If your sense is that road and transit programs almost always cost more than was originally estimated, your sense is correct. A new study by three Danish professors shows that major transportation infrastructure projects built over the past 70 years have habitually gone over budget. To make matters worse, the overruns are not caused by random errors, but rather stem from deliberate distortion and lies to get projects approved. Though not included in the study, cost overruns on several major projects in North Carolina — the Charlotte and Triangle transit systems and the state’s Highway Trust Fund — are remarkably similar to those documented by the Danish researchers.
The paper by Bent Flyvvbjerg, Mette Skamris Holm, and Søren Buhl, titled “Underestimating Cost in Public Works Projects: Error or Lie,” appears in the Summer 2002 issue of the Journal of the American Planning Association. It is the first large sample, statistical analysis comparing forecasted cost to actual cost for transportation-related infrastructure projects.
The core of the paper is a comparison between estimated costs at the time the decisions were made to proceed with the projects and the projects’ final cost. The difference between the estimate and the final cost was the amount of the cost under or over estimation. The authors collected data on 258 transportation infrastructure projects costing a combined $90 billion. They found that costs were underestimated in 86 percent of projects with actual costs, on average, 28 percent higher than estimated costs. For rail projects, the cost underestimation averaged 44.7 percent.
Based upon these findings, the authors conclude that the original cost figures were off not because of some error in estimation, but rather because the forecasters and promoters of projects were engaging in deception and lying to get favored projects started. Often this involved not including all relevant costs in the original estimates.
If rail projects are more likely to have a large cost escalation, then the cost explosion in Charlotte’s transit system should come as little surprise. In 1998, Mecklenburg County voters approved an additional half-cent sales tax to fund a major expansion of the transit system. Theexpanded Charlotte Area Transit System would operate either light rail or buses on dedicated busways on five key corridors in Charlotte. Details of the plan, such as exact routes, location of stops, and even choice of bus or rail for each corridor were not specified. The estimated cost was $831 million.
By 2002, the total cost of the system had ballooned to $2.1 billion. Included in that figure is more than $1 billion in increases in project scope to what was presented to voters just four years earlier. In addition, the costs of four of the five corridors had increased by at least $95 million each.
While Charlotte’s transit plan involves five corridors to bring people to and from its center city, transit in the Triangle has emerged as a means to link cities together. The keystone of the endeavor is a regional rail link between Durham, the RTP, and Raleigh. The first phase of the project was estimated to cost $622 million, cover 35 miles, have 16 stations, and be ready by 2008. After the concept was approved, it proved more difficult, and expensive, to build than was originally thought. As a result, completion of the last five miles (three stations) of the project in north Raleigh will be delayed until at least 2010. Current total cost estimates are at least $754.8 million and will likely be revised upward again later this year.
Cost overruns in North Carolina are not limited to transit projects. In 1989, the General Assembly established the Highway Trust Fund to address some of the state’s road needs. An additional gasoline tax and higher highway use and title fees were to go toward completing 28 designated intrastate projects, seven urban loops, and paving dirt roads. The trust fund, and the additional fees that fund it, were successfully sold to the legislature as being temporary. When the projects were completed, the extra taxes were supposed to have been removed. At the time, completion of the specified projects was estimated to take 13.5 years to complete.
The reality has proven different. A 1998 audit showed that the state had not considered inflation or the possibility of cost overruns in its cost calculations. Planning and engineering costs and the expense of conducting environmental impact studies were also not considered The completion date remains uncertain, though the state auditor projected 2019 in his 1998 audit of the Trust Fund.
The wise use of public funds is based upon a project’s benefits being greater than its costs. All too often — in North Carolina and elsewhere — cost estimates are being used that bare little resemblance to reality. As long as these distortions continue, poor public decision making will inevitably be the result.