Debt-to-GDP tells you a country has a problem. It never tells you when. Andrew Hunt's framing in his recent research note on fiscal sustainability is more useful because it is a cash-flow statement rather than a balance sheet: government borrowing less household savings must be met either by monetisation or by foreign inflows. Put that way, the question stops being how much a government owes and becomes who is writing the cheque this quarter.

On his numbers the answer is uncomfortable. Foreigners fund roughly half the US deficit, perhaps half to three-quarters of the UK's, and the overwhelming majority of France's. That is why France broke first. A rumoured Japanese life office stepping back, ECB quantitative tightening and a falling domestic savings rate were enough to tip arithmetic that had been fragile for years. France is not a special case; it is the most leveraged example of a condition shared across the developed world.

Hunt's simulation of a complete halt to foreign flows into the US produces a ten-year Treasury yield near 7.5%, unemployment at 9% and GDP down three to four per cent. He does not expect it, and the useful part is not the extreme but his observation that the path between here and there is roughly linear. A partial withdrawal produces a proportional repricing, not a partial scare. The tail is less a cliff than a slope, which means monitoring it continuously rather than waiting for it.

From here the indicator he is monitoring is refreshingly concrete: the renminbi and the broader Asian currency complex. Appreciation there means surplus recycling is stalling, because those surpluses are no longer being pushed back into Western bonds. It is a daily observable, which is more than most tail risks offer.

None of this moves our asset allocation, for now. Hunt's own position — diversified, broadly  fully invested, with the running yield on bonds acceptable compensation — is where we have landed. US yields have already risen some seventy-five basis points on bank selling, weaker household saving and seasonal issuance, and now sit near equilibrium while the foreign money still arrives. That supports a modest duration addition, and our preference remains Australian over US duration: the local arithmetic depends less on strangers, and the cheapest implementation is usually an index vehicle.

Where his work does bear on underlying manager exposures is the regime argument. Asian export prices are rising at their fastest since 1973, traded-goods prices are climbing against services and housing, and that favours cash-generative industrials over non-traded services. It is the same conclusion we reached from valuations: US earnings excluding AI have grown strongly this year on a far lower multiple than the index, and the broadening is visible in Australian mid-caps, global small caps and emerging markets outside Korea.

That last distinction matters. Much of the recent emerging-market return has come from Korean semiconductors, which is precisely the concentrated trade the diversification is meant to avoid. Managers whose process excludes it, Aikya and Pzena are the clearest example in our models of the diversification/rotation trade, and they both behaved  very differently risk-off move we are witnessing recently. Advisers may wish to look through to whether an emerging-market allocation is a genuine diversifier or the same AI exposure in another wrapper.