Toll tunnels in Australia

[cpzilliacus]

TOLLROADSnews: Toll tunnel travails down 'unnah - Parsons Brinckerhoff turns tables on litigant in Oz

Toll tunnels in Australia may not be generating the revenues expected from motorists but they're proving goldmines for lawyers. Investors in the tunnels are suing the traffic and revenue forecasters claiming they were misled by their rosy financial scenarios.

Such lawsuits are difficult win because the forecasts are published with carefully phrased caveats about how the traffic and $-number projections are dependent on the assumptions chosen and are not really forecasts.

To win you have to go way further than showing how wrong they turned out. You normally have to show the forecasters deliberately worked to get overoptimistic and false results and published them knowing they were unrealistic. Or you have to demonstrate gross professional negligence in data collection, choice of assumptions or modeling.

[Alps]

Traffic forecasters are not immune. I'm working with PB right now and their work up here is top-notch. Some other forecasters have blatantly skewed, omitted, or plain made up numbers to arrive at the solution that either they or the client wanted. I've had markups that are more red pen than white paper in reviewing such documents. They can choose to heed our warning or plow on ahead, but we have a record so if a lawsuit like this happens, we can say, "Told you so!"

[Chris]

I think most people here are familiar with the megaproject research of Mr. Flyvbjerg of Denmark. He did extensive research on megaprojects over the last couple of decades. One key element with megaprojects is overestimated usage and underestimated cost of construction. Although road traffic forecasts tend to be much more accurate than transit ridership forecasts, usage of planned toll roads is generally overestimated.

Most infrastructure projects have a certain ramp-up period, in which people get familiar with the new options and spatial developments may occur. The ramp-up period tends to be much longer in recent years, due to the weak economy. The problem with toll roads is that they need good results from the very beginning to service its debt and maintain an acceptable credit rating.

[cpzilliacus]

I think most people here are familiar with the megaproject research of Mr. Flyvbjerg of Denmark. He did extensive research on megaprojects over the last couple of decades. One key element with megaprojects is overestimated usage and underestimated cost of construction. Although road traffic forecasts tend to be much more accurate than transit ridership forecasts, usage of planned toll roads is generally overestimated.

Bent Flyvbjerg is indeed Danish, but he's currently working in the UK at Oxford University (details here).

The worst forecasts that I have personally seen have been for passenger rail transit projects.  But sometimes the fault is not with the forecasters, but the data that went in to those forecasts.  The Washington Metrorail system was supposed to be carrying many more riders than it actually did through the 1980's and 1990's, but the forecasts of population did not take into account the crash  in resident population of the District of Columbia that happened during the crack cocaine wars and the overall mismanagement of the city during the Barry years.

Most infrastructure projects have a certain ramp-up period, in which people get familiar with the new options and spatial developments may occur. The ramp-up period tends to be much longer in recent years, due to the weak economy. The problem with toll roads is that they need good results from the very beginning to service its debt and maintain an acceptable credit rating.

This is correct.

[cpzilliacus]

Traffic forecasters are not immune. I'm working with PB right now and their work up here is top-notch.

I have met some very good forecasters that work for PB.

I won't get in to more specifics here (for personal and professional reasons), but forecasts by different technical staff people working for the same firm can be highly variable.

Some other forecasters have blatantly skewed, omitted, or plain made up numbers to arrive at the solution that either they or the client wanted. I've had markups that are more red pen than white paper in reviewing such documents. They can choose to heed our warning or plow on ahead, but we have a record so if a lawsuit like this happens, we can say, "Told you so!"

IMO, bad forecasts are more about the firm not wanting to "disappoint the customer" (and in this context, customer frequently includes elected officials).  Bad forecasts can also be about unreasonable or unrealistic demographic forecasts, or about land use assumptions that fail to come true (land use matters more if the forecast is for future transit patronage as opposed to toll road or "free" road traffic).

[Alps]

Traffic forecasters are not immune. I'm working with PB right now and their work up here is top-notch.

I have met some very good forecasters that work for PB.

I won't get in to more specifics here (for personal and professional reasons), but forecasts by different technical staff people working for the same firm can be highly variable.

Hence "up here" in my OP. Well aware.

Some other forecasters have blatantly skewed, omitted, or plain made up numbers to arrive at the solution that either they or the client wanted. I've had markups that are more red pen than white paper in reviewing such documents. They can choose to heed our warning or plow on ahead, but we have a record so if a lawsuit like this happens, we can say, "Told you so!"

IMO, bad forecasts are more about the firm not wanting to "disappoint the customer" (and in this context, customer frequently includes elected officials).  Bad forecasts can also be about unreasonable or unrealistic demographic forecasts, or about land use assumptions that fail to come true (land use matters more if the forecast is for future transit patronage as opposed to toll road or "free" road traffic).

There are many possible reasons. Our job is to do the best we can with the information we have, and oftentimes that is simply not done. Project I'm working on, we made some fairly aggressive (though justifiable, because they are all planned projects) growth assumptions, and the forecasts are not as high as we might have liked (or even thought, in my case, though the more I learn about it, the less surprised I am). The numbers are what they are, and it's our job to figure out what the benefits will be, and present that to the client. We can recommend to build or not, but it's my professional responsibility not to hide anything in that recommendation.

[cpzilliacus]

There are many possible reasons. Our job is to do the best we can with the information we have, and oftentimes that is simply not done. Project I'm working on, we made some fairly aggressive (though justifiable, because they are all planned projects) growth assumptions, and the forecasts are not as high as we might have liked (or even thought, in my case, though the more I learn about it, the less surprised I am). The numbers are what they are, and it's our job to figure out what the benefits will be, and present that to the client. We can recommend to build or not, but it's my professional responsibility not to hide anything in that recommendation.

You are doing an ethical and honest job.

As a former boss of mine put it, "it's not the  model, it's the inputs to the model."  That especially includes demographics and transportation network assumptions.  Not to say that models cannot come up with crazy outputs, but it is usually not the fault of the model.

Stated another way, IMO, if you feel any of the inputs to the process are "funny" or unreasonable, then I suggest you highlight them to the client, and make sure he or she understands why you feel that way.