Fuel Usage Factors in Highway and Bridge Construction (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.

11 Chapter 1 of this report details the background information, purpose, and the existing literature relevant to NCHRP Project 10-81. The introduction section provides a brief overview of fuel usage factors and the research problem that this report addresses. The purpose section outlines the stated goals and objectives of the project. The literature review presents and describes sources germane to the project, including government sources, state DOT literature, academic studies, and media reports. 1.1 Introduction Price adjustments of selected commodities in highway construction are used in construction contracting as a way of limiting risks to the contractor arising from price fluctuations of these commodities over the life of a contract. The benefits to contracting agencies are bids that better reflect actual construction costs, without added costs for risk of increased commodity cost. Fuel is one commodity for which price adjustments are allowed. Fuel usage factors are commonly applied by state and local agencies in calculating fuel cost price adjustments in a contract specification that permits cost escalation and de-escalation. The original fuel usage factors were published in Highway Research Circular Number 158 by the Highway Research Board (HRB, now the Transportation Research Board) in July 1974. These factors, which were later incorporated in the 1980 Federal Highway Administration (FHWA) Technical Advisory T5080.3, have remained unchanged for over three decades despite the continuous effects of price inflation and changes in construction-dollar purchasing power, construction methods, industry processes, equipment efficiency, and fuel type. HRB Circular 158 established gasoline and diesel fuel usage factors for excavation, aggregate, hot mix asphalt production and hauling, and Portland cement concrete production and hauling. Additionally, HRB Circular 158 established fuel usage factors for structures and miscellaneous construction in gallons per $1,000 of construction cost, with no provision for any adjustment for inflation. 1.2 Purpose The objectives of this research were to 1. Identify present highway construction contract activities that are major consumers of fuel; 2. Prepare fuel usage factors for these activities, including those items of work presented in Attachment 1 of FHWA Technical Advisory T5080.3, for base year 2012; and 3. Develop a recommended practice for state DOTs to implement use of fuel adjustment factors and adjust them for both state-specific conditions and changes in construction costs, methods, and equipment. C h a p t e r 1 Background

12 Fuel Usage Factors in highway and Bridge Construction 1.3 Literature Review This literature review presents the most relevant and helpful sources available. Its purpose is to provide necessary background information while framing the research objectives: factoring inflation and 30 years of technological advancement into a dated set of fuel usage factors, reducing risk for both agencies and contractors, and ensuring the availability of an improved system to state DOTs. The following subsections of this chapter coincide with the major categories of sources utilized during the research process: the original federal guidelines, more recent FHWA/AASHTO surveys outlining state DOT practices, academic research conducted for state DOTs and other entities, information provided in media reports, and citations from inter- national sources. 1.3.1 Original Federal Guidelines The original research on fuel usage factors includes Highway Research Circular Number 158 by the Highway Research Board (now the Transportation Research Board) in July 1974. A mailed survey of 3,000 highway contractors netted 400 responses, and the FHWA compiled and analyzed the data. Factors were computed for construction activities such as excavation, aggregate and asphalt production, and structure construction. Each of these activities received a high, low, and average factor. Both diesel and gasoline were included. This early research did not fully investigate the effects of different terrain and did not account for contingencies such as high altitude. The FHWA incorporated these factors in Technical Advisory T5080.3, originally released in 1980. The advisory contains methods for developing price adjustment provisions such as downward and upward contract provisions, using an average of quotes to avoid manipulation, triggers based on a 5 percent change in fuel price indices, and ad hoc adjustments on fuel usage factors in cases of extreme elevation, rough terrain, etc. It also provides the original fuel usage factors, which are reproduced in Exhibit 1-1, as well as additional fuel usage factors developed by the states. Prior to this study, these factors had not been revisited on the federal level since the issuance of Technical Advisory T5080.3. Since the original survey was conducted, the costs of fuel and structure construction have changed and may be outdated due to changes in technology, work practices, material haul distances and other factors. In addition, the original survey established gallons per $1,000 as the unit of measure for work on structures, yet no adjustments for inflation have been conducted. 