Intriguingly, several news stories noted that the emergence of HP’s Autonomy debacle had been predicted a few weeks earlier at a famous New York investor conference, the Santangel’s Review.
And the analyst who had already detected Autonomy’s problems was not a software expert, had never met any HP or Autonomy executives, and wasn’t even a US resident. He was a former Australian Treasury official who now lived in the beachside Sydney suburb of Bronte.
His name was John Hempton.
Not only had Hempton spotted the looming Autonomy debacle from the other side of the planet, he had done so through accounting.
Hempton specialises in a particular brand of accounting analysis: he’s an expert at spotting fake accounts in public companies. Indeed, he is considered by many to be the world’s leading expert in fraudulent public company accounts. That was why he was at Santangel’s, and that was why he had analysed Autonomy.
“It was dead easy to spot that Autonomy’s statements weren’t right,” Hempton was quoted as saying in The New York Times.
After explaining fakery to the Santangel’s audience, he had explained that it was possible to bet against a company with faked results and yet lose, because someone with deep pockets might buy the company.
The example he gave was Autonomy.
Hempton’s investment firm, Bronte Capital, has turned the spotting of fake accounts into a business. He and colleague Simon Maher sit in their Bondi Junction offices and look at published accounts from all over the world. The accounts suggest to them that, say, an apparently profitable Canadian pharmaceutical company has really been haemorrhaging cash, or that a Chinese forestry company doesn’t seem to actually own very many trees.
How do you make money out of that? The trouble with faking public company accounts is that eventually the faker is found out. So, the way to make money out of spotting a company’s fake accounts is to ‘sell short’ the company’s stock.
Essentially, you make a promise to deliver at some future date, and for a set price, stock you don’t actually own. You sit and wait for people to realise that the company is a dud and for the stock to get really cheap. Then you buy the stock for next to nothing, sell the shares as promised for the higher price you agreed on earlier, and pocket the difference. Simple – except that many things can go wrong, so your risk management has to be exceptional.
As Hempton tells it, his interest in accounting fraud began in the slightly dingy halls of the Federal Treasury’s tax policy section in the early 1990s. There, he set to work as part of a team analysing the complicated tax arrangements of companies like Alan Bond’s empire, which appeared to be paying far too little in tax. Hempton found to his great surprise that a surprising number of these companies weren’t illegally avoiding taxes at all: instead, they were inflating their published earnings.
