February 9, Matthew Knight Arena - Project Audit Oregon University System

Transcription

1 February 9, 2011 Matthew Knight Arena - Project Audit

2 Contents 1. Executive Summary Project Overview Contract Assessment and Audit Hoffman Subcontractor Selection and Subcontracting Practices University Outreach Measures Improvements and Best Practices Appendix A: Audit Scope of Work 58 Appendix B: University Payments Reviewed 60 Appendix C: Alumni Center ASRs 61 Appendix D: Arena Fund Allocation Calculation 62 MRC

3 1 Executive Summary Introduction Marsh Risk Consulting (MRC) was engaged by the (OUS) to provide construction audit and control advisory services related to the construction of the Matthew Knight Arena (Arena or Project) at the University of Oregon (UO or the University). The audit was performed on behalf of the Oregon State Board of Higher Education and coordinated by the Internal Audit Division of the OUS with oversight provided by the Oregon Secretary of State Audits Division. OUS developed a comprehensive audit scope for the project. The full scope is provided as Appendix A of this report. Generally, the scope included an audit of the costs incurred by the University to design, develop, manage and construct the Arena. Unlike an audit of just project costs, OUS also required an in-depth review of the performance of project stakeholders (owner, owner s representative, designer, and contractors). This performance audit was conducted through a comprehensive assessment of the numerous business practices employed by several entities. The basis of evaluation included an assessment of compliance with contract terms and governing statutes or relevant laws. OUS also required the audit to identify where industry best practices were employed, or could be implemented for future projects to improve results on public projects. The audit period mandated by OUS began when the Project was approved for public funding. MRC began field work in August 2010 and concluded in early November Project expenditures were evaluated based on cost incurred and committed to date at the time of field work testing. The audit competencies required by OUS and employed by MRC to perform the assessment included: forensic accounting and fraud examination expertise; construction management experience (including project management, scheduling, cost estimating and budgeting, change management, quality control, documentation, communications, etc); fluency with contract documents and project delivery methods; as well as proficiency working with educational and public entities engaged in large capital projects. MRC 1

4 Methodology In order to perform the audit of the Arena Project, several factors were considered in planning the review, including project level documentation and records, available accounting reports, report summaries, interviews with key personnel, and other information available as audit evidence. In addition, MRC developed an understanding of the Project and the roles of those involved in the project to plan MRC s testing. MRC also conducted detailed interviews with personnel from the Development Manager (JMI, Development Manager, or Owner s Representative), the Contractor (Hoffman), the Architect (Ellerbe), the University, the Developer (National Championship Properties or NCP), the University of Oregon Foundation (Foundation), and the Department of Justice for the State of Oregon to gain an understanding of the project scope, change order process, subcontracting procedures, reporting procedures, the invoice preparation, billing, review, approval and payment processes. A detailed discussion of audit methodology is presented in the body of this report. Summary of Findings In general, MRC found that the Matthew Knight Arena Project has been managed in accordance with the terms of its agreements. As a cost and performance audit, on a scale from Low to High risk, where the risks include material cost exceptions, significant control impairments, or substantial performance deficiencies relative to contract obligations, MRC concluded that the Arena Project has been a Low Risk endeavor for the University as of the close of MRC s field work. At the close of audit field work, MRC had identified no significant quality issues related to the Project and found that the Project was both on-time and on-budget. In fact, the Project had experienced approximately $5.4 million in savings against the contract price. The University also negotiated contract provisions with Hoffman Construction that were favorable to the University with respect to savings; all savings revert back to the Owner (vs. contracts that include sharing provisions that split the savings between the owner and general contractor). MRC found that the success measurements of on-time and on-budget were attributable to those responsible for the development, construction and oversight of the project. MRC identified no material cost exceptions. MRC also found several examples where industry best practices were employed by various stakeholders. There were also several challenges overcome by the University and its project partners during development, as the Project changed from a private to publicly funded job. One of the most significant challenges was the contract reviews and modifications required prior to execution as a public project. While MRC did identify some areas for improvement, MRC concluded that the highly important goals of meeting a budget and schedule for a project with the complexities of the Arena was achieved as a result of effective planning, execution and MRC 2