1.3.2 FHWA and AASHTO Surveys Another important set of literature pertains to the current practices of the states. The Contract Administration Section of AASHTO’s Highway Subcommittee on Construction (AASHTO SOC) maintains a spreadsheet that summarizes the current use of price adjustment clauses for fuel, asphalt, cement, steel, and other highway materials. The 2009 version of a summary spreadsheet includes general information regarding trigger values, indices, Web references, general comments, and state DOT contacts. This set of literature also includes the individual state policies for which the spreadsheet provides Web references. The most recent survey shows that 41 states utilize price adjustments on fuel, while 40 states adjust asphalt pricing. Three states (Arkansas, Michigan, and Texas) adjust neither fuel nor asphalt pricing. Related to these first two items is state DOT research on fuel usage factors. The FHWA advisory provides some of this data, but does not identify sources. In the 2008 AASHTO SOC Survey, New Jersey reported that they were “currently working with industry to review fuel usage factors.” The larger issue of rising costs and uncertainty has been addressed by many state DOTs. The

Background 13 average price per ton of asphalt in Florida, for example, increased from $34.66 in 1990 to $97.04 in 2008. Such increases reinforce concerns with inflationary risks and doubts over the efficacy of current escalator clauses (Prasad 2010). The FHWA’s 2007 report “Growth in Highway Con- struction and Maintenance Costs” points to a nationwide issue, with 42 states reporting large increases in construction costs. The study identifies rising costs of each major commodity input group as the primary cause. The study notes that other potential causes, such as employee wages, insurance and engineering costs, and profit margins, experienced gradual and/or limited growth (Federal Highway Administration 2007). Other FHWA studies, such as the 2006 “Survey on Construction Cost Increases and Competition,” show large majorities of states facing increased bid costs due to rising fuel/asphalt prices (AASHTO 2006). The sponsors of this study believed that providing an updated set of fuel usage factors could alleviate some of these concerns. Another interesting data point is the FHWA Highway Statistics series that, through 2005, provided Table PT-4, “Usage Factors for Major Highway Construction Materials and Labor.” This table provides weighted averages for all federal-aid highway construction contracts over $1 million on the national highway system reported as completed during calendar years 2002, 2003, and 2004. The estimate for petroleum products, defined as fuel and lubricants for equipment and trucks, is 12,279 gallons per million dollars of construction cost, down from 19,909 gallons 3 years earlier. Over the same time period the usage of bituminous material declined only slightly, going from 344 to 329 tons per $1 million. FHWA obtained the data in this table from Form FHWA-47, which FHWA used to develop the FHWA Highway Construction Cost Index. FHWA discontinued this form and the collection of this data after 2004. FHWA has developed a new National Highway Construction Cost Index using data from Oman Systems’ BidTabs data. Exhibit 1-1. Fuel usage factors, Highway Research Circular 158/FHWA Technical Advisory T5080.3. *Estimated due to insufficient data. Low Avg. High Low Avg. High Excava�on Earth 0.27 0.29 0.30 0.11 0.15 0.21 Rock 0.37 0.39 0.42 0.17 0.18 0.22 Other 0.33 0.35 0.38 0.15 0.16 0.18 Aggregates Onsite Produc�on 0.25 0.28 0.36 0.08 0.09 0.11 Aggregate Base 0-10 Mi. Haul 0.24 0.27 0.33 0.22 0.24 0.28 10-20 Mi. Haul 0.35 0.42 0.54 0.27 0.39 0.49 Asphalt Concrete Produc�on 1.75 2.43 3.50 0.07 0.14 0.18 Hauling 0-10 Mi. Haul 0.28 0.33 0.34 0.35 0.34 0.53 10-20 Mi. Haul 0.30 0.49 0.56 0.35 0.58 0.89 Placement 0.06 0.14 0.20 0.08 0.14 0.22 Portland Cement Produc�on 0.15 0.28 0.45 0.12 0.15 0.21 Hauling 0.33 0.48 0.67 0.52* Placement 0.13 0.22 0.31 0.08 0.14 0.22 Structures Gallons/$1,000 10.00 19.00 25.00 10.00 22.00 35.00 Miscellaneous Gallons/$1,000 10.00 19.00 30.00 10.00 19.00 30.00 Gallons/Ton Gallons/C.Y. Item of Work Units Diesel Gasoline Gallons/C.Y. Gallons/Ton

