Legal Aspects of Performance-Based Specifications for Highway Construction and Maintenance Contracts (2013)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4 ration between designers and contractors; 2) public– private partnerships (PPP) and contracts that include not only design and construction services, but also op- eration, maintenance, and financing; and (c) procure- ment practices that focus on best value and life-cycle costing considerations, instead of simply low price. Given the pragmatics of how these delivery and pro- curement systems work, owners find performance speci- fications to be a valuable tool for obtaining creative so- lutions to their stated goals and needs. As a result, the construction industry is seeing more and more real life examples of performance specifying, not only in the highway sector, but also in terms of “green” initiatives (e.g., energy consumption), creative equipment layout, and technology systems that guarantee certain levels of performance. From a liability perspective, performance specifica- tions also offer, at least theoretically, a major advan- tage to an owner. The longstanding legal principle is that the Spearin doctrine does not apply to performance specifications, and the contractor is the party who as- sumes the risk of designing and constructing to meet the performance specification. However, as with many construction law issues, one must look beyond general principles. As noted in the quotation above relative to the Defense Standardization Program, most owners share the perspective that “some detail requirements are necessary in a performance specification.” This cre- ates one of the most thorny issues in dealing with liabil- ity arising out of performance specifications—is the problem actually the result of a defective design specifi- cation embedded within a performance specification, and, if so, who bears the responsibility for that defect? The purpose of this digest is to examine the legal as- pects of performance specifying in the highway sector. The centerpiece of the digest is Section VIII, which fo- cuses on caselaw addressing performance specifications and how the above questions, as well as many others involving performance specifications, have been han- dled to date by courts. Sections II through IV are in- tended to provide context to these legal issues.4 Section II will explain the different types of specifications used in the highway sector, with a particular focus on the differences between “method specifications” and the umbrella of “performance specifications” commonly used (e.g., end-result, quality assurance (QA), and war- ranty specifications). Sections III and IV discuss the 4 Readers should note at the outset that the broad spectrum of issues involving the use and best practices associated with performance-based outcomes on highway and other transporta- tion projects are discussed in myriad industry publications. This is an area that has great interest domestically and inter- nationally, and its implications affect other important issues, particularly in relation to PPPs. While this digest discusses the current use of performance specifications, it is largely contex- tual—aimed at providing the reader background information that will help to orient the reader to the legal issues that are its focus. Readers who would like more information about per- formance contracting will find an abundance of information contained in the publications cited in the footnotes. “state of the practice” in the use and challenges of per- formance specifications in the highway sector. Sections V through VII address some quasi-legal issues relative to specifying performance, focusing on several of the commercial and contractual issues arising from their use, as well as on how performance specifications are used under alternative delivery systems. Finally, Sec- tion IX discusses how to best manage the legal issues associated with performance specifications on highway projects. II. COMPARISONS OF TYPES OF SPECIFICATIONS A substantial amount of technical literature de- scribes and compares various specifications used on highway projects. Depending upon the individual’s per- spective, an author of a particular paper may choose from more than a dozen different terms to describe the universe of specification types, including the following: • Materials and method specifications. • Method specifications. • Recipe specifications. • Design specifications. • Prescriptive specifications. • End result specifications. • QA specifications. • Statistically-based specifications. • Performance specifications. • PBS. • Performance-related specifications (PRS). • Proprietary or proprietary product specifications. • Warranty specifications. • Composite specifications. • Reference standards. Technical literature sometimes explains that certain terms—such as “method specifications,” “recipe specifi- cations,” and “prescriptive specifications”—are synony- mous.5 However, there are times when this is not done, particularly in the discussion of “performance specifica- tions” and related terms (e.g., “performance-based specifications,” “performance-related specifications,” and “quality assurance specifications”). This can lead to confusion over what a particular specification descrip- 5 See, e.g., Glossary of Highway Quality Assurance Terms, Transportation Research Circular E-C074, 3d Update, (Trans- portation Research Board, May 2005), http://onlinepubs.trb. org/onlinepubs/circulars/ec074.pdf; Development and Review of Specifications, Attachment 1, Federal Highway Administration Technical Advisory, Mar. 10, 2010 (referred to herein as “FHWA 2010 Technical Advisory”), http://www.fhwa.dot.gov/ construction/specrevattach1.cfm; Major Types of Transporta- tion Construction Specifications: A Guideline to Understanding Their Evolution and Application (AASHTO Highway Subcom- mittee on Construction, Quality Construction Task Force, Aug. 2003) (referred to herein as “AASHTO 2003 Guidelines”), http://www.fhwa.dot.gov/construction/specs.pdf; New or Emer- ging Technical Specification Definitions, National Highway Specifications Library, at http://fhwapap04.fhwa.dot.gov/ nhswp/browseEmergingSpecs.jsp.