5 communication among the University, Hoffman, JMI, Ellerbe and other divisions of the State. Some inconsistencies and questionable costs were identified through the course of the audit. MRC generally found that there was no significant financial impact from these findings although certain financial and management control weaknesses were noted. MRC s recommendations for improvements in noted areas are discussed in the body of this report. The Arena is not representative of a typical University capital project; its profile changed from a privately developed project to a public project funded by a bond issue. As such, the University negotiated numerous obstacles to finalize the Project s conversion. This renegotiation led to some gaps and ambiguities in contract terms, which led to a few findings of immaterial financial consequence. MRC investigated all areas set forth by OUS in the audit scope and has grouped the results of the audit into two categories: Findings and Conclusions. The report is organized by the company or entity subject to the audit, where Findings are presented first. MRC defines these groupings as follows: Findings: areas where MRC noted exceptions between contract terms and the execution, as well as areas where MRC recommends a review and improvement process be adopted by the University. In the event dollar values were associated with these Findings, and sufficient information was available, these amounts are presented. Conclusions: include the additional areas mandated in the audit scope, where both best practices and areas for improvement were noted by MRC. With respect to areas for improvement, MRC classified these as less significant than a Finding. Any recommendations MRC developed regarding Findings and Conclusions are presented with each grouping. MRC has communicated its Findings and Conclusions with the University and OUS, and received a formal response from the University. The University s Management Response to MRC s findings and conclusions are presented in the body of this report. Findings Generally, MRC s findings relate to language in the contract with the Development Manager related to billing, and its processes around billing. MRC also noted findings related to ambiguous contract language in Hoffman s agreement related to contract price calculations. Finally, MRC includes findings related to the invoice review and approval process used by the University for the Project. Development Manager JMI s contract did not state personnel billing rates, but mandated negotiation of these rates after Contract execution. MRC found that the billing rates were not formally negotiated. Further, despite contract provisions requiring adequate accounting of costs, the MRC 3

6 Development Manager did not employ a time tracking system or provide supporting documentation for hours charged by personnel. The University expanded by amendment its contract with the Development Manager to provide services related to the Parking Garage construction and two other projects. While the University expected the Development Manager s staff would work on both projects, the Development Manager s lack of time tracking and set billing rates created a control weakness, and the potential for overcharges. While the objective of an audit is to quantify exceptions, given the lack of reliable information related to time tracking, no data was available for MRC to assess any impacts. The Development Manager s contract values for the Arena and Garage construction projects total approximately $6M. MRC also notes that the University has a different position regarding its contract with JMI. Specifically, the University views its contract with JMI to be a lump sum for personnel expense. Under such an agreement, the amounts paid to JMI for its work on the Arena and other projects added by Addendum would be appropriate without a detail of time spent. Additionally, the use of a Lump Sum allowed for more predicable project budgeting and risk transfer for price escalation. These areas are discussed in the body of this report. General Contractor The primary finding regarding Hoffman is related to its contract and the late addition of the Warranty Fee provisions to the agreement. The addition of this Contract provision facilitated the completion of contract negotiations with Hoffman. However, MRC noted that the contract language was allowed for more than one interpretation regarding the application of the Warranty Fee, and MRC found that Hoffman s interpretation was not consistent with all of the Contract terms. As a result, Hoffman s GMP calculation incorrectly applied the Warranty Fee to Cost of the Work, which skewed its CM/GC fee proportionately. In addition, the Warranty Fee percentage was misapplied to a reduced Cost of Work in the GMP calculation, and incorrectly included as mark up on Contract Amendments increasing the GMP. Despite these interpretations, the net effect of the variance based on MRC s recalculation was not material (less than $30,000, or 0.02%). University of Oregon MRC s findings relate to project controls employed by the University, including the University s oversight of the Development Manager, and the payment approval process employed by the University. Project controls are processes that are developed to manage construction cost, schedule, scope and quality. The Development Manager did not have formal, documented processes for project controls. MRC found that the absence of these formal processes lessened the accountability of project management decisions and actions (such as the basis for changes, progress updates on schedule or cost). Further, in the absence of formal protocols, such as a standard to define what is an acceptable interval for project reporting or a threshold for providing cost support for construction expenses, the actions are solely based on the subjectivity of the personnel employed. While MRC found that the Development Manager employed competent people, there remains room for improvement in certain areas of project oversight given the use of public funds. Ultimately, the importance of these methods becomes apparent when a project experiences significant MRC 4

7 changes or conflicts when schedule delays, defective work, or cost overruns develop. This did not occur on the Arena Project. Further, the areas noted are presented as mandated by the audit scope defined by OUS and MRC s comments are presented for consideration regarding future improvements; there were no material cost exceptions identified for the Arena Project relative to these areas. The University, Hoffman and the Development Manager also set a threshold that reduced the need for invoice documentation from the general contractor for any contract less than $50,000. This protocol was put in place to reduce the volume of documentation. MRC found this process reduced oversight and created a potential control risk; by creating a $50,000 limit, the parties created the opportunity for expenses to be paid without sufficient verification. Further, as a public project, a best practice would be to require all supporting documentation for a contractor or subcontractor s costs. Implementing dollar value thresholds for invoice review is generally a common practice, however MRC found no evidence that the threshold set saved project administration expenses. MRC understands that the Development Manager spot-checked invoices that fell under this threshold with the contractor, as did the University; however MRC found no evidence to document that these reviews took place. The University put in place significant controls regarding cost control for the project, which required customization in some areas of processing to accommodate the size and complexity of the Arena Project and its contracts. MRC noted several best practices in this regard. However, the payment approval process employed by the University did not adequately validate compliance with the contract terms of the Development Manager s contract, where invoices were paid without adequate support and where billing rates were not questioned. MRC also found that the approval steps in the University s payment process for Arena costs inadequately assumed reliance on others in the process, as evidenced by the payment of the Development Manager s costs without questioning the absence of time detail or approved rates. However, MRC found no financial impact as a result. As previously noted, the University considers the Development Manager s contract to be a Lump Sum price in the area that MRC concluded it was a reimbursable contract. As a Lump Sum price, the areas that MRC noted for improvement would be reduced. Conclusions The audit revealed numerous best practices in the areas of gift-in-kind contributions and subcontractor selection. Immaterial variances due to accounting and documentation processing errors were noted during the course of the audit and discussed with the University and its partners. Additionally, procedural oversights resulting in immaterial control issues on the Arena Project were also identified and communicated to the University. MRC s conclusions are discussed in the body of this report. Conclusion OUS required a comprehensive review of the Arena Project. MRC reviewed 80% to 90% of key contract costs and found no material cost exceptions. MRC conducted more than 100 interviews of multiple parties involved in project. MRC reviewed all performance language of key stakeholder contracts and compared what was agreed to what was done. MRC MRC 5

