MEMORANDUM. Executive Summary.

Transcription

1 11500 WEST OLYMPIC BOULEVARD, SUITE 502 LOS ANGELES, CALIFORNIA TEL: (310) FAX: (310) PUBLIC RESOURCES ADVISORY GROUP MEMORANDUM TO: Mary Lewis, Chief Financial Officer of the City of San Diego FROM: Public Resources Advisory Group ( PRAG ) SUBJECT: San Diego Integrated Convention Center Expansion/Stadium and Tourism Initiative Important Considerations and Risks to the City DATE: August 15, 2016 At the City s request, PRAG has reviewed portions of the: i) San Diego Integrated Convention Center Expansion/Stadium and Tourism Initiative s (the Initiative ) financing approach for a combined convention center and Chargers stadium (the Project ) and ii) sample cash flow analyses prepared by Goldman Sachs & Co. ( GS ), in their capacity as financing and structuring agent for the Chargers Football Company, LLC, that models the ability of the future transient occupancy tax ( TOT ) revenues levied under the Initiative to fund the Initiative s requirements. Our objective was to review, consider and analyze the accuracy, rationale and suitability of the following assumptions and projections for an issuance of TOT backed revenue bonds: TOT tax revenue projections (growth rate); Debt service and bond structuring assumptions; Debt Service Coverage ratio; Sizing and interest rate assumptions; Rating impacts, credit and market access implications for the City; and Project Pro Forma scenarios PRAG s financial analysis excludes certain Project considerations that are also relevant to understanding the full scope of the City s fiscal risks related to the Initiative. One key assumption that PRAG did not independently verify is the reasonableness of the total Project Cost estimate of $1.8 billion. This cost estimate was a base case Project Cost assumed in the cash flows provided to us by GS. Given that the Project Cost is fundamental to all results, it is very important for the Project Cost to encompass not only realistic costs for the core stadium and convention center expansion project but also realistic costs for land acquisition, environmental mitigation, relocation of certain facilities and any ancillary capital infrastructure improvements not currently specified in the Initiative. In addition, our analysis did not consider any broader potential benefits (direct or indirect) of the Project on San Diego s economy or the City s General Fund. Executive Summary. Cash Flow Considerations and Risks. The financing approach incorporated into the Initiative relies on increasing the City s Transient Occupancy Tax rate by 6%. Of the new 6% levy, the first 1% would be transferred off the top to the Tourism Trust Fund ( TTF ). The remaining 5% would then be available for the transfer of a second 1% to the TTF (after the payment of debt service), which together with the

2 Page 2 of 21 (Executive Summary) initial 1%, would replace the current 2% Tourism and Marketing District Assessment ( TMDA ), and for the financing, planning, construction, operations and maintenance ( O&M ) and future capital improvement costs ( CapEx ) of the Project. These TOT revenues would be used on a pay go basis and to pay debt service on TOT backed revenue bonds. In addition, they are required to be used in a certain order, with the City s general fund having the lowest priority in the flow of funds, receiving revenues in a given year only if all of the prior requirements have been met as depicted in the flow of funds chart to the right. Shortfalls in any year must be replenished prior to the City s general fund receiving any money. The ability of these TOT revenues to meet the Initiative s requirements is the primary financial risk factor to the City. As discussed below, based on this flow of funds, there is a significant difference between the ability of the TOT revenues to cover debt service on the bonds and the ability to cover all of the requirements under the Initiative that must be met before any revenues flow to the City s general fund. This will primarily depend on a combination of three variables that have a high degree of uncertainty at this time including the actual: i) Project Cost (combined construction, land acquisition, environmental mitigation, ancillary infrastructure); ii) interest rates for the bonds (including what portion can be taxexempt versus taxable); and iii) future TOT revenue growth. Ultimately, it will be the actual level of each of these three variables together, that over time will determine whether the TOT revenues can initially fund the Project and then meet all future annual debt service, O&M and CapEx expenditures. The City s general fund will only receive TOT revenues if these three priority buckets are satisfied first. As a result, PRAG believes that given their inherent uncertainties, it is important for the City to rely on conservative estimates of the interest rate, TOT growth rate and Project Cost assumptions to determine if any revenues will flow to the general fund under the Initiative. It should be noted that the GS financial model and cash flows assume a tax exempt interest rate on 10 of the bonds. If under tax law at the time of issuance, all or a portion of the bonds were required to be issued on a taxable basis, costs over the life of the bond repayment period may increase significantly based on this single factor alone. Methodology. PRAG reviewed GS analysis and assumptions and was generally comfortable with their bond structuring assumptions given the nature of TOT backed revenue bonds and the flexibility to financially engineer the bond structure closer to the time of sale. However, to provide the City with information on the associated risks, in our view, it was important to analyze GS future revenue growth, interest rate and the assumed Project Cost of $1.8B (for the combined convention center and stadium construction, land acquisition, environmental mitigations and ancillary infrastructure), and to run sensitivity analyses. As a baseline for our analysis, we reviewed GS assumptions from one of their base case scenarios dated April 28 th which assumed an interest rate of 4., a $1.8 billion Project Cost and the City s Five Year Forecast of TOT revenues and a 4% annual growth rate thereafter, which is close to the City s historical TOT revenue growth rate. We note that in our opinion, each of these assumptions and all three of them on a combined basis are not conservative

