Fixing Margin Loan Interest Rate?

A year ago, I fixed most of my CommSec margin loan at a constant rate for the year ahead. The rate I got was 7.54% compared to a variable rate of 9.4%. I just got an email from CommSec asking whether I want to fix my rate again. The fixed rate is now 9.2%, but the variable rate has only increased to 9.65%. Clearly, I made the right choice to fix my loan, but this doesn't look like a good option going forward. Instead, I will reduce my borrowing probably by selling my gold ETF holdings, or at least some of them, especially if the recent budget CGT measures are passed by Parliament. The Greens hold the balance of power, but they think that nominal gains should be taxed at ordinary income tax rates and that there should be no grandfathering, so they might yet derail things for Labor.

Capital Gains Can Vary Radically Depending on the Currency They Are Measured In!

Our SMSF provider only completed our 2024-25 financial year accounts at the end of April. I check these carefully before signing off and paying the ATO. They have made mistakes in capital gains calculations in the past.

I thought that was the case this time too for our investments in the Fidelity Bitcoin ETF and Defi Technologies. So, I challenged their calculation. Their response was that the numbers were correct if I converted the purchases and sales to Australian Dollars using spot exchange rates on the exact days of the transactions. 

I don't use this method in my own tax accounting, as it is complicated. I just take the gain in USD or CAD and multiply by the exchange rate on 30th June of the tax year in question. I thought this was pretty close. It turns out it's not!

The gain on the bitcoin transaction was USD 19k. I downloaded the exchange rates from Pacific Exchange Rate Service, which I use for all my forex calculations, on each transaction day–there were many purchases–and multiplied the USD amounts by those exchange rates. Then I deducted the sum of all the purchases in AUD from the sale amount in AUD. The capital gain turned out to be AUD 53k! 

This is as if the Australian Dollar to US Dollar exchange rate was 35 cents, when it was never below 60 cents. The reason this happened was that I sold when the exchange rate was only 60 US cents but bought at higher exchange rates. Given the USD gain was not that big relative to the size of the transactions, the difference in exchange rates was levered up into a large AUD gain. I never thought something this extreme was possible. 

The Defi Technologies gains were not as radically different in the two currencies because I made more money on those relative to the size of the transactions. The tax bill for the bitcoin trade of AUD 8k is more than half of the SMSF's annual tax bill of AUD 14.4k.

54 Wellington

The Liberman Family ended up liquidating their property fund as a result of developing this office building. And ASA Diversified Property Fund, which our SMSF is a unitholder of, ended up buying it for a bargain price that is much less than replacement cost.

Australian Commonwealth Budget 2026

The budget speech was tonight, and all of the worst stuff that was leaked in the lead up to the budget appears to be in it. The worst is a minimum 30% tax on capital gains instead of up till now a maximum of 23.5%! Australia now has the most favourable tax treatment of dividends in the world and the least favourable treatment of capital gains. I'm sure Labor is thinking about that discrepancy. They were in 2019. I am exactly the type of person worst hit by this budget–a retiree probably in the 30% tax bracket.

The existing capital gains tax discount served two purposes. One is that taxing gains that are just due to inflation seems unfair (though we do that to interest payments), and the second is that companies already pay corporation tax on the profits they reinvest to generate capital gains. While franked dividends in Australia fully pass on to shareholders a credit for tax already paid, the current CGT regime partially does that. Whether or not replacing that with inflation indexation makes sense, a minimum 30% tax on capital gains seems especially unfair. 

I am already thinking about how to adapt to the new regime. I probably will sell my gold ETF holdings and replace them with futures contracts due to the new minimum 30% capital gains tax. And probably put the futures inside our SMSF for good measure. Freeing up the capital will allow me to reduce debt, contribute to super as much as I can and buy some dividend yielding investment instead.

Our Pershing Square Holdings, 3i, Berkshire Hathaway, Masterworks, and Angellist investments are all less attractive. Should also reconsider our accounts (outside the SMSF) with Colonial First State (CFS Imputation Fund and Acadian Global Long-Short Fund) that distribute capital gains. It doesn't make sense to get rid of all investments likely to have real capital gains. Instead, the expected rate of return needs to be high enough post-tax to hold onto those investments. 

Also, it seems that the 30% minimum CGT will only apply to the gains relative to the investment's value at 1 July 2027. So, there is no rush to make changes. This is going to greatly complicate tax calculations.

Pabrai Funds

Monish Pabrai is a well-known value investor who is often interviewed. He recently launched an actively managed ETF: WAGN listed on NASDAQ. After watching the interview and hearing about the ETF, I was curious about his performance. I found online his investor letter from 2020. PIF2, which started in 2000, returned 15.4% per year up to 2019 compared to the S&P 500's 7.1%. This sounds good, but the fund had extreme volatility, and so its information ratio was 0.54 compared to 0.48 for the S&P 500. You might still want to invest on this basis but PIF3, which started in 2002 returned 10.3% vs 8.2% for the S&P 500 with even higher volatility and PIF4 underperformed the S&P 500 also with extreme volatility. PIF4 has continued this pattern since 2020. The mutual fund and ETF have done well since inception in 2023 but I think it is likely that this pattern of high volatility continues. So, for now, I decided not to invest.