8 examined business processes and considered the competencies of the companies and employees charged to execute Arena Project responsibilities. As a result, MRC presents a comprehensive discussion of areas done well, and areas for improvement. MRC respectfully submits these findings, with the context that the Arena construction project has been a successful endeavor for the University based on the UO s goals for project success. Should new information be presented, MRC reserves its right to modify the comments in this report. Management Response Summary University of Oregon The University of Oregon acknowledges the work of Marsh (MRC) and its partners and staff and is pleased to note the conclusion that the Matthew Knight Arena project was completed on time and on budget and in accordance with the terms of the agreements governing its construction, resulting in a conclusion that the arena project has been a low risk endeavor for the university as of the close of the field work. The university is also pleased to note that Marsh identified no significant quality issues related to the project and recognized numerous best practices in the areas of gift-in-kind contributions and subcontractor selection. The university will continue with the best practices it applies as noted in the audit and will incorporate other best practice recommendations in future projects. MRC 6

9 2 Project Overview The Matthew Knight Arena will be a 397,825 square foot, 12,500 seat state-of-the art multipurpose facility for the University of Oregon men s basketball team and women s basketball and volleyball teams, as well as other University and community events. In 2003, National Championship Properties, LLC (NCP) was founded as a fully-owned subsidiary of the UO Foundation to carry out the Arena Project. The idea was that NCP would develop and build the Project and then operate the Arena or donate the Arena to the UO. In late 2003 and the beginning of 2004, NCP contacted Hoffman Construction Company (Hoffman), Ellerbe Becket Architects (Ellerbe) and TVA Architects and started discussions about the development of an Arena. Between the end of 2004 and the beginning of 2006, NCP, the contractor and the architects were working on the financial model of the project as well as a preliminary design. It also became apparent that the chosen site, Howe Field, was inadequate, and the University began to pursue land acquisitions at the current site. In November 2007, it was determined that private development, construction and operation of an arena would not work. Under the Board Administrative Rule in effect at the time, in order to meet the completion date, the UO entered into an emergency contract with NCP for design and pre-construction of the Arena. This allowed NCP to proceed with the design and pre-construction activities, actions taken to avoid cost escalation and delays that could have jeopardized the Project. Since Ellerbe, TVA and Hoffman were involved from the inception of the Project, had unique project knowledge and had developed a collaborative relationship, the University decided that the continued involvement of these parties would lead to an on-time and onbudget project. Therefore, subsequent to its contract with the UO, NCP entered into preconstruction agreements with Hoffman and Ellerbe. In February 2008, the State Legislature approved $200 million of taxable fixed rate bonds for the development and construction of the Arena Project. In March 2008, UO engaged JMI as Development Manager. In June 2008, the State Board of Higher Education approved the Project. MRC 7

10 The UO broke ground on the Arena Project in February The planned completion date noted by MRC during field work was December 29, During and at the close of field work, the Project was on-time and on-budget. OUS Audit The (OUS) issued Request for Proposal # on May 21, 2010, which detailed the scope of work to be conducted for the Matthew Knight Arena Project Audit. After a competitive selection process that involved several responsive firms, OUS engaged Marsh Risk Consulting (MRC) to provide construction audit services for the University of Oregon s Matthew Knight Arena (Arena or Project). In accordance with the Request for Proposal, the audit services performed by MRC were focused on the following tasks: For the contracts used to design, manage and construct the Project: o Determine whether the invoices are in accordance with the Contract documents o Determine whether the change orders are appropriate and priced correctly o Reconcile Project billings to payments made and recorded in the University financial system o Identify potential cost exceptions, overcharges or duplicate payments o Identify potential project control deficiencies and exposure to overcharges on future projects and identify areas of improvement For Hoffman Construction Company (Hoffman) o o o Project Controls o Review Hoffman s subcontracting practices Review the inclusion and application of in-kind donations in the subcontractor s bid Review sole source awards, if any Evaluate the effectiveness of the internal project controls employed by JMI o Evaluate the effectiveness of the internal project controls employed by the UO Review the nature and effectiveness of the University s outreach program Identify any improvements that UO can apply in future projects, to obtain best value for public funds, to ensure fair and responsible practices, and to achieve other public goals To carry out its audit tasks, MRC reviewed the available contemporaneous project documents including but not limited to: MRC 8

