Blog

Weekly Preliminary OS Climate Risk Review (10/14/21)

By October 26, 2021 No Comments

The object of our non-disclosure frustration this week is another infrastructure operator with apparently Nostradamus-level abilities to make multi-decadal financial forecasts, while rejecting the notion of it being possible to project climate risk.

This week’s non-disclosure deadweight cargo:

Georgia Ports Authority Revenue Bonds, Series 2021, $420,150,000

The Bonds

Georgia Ports Authority (GPA) is issuing the Bonds primarily in order to finance four capital improvement projects at the Port of Savannah. Three of the projects are taking place at the Garden City Terminal (risQ Score 3.0), two of which will expand the capacity of the Terminal which has the highest commercial marine vehicle traffic in Georgia with over 1,700 vessel arrivals in fiscal year 2021. The 7 terminals include Garden City Terminal and Ocean Terminal at the Port of Savannah, Colonel’s Island Terminal, Mayor’s Point Terminal, and East River Terminal at the Port of Brunswick, as well as Appalachian Regional Port and Port of Bainbridge. The Bonds are special limited obligations of GPA and secured and payable from Pledged Revenues which primarily comprise net revenues generated from the operations of the terminals. Approximately $156 million of the debt is currently scheduled to come due after 2030.

GPA’s crystal ball doesn’t predict climate risk

The Authority makes it clear how difficult it is to make predictions when referring to various credit risk factors related to labor negotiations, COVID-19, port competition, and of course the impacts of sea level rise: “The Authority cannot predict,” “difficult or impossible to predict,” “predictions…are not possible at this time,” “the Authority is unable to predict,” and so on. Okay, we get it, predictions aren’t easy. In fact, it would be borderline irresponsible for any municipal issuer to claim it has the ability to “predict” anything at all. That being said, in lieu of having access to a crystal ball, the Georgia Port Authority could at least attempt to provide some quantitative insights into the impacts that climate change and sea level rise could have on its terminal locations and revenues. Maybe the Authority just feels uncomfortable communicating decadal projections of climate risk and sea level rise, even if those projections are sourced from climate science domain experts…?

The issuer’s conservative approach to forecasting future events doesn’t appear to apply to its own financial projections. On page 34 of its preliminary official statement (POS), GPA provides projections for pledged revenues out to the year 2032. Various financial metrics including operating revenues, expenses, income, pledged revenues and debt service coverage are forecasted down to the thousands-digit over the next 10 years. That is some impressive level of precision…

The dichotomy between GPA’s aversion to attempt any projection of climate risk vs. its Nostradamus-level ability to forecast revenue beyond a decade into the future is…eyebrow-raising. The section subsequent the Summary of Projected Pledged Revenues provides an overview of a Commercial Analysis for Bond Financing report commissioned in September to Mercator, an advisory firm for the global transportation sector. The entire 100-page midsection of the GPA POS is a copy of Mercator’s assessment, providing “assumptions and forecasts about the Authority’s market position…over the next 30 years” (POS pg. 34). Some highlights:

  • Authority’s revenues are forecasted to increase at a Compound Annual Growth Rate (“CAGR”) of 5.3% over the 30-year forecast period, generating just over $2.8 billion in FY 2052.
  • The Authority’s expenses are forecasted to increase at a CAGR of 4.9% over the 30-year forecast period.
  • Earnings before interest, taxes, depreciation and amortization are forecasted to increase at a CAGR of 5.8% over the 30-year forecast period
  • Actual results likely will differ, perhaps materially so, from those forecasts. Accordingly, the forecasts contained in the Market Assessment are not necessarily indicative of future performance, and neither Mercator nor the Authority assumes any responsibility for the failure to meet such forecasts. In addition, certain assumptions with respect to future business and financing decisions of the Authority are subject to change. If actual results are less favorable than the results forecasted, or if the assumptions used in preparing such forecasts prove to be incorrect, the amount of Net Revenues realized may be materially less than forecasted.

That final bullet serves as an accountability disclaimer off-ramp, and is really just a verbose version of: “30-year financial projections are difficult or impossible to predict.” GPA could have taken at least >0 time to quantify and disclose its worsening climate risk, and then could have appended the same exact disclaimer language as it did for its 2052 revenue projections; instead, the Authority left that job to us.

