© Bent Flyvbjerg Bent Flyvbjerg is a professor at the University of Oxford and co-author of How Great Things Were Created.Bent Flivbjerg, a professor at Oxford University, tried decades ago to track how often large infrastructure projects were completed on time and on budget. He discovered that no one was following this, so he created his own database. What he found was unfortunate: the vast majority of megaprojects — costing at least $1 billion — had neither a budget nor a timeline. It wasn't just about concrete and steel—IT upgrades were one of the worst.
DC had no shortage of projects that didn't fit their original budget and schedule. The second phase of the Silver Line, which extended Metrorail to Dulles International Airport and Loudon County, opened four years late and over budget by at least $250 million. Maryland's Purple Line light rail project is more than $1 billion over budget and 4.5 years behind schedule. Another seven months could be added to that delay as contractors grapple with moving utilities.
With billions of dollars invested in U.S. infrastructure projects under the Infrastructure Act, ensuring that budget and deadlines are met will be critical, especially given inflation and supply chain issues that have made construction more expensive.
In a new book, How Big Things Get Done, Flivbjerg and co-author Dan Gardner outline steps they think will help decision makers (and even owners) avoid the pitfalls that could lead a project down the wrong path. Flivbjerg spoke to the Washington Post about factors that could determine the fate of projects. This interview has been slightly edited for length and clarity.
Q: Tell us about your background and how you got interested in studying megaprojects?
Answer : I am an economic geographer by education. I study the economy of cities and the economy of regions. I quickly realized that large projects are playing an increasingly important role in the development of cities and regions, so I began to study the world through the prism of large projects. I started collecting project data and now we have over 16,000 projects in our database. Over time, we understand better and better what is really happening.
Q: You've gathered some pretty amazing statistics about the percentage of projects that are actually completed on time and on budget. What did you find?
A: Approximately 8.5% of projects were completed on time and within budget. We were surprised because this is a very low figure. But the numbers are strong. We tested it on different sample sizes and it holds up. Now our colleagues have begun to conduct studies similar to ours, and they are getting similar results. Therefore, we believe that the number is valid and reliable.
The second stage of the Silver Line is more sinister than the first: here are a few reasonsQ: There are successful projects from all eras - for example, the Empire State Building. Why do you think we don't learn any of this?
Answer: There is a striking lack of training in construction. My theory is that this is because there is no real competition [because] construction is fragmented in local markets. You have small geographical monopolies. New York would be one market, and Boston would be a completely different market. You don't have a Tesla assembly. If you made a poor-quality car, you would quickly go bankrupt. Cars made in the Soviet Union and Eastern Europe were subject to international competition and the brands disappeared. This is the main statement. You can survive in the local market that produces low-quality buildings.
Q: We tend to think that only major construction projects are behind schedule and over budget, but have you noticed that it's not just construction projects - bridges, roads, subways - and other types of projects that fall into this trap. Why do similar problems haunt megaprojects in other industries?
A: It surprised us. What is like building a skyscraper and building a computer system? Turns out it's because people do all these things, and people have certain characteristics, certain biases, and we bring those biases into everything we do. Some kind of ignorance, some kind of optimism, etc. It's mostly about psychology and strength. That's why we find this commonality between projects. They show the same picture whether or not they are completely different projects.
Q: Let's talk about solutions. How do you make sure a project starts on the right foot?
Answer: You need to check two things: good funding and a good team. Get the right interface by asking, "Why are we doing this project?" Our brain tends to work with the first idea that comes to us, the first idea that comes to mind. “What basic need are we trying to satisfy here? How can we meet this need and so on. You need to address these issues as early as possible and make sure everyone agrees on why you are doing this and on the best solution.
Maryland Purple Line faces another seven-month delayQ: Can you explain what you mean by "think slow, act fast"?
A: I want to emphasize that this doesn't necessarily mean you have to think too long, but it does mean you slow down and try to avoid [prejudices] that might be holding you back. Know why you are doing this project. And once you think slowly, you can act quickly. You can deliver much faster if you have a really good plan. Part of that process is goal setting, which seems so obvious. But in many projects, people get it wrong or miss the answer.
Q: But the goal setting seems so obvious. Why do people fail?
Answer: People may actually disagree with the goal, but they don't know they disagree until they start to implement the project. Then they realize, "Well, that's not what I meant." There can be many different stakeholders who have different ideas of what the project should look like. This happens frequently.
Question: In your book, you mentioned the architect Frank Gehry as an example of a person who uses a certain process to achieve the right goal. Can you explain how this is done?
Answer: When Frank Gehry finds new clients, he doesn't just accept what the client wants. He asks: “Why? For what? For what?' Keep asking that question and move on to the real reasons why people do it. Then, once you understand why, start modeling and iterating the design, first physically using wood, cardboard, and other materials. He then digitizes the models and uploads them to a computer to get a lot more detail and speed up iteration. After hundreds of hundreds, and sometimes thousands of times, he can repeat the building or make different details on it to improve it. This is called the digital twin. This is a very cheap way to do it compared to building. So now when they go to their workplace, it's much easier. Every nut and bolt is pre-specified on the computer, so there are very few surprises.
Q: Do you have any advice for the state transportation departments that allocate money to places to make sure everything runs smoothly, that they have proper oversight?
Answer: You need a governance structure with the right incentives to get real results. If you don't properly incentivize the delivery team and designers, you can't expect them to work. It is very important to have incentives that match your goals. And in fact, the role of the state and federal government is to develop the necessary tools and procedures.
Q: What incentives work?
A: In many ways it's actually very difficult because it's about writing contracts. The secret is to make sure people have skin in the game. For example, you will earn more money if you submit an application, and less money if you do not. But money is not enough. Contracts can actually make the parties involved very antagonistic towards each other, which is why people start talking about a different kind of contract: more about partnerships than contracts. Partnership works better than trying to blame the other parties by saying, "Let's work together to achieve this while minimizing risk."
