Delusion and Deception in Large Infrastructure Projects, Bent Flyvbjerg, et al, California Management Review, Winter 2009

This research report written by Dr. Bent Flyvbjerg, one of the world's leading experts in infrastructure planning, with Massimo Garbuio and Dan Lovallo provides an important overview of the many factors that cause infrastructure plans, especially those for transportation, tend to arrive later and be more costly than anticipated.  The article notes (emphasis added):

There are some phenomena that have no cultural bounds such as maternal love and a healthy fear of large predators. We can add to this list the fact that, across the globe, large infrastructure projects almost invariably arrive late, over-budget, and fail to perform up to expectations. Cost overruns and benefit shortfalls of 50 percent are common; cost overruns above 100 percent are not uncommon. For example, in one study of major projects in 20 countries, nine out of ten projects had cost overruns.4 Similarly, a study of 44 urban rail projects—in North America, Europe, and developing nations, including London’s Tube and the metros in Washington, D.C., and Mexico City—found that the average construction cost overrun in constant prices was 45 percent; for a quarter of the projects, cost overruns were at least 60 percent. In addition, passenger ridership was, on average, 50 percent lower than forecast. Furthermore, for a quarter of the projects, ridership was at least 70 percent lower than estimated.  An appropriate slogan seems to be “over budget, over time, over and over again.”

 The causes are straightforward:

The underlying reasons for all forecasting errors can usefully be grouped into three categories: delusions or honest mistakes; deceptions or strategic manipulation of information or processes; or bad luck.  Bad luck or the unfortunate resolution of one of the major project uncertainties is the attribution typically given by management for a poor outcome. . . .

. . . Both delusion and deception see the high failure rates for ventures as a consequence of flawed decision making. According to the first explanation— delusion—the flaw consists in executives falling victim to what psychologists
call the planning fallacy.  In its grip, managers make decisions based on delusional optimism rather than on a rational weighting of gains, losses, and probabilities. They overestimate benefits and underestimate costs and time. They involuntarily spin scenarios of success and overlook the potential for mistakes and miscalculations. As a result, managers pursue initiatives that are unlikely to come in on budget or on time, or to ever deliver the expected returns.

According to the second explanation—deception—decision making is flawed by strategic misrepresentation or the presence of what economists refer to as principal-agent problems. Whereas the first explanation is psychological, the second is due to the different preferences and incentives of the actors in the system.  In this situation, politicians, planners, or project champions deliberately and strategically overestimate benefits and underestimate costs in order to increase the likelihood that their projects, and not their competition’s, gain approval and funding. These actors purposely spin scenarios of success and gloss over the potential for failure. This results in managers promoting ventures that are unlikely to come in on budget or on time, or to deliver the promised benefits.

This figure from the report highlights the importance of delusion and deception in large infrastructure projects, especially rail projects. 

 
The authors detail various specific causes falling within these three categories and move on to some recommendations about how to limit delusion and deception in large infrastructure projects from both inside and outside perspectives.  Within the latter category, the authors highlight the importance of reference class forecasting.   In brief, this process involves comparing the current proposal with the experience of a group of completed projects of similar size, complexity, etc.  Typically, such a review shows not only the weaknesses in cost and schedule forecasting, but the many specific sources of error that can lead to budget and schedule overruns.  The article notes that the American Planning Association has recommended this procedure for large infrastructure projects. 

Why is this important?

Underestimating the costs and overestimating the benefits of a given project results in an artificially high benefit-cost ratio, which in turn leads to two problems. First, the project may be started despite the fact that it is not economically viable. Second, a project may be started instead of another project that would have yielded higher returns had the actual costs and benefits of both projects been known.

To read the full article, click here.

Who will buy Dulles Toll Road bonds and pay the operating costs?

In a recent post on TollRoadsNews. com, editor Peter Samuel has written a lengthy article regarding Virginia Attorney-General Ken Cuccinelli's opposition to Virginia contributing further funding to the construction of the Silver Line.  Under the LaHood MOA, Gov. McDonnell pledged to propose to the General Assembly that Virginia pay $150 million to help cover debt service costs in the first five years of Dulles rail's Phase 2 debt issuance.  That still has to gain legislative approval.  The article notes Cuccinelli's long-time opposition and says:

Cuccinelli who is from northern Virginia says the whole rail project is a disaster for the region and the state from the huge debts being run up, and the losses that will be incurred and subsidies needed for a facility few will use. High tolls will hit not only Dulles Toll Road users but other residents as traffic diverts.

Late in this article, Samuel raises questions about who will pay for Dulles rail:

There has been little discussion of this but it far from clear that the MWAA will be able to sell $2,700m of toll revenue bonds in the capital markets without further guarantees from area governments.  And even if such guarantees were given state and local budgets could be in such difficult straits as to lack credibility to bond investors.  The financing is far from assured.

Also hardly discussed is who will pay for the operating losses of the rail line.  MWAA and the Toll Road are being stretched to finance capital costs.   The deal is that on opening the line is handed over to the Washington Metropolitan Area Transit Authority (WMATA) to provide service and subsidies to cover the difference between fare revenues and operating costs.  WMATA has major challenges maintaining its existing 5-line 106 mile, 171km network, the oldest parts of which require expensive excavator replacement, trackwork and new signaling and control systems.  Union contracts make its labor costs high.

The Dulles rail line will be WMATA's Silver Line added to Red, Blue, Orange, Yellow and Green Lines and will add over a fifth to its trackage (23 miles added to 106 miles) and an eighth to its number of stations (11 added to 86).

Fares barely cover half of the operating costs of the WMATA system (including buses) and the ability of area governments and the federal government to cover a widening deficit is in question too.

One vital financial topic the article does not address is whether the revenues generated by the much higher tolls on the Dulles Toll Road, if approved by the "funding partners," will be adequate to cover the debt servicing costs, forcing a default on the bonds.  The "optimism bias" in toll and revenue forecasts has been well documented, including by the the NAS Transportation Research Board.  This bias has led to defaults on bond payments and bankruptcies that have forced the sale of the roads to for-profit ventures or led to state and local taxpayer bail outs.

Wilbur Smith Associates (WSA)--the consultant doing the toll and revenue forecasts for MWAA--has not been immune from this outcome.  It repeatedly over-estimated the revenues for the Southern Connector (Greenville, SC) and the San Joaquin Hills (CA) toll roads that was a major factor in their default and bankruptcy.  In Virginia, WSA prepared the toll and revenue forecasts for the Pocahontas Parkway (near Williamsburg) that forced the not-for-profit company managing it to sell to for-profit investors when toll revenues fell far short of WSA forecasts.  Moreover, its revenue projections for the yet-to-be-approved Knik Arm bridge (Anchorage, AK) and the Detroit River International Crossing (DRIC) have been challenged in alternative toll and revenue forecasts as being overly optimistic.

Devolution's Impact on Fairfax Property Taxes

The Potomac Almanac today has an article, "Fairfax to Richmond: Do Us No Harm," discussing the issues in "devolution" of funding for the maintenance and repair of county roads as viewed by local legislators.  The article covers a variety of issues related to devolution, but the money quote (literally) is the following:

Tom Biesiadny, director of Fairfax County’s Department of Transportation, estimated that Fairfax County would have to make a one-time investment of up to $208 million to take local control of local roads, which would mean a property tax increase of about $479 for the average household.

 Click on the article title above for the very good full article.

RA to Consider Indoor Tennis at December 15, 2011, Meeting, GoReston.com

GoReston reports:

RA Board to consider costly indoor tennis facility at Dec. 15 meeting

by Guest Posts on December 13th, 2011

The RA Board will discuss a highly controversial proposal to build an indoor tennis facility at Lake Newport on Thursday, December 15 at 7:00 p.m.  RA is considering spending up to $13 million to build a 46,000 square foot indoor tennis facility to benefit less than five percent of Reston’s residents. RA would spend $75,000 on a referendum if they decide to move forward with the project.  Reston homeowners are encouraged to attend the meeting. Location: RA Headquarters, 12001 Sunrise Valley Drive.

For more on this topic, see Reston Patch's latest article by clicking here.

Will Reston's TOD Housing Demand Follow California's Trend?

The Urban Land Institute (ULI), one of the premier urban design organization in the United States, has just issued a study looking at demand for housing in California.   These are the report's five principal findings:

First, the existing supply of conventional-lot (over one-eighth acre), single-family detached homes exceeds the projected demand for these homes in 2035. This finding does not mean there is no market for new conventional-lot homes in niche markets. It does mean that overall the expansion of the supply of conventional-lot, single-family detached homes would be in excess of current and projected demand (see figure 1).

Second, housing and neighborhood preference surveys indicate that Californians consider transit options to be far more important in choosing a location in which to live than the rest of the nation: 71 percent in California, compared with 47 percent nationally. The demand in 2035 for residences located within one-half mile of public transit stations—called transit-station areas, or TSAs (transit service areas)—will exceed the aggregate amount of current supply plus all new residential units built in these metropolitan areas between 2010 and 2035 (see figure 2 and table 1).