3 Page 3 of 21 (Executive Summary) for a Project with this long of a lead time and are subject to significant change as described further below. As of today, various aspects of the current Project Cost are still being evaluated and are subject to change over time. In addition, GS TOT revenue growth assumptions would be considered optimistic by the rating agencies if relied upon for debt service coverage in a more standard TOT backed revenue bond structure (which lacked the coverage built into this structure to cover O&M, CapEx, etc.). Also, based on a historical analysis, while 4. is a reasonable rate for 30 year tax exempt bonds today, it is not a conservative rate for longer term planning purposes for a $1 billion+ deal that may carry a size penalty, is likely to be sold some years in the future, has the potential to require a taxable portion and may need to be amortized longer to generate more bonding capacity. As a result, PRAG would recommend using a 5% interest rate for any base case scenarios. For the purposes of our analysis, we have used $1.8 billion as the Project Cost but also, in many cases in the Executive Summary, tested a 2 increase to evaluate the impact of higher Project Costs on overall projections. The selection of a 2 increase was done for illustrative purposes only, and PRAG has no knowledge as to whether or not the $1.8 billion Project Cost or a 2 increase in the cost estimate would be appropriate. We note that in our detailed analysis, we provide various other sensitivity runs that reflect Project Cost increases that range between 1. Analysis. PRAG s analysis tested various combinations of these three variables to see if there would be sufficient revenues to: i) pay the debt service costs on the TOT backed revenue bonds ( Debt Service Coverage ) and ii) to meet debt service and all requirements in the Initiative ( Overall Coverage ). Results. Overall, given the flow of funds specified in the Initiative, with only 1% of the 6% of the TOT levy associated with the Initiative having priority over debt service, the credit quality of the TOTbacked revenue bonds is much more protected from revenue shortfalls and cost overruns than Overall Coverage which includes debt service, the second 1% transfer to the TTF, O&M and CapEx, all of which have to be met prior to the City s general fund receiving any of these TOT revenues. PRAG s analysis calculates a breakeven revenue decline percentage in other words, how much can TOT revenues decline in year 1 and still meet Debt Service Coverage and Overall Coverage, assuming that long term TOT revenue growth resumes in year 2. We have provided a myriad of sensitivity analyses (included in the Detailed Analysis which starts on page 6), but have focused just on achieving Overall Coverage in this Executive Summary. (We have also analyzed TOT revenue declines in later years, but those results, as they relate to Overall Coverage, are very similar to assuming a TOT revenue decline in year 1.) A few examples are provided below that reflect a future situation where the City could be deciding whether to proceed with a bond issuance after it has a better sense than it has today of the actual Project Cost and projected interest rate, but future TOT revenues would, of course, still remain uncertain. Using various different assumptions, there are scenarios where Overall Coverage could not be achieved and none of these TOT revenues would flow to the City s general fund. a) What Happens at a Lower Annual TOT Growth Rate Assumption of 2% and a 5% Interest Rate? If we assume a more conservative 2% TOT growth rate beginning in year 2, Overall Coverage is more difficult to meet. For example, at a 5% interest rate and the current Project Cost estimate, first year revenues would need to increase by about 6% to meet Overall Coverage. Assuming just a 2 higher Project Cost, first year revenues would need to be about 3 higher. In fact,

4 Page 4 of 21 (Executive Summary) assuming a lower TOT revenue growth of 2% and a 5% interest rate, no revenues would flow to the general fund even if the Project Cost is today s estimate of $1.8 billion. We modeled additional scenarios where Project Costs are more than 2 higher than $1.8 billion under various revenue growth rates and interest rates which are provided in the Detailed Analysis which starts on page 6. b) What Happens at a Lower Annual TOT Growth Rate Assumption of 3% and a 5% Interest Rate? If we assume a 3% TOT growth rate beginning in year 2, Overall Coverage may also be difficult to meet depending on the interest rate and Project Cost. For example, at a 5% interest rate and the current Project Cost estimate, first year revenues could decline in year 1 by about 1 and meet Overall Coverage. However, at a 5% rate and a 2 higher Project Cost, first year revenues would actually need to increase by about 1 rather than being able to withstand a decline to meet Overall Coverage. c) Historical Average TOT Growth Rate of 4%: What Happens If the Interest Rate is 5% at Different Project Costs? If we assume the City s long term TOT growth rate would be 4% beginning in year 2 and a 5% interest rate, TOT revenues could decline about 22% in year 1 and still meet Overall Coverage. However, if the Project Cost was more than 3 higher, the TOT revenues would be insufficient to meet Overall Coverage. d) Historical Average TOT Growth Rate of 4%: What Happens If the Project Cost is 2 Higher or more across Different Rates? If we assume the City s long term TOT growth would be 4% beginning in year 2 and that the Project Cost is 2 higher and vary interest rates, the rate could be as high as 6% and Overall Coverage could be met. Alternatively, for a rate assumption of 5% and a 2 increase in Project Cost, first year revenues could only decline by about 8% for TOT revenues to still meet Overall Coverage. If the Project Cost is 3 higher with a 5% interest rate, Overall Coverage is barely met with essentially no revenues flowing to the general fund. Other Fiscal Considerations and Risks. In addition to Cashflow Considerations and Risks there are a number of other fiscal considerations that we think are significant. i) Cost Sharing. The GS financial model assumes a $1.8 billion total Project Cost: the Chargers would contribute $650 million to the Project, and 5% of the 6% TOT increase in the Initiative would be available to fund or finance $1.15 billion of Project Costs. Those Project Costs would include convention center construction costs, any allocated costs in the combined convention center/stadium, and all costs for land acquisition, environmental mitigation and ancillary infrastructure. On an ongoing basis, following construction completion, annual TOT revenues would fund pre determined, inflation adjusted annual contributions for O&M and CapEx for both the convention center and the stadium. Assuming 30 year bonds with 28 years after construction completion and an annual 2% inflation adjustment, the TOT revenues would fund $450 million in convention center O&M and CapEx and $637 million in stadium O&M and CapEx. In addition, total debt service on the portion of the Project Cost that would be bonded for assuming 30 year bonds at 5% would cost $2.3 billion in TOT revenues. ii) Post Initiative TOT Levels. Under the Initiative, the total City TOT levy would be 16.5%, one of the highest TOT rates of large urban centers in the State and nationally according to a September 2015 report by a leading lodging consulting firm. TOT revenues represent an important revenue source for the City and increasing the levy for this purpose may have a negative impact on its ability to levy it for other future purposes. While there are many factors that influence industry groups from selecting convention

