We would rather be ruined than changed.

W H Auden, The Age of Anxiety

This is my forty-third portfolio update. I complete this update monthly to check my progress against my goal.

Portfolio goal

My objective is to reach a portfolio of $2 180 000 by 1 July 2021. This would produce a real annual income of about $87 000 (in 2020 dollars).

This portfolio objective is based on an expected average real return of 3.99 per cent, or a nominal return of 6.49 per cent.

Portfolio summary

• Vanguard Lifestrategy High Growth Fund – $726 306
• Vanguard Lifestrategy Growth Fund – $42 118
• Vanguard Lifestrategy Balanced Fund – $78 730
• Vanguard Diversified Bonds Fund – $111 691
• Vanguard Australian Shares ETF (VAS) – $201 745
• Vanguard International Shares ETF (VGS) – $39 357
• Betashares Australia 200 ETF (A200) – $231 269
• Telstra shares (TLS) – $1 668
• Insurance Australia Group shares (IAG) – $7 310
• NIB Holdings shares (NHF) – $5 532
• Gold ETF (GOLD.ASX)  – $117 757
• Secured physical gold – $18 913
• Ratesetter (P2P lending) – $10 479
• Bitcoin – $148 990
• Raiz app (Aggressive portfolio) – $16 841
• Spaceship Voyager app (Index portfolio) – $2 553
• BrickX (P2P rental real estate) – $4 484

Total portfolio value: $1 765 743 (+$8 485 or 0.5%)

Asset allocation

• Australian shares – 42.2% (2.8% under)
• Global shares – 22.0%
• Emerging markets shares – 2.3%
• International small companies – 3.0%
• Total international shares – 27.3% (2.7% under)
• Total shares – 69.5% (5.5% under)
• Total property securities – 0.3% (0.3% over)
• Australian bonds – 4.7%
• International bonds – 9.4%
• Total bonds – 14.0% (1.0% under)
• Gold – 7.7%
• Bitcoin – 8.4%
• Gold and alternatives – 16.2% (6.2% over)

Presented visually, below is a high-level view of the current asset allocation of the portfolio.

Comments

The overall portfolio increased slightly over the month. This has continued to move the portfolio beyond the lows seen in late March.

The modest portfolio growth of $8 000, or 0.5 per cent, maintains its value at around that achieved at the beginning of the year.

The limited growth this month largely reflects an increase in the value of my current equity holdings, in VAS and A200 and the Vanguard retail funds. This has outweighed a small decline in the value of Bitcoin and global shares. The value of the bond holdings also increased modestly, pushing them to their highest value since around early 2017.

There still appears to be an air of unreality around recent asset price increases and the broader economic context. Britain’s Bank of England has on some indicators shown that the aftermath of the pandemic and lockdown represent the most challenging financial crisis in around 300 years. What is clear is that investor perceptions and fear around the coronavirus pandemic are a substantial ongoing force driving volatility in equity markets (pdf).

A somewhat optimistic view is provided here that the recovery could look more like the recovery from a natural disaster, rather than a traditional recession. Yet there are few certainties on offer. Negative oil prices, and effective offers by US equity investors to bail out Hertz creditors at no cost appear to be signs of a financial system under significant strains.

As this Reserve Bank article highlights, while some Australian households are well-placed to weather the storm ahead, the timing and severity of what lays ahead is an important unknown that will itself feed into changes in household wealth from here.

Investments this month have been exclusively in the Australian shares exchange-traded fund (VAS). This has been to bring my actual asset allocation more closely in line with the target split between Australian and global shares.

A moving azimuth: falling spending continues

Monthly expenses on the credit card have continued their downward trajectory across the past month.

The rolling average of monthly credit card spending is now at its lowest point over the period of the journey. This is despite the end of lockdown, and a slow resumption of some more normal aspects of spending.

This has continued the brief period since April of the achievement of a notional and contingent kind of financial independence.

The below chart illustrates this temporary state, setting out the degree to which portfolio distributions cover estimated total expenses, measured month to month.

There are two sources of volatility underlying its movement. The first is the level of expenses, which can vary, and the second is the fact that it is based on financial year distributions, which are themselves volatile.

Importantly, the distributions over the last twelve months of this chart is only an estimate – and hence the next few weeks will affect the precision of this analysis across its last 12 observations.

Estimating 2019-20 financial year portfolio distributions 

Since the beginning of the journey, this time of year usually has sense of waiting for events to unfold – in particular, finding out the level of half-year distributions to June.

These represent the bulk of distributions, usually averaging 60-65 per cent of total distributions received. They are an important and tangible signpost of progress on the financial independence journey.

This is no simple task, as distributions have varied in size considerably.

A part of this variation has been the important role of sometimes large and lumpy capital distributions – which have made up between 30 to 48 per cent of total distributions in recent years, and an average of around 15 per cent across the last two decades.

I have experimented with many different approaches, most of which have relied on averaging over multi-year periods to even out the ‘peaks and troughs’ of how market movements may have affected distributions. The main approaches have been:

• An ‘adjusted income’ approach – stripping out the capital gains components of Vanguard funds to reach an estimate of underlying income generation, both across the entire investment period, and during the sharpest low of the Global Financial Crisis
• A long-term asset class approach – relying on long-term historical data on averages of the income produced by various asset classes
• A ‘tax method’ approach – this derives an income estimate as a percentage of the portfolio by drawing on taxable investment income totals from tax return records
• Simple historical rolling average – this is a rolling three-year measure, based on the actual distributions record of the portfolio
• Average distribution rate approach – this method uses a long-term average of annual distributions received as a percentage of the total portfolio since 1999

Each of these have their particular simplifications, advantages and drawbacks.

