On the Wind – Reviewing the Record of Distributions and Expenses

IMG_20190117_150101_027
But no new findings will ever be made if we rest content with the findings of the past.
Seneca, Letters XXXVII

Measuring distributions and expenses

Over the last three years the investment portfolio has delivered substantial distributions, leading to a brief period in which it appeared that an accidental goal of ‘Credit card FI’ might have been met. Subsequently that prospect receded, due to a sharply lower set of distributions for the half year to December 2018.

Over six months ago Reviewing the Log examined the issue of how my current passive income from distributions compared against both my credit card expenditure and total spending.

This article seeks to update that previous analysis, but also to go further and reach a fuller and more robust picture of overall trends in how distributions and expenses compare over time.

In particular, this article seeks to identify the likely current ‘gap’ between distributions and monthly expenses. This represents a different and arguably more empirical way of viewing and measuring actual day-to-day progress to FI, compared to simply tracking progress to a numerical portfolio goal.

Even so, they are in some senses also different sides of the same coin. This is because the portfolio goals I am aiming for are reverse engineered from target FI income levels, which are translated into lump sum targets, using an assumed average return (currently 4.19 per cent). Each month I report a percentage progress towards these goals. Currently, by this simple lump sum measure, the portfolio is around 90 per cent of the way to Objective #1 and just under 75 per cent of the way to Objective #2.

Re-examining the logs and records

When the monthly record of credit card expenses, total expenses and distributions is examined it is clear that credit card expenditure is volatile, but has a comparatively stable average of around $6000 per month, or around $72 000 per annum. The distributions, on the other hand, have been either stable or growing for most of the past six years, with the exception of the large reduction in the half year to December 2018. During this last half year to December, distributions averaged at around $2 600 a month.

The chart below sets out a ‘credit card only’ (blue) and a ‘total expenses’ series against an averaged measure of monthly portfolio distributions (in red). The green line represents actual credit card expenses, added to an equal monthly contribution of other non-credit card expenses. Total expenses here just includes items such as rates, energy and utility costs, day to day cash, as well as contributions to irregular major expenses such as holidays, house and car repairs, as well as eventual car replacement. Fig 1 - monthly

Note that all segments of the red line reflect annual distributions, except the last period from July 2018 onwards. The red line from July 2018 to the present will need to be revised once the June 2019 half year distributions are known.

This revision is highly likely to lift the currently assumed average distribution for 2018-19 of about $2600 per month. This lift is likely because currently the red line from July 2018 onwards is based on a simple extrapolation or continuation of the traditionally lower December figures. The true underlying level of distributions this financial year may well be higher. In fact, June half year distributions have usually been well above the interim dividend amounts of the December half year.

Depending on the estimation method used, the June 2019 half year distribution could be in the range of $23 000 to $51 000, with an average estimate of around $42 000. This in turn could lead to total annual distributions for financial year 2018-19 being in the range of $39 000 to $66 000 (or between $3 250 to $5 500 per month respectively). For comparison, the five year average of distribution is around $45 000 (or $3 750 per month). The final figure will simply be an unknown factor until early July.

Off-course or temporary shallows?

The same considerations are relevant for examining a second measure of progress. The below figure charts the proportion of total expenses met by annual distributions.Fig 2 - Total Ex DistSince the last update of this graph more than six months ago, the proportion of expenses met by portfolio distributions has fallen, and for the same reason – the low distributions in the half year to December 2018.

Even with this significant fall, from July 2018 to the present, these lower distributions have generally been sufficient to meet between 30 to 40 per cent of total expenses. In overall trend terms, it also suggests the true underlying distributions potential of the portfolio is likely to be sitting at around 60 to 70 per cent (see dotted trend line).

Looking through the weather – adjusting the view

These two ways of viewing progress each have their advantages, but suffer the same disadvantage of being volatile measures of progress. This volatility arises from both monthly variations in expenses, and large variations in distributions between and within years. These variations occur due to a range of factors, such as realisations of capital gains related to rebalancing within some pre-mixed Vanguard retail funds, as well as changes in bond yields or interest rates.