5 Page 5 of 21 (Executive Summary) locations, costs are certainly one of them and a higher TOT is a cost component. As a result, future increases for other purposes above 16.5% would have to consider how high the levy already is relative to other convention locations. iii) City s General Fund Credit and Headline Risk. TOT revenues would be the only revenues legally required to be used to make the debt service payments on the TOT backed revenue bonds. While it is our understanding that special tax supported debt such as TOT backed revenue bonds is included by some of the rating agencies in their calculation of an issuer s direct debt burden, based on preliminary conversations with the rating agencies and other municipalities experience, we would not expect the issuance of TOT backed revenue bonds alone to impact the City of San Diego s general fund ratings. Also, as noted above, if there ended up being related unfunded annual costs that result in general fund expenditure pressure, or if the rating agencies consider the additional TOT levy as reducing the City s revenue raising flexibility, there could be an impact on the City s general fund credit quality. In addition, as a highly visible and very large financing, the issuance of over $1 billion of bonds for this Project would likely attract attention and subject the City to headline risk. iv) Alternative Bond Structures. We would suggest that the City be aware when reviewing GS stress scenarios that bond terms are modified to be able to continue to meet the Initiative s objectives (e.g., higher interest rate or Project Cost, longer bond term). While PRAG is aware of a number TOT backed bonds that have been issued or restructured as 40 year bonds from original issuance, that is considered a relatively long bond term. In addition, while it may be possible to address for example, cost overruns or higher interest rates by extending the bond term or with a delayed financing, which would result in additional upfront TOTs, it would always be at the expense of public funds. We note that it would also be important for the lease with the Chargers to be at least as long as the bond term. Otherwise, in the event the Chargers leave the stadium at the end of the lease and bonds remain outstanding, TOT revenues would be tied up and the City would have a reduced capacity at that time to finance capital expenditures needed to attract another tenant or to otherwise dispose of an under utilized stadium. Summary. Given the excess revenues that are built into the flow of funds with only 1% of the 6% TOT levy having a higher priority than debt service, Debt Service Coverage is much stronger than Overall Coverage. As a result, it is possible to present more conservative revenue projections to the rating agencies and demonstrate sufficient Debt Service Coverage which is primarily what the bond ratings will be based on. However, Overall Coverage must be achieved prior to the City s general fund receiving any revenues. In our analysis that varied GS assumptions, there were various scenarios where Overall Coverage could not be achieved. PRAG believes that given their inherent uncertainties, it is important for the City to rely on more conservative estimates of the interest rate, TOT growth rate and Project Cost assumptions than GS base case assumptions to determine if there will be sufficient TOT revenues to meet Overall Coverage and provide TOT revenues to the general fund. In addition, there is a significant amount of time between now and when bonds would be issued. By that time, with the Project Cost estimate refined and a significant amount of interest rate uncertainty removed, the City should be in a better position to assess the risk parameters of the overall plan using an updated version of the type of analysis summarized above and described in more detail in PRAG s Detailed Analysis starting on the next page.

6 Page 6 of 21 (Detailed Analysis) Detailed Analysis. Objective. PRAG was provided with various scenarios prepared by Goldman Sachs ( GS ) that leverage the TOT revenues through the issuance of non recourse revenue bonds. The scenarios assumed various construction start dates, bond issuance dates, and maturity dates as well as different interest rates based on the various maturity dates, but consistently used the same assumptions in their base cases regarding the Project Cost as of 2018, annual TOT revenue growth, the debt service amortization pattern, and the debt service coverage requirements. The objective of PRAG s detailed financial analysis is to review GS analysis and assumptions, assess their reasonableness, identify the risk factors to the City in the financial model and cash flows provided by GS and provide the City with a sensitivity analysis that varies the important assumptions and demonstrates what may happen to the revenue stream and its ability to meet the annual: i) debt service payments and ii) inflation adjusted O&M and CapEx contributions. Given the Initiative s flow of fund requirements, only amounts in excess of debt service, the transfer to the TTF, O&M and CapEx will flow to the City s General Fund. Pro Forma. While we were provided a number of alternative scenarios prepared by GS, for this purpose, we have modeled our analysis on the scenario provided by GS dated April 28, 2016 that had the following assumptions: Issuance Date: January 2018 Maturity: 30 Years Interest Rate: 4. Tax Status: 10 Tax exempt CPI/CCI Increases: 2% annually for O&M and CapEx Projected Revenues $120,343,000 for the 6% TOTs (based on City s forecast of $210,600,000 for 10.5% for FY2017 (City s 5 Year Forecast) TOT Revenue Annual Growth Rate: Debt Service Annual Growth Rate: Minimum Debt Service Coverage Requirement: Debt Service Reserve Fund: of 5.5%, 5., 5. and 4.5%; 4. thereafter (City s 5 Year Forecast) 1% 1.5x Maximum Annual Debt Service ( MADS ) TOT Backed Revenue Bonds. To inform our analysis, PRAG began by reviewing the rating agencies methodology on special tax bonds, which includes TOT backed revenue bonds, as well as a number of the large TOT backed revenue bond deals that have been issued over the last few years. However, we note that to the best of our knowledge, there have not been any billion dollar TOT backed revenue bond transactions, and a transaction of this size may well carry a size penalty. A general summary of the rating agencies criteria is provided below. At the outset, we note that there is no one standard way to structure hotel or transient occupancy tax revenue bonds. A distinguishing

