Strategy commentary cont'd:

Strategy commentary cont'd:

The US-Iran conflict continued to de-escalate in June in line with our long-held central case that the guts of the conflict had been concluded by the end of March. While there were brief moments where tensions threatened to re-intensify, the dominant market narrative shifted towards resolution after President Trump posted mid-month that the deal with Iran was "now complete". He subsequently authorised an end to the naval blockade, allowing traffic through the Strait of Hormuz to recover. Brent crude fell 20.8% over the month to US$72.92/bbl, leaving it just 0.6% above pre-conflict levels, although prices remained around 20% higher year-to-date. WTI crude similarly declined 20.4% to US$69.50/bbl.

Despite elevated energy prices for much of the year, the US economy remained buoyant, with employment data firm. Combined with the appointment of Kevin Warsh as a more hawkish Fed Chair, this prompted a sharp repricing of US monetary policy expectations. Markets moved from pricing 14bps of Fed rate hikes at the start of the month to 38bps by month-end, driving a bear-flattening of the US rates curve as investors priced a more restrictive policy path. The shift in the Fed backdrop weighed heavily on precious metals, with gold falling 11.7% to US$4,008/oz and silver declining 22% over the month.

Coolabah's theses around re-dollarisation and a globally synchronized hiking cycle have started to grip.

Other major central banks also leaned tighter, with both the ECB and the Bank of Japan raising cash rates by 25bps. Unlike in the US, however, market pricing for year-end policy rates was little changed, and this divergence was reflected in sovereign bond performance.

Strategy commentary cont'd: US Treasuries underperformed most global peers, with the 10-year Treasury yield rising 3bps to 4.47%, while benchmark 10-year yields declined across much of Europe: German Bund yields fell 8bps to 2.86%, UK Gilt yields declined 6bps to 4.76%, and Italian BTP yields fell 2bps to 3.63%. JGB 10-year yields ended the month 1bp higher at 2.67%, while French OAT yields rose 2bps to 3.65%. Sovereign spreads drifted wider, with the OAT/Bund spread widening 10.1bps to 79.2bps and the BTP/Bund spread moving 5.6bps wider to 77.0bps.

The same regional divergence was evident in risk assets. The S&P 500 returned -0.95% in June and the Nasdaq 100 returned -0.12%, while bitcoin fell 20.3% to US$58,642, pressured by the more hawkish US rates backdrop and the re-dollarisation dynamic.

By contrast, European and Japanese equities performed strongly, with the Euro Stoxx 50 returning 4.7%, Euro Stoxx Banks appreciating 7.6%, the FTSE 100 gaining 1.0%, and the Nikkei 225 returning 5.7%.

Credit markets were more stable. In synthetic indices, US CDX IG was little changed at 51bps and CDX HY widened 5.7bps to 306bps, while in Europe iTraxx Main tightened 1.1bps to 52bps, iTraxx Xover rallied 14.3bps to 245bps, and the senior financials index tightened 0.9bps to 54bps.

Cash credit spreads were similarly steady, with US investment-grade corporate spreads widening only 2bps to 74bps, euro aggregate corporate spreads widening 1bp to 79bps, and sterling aggregate corporate spreads 1bp wider at 86bps. Global corporate bond benchmarks nonetheless produced positive returns, with the Global Aggregate Corporate Index, hedged to US dollars, rising 0.32% in June, while the duration-hedged equivalent gained 0.05%. Coolabah's Active Global Bond Fund outperformed, which is benchmarked against the Global Aggregate Corporate Index, returning 0.53% net.

In currency markets, the euro declined 2.0% against the US dollar to 1.142, while the US dollar rose 2.1% against the yen to 162.55, consistent with the more hawkish repricing of the Fed.

June typically marks the beginning of the summer lull in issuance, but primary supply was materially stronger than expected in both the US and Europe, with both markets recording their busiest June on record. In the US, investment-grade supply totalled US$206bn, taking year-to-date issuance to US$1.24tn, around 32% ahead of last year's pace.