Education Bond Analysis

I've now done some spreadsheet simulations of education bonds. Assuming that our marginal personal tax rate will be 30%, the goal is to pay out fund earnings equal to the 30% tax threshold, which is currently $45k (everything in AUD of course) in each of the three years the children are in university. I assume a nominal fund return of 9% and inflation of 3% per year. I increase the amount saved each year by the rate of inflation. When the children finish university, all the contributions are removed.

For Little My, who is now in school year 1, you need to start by saving $17k in the scheme this year. In the end there will be $241k of contributions. The maximum tax saving over the 3 payout years is $40,600 in total and the management fees incurred over the course of the scheme are $18k. The cost of the management fees is a bit higher than this–about $3-$4k–due to compounding. So, the tax benefit is roughly double the extra cost. However, if Little My earned the equivalent of $18,200 per year–the current tax-free threshold–the tax benefit goes down to $16,500!

For Moomin, who is now in school year 5, you need to start by saving $31,750k in the scheme this year. In the end there will be $282k of contributions. The maximum tax saving over the 3 payout years is $36k in total and the management fees incurred over the course of the scheme are $15,500k. The cost of the management fees is again a bit higher than this due to compounding. So, the tax benefit is roughly double the extra cost. However, if Moomin earned the equivalent of $18,200 per year–the current tax-free threshold–the tax benefit goes down to $14,900.

Here is Little My's analysis:

Of course, if they don't go to uni, but work instead, you pay a whole load of management fees and get into maybe a suboptimal investment for nothing. 

Also, up till now I have assumed that all our income is ordinary income. If instead it is all long-term capital gains–for example from selling gold ETF shares–then the tax benefit is halved and the maximum tax benefit is equal to the management fees. Of course, the 50% capital gains discount might not exist in the future–Labor wants to abolish it. If all our income came from franked dividends, then in the 30% tax bracket we would pay no tax on these anyway and so there would be no tax benefit, just management fees!

P.S. 

It seems unlikely that the two of us would earn more than $270k between us outside of super, but just for completeness, I looked at the case where we are in the 37% tax bracket.  In this case the tax benefit for Little My assuming they don't work and keeping all other assumptions the same, rises to $54k or three times the management fee. But again, if all our income came from capital gains that would only be $27k and if all our income came from franked dividends our tax rate on the grossed up dividend is only 7% after the franking credit or 10% of the net dividend. This means there is only a tax benefit on the first $18.2k paid out for education expenses, which would be negated if Little My worked. So, I don't think I am going to do this.

Education Bonds

I just discovered an investment structure I had never heard of: Education Bonds. These are an Australian investment structure that is similar to investment bonds but with some twists. We have an investment bond in Little My's name at Generation Life. We used it to invest the money he inherited from my mother. 

First, I will describe an investment bond again. It is an investment that pays tax in the fund nominally at 30%. If you hold it for 10 years and then withdraw the money you don't pay any additional tax. If you withdraw it before 10 years you owe tax on the earnings at your regular tax rates but get a 30% tax offset. You can reset the 10 year term by contributing a new investment of more than 125% of the previous year's investment. Why would you want to do that? If your tax rate or a child who you made the beneficiary end up having a tax rate below 30%, you'll pay less tax then if you withdraw the money.

Most of this applies to an education bond too. These are the differences:

1. You can withdraw the contributions without tax or penalty at any time. Only the earnings are locked up for 10 years.

2. You can make a claim to pay for education expenses and withdraw earnings to do so. When you do this, you get the tax paid added onto the amount you withdraw. So, there is no tax in the fund on these withdrawals. This can be done at any time, not just after 10 years.

3. The twist is that the beneficiary whose education you are paying for is liable for tax on the earnings. Children under 18 have very high penalty tax rates (one reason we used an investment bond for Little My). So, beyond the tax-free $416 per year this really wouldn't make sense. Once they turn 18, the regular adult rates apply including the tax free threshold.

4. Here is the really interesting part: You can keep any education bills incurred since you started the education bond and claim them in a later year. So, you could claim school tuition from 2026 in 2036 say!

5. If you withdraw all your contributions and then want to withdraw earnings without valid education bills, the standard investment bond rules apply.

6. The downside is you are limited to the investment options the provider has and an additional administration fee. After all, they have to deal with all these education claims... For Australian Unity this additional fee is 0.7% p.a. 

There are only a few providers and so far Australian Unity seems most attractive. Generation Life don't offer this product.

