How do you measure the effectiveness of spending intended to prevent terror attacks?

The lack of terror attacks in the aftermath of such spending is plainly a poor guide, as there may have been no terror attacks in the absence of spending. Alternatively, as we saw last week, some forms of war-on-terror spending have increased the likelihood of terrorist attacks that in the absence of spending would not have occurred.

It’s also unclear whether Australian terror plots that have been uncovered and formed the basis of successful prosecutions, such as those of Abdul Benbrika or the planned attack on Holsworthy Army Barracks, would not have been detected and halted by intelligence and police forces operating at pre-2001 levels of resourcing. Nor is it clear how close to practical implementation the plots came: claims leaked to the mainstream media about the Benbrika plot included a plainly fanciful desire to destroy the Westgate Bridge.

In their paper on the costs and benefits of anti-terror funding, Mueller and Stewart offer a means of assessing how much benefit is gained from anti-terrorism spending. They suggest a formula based on the cost of anti-terrorism measures, the projected “average” cost of terror incidents and an assumed figure for how much anti-terrorism spending has reduced the likelihood of attacks, which enable a calculation of how many attacks would have to be prevented to justify anti-terror expenditure.

There’s much that is arbitrary in such a calculation. Mueller and Stewart suggest, based on statements from US counter-terrorism officials, that much of the reduction in the risk of terror attacks is achieved by law enforcement and intelligence capabilities already in place before 9/11 (in Australia’s case, the AFP, ASIO and Customs). They also suggest naturally increased levels of public awareness post 9/11 have played an important role in providing intelligence and tip-offs about possible attacks. Their assumption is that those two factors reduce by half the risk of terrorist attacks before any additional spending comes into play, but they then assume that counter-terrorism spending reduces the remaining risk nearly to zero, suggesting war-on-terror spending cuts the risk by 45% — a very generous estimate.

They also estimated the average cost of terrorist attacks in the tens of millions — using $100 million as a working figure — on the basis that catastrophic attacks such as 9/11 were highly unusual, and most terror plots since then have been much smaller in scale.

The result is, based on their calculated cost of US anti-terrorism funding of $75 billion a year, that nearly 1700 significant terrorist attacks would have to be stopped per year to justify US spending. In fact, levels of US spending are only justified under the formula when you start talking about a 9/11 every year, a nuclear detonation in a port every six years, or a nuclear detonation in New York every 30 years.

Using Mueller and Stewart’s formula in an Australian context doesn’t yield such large figures, mainly because we spend far less than the United States on the war on terror (though Mueller and Stewart did not add into their calculations the costs of Iraq and Afghanistan, despite the anti-terror purposes of those conflicts, possibly because that would skew their numbers so massively as to render the exercise pointless). But the result is still instructive.

Let’s assume the same figures for Australia: a $100 million impact from a terrorist attack, a 0.45 reduction in risk. All we need is the cost. The fiscal cost of the war on terror has been, we previously estimated, $16.7 billion in current dollars, or $1.67 billion per year on average over the past decade.

However, that’s not the full cost of the war on terror, only the burden borne by taxpayers. Others also pay a cost — primarily in aviation, which has received a disproportionate and possibly tragic focus since 9/11.

First, the cost of baggage and passenger screening at airports is borne by the airlines, who pass it on to passengers in the price of tickets. In 2002, the Board of Airline Representatives told a government inquiry passenger screening charges cost between $3.50 and $5.50 per passenger. Airlines have also told other inquiries of the cost to them of putting in place additional security measures. On a conservative estimate — though using indexation — these figures suggest additional aviation security has cost more than $2 billion over the past decade. Then there’s the cost in time to passengers of enduring additional security — walking through metal detectors, being subjected to explosive detection, enduring epic queues at major airports as people removes belts and shoes (all in an environment that, as the failed Glasgow terror attacks showed, create a target-rich environment for any would-be terrorist). Again costed conservatively, using average weekly earnings inflated slightly to reflect business and executive travel, an average five-minute delay and assuming 90% of annual passenger movements are subjected to security (ignoring non-passengers who are subject to the same delays), that’s a further $172 million over a decade.

Using the most conservative figures possible, that’s an additional $2.24 billion in war-on-terror spending by airline passengers on top of the $16.7 billion.

Plugging those numbers into Mueller and Stewart’s formula suggests our spending on the war on terror would be justified if there were 42 significant terror attacks a year. That is, if each of those attacks would have inflicted $100 million worth of damage, and additional security measures put in place since 2001 had reduced their risk by nearly half, it would only have been worth it if there would have been an attack every nine days. Or let’s assume a really catastrophic attack, inflicting $1 billion — say a passenger jet being brought down. That would have to happen four times a year, year in and year out, to justify spending.

In reality, given the 0.45 figure is so generous and figures are so conservative, this probably significantly understates the disproportion between spending and risk.

The point is not a precise accounting of death and destruction, nor an expectation that politicians and public officials should engage in finely calibrated judgment about how many passenger jets can be lost a year; it is that the return on resources dedicated to the war on terror is far less than if similar resources were dedicated to other programs. Indeed, given such a low return, given that the billions spent on the attack on and occupation of Iraq only served, in the view of experts, to increase the risk of terror attacks on Australia, and given some of key beneficiaries of our spending in Afghanistan are the very sources of private funding for terrorism, it is hard to see any evidence that the nearly $19 billion devoted to the war on terror led to any net reduction in risk to Australians.

Our leaders must necessarily adopt a precautionary approach to such a hard-to-assess risk as terrorism —  although it’s amusing that advocates of more counter-terrorism spending are often to be found opposing taking precautionary action on climate change, a much greater risk that will do far more long-term economic damage than terrorism. But a precautionary approach does not equate to abandoning judgment and spending money with no assessment of its benefits. And there’s no evidence we’ve done anything other than waste billions fighting terrorism to no one’s advantage but terrorists themselves and the bureaucrats and private interests that have grown fat on the spending.