5 tion means and the legal and contractual implications of using such a term. In contrast, when considering construction specifica- tions in the context of legal literature and caselaw, the terminology options narrow substantially. There are generally only three terms that are used. “Design” and “prescriptive” specifications are considered synony- mous, and are generally used to describe a rigid design or construction requirement established by the project owner. Design/prescriptive specifications are distin- guished from “performance specifications,” which is a term used in legal literature and caselaw to describe specifications that give the contractor some (and often complete) discretion in terms of how to design or con- struct a particular element of the work to meet a speci- fied need. The terms “design specification” and “performance specification” will be discussed throughout this digest. The purpose of this section of the digest is to provide a historical perspective of construction specifications in the highway sector, as well as an overview of the at- tributes of and nuances among the more commonly- used construction specification terms on highway pro- jects. A. The Evolution of Construction Specifications in the Highway Sector When road building was in its infancy in the early 1900s, the public was highly reliant on contractors to design, build, and maintain the road system. “Toll roads connected major cities and industrial areas,” and there were many small, unconnected, public road-building agencies scattered around the country. The agencies knew little about what it took to build a successful road. As a result, they generally required their contractors to provide guarantees that they would maintain and re- pair the roads “for a specified time period after con- struction.”6 Things changed as contractors complained about warranties and state governments became more in- formed about road design and construction. Central to this change was the formation of the American Associa- tion of State Highway Officials in 1914, which, among other things, accelerated the movement to uniformity in road specifications and maximum agency control over the design and construction process. All of this led to the development of what has become commonly referred to as “method specifications.” Method specifications, as discussed more fully below, give precise, detailed re- quirements that the contractor must follow, and leave the contractor with no discretion to make any changes. Method specifications have been commonly used since the 1940s.7 However, in the 1960s, industry lead- 6 See Federal Highway Administration, Performance Speci- fications Strategic Roadmap: A Vision for the Future, Spring 2004, updated Apr. 4, 2011; http://www.fhwa.dot.gov/ construction/pssr04tc.cfm. This is generally referred to herein as “Performance Specifications Strategic Roadmap.” 7 AASHTO 2003 Guidelines, supra note 5. ers began questioning whether they were the best way to deliver high quality roads. These questions were prompted by a variety of concerns, including 1) high variability in construction and materials on the com- pleted work; 2) the lack of effective agency sampling and testing programs to determine overall compliance with the specifications; and 3) the inhibition of innova- tion, as method specifications did not reward a contrac- tor for performing better than the minimum require- ment. By the 1970s, industry researchers began promoting the idea of creating new specifications that had an end result in mind—where the contractor would be told what the agency expected for a completed product and would have discretion in how to achieve those expecta- tions. The researchers thought this could lead to more innovation and a more accurate assessment of in-place quality, with the understanding that agencies would have to establish a more structured sampling and test- ing program on the in-place product. This concept led to a variety of questions on how it was to be implemented, including 1) what in-place properties most directly in- fluence product performance; 2) how should these prop- erties be tested; 3) what elements of the work should remain under the control of the agency; and 4) how should noncompliance be evaluated?8 The major output from this research was the crea- tion of a new type of specification that addressed a vari- ety of issues on a statistically-derived basis, such as testing and test variability, sample size, and pay fac- tors. This type was