Ratio that shows share prices at level well above dot-com boom

Shares are trading in US stocks at a level that is higher than it would cost to rebuild the companies, prompting a fresh warning about how high prices are going.

The Tobin Q ratio compares the total market value of US corporate equities to the replacement cost of the underlying companies. A ratio above one indicates the market is paying more than it would cost to rebuild the companies.

The long-run average is 0.85 - it reached 1.47 in 2000 at the dot-com boom. It is now more than 2.1.

Pie Funds chief investment officer Mike Taylor said he would take the signal seriously, but would be cautious about drawing "too mechanical" a conclusion.

"It is a useful long-term valuation gauge, because it asks a sensible question - how much are investors paying for corporate assets relative to what it would cost to replace them?

"On that basis, the US equity market is clearly not cheap, but I am also a skeptic of using Tobin's Q in isolation. The US equity market today is very different from the market of 50 years ago.

"A much larger share of market capitalisation now sits in companies whose real assets are software, intellectual property, data, networks, brands, distribution and scale advantages. Those are genuine economic assets, but they are not captured cleanly in a traditional replacement-cost framework.

"That does not mean the market is immune from valuation risk. A very high Q still tells you that investors are paying a lot for future profitability and that leaves less margin for error.

"If margins normalise, rates stay higher, AI returns disappoint, or competitive and regulatory pressures rise, forward returns could be lower than investors have become used to."

He said it was a yellow light rather than a red one.

"It is not a reason to sell equities wholesale, but it is a reason to be disciplined - manage concentration risk, diversify, be selective and avoid assuming that the returns of the last decade can simply be extrapolated into the next one."

University of Auckland senior finance lecturer Gertjan Verdickt said the concern was warranted and all the standard metrics were stretched. The Shiller PE ratio, which represents price-to-earnings, is at a level only exceeded at the dot-com peak.

"The S&P 500 earnings yield is 3.2 percent, which is now below the 10-year Treasury yield, although it's a real v nominal measure, so on the simplest measure, the ex-ante equity risk premium is actually negative.

"Investors are accepting a lower earnings yield than the risk-free rate, with the whole equity case resting on future earnings growth, but my biggest concern is duration rather than level.

"The market has shifted heavily toward long-duration assets - think AI, semiconductors, technology broadly - where the bulk of expected cashflows sits far out in the term structure.

"Mechanically, low yields mean high duration - an earnings yield of about 3 percent implies an equity duration on the order of 30 years. That makes valuations extremely sensitive to small changes in discount rates or growth expectations.

"A modest move in long rates or a repricing of the AI growth narrative translates into outsized price swings, so the practical implication isn't necessarily an imminent correction. It's that we should expect a more volatile market, with that volatility concentrated in the handful of mega-cap names that dominate the index."

Finsol financial adviser Gareth Dobson said it was something for KiwiSaver members to be aware of. Some active managers had held back on their exposure to those markets, but passive funds that tracked indexes had much higher exposure in most cases.

The strong growth in prices had driven some passive funds to outsize returns, compared to the less-exposed active managers.

"[Some active managers] are basically going, this overexposure to these large US tech companies... we're going to play this cautious, whereas an index is just an index, you have to go where the index goes.

"There's very little human or there's no human element with mitigating overexposure. You can see a lot of these over the last 1-2 years, the growth funds that are indexed and not managed are going really, really well... six months down the track, things could have flipped completely 180."

He said the Q ratio was "a little bit terrifying".

SBS Wealth chief executive Morne Redgard said valuations had been stretched for a while ago.

"I think probably two years ago... we had higher exposures to the Magnificent Seven or the top 10 shares in the US, and slowly, over the last two years, we've kind of neutralised those exposures, so are no longer overexposed in some of those biggest shares.

"It does feel a bit crazy, because it almost feels like 189 months ago, 24 months ago, some of those big shares were only hitting a trillion market cap and some of them have just popped over 5 trillion."

Dean Anderson, founder of Kernel, which offers several passive KiwiSaver funds, said the Q ratio was of interest, but may no longer be appropriate.

"The Q Ratio measures market value against the replacement cost of physical assets - think machinery, inventory, property," he said. "That worked pretty well when most corporate value actually was physical, but look at the most valuable companies today.

"A huge chunk of what makes them worth what they're worth is stuff you can't put a replacement cost on - customer loyalty, network effects, brand, proprietary data, the fact that millions of people open their app every morning out of habit. None of that shows up on a balance sheet.

"These assets aren't free either. Companies spend years and billions building them, but a replacement cost calculation just can't see them.

"This highlights something a bit broader. As services and digital goods make up more and more of the economy, a lot of the traditional metrics we use - Q ratio, price-to-book, the whole growth versus value framing - are showing their age a little. They were built for a different kind of economy.

"To be clear, I'm not making a call on whether markets are cheap or expensive right now. My point is simply that one metric, especially one with a known structural limitation, shouldn't be doing that much work in forming a view.

"The Q ratio hitting a historic high is worth noting, but so is the fact that the economy it's measuring has changed pretty dramatically from the one it was designed for.

"In short, a single data point or a single ratio shouldn't be used to form a whole view of the world."

Sign up for Money with Susan Edmunds, a weekly newsletter covering all the things that affect how we make and spend money