To address this the following chart seeks to account for these multiple sources of variation by adopting a three year moving average for both credit card expenses and distributions. The trade-offs in using this approach is that a three year moving average reduces the time period able to be covered, and can also mute broader emerging trends that should be of concern. Additionally, three years is not close to a complete economic cycle. Thus it is quite possible, for example, for distributions that are abnormally high for two consecutive years to impact this moving average measure.

The advantages of an averaged approach are obvious, however. By reducing the variations and monthly ‘noise’, and taking a relatively conservative assumption (in an increasing portfolio) that the last three years may provide an approximate guide to the true underlying level of distributions, a clearer and more stable picture of overall progress can be gained.Fig 3 no outlineFrom this particular view, a few points emerge:

  • Credit card expenses have remained very stable at around $6 000 across the past three years with no systematic movements up or down
  • From January 2017 onwards distributions increased steadily until they reached around $5 000 per month in the middle of 2018
  • Since that time they have levelled off, and even slightly reduced, as the lower recent distributions form a greater part of the average

The data in the chart suggests a remaining gap of approximately $1 000 per month between distributions and credit card expenses, or distributions being sufficient to meet approximately 80 per cent of credit card expenses or 60 per cent of total expenses. In turn, this means that viewed as a multi-year average, ‘credit card FI’ has not been receding as sharply as volatile month to month figures suggest. It remains, in short, in view if not yet in range. The true average gap measured in these term is likely to continue to gently increase in the lead up to June 2019 distributions, but then potentially either level off or continue to close.

Overall this measure better reflects how the journey has felt so far. A beginning from a firm basis, constant steady progress over the time of the journey, but some significant distance to close yet.

Taking new bearings – an alternative approach

To reach the best view of where one is, it is sometimes useful to use cross-checks that relies on slightly different data.

An alternative approach to reaching a sound estimate which takes into account more stable annual data is to use tax assessment data. The chart below is based on assessed taxable investment income. It is taken from the tax return items of income from partnerships and trusts, foreign source income, and franking credits (i.e. items 13, 20 and 24, excluding capital gains) over the past ten years. This taxable income then given as a proportion of my portfolio objective #1, of $67 000 per year.

Fig 4 - TaxableFrom this chart some observations can be made:

  • For the past three years the equivalent of around 50 to 60 per cent of my first financial independence objective of $67 000 has been met by investment income
  • The past two years have been materially higher than other years – this could perhaps represent an anomaly, however, the overall portfolio that was producing distributions also grew by around 70 per cent since 2015-16, which would tend to support the higher later figures being sustainable
  • Annual variations do occur – with two out of 10 years registering some backward movement

The picture from taxable investment income then seems to support a gradual movement over the past three financial years materially closer to Objective #1, and some confidence that this is more likely than not to be maintained in the current financial year. Taking a three year average it suggests in investment income terms that around 55 to 60 per cent of Objective #1 income is likely to be covered by current distributions.

Summary – on the wind or a voyage becalmed?

Looking at the data highlights a few different points. Progress is not always linear, or exponential, even with compounding effects and well into the FI journey. Yet equally it shows it is possible over the course of several years to go from distributions making a small supporting contribution to ongoing expenses, to the equivalent of paying off the majority of a monthly credit card bill.

From reviewing the records and expanded data it is apparent that ‘credit card FI’ – not exactly a universally recognised stage of FI – is not yet achieved. Longer term progress on the goal will be clearer when June distributions are finalised in the next three months.

Depending on their final levels, between 55 and 90 per cent of annual credit card expenses will be covered by annual distributions. Reviewing past averages of card expenditures and distributions indicates that about 80 per cent of journey may be complete already, leaving a gap of only $1 000 per month.

Moving beyond credit card expenses – the lower distributions over the past six months have been equivalent to only 30 to 40 per cent of total expenses. Using independent tax assessment data indicates that the portfolio is currently generating between 50 and 60 per cent of the total yearly expenditure target under Portfolio Objective #1, with recent portfolio growth meaning the higher end of this range is a more probable guide than the lower.