known as a “quality control specifi- cation.” While these specifications improved the meth- ods for assessing contractor compliance, they did not necessarily address product performance, as the drivers of product performance and the test procedures needed to measure the performance characteristics did not ex- ist. The connection to product performance was devel- oped in the early 1980s, when researchers began focus- ing on life-cycle cost analysis to correlate the relationship between the designed product and future preservation, maintenance, and repair of that product. This focus on life-cycle costing led to consideration and discussion around various types of “performance speci- fications” and attempts to determine analytically the performance aspect of the product.9 Since the 1980s, the industry has spent considerable time attempting to assess how to address performance specifications, and much has been written on the sub- ject.10 In other highway components, such as bridges, 8 See Performance Specifications Strategic Roadmap, supra note 6. 9 Id. 10 See Performance Contracting for Construction: A Guide to Using Performance Goals and Measures to Improve Project Delivery, published by the Federal Highway Administration in August 2012 and generally referred to in this digest as “FHWA’s Performance Contracting for Construction Guide” (available at http://www.fhwa.dot.gov/construction/contracts/ pcfc_2012/pcfc06.cfm#content). It is the first update to FHWA’s 2006 Performance Contracting Framework, fostered by High-

6 there are fewer examples of the use of performance specifications. One resource is a March 2011 report re- leased by the Federal Highway Administration (FHWA) Office of Innovative Program Delivery entitled, Key Per- formance Indicators in Public-Private Partnerships, A State-of-the-Practice Report.11 B. Construction Specifications Generally Specifications are contract documents that are used to communicate a project’s requirements and the crite- ria by which the owner will verify conformance with those requirements: They communicate to bidders prior to contract award, and to the selected contractor thereafter, the definitive direc- tions, procedures, and material and equipment require- ments the State DOT considers necessary for completing the contract work. As a result, they can directly affect the quality of design and construction of every highway prod- uct, as well as the cost of construction and maintenance.12 Specifications enable a project owner to have a stan- dard set of procedures for managing a project, including changes, and also the minimum standards against which to evaluate the contractor’s work, including al- lowable tolerances. In addition to telling a contractor what is expected of it, specifications serve the purpose of informing a contractor of quality and acceptability of work, allowable tolerances, and how payment will be handled.13 As noted earlier, there are a number of ways that specifications have been characterized. For purposes of this section, the specifications have generally been or- ganized into four categories: 1) method,14 2) perform- ance, 3) composite/mixed, and 4) proprietary.15 When ways for LIFE. See also Performance-Based Contracting for the Highway Construction Industry (Battelle, Feb. 2003) (gener- ally referred to herein as “Performance-Based Contracting for the Highway Construction Industry”), http://www.ncppp.org/resources/papers/battellereport.pdf; Per- formance Specifications Strategic Roadmap, supra note 6. 11 FHWA-PL-10-029, http://international.fhwa.dot.gov/pubs /pl10029/pl10029.pdf. This report provides a state-of-the- practice description of domestic and international practices for key performance indicators in PPPs. The report is based on a comprehensive literature review and eight case studies from Australia, British Columbia, the United Kingdom, and the United States. It identifies how government-developed per- formance measures reflecting societal goals such as congestion management or environmental impact are translated through key performance indicators and included in project documents for designing, constructing, operating, and maintaining trans- portation facilities. 12 FHWA 2010 Technical Advisory, supra note 5. 13 Id. 14 Note that in later sections, “method specifications” will be referred to as “design specifications”—consistent with the term used in legal literature. 