The rift over the real cost of Haiti's earthquake

It was called the worst disaster in modern history. A UN spokesman described "historic" devastation "like no other" ever faced by the organisation. News reports heralded scenes of Biblical devastation, complete with piles of corpses in the streets and lorry-loads of bodies being dumped in graves on the outskirts of town.

But how many people really did lose their lives when an earthquake measuring 7.0 on the Richter Scale struck Haiti during the evening rush hour on 12 January, 2010? Was the death toll in the tens of thousands, or hundreds of thousands? And did a mixture of cock-up and conspiracy result in the scale of destruction being dramatically overstated?

The US Agency for International Development (USAID) seems to think so. In a leaked report, the organisation has concluded that the actual number of deaths in last year's disaster was somewhere between 46,000 and 85,000, well short of the estimates of between 200,000 and 300,000 made by most aid groups, and a fraction of the 318,000 claimed by the Haitian government.

Numbers matter, because they help to justify the vast and ongoing aid effort in which around $5.5bn (£3.3bn) has been pledged to the impoverished Caribbean nation by overseas governments (but of which only a portion has actually been delivered). The numbers also continue to inspire private donations to the roughly 3,000 charities and aid agencies which are still invested there.

The US report has not yet been published, but its alleged findings have already been disputed by both Haitian authorities and the UN. Even the State Department in Washington is, for now, reluctant to endorse it, saying that "internal inconsistencies" in some of the statistical analysis are currently being investigated prior to publication.

"We are reviewing these inconsistencies... to ensure information we release is accurate," spokeswoman Preeti Shah told the Associated Press, which on Monday published details of the leaked document.

The draft was compiled for USAID by a private consulting firm, LTL Strategies, which claims to have arrived at its revised death toll by conducting extended interviews in 5,200 homes in Port-au-Prince earlier this year. Respondents were asked more than 100 questions, including how many people died in each building, and where the survivors went. As well as concluding that the death toll was exaggerated, the report claims the number of people made homeless by the disaster – originally estimated at 1.5 million by the UN – was in fact 895,000. And while UN officials estimate that 680,000 Haitians are still living in temporary settlement camps, the real figure is closer to 375,000, the report concludes.

A third key finding suggests that the amount of rubble produced by the quake, which the US Army Corps of Engineers put at 20 million cubic metres, is in fact less than half that amount. The expensive fleet of lorries still working to clear ruins may therefore be finished sooner than previously thought.

The report has the potential to be hugely controversial, since it speaks directly to concerns that international relief efforts create a culture of dependency and corruption. Haiti is already often cited as proof that disasters, and the overseas aid spending they prompt, are all too easily exploited by dishonest locals.

Last January's official death toll has always been eye-opening, since at 318,000, it represents roughly 3 per cent of the entire population of Haiti. Many wondered how officials were able to release precise figures throughout the initial weeks of the catastrophe, even as bodies were being scooped haphazardly off the streets and dumped in mass graves.

Haiti's government has never revealed the methodology behind its figures. And since the majority of public records were destroyed during the quake, they are impossible to confirm. Anyone who was in Haiti during January 2010 can of course attest to the horrifying scale of suffering. In some areas, corpses were piled high on pavements. Field hospitals were overwhelmed, and nearly everyone you spoke to seemed to have lost at least one member of their immediate family.

Enumerating that suffering is not an exact science, though. The UN, for example, lost 96 – roughly 1.4 per cent – of its 7,000 peacekeepers in the country. If the same proportion were to be applied to the entire population of Haiti – 10million – then one might estimate that 140,000 people would have been killed. But UN peacekeepers living in well-built compounds could expect to have had a far higher survival rate than impoverished slum-dwellers who had constructed their homes out of breeze blocks.

As the leaked USAID report is the result of formal research, questions are already being asked as to its integrity. The 5,000 or so people whom researchers surveyed all came from Port-au-Prince, critics have noted. But some of the worst-hit locations were outside the capital, in towns such as Léogâne, Jacmel, and Petit-Goâve. The man who led the survey for LTL Strategies, meanwhile, has a colourful past. Timothy T Schwartz, an author and anthropologist, in 2008 published Travesty in Haiti: a true account of Christian missions, orphanages, fraud, food aid and drug trafficking – a book highly critical of mainstream aid organisations.

In a blog entry which appears to offer a pre-emptive response to the controversy his survey will cause, Dr Schwartz has stressed that, even if the lower end of his estimated death toll is accurate, the January 2010 quake was still a "huge" disaster.

"Intellectually, I really don't care how many people got killed in the earthquake," he wrote. "The draft report for USAID was simply a job I was performing with a team of some 20 university-educated professionals, including two other PhDs. But personally, for me, in terms of the tragedy, less is better. And at about 60,000 dead, that's still a huge tragedy."