Financials accounted for 38% of the month's supply, with deals from more than 16 issuers, including Macquarie, Bank of New Zealand and four Japanese banks, alongside a number of European names. The most notable transactions, however, came from the hyperscalers, with NVIDIA and SpaceX each printing US$25bn of bonds and attracting US$81bn and US$72bn of demand respectively. While both deals initially rallied on the break, they subsequently underperformed broader credit to end the month between 5bps and 30bps wider.

In Europe, investment-grade supply totalled €90bn, of which €42bn came from Financials. Despite supply exceeding expectations, Financial deals averaged a healthy 2.6 times subscription rate, suggesting investor demand remained robust. Issuers appeared mindful of investor reluctance to add significant duration, with issuance more concentrated in the front end of the curve than usual.

Morgan Stanley was one of the first issuers to tap the market, printing €1.5bn of 4NC3 HoldCo bonds with an estimated 6bps of new-issue concession, and the bonds rallied 2bps on the break. DNB, RBC, BMO and NatWest also issued front-end paper during the month. As in the US, the standout transaction came from the corporate space, with Nippon Telegraph and Telephone (NTT) bringing a multi-tranche offering across both euros and sterling, having accessed both markets earlier in the year. The deal priced with a sizeable new-issue concession and all tranches rallied between 1bp and 6bps on the break, with stronger performance in the longer-dated tranches.

Australian credit markets also saw healthy primary activity. Australia recorded approximately A$15.9bn of investment-grade primary supply in June, around 15% above the circa A$13.8bn issued in June 2025, taking year-to-date supply to A$104.4bn compared with roughly A$82.7bn at the same point last year.

Strategy commentary cont'd: There was no major bank issuance, with June a seasonally quieter month heading into the end of the financial year. Macquarie nonetheless issued out of its PUMA RMBS programme, attracting A$2.3bn, while Suncorp Bank issued A$1.75bn of 5-year fixed and floating-rate bonds.

The month was otherwise characterised by Kangaroo bond supply from a range of offshore issuers. Singaporean banks DBS and OCBC issued a combined A$2.0bn of senior unsecured bonds, both via 3-year FRN deals. Among European financials, Societe Generale issued A$650m of 6NC5 fixed and floating senior non-preferred bonds, Lloyds Bank issued A$800m of 6NC5 senior notes, and Commerzbank issued A$1.0bn of 5-year FRN and fixed bonds through an inaugural transaction. RBC issued A$750m of 6NC5 fixed and floating senior paper, and Rabobank issued A$750m of 5-year senior FRNs. In semi-governments, NSWTC, TCV and NTC issued A$4.85bn of syndicated supply.

Australian fixed income and credit markets rallied over the month. The 10-year Australian government bond yield fell 11bps to 4.72%, while 5-year major bank senior spreads tightened 3.5bps to 63.1bps and major bank subordinated spreads tightened 4.4bps to 119.8bps. The AUD iTraxx Index tightened 4.1bps to 68.1bps.

Domestic bond benchmarks were positive, with the AusBond Credit FRN Index rising 0.46% and the AusBond Composite Index climbing 0.96%. The Australian dollar declined 3.7% against the US dollar to 0.6919, while the ASX200 price index rose 0.5% and the ASX200 total return index gained 0.7%.

New Zealand markets also firmed. The 10-year New Zealand government bond yield fell 17bps to 4.18%, while the New Zealand dollar declined 5.2% against the US dollar to 0.5678. Equities performed well, with the NZX50 price index gaining 2.5% and the NZX50 total return index rising 2.9% over the month.

Please note that past performance is not a reliable indicator of future performance. Investors should read the relevant Product Disclosure Statement and Target Market Determination before making any investment decision and consider obtaining advice from an independent financial adviser to determine whether an investment is appropriate for their objectives, financial situation and needs.

AI is driving the biggest capex boom in US history

The scale of the unfolding AI investment boom is unprecedented, eclipsing past US investment booms by a large margin.

AI investment is currently about 4.5 times higher than its recent trough, based on BIS estimates.

The only other boom that came close was the creation of the canals in the early 1800s, when investment peaked at about 4 times its corresponding trough.

The railway boom of the late 1800s saw investment peak at about 2.5 times its trough, while the booms of the roaring 1920s and dotcom bubble of the 1990s saw capex peak at about double their respective troughs.