15 FHWA also considers “reference standards” to be a type of specification. See FHWA 2010 Technical Advisory, supra note 5. These are standards developed by organizations such as AASHTO, ANSI, ASTM, and ACI that provide national stan- dards of performance or measurement. Because of their limited the final product is described in terms of component materials, dimensions, tolerances, weights, and re- quired construction methodology (e.g., equipment type, size, speed), the specifications are commonly known in the highway sector as “method specifications.” When the contractor is given some discretion as to how to per- form the work, the specification structure is commonly described as a “performance specification.” C. Method Specifications Method specifications are also known as “material and method,” “recipe,” “design,” and “prescriptive” specifications. They explicitly identify the materials and work methods or procedures a contractor should use to complete the work included in the contract, placing maximum control in the hands of the specifying agency. The American Association of State Highway and Trans- portation Officials (AASHTO) has characterized method specifications as providing the contractor with a “cook- book” with specific “recipes” to follow.16 “In effect, the Agency rents the Contractor’s personnel and equip- ment.”17 Method specifications are the predominate type of specification used in the United States highway in- dustry.18 As an example, a typical compaction method specifi- cation would detail items such as the 1) moisture con- tent of soils or granular material; 2) minimum rolling temperatures for bituminous materials; 3) maximum thickness of each layer to be compacted; 4) type of roller (e.g., smooth, sheep’s foot, vibratory); 5) weight of the roller; 6) minimum number of passes of the roller; and 7) minimum ambient/surface temperature for bitumi- nous materials. If the contractor followed the “recipe,” then the agency would presumably have an acceptable product. There are several reasons why method specifications have been used so extensively in the highway industry. As noted by FHWA, “Method specifications typically operate on the principle that if the specified materials and methods worked in the past, then the end product is likely to perform well in service so long as the con- tractor strictly adheres to the prescribed require- ments.”19 Consequently, because the details of a method specification are typically based on methods that his- torically provided satisfactory results, they eliminate risk associated with newer, less proven methods as well as the risk associated with varying contractor perform- ance. They are also familiar, straightforward to write, and can be implemented with minimal agency involve- ment. This is particularly helpful to those agencies that application to this digest, they will not be discussed any fur- ther than this footnote. 16 AASHTO 2003 Guidelines, supra note 5, at 2. 17 Id. at 3. 18 PERFORMANCE SPECIFICATIONS STRATEGIC ROADMAP, supra note 6. 19 FHWA 2010 Technical Advisory, supra note 5.

7 lack the expertise and resources required to use per- formance specifications.20 Method specifications have several key disadvan- tages. The contractor has little, if any, opportunity to deviate from the specifications. It has no responsibility for performance deficiencies in the end product and no incentive to innovate or use better, more efficient con- struction methods. The contractor’s only responsibility is to follow the specifications. Another major disadvantage of method specifications is the associated inability of an agency to determine the actual quality of the contractor’s work. Method specifi- cations typically base acceptance on the “reasonable conformance” or “substantial compliance” of the work with the specification requirements, based on the agency’s inspection of the work. Individual or represen- tative field sample results may not recognize the inher- ent variability in construction materials—particularly when the sample is intended to address quantities of materials. Consequently, this can create some prob- lems, where one field sample could be truly representa- tive of the in-place conditions and another sample might be an aberration of the actual in-place conditions. This can lead to conflicts between the contractor and agency if work is rejected. Most method specifications do not indicate whether or how contract prices should be adjusted for noncon- forming work. In fact, a contractor’s payment is not linked to product quality or long-term performance—it is simply tied