In the first examination of these trends more than six months ago I observed the inevitable issue of volatility and noted that is was not impossible for future periods of higher expenditure to coincide with lower portfolio income. This could still occur, and clear precedents exist for it. Averages and forecasts have the power to mislead as well as guide.

Yet overall, looking back at the record puts some firm underpinnings to the progress already made – and leads me to strain forward for the next set of bearings.

Monthly Portfolio Update – March 2019

IMG_20190121_140755_576
Everyone complains of his memory, none of his judgement.
La Rochefoucauld, Maxims

This is my twenty-eighth portfolio update. I complete this update monthly to check my progress against my goals.

Portfolio goals

My recently revised objectives are to reach a portfolio of:

  • $1 598 000 by 31 December 2020. This should produce a real income of about $67 000 (Objective #1)
  • $1 980 000 by 31 July 2023, to produce a passive income equivalent to $83 000 (Objective #2)

Both of these are based on an expected average real return of 4.19%, or a nominal return of 7.19%, and are expressed in 2018 dollars.

Portfolio summary

  • Vanguard Lifestrategy High Growth Fund – $732 134
  • Vanguard Lifestrategy Growth Fund  – $42 428
  • Vanguard Lifestrategy Balanced Fund – $76 692
  • Vanguard Diversified Bonds Fund – $104 802
  • Vanguard Australia Shares ETF (VAS) – $78 091
  • Betashares Australia 200 ETF (A200) – $216 609
  • Telstra shares (TLS) – $1 769
  • Insurance Australia Group shares (IAG) – $13 393
  • NIB Holdings shares (NHF) – $6 288
  • Gold ETF (GOLD.ASX)  – $83 212
  • Secured physical gold – $13 437
  • Ratesetter (P2P lending) – $26 147
  • Bitcoin – $63 947
  • Raiz app (Aggressive portfolio) – $14 491
  • Spaceship Voyager app (Index portfolio) – $1 751
  • BrickX (P2P rental real estate) – $4 621

Total value: $1 479 910 (+$40 302)

Asset allocation

  • Australian shares – 41.6% (3.4% under)
  • Global shares – 23.6%
  • Emerging markets shares – 2.7%
  • International small companies – 3.5%
  • Total international shares – 29.9% (0.1% under)
  • Total shares – 71.5% (3.5% under)
  • Total property securities – 0.3% (0.3% over)
  • Australian bonds – 6.1%
  • International bonds – 11.2%
  • Total bonds – 17.3% (2.3% over)
  • Cash – 1.2%
  • Gold – 6.5%
  • Bitcoin – 4.3%
  • Gold and alternatives – 10.9% (0.9% over)

Presented visually, below is a high-level view of the current asset allocation of the portfolio.Mar 19 alloc

Comments

This month saw the total portfolio reach and exceed the original portfolio objective set at the commencement of this journey of $1 476 000.

Since that time, portfolio goals have been updated, but nonetheless it feels as though a significant milestone has passed. Measured over the past twelve months, strong progress has resumed, that being in part a function of the dull echoes of ‘Bitcoin bubble’ of late 2017 falling out of the time period.

Mar 19 Monthly valueThe portfolio increased by a significant $40 000 this month. Part of this was new investments in Betashares A200, and a majority of this gain is attributable to the second instalment from the lowering of my emergency fund discussed here being invested. In the end, averaging two parts of this lump sum into the equity market around three months apart did not make much difference, except perhaps a mild psychological benefit. Together these moves made up the majority of the total portfolio gains. The value of the small Bitcoin holding has also increased slightly, despite its volatility having substantially reduced over the past year.

Mar 19 mnth chnge

Another milestone this month has been the finalisation of my first significant March quarter dividend from A200, which will total around $1900. This is lower than expected, being equivalent to around 0.9% for the quarter, and lower than the expected distribution rate of the broadly equivalent Vanguard VAS ETF. It is quite possible that the end of financial year results will be better, however, and on a total returns basis the A200 ETF has still tracked its benchmark closely.