History does not necessarily repeat, but past booms suggest that the expansion of AI probably has another 2-3 years before it tops out.

Moreover, the peak in AI investment will mechanically be higher because investment in tech is exaggerated by the rapid depreciation of software and computers.

Software – which is driving the current expansion – depreciates in value by almost half after a year, with depreciation of computers running at about one-third, and research and development depreciating at an annual rate of 15%, all according to Fed calculations.

CCI's analysis has already shown that the tech sector is driving the US economy at the moment, accounting for about three-quarters of recent economic growth.

Strategy commentary cont'd: The tech boom will boost productivity for some time, but the question for markets is whether companies can recoup an adequate return on this massive investment and if equity investors have paid too much for shares.

On the latter, long-term valuation measures – such as the cyclically-adjusted PE ratio – have shown US equities as extremely expensive over recent years, only beaten by the dotcom peak of 1999-2000.

Equities are also offering the smallest premium over bonds in recent history, except, again, for the dotcom episode.

While these valuation measures are not designed to provide a signal on timing the market, they underscore the ongoing risk of a correction in stocks, particularly now that Fed seems likely to raise interest rates and with real bond yields at their highest point since before the global financial crisis.

Strategy commentary cont'd:

US inflation is still too high even as upside risks abate

US core inflation was in line with market forecasts in May, tracking a fraction above the Fed's forecast profile, with annualised core PCE inflation of 3.75% in trend terms still well above the bank's 2% target.

The best news for inflation globally is that the upside risks from the energy shock have abated, while in the US, the boost to goods prices from tariffs should fade from around the middle of this year.

However, the goods news on the energy shock also means that the downside risks to growth have dissipated, which means that most central banks are left with core inflation that is still too high relative to targets.

Some central banks have started raising rates, including the ECB, BoJ, and RBA, and the Fed seems likely to follow over the coming months.

The current average policy rate in the US is 3.6% and policy rules now suggest that it should be about 4.5% (previously 4%). This underscores the deep division among policymakers on the FOMC, where about half the committee is happy to keep rates on hold and the other half would prefer to raise rates at least once this year.

Fed Chair Warsh has not shown his hand, abruptly refraining from offering forward guidance to the market and repeating that the Fed will achieve price stability without spelling out how that will happen.

Warsh is speaking at the ECB's annual policy conference next week and there will be little point to his participation if he repeats his press conference strategy of not answering questions.

That said, his confirmation hearing and recent actions suggest that he probably wants the Fed to keep rates steady heading into the mid-term elections by persuading his colleagues to focus on lower trimmed mean inflation rather than the better-known core PCE measure and taking up their time with a number of new taskforces.

While CCI's analysis has shown that trimmed mean inflation does a good job at approximating the trend in inflation, even there the story is not as favourable as Warsh might hope.

That is, annualised trimmed mean inflation briefly touched the 2% target late last year, but picked up again and is currently running at about 2.75%, which is still unacceptably high, reinforcing the risk of higher interest rates.

Strategy commentary cont'd:

Strategy commentary cont'd: The tech sector's role in driving US growth

The tech sector is driving the US economy, accounting for the bulk of recent growth. Outside of tech, the economy is not as good, while the strength in tech investment is exaggerated by the rapid depreciation of software and computers. Demand for tech is very narrowly based, and the productivity payoff is highly uneven. With the tech sector driving growth, the speed limit for the US economy is about 2.5%, higher than the FOMC estimate of potential growth of about 2%. Potential growth is a key influence on the neutral policy rate, so this supports the view that the neutral rate is likely higher than the FOMC estimate of about 3%.

1. The US is the best performing economy in the post-COVID period.

The pandemic has had lasting effects on several advanced economies, with GDP well below pre-COVID levels in the EA, JPN, the UK, and NZ, something you might ordinarily expect in the wake of a financial crisis. In contrast, the US, CAN and AUS have tracked closely to their simple pre-COVID trends, with the US narrowly outperforming.

2. Three-quarters of recent US economic growth has been driven by the tech sector.

The role of tech in driving the US economy can be approximated by estimating the contribution to growth from the tech capital stock and technological progress, where tech covers IT equipment, software and research and development and technological progress is proxied by multifactor productivity.