to a demonstration that the contractor met the specifications it was furnished. As a result, con- tractors usually receive 100 percent payment for the work completed, regardless of the level of quality.21 The use of method (i.e., design and prescriptive) specifications is certainly not unique to the highway industry. However, it should be noted that other con- struction sectors seem to afford construction contractors far more latitude in terms of means and methods of construction. This is one reason that other construction sectors tend to use the terms “design” and “prescriptive” specifications, rather than “method” specifications, as their nomenclature for this category of specification. For example, while owners in the water and wastewater sector have historically been quite prescriptive in terms of specifying the type and layout of equipment, they largely leave the means and methods for installation of process piping to the contractor. Similarly, building contractors have substantial discretion over most of the means and methods of the construction process, such as support of excavation and shoring of concrete slabs, even though they are bound to comply with design specifications relating to these activities. D. Performance Specifications The term “performance specification” is an “um- brella” concept that incorporates end result specifica- 20 AASHTO 2003 Guidelines, supra note 5. 21 AASHTO 2003 Guidelines, supra note 5; FHWA 2010 Technical Advisory, supra note 5. tions, PRS, PBS, QA specifications, and warranty speci- fications. A performance specification describes the de- sired final product in terms of operational characteris- tics or ultimate use and gives the contractor significant latitude in how it achieves the final product. As an ex- ample, with respect to concrete pavement, one might see a performance specification tied to strength (e.g., 28-day compressive strength), slab thickness, and smoothness levels (e.g., mean profile index). For soil used as fill in an embankment, one might see a compac- tion performance specification tied to maximum air voids content, without dictating the type of roller or number of roller passes. Note that results may also be expressed in terms of time, safety, work zone manage- ment, quality, and cost. With all performance specifica- tions, it is critical to identify criteria for verifying com- pliance and the consequences if the contractor fails to comply. It is beyond the scope of this digest to thoroughly re- view the nuances, policies, and “how-to’s” associated with the terms falling under the performance specifica- tion umbrella.22 However, it is important to have a general understanding of what distinguishes one from another: • End-Result Specifications. These specifications re- quire the contractor to take the entire responsibility for producing and placing a product. The agency’s respon- sibility is to either accept or reject the final product or to apply a price adjustment commensurate with the degree of noncompliance with the specifications. Agency acceptance is based on sampling and testing of the final in-place product. Because they make the contractor completely responsible for supplying a product or an item, these specifications offer the contractor the great- est degree of flexibility in exercising options for develop- ing new techniques and procedures to perform the work and to improve the quality of the end product. While there may be some prescriptive elements to the specifi- cation, they are generally minimized.23 • Quality Assurance Specifications. QA specifications require contractor quality control and agency accep- tance activities throughout the production and place- ment of a product. Final acceptance of the product is usually based on a statistical sampling of the measured quality level for key quality characteristics. Stated dif- ferently, QA specifications establish acceptance by iden- tifying, for each quality characteristic, the percentage of measured materials within a lot that must be within specified limits. This has the benefit of providing an objective process for assessing acceptance. The critical 22 For further information on these terms, see generally Per- formance Specifications Strategic Roadmap, supra note 6; FHWA 2010 Technical Advisory, supra note 5; AASHTO 2003 Guidelines, supra note 5; Performance-Based Contracting for the Highway Construction Industry, supra note 10; FHWA’s Performance Contracting for Construction Guide, supra note 10. 23 Id. at 6.