7 Page 7 of 21 (Detailed Analysis) characteristic of TOT revenues called out by the rating agencies is that it is a narrow and relatively volatile revenue stream. Given the City s own history with TOT revenues, it is familiar with the cyclicality of these revenues. As a result of the nature of these revenues, issuers have approached structuring TOT backed revenue credits in a number of different ways including funding full and partially funded debt service reserve funds as well as additional reserve funds on a taxable basis, providing back up pledges of other revenues, and senior and subordinate structures. In addition, issuers have structured both level debt and ascending debt service and have assumed both no revenue growth and revenue growth to cover debt service to best meet their own particular goals. Issuers goals vary widely from wanting to maximize bond proceeds to targeting a maximum contribution that allows for more conservative bond structures. General Rating Criteria. TOT backed revenue bonds are classified as special tax bonds in the rating agencies criteria. Within this category, TOT revenues are considered a tourist related, relatively narrow, economically sensitive, and volatile revenue source. The ratings on TOT backed revenue bonds generally depend on an analysis of the following: Tax Base: Size, depth, strength, and stability of the economic base; Historic Performance: Stability/volatility of the revenues, historical growth rates, and tourism characteristics; Financial Metrics: Debt service coverage ratio (based on annual debt service and MADS); recognizing the uncertainties of forecasting new growth, the ratings are typically based on historical revenues coverage of MADS; however, pledged tax growth rates are examined and some credit may be given for future growth; Legal Covenants: Debt service reserve funds ( DSRF ), additional bonds test ( ABT ), where excess revenues flow after debt service (to either accelerate repayment or to another purposes such as the general fund); PRAG s Analysis Approach. GS scenarios incorporate various bond structuring assumptions including debt service growth, the length of the amortization period, and the size of the debt service reserve fund as well as assumptions about future revenue growth, interest rates, annual CPI/CCI changes, and Project Costs. As noted above, there is not a standard way to structure hotel or transient occupancy tax revenue bonds. GS bond structuring assumptions in the scenario that we have focused on including a 30 year amortization period, 1% debt service growth and a DSRF, are relatively reasonable and can and would likely be refined in the course of structuring the bonds. For example, it may be that senior and subordinate bonds would result in an overall lower interest rate, debt service growth could be used on one rather than on both liens or differ between liens, or a longer amortization period may provide the City or the structure with some additional benefits. Given that GS assumptions were reasonable and will likely be subject to further financial engineering prior to issuance, we were comfortable incorporating them into our analysis. In our view, given their inherent uncertainties, it is GS other assumptions regarding future TOT revenues, interest rates and Project Costs that are very important and should be analyzed to provide the City with information on the risks associated with the Initiative. To address some of

8 Page 8 of 21 (Detailed Analysis) this, GS has over time run a variety of scenarios, including most recently, Downside Revenue Scenarios wherein different assumptions for future TOT revenues were used. Our analysis focused on what the impact will be on debt service coverage, Overall Coverage (debt service, the second 1% transfer to the TTF, O&M, CapEx) and general fund revenues at different TOT revenue growth rates, interest rates and Project Cost levels. At the outset: We note that the future TOT revenue growth assumptions, the current Project Cost estimate and a 4. all in interest rate reflected in GS numbers underpin the ability of the Initiative to pencil out financially. As discussed further below, in our opinion, each of these assumptions and all three of them on a combined basis are not conservative for a Project with this long of a lead time and are subject to significant risk of change. Given the flow of funds specified in the Initiative, with only 1% of the 6% of TOT levy associated with the Initiative having priority over debt service, the ability to repay debt service on the bonds ( Debt Service Coverage ) is much more insulated from revenue shortfalls and cost overruns than Overall Coverage which includes not only debt service but also the second 1% transfer to the TTF, and inflation adjusted O&M and CapEx contributions. The City s general fund receives no funds unless all three buckets are filled first, including funding up any prior year shortfalls on a priority basis. In each scenario we analyzed, we therefore looked at both Debt Service Coverage and Overall Coverage. As demonstrated in the charts to come, achieving Overall Coverage is significantly more difficult to achieve than just Debt Service Coverage. A simple way to think of the fundamental risk to the City of revenue projections is that long term TOT revenue growth is essential to fully fund all of the buckets set forth in the Initiative, including debt service, the transfer to the TTF and O&M and CapEx set asides. The table below uses an example from GS analysis to show why this growth is essential. First, there are little excess revenues left after funding all buckets, even assuming the relatively healthy TOT revenue growth rates reflected in the City s Five Year Forecast. Second, over time, each of the uses is expected to continue to grow, albeit at different growth rates, with only one of these growth rates tied to the TOT revenue growth rate. If TOT revenue growth is slower than the sum of the growth of the other uses, the amounts needed to fill the buckets will quickly overtake available revenues. Amounts in Year 4* Long Term Growth Rate 5% TOT Revenue $116,619,000 less Debt Service (60,777,000) 1% per year less 2 nd 1% to TTF (23,324,000) Equal to TOT Revenue Growth less Stadium O&M/CapEx (18,035,000) CPI & CCI (each 2% in GS run) less CC O&M/CapEx (12,731,000) CPI & CCI (each 2% in GS run) Net to General Fund $1,752,000 * Based on amounts from GS analysis for 30 year bond financing at 4. issued in 2018, following assumed growth of 5.5%, 5% and 5% in years 2, 3 and 4, respectively, CPI and CCI annual increase of 2%. By the time the City would be ready to issue bonds, with the Project Cost estimate refined and a significant amount of interest rate uncertainty removed, the City should be in a better position

