Investing in 2025: beyond the laws of reality

In Tom Stoppard's play Rosencrantz and Guildenstern are Dead, the titular characters become disquieted when a coin they are tossing between them lands on 'heads' seventy-six times in a row. Since the law of probability is clearly not working, they conclude they must be operating beyond the laws of reality.

As an active fund manager, I have felt a similar sense of foreboding this year. Unprofitable companies are flying (up 40 per cent in the last six months), low returning businesses are dramatically outperforming high returning businesses, and companies with the highest leverage are outperforming companies with the lowest leverage. 

For evidence that we're in broken coin territory, we need look no further than the rise and rise of Australian banks. If you added together the underlying cash earnings per share produced by the Big Four in financial 2023, it amounted to $13. In FY25, that figure is $12.43, a four per cent decline.

Yet if you add together the share prices of the Big Four, you'd have paid $178 for those earnings two years ago versus $264 today, a whopping 48 per cent increase. Rising interest rates? Heads, good for banks. Falling interest rates? Heads, good for banks. Declining earnings? Heads, good for banks. Matt Comyn himself said it best at a recent parliamentary hearing: "When you have, as we do today, many banks earning below the cost of capital in the most benign credit cycle any of us have seen in our lifetimes, I'm not sure that augurs well".

Even things that have historically been indisputably not good for banks are somehow good for banks in 2025. Death, taxes and a major bank failing to deliver on a stated cost-out target are the certainties of life. Yet ANZ has rallied 30 per cent this year, as new CEO Nuno Matos aims to aggressively cut costs (while somehow not leaking revenue). Merely announcing a cost-out? Heads, good for banks. 

Meanwhile, Westpac's CEO Anthony Miller is finally, laudably, tackling the herculean challenge of uniting the spaghetti systems holding Westpac's various brands together, through Project Unite. Any shareholder who has read Bent Flyvbjerg's excellent book How Big Things Get Done should be trembling.

Flyvbjerg studied 1,600 large projects across 25 different categories (from roads, bridges and railways to Olympic Games) and identified what he deemed the "Iron Law of Megaprojects": they pretty much all come in massively over-budget, late, and under-deliver on benefits. Large IT projects are amongst the riskiest, ranking right below hydroelectric dams as the most likely to encounter massive cost blowouts (someone should've sent a copy to Malcolm Turnbull).

We saw this at Metcash as management attempted to replace nine IT systems with one new Microsoft system, and costs blew out from $80 million to $300 million and counting. Westpac's Project Unite is an order of magnitude more complex. If you think they can painlessly deliver a circa $3.5 billion upgrade to consolidate 180 systems into 60, all while maintaining a seamless customer experience and adequate cybersecurity, I've got a bridge to sell you (delivered on time and on budget). Despite this burgeoning risk, Westpac shares have risen 20 per cent this year. Multi-billion dollar IT project? Heads, good for banks.

First, Do No Harm

Adding to the sense of unreality were the baffling capital allocation decisions of some management teams this year. We investors are now so riddled with the scars of large-scale M&A that we need them to swear a version of the Hippocratic oath: First, Do No Harm. Forget upgrades and earnings 'beats', just don't do anything dumb. Investors understand that the world is complex, that luck plays a role in the business cycle. Frequently, things go wrong for companies that are outside of their control. 

By way of example, we would not have excoriated James Hardie's management team for delivering an earnings downgrade in the face of a rapidly deteriorating housing market in the US. In fact, in an alternate universe, we would've welcomed the opportunity to buy shares in an excellent business with a pristine balance sheet at the bottom of the cycle. Unfortunately, that universe must remain alternate, as management instead opted to become the face of inexplicable value destruction with the US$8.75 billion ($13 billion) acquisition of composite decking company Azek, thereby gearing up the balance sheet right into a macro storm. 

James Hardie shares have halved since the announcement of the deal in March. Having learnt their lesson watching Boral unpick Headwaters, CSL overpay for Vifor, Orora downgrade Saverglass, Viva struggle with On The Run and Ramsay languish under the weight of everything it ever bought in Europe, shareholders have simply voted with their feet on day one rather than sticking around to see whether these deals create their purported value. 

From Hardie's self-immolation, to Xero's $4 billion bet on loss-making US payments company Melio, which has seen its share price fall nearly 40 per cent, 2025 has hopefully taught corporate Australia that investors are simply jack of it: jack of empire building, jack of EPS accretion stories, jack of eye-watering fees to investment bankers and jack of valuations propped up by rubbery revenue synergies. 

While the mood of the market is that acquirers are guilty until proven innocent, some companies seem to have earned the benefit of the doubt. Ampol's proposed acquisition of EG Group is cheap and underpinned by cost synergies, Carsales has made a rare go of it overseas, Eagers Automotive appears to have bought the Canadian tuxedo version of itself. It wasn't E2open that undid WiseTech this year, but a case of the founder being too open with his… share sales. I suspect the market will buy whatever Ryan Stokes does next at SGH. The through-line is a track record of execution, and acquiring in your own lane.