8 aspect of developing a QA specification is to identify the material attributes that are essential to good perform- ance and the associated limits within which the work can be produced to suggest good performance over the design life of the product. For asphalt pavements, qual- ity characteristics might include asphalt content, den- sity of the compacted pavement, and pavement smooth- ness.24 QA specifications also link the measured quality levels to payment. For example, the contractor might receive an increased payment (e.g., 1 to 105 percent) for superior quality work, and a reduced payment (0 to 99 percent) for lesser quality work. Work below a mini- mum quality level may be subject to removal and re- placement or another corrective action.25 • Performance-Related Specifications. PRS are QA specifications that use quality characteristics and life- cycle cost relationships that are correlated to product performance through mathematical models. The quality characteristics are amenable to acceptance testing dur- ing construction. For asphalt pavements, these charac- teristics might include total in-place air voids or ride smoothness; for concrete pavements, they might include concrete permeability and strength. The models at- tempt to predict when and to what extent a construc- tion product (e.g., pavement) will exhibit a given type of distress (e.g., fatigue cracking or joint spalling), as well as the post-construction life-cycle cost for maintenance and rehabilitation. PRS may use empirical data, engi- neering judgment, and life cycle costing as a basis for acceptance and pay adjustments.26 • Performance-Based Specifications. PBS are con- cerned with the performance of the final in-place prod- uct, not how it was built.27 These are QA specifications that describe the desired levels of fundamental engi- neering properties (e.g., resilient modulus, creep prop- erties, and fatigue properties) that are predictors of performance and appear in primary prediction relation- ships (i.e., models that can be used to predict pavement stress, distress, or performance from combinations of predictors that represent traffic, environmental, road- bed, and structural conditions). They differ from PRS in that they specify the desired levels of fundamental en- gineering properties rather than key quality character- istics.28 • Warranty Specifications. These are performance specifications where the condition of the product is measured after some predetermined time. FHWA has defined them as “a guarantee of the integrity of a prod- 24 FHWA 2010 Technical Advisory, supra note 5. 25 AASHTO 2003 Guidelines, supra note 5. 26 FHWA 2010 Technical Advisory, supra note 5. As of the date of the Advisory, the sole PRS has been piloted on concrete pavements, with a notation that research is being performed on asphalt. 27 AASHTO 2003 Guidelines, supra note 5. 28 Id. As of the date of the Advisory, complete PBS had not been applied in highway construction, “primarily because the most fundamental engineering properties are only now becom- ing amenable to timely acceptance testing.” uct and of responsibility for the repair or replacement of defects by the contractor.”29 For example, the agency could use a warranty specification to specify pavement performance (e.g., rut depths, transverse or longitudi- nal cracking) or metrics such as the Pavement Condi- tion Index, pavement smoothness, and surface fric- tion.30 The agency could require the contractor to warrant this performance over a period of time (e.g., 2 to 10 years). The agency monitors pavement perform- ance during the warranty period, and any defects at- tributable to construction are to be repaired by the con- tractor. This can be an advancement to a typical end- result specification, in that it can specify actual per- formance of the pavement rather than quality charac- teristics that are indicative of performance. Warranty specifications do have some major challenges, however, relative to the financial burdens that accompany partial or complete product failures. Each of the above-referenced performance specifica- tion terms gives some discretion to the contractor over how to perform the required work. As is evident from the above descriptions, the primary differences among them lie in how and when contract compliance is meas- ured, as well as the consequences for failing to meet the contract’s requirements. To accomplish the goals behind performance specify- ing, a well-drafted performance specification needs to consider a number of items. While these items are ad- dressed further in Section IV below, they include 1) an identification of the agency’s needs or goals; 2) perform- ance parameters that can be measured or tested to en- sure that the goals are satisfied (e.g., pavement smoothness); 3) measurement or testing techniques (e.g., using a high-speed profilograph to measure pave- ment smoothness); 4) performance values for each per- formance parameter (e.g., maximum International Roughness Index in inches per mile); 5) inspection and verification testing regimens for both the contractor and the agency; and 6) price adjustments to reflect the range of acceptable work.31 E. Composite/Mixed Specifications As will be discussed in detail in the caselaw review in Section VIII, it is rare that a performance specifica- tion will not contain some prescriptive requirements. For example, if a bridge project involved the driving of concrete cylinder piles, a performance specification might say, “Drive the 50-foot diameter piles to a mini- mum tip elevation of -55 feet and to a bearing capacity of 650 tons.” However, the agency might choose to im- pose some specific restraints on the contractor’s flexibil- ity in achieving these results. Consequently, it may add 29 National Highway Specifications Library, supra note 5. 30 Design-Build Contracts: Performance Specifications, FHWA, Highways for Life, http://www.fhwa.dot.gov/hfl/ innovations/designbuildperformspecs.cfm. 31 See generally FHWA 2010 Technical Advisory, supra note 5.