11 Contracts, Amendments and Addendums to the Contracts between the parties involved OUS policies and procedures, Oregon Revised Statutes (ORS) and Oregon Administrative Rules (OAR) Project invoices from Hoffman, Ellerbe, JMI, Clair and other miscellaneous vendors and consultants Monthly Project Reports from Hoffman and JMI UO financial system outputs Project budgets (planned and current) Cost Proposal Notifications (CPNs) and supporting documents Documents relative to subcontractor selection Certified payroll reports In addition to document reviews, MRC conducted detailed interviews with personnel from: Hoffman JMI UO Capital Construction UO Business Affairs UO General Counsel UO Finance and Administration NCP University of Oregon Foundation Department of Justice for the State of Oregon Ellerbe Project Manager and Accountant In its review, MRC also assessed the reasonableness of certain costs through comparisons to market and industry guides. For example, equipment rental prices billed in contractor or subcontractor change orders. MRC s detailed findings and conclusions are discussed in Sections 3 through 7 of the report. MRC 9

12 3 Contract Assessment and Audit A fundamental purpose of the audit was to determine if the University of Oregon was billed correctly and received what it contracted and paid for. This evaluation was accomplished through an assessment of the following areas: Contract Billings Change Orders University Project Payments Project Controls The contracts subject to audit included:...the UO s project management, architectural services, construction and consulting contracts related to design and construction of the Arena. 1 Specifically, these firms included: 2 JMI Sports, Inc. Development Manager / Owner s Representative Ellerbe Becket Architects Design and Architectural Services Hoffman Construction Construction Manager / General Contractor MRC reviewed the processes, protocols, procedures, other controls, and documentation employed to assess and pay for the performance of the contracting parties. MRC conducted interviews with the on-site management and administration personnel responsible for contract billings, as well as off-site audit field work consisting of document reviews and interviews with the home office personnel of JMI in San Diego, CA and Hoffman in Portland, OR. Discussions were also conducted with Ellerbe. MRC also conducted interviews of several University, State and Foundation employees involved or familiar with the project s contracting and execution phases, and MRC was regularly on- 1 Request for Proposals RFP # Matthew Knight Arena Project Audit. May 21, Page 15, Scope of Work, A Nature of Services, part 1 2 As identified in the Request for Proposals RFP # Matthew Knight Arena Project Audit. May 21, Page 8, Background. Pages 8-15 MRC 10

13 site at the Arena for the few months that entailed audit field work, where interviews were conducted and independent observations were made. MRC s observations, findings and recommendations are presented below. JMI Sports JMI serves as the Project s Development Manager providing development oversight and management of the Project team on behalf of the University. In its role as Development Manager, JMI effectively served as the Owner s Representative in conjunction with the University s Capital Projects division. Given the size, scope, schedule and complexity of the Arena, MRC found that JMI s involvement was critical to the success of the Project. However, MRC noted inconsistencies regarding the manner in which JMI s contract, and contract billings were managed. MRC notes that JMI and its site and home office teams were supportive of the audit. MRC advised JMI of the observations regarding contract billings (discussed below) and JMI readily provided information and answers to MRC s inquiries. Fees and reimbursable expenses paid to JMI for its services are specified in its contract. 3 Specifically, JMI s billings are to include: reimbursable personnel and overhead expenses, reimbursable business expenses at a lump sum, reimbursable insurance expenses and a development management fee. Finding 1: Recommendation 1: JMI s contract billing practices varied from contract language. The University should consider revising future contracts to explicitly define contract billing terms. Management Response 1: Management acknowledges that while both the university and JMI had a common understanding of the contract language, which was implemented as understood by both parties, future contracts of a similar nature should include more explicit language. JMI s rates were not negotiated as required by contract. Rates were billed based on projections of salary cost. These projections were used to develop the estimate of payroll cost that was provided in JMI s contract, dated March JMI s contract states: The University and Development Manager shall mutually agree upon personnel billing rates and that the University will reimburse JMI for personnel costs based on these rates. MRC found no documentation to substantiate that the University and 3 March 1, 2008 Development Management Agreement between The State of Oregon acting by and through The State Board of Higher Education on behalf of The University of Oregon and its Athletic Department Section 5 MRC 11