14 Fuel Usage Factors in highway and Bridge Construction 1.3.3 Academic Research Another set of literature consists of university-based research studies conducted for state DOTs on various aspects of price adjustment clauses. One such study is the June 2007 “Best Practice for Developing the Engineer’s Estimate,” a SCDOT research project with FHWA fund- ing, written by Karl Niedzwecki, Greaton Sellers, and Lansford Bell of Clemson University. The study is concerned mostly with comparing two methods of project cost estimation: the unit cost line item approach and the cost-based approach. The authors reached the conclusion that cost-based estimation requires impractical investments in time and expertise and cannot be broadly adapted (Niedzwecki and Bell 2007). This conclusion stands in contrast to the opinions of George Bradfield, chief estimator of the Georgia DOT, who criticizes the inclusion of low bids in the historical data and asserts that a cost-based estimate is more accurate and ultimately more cost effective when applied to the project at hand (Bradfield). Also of note in this study is a figure (see Exhibit 1-2) that presents the results of a survey question concerning the data source fuel cost adjustments. Most states are currently using factors they developed over older FHWA estimates. The Associated General Contractors of America (AGC) is currently absent from these processes and the states acquire their fuel cost adjustments through other entities. The second volume of the report uses the SCDOT bid data to conduct statistical analysis of the influence of fuel price fluctuations on bid prices. The authors note, “the SCDOT Research Steering Committee identified a total of 44 different pay items, also referred to as Unit Cost Line Items, which were believed to be impacted by fuel and asphalt price” (Sellers and Bell 2007). The authors conclude in the first volume that the unit cost line item is preferable to the cost-based method. Two of the most substantial limitations relate to the long-term nature of the unit cost line item: prices can be affected by past unbalanced bids, while database prices for items could have been affected by now-irrelevant economic conditions. With the prices of 44 items being affected by fuel cost, which historically fluctuates quite rapidly, state DOTs that use the unit cost line item (currently 30 out of 50) seem to be at a disadvantage. It is later demonstrated that the engineer’s estimate can often take up to a year to adjust to fluctuations in the low bid. Sellers and Bell believe this effect is caused by the unit cost line item methodology, as its historically averaged price indexes would be unresponsive to rapid changes. They conclude, “Many of the unit cost line items examined in this research have bid prices correlated with either the fuel price index or bidding volume. Many of the items tended to rise or fall with the cost of fuel as price trended up or down.” The use of statistical analysis of bid data on pay items with fuel price indexes is significant, because this is one of the methodological approaches examined in this study. Another university study is the “Evaluation of Fuel Usage Factors in Highway Construction in Oregon” by Ken Casavant, Professor at Washington State University, with co-authors Eric Jessup and Mark Holmgren. This study confronts many of the same research problems addressed in this study. This analysis compiles information regarding how other states address the issue of inflation in fuel factors and develops an approach to updating the estimation of fuel factors used for various types of structures. The authors present three major errors with the current fuel adjustment system. The first is the effects of inflation on construction costs exacerbated by the Exhibit 1-2. How states obtain fuel cost adjustments. Source Bid History Cost-based/Combo Through AGC contacts or resources 0% 0% Use of quoted FHWA adjustment factors 23% 0% Use of US DOT resources 0% 13% Use of state DOT factors developed through self-determined inves�ga�on 38% 13% Other 38% 38%