9 Page 9 of 21 (Detailed Analysis) to assess the remaining risk parameters of the overall plan using an updated version of the type of analysis described below, focusing primarily on risks related to TOT revenue projections. Methodology: To isolate the impact that the various assumptions can have on Debt Service Coverage and Overall Coverage, PRAG developed a number of sensitivity analyses for the various factors discussed above including future TOT revenue growth rates, Project Costs, and interest rates. Our approach was to calculate a break even revenue number for year 1 or in other words, how much can revenues decline or how much would revenues actually have to increase from the current FY 2017 TOT revenue forecast to meet the requirements under the Initiative. We have also included an analysis that compares the results of revenue declines in later years rather than initial declines in year 1 to illustrate the impact of the timing of declines changing. I. Sensitivity #1: TOT Revenue Growth Rates Impact on Debt Service Coverage and Overall Coverage Question: How much can TOT revenues decline in year 1 assuming various TOT revenue growth rates beginning in year 2, GS 4. interest rate assumption and the current Project Cost estimate? As the City is aware, TOT revenues are a relatively volatile revenue stream but over the long run have shown substantial growth. In fiscal 2009 and 2010, the City s TOT revenues declined precipitously by 11.7% and 11.9%, respectively for a total decline over two years of 22.3%. However, from fiscal 2010 through fiscal 2015, the City s TOT revenues have increased by an average annual growth rate of 8.6%. Given the nature of these revenues, we believe that an important sensitivity analysis should focus on how much TOT revenues could decline in the first year and still achieve Debt Service Coverage and Overall Coverage in full under various long term TOT revenue growth rates. We calculated the break even revenue decline or increase in the first year assuming various TOT revenue growth rates beginning in year 2. This analysis looked at various long term TOT revenue annual growth rates in 0.5% increments from 0 4% starting the year after the initial decline. We note that GS analysis assumes the City s Five Year Forecast of TOT revenues followed by 25 years of 4% annual growth for an average annual growth rate of 4.2%. TOT Revenue Growth Rates and Coverage. As you can see in Chart 1A on the left of the following page, under this scenario, assuming TOT revenue growth beginning in year 2 through year 30, revenues could decline in the first year by approximately 2 and achieve Debt Service Coverage. However, if revenues increase annually by 0.5% starting in year 2, revenues in the first year could decline further, by about 31%. From 1% 4% annual TOT revenue growth beginning in year 2, revenues could decline by about 4 in year 1. (We note that the reason the breakeven revenue decline levels off at 4 is that the debt service shortfall occurs in year 1 at this level of decline, prior to any assumed increases in later years.) Next, PRAG calculated how much TOT revenues could decline or need to increase in the first year to achieve Overall Coverage, assuming no excess funds flow to the general fund under various long term TOT revenue growth rates that start in year 2. Similarly, this analysis was also run assuming long term TOT revenue annual growth rates in 0.5% increments from 0 4% starting the year following the initial decline. As you can see in Chart 1B, under this scenario, if revenues

10 Page 10 of 21 (Detailed Analysis) were flat to less than 2% higher in year 2, revenues would actually need to be higher in year 1 to achieve Overall Coverage. This makes sense given GS 2% CPI/CCI growth assumption applied to the annual O&M and CapEx contributions. If, however, TOT revenue growth averages 4% beginning in year 2, revenues could decline by about 26% in year 1. These two charts illustrate how much harder it is to meet Overall Coverage than only debt service coverage and how important long term TOT revenue growth is to meeting Overall Coverage. Chart 1A Maximum Initial Revenue Decline Debt Service Coverage at 1x Chart 1B Maximum Initial Revenue Decline Overall Coverage at 1x 33.3% 23.8% 14.9% 6.5% % 31.1% 40.4% 40.4% 40.4% 40.4% 40.4% 40.4% 40.4% 6.7% 12.9% 18.9% 26.3% Long Term TOT Growth Rates Long Term TOT Growth Rates II. Sensitivity #2: Interest Rates and Project Costs Impact on Debt Service Coverage and Overall Coverage at a 2% Long Term TOT Revenue Growth Rate. Question: How much can TOT revenues decline in year 1 assuming a constant 2% TOT revenue annual growth rate beginning in year 2 based on various Project Cost increases and interest rate levels? In addition to uncertain revenues, the other important uncertainties today are what the actual Project Cost will be and what interest rates will be when the time comes to issue the bonds. To analyze these assumptions, we held the annual TOT revenue growth starting in year 2 constant at 2%. We then calculated how much TOT revenues could decline in the first year and still achieve Debt Service Coverage and Overall Coverage under various interest rate and Project Cost scenarios. These scenarios reflect a situation where the City has to decide whether to proceed with a bond issuance at a later date when it has gotten a better sense than it has today of the actual Project Cost and projected interest rates. This analysis asks the question how much could revenues decline in year 1, assuming a 2% annual growth rate in TOT revenues beginning in year 2 based on different Project Costs and interest rates? We calculated the breakeven decrease or increase in revenues in year 1. For illustrative purposes, we have shown four cases below. The first two (2A and 2B) analyze Debt Service Coverage and Overall Coverage based on a specific increased Project Cost and different interest rates and the second two (2C and 2D) analyze debt service coverage and Overall Coverage based on a specific interest rate and different Project Costs.