14 JMI negotiated the rates for JMI s personnel. Lacking a negotiated rate, JMI s billings reflect a Lump Sum invoicing methodology based on JMI s estimate of personnel cost that was included in JMI s contract 4 dated March 1, The contract billing stipulation appears in JMI s contract, Article 5.1, as follows: 5 The University shall reimburse the Development Manager for cost associated with the Development Manager s Project personnel and general overhead during the Agreement Term. The University and Development Manager shall mutually agree upon personnel billing rates and the time allocation for each Development Manager employee assigned to the Project. On a monthly basis, the University shall reimburse the Development Manager 135% of the agreed upon personnel rates, as compensation for Project related salaries, payroll burden and general overhead. Attached hereto is the Development Manager Cost Summary, which provides an annual estimate of these total costs. Based on discussions with the University, MRC understands that JMI s contract was intended to be a Lump Sum agreement. It is beyond the scope of MRC s work to make a similar legal interpretation. MRC considered the contract wording when evaluating JMI s billing practices. Regarding personnel expense, JMI s contract language incorporates wording such as reimburse and cost associated. The contract refers to an estimate of these costs but does not state that the estimate established a fixed fee to be used for billing purposes. The contract is silent regarding an agreement constituting a fixed price or lump sum for personnel cost. Other JMI contract costs, such as Reimbursable Business Expense, are more clearly defined as a fixed rate with a numerical cost defined directly in the language of the contract. Based on these contract terms, MRC found that JMI s contract wording with respect to personnel expense was not followed. MRC notes that as a lump sum or fixed price agreement for personnel expense, which is the University s position, then JMI billed according to its estimated costs provided at contract inception. MRC found that JMI s estimate of cost provided in March of 2008 included guaranteed increases for personnel expense through This is a business decision for the University, where the need to confirm pricing for development and construction costs to attain budget approval must be weighed against the risk of future price escalation. MRC understands that these considerations were made by the University. MRC recommends that the University incorporate specific language in future contracts to explicitly classify contract expenses as lump sum. In the event the contract requires 4 Appendix A of JMI s contract 5 March 1, 2008 Development Management Agreement between The State of Oregon acting by and through The State Board of Higher Education on behalf of The University of Oregon and its Athletic Department Section 5.1 MRC 12

15 agreement or negotiation of rates or costs, MRC recommends that the University document these agreements. Finding 2: Recommendation 2: JMI s Fixed Price $1.6M Contract Addendum for adjacent University projects cites potential cost savings as the basis for expanding JMI s scope; however, MRC found that the absence of time tracking for the hours spent by JMI employees prevented the identification or measurement of these savings. The University should consider requiring actual time detail as part of contract addendum pricing negotiations on future development manager or other professional services contracts. Management Response 2: Management acknowledges the comment and understands the recommendation. After the Arena Project was underway, the UO started other projects adjacent to the Arena: the Parking Garage, the Triangle Area, and the Alumni Center. The University amended JMI s Contract for the additional work. Addendum #1 was executed March 1, 2009 for a lump sum amount of $1,605, MRC interviews with project personnel indicated JMI had a designated Senior Project Manager for the Parking Garage and the Alumni Center who worked from JMI s home office of San Diego. Interviews and the project records indicated that some of the personnel assigned to the Arena Project also supported the adjacent projects. JMI s lack of time tracking and documentation of personnel costs (agreed upon rates billed at actual time) by project prevented MRC from verifying JMI s payroll costs on the Arena versus other University projects. MRC found that some of JMI s Arena personnel included in the Arena estimate at 100% utilization performed work on the other University projects. Accordingly, MRC questioned how JMI employees could be full-time on the Arena and also bill the other projects. JMI explained that its billing practice is based on a forty-hour week, where 40 hours represents 100% of time spent. For those employees that work on other University projects and exceed 40 hours, JMI bills in excess of 100% of the rates, and allows for billing in excess of 40 hours. JMI explained its billing practice to MRC, as follows: 7 JMI Sports does not track the exact time that employees spend on each project, given that we do not invoice our clients on an hourly basis. Instead, average time allocations are estimated for each project team member, based upon a 40 hour work week. Those 6 March 1, 2009 Addendum #1 to Development Management Agreement Between The State of Oregon acting by and through The State Board of Higher Education on behalf of The University of Oregon and its Athletic Department and JMI Sports LLC for Arena Project October 6, 2010 JMI Sports Letter Response to an MRC audit request MRC 13

