are willing to take and run your eye northwards to find a mix of assets that maximises your return given this constraint.
There are loads of assumptions behind these outcomes. So many in fact that most academics now reckon the whole approach is bollocks. For starters, expected returns are based on long-run historical averages. These can differ from reality for years.
Measuring risk is also insanely hard. Modern portfolio theory uses the volatility (variance) of each asset class as a proxy, analysing how dispersed returns are versus historical averages. Again, numbers can deceive. Lots of small jitters are mathematically equal to a long calm followed by a whopping move. Upward volatility is considered as risky as deviations to the downside.
Another problem with modern portfolio theory is while it is clever at assessing the volatility of different combinations of asset classes, it fails to comprehend systemic risk. That is, when a major event sends everything off the rails.
Last year is a good example, when inflation forced central banks to increase rates quickly and stocks and bonds crashed in tandem. A nicely-balanced fund on modern portfolio theory lines wouldn’t have prevented you from losing a fortune.
This is why the likes of Warren Buffett remind us that shares beat all in the long run, so why bother? Others say risk-adjusted returns matter for investors with shorter timescales — and diversified portfolios are superior on this metric. Besides, a 60:40 allocation only trailed an equity-only one by one per cent per annum over the past half century.
I’m not going to rush to tweak my Sipp. However it would be useful to build an optimisation model that all readers can use online. Another for my to-do list.
The author is a former portfolio manager. Email:stuart.kirk@ft.com; Twitter:@stuartkirk__