11 Page 11 of 21 (Detailed Analysis) Scenarios 2A and 2B: Increased Project Cost at Different Interest Rates. Chart 2A assumes a 2 increase in Project Costs and analyzes the impact that interest rates between have on Debt Service Coverage. It calculates how much revenues could decrease or would actually have to increase in year 1 based on a 2% annual growth rate thereafter at different interest rates and still achieve Debt Service Coverage. For example, as you can see, to cover debt service at a 4. interest rate, the initial revenue decline could be about 2 but at a 5% interest rate, the decline would drop to 12.5% and by 6., initial revenues would have to be modestly higher. Next, Chart 2B also assumes the 2 increase in Project Costs and interest rates between and analyzes Overall Coverage assuming 2% annual TOT revenue growth thereafter. As you can see, this analysis has a very different result than in Chart 2A. Assuming 4., the first year s revenues would actually have to be 22% higher (they could not be lower) than the current forecast. As interest rates increase, first year revenues would need to increase further. For example, a 5% interest rate combined with the 2 Project Cost increase would require first year revenues to be about 3 higher to achieve Overall Coverage. We do note that in their scenarios, GS proposes that principal amortization could be extended beyond 30 years to compensate for higher interest rates. However, as we have shown, for more modest forecasts of long term TOT revenue growth (such as 2% per year), even at current record low interest rates of about 4., initial TOT revenues would have to be higher to fully fund all of the Initiative s requirements. Chart 2A Maximum Initial Revenue Decline Debt Service Coverage at 1x +2 Project Cost; 2% TOT Growth Year 2 and After Chart 2B Maximum Initial Revenue Decline Overall Coverage at 1x +2 Project Cost; 2% TOT Growth Year 2 and After 21.6% 24.5% 27.4% 30.4% 33.4% 36.5% 39.6% 42.8% % 17.9% 15.2% 12.5% 9.8% 6.9% 4.1% 1.2% 1.8% Interest Rates Interest Rates The two charts above also illustrate how much harder it is to meet Overall Coverage than just debt service coverage. We do note however, that higher annual TOT revenue growth rates would significantly improve the results. We have provided a scenario based on 4% TOT revenue growth in Sensitivity #3 below, which represents a proxy for the long term actual historic average of the City s TOT revenue growth rate. Scenarios 2C and 2D: Increased Interest Rate at Different Project Costs. Chart 2C assumes a higher interest rate of 5% (versus GS s assumption of 4.) and looks at the impact of different Project Cost increases between 0 and. It calculates how much revenues

12 Page 12 of 21 (Detailed Analysis) could decrease in year 1 assuming the 2% annual TOT revenue growth rate thereafter and a 5% interest rate at different Project Costs and achieve Debt Service Coverage. For example, at no change to the Project Cost, the initial decline could be about 34%; however, with about a 3 increase in Project Costs, TOT revenues could not be lower than the current projection for year 1 and still achieve Debt Service Coverage. Chart 2D. Increased Interest Rate at Different Project Costs (Overall Coverage). Chart 2D assumes the higher interest rate of 5% (versus Goldman s 4.) and looks at the impact of different Project Cost increases between 0 and. It calculates how much revenues could decrease or would have to increase in year 1 assuming the 2% annual TOT revenue growth rate thereafter and still achieve Overall Coverage, assuming no excess funds flow to the general fund. For example, as you can see, at 5%, even at the current Project Cost, revenues would have to be higher in year 1 to achieve Overall Coverage. 34.3% Chart 2C Maximum Initial Revenue Decline Debt Service Coverage at 1x 5% Interest Rate; 2% Growth Year 2 and After 23.4% 12.5% 1.6% 9.3% 20.2% 6.5% Chart 2D Maximum Initial Revenue Decline Overall Coverage at 1x 5% Interest Rate; 2% Growth Year 2 and After 18.5% 30.4% 42.3% 53.2% 63.4% Project Costs Project Costs Scenarios 2C and 2D can be run based on different annual TOT revenue growth rates (higher or lower than the 2% assumed here). One key observation is that low long term TOT revenue growth rates, as provided in the 2% scenario above, would have a material impact on the ability to fully fund all costs for even modest increases in the Project Cost or bond interest rate. In Sensitivity #3 below, we also provided a scenario that is based on a 4% TOT revenue growth rate which is a proxy for the City s long term historic TOT revenue average growth rate. III. Sensitivity #3: Interest Rates and Project Costs Impact on Overall Coverage at 3% and 4% Long Term TOT Revenue Growth Question: How much can TOT revenues decline in year 1 assuming higher TOT revenue annual growth beginning in year 2 based on various Project Cost increases and interest rate levels rather than the 2% TOT revenue growth rate assumed in Sensitivity #2? Scenarios 3A and 3B: Increased Project Cost at Different Interest Rates (Overall Coverage). Chart 3A assumes a 2 increase in Project Costs and interest rates between with a 3% TOT revenue growth rate in year 2 and thereafter. We analyzed Overall Coverage and it is