Digging into the Authority’s boilerplate climate risk disclosure —

Climate Change and Storms

The Authority is unable to predict whether a rise of sea-level or any other impacts of climate change will occur while the 2021 Bonds are outstanding or, if any such events occur, whether there will be an adverse impact, material or otherwise, on the Port Facilities or the Gross Revenues. In addition, the State is susceptible to hurricanes and similar storms in which winds and tidal surges can be powerful enough to cause severe destruction. The Authority’s facilities have suffered damage occasionally from previous hurricanes and tropical storms that have impacted the State; however, such storms have not had a material effect on the operations or revenues of the Authority. The Authority has adopted a hurricane response plan that, among other things, establishes protective measures designed to make its facilities safer in the event of a hurricane. In the event it is necessary for ships or cargo to be directed to other ports due to hurricane or other weather-related damages, such damage or interruption could materially adversely affect the Gross Revenues of the Authority. The Authority expects to be able to manage its operations and activities, including any new or ongoing construction projects, in a manner intended to minimize the potential future impacts of these types of occurrences; however, no assurance can be given that climate change, rising sea levels, storms, severe droughts, or other weather events will not adversely affect the Gross Revenues of the Authority.

The “hurricane response plan” that the Authority references is embedded in its Severe Weather Plan released in July 2020, a 24-page document that covers hurricane, tornado, high wind, lightning, ice and snow perils. The Plan provides recommendations on preparedness, response and recovery strategies with lead times of 12-120 hours prior to an event taking place. While it serves as useful procedural documentation, the Plan provides no mention of the need for climate adaptation or resilience. It appears GPA does not have a climate action plan and is not currently considering the risks of long term climate change in any of its infrastructure designing or planning whatsoever.

The climate risk disclosure omission and adaptation negligence is about as pronounced as the Authority’s climate risk. Below we have a view of each of the 7 terminals’ climate risk in the year 2031 within a 6-minute drive time band. We use the 6-minute radius as it gives us a facility-specific view of operational risk for port terminals. risQ Scores for the Port sector are calculated with an even weighting attributed to GDP impairment and property value-at-risk. GDP impairment is calculated by considering the downtime / rebuilding time of a facility in the wake of a climate event, and is provided as a percentage of the total annual GDP produced within a given location. Loss of business activity and impacts to GDP health (e.g. freight-based revenues), as well as physical impacts to property will flow into port authority revenue streams, regardless of the type of cargo that is entering and exiting each terminal, whether it is container cargo (e.g. televisions, meats, computers), liquid bulk (e.g. petrol, vegetable oils), dry bulk (e.g. coal, sugar, cement), breakbulk (e.g. steel girders, equipment) or roll-on/roll-off (e.g. shiny new red convertibles).

Bainbridge and Appalachian Regional terminals are both located inland, with Bainbridge bordering Florida and Appalachian Regional located about 300 miles due north at the other end of Georgia, bordering North Carolina. While Appalachian is positioned far enough inland to benefit from insulation from material hurricane or flood risk, Bainbridge’s Scores are a bit higher given its exposure to hurricane tracks traveling up Florida’s western Gulf Coast. Flooding from a hurricane of any magnitude is expected to cause 16% GDP impairment within Bainbridge’s 6-minute drive time band. The coastal terminals have materially higher climate risk than both inland terminals, with the second highest-traffic Colonel’s Island Terminal (Flood risQ Score 4.9) at the Port of Brunswick appearing to have the highest exposure. Colonel’s Island terminal is projected to experience 17% property losses and 30% GDP impairment from coastal flooding events alone by 2031, ranking it in 94th and 97th percentiles nationally within the port cohort. Those value-at-risk metrics for Colonel’s pop off the page when one drags the Year slider in our UI out to GPA’s favorite time horizon of 2052 — total GDP impairment within Colonel’s 6-minute drive time band is projected at 154% by 2052, driven largely by coastal flooding, hurricane flooding and storm surge, and inland flooding events. (*any value at risk percentage above 100% means that risQ projects the location could sustain cumulative losses above 100% of current property or GDP value between now and the specified time horizon).