On this basis, the tech sector has accounted for 75% of recent US economic growth. This is basically the same aggregate contribution that was made during the dot-com bubble of the late 1990s. Overall economic growth was much stronger during the 1990s because the labour market was also booming.

Strategy commentary cont'd:

3. Outside the tech sector, the news is not as good.

Outside the tech sector, the economy is not doing that well. Annual growth in the non-tech capital stock is weak at 1% and has recently slowed. Growth at this rate is close to the low points reached after the recessions of the 1980s, 1990s and 2000s.

4. The tech sector's strength is exaggerated.

While the tech sector is driving US economic growth, the strength of investment is flattered by high rates of depreciation, where rising investment has to account for the rapid wear and tear of existing tech assets. As a result, annual growth in the tech capital stock is also historically weak, although, unlike the non-tech sector, growth has accelerated recently from 6 to 8%.

Strategy commentary cont'd:

5. The demand for tech in the US is narrowly based and is dominated by the IT and finance sectors.

The demand for software, which drives tech investment, is narrowly based on the US. Half of all spending on software is accounted for by the IT and finance sectors, with professional services bringing the total share to almost two-thirds.

6. Tech investment has a very uneven payoff on productivity.

Among the three sectors accounting for the bulk of tech investment, the positive payoff in terms of productivity is very clear for the IT and professional services sectors. However, the finance sector – which is dominated by banks and insurers – has seen productivity stagnate since the global financial crisis.

Strategy commentary cont'd:

7. The speed limit for US growth looks higher than thought by the Fed.

With the tech sector driving growth, we estimated that the speed limit for the US economy is 2.5%, based on a production function approach. This growth rate for potential output is higher than the roughly 2% median of FOMC estimates and has been higher than the Fed's calculation for about the past ten years.

8. Higher potential growth supports the view that the neutral funds rate is higher than the FOMC estimate of about 3%.

Potential growth is an important influence on the neutral policy rate. With the estimated 2.5% speed limit for US growth exceeding the FOMC's 2% calculation, this points to the neutral funds rate being above the median FOMC estimate of about 3%, where Fed staff models put it at around 3.5% and average market pricing is about 4%.

Strategy commentary cont'd:

Negative gearing and house prices, plus a case study of the 1980s

The government is limiting negative gearing for investment properties to new homes from July next year, returning to cost-base indexation for the taxation of capital gains, and imposing a minimum tax rate of 30% for inflation-indexed gains.

The government does not think these measures will raise much revenue or have much of an economic effect. However, negatively-geared investor loans for existing properties account for an estimated 20% of all new home loans, such that house prices – which had broadly stalled prior to the announced changes – could fall by 4-9% and credit growth should slow from 7.5% to about 4.5%. Policy rules point to tighter monetary policy, but tax changes could result in a lower peak in the cash rate.

The only comparable tax changes where in the mid 1980s, when negative gearing was all but scrapped and capital gains were taxed for the first time, with the changes to negative gearing reversed under political pressure after only two years. House prices fell in real terms, while real rents rose, real credit growth slowed sharply and new home loans slumped. However, many of these trends commenced prior to the tax changes when the RBA was aggressively raising interest rates.

Major changes to the taxation of the housing market.

The government has legislated two major tax changes that will significantly affect the housing market, which were announced in the May Budget and take effect in July 2027.

1. Significant limits to negative gearing for investment properties.

   Negative gearing – which allows interest payments to be deducted from total taxable income – will be limited to investments in new homes. Existing investments will be exempt from these changes, while interest payments on existing homes purchased between last month and July 2027 can only be deducted from income from residential properties (albeit where there is a carry-forward of losses).

2. A return to inflation indexation of the capital gains tax base and the imposition of a 30% minimum tax rate on capital gains.

   Capital gains – net of any current losses or past capital losses that have been carried forward – are currently halved under a concessional 50% discount and added to total taxable income. The 50% discount will be replaced by the cost-based indexation of capital gains, which is a return to pre-1999 arrangements. A minimum tax rate of 30% will be imposed on inflation-adjusted capital gains.