13 Page 13 of 21 (Detailed Analysis) analogous to Chart 2B. For example, assuming a 4. interest rate, the first year s revenues would need to be about 2% higher to achieve Overall Coverage. At a higher 5% interest rate, first year s revenues would need to be even higher, increasing by about 1 to achieve Overall Coverage. Chart 3B uses the same 2 increase in Project Cost and interest rate range as Chart 3A, but at a higher long term TOT revenue growth rate of 4%. As a result, assuming a 4. interest rate, revenues could decline by about 13% and achieve Overall Coverage. If the decline was less than about 13% with the 4. rate, excess revenues would flow to the general fund. Rates could increase to about 6%, and the financing structure could provide Overall Coverage. Chart 3A Maximum Initial Revenue Decline Overall Coverage at 1x Chart 3B Maximum Initial Revenue Decline Overall Coverage at 1x +2 Project Cost; 3% TOT Growth +2 Project Cost; 4% TOT Growth 2.1% 4.5% % % 17.2% 19.8% 22.5% 13.4% 11.7% % 6.4% 4.5% 2.7% 0.8% 1.1% Interest Rates Interest Rates Scenarios 3C and 3D: Increased Interest Rate at Different Project Costs (Overall Coverage). Chart 3C assumes the higher interest rate of 5% (versus GS 4.) and a TOT revenue growth rate of 3% and looks at the impact of different Project Cost increases between 0 and and is analogous to Chart 2D. It calculates how much revenues could decrease or would have to increase in year 1 assuming a 3% annual TOT revenue growth rate in year 2 and thereafter and a 5% interest rate at different Project Costs and still achieve Overall Coverage, assuming no excess funds flow to the general fund. For example, at a 2 higher Project Cost, a 5% all in interest rate and a 3% TOT revenue growth, revenues would need to be about 1 higher to achieve Overall Coverage. At a 4. interest rate, the breakeven Project Cost could be about 1 higher and the financing structure could provide Overall Coverage based on current TOT revenues. Chart 3D is similar to Chart 3C but assumes higher TOT revenue growth of 4%. In this scenario, TOT revenues could decline by about 8% in year 1 and achieve Overall Coverage. If the decline was less than about 8% at a 5% rate, excess revenues would flow to the general fund. At a 5% interest rate, the breakeven Project Cost could be by about 3 higher and the financing structure could provide Overall Coverage.

14 Page 14 of 21 (Detailed Analysis) Chart 3C Maximum Initial Revenue Decline Overall Coverage at 1x Chart 3D Maximum Initial Revenue Decline Overall Coverage at 1x 5% Interest Rate; 3% TOT Growth 5% Interest Rate; 4% TOT Growth 9.5% 19.4% 29.4% 39.4% 6.1% 13.2% 10.5% 0.5% 22.4% 15.3% 8.2% 1.1% Project Costs Project Costs As you can see, with a 4% TOT revenue growth rate assumption beginning in year 2, the Initiative parameters perform significantly better. However, the actual Project Cost and bond interest rate at the time of sale will also be significant determinants of the ability to meet the annual flow of funds in the Initiative. IV. Sensitivity #4: Timing of Revenue Decline Impact on Overall Coverage at 4% Long Term TOT Revenue Growth Rate. Question: How would modifying the timing of the TOT revenue decline affect the results shown in earlier sensitivity analyses? Scenario 4: Increased Project Cost and Interest Rate with Different Timing of TOT Revenue Decline (Overall Coverage) To address the issue of the timing of the assumed revenue decline, we analyzed the changes in results for one specific case in Scenarios 3B and 3D with different timing for the one time TOT revenue decline. For this purpose, we have selected the case where there would be 2 of additional Project Cost and a 5% interest rate. In addition, we assumed a 4% annual increase in TOT revenues in the intervening years prior to the one time decline, as well as 4% TOT revenue growth after the decline. This choice is again deliberate and intended to reflect the City s historical long term TOT revenue performance. The analysis then focuses on both achieving Debt Service Coverage only and achieving Overall Coverage.

15 Page 15 of 21 (Detailed Analysis) Chart 4A Maximum One Time Revenue Decline Debt Service Coverage at 1x +2 Project Cost; 5% Interest Rate 4% TOT Growth All Other Years Chart 4B Maximum One Time Revenue Decline Overall Coverage at 1x +2 Project Cost; 5% Interest Rate; 4% TOT Growth All Other Years 12.5% % 19.9% 22.2% 26.6% 7.1% 7.3% 7.6% 7.8% % 9.7% 32.8% Year of Decline Year of Decline Longer delays in the timing of the decline in TOT revenues produce better results as long as TOT revenue growth was healthy leading up to the decline. As you can see in Chart 4A, a change in the timing of the assumed revenue decline has a very material impact on the ability of the structure to withstand declines in revenues and still achieve Debt Service Coverage, but as you can see on Chart 4B, a much smaller impact on the ability to do so with respect to Overall Coverage. This reflects the fact that debt service depends on each year s annual revenues, whereas Overall Coverage depends on cumulative revenues over a longer period of time. Overall, the improvement in aggregate cash flow for delayed declines in revenues is limited, and the Project would still be vulnerable to interest rate and Project Cost increases above the assumptions used in GS analysis. Results, Assumptions and Other Considerations. The discussion below analyzes these results, considers the reasonableness of GS interest rate and revenue growth rate assumptions and discusses other possible other considerations. Analyzing the Results. As discussed above, the results of the analysis will depend heavily on the combination of assumptions about i) Project Cost, ii) future TOT revenues and iii) bond interest rates. At this time, there is significant uncertainty as it relates to each of these variables. Together, this is a three dimensional risk factor because ultimately, it will be the actual levels of each of these three variables over time that will determine whether the TOT revenues can initially fund the Project and then fund all future requirements that must be met before the City receives residual TOT revenues. As shown above, there can be many scenarios where Overall Coverage would not be achieved based on different assumptions for the three variables. It is possible to construct an endless number of different combinations of factors to demonstrate such results. To focus in on each factor, we have determined breakeven levels for one variable holding the other variables constant and highlighted two straightforward approaches to looking at the results for any scenario. The first approach is to know how large a change in one assumption (i.e., Project Cost or interest rate) could be until Overall Coverage is no longer achievable, and the second is to know how much of a decline or increase in TOT revenues be needed before the City s general fund would receive any residual TOT revenues over the life of the bonds (i.e., 30 years). At the top of the following page are selected examples of our analysis above that focus on Overall Coverage, looking at the results using the two approaches just described.

