Tuesday, July 22, 2014

MH17. Why planes and financial systems crash

What does the crash of Malaysia Airlines flight MH17 have to do with the global financial crisis?

One was destroyed by a surface-to-air-missile, the other came about because huge numbers of American housing loans became worthless at once. Enabling each was a bet that the unlikely wouldn’t happen.

It’s usually a good bet.

Qantas, Korean Air, and Taiwan's China Airlines weren't prepared to take it. They rerouted their flights to avoid the Ukraine months ago. Their caution cost them fuel, travelling time and profits.

Airlines such as Malaysia, Singapore and Lufthansa took a punt.

“What logic, what lack of sensitivity, and what lack of basic decency influenced Singapore Airlines and Malaysia Airlines and others to expose their passengers to these risks?” asked aviation journalist Ben Sandilands on his blog Plane Talking.

The logic was that the unlikely probably wouldn’t happen, or at least wouldn’t to them. Being slaughtered while flying well above a war zone is what experts call a low-probability, high-impact event.

Coldly risking something catastrophic in the knowledge that it almost certainly won’t happen (at least not to you) is a way to deliver superior financial returns, right up the point when it is not. And it’s rife in the finance industry.

Fund managers get paid for performance. Well ahead of the financial crisis in early 2008 two academics from Oxford and Pennsylvania universities demonstrated that it was possible for a fund manager to consistently deliver superior performance by betting the fund that an unlikely catastrophic event wouldn’t happen.

If, as was highly likely, the catastrophe never occurred the bet would pay off and they would be rewarded for their superior performance. If it eventually did occur they would have already received their bonuses and could leave the fund to collapse, moving on to a new job.

Because fund managers keep their methods secret professors Peyton Young and Dean Foster said it was “virtually impossible to set up an incentive structure that rewards skilled hedge fund managers without at the same time rewarding unskilled managers and outright con artists”.

As they put it, “anyone can cobble together a car that delivers apparently superior performance for a period of time and then breaks down completely”. Airlines can do it, privatised electricity suppliers can do it by not investing in maintenance as Victoria has discovered to its cost during brownouts, and state governments can do it by continuing to allow building in flood prone locations as Queensland did before its most recent devastating flood.

The entire world can do it by acting as if climate change won’t be too serious (although that’s probably better described as a medium to high probability high-impact event).

And the manufacturers of financial products can do it...

In the leadup to the global financial crisis they created products sprinkled with loans that could never be repaid if housing prices fell. But they bet that prices wouldn’t fall, not all at once. Compliant ratings agencies produced estimates of how unlikely such an event was. When it happened the products and the financial institutions that created them became worthless. The government rescued the important ones and much of the world slid into recession.

There’s no quick fix to stop it.

Part of the solution is better regulation, something the world’s financial authorities are on to after the global financial crisis. But regulation usually only closes a door after a crisis. Then a different unforeseeable event occurs creating another crisis creating another regulation.

As strange as it seems rewards for performance are probably a bad idea. They are what encouraged reckless practices in the United States. If Qantas schedulers were paid bonuses for speed they might have been keener to flirt with danger.

On the other hand real ownership could help. Paying employees in shares that couldn’t be cashed in for years would encourage them to be careful, as would requiring maintenance engineers to engrave their names of the fuselage of the planes they repair - a practice that is said to take place in Japan.

The best antidote is probably a rigorous cost benefit and risk analysis performed by someone whose pay cheque doesn’t depend on the next month’s profit.

The Coalition embraced such an idea in its September election policy. All Commonwealth infrastructure spending exceeding $100 million was to be “subject to analysis by Infrastructure Australia to test cost-effectiveness and financial viability”. Even state government projects only partially supported by the Commonwealth were to face Infrastructure Australia scrutiny.

No longer. On Thursday the government rejected a Senate resolution that would have given effect to its own policy. Labor moved that the reward payments made to states that privatise assets and then use the proceeds for new projects be subject to Infrastructure Australia cost benefit analysis. It would have covered the Metro Rail project and anything else funded by the sale of right to use the Port of Melbourne.

The Coalition said no.

Commonwealth cost benefit statements were “red tape with no additional benefit”. They risked delaying “the delivery of critical infrastructure”. They would “stand in the way of the government building a stronger more prosperous economy and investing in new infrastructure,” according to the finance minister Mathias Cormann.

But when lots of money or lives are at stake, delay is often a good idea. There’s a lot to be said for caution.

In The Age and Sydney Morning Herald


Related Posts

. 2008. Getting the truth out of Qantas

. 2008. The sub-prime primer

. "Things were right on the edge" - the crisis as it unfolded