Strategy commentary cont'd: The government thinks the changes won't raise much revenue and won't have much impact on the economy.

The government does not expect that the two changes will raise much revenue, with the main budget saving coming from planned reforms to the National Disability Insurance, which are recycled into more spending on hospitals, defence, an income tax cut, health and other expenditure.

Instead, the treasurer hopes that the budget will "encourag[e] investment in new housing supply" and "[provide] a fair go for first home buyers". History suggests that these worthy aims will probably not be realised. No government has succeeded with either goal because they are unwilling to pay off the states to reform the supply of housing to allow it to catch up with demand, while repeatedly turning to subsidies to first home-buyers that end up capitalised in higher house prices.

In terms of the economic impact, Treasury's modelling claims the tax changes will:

• Reverse the decline in the home ownership rate for owner-occupiers over the past 10 years by boosting the number of owner-occupiers by 75,000 over the next ten years;
• Temporarily reduce house price growth by about 2pp over a couple of years, which will have a "small [negative] impact on housing supply"; and
• Raise rents by about $2 per week, which is a forecast rounding error compared with the median weekly rental rate of about $670 per week.

The tax changes affect a material share of the housing market.

While not expected to raise much revenue, the new policies will affect a material share of the housing market given private rentals account for about one-quarter of the stock of homes.

• Private rental properties – which encompass the homes affected by the tax changes – account for 24% of the total number of homes in Australia, including vacant properties.
• Home ownership has declined, but still dominates the housing market, accounting for 66% of homes (this includes the 9% of homes that are empty).
• Public rental properties account for 3% of homes, although this greatly understates the role of government in the housing market because about half of all renting households receive Commonwealth rent assistance.
• The remainder of 7% of homes includes hotels, aged care, hospitals, etc., as well as homes where details are not available.

Strategy commentary cont'd: In the private rental market, more than half of investment properties are negatively geared. The split between taxpayers with a rental property who are negatively geared and those who are either neutrally or positively geared was 54%/46% in 2023-24, based on Tax Office data. More recently, the split has probably shifted to about 60%/40% given the share of negatively-geared properties is largely driven by the level of mortgage rates.

Investors do buy new homes, which can still be negatively geared, but have favoured existing homes over recent years. The split of investor loans in dollar terms between new and existing homes is currently 18% new / 82% existing. This means that new investor loans for existing properties account for one-third of all new owner-occupier and investor loans.

On the strong assumption that 60% of investor loans for existing homes are negatively geared, this would amount to about 50% of new investor loans, or around 20% of all new home loans.

The tax changes could reduce house prices by 4-9% over the next couple of years.

Home prices – which have been at an extreme for some time as a multiple of household income – have recently stalled at the national level as the RBA has raised rates, taking back all of last year's rate cuts.

The government's new policies are likely to weigh on prices, although estimating the economic impact of the tax changes is very difficult. This is because macro models do not adequately incorporate taxes and the only comparable changes to the tax treatment of the housing occurred over forty years ago.

Strategy commentary cont'd: One approach is to broadly equate the tax change to an increase in mortgage rates, taking a leaf from research by Trent Saunders at CBA. The estimated increase in the mortgage rate can then be combined with results from the RBA's Saunders-Tulip model of the housing market to back out a potential impact on house prices.

On this basis, the tax changes equate to an increase in investor mortgage rates for negatively-geared properties of about 2-2.75%, depending on the tax bracket of the investor. This is calculated by backing out the increase in the mortgage rate required to raise loan repayments by the same amount as the lost tax concession on Cotality-derived estimates of an average national property worth $1 million with a 3.6% rental rate and relying on the APRA estimate of an investor mortgage rate of 6.4% (the latter assumes that the latest rate rise was fully passed on to new borrowers).

Given that negatively-geared investors account for about 20% of new home loans, this broadly equates to a rounded increase in the mortgage rate on all new home loans of about 50bps.

The Saunders-Tulip model suggests a rate rise of this magnitude would lower real house prices by about 3-5% over one year and by a cumulative 4-9% over two years, with the range reflecting uncertainty over whether the changes are truly permanent.