With Australian equities continuing to stay close to their long term price-earnings ratio of 15, Australian equity ETFs will likely remain the primary focus of investment over the next few months. This has been one of the most dominant trends of the journey so far, with total Australian equity holdings growing from around $277 000 in January 2017, and 28 per cent of the portfolio, to around $600 000 this month, and over 40 per cent.

A small action this month has been passing up the option of further investment in Australian real estate through BrickX. Distributions had built up to a level to allow a further small fractional investment. The Australian residential debate continues in full force, however, I cannot justify even small further incremental investments at current low yields, especially given my view of the likelihood of further capital losses.

Overall, expenses continue to track at steady levels. The low red distributions line from July 2018 onwards is a product of low December half distributions, and may be able to be revised upwards once June distributions are known. This would be a welcome revision, as ‘credit card’ FI seemed to come into view through the last two years, and then disappear in a discouraging way with the December 2018 distributions.Mar 19 - Card

Progress

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

Measure Portfolio All Assets
Objective #1 – $1 598 000 (or $67 000 pa) 92.6% 128.8%
Objective #2 – $1 980 000 (or $83 000 pa) 74.8% 104.0%
Credit card purchases – $73 000 pa 85.0% 118.2%
Total expenses – $96 000pa 64.6% 89.9%

Summary

This month has brought the portfolio to approximately three-quarters of the way to my Objective #2, and Objective #1 also draws appreciably closer.

Over the past month, as progress has accumulated, I have found myself meditating more fully on the nature and value of time and freedom. What has surprised is the powerful but gradual feeling of decompression that knowledge of the increasing proximity to the goals has brought. Fewer external events, daily stresses, impinge on my daily outlook.

This recent podcast from the Econtalk series, crystallised some of these thoughts, the first 10 minutes contains the best economic and empirical analysis I have encountered on the intersection of time, money, leisure and work. One of its key points – relevant for seekers of FI – is that our growing wealth over time affects how we see and value leisure time itself, and it also has some useful reflections on the concept of ‘busyness’.

This month has seen some of this more valuable leisure time used looking at the summary version of the Credit Suisse Global Investment Returns Yearbook, released last month. This provides updated data from the single best long-term series on equity and bond returns across the world. One interesting aspect of this years updated estimate of long-term historical global equity returns (of 5 per cent) is that it includes for the first time markets that suffered total losses (Russia and China following revolutions in 1917 and 1949) – addressing the issue of survivorship bias. The report argues for significant modesty in expectations of future returns.

A practical implication of this is that my conservative long-term return assumption for global equities (of 4.5 per cent) may be marginally less conservative than when it was made at the start of the year. This podcast from Bloomberg, interviewing Yale Professor of Finance Roger Ibbotson – a key figure in the collection and analysis of historical financial market returns – will provide more related food for thought.

A further intriguing crossover from recent economic literature to FI issues is the release last week of this paper The Power of Working Longer by the National Bureau of Economic Research, which studies the relationship between the decision to work for longer, compared to investing more, prior to retirement. The intriguing summary finding is that delaying retirement by 3-6 months is equivalent to the effect of one percentage point of higher wages over a 30 year working life.

The Australian FI community has also been full of interesting content this month, with Aussie Firebug laying out the basics of FI in an excellent introductory podcast, and Strong Money Australia doing a short summary of his current progress in transitioning from property investment dominated portfolio to equities. Australian investor’s benefits  from higher dividend rates also got a mention in Big ERN’s comprehensive safe withdrawal rates series. It was also great to see the appearance and progress of other new Australian FI bloggers, such a AFamilyOnFire.

With the month closed, the focus will now be shifting to awaiting and re-investing the quarterly dividends due, and contemplating that a further three months comparable to the last three – an unlikely but possible scenario – could see Objective #1 reached much earlier than my judgement had anticipated.