Background 15 failure to correct for the effects of inflation on the 1980 fuel adjustment factors for structures and miscellaneous costs. The second is improvements in construction practices and fuel efficiency. Lastly, fuel preferences have shifted, with the change from diesel to natural gas in asphalt plants being the most notable (Holmgren et al. 2010). The study proceeds with an overview of state practices for formulating fuel adjustments and a survey of state DOTs, which found that most consider their current fuel adjustments to be fair despite contractor complaints and recently implemented or planned changes in many of their fuel adjustments. Two primary recommen- dations are presented. The first is to cut the fuel usage factors for structures approximately in half—from 19 to 9 for cast-in-place and from 10 to 5 for pre-cast. A review and recalculation of fuel usage factors every 3 years is also suggested. A third study in this group is a 2009 paper, “Materials Risk Management—Beyond Escalation Clauses and Price Indexing,” by Larry Redd of a private firm and Tim Hibbard, Assistant Chief Engineer, Operations, Wyoming Department of Transportation (WYDOT). The paper discussed the WYDOT study, “Asphalt Risk Management at WYDOT.” That study examined outcomes after 3 years of an escalation “option” for contractors. The WYDOT escalation clause stipulates that contractors must opt in within 10 days of the pre-construction conference. Triggering of the escalation clause occurs after a 10 percent change in the base price of the commodity. The contractor will then be reimbursed for 90 percent of upward index movement, whereas they would pay back 90 percent of downward movement to WYDOT. The study encompassed a 3-year period from 2006–2008 with careful attention paid to summer 2008, when a pronounced spike in oil prices resulted in remarkable volatility in asphalt price and availability (the price of asphalt ballooned from approximately $350 per ton to $700 per ton during the construction season). Although the agency disbursed over $7 million in repayments during the summer of 2008, assum- ing a substantial amount of contractor risk in the process, WYDOT expressed its satisfaction with its current mechanisms for price adjustment (Redd and Hibbard 2009). Contractors did not achieve full protection from the volatility of the market. Some contractors ended up paying more to their suppliers than the adjusted price due to short-term increases in the supplier’s pricing. This difference was not covered by WYDOT because the suppliers’ prices were already over the index amount. One contractor received a higher-than-expected bill from his supplier, who was calculat- ing cost based on the previous month’s index even though commodity prices had begun to decline. Additionally, WYDOT does not cover adjustments when a contractor has a fixed price agreement with a supplier. In this case, contractors have to hope that suppliers will honor the agreed upon price. In addition, concerns arose that price indexing may have long-term adverse consequences in the asphalt market regarding price competition. “Fuel Price Adjustment Techniques: A Review of Industry Practice,” prepared by Rutgers University at the behest of the Monmouth County (NJ) Department of Human Services, provides a useful overview of the different types of fuel adjustments as well as their varying implications (Rutgers University 2004). The main methods presented are contract pricing, fixed price with adjustment, direct refueling using agency-operating fueling facilities, and floating price-direct cost reimbursement. The study also offers several observations about the use of fuel price provisions in bidding and construction. For example, fluctuation in fuel price creates risk that is detrimental to all parties involved, and expecting the contractor to bear all the risk will often portend inflated costs as a means to reduce liability. The authors also have a favorable opinion of escalator triggers and re-adjustment as well. Prepared for AASHTO by researchers at Arizona State, “Project Cost Estimating: A Synthesis of Highway Practice” is a broad survey of current cost estimation practices by the states. Of particular concern to the authors is the tendency for the actual costs of large transportation projects to exceed cost estimations during planning and even the beginning of construction. The frequency and magnitude of estimation errors remains analogous to projects from 70 years ago despite

16 Fuel Usage Factors in highway and Bridge Construction ostensible improvements in estimation methodology (Schexnayder et al. 2003). The authors set forth several recommendations, such as the inclusion of contingency budgeting and annual adjustments on inflation so costs would be in current-year amounts. An update system of factors could improve estimation accuracy and make fuel costs more predictable. The final university study is the Georgia Tech Research Institute’s “A Study of Liquid Asphalt Price Indices Applications to Georgia Pavement Contracting.” This study included a survey of state DOTs and conducted statistical analysis of price trends using price indices. The authors weighed the costs and benefits of implementing an asphalt index pricing system in