As a crosscheck, a second much simpler and mechanical approach relies on balance sheet data. The stock of all residential mortgages is $2.6 trillion and estimated negatively-geared new investor loans for existing homes was an annualised $80 billion in Q1. Mortgages are currently growing at an annual rate of 7.5%, so the tax changes could take just over 3pp off growth by ending the demand for these loans, which could be interpreted as an upper bound for the impact on growth in house prices.

Strategy commentary cont'd: House price declines could see a lower peak in the cash rate.

Policy rules currently point to a peak in the cash rate of about 4.75-5% given high inflation and unemployment that is still below the estimated NAIRU. A material decline in house prices could see a lower peak in the cash rate, although it should be stressed that the circa 50bp estimate of the increase in the overall mortgage rate does not equate to a cash rate increase of 50bps. This is because other channels of the transmission mechanism of monetary policy are not affected by the tax changes.

In addition, there is the possibility that banks partly counter the tax changes by lowering mortgage rates for owner-occupier loans and investor loans for new homes to make up for lost revenue. APRA publishes the best data on mortgage rates actually paid by borrowers with a lag, although there are unconfirmed reports that banks have recently trimmed owner-occupier rates.

Haven't we seen this movie before? What happened in the 1980s?

The only other time that the taxation of housing and capital gains was changed in such a significant way was in 1985. That was when recommendations made in the Hawke-Keating government's White Paper on tax reform were adopted, with the notable exception of a GST, which was not introduced until 2000.

1. Negative gearing on investment properties was all but scrapped.

   Negative gearing was heavily curtailed for both new and existing homes by not allowing interest payments to be deducted from total taxable income and restricting deductions to income from residential properties (albeit where there was a carry-forward of losses).

2. Capital gains on assets bought after 1985 were taxed for the first time.

   Assets bought before 1985 remained exempt from tax, but inflation-adjusted capital gains on assets bought after 1985 were taxed when the assets were sold. Capital losses could not be deducted from taxable income. Instead, they could be deducted from any other current or future capital gains. This arrangement prevailed until it was replaced in late 1999 by the above-mentioned 50% discount on capital gains tax liabilities.

The economic impact of the tax changes is still debated today, with most focus on higher surveyed rents, particularly in Sydney and Perth. These rent rises, or more perhaps effective political campaigning by the housing industry, caused the government to completely the reverse changes to negative gearing in 1987, although the new capital gains tax was retained.

While the rental market share of the housing stock in the 1980s is similar in size to today, the impact of the tax changes is hard to disentangle from the impact of higher interest rates, which look to have been the driving force behind the weaker housing market on many metrics.

• The RBA raised the cash rate from 10% to 18% over 1983-85 (or 2% to 10% in real terms). However, the mortgage rate – where a ceiling was in place until 1986 – rose by much less, up from 12% to 16%, and the increase in the real mortgage rate was only 1pp given high inflation at the time.

Strategy commentary cont'd:

With this key caveat in mind:

• House prices fell in real terms across the country, although they were weakening prior to the tax changes.

  National house prices declined by about 4% in real terms over the period, with similar declines across the country. Prices were weaker prior to the tax changes, likely reflecting the impact of higher interest rates.


• Rents rose in real terms, although it is not clear whether tax changes were responsible.

  Nationally, rents rose by about 7% in real terms, with some lag, given that the CPI measures the stock of rents and not every lease is regularly updated. The increase was not uniform, rather driven by Sydney and Perth, which raises questions about whether the tax changes were behind the increases. Vacancy rates were extremely low in both cities, and were extremely low in Sydney well before the tax changes.

Strategy commentary cont'd:

• Credit growth slowed sharply in real terms, cooling before the tax changes, while much the same fall in new owner-occupier and investor home loans suggests interest rates were more the driver.

  Annualised growth in stock of residential mortgages halved from 18% to 9%, with the slowdown in real terms from 12% to -1% commencing well before the announcement of the tax changes. The value of new investor and owner-occupier loans both fell by about 30%. The fact that owner-occupier loans – which were not affected by the tax changes – fell at the same rate suggests that higher interest rates had a greater role than tax policy.

Strategy commentary cont'd: