Those who fail to learn from history are condemned to repeat it, goes the famous saying attributed to Winston Churchill and philosopher George Santayana.
Fletcher Building's near $1 billion in construction losses in New Zealand over 18 months are still fresh in investors' minds, but the company also has a long history of losing money in Australia since it was split out of Fletcher Challenge in 2001.
The $222 million of write-offs of the Tradelink and Iplex Australia businesses announced in the 2017 annual results, besides being overshadowed by the construction division's woes, were just the latest in a long string of write-offs.
By my count after trawling though the company's annual reports, accumulated write-offs and restructurings of its Australian businesses total well over $500m since 2001.
The long, slow burn started in its first year as a stand-alone company when it sold its Australian construction business for an $8m loss.
Losses on acquisitions ranging from the $800m Laminex purchase in 2002 and the $581m Amatek purchase in 2005 to the $1b Crane takeover in 2013 are sprinkled through the various annual reports.
Fletcher Building's earnings record in Australia has also been a dismal affair.
When Fletcher Building was formed in 2001, Australia accounted for 9 per cent of group sales, or $272m on a pro forma basis, and delivered an earnings before interest and tax (ebit) loss of $43m.
The following year, the first full-year in its current incarnation, Australian sales had advanced to $315m while ebit turned positive at $4m, a skinny 1.3 per cent margin.
The company's best ever year in terms of margin was the year ended June 2006 when ebit of $211m on sales of $1.86b resulted in an ebit margin of 11.35 per cent.
With all the acquisitions in Australia the company has made, the Australian operations accounted for 32 per cent of sales in the six months ended December 2018.
At its investor day last month, the company said Australian ebit for the year ended June this year would be just $55m, down from $114m the previous year and with a margin of just 2 per cent, half the margin it forecast at the 2018 investor day. The New Zealand operations achieved a 7.5 per cent margin in the six months ended December 2018.
In 2018, the company said it was aiming to double the 2017 Australian ebit of $117m by 2023, lifting the margin to 7 per cent, reflecting the lower-than-expected starting point.
This year's investor day kept the same margin aspiration but has pushed its achievement out a year to 2024.
At that level, chief financial officer Bevan McKenzie said the company would be achieving above 10 per cent return on funds employed – that's shy of the company's company-wide target of 15 per cent.
The market doesn't appear to believe Fletcher will be able to achieve that target – its share price has dropped another 10 per cent since the investor day while the benchmark S&P/NZX 50 Index has gained 3.5 per cent.
Chief executive Ross Taylor acknowledged his company's credibility problem, saying near the end of the investor day that "there's a lot of cynicism around our Australian turnaround."
In Craigs Investment Partners analyst Grant Swanepoel's telling, when Fletcher's balance sheet was under pressure last year and management decided to sell the Formica business, it also considered selling the Australian operations.
"The strategy update last year made it clear that the board thought that Australia was a turnaround story and hence sold Formica," Swanepoel says in his note after this year's investor day.
Swanepoel clearly still believes management should be thinking about ditching the Australian businesses.
"The way Fletcher has restructured their businesses into Australia and NZ separately makes it obvious that there are limited strategic synergies in having both countries."
His working assumption is that the Australian operations may achieve only about a 5.1 per cent ebit margin by 2024 and he's also assuming that will be the top point of the cycle – mid-cycle margin would be about 3.7, Swanepoel says.
Other analysts share his scepticism. Matt Henry at Forsyth Barr says the halving of Australian ebit in the year just gone "highlights our near-term concerns about these high operating leverage businesses in a challenging cyclical and competitive environment. We remain to be convinced."
Henry also reflects on the company's history, saying it has a track record of "cost savings targets that never obviously materialise into earnings."
Arie Dekker at Jarden says any improvement in Australia in the year ending June next year is likely to be minimal, with any signs of Fletcher's initiatives starting to pay off being offset by the still difficult market.
As for Fletcher's decision to edge back into the high-rise part of the construction market, the analysts are sceptical about that too.
Dekker thinks the Building + Interiors unit is unlikely to contribute to earnings until the 2021 financial year at the earliest.
Swanepoel says that "the pull-through revenue from NZ core businesses from having NZ construction is not entirely clear as to whether it has been a benefit or a burden."
He says the "NZ construction risk is tarnishing the value that would otherwise be applied to the strong core five NZ businesses."
The one thing saying – not necessarily doing – that B+I is back in the market for new projects might help Fletcher with is completing the remaining backlog of projects.
Taylor made a strategic error in early 2018 when he said B+I would cease bidding for new contracts because that meant retaining key staff required to complete the existing projects became exponentially more difficult.
He told the investor day that the value of outstanding work has dropped from $1.4b in February 2018 to $400m now and that only the Commercial Bay and Sky City convention centre projects will continue into 2020.
As Henry says, "Whilst the consensus view is that the company has likely now 'kitchen sinked' its loss provisions, the risk will not be extinguished until the projects are complete in the second half of calendar 2019. Anecdotes of staff losses and project issues persist."