In Texas, a DOT report prepared by PricewaterhouseCoopers (PwC)144 generated charges of statistical manipulation from the Consulting Engineers Council of Texas (CECT). The CECT offered to sponsor the study, but the offer was refused. Upon release of the study, the CECT said of the PwC report: “This study has been from the first step less of an inclusive effort to review a controversial problem than a set up process aimed at a desired outcome.”145 Analysis of the PwC report by Tri Dimension Strategies substantiated CECT claims of statistical deficiencies. This revealed that PwC did not follow the comparison guidelines prescribed by OMB Circular A-76 and that PwC’s calculation of indirect costs associated with in-house design are skewed as a result.146 PwC attributed discrepancies to difficulties in obtaining the source data from the Texas DOT.
The PwC study’s deepest flaw was lumping together projects that are similar only at the grossest level. It made no effort to look at best value, to see if more-expensive projects had features that would justify higher
The Tri Dimension Strategies review highlights this point as well, pointing out that on average, for
larger and more-complex projects, using consultants is considerably more efficient than in-house design and
preliminary engineering. The analysis claims:
Throughout its report, PwC states that outsourced [engineering] projects cost more than projects conducted by in-house staff. There are two major methodology problems with this claim. First, PwC did not recognize the disproportionate number of projects classified as an in-house project. Because of the significant variation in size, PwC’s analysis left unexplained the large number of higher-cost PE projects done by in-house staff. Second, PwC’s report never addressed the issue of whether there is a correlation between the complexity of a design project and its design costs. PwC’s methodology completely ignored this strong possibility.148
Discrepancies over cost accounting also plague a study by the New York comptroller.149 An analysis done by PSMJ Resources casts doubt on the study’s claim that private-sector engineering firms have higher overhead costs.
The comparison is apparently based on different methods of calculating overhead. As a result of using different methods, the results are not comparable. It is impossible, with the limited data provided in the
report, to show that if the differences in methods were eliminated, in-house overhead rates may in fact be higher than private-sector overhead rates.150
PricewaterhouseCoopers, Highway Design Cost Comparison.
Consulting Engineers Council of Texas, “Perspective of the 1999 TxDOT/PricewaterhouseCoopers Cost Study,” CECT Newsletter (November 1999), p. 1.
Cynthia Thomas, Analysis of Highway Design Cost Comparison (Dallas: Tri Dimension Strategies, November 1999), p. 2.
The Texas DOT responded to this criticism: “The PwC study did not address many factors that could explain the cost differences found between the use of consultants and in-house staff to do engineering design. Those factors, such as quality considerations, review of consultant plans, timeliness, the varying complexity of projects designed, etc., are valid reasons to suggest that cost should not be the only factor used in determining whether to use consultants. However, the PwC study has never been purported to be anything more than a cost comparison study prepared by independent experts in the field of accounting.” The DOT went on to assert that PwC adhered to accepted accounting practices and questioned the competence of Tri Dimension Strategies to critique PwC’s methods. James Bass, director of finance, Texas Department of Transportation, correspondence with authors, July 2000.
Cynthia Thomas, Analysis of Highway Design Cost Comparison, p. 3.
New York Office of the State Comptroller, New York State Department of Engineering, Report 97-S-12 (Albany NY: State of New York, April 1998).
PSMJ Resources, Comments on Audit Report 97-S-12, State of New York, Office of the State Comptroller (Newton, Mass.: PSMJ Resources, 1997) p. 2.