16 who are solely working on the arena project, or that spend at least 40 hours per week on the project, are allocated at 100%. Those employees who also devote considerable time to other projects, carry an allocation that reflects the average amount of time that we (with their input) assess they are spending on the arena project. For these individuals, their actual time spent working on the arena project varies from week to week, and month to month; however, our assigned allocations are conservative in the aggregate. To meet the demands of our projects, a typical work week for most employees exceeds 40 hours; therefore, this billing strategy based on estimated percentages of a 40 hour work week, versus actual hours worked, is of benefit to the University. MRC questions this billing methodology given the absence of time tracking, which would be a practice that would allow JMI to both know and show that an employee actually worked 40 hours on a project. Further, without the recording of time, it is not possible to evaluate how much substantial time JMI employees spent on other University projects after working full-time on the Arena Project. Last, because the billing allocation JMI uses in invoicing the University is based on the premise that employees charging full time to the project worked at least 40 hours, the absence of any records to show this time prevents verification by the University, or its auditors. JMI provided MRC with an analysis of the billing rates developed for the Arena Project. These rates were based on the projected average annual compensation (salary plus bonuses) of JMI s employees. Given this computational basis, JMI is compensated for its entire annual costs when an employee is billed at 40 hours per week; JMI does not incur additional cost when an employee works more than 40 hours in a week. Further, JMI s estimated rates include projected bonuses and raises. Since there is no incremental cost to JMI for work over 40 hours per week, the University should consider the potential of paying in excess of JMI s actual personnel costs when implementing lump sum agreements. Additionally, while the University endeavored to save cost by leveraging JMI s knowledge of UO and its existing mobilization on-site, MRC recommends that the University consider putting these contracts out for competitive bid in the future to validate savings. Based on a review of the scope of work in JMI s Contract addendum, MRC also recognizes JMI s financial risk given the range of work that may be required. Specifically, the Addendum states: The Development Manager will provide the services indentified in Section 2 of the Agreement. 8 Any underestimated level of effort could create an increased personnel expense to JMI. In this regard, the University effectively shifted the risk of increased costs to JMI by using a fixed price for the Addendum work. However, given the absence of time tracking, the University has no way of verifying which projects that JMI personnel spent time. In the event of a change order for increased costs, the University would not be able to assess the overruns or underruns of JMI s hours across three projects, on one contract, without time detail. There were no JMI cost overruns during the Project, and MRC presents these comments as a best practice for consideration on future jobs. 8 March 1, 2009 Addendum #1 to Development Management Agreement Between The State of Oregon acting by and through The State Board of Higher Education on behalf of The University of Oregon and its Athletic Department and JMI Sports LLC for Arena Project 2.1 MRC 14

17 Finding 3: Recommendation 3: Lump Sum billing set in 2008 for the personnel cost of the University did lock in pricing risk and allow for accurate project budgeting; however, actual market conditions indicate a lump sum structure was not required. UO should consider using guaranteed maximum prices or target price contracts, along with lump sum pricing for future development manager contracts. Under these contracting structures, the University can develop long term (over 1 year) project budget amounts, include shared savings provisions if appropriate, and better align the cost of development services with market conditions. Management Response 3: Management acknowledges the comment and understands the recommendation and will consider this when balancing risk in future contract situations. MRC s interpretation of JMI s contract was that JMI and the University were required to negotiate its billing rates for the cost of personnel charged to the project and bill actual amounts or agreed allocations. MRC did not find documentation beyond JMI s March contract to demonstrate these negotiations occurred. The University s position differs from MRC, asserting that personnel cost was agreed upon as a lump sum. Given that the University was a party to the contract, and MRC was not, comments in this area are noted for the University s consideration in drafting future contracts. In the event the contract for JMI s personnel is a lump sum agreement, MRC notes that at the time of the agreement (March of 2008), JMI had yet to hire the majority of its employees used for the Project. This was a primary factor that MRC considered when presented with the University s contract interpretation; it has not been MRC s experience to see pricing locked in for 3 years for employees yet to be hired as Development Managers. As a best practice, and at a minimum, MRC recommends that the qualifications, years of experience and detailed duties for each planned hire be clearly determined and documented before agreeing to long-term fixed pricing. MRC also recommends that the University negotiate these employee rates based on current market conditions for each hire, as allowed by contract with JMI, but not done during the Arena Project. The projection of JMI s payroll cost for 2008, 2009, 2010, and 2011 was submitted in March of 2008 and billed approximately as presented. JMI provided MRC an analysis of employee cost for these years. This analysis showed that generally, JMI s projections and subsequent payments to employees included both raises and bonus amounts each year. Given that JMI billed the Project pursuant to the projections in 2008 that included raises and bonuses for future years, JMI locked in increases for compensation into Granting raises and awarding bonuses are a normal practice in the construction industry. However, there have been significant increases in construction unemployment rates during the period of JMI s contract, which can become a significant factor in compensation MRC 15