Georgia’s coastal pocket of wildfire risk

Hurricanes, inland and coastal flooding events are the main climate culprits for GPA losses; however, the Authority’s wildfire risk is what really caught our eye. The map of Georgia above shows that it’s the counties along the Atlantic coast that have the most exposure to wildfire events, in particular the low probability / high severity wildfire events that can effect losses at an unprecedented scale. Three of the Authority’s ports are near coastal Brunswick in Glynn County. Glynn’s western half is largely forested, and according to the County’s Wildfire Mitigation Plan, the risk of wildfire along the wildland-urban interface is present throughout the County including on the edge of Brunswick. Between 1998-2018, Glynn County averaged 45 wildland fires per year, which burned an average of 217 acres annually (pg. 8 of the Glynn County Plan). This fire risk is reflected in the 6-minute drive time band Wildfire risQ Scores for the Brunswick-based ports: East River is 3.5, Mayor Point is 3.5 and Colonels Island is 3.4. While a 100-year event isn’t expected to impact these terminal locations, a rare 500-year event could have potentially devastating consequences, leading to property losses >50% and GDP impairment >29% at each of the three wildfire-prone terminals.

Carbon transition risk kicker

Five of the Port Authority’s Top Ten Vessel and Cargo Customers (pg. 28) — ONE, Maersk Inc, Mediterranean Shipping, Hapag Lloyd, and Zim American Shipping — have signed on to the Getting to Zero Coalition, a pledge by maritime stakeholders to develop and deploy zero emission vessels by 2030 and cut emissions from the global shipping sector by one-half by 2050. Based on the NASA’s estimates of CO2 emissions viewable in the risQ UI, the Colonels Island, East River, Garden City, Mayors Point, and Ocean Terminal facilities’ maritime operations produce an estimated aggregate of 950,000 metric tons of emissions per year, yet there’s absolutely zero mention of carbon emissions in the GPA POS. To best align with the needs of the cargo vessels they’re serving, especially the ones with visionary goals, the Port Authority will have to think innovatively about modifications they can make to their operations to better accommodate the needs of cargo ships. Along these lines, GPA might be able to learn from the Port of Miami which implemented a sustainability project which upgraded the electricity capacity available for docked vessels, resulting in zero-fuel-emissions when in port.

It is impossible to make perfect predictions. That being said, we were able to disclose Georgia Ports Authority’s climate risk in < 8 hours more comprehensively than the Authority apparently has done in its entire existence, let alone in its 344 page preliminary official statement. But how? Our secret sauce: risQ doesn’t make predictions. We project what *could* and what is *most likely* to happen, rather than what is absolutely going to happen. Word of advice for GPA’s next issuance: make at least some effort to disclose risk, and throw some error bands around your projections.

Honorable Mentions

San Francisco Tax District No. 2020-1 (Mission Rock), CA (risQ Score 3.5, Flood risQ Score 5.0) $64,900,000

Mission Rock is flanked by San Francisco Bay, on part of the lowest-lying areas of Mission Bay. Currently, Mission Rock is a parking lot, but these bonds partially finance the transformation of ~1 million square feet of land into office, retail, and residential units (pg. 2). Unfortunately, Mission Rock sits in the 97th percentile for property VaR, with a 100-year flood driving 33% of property losses. The POS isn’t shy about disclosing this risk, dedicating 2.5 pages to climate change, sea level rise, and flooding. The POS notes that flood events may be exacerbated by climate change and sea level rise (which have risen 9 inches since the mid 19th century) (pg. 79). Additionally, much of the Bay is built on fill, meaning that it is susceptible to subsidence and liquefaction (pg. 80). These risks could result in the loss of tax revenue, and the displacement of residents and businesses (pg. 79). 

In anticipation, the City has released a Sea Level Rise Vulnerability and Consequences Assessment, a Sea Level Rise Action Plan, and a Regional Sea Level Rise Vulnerability and Adaptation Study. The POS notes that the city has incorporated “site specific adaptation plans” at Mission Rock, which appears to be limited to raising the grade of the land 4 feet. The City’s $400+ million, 3 mile Embarcadero Seawall along the Bay coast (pg. 80) unfortunately doesn’t cover Mission Rock

Although the area has a low Social Impact Score (1.4), Mission Rock has a Minority/Nonwhite Score of 92 and a Percent of Income Spent on Housing Score of 76. 40% of the residential units at Mission Rock will be affordable (pg. 2). San Francisco has made strides to make the city resilient, let’s hope the elevation of Mission Rock will prove sufficient.

Penn Highland Healthcare, PA (risQ Score 1.5, Flood risQ Score 3.2) $43,000,000

Pennsylvania probably isn’t the first place that comes to mind when you think of flood risk, but indeed north-central PA has a concentration of it. Jefferson and Clearfield Counties are part of the System’s primary service area (pg. A-9) and they have flood risk — past events provide anecdotal examples, such as a 2013 storm which flooded homes and businesses in Jefferson and Clearfield Counties. Penn Highland Healthcare is issuing this series to acquire the Monongahela Valley facility, and since the POS is completely absent of any mention of flood risk in their hospital service areas, we take a closer look at how the risQ plays out across their portfolio of hospitals.

3 of Penn Highland’s 7 hospitals have Flood risQ Scores ranging from 3.53.7 at the 30-min radius, while the other four hospitals are below 3.2 Flood Score. Three facilities rank above the 88th percentile nationally for GDP impairment from combined perils. Hospital revenues in part come from elective procedures, and elective services are in part driven by patients’ discretionary income. Climate-related events have the potential to disrupt discretionary income when funds must be diverted for property damage repairs or diminished by instability in the local economy. Therefore, damages from severe weather have a bearing on hospitals’ revenue impairment risk. The cumulative expected GDP impairment out to the maturity date of 2041 in the 30-minute service area for the top-three risQiest hospitals is as follows: Clearfield 27.5%, DuBois 26.3%, and Huntingdon 25%. Beyond climate-related impairment, if a carbon tax is implemented and energy use becomes more expensive, discretionary income could also be impacted. All of the hospitals’ 30-minute service areas rank between the 88th – 95th percentile in per capita emissions from the electricity sector, which means they are carrying relatively high liabilities in regards to Carbon Transition Risk

When evaluating hospital systems, it’s worth considering the communities they serve and the demographic profiles of the residents. Looking at the Monongahela Valley, the area within 30-minutes of the hospital, ranks in the 90th percentile for health insurance coverage – with 95% covered. The Huntingdon facility has the highest Social Impact Score among the Penn cohort (60) and its service area has widespread health insurance coverage with 92% insured, while at the same time ranking in the ~70th percentile nationally for percentage of population with pulmonary disease, cancer, and heart disease.

Overall, Penn Highland serves communities that have relatively good access to the health care system, but at the same time 3 of 7 of their hospitals’ service areas are exposed to flooding risk that could impair patients’ discretionary incomes. 

Savannah State University, GA (risQ Score 3.0, Flood risQ Score 3.7) $36,170,000

$36 million for building student housing facilities for Savannah State University (SSU), the oldest historically Black public university (HBCU) in Georgia, upholding rich traditions deeply rooted in American history. However, what this particular POS fails to uphold is its duty to disclose the climate and socioeconomic risks that have plagued the area since inception and will continue to do so well beyond the bonds’ maturity date in 2041. 

SSU (Nonwhite/Minority Population Score 90.8) has ~3,700 students and employs 700 administrative staff members. It sits 5 miles from the center of Savannah, a city known for both triumph and tragedy. 16 percent of the city’s population lives below the poverty line, which has recently fallen from 24% due to anti-poverty programs and job training initiatives. Unfortunately, the story doesn’t end there. Savannah’s population is 33% Black and lacks access to healthcare (Persistent Health Obstacles Score 90.7). Black homeownership is below 42%, with majority Black neighborhoods valued 23 percent less than majority White neighborhoods (Percent of Income Spent on Housing Score 91). This equates to more than $156 billion in losses.

This brings us to climate. Don’t bother searching for climate or flood in all of the 268-page POS. In fact, there is only one instance of storm and hurricane, both located in a single boilerplate sentence on page 240 relating to property damage. Within a 20-minute drive time radius, SSU has a Total risQ Score of 3.0 and a Flood risQ Score of 3.7. With a large majority of the campus located on marshland, this area is ranked 96th percentile nationally for property losses from physical risk. Storm surge from a Cat 2 event projects ~18% property losses. In 2017, Tropical Storm Irma brought a historic tidal surge to the area. By 2041, these numbers are likely to increase along with perils. Carbon Transition Risk is also a great concern—92nd percentile in per capita emissions from electricity production. SSU’s motto is “Lux Et Veritas.” Light and Truth. In light of the climate and socioeconomic risks presented here, which are not mentioned in the POS, the issuer needs to disclose the truth; there needs to be strong climate adaptation/mitigation strategies in place before $36 million is used to build in and around highly flood-prone marshland.

Leave a Reply