Role Of Luck

Further, some companies have seemingly gone from guilty to innocent in 2025. Few would've picked Mineral Resources to rebound 250 per cent off its April lows within six months. From a haul road that was literally washing away in the rain at the start of the year to receiving its bonus payment from Morgan Stanley Infrastructure Partners for hitting a 35 million tonne run-rate for iron ore delivery, MinRes deserves credit for its successful ramp of the Onslow iron ore project. 

However, the 2025 MinRes story highlights not only the role of luck in investing, but clearly the short memories of market participants. Had the iron ore price dropped $10 instead of rising $10, had we bounced along the bottom of the lithium cycle for another year, there would likely have been a nasty recapitalisation of the company as its mountain of debt came due. Forget the Australian Taxation Office investigation for tax evasion. Forget the Australian Securities and Investments Commission looking into whether the CEO's mother-in-law front-ran the company's purchase of Kali Metals stock. Forget the travertine tiles flown in from Italy. Investors flocking back to the stock have shown that if profits are up, you can do what you want. 

This year also provided another great example of my favourite kind of event: a good old-fashioned market meltdown. Leo Tolstoy opened Anna Karenina with the immortal line that "Happy families are all alike; every unhappy family is unhappy in its own way". Similarly, unhappy markets always have their own idiosyncratic causes: a pandemic, a war, subprime mortgages. That's a feature not a bug, and the unique nature of the cause enhances the sense of uncertainty, the lack of precedent, the sense the coin is broken. 

In April this year, it was Trump's wild tariffs that sent shockwaves through global markets. The S&P 500 fell 19 per cent over six weeks, while the ASX 200 fell 14 per cent over the same time frame. As they always do, this event turned out to be a great buying opportunity, with the ASX 200 rallying 23 per cent over the next six months. As Buffett says, if you're a net purchaser of equities you should want to see share prices lower. A good litmus test is: if your mum's whinging that her super has gone down, and your broker is emailing you a list of defensive stocks to buy, the bottom is probably in.

In fact, a critical part of investing is having a vaguely accurate idea of where you are in the cycle. Not even the broader economic cycle, because just like five o'clock, it's always the top of the market somewhere. 

From tech stocks to lithium, from uranium to rare earths, and even the utterly bizarre cycle that played out in the Commonwealth Bank's share price, the bottom and top of a cycle are clear only in hindsight, yet there are common markers. 

At the bottom of a cycle there are many reasons why prices will never rise again. Similarly, at the top, what usually emerges is a very specific reason why this time it's different; that this time, the laws of reality will remain suspended. The last commodities super cycle that peaked in 2011 was driven by a belief that the rise of China meant this time it was different. Spodumene at over US$8000 per tonne in 2022 was driven by the belief that the energy transition meant this time it was different. As the CBA share price blew through $170, then $180, then $190, many in the market posited that passive flows and a fixed retail register meant this time it was different. Write it on a Post-it note and stick it to your computer: it's never different. 

As my daily commute takes me past the queue of people lining up at Martin Place to buy physical gold from ABC Bullion, I can't help but feel the "central banks are forced buyers/government deficits are out of control" thesis that underwrites any gold price feels a little like This Time It's Different thinking. Then again, maybe ABC Bullion is just too close to the former Peloton store and I'm seeing bubbles everywhere.

Speaking of bubbles, a review of the year would be incomplete without touching on the debate that has dominated global discourse as investors scramble to figure out where generative artificial intelligence will land on the scale of technological transformations. Is it akin to the internet? The steam engine? Or is it this generation's 3D printer? I've read so many compelling arguments for and against the idea that we're in an AI bubble that at this point my whole personality is the last think piece I read on Substack. 

Whatever the outcome, the share of mind that AI has already garnered has certainly added to the sense of disquiet this year. Anyone who has dabbled in the dark arts of DCF would know that so much of a company's value per share lies in its terminal value – a nebulous bucket of cash flows generated far out in the future, discounted back to today. A potentially life-as-we-know-it altering technological shift has large implications for the terminal value of pretty much every company, and markets will be quick and savage in marking down perceived losers. We began to see this play out in 2025, as the ASX tech sector darlings have derated under a "SaaS is dead" narrative. This year's losers may well prove a fertile hunting ground for next year's winners, particularly if the AI revolution shows any signs of stalling.

Looking to the future: there is no such thing as a normal year in investing, but there are some soothing signposts I look forward to in 2026. Comforting and expected, almost quaint, like an impairment from Lendlease, a production downgrade from a gold producer or REA writing off an attempt to establish itself in an Asian market (and the market shrugging it off as the company launches a new depth product). All signs that the laws of reality are operating once more, that a coin tossed in the air is as likely to land heads as tails, that order is restored to the universe, and that share prices will follow earnings, not narratives, macro fears, frenzies or vibes.

Emma Fisher is the deputy head of equities at Airlie Funds Management.