The gold medal holder in Olympic cost overrun in dollar terms remains Sochi (2014). The final price tag for Vladimir Putin’s folly was a staggering $16.3 billion higher than the planned $5.6 billion. Silver is retained by Barcelona (1992) with a cost overrun of $7 billion while
London (2012) claims bronze with a $6.5 billion overspend.
Rio, with an overrun of $1.5 billion, ranks as a mid-level Olympic blowout. However, cautions Bent Flyvberg, “This comes at a time when Brazil can least afford it, given that it’s facing its worst economic and political crisis since the 1930s.”
Flyvberg, chair of Major Project Management at Oxford’s Said Business School, defines mega projects as those costing more than $1 billion and affecting more than 1 million people. With sports-related costs ranging up to $21.9 billion (Sochi), the Olympic Games certainly fall into that category.
But all mega-projects have the potential to be mega-disasters. Berlin’s Brandenburg Airport is one famed embarrassment. Costs have quadrupled to €8 billion ($9 billion) and the futuristic airport remains unopened even though it was originally scheduled for 2012. Berlin’s farce is only an extreme version of the problems that plague many projects.
“Most mega-projects fail,” says Flyvberg. Most is nine out of 10. Nine out of 10 have cost overruns, benefit shortfalls or are not completed on time. To obscure the issue, some projects fail financially but are still considered a success. Certainly passengers zipping along the seabed of the English Channel appreciate the speed, simplicity and lack of seasickness of the transit. Few would recall that the project was initially insolvent and had a return on investment of -14.5%.
Mega-project failures are so common that Flyvberg has over 200 from 20 countries and five continents in his database. Conversely, there are too few successes to complete reliable analysis. And yet we continue to start new projects.
As populations grow and cities develop, mega-projects can provide solutions to some of the most pressing transit, water security and natural resource management problems, as well as providing the information infrastructure required for today’s world. Unfortunately, mega-projects as they currently work are a poor delivery model.
THE BIRTH OF A MEGA-PROJECT
The question arising from Flyvberg’s research is why we continue to be so bad at building the things we so desperately need. Infrastructure planning, he explains, is based on the business case, which sums up benefits against costs. When benefits outweigh costs, building begins, but planners are notoriously inaccurate in estimating costs. Business cases are riddled with errors, with costs creeping up as scope increases. In rail projects, for example, costs are on average 44.7% over forecast whereas ridership only meets 51% of forecasts.
Planning errors would be forgivable if the industry showed improvement. However, as his database shows, over the last 70 years there has been none. Without improvement, Flyvberg dismisses incompetence as an explanation. He points out more cynical reasons why projects cost more and benefits are less than expected.
Mega-projects often require a hybrid of public sector support and private sector input, with neither having the capacity to complete the project without the other. It can be one of the rare times when co-operation between trade unions, private companies, consultants, citizens and politicians is not only essential but in everyone’s interest. These stakeholders all want the mega-project to begin, and they interact in a way that forces it off the drawing board and into the public sphere.
Unfortunately, those who benefit most are often the ones identifying costs, developing timelines and forecasting benefits. With so much support for mega-projects, the business case is often pre-determined with little critique from insiders. Outsiders are excluded entirely.
In reality, benefits and costs are more complex. Large projects often have long-planning horizons or non-standard design, or they are the first of their kind. Benefits are sometimes financial, but often represent improvements in quality of life. This makes forecasting more art than science and therefore prone to abuse, with costs and benefits adjusted as needed.
Those involved are not naïve, but former mayor of San Francisco,
Willie Brown, was more brazen than most when he dismissed questions on the Transbay Terminal cost overrun. Speaking at the time that the first budget shortfall became apparent, he said, “News that the Transbay Terminal is something like $300 million over budget (now $700 million) should not come as a shock to anyone. We always knew the initial estimate was way under the real cost.”
He then confirmed the real issue: “The idea is to get going. Start digging a hole and make it so big there’s no alternative to coming up with the money to fill it in.”
OLYMPICS – A MODEL FOR DISASTER
Improving mega-project planning and delivery is possible. Flyvberg advocates a move away from current forecasting methods. Rather than allowing planners to be influenced by excitement and optimism, he recommends reference class forecasting. This ignores specific details of any one project and instead looks at the timelines, costs and benefits of similar project types, which allows for a less biased and more realistic representation.
He also highlights the role of the private sector, which typically delivers mega-projects. Procurement methods such as cost plus (where governments pay the costs of a project and a percentage as profit) dis-incentivize cost reduction and allow companies to shift the risk onto the government. An open and transparent bidding process with financial responsibility shared between the public and private sectors is essential to managing costs.
If these are some of the elements that make successful mega-projects, it is easy to see why the Olympics are unsuccessful. They represent once in a lifetime events for most locations. Those involved have no experience in planning mega-projects, local industry has never built anything on a similar scale, the bidding process is often rushed, non-transparent and corrupt, and cost responsibility is entirely borne by the host city.
“If you wanted to make it as difficult as possible to deliver a mega-project on budget, you would follow the template of the Games,” says Flyvberg.
Aware of a growing reluctance by potential hosts to bid for the Games because of costs, the
International Olympic Committee developed a Knowledge Management Program in the early 1990s. This shares knowledge between past and future host cities. Since the program’s beginning, the median cost overrun has fallen from 231% to 51%. However, the average cost of the Olympics has increased, with London being the most expensive summer games and Sochi the most expensive winter games ever.
Cost overruns aside, the Olympic Games are also a significant burden on hosts, particularly in developing countries. To put that into perspective, the average Games cost $500,000 per athlete. Rio was lower, at a cost of $400,000 per athlete, but it was still more than ten times the average annual salary in Brazil. If the country had invested that money in its citizens, it may not be experiencing its current political and economic turmoil.