Developing new navigation tools

Over the past month I have also developed more fully an alternate ‘model’ for estimating returns.

This simply derives a median value across a set of historical ‘cents per unit’ distribution data for June and December payouts for the Vanguard funds and exchange traded funds. These make up over 96 per cent of income producing portfolio assets.

In other words, this model essentially assumes that each Vanguard fund and ETF owned pays out the ‘average’ level of distributions this half-year, with the average being based on distribution records that typically go back between 5 to 10 years.

Mapping the distribution estimates

The chart below sets out the estimate produced by each approach for the June distributions that are to come.

Some observations on these findings can be made.

The lowest estimate is the ‘adjusted GFC income’ observation, which essentially assumes that the income for this period is as low as experienced by the equity and bond portfolio during the Global Financial Crisis. Just due to timing differences of the period observed, this seems to be a ‘worst case’ lower bound estimate, which I do not currently place significant weight on.

Similarly, at the highest end, the ‘average distribution rate’ approach simply assumes June distributions deliver a distribution equal to the median that the entire portfolio has delivered since 1999. With higher interest rates, and larger fixed income holdings across much of that time, this seems an objectively unlikely outcome.

Similarly, the delivery of exactly the income suggested by long-term averages measured across decades and even centuries would be a matter of chance, rather than the basis for rational expectations.

Central estimates of the line of position

This leaves the estimates towards the centre of the chart – estimates of between around $28 000 to $43 000 as representing the more likely range.

I attach less weight to the historical three-year average due to the high contribution of distributed capital gains over that period of growth, where at least across equities some capital losses are likely to be in greater presence.

My preferred central estimate is the model estimate (green) , as it is based in historical data directly from the investment vehicles rather than my own evolving portfolio. The data it is based on in some cases goes back to the Global Financial Crisis. This estimate is also quite close to the raw average of all the alternative approaches (red). It sits a little above the ‘adjusted income’ measure.

None of these estimates, it should be noted, contain any explicit adjustment for the earnings and dividend reductions or delays arising from COVID-19. They may, therefore represent a modest over-estimate for likely June distributions, to the extent that these effects are more negative than those experienced on average across the period of the underlying data.

These are difficult to estimate, but dividend reductions could easily be in the order of 20-30 per cent, plausibly lowering distributions to the $23 000 to $27 000 range. The recently announced forecast dividend for the Vanguard Australian Shares ETF (VAS) is, for example, the lowest in four years.

As seen from chart above, there is a wide band of estimates, which grow wider still should capital gains be unexpectedly distributed from the Vanguard retail funds. These have represented a source of considerable volatility. Given this, it may seem fruitless to seek to estimate these forthcoming distributions, compared to just waiting for them to arrive.

Yet this exercise helps by setting out reasoning and positions, before hindsight bias urgently arrives to inform me that I knew the right answer all along. It also potentially helps clearly ‘reject’ some models over time, if the predictions they make prove to be systematically incorrect.

Progress

Progress against the objective, and the additional measures I have reached is set out below.

Measure	Portfolio	All Assets
Portfolio objective – $2 180 000 (or $87 000 pa)	81.0%	109.4%
Credit card purchases – $71 000 pa	98.8%	133.5%
Total expenses – $89 000 pa	79.2%	106.9%

Summary

The current coronavirus conditions are affecting all aspects of the journey to financial independence – changing spending habits, leading to volatility in equity markets and sequencing risks, and perhaps dramatically altering the expected pattern of portfolio distributions.

Although history can provide some guidance, there is simply no definitive way to know whether any or all of these changes will be fundamental and permanent alterations, or simply data points on a post-natural disaster path to a different post-pandemic set of conditions. There is the temptation to fit past crises imperfectly into the modern picture, as this Of Dollars and Data post illustrates well.

Taking a longer 100 year view, this piece ‘The Allegory of the Hawk and Serpent‘ is a reminder that our entire set of received truths about constructing a portfolio to survive for the long-term can be a product of a sample size of one – actual past history – and subject to recency bias.

This month has felt like one of quiet routines, muted events compared to the past few months, and waiting to understand more fully the shape of the new. Nonetheless, with each new investment, or week of lower expenditure than implied in my FI target, the nature of the journey is incrementally changing – beneath the surface.

Small milestones are being passed – such as over 40 per cent of my equity holdings being outside of the the Vanguard retail funds. Or these these retail funds – which once formed over 95 per cent of the portfolio – now making up less than half.

With a significant part of the financial independence journey being about repeated small actions producing outsized results with time, the issue of maintaining good routines while exploring beneficial changes is real.

Adding to the complexity is that embarking on the financial journey itself is likely to change who one is. This idea, of the difficulty or impossibility of knowing the preferences of a future self, is explored in a fascinating way in this Econtalk podcast episode with a philosophical thought experiment about vampires. It poses the question: perhaps we can never know ourselves at the destination? And yet, who would rationally choose ruin over any change?