16 Page 16 of 21 (Detailed Analysis) Scenario 1B: For + Project Cost and 4. interest rate, long term TOT revenues must grow at least 2% per year, or revenues could decline 26.3% in year 1 if long term revenue growth were 4% per year. Scenario 2B: For +2 Project Cost and 2% long term TOT revenue growth, interest rates must be less than 4., or revenues would need to increase 30.4% in year 1 at an interest rate of 5%. Scenario 2D: For a 5% interest rate and 2% long term TOT revenue growth, Project Cost must be less than the current estimate, or revenues would need to increase 6.5% in year 1 at the current Project Cost estimate. Scenario 3A: For +2 Project Cost and 3% long term TOT revenue growth, interest rates must be less than 4., or revenues would need to increase 9.5% in year 1 at an interest rate of 5%. Scenario 3B: For +2 Project Cost and 4% long term TOT revenue growth, interest rates could be 6%, or revenues could decline 8.2% in year 1 at an interest rate of 5%. Scenario 3C: For a 5% interest rate and 3% long term TOT revenue growth, the Project Cost could be increased by 1, or revenues could decline 10.5% in year 1 at the current Project Cost estimate. Scenario 3D: For a 5% interest rate and 4% long term TOT revenue growth, Project Cost could be increased by 3, or revenues could decline 22.4% in year 1 at the current Project Cost estimate. Interest Rate and Revenue Growth Assumptions. 4. Interest Rate Assumption. Based on current low interest rates and the results of other TOTbacked revenue bond transactions, a 30 year all tax exempt financing with the GS structure and A level ratings could reasonably be expected to have an aggregate interest of 4. or less today the rate GS assumed in their base cases. However, to the best of our knowledge, there have not been any billion dollar TOT backed revenue bond transactions in the US, and a transaction of this size may well carry a size penalty. In addition, a clear risk is where general municipal market interest rates would be when the City was ready to bring the transaction to market. In addition, if all or a portion of the Project financing was required to be taxable, the interest rate increase could add significant cost to the financing. Rather than trying to guess at future interest rates, we used historic data to provide a framework for measuring potential changes in future interest rates. Our analysis overlays current estimated spreads for an A rated tax exempt TOT backed revenue bond financing onto the industry benchmark of the AAA municipal market data (MMD) indices, weighted based on the GS structure to calculate the historic aggregate interest cost for such a financing. The result is a series of rates that could be considered proxies for the interest cost of the GS structure at different points in time in the past. The table at the top of the following page looks at the percent of time over the last 5, 10 and 20 year periods through May 24, 2016, as well as since 1990, that the calculated rates in the series would have been below certain levels namely, 4., 5% and 6.. In addition, it provides the median rate for each time period.

17 Page 17 of 21 (Detailed Analysis) Percentile Rank Median Rate 5 year history 52% % 10 year history 26% 53% 98% 4.96% 20 year history 13% 26% % Since % As shown in the table above, 4. would be equal to or higher than the calculated rates 52% of the time during the last five years. In other words, 4. is very close to the median rate over this period. For longer periods of analysis, such as over the past 10 years, 4. would be equal to or higher than only 26% of the calculated rates. A 5% interest rate would be a more conservative estimate of future rates based on the last five years capturing more than 9 of the calculated rates over the past five years and 53% of the rates over the last 10 years. Given these results, PRAG would recommend that a base case scenario have a higher rate than 4. for any preliminary analyses. In addition, we note that if taxable bonds were required to be issued to fund some portion of the Project Cost, the interest cost could also increase. The extent of such an increase will depend on the proportion of tax exempt versus taxable bonds to be issued, as well as the amortization pattern of each type of bond. Furthermore, to the extent that the amortization of the bonds must be extended to increase capacity to finance additional Project Costs or to compensate for other financial stresses, the interest rate could reasonably be expected to increase as well. Five Year Forecast Followed by 4% Annual Growth Rate. As highlighted in the analysis above, (Charts 1A and 1B) the risk of lower TOT revenue growth than projected is much greater on Overall than on just Debt Service Coverage. Given the Initiative s flow of funds, while the rating agencies would certainly use more conservative TOT revenue projections in analyzing Debt Service Coverage, they would likely not be concerned about the ability of the revenues to pay debt service. We note that the rating agencies could be provided with more conservative revenue projections than the scenarios presented to demonstrate residuals revenues flowing to the City. However, to the extent that the numbers in the Initiative accurately reflect the City s future O&M and CapEx costs and they run scenarios that test Overall Coverage, the rating agencies may be concerned that these additional costs may put expenditure pressure on the City s general fund. Other Fiscal Considerations and Risks. As noted in the Executive Summary, in addition to Cash Flow Considerations and Risks there are other considerations that we think are significant. Alternative Bond Structures. We would suggest that the City be aware when reviewing GS stress scenarios that bond terms are modified to be able to continue to meet the Initiative s objectives (e.g., higher interest rate or Project Cost, longer bond term). While PRAG is aware of a number TOT backed revenue bonds that have been issued or restructured as 40 year bonds from original issuance, that is considered a relatively long bond term. In addition, while it may be possible to address for example, cost overruns or higher interest rates by extending the bond term or with a delayed financing, which would result in additional upfront TOTs, it would always be at the expense of public funds. We note that it would also be important for the lease with the Chargers to be at least as long as the bond term. Otherwise, in the event the Chargers leave the stadium at the end of the lease