18 increases. Accordingly, MRC found that locking in JMI s payroll costs for the term of the 3 year contract was not necessarily advantageous to the University given the actual market conditions. JMI did prepare two reconciliations of its payroll costs. While MRC was able to verify select compensation amounts, some conflicting amounts were noted. Further, in the absence of appropriate time tracking and documentation such as hours billed to a project code MRC could not rely on the analysis of billings prepared by JMI to validate the costs JMI actually incurred for personnel working on the project. Given the absence of corroborative information, it is not appropriate for MRC to quantify the actual payroll costs, increases, and bonuses paid by the University since a reasonable comparison of actual costs to billed costs would not result. Conclusion 4: JMI managed the Project effectively but MRC recommends improvements to formalize and document project control policies and procedures. Recommendation 4: The University should implement a pre-qualification process prior to engaging an Owner s Representative to evaluate the methods, processes and management tools the firm will use to align with the goals of a public project. Management Response 4: Management acknowledges the conclusion but believes that appropriate due diligence was performed as evidenced by the successful completion of a major public works project that was on time and on budget. JMI met or exceeded contract obligations in several areas. MRC found that JMI s involvement facilitated the resolution and management of several project challenges, from development through construction, as well as the on-time, on-budget status. MRC also found that JMI established and followed informal project control procedures during the Project, where formalized processes would constitute a best practice. Accordingly, MRC noted some deficiencies in project management. However, the impact of these observations was immaterial. MRC s comments are provided to illustrate areas for on-going development for JMI, and opportunities for the University to implement additional best practices when setting prequalification and contract criteria for Development Managers. MRC s discussion of project management performance, including effective practices and areas for development, are below. Project Accounting JMI employs competent colleagues in the areas of project accounting, as well as the other project control areas noted below. MRC found JMI employed appropriate methods to MRC 16

19 segregate duties in financial areas, and had several informal practices for oversight and quality control, including regular oversight from JMI s executives. However, MRC noted areas where project accounting practices could be developed to strengthen reporting and controls. MRC notes, these areas were not found to be materially deficient. MRC s recommendations in project accounting include: Implement a documented reporting procedure to verify Owner expenditures as a best practice MRC found that JMI complied with its responsibilities to develop and maintain the overall project budget. JMI s Accounting Project Manager and UO Accounts Service Manager communicated to ensure JMI s records reconciled to the separate reporting system maintained by UO. However, communication was over the phone and no formal report was issued to UO s Accounts Service Manager. MRC recommends that, in addition to the cross-check of the financial statements and records, JMI should document the control process itself and enhance the reporting around project budget monitoring. This would enhance accountability and assist in identification of any budget-to-actual variances in University spending, as well as any deficiencies in the UO s reporting systems. While system defects are less common, as a financial system becomes customized for a particular use, such as for a construction project, human error is possible. Incorporate a system of daily time tracking JMI did not use a time tracking or accounting system to capture actual hours spent by its personnel on the Project. As such, no contemporaneous records (i.e., documentation that was generated when the cost or time was incurred) exist to verify the actual time spent by each individual billed to multiple University projects. While this record keeping can vary, it may consist of: A system that includes a designated project code where employees can charge daily time (hours) Daily logs that demonstrate the time spent on a given project by date JMI did provide budgeted and actual summaries of payroll cost allocations in Excel spreadsheets. These summaries do not adequately support the time spent on the Arena Project but did adequately verify for the project period that the employees worked for JMI full or part time. Include backup for actual costs or actual against estimated costs in each invoice for any reimbursable expense. Thresholds can be set given the needs of the University and materiality of the charge MRC s review of JMI s monthly billings found that JMI s invoices did not include backup for the cost of personnel that were on-site or working on Arena business off-site. As a reimbursable contract, this information would be required to substantiate costs. Invoice documentation also serves the University in the event of change orders or contract addendums, where a cost history serves to validate future pricing. MRC 17

20 Project Reporting JMI submitted monthly project reports to NCP and the UO. The monthly project reports complied with the Contract requirements and included budget, project status, expenditures, a milestone project schedule, and cash flow projections. Require consistent and detailed reporting regarding project progress Based on its contract 9 and as typical for every large construction project, Hoffman was required to submit a monthly project report to the Owner s Representative. Hoffman submitted three monthly reports to JMI after which it was mutually agreed that Hoffman did not need to submit more because it was redundant with the reports that JMI was submitting to NCP and UO. MRC reviewed Hoffman s reports and noted that Hoffman s reports included a much higher level of detail than the JMI reports regarding the progress of the project. Given the scope and dollar value of the project, and the availability of this information, MRC found that Hoffman s reporting should have been continued and provided to the University. MRC has commented here, under JMI s section, given JMI s responsibilities on behalf of the University for oversight in this area. Appropriate practices regarding meeting minutes; best practice recommendation: obtain signed copies of meeting minutes as part of project reporting The project team (Hoffman, JMI, and UO) conducted weekly project meetings. These meetings were documented by Hoffman and made available via a project website. MRC found this to be an appropriate practice. MRC noted one area for improvement that would benefit the University: the UO (or its Development Manager) should obtain signed copies of the meeting minutes each week. A best practice would be to have all project stakeholder representatives at the weekly meetings sign the meeting minutes, then the University maintains these documents. This memorandum can become important should there be a disagreement regarding discussions regarding project actions later in the job. MRC understands that the main discussions and communications between the parties were held during the Weekly Construction Coordination Meetings, where representatives from Hoffman, UO, JMI, Ellerbe, TVA, and sometimes the engineers (Haris, HEI) attended. Hoffman kept detailed meeting minutes which were provided to JMI and UO. MRC noted there was no documented commentary back from JMI or the UO. The construction coordination meeting minutes include information on new, resolved and unresolved items, 4-week look-ahead schedule, RFI log, submittal log, and design items that needed to be addressed. Effective communication by JMI MRC understands from various interviews with key project participants that the JMI Executive Project Manager was the focal point for coordination and reporting activities 9 February 6, 2009 CM/GC Contract, Article 3-Work of the Contract, Section MRC 18

