There is good, bad and ugly in terms of climate risk disclosure by municipal bond issuers and its fair to say the distribution skews to the dark side of that spectrum. This week offers up a chance to compare Preliminary Official Statements from two counties, Harris County, TX (via $437M in toll revenue bonds) and Suffolk County, NY (via $410M in tax anticipation notes) posted on MuniOS on December 8th and December 7th, respectively.
Before we pop the hood on the POS documents, let’s lay the risQ metric foundations. Most will be aware of Harris County’s recent history, especially with Hurricane Harvey. Accordingly, the county is 90th percentile nationally for Property VaR (16% cumulative VaR to 2030), with the main risk components being inland flooding and hurricane flooding in a 2:1 ratio, and primarily with respect to residential property. The GDP Impairment Risk is only 55th percentile nationally, dominated by hurricane flood risk (i.e. extreme precipitation akin to Hurricane Harvey). It will come as a surprise to many that Suffolk County, NY has higher climate risk being 93rd percentile for Property VaR (20% cumulative VaR to 2030) and 88th percentile for GDP Impairment Risk. In this case the key perils are a 1:1 mix of coastal and inland flood for Property VaR, while inland flood dominates the GDP risk.
Per our demographic data, note that with 3 times the residential population, the absolute Property VaR in dollars will be higher for Harris County (even accounting for a 2x higher median property value in Suffolk), and the GDP risk in Harris County will certainly be higher in absolute terms given the relative workforce sizes. Its impact relative to the local economy that’s important though when it comes to the capability and timeline for recovery. On that measure, Suffolk County has more systemic climate risk mitigation issues to face.
|Median House Value||Residential Population||Worker Population|
Now for the Preliminary Official Statements:
Harris County: The document has solid accounting for Hurricane Harvey on page 47. This includes discussion of specific damage costs to the toll road system as well as revenue impacts in the aftermath. The same paragraph also openly acknowledges that FEMA did not cover those costs. This is a more common occurrence than many realize, in part because most issuers fail to disclose this fact. Full credit to Harris in this case. The next two paragraphs discuss weather events and flooding in further detail. Included are specific references to engagement in the FEMA NFIP program by the county and many municipalities therein, enabling residents to get subsidized flood insurance. Indeed, we can confirm, for example, that the City of Houston has a 5 Rating in the NFIP Community Rating Survey, one of only 3 cities in Texas with a rating that high, and requiring genuine flood mitigation efforts to be under way. Harris County is also listed (7 Rating). In the subsequent paragraph, there is acknowledgment that flood plain designations may change and this may cause a loss of value for property that falls within newly created boundaries. All-in-all, “flood” is mentioned no less than 23 times in the document.
Suffolk County: Don’t worry, this won’t take long. There’s a single five line paragraph on page 6 covering environmental factors in distinctly uninformative boilerplate language. In two distinct place in Appendix A a Coastal Resiliency Initiative is discussed, but this is limited to efforts to connect 6,500 cesspool to sewer infrastructure, which is worthy but nowhere near enough in terms of aggregate risk. There is no other discussion of the systemic vulnerabilities to property overall, and no reference to inland flood risk that is just as prevalent as coastal flooding. We went looking for NFIP CRS registrations on our own. Nothing for Suffolk, and only Freeport showed up as being active (a 7 rating) among the towns and boroughs. That’s weak. In case anyone is thinking that this is just a specific instance associate with a single year tax anticipation note OS and the serial bond issuances with out-year maturities are better, we checked…they’re not.
You could reasonably argue that Harvey’s impact has brought flooding issues front and center for Harris County and actions that have followed. Over 3,100 properties have been bought out by FEMA (retiring them from the ad valorem tax base), but this is actually an indication of willingness to engage in flood mitigation. NFIP flood claims/capita/year are a remarkable $55,000, in the worst 2% of counties nationally, but the CRS efforts both in mitigating risk as well as generating discounts for residents are evidence of higher level ESG thinking. In contrast, a paltry 10 properties have been bought out in Suffolk County, but the residents have gladly sucked up over $24,000 per capita/year in NFIP claims – still good enough to be in the worst 3% of counties nationally – while not enjoying the insurance discounts that residents in Harris can enjoy. Suffolk might not have had the front-page, direct hit event that Harvey represented for Harris, but the financial flood data doesn’t lie.
So in the big picture, you have an issuer with significant climate risk that discloses that risk in concrete ways but then points to efforts to mitigate it. On the other hand, you have an issuer with comparable – if not higher – climate risk that doesn’t. Apply your ESG filter and take your pick.