 

 

Third, through modest redevelopment that will happen anyway, existing developed land with nonresidential uses could be sufficient to accommodate all new jobs created over this period. In particular, existing and potential TSA development may have sufficient capacity to accommodate 7 million jobs, or more than enough to absorb all new jobs between 2010 and 2035 (see table 1).

Fourth, changing demographics in combination with changes in home mortgage finance will reduce the rate of homeownership in California by up to 5 percent from 2010 levels and perhaps by as much as 10 percent over the long term. A 5 percent  reduction represents a market condition where three-quarters of the demand for new housing in the state’s largest metropolitan planning organization (MPO) areas will be for rental housing. This demand should lead to an increase in existing residential units being used to house multiple or intergenerational households
as well as to a variety of hybrid or new housing formats, such as accessory dwelling units or new nontraditional multifamily housing options.

Fifth, these long-term market trends represent a directional alignment between the real estate preferences expressed by consumers and the greenhouse gas reduction objectives expressed by the state of California in the form of Senate Bill (SB) 375.

The vital points from these findings is that ULI anticipates an eight-fold demand in residential housing in TOD (or TSA) areas in the next 25 years (9.2M HUs--18.4M residents), but supply will only triple (about 3.5M HUs--7M people) in that timeframe.  In contrast, employment growth over the next 25 years is expected to slightly more than double (3.0M to 6.5M). 

If this kind of situation applies in Reston, current proposals to roughly balance incremental housing unit (HU) and employment construction could lead to a serious housing shortage and possibly an over-supply of commercial space in Reston's TOD areas.

UPDATE:  On December 14, KPBS, San Diego, published an article on the ULI report above titled, "Study: Single-Family Homes May Be History."  A couple of key quotes:

The study looked at California’s major metropolitan areas — including San Diego — and found that by 2035 the supply of homes in conventional subdivisions will far exceed demand. . .

 On the flip side, Nelson's study predicts that demand for town homes, multifamily units and small lots will far exceed the supply of these types of housing over the next 25 years. Demand for housing that's close to public transit will also outstrip demand, according to the study.

"The bottom line is that as many as 9 million households would like the option to live in locations served by public transit," Nelson wrote. "But today, only about 1.2 million California households can claim to have it."

The study relied on demographic, employment and housing trends, and surveys to develop its prediction for the future. Nelson said California is a bellwether for the rest of the country. . . .

 Click on the title above to read the rest of this article. The New California Dream, Nelson, ULI, 2011 

Rigging reports for rail - even left-Dems see bias, Dulles Toll Road cited, TollRoadNews.com, December 12, 2011

The website TollRoadNews today published an article that provides an extensive overview of a May 2011 report criticizing bias in transit decisions toward selecting rail options.  The report focuses on the effectiveness of bus rapid transit (BRT) as an alternative.  The article begins:

A report published by the Institute for Transportation & Development Policy says that the transit selection process is rigged in favor of less effective rail projects and seriously biased against bus and road. The report "Recapturing Global Leadership in Bus Rapid Transit: a Survey of Select US Cities" is written by  Annie Weinstock, Michael Replogle and Ramon Cruz. Replogle at least has been a vocal opponent of roads and a transit supporter, employed for years at the Environmental Defense Fund. The ITDP report has an Introduction by longtime transit enthusiast Congressman Earl Blumenaeuer, Democrat-Oregon.

Dulles Toll corridor prime example of bias & mis-selection

Cited as an example of rigged selection against bus is the Dulles Toll Road corridor where an approximate $6 billion rail transit line is currently under construction, the financing being heavily based on toll revenue bonds.

The financing plan for Dulles Rail assumes huge surpluses that supposedly can be generated by the Toll Road with large increases in toll rates.

A problem is that estimates of the revenue yield of high toll rates are especially uncertain in the case of several fold increases in tolls. At some point a revenue maximizing toll rate is hit - beyond which further increases in toll rates actually reduce revenue. Where that revenue maximizing toll point is located is very difficult for modelers to estimate.

If less expensive bus rapid transit had been chosen instead of rail the huge risk being taken with toll revenues of the Dulles Toll Road would not have been necessary.

Read the full article by clicking here. 

The study itself is a huge PDF file--some 26MBs--and can not be stored on available document archive sites.  To access this full report, click on this link.