21 between Hoffman and NCP and UO. The Executive Project Manager received updates on the project status from JMI s Senior Project Manager. JMI s San Diego-based Project Manager was responsible for accounting, and for updating the Executive Project Manager on project budget, cash flow and commitment reports. MRC found these JMI practices to be effective. No communication gaps were identified between the JMI team, Hoffman and UO. University Payment Authorizations Increase documentation to substantiate reviews and approvals and consider the use of earned value metrics to audit project completion for payment purposes There was an established process for review and approval of payment applications but the accountability of the authorizations in the process could be strengthened with improved documentation and written policies and procedures. UO procedures should mandate an Earned Value assessment of costs, and uniform procedures for budgeting and payment tracking that would allow independent verification of budget status. Section 2.1 of JMI s contract stated: 10 The Development Manager shall develop and implement procedures for the timely review and processing of applications by the Contractor and other consultants for progress and final payments. There was an established process between JMI and UO for review and approval of payment applications. The progress payments were based upon percent completion of the work. The percent complete was visually verified by JMI and UO s representative on-site. While visual verification of percent of the work is appropriate, it should not be the sole factor used to determine the project status. An Earned Value (EV) calculation is an industry best practice and should be used for verification of the project status against cost and schedule. Scope Management / Change Orders Enhance documentation regarding change order discussions and decisions JMI did not have formal project control policies and procedures regarding change orders. Formal policies, processes and procedures, and appropriately tracking the nature of the changes would increase accountability for the scope/change process. Contractor s CPNs should be rigorously reviewed in order to assure that the application of fees and mark-ups are correct. However, MRC concluded based on its interviews and review of related information that JMI and Hoffman did review changes on an adequate basis. MRC documents its understanding below. 10 March 1, 2008 Development management Agreement between The state of Oregon acting by and through The State Board of Higher Education on behalf of The University of Oregon and its Athletic Department Section 2.1 (k) MRC 19

22 JMI did not maintain a CPN log that identified the reason for the change. Starting with Amendment 6, JMI maintained CPN backup information used for the Amendment which included the amount and description of the CPNs, the related documents (RFI, CCD) and the approval entity. The CPN log provided by JMI which was used for Amendments 3-5 included only the cost breakdown for each CPN. It is industry best practice for the Owner or Owner s Representative to maintain a change order log identifying the reason for the change as a check against the Contractor s log. Based upon discussions with JMI s team and review of contemporaneous documents, MRC understands that all project documentation was entered in online management software. JMI routinely interacted with the design team and discussed alternate solutions to a problem, and helped determine if a change was appropriate for the Owner s needs and standards. When a CPN arrived, the Senior Project Manager (SPM) received the CPN and reviewed the change together with Hoffman and Ellerbe; the SPM then discussed and questioned the costs as appropriate. After the JMI SPM approved the cost, the CPN was sent to the JMI Executive Project Manager (EPM), who reviewed it again with the SPM. MRC found that the SPM performed an extensive review and as necessary, made reductions/adjustments to CPNs that benefited the University. As discussed during an interview with MRC, the CPN review was always based on the intent of GMP drawings. The EPM also noted that there were many design corrections due to the fast-track nature of Project and coordination problems between design disciplines. MRC found this to be a valid statement based on the complexity of the Project. JMI was responsible for the final CPN cost approval. As noted in other sections of this report, there were some inappropriate mark-ups which JMI failed to identify. However, these markups were not material and the differences went both ways in favor and not in favor of the UO. As a best practice, MRC s recommendation is to enhance the documentation to memorialize the progress and changes during the job, which can be developed from effective processes already employed by JMI in managing project change. Schedule Control Incorporate contemporaneous reviews of electronic schedules in the project oversight process Both JMI and Hoffman managed the project schedule successfully. A complex Arena Project was on-time at the end of MRC s field work. MRC s comments relate to opportunities for improvement. JMI was responsible to develop the overall project schedule which should include Hoffman s construction schedule, plus the design and permitting activities. JMI included a milestone schedule in its monthly project reports for the UO. After review of selected schedules, MRC found that JMI s scheduling process did not meet industry best practices for scheduling since its milestone schedule did not use relationship logic to determine project status but applied mandatory date constraints to its activities. MRC s position is that a review of relationship logic provides a more accurate and MRC 20