Georgia. Expected costs included: high start-up costs, increased costs and labor to generate the index, the manpower required to calculate adjustments in the field, a higher price paid for asphalt when the market price increases over the period of the contract, possible price manipulation by suppliers, and possibly reduced contractor payments in the event of a decline in liquid asphalt price. Anticipated benefits included: a lower price for asphalt when market prices decline, more rapid completion of contracts, no assignment of risk for contractors at the time of the bid, more competition in the market from price risk reduction, and regional uniformity for Georgia and its neighbors (Eckert and Eger III 2005). Finding that the Georgia DOT had a lower quoted price on liquid asphalt than any of its neighbors, each of which employs a price index, the authors recommended that the Georgia DOT retain its existing protocols. 1.3.4 Media Reports Another set of literature is comprised of news articles. Media items can provide useful opinions and identify contacts and data sources. For example, the Albany (New York) Business Journal ran an article in 2008, entitled “Asphalt Costs, Tied to Climbing Oil Prices, put the Squeeze on Paving Contractors.” Written at the height of the surge in fuel costs during the summer of 2008, this article examines the effects of rising asphalt prices on contractors. The New York State index price for liquid asphalt rose from $335 in July 2007 to $588 in July 2008. Subsequently, the price of a ton of blacktop jumped from $65 in April to $73 in June (Business Review [Albany] 2008). As a result, contractors had to weather both increased costs and falling demand as property owners increasingly canceled or postponed jobs. The article included an interview with a state DOT official. More recent media coverage further illustrates the volatility inherent in construction con- tracting. The Illinois bituminous index rose 35 percent between October 2009 and March 2010. Upward movement on most materials is probable for the short term (Associated General Con- tractors of America 2010). One effect of the ongoing recession has been plummeting demand for construction contracts, especially in the private sector. Contractors are increasingly submit- ting bids that are lower than normal for public projects in an effort to secure work. A new toll plaza on the Florida Turnpike, originally estimated to cost $37,000,000, received a low bid of $17,000,000. Broward County engineer Richard Tornese comments that bids for projects in 2008 and 2009 were 10 percent below budget estimates (Streeter 2010). The competition is so intense in Louisiana that many contractors have resorted to examining winning bids for errors in an effort to reopen bidding on public contracts. The decline of private construction is one underlying explanation for this glut of contractors. Although several recent hurricanes have necessitated a large number of public projects in Louisiana, future construction will slow as the state’s large debt begins to limit the amount of money it can borrow for construction (Roberts 2010). 1.3.5 International Sources One of the aims of this study was to examine whether the international community had conducted any research on fuel factors or fuel usage. However, little relevant literature was found,

Background 17 despite numerous contacts with entities such as the United Kingdom’s Highways Agency. Perhaps the best source currently available is the “International Construction Cost Survey 2009,” published by Turner & Townsend. This report has limitations, such as not including international pricing for fuel and petroleum-based commodities like asphalt, and is best viewed as a survey of general international economic trends. The near disappearance of inter-bank lending and granting of loans forced the postponement or cancellation of many planned projects. Many of the hardest hit countries are in Europe: construction costs in Scotland were predicted to fall by 8 percent in 2009, and the construction sector’s contribution to the Irish economy fell from 14–16 percent to 5–6 percent in 2009 (Emmett 2010). The most common attempt to rectify the crisis has been to inject massive amounts of public funds into the construction industry with the intent of creating jobs and improving infrastructure. An article from Uruguay details the rise of construction costs in both Uruguay and Argentina. In 2009, overall construction costs rose by 10.8 percent, mainly due to increases in the prices of materials. Inflation on materials and increased labor costs are expected to further raise costs (Sainz 2010). There is no mention of the contribution of bridges and highways to that figure.