"Excluding" Partner Funds Could Lead to Shortfalls in Flood-Mitigation Funding

The new Harris County Administrator has proposed “excluding” potential partner funds from consideration in the allocation of flood-bond and flood-resilience trust dollars. There’s only one problem with that. Without partner funds, there won’t nearly be enough money to cover all the proposed flood-mitigation projects in the bond, the trust, or future bonds.

The conclusions above come from comparing projected project totals in the 2018 flood bond with the December 2021 flood-bond update that Harris County Flood Control District (HCFCD) provided to County Commissioners. The exercise also illustrates why people who are TRULY interested in making informed decisions about HCFCD’s Bond Program need to dig deep into the data.

Making conclusions based on hearsay or a glance at a chart could be self-defeating. There is more to the story.

Partner Funds Make It All Possible

We anticipated 43% of the dollars in the flood bond would come from partners such as FEMA, HUD and the TWDB. We also anticipated it would take another 16% in local matching funds to attract the 43%. So 59% of the flood bond revolved around partner funds. Only 41% was local cash to pay for projects totally out of pocket.

*From the project spreadsheet approved by voters in 2018.*

Excluding Partner Funds Could Accelerate Construction in LMI Neighborhoods, Deny Others

David Berry, the County Administrator, proposed the partnership exclusion to accelerate construction of projects in Low-to-Moderate Income (LMI) watersheds, such as Halls. Halls, in particular, has waited on grant awards from the US Department of Housing and Urban Development (HUD) longer than most.

A HUD decision is expected sometime in January, according to the Texas General Land Office, which distributes HUD grants in Texas. So it’s not clear how much residents gain by Berry’s proposal. And they could lose big.

Most of the HUD grant applications for Halls are on a 90:10 basis, meaning the local share is only 10%. So excluding these grants means increasing the local contribution for that portion of the budget by 9X. That could cost local taxpayers hundreds of millions of dollars. For instance, HCFCD budgeted $500 million for Halls drainage alone. 90% of that is $450 million…to cover 2.4% of the county!

The effect would be to take money from affluent watersheds – which don’t qualify for HUD 90:10 grants – and shift it to LMI watersheds. No one then would get the grants and something would have to give somewhere down the road.

One Third of Way Through Flood Bond: Good Time to Take Stock

At the end of this month, exactly one third (3 years, 4 months) of the 10-year flood bond will have expired. So this is a good time to review spending versus projections.

Thirty-three percent of the way in, we’ve expended a little more than 16% of the flood-bond funds. While that may sound like a slow start, one must consider project lifecycles. Projects start with studies (feasibility, preliminary engineering, final engineering, design). These determine and validate cost projections. They also form the basis for grant applications, a plan and bids. But they are the least expensive part of a project. Together, they comprise only one eighth of project costs.

The expensive parts follow. They include right of way acquisition and construction. Those comprise more than three-quarters of all project costs. See the pie chart below which shows averages for the last two decades.

*Average percent of costs in various project stages since 2000. ROW (Right of Way) Acquisition includes purchase of land upon which projects will be built.*

For most flood-bond projects, we’re just now getting to the expensive phases. So I wouldn’t worry too much about that 16% overall average right now.

Spent/Unspent Funds by Watershed Gives Greater Insight

You can gain more insight by looking at spent and unspent dollars in each watershed.

*Height of bars shows total amount budgeted per watershed in flood bond. Blue areas show dollars spent to date. Does not include any funds spent prior to flood bond.*

From the charts above and below, you can see that spending rates vary widely among watersheds. Brays has consumed 57% of its budget already. On the opposite end of the spectrum, Sims has consumed just 1.6% of its. Why the wide variation?

To understand, we need to look at unique circumstances in each watershed. The chart below makes it easier to see actual spending as a percent of the budget that voters approved for each watershed in the flood bond.

*Percentages represent the portion of budget spent to date. See discussion below for explanations of ranks.*

The height of some of the bars above could be “predicted” by referring to the flood bond equity prioritization framework. Brays and Greens Bayous, for instance, are two watersheds with high percentages of LMI residents (58% and 57% respectively).

But others cannot. More than half the residents in Goose Creek/Spring Gully, Hunting, White Oak, Halls, Vince and Sims Watersheds also qualify as LMI. But dollars spent to date in those watersheds are far lower as a percent of the total budget. To see why, you need to put the numbers in a bigger context that includes:

Investment Prior to Flood Bond
Size of Total Budget for Each Watershed
Percentage of Partner Funds
Grant Application Status
Stage of Project Lifecycle

Bigger Context Shows Reasons for Variance

Brays and Greens had a large number of shovel-ready projects that had already been studied and approved when the flood bond passed. They were just waiting for dollars to become available. So they had head starts.

Other factors explain LMI watersheds further down the curve:

Sims received $380 million in federal funding for 23 Army Corps projects that finished construction by 2015. As a result, Sims was the only bayou in Harris County that stayed within its banks during Harvey. None of that spending shows up in the Bond Program charts. Because it’s already done!
White Oak received full funding in the Bi-Partisan Budget Act of 2018. The Army Corps also started addressing many projects there before the flood bond.
Vince lies wholly within the City of Pasadena and is primarily the City’s responsibility.
Halls is the poorest watershed (71% LMI) and has only spent 6% of its projected budget to date. Lest you attribute this to racism, understand that the bond allocated more than HALF A BILLION DOLLARS to Halls. In percentage terms, the $29 million dollars spent to date looks small. But in absolute dollars, it outranks 15 other watersheds.
Far more affluent watersheds – such as Buffalo Bayou, Cedar Bayou, the San Jacinto River, Barker, Willow Creek, Armand Bayou, Galveston, Luce Bayou and Jackson Bayou – have each received fewer dollars from the flood bond than Halls.

HCFCD had just finished a watershed plan for Halls Bayou in 2018 when the flood bond passed. That explains the size of the watershed budget as well as the late start compared to Brays, White Oak, Greens, and Sims.

Other factors also explain affluent watersheds further up the spending curve, such as Little Cypress. HCFCD started working on that watershed long before the flood bond, too. Dollars spent to date on Little Cypress primarily reflect right-of-way acquisition costs, not construction. It’s also important to understand that the total budget for Little Cypress is only 37% of the total budget for Halls.

Creating a Win-Win For Everyone

In another three years and four months, these charts will look totally different than the ones you see today. Construction costs will surge for some and be long gone in the rear view mirror for others.

In my opinion, we need to stop creating chaos with endless tinkering in the bond program. The people have spoken. Leaders should listen. Let’s stop changing the allocation formula, focus on construction, and work like hell to win those grants. Then everybody wins.

Posted by Bob Rehak on 12/28/2021

1582 Days since Hurricane Harvey