Basically, this is a way of tax-sheltering some investment income in a similar way to income splitting through a family trust. But it is much more restrictive on investments and possibly has higher fees (our SMSF pays 0.3% p.a.). You are only really going to be directly paying for higher education expenses using this.

For someone in my position, it might make sense after you have maxed out your tax free super pension and you are already above the tax-free bracket of income tax on your non-super earnings, which is true in my case. The problem is that actually trying to reclaim all the children's private school fees during the 3 or so years they are in Uni would push them into the 30% marginal tax bracket, which is probably where I will be myself. If they are working part time they might already use up the tax free allowance (currently AUD 18.2k), which would make the tax savings small. And this is assuming they go to Uni. With these considerations, the 0.7% annual fee, and limited investment options, I am undecided if this is worthwhile.

 

April 2026 Report

Markets rebounded this month, but they rebounded a lot less in Australia than in the rest of the world. This was partly because of a strong rebound in the Australian Dollar from USD 0.6880 to USD 0.7179. Gold fell a little in USD terms and quite a lot in AUD terms. Here is the performance of our benchmarks (total returns including dividends):

US Dollar Indices

MSCI World Index (gross): 10.21%

S&P 500: 10.49%

HFRI Hedge Fund Index (forecast): 2.71%

Australian Dollar Benchmarks

ASX 200: 2.19%

Target Portfolio (forecast): 1.97%

Australian 60/40 benchmark: 2.44%

In Australian Dollar terms we gained 2.88% and in US Dollar terms we gained 7.35%. So we outperformed all AUD benchmarks and HFRI but underperformed relative to the two USD stock indices. Our SMSF gained 4.62%. Unisuper gained 4.49% and PSS(AP) 2.34%. So, we outperformed one of our superannuation benchmarks.

Here is a report on the performance of investments by asset class:

The asset class returns are in currency neutral terms as the rate of return on gross assets and do not include investment expenses such as margin interest, and so the total differs from the Australian Dollar returns on net assets mentioned above. Gold and Australian small cap lost money but all other asset classes gained. Futures were the best performer while hedge funds made the greatest contribution. 

Things that worked well this month:

• Nine investments gained more than AUD 10k: L1 Global Long-Short (GLS.AX, 42k), Unisuper (35k), Australian Dollar Futures (35k), Pershing Square Holdings (PSH.L, 19k), Acadian Global Long-Short (19k), PSS(AP) (15k), Pengana Private Equity (PE1.AX, 15k),  CREF Social Choice (11k), and Hearts and Minds (HM1.AX, 11k).

What really didn't work:

• Only three investments lost money with gold losing 28k.

We moved towards our target allocation. Our actual allocation currently looks like this:

Almost 70% of our portfolio is in what are often considered to be alternative assets: real estate, art, hedge funds, private equity and credit, gold, and futures. A lot of these are listed investments or investments with daily liquidity, so our portfolio is not as illiquid as you might think.

Moominmama receives employer superannuation contributions every two weeks. We also make monthly concessional contributions to Moominmama's superannuation to reach the annual cap on contributions. There will still be capital calls from Aura Venture Fund II and III. I am receiving monthly pension payments from both Unisuper and our SMSF totalling AUD 5,150 per month. I was again less active in the market, making the following investment and trade moves this month:

• I reinvested a distribution at Masterworks and invested USD 2k in another startup at Unpopular Ventures on Angellist. We are no longer subscribing to their rolling fund but will invest a little in promising startups that they syndicate.
• I sold 1,000 shares of PMGOLD.AX (10 ounces) and then bought back in again at a lower price, but not low enough, as the price has fallen more.
• I bought another 10k shares in WAM Alternatives (WMA.AX).

Here are the income and spending accounts * for this month:

Other income includes Moominmama's salary and employer superannuation contributions and totalled AUD 4k as usual. It was a big spending month at AUD 19k due to school fees. This number does not include our mortgage payments, which are regarded here as saving and investment costs. Dissaving amounted to AUD 15k, within the 4% rule limit of AUD 23k. We gained AUD 193k investing, 2/3 of which were was in retirement accounts. They performed better both on the way down and the way up from the March correction than our non-retirement accounts and are now higher than in February. We received lots of dividends with associated franking credits this month. As a result of all this, net worth rose by AUD 162k to AUD 8.309 million. We are about AUD 100k above the beginning of the year, but this is mainly due to the increase in the value of our house.

* Results are shown separately for retirement and non-retirement accounts as well as housing, which nowadays doesn't have much activity. The grey shaded rows are additional notes. Total investment income is split into investment income before exchange rate moves and the contribution of exchange rates. Other income is non-investment income including salaries, employer superannuation contributions, and net tax returns. Investment income is shown pre-tax. Tax credits include franking credits on Australian Dividends and imputed tax on industry superannuation returns and and actual SMSF tax. These are taken away from investment income to get changes in actual net worth. Inheritances include gifts from relatives. Saving is from non-investment income, transfers, and inheritances.