Adelaide Steamship: John Spalvins' house of mirrors

The Adelaide Steamship Company, or Adsteam, produced a group structure chart that simply doesn't fit on one page. Arrows loop back on themselves. Companies own slices of companies that own slices of the company that owns them. If you've ever tried to explain a cross-shareholding structure to a first-year finance student, you'll understand why such a chart becomes a cautionary artefact.

By 1990 Adelaide Steamship was one of the largest and most admired industrial groups in the country. Four years later it was effectively gone, and its principal architect, John Spalvins, had gone from being the subject of glowing profiles to a case study in what happens when leverage meets a recession the market didn't see coming. This is one of the essential entries in the Great Australian Collapses catalogue, not because the fraud was spectacular (there wasn't really any) but because the structure itself was the problem.

Adelaide Steamship: from shipping line to industrial sprawl

Adelaide Steamship Company started life, as the name suggests, as a shipping business, founded in South Australia in the nineteenth century. By the time Spalvins took the reins in the early 1980s it had long since diversified, but nothing prepared the market for what came next. Over the following decade Spalvins built, through a mix of takeovers and strategic stakes, a conglomerate that touched an enormous share of everyday Australian commerce: David Jones, Woolworths supermarkets (the Australian one, not the UK entity of similar name), Petersville, Tooth & Co brewing interests, National Consolidated, and a rotating cast of finance and industrial subsidiaries.

The genius, and later the poison, was in how it was financed. Rather than Adsteam simply owning these businesses outright, the group built a web of cross-shareholdings: Adsteam owned a stake in Company A, which owned a stake in Company B, which owned a stake back in Adsteam or in another group entity. Each entity could book the others' equity as an asset, gearing ratios looked more comfortable than they were, and a relatively modest amount of Spalvins-controlled capital could command effective control over billions of dollars of underlying industrial assets. It was leverage dressed as diversification, and for most of the 1980s it worked spectacularly. Spalvins was lionised in the business press, feted as a strategic genius, and Adsteam shares were a market darling.

The mechanics of a cross-shareholding empire

To understand why this structure was fragile rather than merely complicated, you need to think about what happens on the way down rather than the way up. In a rising market, cross-shareholdings are a wealth-multiplication machine: asset values rise, everyone's balance sheet looks stronger, and the debt used to acquire the stakes in the first place seems trivial against the paper gains. Banks lend against that paper. More stakes get bought. The circle widens.

But the same mechanism runs in reverse with brutal efficiency. If Company B's share price falls, Company A's balance sheet weakens because its major asset, the stake in B, is worth less. That weakens Adsteam's own accounts, because Adsteam's asset is a stake in A. There is no genuine diversification benefit, because everything is correlated to everything else through the same handful of hands. A conglomerate that looks like six different businesses in six different sectors can, structurally, behave like one single leveraged bet on the group's own share price.

This is the same underlying lesson that turns up, in different clothing, in the Bell Group collapse: complexity in corporate structure is not neutral. It tends to redistribute risk toward whoever is least able to see it coming, usually outside shareholders and, eventually, the banks holding the debt.

The recession finds the seams

Adsteam's structure was built for a world of rising asset prices and cheap, available credit. Australia in 1989 and 1990 supplied neither. The Reserve Bank had pushed interest rates to punishing levels to choke inflation, and what followed was, in Paul Keating's famous phrase, "the recession we had to have." Commercial property values fell, retail spending softened, and corporate Australia's most heavily geared conglomerates, Adsteam among them, alongside contemporaries such as Qintex and Bond Corporation, found that the debt taken on during the boom didn't shrink just because the assets underneath it had.

Reported profits at Adsteam group companies weakened through 1990 and 1991. Share prices across the group followed, and because of the cross-shareholding structure, a fall in one component company's price mechanically dragged down the reported net asset backing of the others. The Reserve Bank's own retrospective analysis of the period, published in its research on the early-1990s recession, points to exactly this pattern across the highly geared conglomerates of the era: asset values and debt levels that had seemed manageable at 1988 prices looked entirely different once credit tightened and the banks began reassessing security values line by line.

By 1991, Adsteam's total debt was widely reported to run into the billions, and the group's bankers, a syndicate that included the major Australian trading banks, were no longer prepared to simply roll it over on the old terms. What followed was less a single collapse than a prolonged, grinding disassembly.

The long unwind

Unlike some of the era's more theatrical failures, Adsteam didn't go into administration in a single dramatic weekend. Instead, from 1991 through the following several years, the group underwent what amounted to a negotiated deconstruction, overseen by its bankers rather than by an insolvency practitioner in the conventional sense. Non-core assets were sold. Stakes in listed subsidiaries were progressively unwound or sold down. David Jones was eventually separated out and refinanced as a standalone entity. The web of cross-shareholdings that had been the source of Adsteam's apparent strength was, piece by piece, dismantled under the direction of the banking syndicate, whose priority was recovering as much of their exposure as possible rather than preserving Spalvins' industrial vision.

Spalvins himself resigned from his executive roles in 1991, and the restructuring that followed occupied the group's bankers and advisers for years. It was, in the accounting and insolvency trade, remembered less as a bankruptcy than as a masterclass in how much value can leak out of a conglomerate when a workout drags on: legal fees, advisory fees, distressed asset sales at a discount to underlying value, and the slow erosion of businesses that, standing alone, had been perfectly viable.

I'll admit a mild bias here: I think the Adsteam unwind is underrated next to the more famous 1990s corporate failures, precisely because nobody went to prison and there was no single dramatic collapse morning. But in terms of value destroyed and lessons for how conglomerates get financed, it's arguably more instructive than flashier cases. The fraud cases teach you to watch for liars. Adsteam teaches you to read a balance sheet.

Why regulators and analysts still cite it

Adsteam's implosion fed directly into a broader Australian regulatory conversation about corporate governance and disclosure that ran through the 1990s. The Australian Securities Commission, the predecessor to today's ASIC, took a particular interest in cross-shareholding structures and related-party transactions in the years that followed, and accounting standard-setters tightened requirements around consolidated reporting and equity accounting for exactly the kind of circular holdings that had let Adsteam obscure its true gearing. The corporate law reforms and continuous disclosure tightening of the mid-to-late 1990s, while not attributable to any single collapse, were shaped by a string of conglomerate failures in which Adsteam sat alongside Bond Corporation and Qintex as prime exhibits.

For students of insolvency, the case remains a staple teaching example in corporate finance courses precisely because the mechanism is so clean: no fraud required, just leverage multiplied through a structure that made genuine risk concentration look like diversification. ASIC's own historical materials on corporate regulation reform reference the early-1990s conglomerate failures as a formative period for Australian disclosure law, and the pattern shows up again, in different form, whenever a highly geared group with circular ownership hits a downturn.

What Adsteam still has to teach

The comparison I keep coming back to is how differently value gets destroyed depending on the structure of the failure. One.Tel's collapse burned cash at a rate that was almost breathtaking to watch in real time. Adsteam's destruction was slower, quieter, and in some ways more insidious: value didn't vanish so much as leak away through years of forced sales and advisory fees while a structure too complicated for its own directors to fully model came apart in slow motion.

Spalvins went from Australian Financial Review Businessman of the Year to a cautionary footnote inside three years, without ever being found by a court to have done anything improper. That's worth sitting with. The Adsteam story isn't really about a villain. It's about what happens when a genuinely talented dealmaker builds a structure so intricate that even a mild recession, rather than a catastrophic one, is enough to bring the whole thing down. The tangled group chart isn't a curiosity. It's the whole lesson, drawn as a diagram.

Ray Petrakis, Corporate collapses & insolvency correspondent